Hormonal Influence on Insulin Transport Through the Blood-Brain

Hormonal influence on insulin transport through the blood- brain barrier and hypothalamic inflammation

A dissertation submitted to the Graduate School of the University of Cincinnati In partial fulfillment of the requirements for the degree of

DOCTORATE OF PHILOSOPHY

In the Graduate Program of Pathobiology and Molecular Medicine of the College of Medicine

December 2016

Aaron A. May

B.S. Agricultural Biochemistry, Iowa State University, Ames, IA

Committee chair: Min Liu, PhD.

ABSTRACT

Insulin is an important effector of energy balance, and it reduces food intake and body weight via its actions in the hypothalamus. Because little or no insulin is produced by the brain, the majority of insulin’s effects in the central nervous system (CNS) are dependent on the transport of insulin through the blood-brain barrier (BBB). My research tests the hypothesis that some of the hormones involved in maintaining energy homeostasis exert their effects by influencing the transport of insulin into the CNS. Specifically, we examined the intestinal hormone, cholecystokinin (CCK), and the gonadal hormone, estrogen, due to their common catabolic actions and their pronounced interactions with central insulin.

After observing that CCK receptors and estrogen receptors are both co-expressed with insulin receptors in endothelial cells of the BBB, we found that CCK, but not estrogen (E2), increased the transport of insulin into the CNS. These findings spawned an additional investigation into how E2 mediates its interaction with insulin signaling in the hypothalamus.

Although insulin signaling was not increased by E2, we observed a significant reduction of inflammatory signaling in the hypothalamus of these E2-treated rats. Further, we found that insulin, itself, can influence inflammation. This work identifies novel mechanisms that influence energy balance and offers a potential approach for enhancing drug delivery into the CNS.

This dissertation is based on the following papers, referred to by their Roman numeral:

I. CCK increases the transport of insulin into the brain II. Estrogen and insulin transport through the blood-brain barrier

A figure in this dissertation also appears in the manuscript:

Using Cerebrospinal Fluid to Understand ingestive Behavior

ACKNOWLEDGEMENTS

Advisors

Min Liu (primary advisor): Thank you for all that you’ve taught me and for always being willing to help or discuss, no matter how busy you were. I have learned so much from you, including your unprecedented ability to plan out experiments in detail and execute them with precision. I am also inspired by your work ethic, your organizational skills, and your attention to detail, which have all been immensely formative. I am very grateful for our friendship and daily conversations about science and life; you have been so thoughtful, kind, and generous. And, of course, I will also always be indebted to you for introducing me to Linda!

Stephen Woods: Thanks for teaching me so many valuable life lessons and for being such a great friend. I’ve thoroughly enjoyed our discussions about science, life, birding, and all things esoteric during our meetings and lunch outings. You always managed to find ways to build my confidence, while also challenging me to keep improving in every way possible and to become a well-rounded scientist. Most importantly, you lead by example; even during some of the roughest phases of health, you went above and beyond in order to help me in any way you could. Your level of perseverance and commitment is beyond inspirational to me.

Committee

Patrick Tso: Thanks for everything you’ve done for me and our lab. You have provided so much sage advice during lab meetings and lab outings. Thanks also for enabling me to meet with visiting faculty and postdoctoral candidates; I’ve had many memorable experiences.

Yve Ulrich-Lai: I have learned so much from you and your lab. Thanks for your willingness to meet whenever I needed another perspective. Your suggestions have always been spot-on. Your scientific rigor is also very impressive and I have the upmost respect for you. SfN dinner outings with you and the lab have been a highlight and I enjoy your witty humor.

Silvana Obici: Thank you for making time to meet despite your demanding schedule and providing keen insights in experimental design. I have also cited your publications in every manuscript I’ve written! It has been a pleasure having you as a member of my committee.

PMM Graduate Program

Laura Woollett: I am so grateful that Dr. Beitz sent me your way. Go ISU! You have been such a great mentor, friend, and inspiration. Your encouragement to pursue community service and to “bloom wherever you’re planted” has enriched my life in numerous ways.

David Askew: Thanks also for all you do with teaching, mentoring, and helping students. It is inspiring. I also appreciate that you let me guest lecture and provided such helpful feedback. Last but not least, thank you for hosting such great parties and inspiring me to cook more.

Jason Blackard: Thanks for your all of your effort in running the program during my first 2 years here. Most importantly, thanks for accepting me into the PMM program, despite that I “made a big mistake wearing jeans to the interview.”

iii

Heather Anderson: Thank you for all of your hard work and dedication to the program. You have been so kind and helpful and it was reassuring to know you were there if I ever needed help.

Mentors

Deborah Clegg: Thanks for being such a strong mentor during a pivotal time of my research, which helped me push through to the finish line. I learned so many valuable insights from our phone conversations and your unique perspective helped me see things in a new light.

Anja Jaeschke: I am so appreciative of your help and assistance during the most challenging time of my research. Your ability to brainstorm and deconstruct my complicated thoughts about molecular biology into concise, testable experiments is unlike anyone I’ve met.

Phil Howles: It has been great getting to know you over the years while carpooling to main campus. I’ve enjoyed our conversations about science and life. You helped me regain an appreciation for my roots as a biochemist.

Matt Wortman: Thanks for your continual support; chapters 5 and 6 of this thesis most likely wouldn’t exist if it wasn’t for you. You always provided such great insights, as well as hilarity.

Diego Perez-Tilve: I greatly admire your thoughtfulness and for being the first person to deeply instill in me what it means to test a hypothesis in the most effective way possible.

Amy Packard: Thank you for always being so kind to meet with me even when you were busy. You provided so many invaluable insights and have always been very thoughtful.

Fellow lab members & troubleshooters; Ling Shen, Yin Liu, Nick Bedel, Alfor Lewis:

Thanks for your unfailing guidance and support. You have all taught me so much. I’ve thoroughly enjoyed the deep life discussions, funny stories, and other antics over the years.

Friends

Thanks to all of you, especially those in the PMM program, namely Gabe and Logan for being so supportive during the tough times and for always providing a good laugh.

Family

To my mom, dad, Marita, Tony and Lisa, cousins, aunts, uncles, and grandparents: Thank you for your unconditional love, support and unwavering confidence in me. I am beyond blessed to have you all encouraging me to pursue my dreams and cheering me on the whole way.

Linda: Life has stepped up to another dimension since I’ve known you. It has been an incredible journey with you so far and I am beyond fortunate to be able to marry the love of my life in April and share the rest of my life with you. You inspire me every single day and have taught me so many invaluable lessons. You have been there with your energy and enthusiasm on so many days when I needed it the most and have been my #1 supporter. No matter where the road ahead leads us, I know it is going to be awesome!

TABLE OF CONTENTS

Abstract………………….…………………………………………………………………….. i

Acknowledgements…………………………………………………………………………...ii

Table of Contents…………………………………………………………………….……… 1

List of Tables and Figures.……………………………… ………………….……………….3

1. Introduction……………………………………………………………………...... 4

1.1 The Obesity Epidemic

1.2 Catabolic Effects of Insulin in the Brain

1.3. The Blood-Brain Barrier

1.4 Aims of Investigation

2. Cholecystokinin Increases the Transport of Insulin into the CNS……………...20

2.1 The Discovery of CCK

2.2 CCK Receptors

2.3 Mechanism of Action for CCK

2.4 CCK Increases the Transport of Insulin into the Central Nervous System (Paper I)

3. Estrogen and Insulin Transport Through the Blood-Brain Barrier…..…….…..30

3.1 Background

3.2 Estrogen and Energy Balance

3.3 Estrogen and Insulin Transport into the CNS (Paper II)

4. Estradiol Prevents Diet-Induced Hypothalamic Inflammation………………… 48

4.1 Sex Differences in the Development of Cardiovascular Disease

4.2 Role of Hypothalamic Inflammation in Obesity and Cardiovascular Disease

4.3 Estradiol Inhibits Hypothalamic Inflammation

4.4 Influence of Inflammation on ERα Expression

5. Insulin Attenuates Hypothalamic Inflammation……………………………….…...61

5.1 Inflammatory Effects of Insulin

5.2 Anti-Inflammatory Effects of Insulin

6. Summary and General Discussion…………………………………………..…....….81

6.1 CCK Increases Insulin Transport into the CNS

6.2 Estrogen and Insulin Transport into the CNS

6.3 Interactions of Estradiol With Insulin Signaling in the Hypothalamus

6.4 Insulin and Hypothalamic Inflammation

6.5 Conclusions

Appendices…………………………………..………………………………………..…....….92

I. Table 1: Antibodies used in the current studies

Table 2: qPCR assays used in the current studies

II. CCK Increases the Transport of Insulin into the CNS (Paper I)

III. Estrogen and Insulin Transport Through the Blood-Brain Barrier (Paper II)

LIST OF TABLES AND FIGURES

Chapter 1

Figure 1.1. Major insulin signaling pathways of the hypothalamus

Chapter 2 - N.A.

Chapter 3

Figure 3.1. Simplified Aims of Chapter 3.

Figure 3.2. Colocalization of ERα and insulin receptors in the mediobasal hypothalamus.

Figure 3.3. Chronic E2 treatment does not increase proximal or distal insulin signaling pathways in the mediobasal hypothalamus.

Chapter 4

Figure 4.1. Diagram of the IKKβ-NF-κB inflammatory signaling pathway.

Figure 4.2. E2 prevents the activation of the hypothalamic IKKα/β signaling pathway during chronic maintenance on a HFD.

Figure 4.3. Proposed mechanism of the anti-inflammatory, cardioprotective effects of E2-ERα signaling.

Figure 4.4. Effects of IKKβ inhibition on the reduction of Pgc1α and ERα mRNA expression during treatment of cultured hypothalamic neurons with palmitate.

Chapter 5

Figure 5.1. Insulin induces inflammation in cultured hypothalamic neurons.

Figure 5.2. Inflammatory cytokine expression at 1.5 hours following the administration of insulin into the third ventricle.

Figure 5.3. Inflammatory cytokine expression at 4 hours following the administration of insulin into the third ventricle.

Figure 5.4. Acute HFD feeding does not induce hypothalamic inflammation.

Figure 5.5. LPS increases hypothalamic inflammation within 4 hours of LPS treatment.

Figure 5.6. Central insulin does not attenuate inflammation of the hypothalamus or hippocampus.

Chapter 6 - N.A.

Chapter 1:

Introduction

1.1 The Obesity Epidemic

The World Health Organization recently estimated that more than 600 million people worldwide are obese [1]. This is alarming, considering that obesity substantially increases the risk of cardiovascular disease, type II diabetes, cancer and liver diseases [2–5]. It has been recently reported that obese individuals have medical costs that exceed those of normal weight individuals by over $1,500 each year [6]. In the United States alone, obesity accounts for an estimated $200 billion in annual medical expenditures and an additional $66 billion per year in indirect costs to the economy [7]. Despite a large concerted effort to develop antiobesity drugs and therapies, the majority of approaches have proven largely ineffective [8]. On the surface, the energy balance equation seems simple; in order to maintain a stable body weight, calories consumed must be equal to the calories expended. However, it is clear that our society is consuming more energy than it is expending. On average, adults in the United States have gained around a pound per year since the 1990’s [9; 10]. If we consider that an average man consumes 912,500 calories per year, then gaining one pound of fat is equivalent to an extra

3,500 calories. This equates to overconsuming a mere 8 calories more than what is expended each day, which is less than a single potato chip or a peanut M&M. This puts in perspective how exquisitely sensitive the energy balance system must be in order to maintain a stable body weight over time.

Unsurprisingly, the energy balance system is complex and is coordinated by numerous signals, many of which are redundant [11–13]. Because all food intake behavior is ultimately coordinated by the brain, it is important to understand how these signals are integrated by their individual actions throughout the body and/or brain. Every time a meal is consumed, the quantity and composition of the food is reported to the brain in real-time [14]. As nutrients pass through the gastrointestinal (GI) tract, they stimulate the secretion of satiation peptides from specialized enteroendocrine cells in the intestines [15–17]. Satiation signals act locally or over

longer distances to inform the brain about the relative number of calories that have been consumed [15–17]. Ingested and stored nutrients can also be sensed directly by the nervous system [12; 14]. A third class of signals, termed adiposity signals, is secreted in direct proportion to the amount of stored fat [18]. Two well-known examples are insulin and leptin, which exert their effects within the central nervous system (CNS) after being transported by a receptor-mediated process through the blood-brain barrier (BBB) [19; 20]. Gonadal hormones can also interact with these energy balance systems in order to ensure that adequate levels of nutrients are consumed and to optimize reproductive success [21]. Collectively, these signals interact among themselves in numerous ways to influence energy balance and the timing of meals [22].

Given that such effective regulatory systems are in place to maintain energy balance, why is obesity such a problem in modern society? It is abundantly clear that diet and lifestyle have the greatest impact on obesity [23–25], but genetic factors place some individuals at a much greater risk for developing obesity than others, when placed in the same obesogenic environment [26; 27]. While mutations in single genes, such as that of leptin [28; 29], the leptin receptor [30], and/or the melanocortin system [31] are known to cause monogenic obesity, these mutations represent a rather small segment of the obese population [32]. The genetic underpinnings of common obesity are polygenic, without a simple Mendelian inheritance pattern

[26], making it difficult to develop targeted solutions.

There are numerous dieting and exercise programs that are effective at reducing body weight in obese individuals [33–37], but in many cases, these programs are ineffective at maintaining weight loss over the long-term [38; 39]. The lack of adherence to the diet can be a major contributor to the inefficacy of these programs [40]. For this reason, other options are needed in order to mitigate obesity and its many associated risk factors. To this end, many antiobesity drugs have been created with the goal of promoting long-term weight loss. When

considering that obesity involves biochemical, neurobehavioral, microbial, and genetic components (among many others) [12; 41–43], it is no surprise that the development of reliable weight-loss drugs has been a difficult endeavor. Hundreds of drugs have failed to obtain FDA approval, due to their inefficacy or adverse side-effects [8], and the few that remain carry many risks or are unable to promote sustained weight loss when treatment is discontinued [8].

Additionally, most weight-loss drugs require a prescription that is restricted to obese individuals

(BMI ≥ 30) or overweight individuals (BMI ≥ 25) with at least one obesity-related comorbidity, such as type II diabetes or high blood pressure due to the risk of side-effects [44; 45]. Despite these challenges, drug development remains in high demand, due to a large preference for non- surgical options for treating obesity [46].

We have recently focused our efforts in therapeutically-targeting the BBB, because it functions as a gateway for many molecules that play important roles in energy balance [47].

Therapeutics that act on the BBB are more likely to be safe and reliable, and to promote sustainable weight loss, because they do not rely on the direct manipulation of neurotransmission [8; 47]. We are particularly interested in increasing the delivery of the pancreatic hormone, insulin, through the BBB, due to its pronounced catabolic action in the brain [48; 49] and because of the abundant evidence that insulin transport into the brain is impaired during obesity [50–52].

1.2 Catabolic Effects of Insulin in the Brain

Insulin, which is secreted by β-cells of the pancreas, is most well-known for facilitating the uptake and storage of glucose and other nutrients into peripheral tissues. In the early 1970s, parabiosis experiments conducted using obese mice and lean mice indicated that a circulating factor was involved in regulating food intake and body weight. It was discovered soon after this time that when insulin is administered intravenously, it appears at elevated levels in the

cerebrospinal fluid (CSF) of dogs, within a short timeframe [53]. Prior to this time, it was thought that proteins did not cross the BBB [47]. In addition, insulin was found to have no effect on the uptake of glucose within the nervous system [54], in contrast to its profound effects on glucose uptake in systemic circulation. The puzzling question at this point was: what is insulin doing in the CNS [53]? Because insulin is secreted in direct proportion to the amount of adiposity [55], it was hypothesized that insulin was a humoral factor that informs the brain of the relative amount of stored fat [55]. As first demonstrated in 1979, central insulin infusion for 2–3 weeks reduces food intake and body weight of baboons in a dose-dependent manner [48]. Following cessation of insulin infusion, food intake returns to its initial values [48; 56]. Numerous studies have since corroborated the catabolic role of central insulin infusion in rodents [57–60], chicks [61], sheep

[62] and humans (24) and have found that its weight-lowering effects are not due to malaise [63;

64]. Because of its role in reporting the level of stored fat to the brain, insulin was later classified as an “adiposity signal” [55].

Insulin exerts its catabolic effects in the CNS via insulin receptors expressed in neurons and glia in the hypothalamus and other regions [65]. When insulin receptors are genetically knocked-out from the CNS, both male and female mice develop obesity [66]. Similarly, site- specific knockdown of insulin receptors in the hypothalamus results in substantial weight gain

[67; 68]. Intrahypothalamic infusion of insulin antibodies disrupts insulin signaling and similarly leads to weight gain [58] and impairs hepatic glucose regulation [69]. The brain regions found to mediate the most profound effects of insulin on energy balance include the arcuate (Arc) and ventromedial (VMH) nuclei of the mediobasal hypothalamus (MBH) and the solitary tract nucleus of the brainstem (NST) [67; 70–76]. It was later demonstrated that insulin’s effects on energy balance and glucose homeostasis are mediated by signaling through phosphoinositol-

4,5-bisphosphate 3-kinase (PI3K) [75; 77; 78], MAPK/Erk-1/2 [79], and other signaling molecules (Fig 1.1).

Figure 1.1. Major insulin signaling pathways of the hypothalamus. Upon the binding of insulin to its receptor, insulin receptor substrates (IRS) are phosphorylated, leading to the phosphorylation and activation of ERK-1/2 and PI3K/Akt signaling pathways, which are known to exert catabolic signaling effects in the hypothalamus that reduce body weight [75; 108].

In contrast, insulin’s action in the periphery is anabolic (leads to weight gain) by facilitating the uptake of glucose from the blood for storage as fat in adipose tissue [80–82]. By shunting available nutrients into storage, insulin promotes weight gain and increased food intake, such that even when rodents administered systemic insulin are pair-fed, they still gain more weight than controls [80]. Consistently, individuals with a relatively lower proportion of insulin in the CNS are more likely to have obesity and the metabolic syndrome [50; 52; 83].

1.3. The Blood-Brain Barrier

In mammals, very little if no insulin is synthesized in the brain, such that the majority of insulin’s effects in the CNS are dependent on the transport of insulin from the blood into the brain. Insulin is transported through the BBB via a saturable, insulin receptor-mediated transport process through brain capillary endothelial cells that comprise the BBB [12; 20; 84;

85]. The primary barriers between the CNS and systemic blood circulation are the BBB and the blood-CSF barrier (BCSFB). These barriers protect the central nervous system (CNS) from pathogens, toxins and other components of the blood, thereby maintaining neuronal function

[86]. BBB endothelial cells form a physical barrier between the blood and the brain interstitial fluid, while the choroid plexus serves the same function in the blood-cerebrospinal-fluid barrier

(BCSFB) and produces CSF by filtering blood through a specialized ependymal cell layer [87].

This dissertation highlights the BBB because insulin transport into the CNS is known to be largely specific to the BBB and is independent of the choroid plexus; modeling studies have revealed that the bulk flow of insulin in the CNS is derived from blood-borne insulin that crosses through the BBB and then is released into the brain interstitial fluid, where it then gains access to insulin receptors expressed throughout the CNS [84; 87].

Endothelial cells of the BBB are connected to one another by intercellular tight junctions that form a highly-selective permeability barrier [86]. The BBB is present in all vessel types within the CNS, including arteries, arterioles, capillary beds, postcapillary venules, draining venules and veins [88]. Capillary beds that come in close contact with neurons are specialized to exchange nutrients, while the postcapillary venules facilitate leukocyte trafficking and influnce immune modulation [89; 90].

Astrocytes and pericytes are also incorporated into the BBB and together comprise the neurovascular unit, which maintains proper function and structure of brain endothelial cells.

Pericytes wrap around the abluminal surfaces of brain microvessels and deposit components of the basement membrane [86]. Pericytes also influence vessel function, stability and integrity, while also regulating capillary diameter and blood flow [91; 92]. Astrocytes extend perivascular end feet, which surround more than 95% of the the abluminal surface of brain vessels. Through these connections, astrocytes promote BBB maintenance and can influence the transport of water, potassium, and other molecules from the endothelium into the CNS [93]. Proteomics

studies have found that 10-15% of all proteins in the neurovascular unit are transporters [94].

Further, astrocytes provide nutrients and growth factors to endothelial cells and thereby influence BBB integrity [86; 95].

Insulin transport into the CNS

Insulin predominately enters the CNS via the BBB [87; 96]. Because the BBB generally only allows the passage of particular proteins that are smaller than 1 nM in diameter [97], the

BBB is equipped with transporter systems that facilitate the entry of specific larger molecules into the CNS [19; 49]. Receptor-mediated transport is used to facilitate the entry of insulin, leptin, and other molecules into the CNS that have key functions in the brain and are too large to freely diffuse [98; 99]. After insulin binds to insulin receptors expressed along the luminal membrane of brain capillary endothelial cells, the ligand-receptor complex is endocytosed and trafficked within vesicles and endosomes until fusing with the membrane on the abluminal

(brain) side of the endothelial cell [49; 100]. The ligand can then be released down its concentration gradient into the brain interstitial fluid [87].

Insulin receptors are also present in particular regions of the brain that are not fully protected by the BBB, referred to as circumventricular organs. These brain regions could potentially sample the levels of insulin in the blood directly without requiring insulin to be transported through the BBB [101; 102]. While insulin receptors are expressed in several circumventricular organs, including the median eminence, subfornical organ and area postrema

[103], the expression of insulin receptors is low relative to that in the rest of the brain [104].

Consistently, circumventricular organs are generally considered to have a relatively minor role in insulin’s effects on energy balance [105], although this remains to be definitively tested. Some specific brain nuclei, such as the Arc of the hypothalamus, express a complete BBB proteome, but have an incomplete brain-CSF barrier [101]. This arrangement allows the Arc to sample the

levels of insulin and other molecules and nutrients in the CSF, while retaining a private milieu from the adjacent median eminence [101]. In this case, the previously-mentioned studies of insulin transport suggest that the concentration of insulin in these regions would be higher than that of the concentration of insulin in the CSF [87; 96]. Therefore, the majority of hypothalamic insulin signaling is expected to be derived from BBB-transport of insulin [49].

1.4 Aims of Investigation

Insulin acts in a dose-dependent manner within the CNS, such that the more insulin is present in the brain, the greater the catabolic effect. Because the primary route of entry for insulin into the CNS is via the BBB, our working hypothesis is that increasing the transport of insulin into the BBB, thus increasing the amount of insulin in the CNS, will be beneficial in ameliorating obesity. We have already explored a related mechanism using the insulin analog, insulin detemir [106], which passes through the BBB to a comparable extent as regular insulin but has a longer half-life within the brain. In this way, the proportion of insulin detemir in the

CNS is increased relative to that of regular insulin and insulin glargine, another long-acting insulin analog. Consistent with our hypothesis, we found that insulin detemir was associated with lower body weight relative to both regular insulin and glargine.

The mechanisms influencing the transport of insulin through the BBB remain poorly understood. My research investigated the possibility that some of the hormones involved in maintaining energy balance exert their effects by influencing the transport of insulin into the

CNS. Specifically, we tested the hypothesis that 1) the intestinal hormone, cholecystokinin

(CCK) exerts its satiating effects in part by increasing insulin transport through the BBB and that

2) the gonadal hormone, estradiol (E2), exerts its catabolic effects and reverses metabolic syndrome by increasing endogenous insulin signaling in the hypothalamus via enhancing the

transport of insulin through the BBB and/or directly interacting with insulin signaling pathways in the hypothalamus.

In the process of identifying the mechanism by which E2 affects hypothalamic insulin signaling, we found that E2 reduces diet-induced hypothalamic inflammation. This observation was intriguing, because a recent report demonstrated that females, but not males are protected from the loss of hypothalamic estrogen receptor-α (ERα) expression and from the impairment of cardiovascular function during the progression of diet-induced obesity (DIO) [107]. Based on this information, we tested the further hypothesis that 3) the inflammatory signaling molecule,

IKKβ, is responsible for the loss of hypothalamic ERα during DIO and that E2 protects ERα expression by inhibiting inflammatory signaling in the hypothalamus. Lastly, we found that insulin also impacts hypothalamic inflammation and we tested the hypothesis that 4) central insulin administration may be beneficial in attenuating diet-induced inflammation. As we continue to understand how the transport of molecules through the BBB is influenced by hormones under normal conditions and in metabolic diseases, this will open the door to new therapeutics that can safely ameliorate obesity and its co-morbidities.

REFERENCES

[1] WHO (2015). Obesity and overweight. Fact sheet number 311. 2015.

[2] Marchesini G, Brizi M, Bianchi G, Tomassetti S, Bugianesi E, Lenzi M, et al. Nonalcoholic fatty liver disease: a feature of the metabolic syndrome. Diabetes 2001; 50 (0012–1797 (Print)): 1844– 1850.

[3] Bastard JP, Maachi M, Lagathu C, Kim MJ, Caron M, Vidal H, et al. Recent advances in the relationship between obesity, inflammation, and insulin resistance. Eur.Cytokine Netw. 2006; 17 (1148–5493 (Print)): 4–12.

[4] Allison DB, Fontaine KR, Manson JE, Stevens J, VanItallie TB. Annual deaths attributable to obesity in the United States. JAMA 1999; 282 (0098–7484 (Print)): 1530–1538.

[5] Grundy SM. Metabolic complications of obesity. Endocrine 2000; 13 (2): 155–165.

[6] Finkelstein EA, Trogdon JG, Cohen JW, Dietz W. Annual medical spending attributable to obesity: payer-and service-specific estimates. Heal. Aff.(Millwood.) 2009; 28 (1544–5208 (Electronic)): w822–w831.

[7] Spieker EA, Pyzocha N. Economic Impact of Obesity. Prim. Care Clin. Off. Pract. 2016; 43 (1): 83– 95.

[8] Rodgers RJ, Tschöp MH, Wilding JPH. Anti-obesity drugs: past, present and future. Dis. Model. Mech. 2012; 5 (5): 621–6.

[9] Olgden C, Carroll M, Kit B, Flegal K. Prevalence of Obesity in the United States, 2009–2010. NCHS Data Brief No. 82.

[10] Flegal KM, Carroll MD, Ogden CL, Curtin LR. Prevalence and trends in obesity among US adults, 1999-2008. JAMA 2010; 303 (1538–3598 (Electronic)): 235–241.

[11] Beck B. Neuropeptides and obesity. Nutr. 2000.Oct.;16.(10)916.-23. 2000; 16: 916–923.

[12] Schwartz MW, Woods SC, Porte Jr. D, Seeley RJ, Baskin DG. Central nervous system control of food intake. Nat. 2000.Apr 6.;404.(6778.)661.-71. 2000; 404 (0028–0836): 661–671.

[13] Woods SC, Seeley RJ, Porte DJ, Schwartz MW. Signals That Regulate Food Intake and Energy Homeostasis. Science (80-. ). 1998; 280 (5368): 1378–1383.

[14] Morton GJ, Meek TH, Schwartz MW. Neurobiology of food intake in health and disease. Nat. Rev. Neurosci. 2014; 15 (6): 367–78.

[15] Grundy D. Signalling the state of the digestive tract. Auton.Neurosci. 2006; 125 (1566–0702 (Print)): 76–80.

[16] Cummings DE, Overduin J. Gastrointestinal regulation of food intake. J Clin.Invest 2007; 117 (0021–9738 (Print)): 13–23.

[17] Bray GA. Afferent signals regulating food intake. Proc.Nutr.Soc. 2000; 59 (0029–6651 SB–IM): 373–384.

[18] Woods SC, Seeley RJ. Adiposity signals and the control of energy homeostasis. Nutrition 2000; 16 (10): 894–902.

[19] Banks WA. Role of the blood-brain barrier in the evolution of feeding and cognition. Ann. N. Y. Acad. Sci. 2012; 1264 (1): 13–19.

[20] Banks WA, Owen JB, Erickson MA. Insulin in the brain: there and back again. Pharmacol. Ther. 2012; 136 (1): 82–93.

[21] Wade GN. Gonadal hormones and behavioral regulation of body weight. Physiol Behav. 1972; 8 (0031–9384 (Print)): 523–534.

[22] Strubbe JH, Woods SC. The timing of meals. Psychol.Rev. 2004; 111 (0033–295X (Print)): 128– 141.

[23] Neel J V. The “Thrifty Genotype” in 1998. Nutr. Rev. 2009; 57 (5): 2–9.

[24] Bennett B, Sothern MS. Diet, exercise, behavior: the promise and limits of lifestyle change. Semin. Pediatr. Surg. 2009; 18 (3): 152–158.

[25] Hall KD. Diet versus exercise in “the biggest loser” weight loss competition. Obesity 2013; 21 (5): 957–959.

[26] Walley AJ, Asher JE, Froguel P. The genetic contribution to non-syndromic human obesity. Nat. Rev. Genet. 2009; 10 (7): 431–442.

[27] Stunkard AJ, Foch TT, Hrubec Z, Bray GA Y DA, CB D, L A, et al. A Twin Study of Human Obesity. JAMA J. Am. Med. Assoc. 1986; 256 (1): 51.

[28] Strobel A, Issad T, Camoin L, Ozata M, Strosberg AD. A leptin missense mutation associated with hypogonadism and morbid obesity. Nat. Genet. 1998; 18 (3): 213–215.

[29] OʼRahilly S, Montague CT, Farooqi IS, Whitehead JP, Soos MA, Rau H, et al. Congenital leptin deficiency is associated with severe early-onset obesity in humans. Nature 1997; 387 (6636): 903–908.

[30] Froguel P, Clément K, Vaisse C, Lahlou N, Cabrol S, Pelloux V, et al. A mutation in the human leptin receptor gene causes obesity and pituitary dysfunction. Nature 1998; 392 (6674): 398–401.

[31] Krude H, Biebermann H, Luck W, Horn R, Brabant G, Gruters A. Severe early-onset obesity, adrenal insufficiency and red hair pigmentation caused by POMC mutations in humans. Nat.Genet. 1998; 19 (1061–4036): 155–157.

[32] Santini F, Maffei M, Pelosini C, Salvetti G, Scartabelli G, Pinchera A. Melanocortin-4 receptor mutations in obesity. Adv. Clin. Chem. 2009; 48: 95–109.

[33] Jensen MD, Ryan DH, Apovian CM, Ard JD, Comuzzie AG, Donato KA, et al. 2013 AHA/ACC/TOS Guideline for the Management of Overweight and Obesity in Adults. J. Am. Coll. Cardiol. 2014; 63 (25): 2985–3023.

[34] Carlson J. Of the Most Common Weight-loss Programs, Weight Watchers, Jenny Craig, Nutrisystem, and Medifast, which is the most effective? 2014 Award. Excell. Student Res. Creat.

Act. Doc. 2014;

[35] Gudzune KA, Doshi RS, Mehta AK, Chaudhry ZW, Jacobs DK, Vakil RM, et al. Efficacy of commercial weight-loss programs: an updated systematic review. Ann. Intern. Med. 2015; 162 (7): 501–12.

[36] Halseth AE, Bracy DP, Wasserman DH. Overexpression of hexokinase II increases insulinand exercise-stimulated muscle glucose uptake in vivo. Am. J. Physiol. 1999; 276 (1 Pt 1): E70-7.

[37] Eckel RH, Jakicic JM, Ard JD, de Jesus JM, Houston Miller N, Hubbard VS, et al. 2013 AHA/ACC guideline on lifestyle management to reduce cardiovascular risk: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation 2014; 129 (25 Suppl 2): S76-99.

[38] Schwartz MW. An inconvenient truth about obesity. Mol. Metab. 2012; 1 (1–2): 2–4.

[39] Knowler WC, Fowler SE, Hamman RF, Christophi CA, Hoffman HJ, Brenneman AT, et al. 10-year follow-up of diabetes incidence and weight loss in the Diabetes Prevention Program Outcomes Study. Lancet (London, England) 2009; 374 (9702): 1677–86.

[40] Makris A, Foster GD. Dietary approaches to the treatment of obesity. Psychiatr. Clin. North Am. 2011; 34 (4): 813–27.

[41] Beck B. KO’s and organisation of peptidergic feeding behavior mechanisms. Neurosci.Biobehav.Rev. 2001; 25 (0149–7634 SB–IM): 143–158.

[42] Zeevi D, Korem T, Zmora N, Israeli D, Rothschild D, Weinberger A, et al. Personalized Nutrition by Prediction of Glycemic Responses. Cell 2015; 163 (5): 1079–94.

[43] Hebebrand J, Albayrak Ö, Adan R, Antel J, Dieguez C, de Jong J, et al. “Eating addiction”, rather than “food addiction”, better captures addictive-like eating behavior. Neurosci. Biobehav. Rev. 2014; 47: 295–306.

[44] Jain SS, Ramanand SJ, Ramanand JB, Akat PB, Patwardhan MH, Joshi SR. Evaluation of efficacy and safety of orlistat in obese patients. Indian J. Endocrinol. Metab. 2011; 15 (2): 99–104.

[45] Weigle DS. Pharmacological therapy of obesity: past, present, and future. J.Clin.Endocrinol.Metab 2003; 88 (0021–972X (Print)): 2462–2469.

[46] Madura JA, Dibaise JK. Quick fix or long-term cure? Pros and cons of bariatric surgery. F1000 Med. Rep. 2012; 4: 19.

[47] Banks WA. Brain meets body: the blood-brain barrier as an endocrine interface. Endocrinology 2012; 153 (9): 4111–9.

[48] Woods SC, Lotter EC, McKay LD, Porte Jr. D. Chronic intracerebroventricular infusion of insulin reduces food intake and body weight of baboons. Nature 1979; 282 (0028–0836 (Print)): 503– 505.

[49] Begg DP, Woods SC. The central insulin system and energy balance. Handb.Exp.Pharmacol. 2012; 209 (0171–2004 (Print)): 111–129.

[50] Begg DP, Mul JD, Liu M, Reedy BM, D’Alessio DA, Seeley RJ, et al. Reversal of diet-induced obesity increases insulin transport into cerebrospinal fluid and restores sensitivity to the anorexic action of central insulin in male rats. Endocrinology 2013; 154 (3): 1047–54.

[51] Urayama A, Banks WA. Starvation and triglycerides reverse the obesity-induced impairment of insulin transport at the blood-brain barrier. Endocrinology 2008; 149 (7): 3592–3597.

[52] Kern W, Benedict C, Schultes B, Plohr F, Moser A, Born J, et al. Low cerebrospinal fluid insulin levels in obese humans. Diabetologia 2006; 49 (11): 2790–2.

[53] Woods SC, Porte D. Relationship between plasma and cerebrospinal fluid insulin levels of dogs. Am. J. Physiol. 1977; 233 (4): E331-4.

[54] Rafaelsen OJ. Action of insulin on glucose uptake of rat brain slices and isolated rat cerebellum. J. Neurochem. 1961; 7 (1): 45–51.

[55] Woods SC, Seeley RJ. Insulin as an adiposity signal. Int.J.Obes.Relat Metab Disord. 2001; 25 Suppl 5 (0307–0565 (Print)): S35–S38.

[56] Woods SC, Jr PD. Insulin and the set-point regulation of body weight. In: Hunger: basic mechanisms and clinical implications. Raven, New York; 1976. pp. 273–280.

[57] Brown LM, Clegg DJ, Benoit SC, Woods SC. Intraventricular insulin and leptin reduce food intake and body weight in C57BL/6J mice. Physiol Behav. 2006; 89 (0031–9384 (Print)): 687–691.

[58] McGowan MK, Andrews KM, Grossman SP. Chronic intrahypothalamic infusions of insulin or insulin antibodies alter body weight and food intake in the rat. Physiol. Behav. 1992; 51 (4): 753– 766.

[59] Clegg DJ, Riedy CA, Smith KAB, Benoit SC, Woods SC. Differential sensitivity to central leptin and insulin in male and female rats. Diabetes 2003; 52 (3): 682–687.

[60] Clegg DJ, Brown LM, Woods SC, Benoit SC. Gonadal hormones determine sensitivity to central leptin and insulin. Diabetes 2006; 55 (4): 978–987.

[61] Honda K, Kamisoyama H, Saneyasu T, Sugahara K, Hasegawa S. Central administration of insulin suppresses food intake in chicks. 2007.

[62] Miller DW, Bennett EJ, Harrison JL, Findlay PA, Adam CL. Adiposity and plane of nutrition influence reproductive neuroendocrine and appetite responses to intracerebroventricular insulin and neuropeptide-Y in sheep. Reprod. Fertil. Dev. 2011; 23 (2): 329–38.

[63] Chavez M, Seeley RJ, Woods SC. A comparison between effects of intraventricular insulin and intraperitoneal lithium chloride on three measures sensitive to emetic agents. Behav.Neurosci. 1995; 109: 547–550.

[64] Woods SC, Chavez M, Park CR, Riedy C, Kaiyala K, Richardson RD, et al. The evaluation of insulin as a metabolic signal influencing behavior via the brain. Neurosci.Biobehav.Rev. 1996; 20: 139– 144.

[65] Marks JL, Porte Jr. D, Stahl WL, Baskin DG. Localization of insulin receptor mRNA in rat brain by in situ hybridization. Endocrinology 1990; 127 (0013–7227 (Print)): 3234–3236.

[66] Bruning JC, Gautam D, Burks DJ, Gillette J, Schubert M, Orban PC, et al. Role of brain insulin receptor in control of body weight and reproduction. Science (80-. ). 2000; 289 (0036–8075): 2122–2125.

[67] Obici S, Feng Z, Karkanias G, Baskin DG, Rossetti L. Decreasing hypothalamic insulin receptors causes hyperphagia and insulin resistance in rats. Nat. Neurosci. 2002; 5 (6): 566–572.

[68] Grillo CA, Tamashiro KL, Piroli GG, Melhorn S, Gass JT, Newsom RJ, et al. Lentivirus-mediated downregulation of hypothalamic insulin receptor expression. Physiol Behav. 2007; 92 (0031–9384 (Print)): 691–701.

[69] Obici S, Zhang BB, Karkanias G, Rossetti L. Hypothalamic insulin signaling is required for inhibition of glucose production. Nat. Med. 2002; 8 (12): 1376–1382.

[70] Havrankova J, Brownstein M, Roth J. Insulin and insulin receptors in rodent brain. Diabetologia 1981; 20 Suppl (0012–186X (Print)): 268–273.

[71] Hopkins DF, Williams G. Insulin receptors are widely distributed in human brain and bind human and porcine insulin with equal affinity. Diabet. Med. 1997; 14 (12): 1044–50.

[72] Chen B-R, Ching-Jiunn T, Lu P-J. Neuronal insulin resistance in the nucleus tractus solitarii induces hypertension in fructose-fed rats.

[73] Huang H-N, Lu P-J, Lo W-C, Lin C-H, Hsiao M, Tseng C-J. In situ Akt phosphorylation in the nucleus tractus solitarii is involved in central control of blood pressure and heart rate. Circulation 2004; 110 (16): 2476–83.

[74] Filippi BM, Yang CS, Tang C, Lam TKT. Insulin Activates Erk1/2 Signaling in the Dorsal Vagal Complex to Inhibit Glucose Production. Cell Metab. 2012; 16 (4): 500–510.

[75] Niswender KD, Morrison CD, Clegg DJ, Olson R, Baskin DG, Myers Jr. MG, et al. Insulin activation of phosphatidylinositol 3-kinase in the hypothalamic arcuate nucleus: a key mediator of insulin- induced anorexia. Diabetes 2003; 52 (0012–1797 (Print)): 227–231.

[76] Diggs-Andrews KA, Zhang X, Song Z, Daphna-Iken D, Routh VH, Fisher SJ. Brain insulin action regulates hypothalamic glucose sensing and the counterregulatory response to hypoglycemia. Diabetes 2010; 59 (9): 2271–80.

[77] Gelling RW, Morton GJ, Morrison CD, Niswender KD, Myers MG, Rhodes CJ, et al. Insulin action in the brain contributes to glucose lowering during insulin treatment of diabetes. Cell Metab 2006; 3 (1550–4131 (Print)): 67–73.

[78] Hill JW, Elias CF, Fukuda M, Williams KW, Berglund ED, Holland WL, et al. Direct Insulin and Leptin Action on Pro-opiomelanocortin Neurons Is Required for Normal Glucose Homeostasis and Fertility. Cell Metab 2010; 11 (1932–7420 (Electronic)): 286–297.

[79] Mayer CM, Belsham DD. Insulin directly regulates NPY and AgRP gene expression via the MAPK MEK/ERK signal transduction pathway in mHypoE-46 hypothalamic neurons. Mol. Cell. Endocrinol. 2009; 307 (1): 99–108.

[80] Torbay N, Bracco EF, Geliebter A, Stewart IM, Hashim SA. Insulin increases body fat despite

control of food intake and physical activity. Am. J. Physiol. 1985; 248 (1 Pt 2): R120-4.

[81] Bluher M, Michael MD, Peroni OD, Ueki K, Carter N, Kahn BB, et al. Adipose tissue selective insulin receptor knockout protects against obesity and obesity-related glucose intolerance. Dev.Cell 2002; 3 (1534–5807 (Print)): 25–38.

[82] Carver C. Insulin treatment and the problem of weight gain in type 2 diabetes. Diabetes Educ. 2006; 32 (6): 910–7.

[83] Owen OE, Reichard GA, Boden G, Shuman C. Comparative measurements of glucose, beta- hydroxybutyrate, acetoacetate, and insulin in blood and cerebrospinal fluid during starvation. Metabolism 1974; 23 (1): 7–14.

[84] Baura GD, Foster DM, Porte Jr. D, Kahn SE, Bergman RN, Cobelli C, et al. Saturable transport of insulin from plasma into the central nervous system of dogs in vivo. A mechanism for regulated insulin delivery to the brain. J.Clin.Invest 1993; 92 (0021–9738 (Print)): 1824–1830.

[85] Schwartz MW, Figlewicz DP, Baskin DG, Woods SC, Porte Jr. D. Insulin in the brain: a hormonal regulator of energy balance. Endocr.Rev. 1992; 13 (0163–769X (Print)): 387–414.

[86] Obermeier B, Daneman R, Ransohoff RM. Development, maintenance and disruption of the blood-brain barrier. Nat. Med. 2013; 19 (12): 1584–1596.

[87] Schwartz MW, Bergman RN, Kahn SE, Taborsky GJ, Fisher LD, Sipols AJ, et al. Evidence for entry of plasma insulin into cerebrospinal fluid through an intermediate compartment in dogs. Quantitative aspects and implications for transport. J. Clin. Invest. 1991; 88 (4): 1272–81.

[88] Daneman R. The blood-brain barrier in health and disease. Ann. Neurol. 2012; 72 (5): 648–672.

[89] Aird WC. Phenotypic Heterogeneity of the Endothelium: I. Structure, Function, and Mechanisms. Circ. Res. 2007; 100 (2): 158–173.

[90] Aird WC. Phenotypic Heterogeneity of the Endothelium: II. Representative Vascular Beds. Circ. Res. 2007; 100 (2): 174–190.

[91] Allt G, Lawrenson JG. Pericytes: Cell Biology and Pathology. Cells Tissues Organs 2001; 169 (1): 1– 11.

[92] Winkler EA, Bell RD, Zlokovic B V. Central nervous system pericytes in health and disease. Nat. Neurosci. 2011; 14 (11): 1398–1405.

[93] Neuhaus J. Orthogonal arrays of particles in astroglial cells: Quantitative analysis of their density, size, and correlation with intramembranous particles. Glia 1990; 3 (4): 241–251.

[94] Enerson BE, Drewes LR. The rat blood-brain barrier transcriptome. J. Cereb. Blood Flow Metab. 2006; 26 (7): 959–973.

[95] Gee JR, Keller JN. Astrocytes: regulation of brain homeostasis via apolipoprotein E. Int.J Biochem. Biol. 2005; 37 (1357–2725 (Print)): 1145–1150.

[96] Schwartz MW, Sipols A, Kahn SE, Lattemann DF, Taborsky GJ. J, Bergman RN, et al. Kinetics and specificity of insulin uptake from plasma into cerebrospinal fluid. Am J Physiol Endocrinol Metab

1990; 259 (3): E378-383.

[97] Chen H, Konofagou EE. The size of blood-brain barrier opening induced by focused ultrasound is dictated by the acoustic pressure. J. Cereb. Blood Flow Metab. 2014; 34 (7): 1197–204.

[98] Pardridge WM. CNS drug design based on principles of blood-brain barrier transport. J Neurochem. 1998; 70: 1781–1792.

[99] Zlokovic B V. Cerebrovascular permeability to peptides: manipulations of transport systems at the blood-brain barrier. Pharm.Res 1995; 12 (0724–8741 SB–IM): 1395–1406.

[100] Duffy KR, Pardridge WM. Blood-brain barrier transcytosis of insulin in developing rabbits. 1987.

[101] Rodriguez EM, Blazquez JL, Guerra M. The design of barriers in the hypothalamus allows the median eminence and the arcuate nucleus to enjoy private milieus: the former opens to the portal blood and the latter to the cerebrospinal fluid. Peptides 2010; 31 (1873–5169 (Electronic)): 757–776.

[102] Norsted E, Gömüç B, Meister B. Protein components of the blood–brain barrier (BBB) in the mediobasal hypothalamus. J. Chem. Neuroanat. 2008; 36 (2): 107–121.

[103] Hill JM, Lesniak MA, Pert CB, Roth J. Autoradiographic localization of insulin receptors in rat brain: prominence in olfactory and limbic areas. Neuroscience 1986; 17 (0306–4522 (Print)): 1127–1138.

[104] Havrankova J, Roth J, Brownstein M. Insulin receptors are widely distributed in the central nervous system of the rat. Nature 1978; 272 (0028–0836 (Print)): 827–829.

[105] Plum L, Schubert M, Bruning JC. The role of insulin receptor signaling in the brain. Trends Endocrinol.Metab 2005; 16 (1043–2760 (Print)): 59–65.

[106] Begg DP, May AA, Mul JD, Liu M, D’Alessio DA, Seeley RJ, et al. Insulin Detemir Is Transported From Blood to Cerebrospinal Fluid and Has Prolonged Central Anorectic Action Relative to NPH Insulin. Diabetes 2015; 64 (7): 2457–66.

[107] Morselli E, Fuente-Martin E, Finan B, Kim M, Frank A, Garcia-Caceres C, et al. Hypothalamic PGC- 1α protects against high-fat diet exposure by regulating ERα. Cell Rep. 2014; 9 (2): 633–45.

[108] Mirshamsi S, Laidlaw HA, Ning K, Anderson E, Burgess LA, Gray A, et al. Leptin and insulin stimulation of signalling pathways in arcuate nucleus neurones: PI3K dependent actin reorganization and KATP channel activation. BMC.Neurosci. 2004; 5 (1471–2202 (Electronic)): 54.

[109] Bakkar N, Guttridge DC. NF-kappaB signaling: a tale of two pathways in skeletal myogenesis. Physiol. Rev. 2010; 90 (2): 495–511.

Chapter 2:

Cholecystokinin Increases the Transport of Insulin into the CNS

2.1 The Discovery of CCK

Up until the early 1970s, the critical factor(s) responsible for the termination of meals were still unknown. The glucostatic theory was the preeminent model for influencing food intake behavior at the time, which reasoned that as levels of glucose increase during a meal, this increases feelings of fullness; similarly, when blood glucose levels decrease, this increases feelings of hunger [1]. In 1972, Gerard Smith et al. provided strong evidence against this theory for initiating normal meals [2]. The other prevailing theories at the time reasoned that gastric-contraction was responsible for the initiation of food intake and that gastric expansion facilitated the termination of meals [3]. However, these theories were largely disproven by the early 1970s [4; 5].

It was postulated by Gibbs and Smith that the termination of meals would most likely involve some sort of interaction of nutrients with the digestive tract, because the passage of ingested food through the stomach and small intestine stimulated the release of many hormones. Many of these hormones had been individually isolated, identified and synthesized by that time. Importantly, these gastrointestinal hormones had the potential to influence the size of individual meals, because they were released before the meal was terminated.

The first candidate peptide examined was cholecystokinin (CCK), which is secreted by I cells within the mucosa of the upper small intestine in response to ingested fats and proteins [6;

7]. CCK was discovered long before this time, in 1902, as a substance from dog intestinal extracts that stimulates pancreatic secretion [8] and gallbladder contraction [9]. It was given its name because of its effects on the gallbladder, with Latin routes of chole (bile), cysto (sac), kinin

(movement or contraction) [9].

When male Sprague Dawley rats were injected with impure extracts of CCK and synthetic CCK-8, they ate significantly smaller meals in a dose-dependent manner [6]. These

effects were specific to food intake and did not induce malaise or alter water intake [10]. It was uncertain, however, whether endogenous CCK played a physiological role in satiation [11].

Subsequent experiments found that antagonists of the CCK-1 receptor (CCK-1R) prolonged meal duration and increased food consumption within the meal [12; 13], supporting the role of endogenous CCK in inducing feelings of satiation. These findings were eventually confirmed in humans [14].

2.2 CCK Receptors

CCK predominately acts through two receptors; CCK-1R (originally named CCKAR; discovered in the alimentary tract) and CCK-2R (originally named CCKBR; discovered in the brain). CCK-1R is reportedly expressed at high levels in the duodenum, jejunum and acinar cells of the pancreas, and is also expressed in nerves, smooth muscles, aortic endothelial cells, mucosal cells, and the brain [15; 16]. CCK-2R is primarily expressed in the brain, but is also expressed in the stomach, intestinal tract, and other regions.

CCK-1R and -2R share 50% homology and are both G-coupled protein receptors, which bind CCK at comparable affinities [9]. Studies utilizing CCK receptor antagonists have defined

CCK-1R as necessary for the satiating effects of CCK. CCK-1R antagonists block the anorectic effects of CCK-8 and increase the size of normal meals [12; 14; 17], whereas CCK-2R antagonists have no such effects [14; 17]. CCK-1R antagonists also attenuate the satiating effects of intestinal infusions of lipid and protein [15]. Further, CCK-8 is unable to reduce food intake in CCK-1R-KO mice [18] and in OLETF rats that harbor a deletion of the CCK-1R [19]. In contrast, CCK-2R KO mice are fully responsive to the anorectic effect of CCK-8, indicating that

CCK-2R does not contribute to the effects of CCK-induced satiation [20].

2.3 Mechanism of Action of CCK

An important question is how endogenous CCK acts through CCK-1R to exert its catabolic effects. In response to ingested food, CCK is secreted from the duodenum and acts locally on vagal nerve fibers that express CCK-1Rs [15]. When stimulated by CCK, these nerve branches send signals to the hindbrain, which then relays the satiating signal to other brain regions, including the hypothalamus [21–23]. When the vagus nerve is severed near the duodenum [21; 24] or along the afferent fibers feeding into the hindbrain [25], this significantly reduces the ability of CCK to reduce meal size. CCK may also potentiate feelings of satiation by directly binding to CCK-1Rs expressed on intestinal smooth muscle and adjacent nerves to slow gastric emptying [9]. CCK can also act in particular regions of the brain that do not express a complete BBB-proteome, such as the area postrema [9].

Our lab and others have found that when insulin is elevated in the brain prior to a meal, it increases the anorectic effect of CCK in a dose-dependent manner in rats [26] and baboons

[27]. Similarly, leptin enhances the satiating potency of CCK [28–31]. This implies that normally, satiation signals and adiposity signals cooperate to exert their catabolic action. When an individual gains fat, this increases the amount of insulin and leptin within the circulation, and these then enter and stimulate the brain. Thus, when this individual begins eating a meal, the secreted satiation signals will reduce meal size to a greater extent, such that body weight will be brought back toward normal. This is consistent with homeostatic models of energy balance [32;

33]. Importantly, CCK can also enhance the anorectic effects of leptin [29; 30; 34], indicating that satiation signals can also exert their effects, in part, by enhancing the strength of adiposity signals.

These interactions could theoretically occur anywhere along the path of vagal afferent fibers to the NST and the rest of the brain. Recent findings suggest that a portion of this cooperative catabolic interaction between CCK and leptin results from increasing the entry of leptin into the CNS [35; 36]. In order to determine whether the effect of CCK on transport

through the BBB is specific to leptin, or extends to other adiposity signals, we tested the hypothesis that CCK enhances the transport of insulin into the CNS. This investigation became the basis of Paper I.

2.4 CCK Increases the Transport of Insulin into the Central Nervous System

We first examined the possibility that CCK acts directly at the BBB in order to exert its effects. The expression of CCK-1R was assessed, because this receptor is responsible for the satiating effects of CCK, and when CCK-1R is antagonized, it prevents the ability of CCK to increase leptin transport into the CNS [14; 17]. Brain microvessels comprising the BBB were isolated using modifications to previously published techniques [37; 38], as fully described in

Paper I, Methods, Section 2.2. In brief, microvessels were isolated from rat forebrains, because the hindbrain has a relatively lower density of microvessels and a higher proportion of lipids [37].

This was performed using a Brain Block™ (Stoelting, Chicago, IL), by making a coronal slice, caudal to the mammillary bodies. In this process, we also removed the cerebellum, which does not play a major role in energy balance [39]. The large choroid plexi of the lateral ventricles were removed from the forebrain prior to homogenization, in order to avoid interference of the choroid plexus in downstream gene expression applications. Because the choroid plexus of the fourth ventricle was also removed during the disposal of the hindbrain, there was only one remaining segment of choroid plexus (in the third ventricle) that was not removed. However, this constitutes a relatively negligible fraction of microvessels relative to the entire amount present in the brain.

Microvessels were then isolated from the forebrain of male rats after a process of gentle homogenization, centrifugation, and filtering through nylon mesh to obtain microvessels with a vessel diameter in the range of 20-100 nm. After verifying the relative purity of the microvessel preparations, as previously demonstrated (Paper II, Fig 1A), the expression of CCK-1R was

assessed using immunohistochemistry, Western blotting, and qPCR. We observed substantial expression of CCK-1R mRNA and protein at the BBB (Paper I, Fig 1A,B,E,G), which was comparable to that of the leptin receptor transcript (Paper I, Fig 1L). CCK-1R is therefore positioned to directly influence the transport of leptin (and potentially insulin) through the BBB.

CCK-1R expression was also observed at the choroid plexus (Paper I, Fig 1F,H), which contributes to leptin transport [35; 40; 41], but is not the major contributor for the transport of insulin into the CNS [42; 43].

In order to test the hypothesis that CCK enhances insulin transport into the CNS, we then used our well-established CSF-collection technique [44], which is described fully in in

Paper I, Methods, Section 2.3. Rats were administered 0.3 U/kg of insulin NPH (Novo Nordisk,

Denmark) 30-minutes prior to CSF collection. 15 minutes later, half of the rats received ip CCK-

8 (10 μg/rat) or vehicle (saline). After 10 minutes, rats were anesthetized with isoflurane and underwent CSF collection exactly 30 minutes after injection. Blood was then immediately withdrawn. We found that insulin levels in the CSF in CCK-treated rats were more than twice that of controls that received saline (Paper I, Fig 2B), and that there were no differences in the levels of plasma insulin at this time (Paper I, Fig 2A).

Considering the short 15-minute time frame in which CCK effectively increased the appearance of insulin into the CNS, it is most likely that CCK acts directly at the BBB. Blood- borne CCK cannot cross the BBB [11], ruling out the possibility that CCK prevents the degradation of insulin by insulin degrading enzyme in the CNS. It is also possible that CCK could interact with IDE to prevent insulin from being degraded during its transport into the CNS

[45; 46]. However, this interaction is unlikely, considering that the downregulation of IDE activity has not been reported to occur in such a rapid timeframe. CCK also did not affect the plasma levels of insulin in this study, eliminating the possibility that CCK increased insulin transport into the CNS by increasing the levels of circulating insulin. While it has been reported that CCK

stimulates insulin secretion [47; 48], these insulinotropic effects can be diminished after fasting

[35], as appeared to be the case in this study. Collectively, these findings allowed the strong conclusion that CCK enhances the transport of insulin into the CNS and that this is potentially mediated by CCK-1R that is expressed by endothelial cells of the BBB [49]. This is consistent with the hypothesis that CCK mediates its effects on satiation by increasing the entry of catabolic signals into the CNS.

A major question arising from this work is whether CCK increases BBB permeability, in general, or whether this effect is specific to certain molecules. Increasing the speed at which proteins are trafficked through the BBB following a meal might allow cytokines or other molecules to enter the brain more rapidly, which could also contribute a portion of CCK’s satiating effects. However, another possibility is that CCK reduces BBB integrity and thereby increases the permeability of the BBB to a wider array of molecules, which could be detrimental.

It will be important to rule out the latter possibility in the event that detrimental effects may arise.

Neuronal activity can also influence the permeability and transport systems of the BBB by directly communicating with the endothelium [50]. There is a possibility that CCK increased insulin transport in this manner, by acting via afferent nerves that extend from the duodenum or other peripheral organs. Histamine is one of very few neurotransmitters that can influence BBB permeability at physiological levels, but histaminergic neurons have a rather limited distribution in the hypothalamus [50]. Thus, this mechanism seems unlikely to increase the entry of insulin into the CNS to the extent observed in our studies. Further, the expression of c-Fos in the brain following systemic CCK administration occurs in relatively discrete patterns [51] that are not consistent with a seemingly global increase in insulin transport throughout the brain that were observed in our studies.

In the event that the transport rate of specific molecules is effected by CCK, an important follow-up study involves determining the critical CCK receptor(s) involved in enhancing insulin transport through the BBB. Considering that the antagonist for CCK-1R, but not for CCK-2R, blocks the ability of CCK to increase leptin transport into the CNS [35] and also inhibits the anorectic effects of CCK [17], we hypothesize that CCK-1R is the receptor involved in enhancing the transport of insulin and leptin into the CNS. Identifying the mechanisms by which

CCK enhances the transport of adiposity signals into the CNS could aid in the development of novel antiobesity therapeutics. Further, this system could be targeted in order to improve the efficacy of drug delivery into the CNS [52].

REFERENCES

[1] Van Itallie TB, Beaudoin R, Mayer J. Arteriovenous glucose differences, metabolic hypoglycemia and food intake in man. Am. J. Clin. Nutr. 1953; 1 (3): 208–217.

[2] Smith GP, Gibbs J, Strohmayer AJ, Stokes PE. Threshold doses of 2-deoxy-D-glucose for hyperglycemia and feeding in rats and monkeys. Am. J. Physiol. 1972; 222 (1): 77–81.

[3] Cannon WB, Washburn AL. An explanation of hunger. 1911. Obes. Res. 1993; 1 (6): 494–500.

[4] Penick SB, Smith GP, Wieneke K, Hinkle LE. Experimental Evaluation of the Relationship Between Hunger and Gastric Motility. Am. J. Physiol. 1963; 205: 421–6.

[5] Stunkard AJ, Fox S. The relationship of gastric motility and hunger. A summary of the evidence. Psychosom. Med. 33 (2): 123–34.

[6] Gibbs J, Young RC, Smith GP. Cholecystokinin decreases food intake in rats. J. Comp. Physiol. Psychol. 1973; 84 (3): 488–95.

[7] Gibbs J, Young RC, Smith GP. Cholecystokinin elicits satiety in rats with open gastric fistulas. Nature 1973; 245 (0028–0836 (Print)): 323–325.

[8] Bayliss WM, Starling EH. The mechanism of pancreatic secretion. J. Physiol. 1902; 28 (5): 325.

[9] Sayegh AI. The role of cholecystokinin receptors in the short-term control of food intake. Prog. Mol. Biol. Transl. Sci. 2013; 114: 277–316.

[10] Pi-Sunyer X, Kissileff HR, Thornton J, Smith GP. C-terminal octapeptide of cholecystokinin decreases food intake in obese men. Physiol. Behav. 1982; 29 (4): 627– 30.

[11] Baldwin BA, Parrott RF, Ebenezer IS. Food for thought: a critique on the hypothesis that endogenous cholecystokinin acts as a physiological satiety factor. Prog. Neurobiol. 1998; 55 (5): 477–507.

[12] Weatherford SC, Chiruzzo FY, Laughton WB. Satiety induced by endogenous and exogenous cholecystokinin is mediated by CCK-A receptors in mice. Am.J Physiol 1992; 262 (0002–9513 (Print)): R574–R578.

[13] Brenner L, Ritter RC. Peptide cholesystokinin receptor antagonist increases food intake in rats. Appetite 1995; 24 (1): 1–9.

[14] Moran TH, Ameglio PJ, Peyton HJ, Schwartz GJ, McHugh PR. Blockade of type A, but not type B, CCK receptors postpones satiety in rhesus monkeys. Am. J. Physiol. 1993; 265 (3 Pt 2): R620-4.

[15] Cummings DE, Overduin J. Gastrointestinal regulation of food intake. J Clin.Invest 2007; 117 (0021–9738 (Print)): 13–23.

[16] Konturek SJ, Konturek JW, Pawlik T, Brzozowski T. Brain-gut axis and its role in the control of food intake. J.Physiol Pharmacol. 2004; 55 (0867–5910): 137–154.

[17] Moran TH, Ameglio PJ, Schwartz GJ, McHugh PR. Blockade of type A, not type B, CCK receptors attenuates satiety actions of exogenous and endogenous CCK. Am. J. Physiol. 1992; 262 (1 Pt 2): R46-50.

[18] Bi S, Moran TH. Actions of CCK in the controls of food intake and body weight: lessons from the CCK-A receptor deficient OLETF rat. Neuropeptides 2002; 36 (0143–4179 (Print)): 171–181.

[19] Moran TH, Bi S. Hyperphagia and obesity in OLETF rats lacking CCK-1 receptors. Philos. Trans. R. Soc. Lond. B. Biol. Sci. 2006; 361 (1471): 1211–8.

[20] Kopin AS, Mathes WF, McBride EW, Nguyen M, Al-Haider W, Schmitz F, et al. The cholecystokinin-A receptor mediates inhibition of food intake yet is not essential for the maintenance of body weight. J. Clin. Invest. 1999; 103 (3): 383–91.

[21] Schwartz GJ, Moran TH. CCK elicits and modulates vagal afferent activity arising from gastric and duodenal sites. Ann.N.Y.Acad.Sci. 1994; 713 (0077–8923 (Print)): 121–128.

[22] Kobelt P, Paulitsch S, Goebel M, Stengel A, Schmidtmann M, van I der V, et al. Peripheral injection of CCK-8S induces Fos expression in the dorsomedial hypothalamic nucleus in rats. Brain Res 2006; 1117 (0006–8993 (Print)): 109–117.

[23] Glatzle J, Kreis ME, Kawano K, Raybould HE, Zittel TT. Postprandial neuronal activation in the nucleus of the solitary tract is partly mediated by CCK-A receptors. Am.J Physiol Regul.Integr.Comp Physiol 2001; 281 (0363–6119 (Print)): R222–R229.

[24] Moran TH, Baldessarini AR, Salorio CF, Lowery T, Schwartz GJ. Vagal afferent and efferent contributions to the inhibition of food intake by cholecystokinin. Am.J.Physiol 1997; 272 (0002–9513): R1245–R1251.

[25] Edwards GL, Ladenheim EE, Ritter RC. Dorsomedial hindbrain participation in cholecystokinin-induced satiety. Am.J.Physiol 1986; 251 (0002–9513 (Print)): R971– R977.

[26] Riedy CA, Chavez M, Figlewicz DP, Woods SC. Central insulin enhances sensitivity to cholecystokinin. Physiol. Behav. 1995; 58 (4): 755–760.

[27] Figlewicz DP, Sipols AJ, Seeley RJ, Chavez M, Woods SC, Porte Jr. D. Intraventricular insulin enhances the meal-suppressive efficacy of intraventricular cholecystokinin octapeptide in the baboon. Behav.Neurosci. 1995; 109: 567–569.

[28] Williams DL, Baskin DG, Schwartz MW. Hindbrain leptin receptor stimulation enhances the anorexic response to cholecystokinin. Am.J Physiol Regul.Integr.Comp Physiol 2009; 297 (1522–1490 (Electronic)): R1238–R1246.

[29] Matson CA, Reid DF, Cannon TA, Ritter RC. Cholecystokinin and leptin act synergistically to reduce body weight. Am.J.Physiol Regul.Integr.Comp Physiol 2000; 278 (0363–6119 (Print)): R882–R890.

[30] Barrachina MD, Martinez V, Wang L, Wei JY, Tache Y. Synergistic interaction between leptin and cholecystokinin to reduce short-term food intake in lean mice. Proc.Natl.Acad.Sci.U.S.A 1997; 94 (0027–8424 (Print)): 10455–10460.

[31] Emond M, Schwartz GJ, Ladenheim EE, Moran TH. Central leptin modulates behavioral

and neural responsivity to CCK. Am.J.Physiol 1999; 276 (0002–9513 (Print)): R1545– R1549.

[32] Schwartz MW, Woods SC, Porte Jr. D, Seeley RJ, Baskin DG. Central nervous system control of food intake. Nat. 2000.Apr 6.;404.(6778.)661.-71. 2000; 404 (0028–0836): 661–671.

[33] Begg DP, Woods SC. The central insulin system and energy balance. Handb.Exp.Pharmacol. 2012; 209 (0171–2004 (Print)): 111–129.

[34] Matson CA, Reid DF, Ritter RC. Daily CCK injection enhances reduction of body weight by chronic intracerebroventricular leptin infusion. Am.J.Physiol Regul.Integr.Comp Physiol 2002; 282 (0363–6119 (Print)): R1368–R1373.

[35] Cano V, Merino B, Ezquerra L, Somoza B, Ruiz-Gayo M. A cholecystokinin-1 receptor agonist (CCK-8) mediates increased permeability of brain barriers to leptin. Br. J. Pharmacol. 2008; 154 (5): 1009–15.

[36] Cano V, Ezquerra L, Ramos MP, Ruiz-Gayo M. Regulation of leptin distribution between plasma and cerebrospinal fluid by cholecystokinin receptors. Br. J. Pharmacol. 2003; 140 (4): 647–52.

[37] Yousif S, Marie-Claire C, Roux F, Scherrmann J-M, Declèves X. Expression of drug transporters at the blood-brain barrier using an optimized isolated rat brain microvessel strategy. Brain Res. 2007; 1134 (1): 1–11.

[38] Schwartz MW, Figlewicz DF, Kahn SE, Baskin DG, Greenwood MRC, Porte D. Insulin binding to brain capillaries is reduced in genetically obese, hyperinsulinemic Zucker rats. Peptides 1990; 11 (3): 467–472.

[39] Plum L, Schubert M, Bruning JC. The role of insulin receptor signaling in the brain. Trends Endocrinol.Metab 2005; 16 (1043–2760 (Print)): 59–65.

[40] Zlokovic B V, Jovanovic S, Miao W, Samara S, Verma S, Farrell CL. Differential regulation of leptin transport by the choroid plexus and blood-brain barrier and high affinity transport systems for entry into hypothalamus and across the blood-cerebrospinal fluid barrier. Endocrinology 2000; 141 (4): 1434–41.

[41] Thomas SA, Preston JE, Wilson MR, Farrell CL, Segal MB. Leptin transport at the blood-- cerebrospinal fluid barrier using the perfused sheep choroid plexus model. Brain Res. 2001; 895 (1–2): 283–90.

[42] Schwartz MW, Bergman RN, Kahn SE, Taborsky GJ, Fisher LD, Sipols AJ, et al. Evidence for entry of plasma insulin into cerebrospinal fluid through an intermediate compartment in dogs. Quantitative aspects and implications for transport. J. Clin. Invest. 1991; 88 (4): 1272–81.

[43] Baura GD, Foster DM, Porte Jr. D, Kahn SE, Bergman RN, Cobelli C, et al. Saturable transport of insulin from plasma into the central nervous system of dogs in vivo. A mechanism for regulated insulin delivery to the brain. J.Clin.Invest 1993; 92 (0021–9738 (Print)): 1824–1830.

[44] Liu M, Shen L, Begg DP, D’alessio DA, Woods SC. Insulin increases central

apolipoprotein E levels as revealed by an improved technique for collection of cerebrospinal fluid from rats. J. Neurosci. Methods 2012; 209 (1): 106–12.

[45] Gao W, Eisenhauer PB, Conn K, Lynch JA, Wells JM, Ullman MD, et al. Insulin degrading enzyme is expressed in the human cerebrovascular endothelium and in cultured human cerebrovascular endothelial cells. Neurosci. Lett. 2004; 371 (1): 6–11.

[46] Lynch JA, George AM, Eisenhauer PB, Conn K, Gao W, Carreras I, et al. Insulin degrading enzyme is localized predominantly at the cell surface of polarized and unpolarized human cerebrovascular endothelial cell cultures. J. Neurosci. Res. 2006; 83 (7): 1262–70.

[47] Rushakoff RJ, Goldfine ID, Carter JD, Liddle RA. Physiological concentrations of cholecystokinin stimulate amino acid-induced insulin release in humans. J. Clin. Endocrinol. Metab. 1987; 65 (3): 395–401.

[48] Ahrén B, Pettersson M, Uvnäs-Moberg K, Gutniak M, Efendic S. Effects of cholecystokinin (CCK)-8, CCK-33, and gastric inhibitory polypeptide (GIP) on basal and meal-stimulated pancreatic hormone secretion in man. Diabetes Res. Clin. Pract. 1991; 13 (3): 153–61.

[49] May AA, Liu M, Woods SC, Begg DP. CCK increases the transport of insulin into the brain. Physiol. Behav. 2016; 165: 392–397.

[50] Abbott NJ. Inflammatory mediators and modulation of blood-brain barrier permeability. Cell. Mol. Neurobiol. 2000; 20 (2): 131–47.

[51] Wang L, Martıneź V, Barrachina MD, Taché Y. Fos expression in the brain induced by peripheral injection of CCK or leptin plus CCK in fasted lean mice. Brain Res. 1998; 791 (1): 157–166.

[52] Pardridge WM. Drug transport across the blood-brain barrier. J. Cereb. Blood Flow Metab. 2012; 32 (11): 1959–72.

Chapter 3:

Estrogen and Insulin Transport Through the Blood-Brain Barrier

3.1 Background

As discussed in the previous chapter, hormones can interact at the BBB to alter the transport of other molecules into the CNS. Likewise, during conditions of the metabolic syndrome, the transport systems of the BBB can become impaired [1–3]. For example, insulin binds less effectively to insulin receptors expressed at the BBB of genetically obese Zucker rats relative to their lean littermates [4]. Obese individuals also have proportionally less insulin and other catabolic signals in the CSF relative to their lean counterparts [3; 5]. Recently, we and others found that that high-fat diet-induced obesity (HFD-DIO) impairs the entry of insulin into the CNS [1; 2], and that this impairment is completely reversible if the animals are maintained on a low-fat diet [1]. Overall, these findings indicate that during the metabolic syndrome, signals that report the amounts of stored, circulating, and consumed energy are under-represented to the brain [1; 6].

Factors that improve the sensitivity to insulin and other catabolic signals would be highly desirable for preventing the energy imbalances that further progress into obesity and its co- morbidities. Estradiol (E2) is well-known to enhance the effects of other catabolic signals, including leptin [7], CCK [8; 9], apolipoprotein A-IV [10; 11], and GLP-1 [12], with a large portion of these interactions occurring within the CNS. E2 can reverse the metabolic syndrome in both sexes [12] and has been demonstrated to significantly increase systemic insulin sensitivity in men, women, and rodents [12–17]. We were therefore surprised to find that E2 reduces the catabolic effect of insulin that is administered within the third ventricle of the brain, despite enhancing the effect of leptin in the same study [7; 18]. It was later confirmed that insulin administered into the CNS of humans via the intranasal route suppresses food intake and adiposity in men, but not in women [19]. These findings were unexpected, because generally, treatments which improve systemic insulin sensitivity are positively associated with central insulin sensitivity and vice versa [1; 20–23].

The apparent reduction of central insulin sensitivity following E2 treatment does not necessarily imply that E2 impairs insulin signaling in the hypothalamus; if insulin signaling was already maximally activated in the presence of E2, then adding additional insulin would have no further effect on hypothalamic insulin signaling and food intake. For example, obese rats are considered less sensitive to leptin than their lean counterparts because they do not respond to the anorectic effects of exogenous leptin [24; 25]. However, it was recently found that obese mice still have fully-functional endogenous leptin signaling [26]. Therefore, the lack of response to insulin administered into the 3rd ventricle may not be due to impaired insulin sensitivity, but rather, may be due to increased endogenous insulin signaling. While E2 signals through several catabolic signaling pathways that are important for insulin signaling (such as phosphoinositide 3-kinase [PI3K] and extracellular signal-regulated kinase [Erk1/2], etc.), it is unlikely that E2 overrides the effects of insulin by signaling through these pathways. On the contrary, E2 enhances the effects leptin and other molecules [7; 18; 27] which also act through

PI3K and Erk1/2 signaling pathways to exert their catabolic effects [28–30]. Therefore, it is most likely that a different mechanism underlies this effect.

Insulin acts dose-dependently within the CNS, such that at higher concentrations there is a greater catabolic effect [31]. E2 may increase the transport of insulin into the brain at levels that saturate hypothalamic insulin receptors, leading to a decreased response to exogenous insulin that is administered into the brain of females or males treated with E2 [7]. We first tested the hypothesis that E2 exerts its catabolic effects, in part, by increasing the transport of insulin into the CNS (Fig 3.1). This became the basis for Paper II (Chapter 3b).

3.2 Estrogen and Energy Balance

Estrogens were discovered by

Edward Doisy in 1929, who found that

extracts from sow ovaries led to the

production of cornified cells in the vagina

of ovariectomized mice [32]. Between

1929-1936 Doisy proceeded to discover

and characterize estrone (E1), estradiol

(E2), and estriol (E3), which are the three

Figure 3.1. Simplified Aims of Chapter 3. primary isoforms of estrogen in humans. Of The red box outlines the mediobasal hypothalamus, which has high pertinence to these estrogens, E2 has the highest affinity the effects of insulin on energy balance for estrogen receptors and is also the most abundant estrogen in circulation in premenopausal women. Specifically, plasma concentrations of E2 are at least 2-4 times that of E1 and 80-100 times that of E3 [33; 34]. E2 has a binding affinity for estrogen receptors that is approximately double that of E1 and has approximately 10 times the receptor affinity of E3 [35; 36]. After menopause, the levels of E2 drop to the levels found in males and E1 becomes the most abundant estrogen in circulation. E3 is only present at a substantial level during pregnancy, because it is synthesized mostly by the placenta [33].

Thus, in premenopausal women, E2 is considered the most active form of estrogen in circulation [33; 35].

E2 is most abundantly produced in the theca interna of ovarian follicles during the follicular phase (termed “proestrus” in rodents). Interestingly, the period of highest E2 secretion is followed by significant reductions of food intake and body weight [27]. When female rodents

are ovariectomized (OVX), the levels of circulating E2 decrease to the levels found in males

[37], and weight gain occurs at a rate comparable to that of males [10]. By supplementing these rodents with physiologic, “cyclic” doses of E2, it was found that E2 is directly responsible for the variations in food intake and body weight during the ovarian cycle [38]. In contrast, progesterone does not affect energy balance [27], and its peak during the ovarian cycle coincides with the time at which food intake is highest. While the mechanisms for how E2 induces catabolism have been thoroughly investigated, it is still not fully understood why E2 impacts energy balance in such a profound way.

E2 can also be synthesized by many other tissues, such as adipose, muscle, and many regions of the brain via the conversion of testosterone to E2 by aromatase [33; 39–41]. E2 can therefore act locally as a paracrine or autocrine factor to influence energy balance and reproductive behavior [35]. It is now well-appreciated that E2 also plays important roles in metabolic and sexual function of men [42]. More than 80% of circulating E2 in men is converted from testosterone by the aromatase enzyme and healthy men (age 20-50) treated with aromatase inhibitors become E2-deficient and develop greater adiposity than controls [42].

These findings are consistent with aromatase-KO mice, in which KO’s of both sexes have greater body weight, adiposity, and hepatic fat accumulation than controls [43]. Taken together, it is clear that estrogens strongly influence metabolism and energy balance in both sexes.

Estrogen Receptors: Metabolic functions

E2 acts through estrogen receptors to mediate the majority of its effects [44]. There are three primary estrogen receptors that share some functional overlaps, but have many distinct physiological effects, as revealed by knock-out studies [45]. The first estrogen receptor to be identified was ERα [46] and knockout studies led to the subsequent discovery of ERβ in 1996

[44; 47]. Later, a G-protein-coupled estrogen receptor (GPER), was discovered, which acts

exclusively through “non-classic” signaling pathways [48]. There has been much debate as to which ER(s) is most involved in maintaining energy balance. Initial studies examining the genetic knockout (KO) of ERα throughout the whole body of mice noted a pronounced obesity phenotype [49]. In comparison, ERβ-KO mice have normal body weight [27]. Similarly, ERα- selective agonists reduce food intake and body weight, while ERβ-selective agonists do not exert such effects [50]. Several GPER-KO mouse lines have been developed, with earliest reports finding no effect on energy balance, although it was later found that a different line did indeed develop a reliable obesity phenotype [48]. However, the discrepancies between KO- models of GPER are still not understood [48]. Considering the clear-cut role of ERα in energy balance and metabolism and that E2 has the strongest affinity for ERα out of all estrogen receptors [33], we focused our efforts on further understanding the role of ERα the prevention and progression of metabolic syndrome.

3.3 Estrogen and Insulin Transport into the CNS

In order to determine whether E2 could act directly on the BBB to influence insulin transport, we first isolated rat brain microvessels (which comprise the BBB), as described in

Chapter 2 [51]. We conducted immunohistochemistry for ERα and insulin receptor after verifying the relative purity of our microvessel isolates (Paper II, Fig 1A). These antibodies were also validated by Western Blotting (Paper II, Fig 1C,D,F). Substantial co-localization of ERα with insulin receptors was observed in brain microvessels (Paper II, Fig 1E), indicating that E2 has the potential to influence insulin transport by acting directly through ERα at the BBB.

Next, we examined whether acute E2 treatment increases the transport of insulin into the CNS, using ovariectomized (OVX) female rats. For this study, female Long-Evans rats (14 weeks of age, Envigo, Indianapolis, IN) underwent OVX surgery, as described [10]. Briefly, rats were anesthetized with an intraperitoneal injection of ketamine/xylazine (60 mg/kg and 10

mg/kg, respectively) and were administered buprenorphine (0.05 mg/kg) subcutaneously (sc) prior to surgery. After shaving and disinfecting the surgical site, a small incision was made along the midline of the abdomen. The ovaries were individually removed and excised after tying off the fallopian tubes with suture (7.0, Silk, Ethicon, New Jersey). The incision site was then sutured (4.0, Vicryl Rapide, Ethicon, New Jersey) and after administering post-operative meloxicam (4 mg/kg, sc), rats were allowed to recover for two weeks prior to the study. OVX rats were then administered a single dose of E2 one day prior to assessing insulin transport into the CNS. Although E2 treatment reduced body weight relative to the vehicle-treated group, providing a positive control (Paper II, Fig 2E), E2 did not increase insulin transport into the CNS

(Paper II, Fig 2B,C). The expression of insulin receptor mRNA at the BBB was also unaffected by E2 treatment (Paper II, Fig 2D).

We next examined the effect of acute E2 treatment on insulin transport of male rats, because males are relatively more sensitive to central insulin administration than OVX females.

As long as 4 weeks following surgery, OVX females are still relatively insensitive to intracerebroventricular insulin [7]. A clinical study similarly found that postmenopausal women remain insensitive to intranasal insulin when administered the same dose that is effective at reducing food intake and adiposity in males [52]. Further, E2 completely blunts the anorectic effects of insulin in male rats [7]. For this reason, the size of E2’s effect on increasing the transport of insulin into the CNS was therefore anticipated to be much greater in males than

OVX rats. We first verified that that acute E2 treatment effectively reduces food intake in male rats in a pilot study (Paper II, Fig 3D). Nonetheless, E2 treatment in male rats did not increase the transport of insulin into the CNS (Paper II, Fig 3A-C).

Although E2 can acutely reduce food intake and body weight, a longer treatment period may be required for E2 to fully exert its effects on insulin transport into the CNS. As mentioned, we previously found that insulin transport through the BBB is impaired during diet-induced

obesity, but that insulin transport was restored in association with increased systemic insulin sensitivity and reduced body weight following maintenance on a low-fat diet [1]. E2 treatment promotes similar benefits, by increasing insulin sensitivity and reducing body weight [12]. We therefore hypothesized that the ability of E2 to reverse metabolic syndrome in DIO rodents [12] is partly due to increasing the transport of insulin into the CNS.

In this study, male Long-Evans (250 g) were fed a 40% HFD (Research Diets,

D03082706) for 10 weeks prior to the study. They were then administered chronic, cyclic injections of E2 or vehicle for one month, while being maintained on the HFD. By the end of the treatment period, the E2-treated group weighed 10% less than controls after being maintained on HFD (Paper II, Fig 4A). Despite the reversal of 10% body weight loss, insulin transport into the CNS was not improved in E2-treated rats, nor in weight-matched controls (Paper II, Fig 4).

The concentration of insulin in the blood is directly proportional to the amount of insulin in the CSF, up to the point of transporter saturation [53]. When animals are injected with the same dose of insulin, as in this study, individual variability can lead to differences in the levels of insulin in the blood and CSF among rats of the same group [54]. In theory, if all rats in a group have similar insulin transport rates, a rat with the highest levels of insulin in the blood would be expected to have the highest CSF insulin levels relative to other rats in the group. If transport does not occur at an appreciable rate, then no matter how much insulin is present in the blood, there will be no change in the concentration of insulin in the CSF [54; 55]. For this reason, we conducted Pearson analysis of blood vs. CSF insulin, as previously described [1]. No significant correlations were detected in any individual group, nor when all data were combined into the same plot, which is consistent with impaired insulin transport in these chronic HFD-DIO rats

(Paper II, Fig 4H). E2 therefore does not mediate the reversal of HFD-DIO by increasing insulin transport into the CNS.

We conducted a final experiment to address the possibility that although E2 is unable to restore insulin transport into the CNS during the reversal of metabolic syndrome, it may be capable of protecting insulin transport into the CNS if provided at the onset of HFD feeding. In this study, male rats that were never previously exposed to HFD were treated with cyclic injections of E2 at the onset of HFD feeding. After one month, E2-treated rats had gained ~13% less body weight than controls (Paper II, Fig 5A). Peripheral insulin sensitivity was significantly improved (Paper II, Fig 5D-E), indicating that the metabolic syndrome was prevented with E2 treatment. However, E2-treatment still did not increase insulin’s appearance in the CSF (Paper

II, Fig 5G-I). Surprisingly, Pearson analysis revealed a significant correlation between insulin levels in the plasma and CSF in pair-fed rats, but this was not observed with E2 treatment

(Paper II, Fig 5I). This led to the conclusion that E2-treatment may actually impair insulin transport into the CNS under these conditions.

Considering that insulin appearing in the CSF must first pass through the BBB and brain interstitial fluid, it is possible that differences in insulin binding to receptors throughout the brain in E2-treated rats could reduce the appearance of insulin in the CSF. However, it was previously found that the brains of E2-treated rats have no difference in the binding of insulin to brain tissue, relative to controls [56], limiting the possibility that we failed to detect an increase of insulin transport. Alternatively, insulin may be degraded more rapidly in the brain of E2-treated rats relative to controls, which could similarly mask an increase of insulin transport into the CNS

[57]. To address this possibility, we measured the levels of insulin degrading enzyme in the hypothalamus and cortex via Western blotting (Paper II, Fig 5J,K) and detected no significant differences among the groups, making it unlikely that insulin degradation is a major influence on insulin levels in the CNS. From these results, we concluded that the reversal and prevention of

HFD-DIO by E2 treatment in rats occurs entirely independent of increasing insulin transport into the CNS [51].

Visceral adiposity has long been associated with impaired insulin transport into the CNS

[5; 58]. In the current study, we found that E2 did not alter the adiposity relative to the control group, but that it did lead to a significantly higher accumulation of visceral fat than weight- matched controls (Paper II, Fig 5B). These observations may further explain why insulin transport appeared to be effective in pair-fed, but not E2-treated rats.

After eliminating the possibility that E2 increases endogenous hypothalamic insulin signaling by increasing the levels of insulin in the CNS, we next tested the hypothesis that E2 increases the activation of insulin signaling pathways within the hypothalamus. It is known that

E2 can freely cross the BBB [59] and exerts catabolic effects within the hypothalamus via ERα.

OVX or male rats injected into the third ventricle with small doses of E2 or vehicle, so as to preferentially target the hypothalamus, maintain a comparable body weight difference relative to groups that were injected subcutaneously with E2 or vehicle [7]. ERα is highly expressed by neurons and glia in the hypothalamus, with strongest expression in the VMH and Arc [60].

Genetic silencing of ERα expression in VMH adult mice and rats results in a rapid weight gain, adiposity and insulin resistance [61]. Consistently, male and female mice with a CNS-specific deletion of ERα develop obesity and insulin resistance [62]. The metabolic outcomes of these studies are similar to analogous studies in which the insulin receptor was silenced in the hypothalamus or deleted from the CNS [63; 64]. Further, ERα co-localizes with insulin receptors in the hypothalamus (Fig 3.1), indicating the potential for a direct interaction between these signaling pathways.

Figure 3.2. Co-localization of ERα and insulin receptors in the mediobasal hypothalamus. A) Dual-label immunohistochemistry for estrogen receptor α (ERα, green, Alexa-488) and B) insulin receptor-β (IR, red, Cy3) in neurons of the arcuate nucleus. Single-labeled neurons mostly expressing ERα are marked with green arrows. Single-labeled neurons mostly expressing IR are marked with red arrows. C) An overlay of the images, with yellow arrows pointing to dually-labeled neurons. Scale bar, 10 μm.

We therefore examined insulin signaling in the hypothalami of male rats that were treated with cyclic E2 for one month in order to prevent HFD-DIO. Specifically, the MBH was selected, because of the importance of this region for the effects of insulin on energy balance

[21; 63; 65; 66]. The MBH was micropunched from frozen brains using a cryostat (Leica).

Samples were subsequently prepared for Western blotting as described in Paper II, Methods,

Section 2.10, and protein expression was assessed by densitometric analysis using Image J software (NIH). Antibodies used in these studies are listed in Table 1 of Appendix I. Despite substantially increasing systemic insulin sensitivity and reducing body weight (Paper II, Fig 5A) as previously described, E2 did not increase the activation of proximal or distal insulin signaling pathways (Fig 3.3). Taken together, we conclude that E2 ameliorates the symptoms of the metabolic syndrome independent of affecting insulin transport or altering the activation of proximal or distal insulin signaling in the hypothalamus under these conditions.

Figure 3.3. Chronic E2 treatment does not increase proximal or distal insulin signaling pathways in the MBH. Rats were treated chronically with E2 or vehicle to during maintenance on a HFD, in the manner described earlier. MBH were dissected and protein expression was analyzed via Western blot using antibodies for phosphorylated insulin receptor (pIR-β) (A), IRS-1/2 (pIRS-1/2) (B), and CREB (pCREB) (C). Blots were quantified via densitometry (Image J) and analyzed using one-way ANOVA with a Holm-Sidak test for post-hoc multiple comparisons. There were no statistically significant differences among controls, E2-treated, and pair-fed rats for phosphorylated or total protein levels; ratio of phosphorylated/total protein levels also did not differ among groups for these proteins. Actin was used as a loading control. Mean ± SEM, P<0.05, n=5-7.

The above figure appears in: Woods and May et al. (2016), Using the Cerebrospinal Fluid to Understand Ingestive Behavior. Physiology and Behavior.

When re-examining the mechanism by which E2 inhibits the anorectic effect of exogenous insulin, it appears that that E2 exerts this effect without increasing endogenous insulin signaling. Instead, E2 may override critical insulin signaling pathways, such as PI3K/Akt or Erk (however unlikely) or it may alter the responsiveness of neurons or glia to insulin in such a way that reduces the anorectic effects of central insulin.

Although its specific effects can vary, hypothalamic inflammation alters the activation of neurons following insulin treatment [65; 67] and impairs systemic insulin sensitivity [67].

Because E2 is known to attenuate inflammation in the brain and numerous cell types, we hypothesized that E2 may indirectly influence insulin’s activity in the hypothalamus by inhibiting inflammation. Western blotting revealed that E2 significantly reduced the activation of the signaling molecule, IKKβ, which is a major effector in the inflammatory NF-κB inflammatory pathway (Chapter 4, Fig 1). This finding provided the basis for the studies described in Chapter

REFERENCES

[1] Begg DP, Mul JD, Liu M, Reedy BM, D’Alessio DA, Seeley RJ, et al. Reversal of diet- induced obesity increases insulin transport into cerebrospinal fluid and restores sensitivity to the anorexic action of central insulin in male rats. Endocrinology 2013; 154 (3): 1047– 54.

[2] Urayama A, Banks WA. Starvation and triglycerides reverse the obesity-induced impairment of insulin transport at the blood-brain barrier. Endocrinology 2008; 149 (7): 3592–3597.

[3] Kern W, Benedict C, Schultes B, Plohr F, Moser A, Born J, et al. Low cerebrospinal fluid insulin levels in obese humans. Diabetologia 2006; 49 (11): 2790–2.

[4] Schwartz MW, Figlewicz DF, Kahn SE, Baskin DG, Greenwood MRC, Porte D. Insulin binding to brain capillaries is reduced in genetically obese, hyperinsulinemic Zucker rats. Peptides 1990; 11 (3): 467–472.

[5] Owen OE, Reichard GA, Boden G, Shuman C. Comparative measurements of glucose, beta-hydroxybutyrate, acetoacetate, and insulin in blood and cerebrospinal fluid during starvation. Metabolism 1974; 23 (1): 7–14.

[6] Morton GJ, Meek TH, Schwartz MW. Neurobiology of food intake in health and disease. Nat. Rev. Neurosci. 2014; 15 (6): 367–78.

[7] Clegg DJ, Brown LM, Woods SC, Benoit SC. Gonadal hormones determine sensitivity to central leptin and insulin. Diabetes 2006; 55 (4): 978–987.

[8] Geary N, Trace D, McEwen B, Smith GP. Cyclic estradiol replacement increases the satiety effect of CCK-8 in ovariectomized rats. Physiol Behav. 1994; 56 (0031–9384 (Print)): 281–289.

[9] Butera PC, Bradway DM, Cataldo NJ. Modulation of the satiety effect of cholecystokinin by estradiol. Physiol Behav. 1993; 53 (0031–9384 (Print)): 1235–1238.

[10] Shen L, Wang DQH, Lo CM, Tso P, Davidson WS, Woods SC, et al. Estradiol increases the anorectic effect of central apolipoprotein A-IV. Endocrinology 2010; 151 (7): 3163– 3168.

[11] Shen L, Liu Y, Wang DQH, Tso P, Woods SC, Liu M. Estradiol stimulates apolipoprotein A-IV gene expression in the nucleus of the solitary tract through estrogen receptor-α. Endocrinology 2014; 155 (10): 3882–90.

[12] Finan B, Yang B, Ottaway N, Stemmer K, Müller TD, Yi C-X, et al. Targeted estrogen delivery reverses the metabolic syndrome. Nat. Med. 2012; 18 (12): 1847–56.

[13] Sandoval DA, Ertl AC, Richardson MA, Tate DB, Davis SN. Estrogen Blunts Neuroendocrine and Metabolic Responses to Hypoglycemia. Diabetes 2003; 52 (7): 1749–1755.

[14] Barrett-Connor E, Laakso M. Ischemic heart disease risk in postmenopausal women. Effects of estrogen use on glucose and insulin levels. Arteriosclerosis 1990; 10 (0276– 5047 (Print)): 531–534.

[15] Riant E, Waget A, Cogo H, Arnal J-F, Burcelin R, Gourdy P. Estrogens protect against high-fat diet-induced insulin resistance and glucose intolerance in mice. Endocrinology 2009; 150 (5): 2109–17.

[16] Shen M, Kumar SPDS, Shi H. Estradiol regulates insulin signaling and inflammation in adipose tissue. Horm. Mol. Biol. Clin. Investig. 2014; 17 (2): 99–107.

[17] Alonso A, Fernandez R, Moreno M, Ordonez P, Gonzalez-Pardo H, Conejo NM, et al. Positive Effects of 17β-Estradiol on Insulin Sensitivity in Aged Ovariectomized Female Rats. Journals Gerontol. Ser. A Biol. Sci. Med. Sci. 2006; 61 (5): 419–426.

[18] Clegg DJ, Riedy CA, Smith KAB, Benoit SC, Woods SC. Differential sensitivity to central leptin and insulin in male and female rats. Diabetes 2003; 52 (3): 682–687.

[19] Hallschmid M, Benedict C, Schultes B, Fehm HL, Born J, Kern W. Intranasal insulin reduces body fat in men but not in women. Diabetes 2004; 53 (11): 3024–3029.

[20] Obici S, Feng Z, Morgan K, Stein D, Karkanias G, Rossetti L. Central administration of oleic acid inhibits glucose production and food intake. Diabetes 2002; 51 (0012–1797 (Print)): 271–275.

[21] Obici S, Zhang BB, Karkanias G, Rossetti L. Hypothalamic insulin signaling is required for inhibition of glucose production. Nat. Med. 2002; 8 (12): 1376–1382.

[22] Carvalheira JBC, Ribeiro EB, Araújo EP, Guimarães RB, Telles MM, Torsoni M, et al. Selective impairment of insulin signalling in the hypothalamus of obese Zucker rats. Diabetologia 2003; 46 (12): 1629–40.

[23] Porte Jr. D, Seeley RJ, Woods SC, Baskin DG, Figlewicz DP, Schwartz MW. Obesity, diabetes and the central nervous system. Diabetologia 1998; 41: 863–881.

[24] Munzberg H, Myers Jr. MG. Molecular and anatomical determinants of central leptin resistance. Nat.Neurosci. 2005; 8 (1097–6256 (Print)): 566–570.

[25] Enriori PJ, Evans AE, Sinnayah P, Jobst EE, Tonelli-Lemos L, Billes SK, et al. Diet- induced obesity causes severe but reversible leptin resistance in arcuate melanocortin neurons. Cell Metab 2007; 5 (1550–4131 (Print)): 181–194.

[26] Ottaway N, Mahbod P, Rivero B, Norman LA, Gertler A, D’Alessio DA, et al. Diet-induced obese mice retain endogenous leptin action. Cell Metab. 2015; 21 (6): 877–82.

[27] Asarian L, Geary N. Sex differences in the physiology of eating. Am. J. Physiol. Regul. Integr. Comp. Physiol. 2013; 305 (11): R1215-67.

[28] Rahmouni K, Sigmund CD, Haynes WG, Mark AL. Hypothalamic ERK mediates the anorectic and thermogenic sympathetic effects of leptin. Diabetes 2009; 58 (3): 536–42.

[29] Xu AW, Kaelin CB, Takeda K, Akira S, Schwartz MW, Barsh GS. PI3K integrates the action of insulin and leptin on hypothalamic neurons. J.Clin.Invest 2005; 115 (0021–9738 (Print)): 951–958.

[30] Donato Jr. J, Frazao R, Elias CF. The PI3K signaling pathway mediates the biological effects of leptin. Arq Bras.Endocrinol.Metabol. 2010; 54 (1677–9487 (Electronic)): 591– 602.

[31] Chavez M, Kaiyala K, Madden LJ, Schwartz MW, Woods SC. Intraventricular insulin and the level of maintained body weight in rats. Behav.Neurosci 1995; 109 (0735–7044 (Print)): 528–531.

[32] Simoni RD, Hill RL, Vaughan M. The Discovery of Estrone, Estriol, and Estradiol and the Biochemical Study of Reproduction. The Work of Edward Adelbert Doisy. J. Biol. Chem. 2002; 277 (28): e17–e17.

[33] Gruber CJ, Tschuggel W, Schneeberger C, Huber JC. Production and actions of estrogens. N. Engl. J. Med. 2002; 346 (5): 340–352.

[34] Baird DT, Fraser IS. Concentration of oestrone and oestradiol in follicular fluid and ovarian venous blood of women. Clin. Endocrinol. (Oxf). 1975; 4 (3): 259–66.

[35] Nelson LR, Bulun SE. Estrogen production and action. J Am.Acad.Dermatol. 2001; 45 (0190–9622 (Print)): S116–S124.

[36] Gruber CJ, Tschugguel W, Schneeberger C, Huber JC. Production and actions of estrogens. N. Engl. J. Med. 2002; 346 (5): 340–52.

[37] Clegg DJ, Brown LM, Zigman JM, Kemp CJ, Strader AD, Benoit SC, et al. Estradiol- dependent decrease in the orexigenic potency of ghrelin in female rats. Diabetes 2007; 56 (0012–1797 (Print)): 1051–1058.

[38] Wade GN. Gonadal hormones and behavioral regulation of body weight. Physiol Behav. 1972; 8 (0031–9384 (Print)): 523–534.

[39] Prange-Kiel J, Rune GM. Direct and indirect effects of estrogen on rat hippocampus. Neuroscience 2006; 138 (3): 765–72.

[40] Remage-Healey L, Dong S, Maidment NT, Schlinger BA. Presynaptic control of rapid estrogen fluctuations in the songbird auditory forebrain. J. Neurosci. 2011; 31 (27): 10034–8.

[41] Kretz O, Fester L, Wehrenberg U, Zhou L, Brauckmann S, Zhao S, et al. Hippocampal synapses depend on hippocampal estrogen synthesis. J. Neurosci. 2004; 24 (26): 5913– 21.

[42] Finkelstein JS, Lee H, Burnett-Bowie S-AM, Pallais JC, Yu EW, Borges LF, et al. Gonadal Steroids and Body Composition, Strength, and Sexual Function in Men — NEJM. N. Engl. J. Med. 2013; 369 (11): 1011–1022.

[43] Jones ME, Thorburn AW, Britt KL, Hewitt KN, Wreford NG, Proietto J, et al. Aromatase- deficient (ArKO) mice have a phenotype of increased adiposity. Proc. Natl. Acad. Sci. U. S. A. 2000; 97 (23): 12735–40.

[44] Cooke PS, Buchanan DL, Lubahn DB, Cunha GR. Mechanism of estrogen action: lessons from the estrogen receptor-alpha knockout mouse. Biol. Reprod. 1998; 59 (3): 470–5.

[45] Cheskis BJ, Greger JG, Nagpal S, Freedman LP. Signaling by estrogens. J.Cell Physiol 2007; 213 (1097–4652 (Electronic)): 610–617.

[46] Lubahn DB, Moyer JS, Golding TS, Couse JF, Korach KS, Smithies O. Alteration of

reproductive function but not prenatal sexual development after insertional disruption of the mouse estrogen receptor gene. Proc. Natl. Acad. Sci. 1993; 90 (23): 11162–11166.

[47] Kuiper GG, Enmark E, Pelto-Huikko M, Nilsson S, Gustafsson JA. Cloning of a novel receptor expressed in rat prostate and ovary. Proc.Natl.Acad.Sci.U.S.A 1996; 93 (0027– 8424 (Print)): 5925–5930.

[48] Shi H, Kumar SPDS, Liu X. G protein-coupled estrogen receptor in energy homeostasis and obesity pathogenesis. Prog. Mol. Biol. Transl. Sci. 2013; 114: 193–250.

[49] Heine PA, Taylor JA, Iwamoto GA, Lubahn DB, Cooke PS. Increased adipose tissue in male and female estrogen receptor-alpha knockout mice. Proc. Natl. Acad. Sci. U. S. A. 2000; 97 (23): 12729–34.

[50] Roesch DM. Effects of selective estrogen receptor agonists on food intake and body weight gain in rats. Physiol Behav. 2006; 87 (0031–9384 (Print)): 39–44.

[51] May AA, Bedel ND, Shen L, Woods SC, Liu M. Estrogen and insulin transport through the blood-brain barrier. Physiol. Behav. 2016;

[52] Krug R, Benedict C, Born J, Hallschmid M. Comparable Sensitivity of Postmenopausal and Young Women to the Effects of Intranasal Insulin on Food Intake and Working Memory. J. Clin. Endocrinol. Metab. 2010; 95 (12): E468–E472.

[53] Baura GD, Foster DM, Porte Jr. D, Kahn SE, Bergman RN, Cobelli C, et al. Saturable transport of insulin from plasma into the central nervous system of dogs in vivo. A mechanism for regulated insulin delivery to the brain. J.Clin.Invest 1993; 92 (0021–9738 (Print)): 1824–1830.

[54] Begg DP, May AA, Mul JD, Liu M, D’Alessio DA, Seeley RJ, et al. Insulin Detemir Is Transported From Blood to Cerebrospinal Fluid and Has Prolonged Central Anorectic Action Relative to NPH Insulin. Diabetes 2015; 64 (7): 2457–66.

[55] Begg DP, Woods SC. The central insulin system and energy balance. Handb.Exp.Pharmacol. 2012; 209 (0171–2004 (Print)): 111–129.

[56] Baskin DG, Gierke EP, Wilcox BJ, Matsumoto AM, Schwartz MW. Food intake and estradiol effects on insulin binding in brain and liver. Physiol Behav. 1993; 53 (0031–9384 (Print)): 757–762.

[57] Zhao L, Yao J, Mao Z, Chen S, Wang Y, Brinton RD. 17β-Estradiol regulates insulin- degrading enzyme expression via an ERβ/PI3-K pathway in hippocampus: relevance to Alzheimer’s prevention. Neurobiol. Aging 2011; 32 (11): 1949–63.

[58] Woods SC, Gotoh K, Clegg DJ. Gender differences in the control of energy homeostasis. Exp.Biol.Med.(Maywood.) 2003; 228 (1535–3702 (Print)): 1175–1180.

[59] Banks WA. Brain meets body: the blood-brain barrier as an endocrine interface. Endocrinology 2012; 153 (9): 4111–9.

[60] Lee H, Kim D-W, Remedios R, Anthony TE, Chang A, Madisen L, et al. Scalable control of mounting and attack by Esr1+ neurons in the ventromedial hypothalamus. Nature 2014; 509 (7502): 627–32.

[61] Musatov S, Chen W, Pfaff DW, Mobbs C V, Yang X-JJ, Clegg DJ, et al. Silencing of estrogen receptor alpha in the ventromedial nucleus of hypothalamus leads to metabolic syndrome. Proc.Natl.Acad.Sci.U.S.A 2007; 104 (0027–8424 (Print)): 2501–2506.

[62] Xu Y, Nedungadi TP, Zhu L, Sobhani N, Irani BG, Davis KE, et al. Distinct hypothalamic neurons mediate estrogenic effects on energy homeostasis and reproduction. Cell Metab. 2011; 14 (4): 453–65.

[63] Grillo CA, Tamashiro KL, Piroli GG, Melhorn S, Gass JT, Newsom RJ, et al. Lentivirus- mediated downregulation of hypothalamic insulin receptor expression. Physiol Behav. 2007; 92 (0031–9384 (Print)): 691–701.

[64] Bruning JC, Gautam D, Burks DJ, Gillette J, Schubert M, Orban PC, et al. Role of brain insulin receptor in control of body weight and reproduction. Science (80-. ). 2000; 289 (0036–8075): 2122–2125.

[65] Zhang X, Zhang G, Zhang H, Karin M, Bai H, Cai D. Hypothalamic IKKbeta/NF-kappaB and ER stress link overnutrition to energy imbalance and obesity. Cell 2008; 135 (1): 61– 73.

[66] Obici S, Feng Z, Karkanias G, Baskin DG, Rossetti L. Decreasing hypothalamic insulin receptors causes hyperphagia and insulin resistance in rats. Nat. Neurosci. 2002; 5 (6): 566–572.

[67] Tsaousidou E, Paeger L, Belgardt BF, Pal M, Wunderlich CM, Brönneke H, et al. Distinct Roles for JNK and IKK Activation in Agouti-Related Peptide Neurons in the Development of Obesity and Insulin Resistance. Cell Rep. 2014; 9 (4): 1495–506.

Chapter 4:

Estradiol Prevents Diet-Induced Hypothalamic Inflammation

4.1 Sex Differences in the Development of Cardiovascular Disease

Cardiovascular disease remains the leading cause of morbidity and mortality across the globe and is responsible for more than 17.3 million deaths each year [1]. Cardiovascular disease is strongly associated with obesity [2], with incidence in premenopausal women that is significantly lower than in age-matched men [2–4]. It is important to better understand the source of this sex difference in order to identify new treatment strategies that could be efficacious in both sexes. Recently, it was found that female, but not male mice, are protected against hypothalamic inflammation and impaired cardiovascular function resulting from 16 weeks of HFD feeding [5]. It was also observed that the expression of ERα was reduced in the hypothalamus of males, but not females, leading to the hypothesis that the protection of hypothalamic ERα during chronic HFD feeding plays an important role in the protection of cardiovascular function.

Although E2 prevents against the loss of ERα in cultured hypothalamic neurons treated with palmitic acid (which closely mimics the hypothalamic effects of HFD feeding), E2 treatment cannot not be considered for most men, due to the undesirable effects of E2 on secondary sex characteristics [6]. Identifying the mechanism(s) by which E2 protects hypothalamic ERα against HFD feeding could enable the development of alternative therapeutics to protect ERα signaling without any adverse side-effects.

4.2 Role of Hypothalamic Inflammation in Obesity and Cardiovascular Disease

It is now well-appreciated that the inflammatory transcription factor, NF-κB (nuclear factor- κB), and its upstream signaling molecules, directly promote weight gain during HFD feeding (Fig 4.1). When one such signaling molecule, IKKβ (inhibitor of the κB kinase), is specifically knocked out in the brain or MBH of mice, these mice are protected from HFD- induced obesity relative to controls [7]. Consistent with this, viral overexpression of constitutively-active IKKβ in the mediobasal hypothalamus is sufficient to impair hypothalamic

insulin sensitivity [7; 8], induce hypothalamic ER stress [7], and increase sympathetic tone to the heart [9]; these symptoms are consistent with the obese condition [7; 9]. Other studies have confirmed these findings by manipulating other components of the NF-κB pathway; overexpression of the inhibitor of κB (IκB, which inhibits NF-κB) in the hypothalamus prevents against the cardiovascular effects of HFD-DIO, including hypertension and increased sympathetic tone [9].

Figure 4.1. Diagram of the IKKβ-NF-κB inflammatory signaling pathway. The IKKβ complex can become phosphorylated and activated by over-nutrition, inflammatory cytokines, and other cellular stressors. IKKβ activation promotes the phosphorylation of IκB [7; 29]. Under unstimulated conditions, IκB holds the NF-κB complex (herein represented as p50/p65) in the nucleus by blocking its nuclear localization signals, thereby inhibiting the transcriptional activity of NF-κB. When IκB is phosphorylated, this promotes its ubiquitination and subsequent degradation [29]. As the levels of IκB are reduced in this manner, this results in increased nuclear localization and transcriptional activity of NF-κB.

4.3 E2 Inhibits Hypothalamic Inflammation

The physiological effects of hypothalamic inflammation are in opposition to E2-ERα signaling, and when E2 is administered, it is able to prevent or completely override the effects of inflammation [5; 10]. In vitro studies have determined that E2-ERα signaling can directly inhibit

NF-κB transcription in various cell lines [11–13] and can attenuate inflammation induced by saturated fatty acid treatment in primary cultured hypothalamic neurons and astrocytes [5]. ERα was also found to play a strong anti-inflammatory role in microglia in vivo [10], although this effect may be indirect [5].

Although the anti-inflammatory role of E2 is well-established, the mechanism by which

E2 reduces hypothalamic inflammation is not understood [5]. In order to determine the mechanism involved in this process, we assessed the effect of chronic E2 treatment on the

IKKα/β signaling pathway during maintenance on a HFD. This led to the observation that E2 prevents the activation of the inflammatory IKKβ/NF-κB inflammatory signaling pathway in the hypothalamus of male rats maintained on HFD for one month. Specifically, E2-treated rats had a lower proportion of active IKKα/β relative to control rats or pair-fed rats that were weight- matched to the E2 group (Fig 4.2A), and this was not due to a change in total protein levels of

IKKα/β (Fig 4.2A). The results from Western blotting were consistent between two replicate experiments, demonstrating the repeatability of this finding. Intriguingly, pair-fed rats had significantly higher levels of pIKKα/β than E2-treated rats, suggesting that these effects were

E2-specific and not solely due to a reduction in body weight. This finding established that E2 reduces diet-induced hypothalamic inflammation by inhibiting signaling of the NF-κB cascade.

As previously mentioned and depicted in Fig 4.1, IKKβ is upstream NF-κB and therefore regulates its activity as a transcription factor. We therefore examined the expression of the total/active levels of NF-κB via Western blot, but no differences were observed in the levels of phosphorylated or total NF-κB among E2-treated, controls, and pair-fed rats (Fig 4.2B).

However, we cannot come to a definitive conclusion about the function of the NF-κB signaling pathway based on these results alone, because our approach may not be sensitive enough to detect differences in NF-κB activity [7] and further, changes in the phosphorylation status of the

RelA subunit are not always coupled to the transcriptional output of the NF-κB pathway [14]. A more sensitive NF-κB transcriptional assay would need to be used in future studies to definitively test the hypothesis that E2 inhibits the hypothalamic NF-κB pathway during maintenance on a HFD [7].

Alternatively, there many reported instances in which IKKβ is strongly activated, without any resultant differences in NF-κB transcriptional activity [14; 15]. Depending on its mode of activation, IKKβ can interact with many other pathways and is not always associated with NF-κB activity during maintenance on a HFD in all cell types [14; 16; 17]. IKKβ signaling is very flexible and therefore a broad range of players could be involved, including interactions with signaling pathways involved in ER stress and oxidative stress [7; 14]. Regardless of the actual mechanism involved, IKKβ promotes weight gain, type II diabetes and cardiovascular disease in the context of HFD-induced obesity [7–9].

We next asked whether the anti-inflammatory effects of E2 are specific to IKKα/β or if other pathways in the hypothalamus are involved. Another signaling molecule that is a known contributor to HFD-induced hypothalamic inflammation is c-Jun-N-terminal kinase (JNK) [18–

21]. Western blotting for phosphorylated/total JNK in the hypothalamus revealed that there were no differences in the amount of active JNK, nor in the total levels of JNK nor in the proportion of active/total JNK among groups (Fig 4.2C). For this reason, we focused our attention on the IKKβ/NF-κB pathway.

Figure 4.2. E2 prevents the activation of the hypothalamic IKKα/β signaling pathway during chronic maintenance on a HFD. Rats were prepared in the manner described in Chapter 3. MBH were Western blotted using antibodies for phosphorylated IKKα/β (pIKKα/β) and IKKβ. Blots were quantified via densitometry (Image J) and analyzed using one-way ANOVA with a Holm-Sidak test for post-hoc multiple comparisons. E2 led to a significant reduction of pIKKα/β without affecting total levels of IKKβ (A). There was no difference in the levels of phosphorylated JNK (pJNK) (B) or of the phosphorylated pRelA subunit of NF-κB (pNF-κB) (C). The total protein levels and the ratio of phosphorylated/total protein levels also did not differ among groups for these proteins. Actin was used as a loading control. Mean ± SEM, P<0.05, n=5-7.

4.4 Influence of Inflammation on ERα Expression

When cultured hypothalamic neurons are treated with palmitate in the absence of E2, inflammation occurs that is associated with reduced ERα expression [5]. Co-treating these cells with E2 prevents the increase of inflammation and fully protects the expression of ERα [5].

Given the inverse association between inflammation and the levels of ERα, we asked whether inflammatory signaling might be directly responsible for the reduced expression of ERα. Based on the well-established role of hypothalamic IKKβ in the progression of obesity and impaired cardiovascular function [5], and our own data that E2 significantly inhibits IKKβ activity (Fig 4.2), we sought to test the hypothesis that IKKβ negatively influences the expression of ERα. This

effect could result from reduced transcription of the ERα gene or by increasing the degradation of ERα protein or mRNA transcripts.

We first chose to examine the involvement of the transcription factor, PGC-1α

(peroxisome proliferator-activated receptor γ coactivator 1-α), in this process, based on its roles in maintaining ERα expression in the hypothalamus and exerting strong effects on metabolism

[5]. Based on evidence that Pgc1α expression is protected in the hypothalamus of females, but not males, during chronic HFD feeding [5], we tested the hypothesis that IKKβ reduces hypothalamic ERα expression by negatively regulating its transcription via Pgc1α. In turn, we expected that the anti-inflammatory effects of E2-ERα signaling thereby maintain a positive- feedback loop with PGC-1α on a HFD (see proposed mechanism in Fig 4.3). Our initial findings demonstrate that E2 can specifically inhibit the IKKβ inflammatory pathway (Fig 4.2), and imply that reducing inflammatory signaling via IKKβ by E2 may mediate the protection of Pgc1α and

ERα expression in response to HFD feeding. The involvement of Pgc1α in this process has not yet been established.

A plausible model for this system (Fig 4.3) is that in conditions of lower hypothalamic

E2/ERα signaling (such as occurs in males), particular saturated fatty acids are synthesized or transported into the brain to a greater extent [5; 22]. In excess, these fatty acids promote the activation of IKKβ [8], which may promote the loss of Pgc1α expression, thereby leading to a proportional reduction in the levels of ERα [5]. The expectation is that below a certain level of hypothalamic ERα expression, cardiovascular protection will be lost during maintenance on a

HFD. The loss of hypothalamic ERα is expected to result in a vicious circle of continually- elevated hypothalamic IKKβ signaling and reduced ERα expression.

Figure 4.3. Proposed mechanism of the anti- inflammatory and cardioprotective effects of E2-ERα signaling. HFD and palmitate (PA) stimulate the increased activation and expression of IKKβ [5; 7]. Inflammation reduces the levels of PGC-1α [5], and we anticipate that this negative regulation occurs via IKKβ. ERα expression in the hypothalamus is maintained by Pgc1α in the presence of E2 [5]. Further, we have found that E2 prevents the activation of IKKβ (Fig 4.2). Hypothalamic ERα is hypothesized to confer cardiovascular protection [5] and when ERα expression is reduced, this is associated with metabolic disease [5; 31].

Inflammation regulates the levels of Pgc1α and ERα

We next directly tested our hypothesis that IKKβ

activation is responsible for the loss of ERα during

chronic maintenance on a HFD. This question was most easily assessed using cultured mouse hypothalamic neurons (N-43, Cedar Lane Labs,

Cellutions). In order to mimic the effects of HFD-feeding on brain cells, we treated cells with the saturated fatty acid, palmitate, because it is the most abundant lipid constituent in our HFD and others [5; 23], and because it is increased in the CNS during diet-induced obesity [5; 8; 22].

Importantly, palmitate recapitulates many of the hypothalamic effects of chronic HFD-feeding [5;

8; 22; 24].

Cells were maintained in Dulbecco’s Modified Eagle Medium (DMEM, Gibco) with 10% fetal bovine serum (FBS, Gibco) until reaching 80-90% confluency. At this time, cells were plated at a density of 5 x 104 cells/2 mL/ well in 12-well sterile culture plates (Falcon®, Corning,

New York). After 24 hours, the medium was removed from each well and replaced with phenol- red-free DMEM (Gibco, so as to avoid interactions of phenol red with estrogen receptors) with

5% charcoal-stripped FBS (Gemini, which removes steroid hormones, lipids, and cytokines).

First, cultured N-43 cultured hypothalamic neurons were treated with palmitate

conjugated to BSA (5%) or with 5% BSA as a control. We found that palmitate reliably reduced

the expression of ERα and Pgc1α, as assessed via qPCR (Fig 4.4A,B). After establishing this

consistent effect, we treated cells with the highly selective IKKβ inhibitor, ML120B (Sigma) or

vehicle (DMSO, Sigma). MLN120B binds competitively at the ATP-binding site of IKKβ, which is

required for the phosphorylation and activation of IκB and NF-κB [25–27] (Fig 4.1). We

assessed the mRNA expression of Pgc1α and ERα, in order to determine the impact of IKKβ

inhibition on the levels of these proteins following PA treatment. Interestingly, IKKβ inhibition

completely protected Pgc1α expression (Fig 4.4A), but only partially protected ERα expression

(Fig 4.4B). This indicates that while IKKβ is fully responsible for the reduction of Pgc1α

expression by PA, it is not fully account for the loss of ERα expression following PA treatment.

Further, Pgc1α does not appear to be fully responsible for maintaining ERα expression during

palmitate treatment, as proposed [5].

Figure 4.4. Effects of IKKβ inhibition on the reduction of Pgc1α and ERα mRNA expression during treatment of cultured hypothalamic neurons with palmitate. Palmitate (PA) treatment led to the reduced expression of Pgc1α (A) and ERα (B) mRNA relative to treatment with 5% BSA, alone. Addition of the inhibitor, ML120B, fully protected against the loss of Pgc1α expression, but only partially restored ERα expression. Data were analyzed using One Way ANOVA and Holm-Sidak test for multiple comparisons. Mean ± SEM, P<0.05, n=6 from 3 separate experiments, in total.

Nonetheless, it is possible that the reduction of ERα levels in the hypothalamus of males maintained on a chronic HFD may not be entirely responsible for the impairment of cardiovascular function. For example, there is some evidence that hypothalamic inflammation can influence cardiovascular function over the short-term [9], an effect which could potentially be independent of changes in ERα expression. In this case, the anti-inflammatory potential of hypothalamic E2-ERα would still be expected to be beneficial in protecting cardiovascular function [5]. Interestingly, a novel E2-selective-pathway agonist has been recently developed that selectively inhibits the activation of NF-κB and other inflammatory pathways [28]. If this compound is efficacious at preventing hypothalamic inflammation in males without inducing any undesirable secondary sex characteristics or other side effects, this could provide many benefits for managing obesity and cardiovascular disease.

REFERENCES

[1] Mozaffarian D, Benjamin EJ, Go AS, Arnett DK, Blaha MJ, Cushman M, et al. Heart Disease and Stroke Statistics-2015 Update: A Report From the American Heart Association. Circulation 2014; 131 (4): e29-322.

[2] Flegal KM, Carroll MD, Ogden CL, Curtin LR. Prevalence and trends in obesity among US adults, 1999-2008. JAMA 2010; 303 (1538–3598 (Electronic)): 235–241.

[3] Lloyd-Jones D, Adams RJ, Brown TM, Carnethon M, Dai S, De Simone G, et al. Heart disease and stroke statistics--2010 update: a report from the American Heart Association. Circulation 2010; 121 (7): e46–e215.

[4] Mikkola TS, Gissler M, Merikukka M, Tuomikoski P, Ylikorkala O. Sex differences in age- related cardiovascular mortality. PLoS One 2013; 8 (5): e63347.

[5] Morselli E, Fuente-Martin E, Finan B, Kim M, Frank A, Garcia-Caceres C, et al. Hypothalamic PGC-1α protects against high-fat diet exposure by regulating ERα. Cell Rep. 2014; 9 (2): 633–45.

[6] Reubinoff BE, Wurtman J, Rojansky N, Adler D, Stein P, Schenker JG, et al. Effects of hormone replacement therapy on weight, body composition, fat distribution, and food intake in early postmenopausal women: a prospective study. Fertil.Steril. 1995; 64 (0015– 0282 (Print)): 963–968.

[7] Zhang X, Zhang G, Zhang H, Karin M, Bai H, Cai D. Hypothalamic IKKbeta/NF-kappaB and ER stress link overnutrition to energy imbalance and obesity. Cell 2008; 135 (1): 61– 73.

[8] Posey KA, Clegg DJ, Printz RL, Byun J, Morton GJ, Vivekanandan-Giri A, et al. Hypothalamic proinflammatory lipid accumulation, inflammation, and insulin resistance in rats fed a high-fat diet. Am. J. Physiol. Endocrinol. Metab. 2009; 296 (5): E1003-12.

[9] Purkayastha S, Zhang G, Cai D. Uncoupling the mechanisms of obesity and hypertension by targeting hypothalamic IKK-β and NF-κB. Nat. Med. 2011; 17 (7): 883–7.

[10] Vegeto E, Belcredito S, Etteri S, Ghisletti S, Brusadelli A, Meda C, et al. Estrogen receptor-alpha mediates the brain antiinflammatory activity of estradiol. Proc. Natl. Acad. Sci. U. S. A. 2003; 100 (16): 9614–9.

[11] Ghisletti S, Meda C, Maggi A, Vegeto E. 17beta-estradiol inhibits inflammatory gene expression by controlling NF-kappaB intracellular localization. Mol. Cell. Biol. 2005; 25 (8): 2957–68.

[12] Guzeloglu-Kayisli O, Halis G, Taskiran S, Kayisli UA, Arici A. DNA-binding ability of NF- kappaB is affected differently by ERalpha and ERbeta and its activation results in inhibition of estrogen responsiveness. Reprod. Sci. 2008; 15 (5): 493–505.

[13] Marino M, Galluzzo P, Ascenzi P. Estrogen signaling multiple pathways to impact gene transcription. Curr.Genomics 2006; 7 (1389–2029 (Print)): 497–508.

[14] Oeckinghaus A, Hayden MS, Ghosh S. Crosstalk in NF-κB signaling pathways. Nat. Immunol. 2011; 12 (8): 695–708.

[15] Chariot A. The NF-κB-independent functions of IKK subunits in immunity and cancer. Trends Cell Biol. 2009; 19 (8): 404–413.

[16] Tsaousidou E, Paeger L, Belgardt BF, Pal M, Wunderlich CM, Brönneke H, et al. Distinct Roles for JNK and IKK Activation in Agouti-Related Peptide Neurons in the Development of Obesity and Insulin Resistance. Cell Rep. 2014; 9 (4): 1495–506.

[17] Valdearcos M, Robblee MM, Benjamin DI, Nomura DK, Xu AW, Koliwad SK. Microglia Dictate the Impact of Saturated Fat Consumption on Hypothalamic Inflammation and Neuronal Function. Cell Rep. 2014; 9 (6): 2124–2139.

[18] Prada PO, Zecchin HG, Gasparetti AL, Torsoni MA, Ueno M, Hirata AE, et al. Western diet modulates insulin signaling, c-Jun N-terminal kinase activity, and insulin receptor substrate-1ser307 phosphorylation in a tissue-specific fashion. Endocrinology 2005; 146 (0013–7227 (Print)): 1576–1587.

[19] Belgardt BF, Mauer J, Wunderlich FT, Ernst MB, Pal M, Spohn G, et al. Hypothalamic and pituitary c-Jun N-terminal kinase 1 signaling coordinately regulates glucose metabolism. Proc. Natl. Acad. Sci. U. S. A. 2010; 107 (13): 6028–33.

[20] Velloso LA, Schwartz MW. Altered hypothalamic function in diet-induced obesity. Int. J. Obes. (Lond). 2011; 35 (12): 1455–65.

[21] Vernia S, Cavanagh-Kyros J, Barrett T, Jung DY, Kim JK, Davis RJ. Diet-induced obesity mediated by the JNK/DIO2 signal transduction pathway. 2013; 27 (21): 2345–55.

[22] Rodriguez-Navas C, Morselli E, Clegg DJ. Sexually dimorphic brain fatty acid composition in low and high fat diet-fed mice. Mol. Metab. 2016; 5 (8): 680–9.

[23] Woods SC, Seeley RJ, Rushing PA, D’Alessio D, Tso P. A controlled high-fat diet induces an obese syndrome in rats. J Nutr. 2003; 133 (0022–3166 SB–IM): 1081–1087.

[24] Ji H, Friedman MI. Fasting plasma triglyceride levels and fat oxidation predict dietary obesity in rats. Physiol Behav. 2003; 78 (0031–9384 (Print)): 767–772.

[25] Uitdehaag JCM, Verkaar F, Alwan H, de Man J, Buijsman RC, Zaman GJR. A guide to picking the most selective kinase inhibitor tool compounds for pharmacological validation of drug targets. Br. J. Pharmacol. 2012; 166 (3): 858–76.

[26] Davis MI, Hunt JP, Herrgard S, Ciceri P, Wodicka LM, Pallares G, et al. Comprehensive analysis of kinase inhibitor selectivity. Nat. Biotechnol. 2011; 29 (11): 1046–51.

[27] Nottingham LK, Yan CH, Yang X, Si H, Coupar J, Bian Y, et al. Aberrant IKKα and IKKβ cooperatively activate NF-κB and induce EGFR/AP1 signaling to promote survival and migration of head and neck cancer. Oncogene 2014; 33 (9): 1135–47.

[28] Chadwick CC, Chippari S, Matelan E, Borges-Marcucci L, Eckert AM, Keith JC, et al. Identification of pathway-selective estrogen receptor ligands that inhibit NF- B transcriptional activity. Proc. Natl. Acad. Sci. 2005; 102 (7): 2543–2548.

[29] Alba Gonzalez-Franquesa VDNCL and PMG-R. Skeletal Muscle - From Myogenesis to Clinical Relations. InTech; 2012.

[30] Bakkar N, Guttridge DC. NF-kappaB signaling: a tale of two pathways in skeletal

myogenesis. Physiol. Rev. 2010; 90 (2): 495–511.

[31] Musatov S, Chen W, Pfaff DW, Mobbs C V, Yang X-JJ, Clegg DJ, et al. Silencing of estrogen receptor alpha in the ventromedial nucleus of hypothalamus leads to metabolic syndrome. Proc.Natl.Acad.Sci.U.S.A 2007; 104 (0027–8424 (Print)): 2501–2506.

Chapter 5:

Influence of Insulin on Hypothalamic Inflammation

5.1 Inflammatory Effects of Insulin

As described in the previous chapter, it is clear that hypothalamic inflammation promotes weight gain and insulin resistance and impairs cardiovascular function during conditions of diet- induced obesity. However, when hypothalamic inflammatory signaling is exogenously activated, the effect is generally catabolic [1], with reductions in food intake and body weight [2–4].

Indeed, the link between inflammatory cytokines and cachexia resulting from cancer [5] and infections [6] has been known for nearly 40 years. Recently, it was demonstrated that leptin, a member of the cytokine family [7; 8], exerts its catabolic effects by activating the inflammatory

IKKβ-NF-κB signaling pathway in the hypothalamus [9]. While conducting the studies described in the previous chapter, we found that insulin increases inflammatory IKKα/β signaling in cultured hypothalamic neurons (Fig 5.1A), raising the possibility that insulin may partly exert its catabolic effects by promoting hypothalamic inflammation.

In these studies, immortalized rat hypothalamic neurons (IV-B cells, a generous gift from

Renu Sah) were maintained as described in Chapter 5 and were plated at a density of 5 x 104 cells/well in 12-well culture plates. After 24 hours, the culture medium was aspirated from each well and was replaced with DMEM with 5% charcoal-stripped FBS. Cells were treated 24 hours later with insulin or vehicle (PBS, Gibco). After the treatment period was complete, cells were collected on ice using protein lysis buffer (Thermo Fisher) with a combination of phosphatase inhibitors (Fisher) and protease inhibitors (Fisher). After homogenizing the lysates on ice, samples were analyzed for protein content using BCA reagent (Thermo Fisher) and were then prepared for Western blotting, as described in Chapter 4.

We were surprised to find that while insulin exerted its typical effects by promoting the phosphorylation of insulin signaling (data not shown), it also stimulated the phosphorylation of the inflammatory IKKα/β signaling protein (Fig 5.1B), which is involved in the NF-κB signaling

pathway, as described in Chapter 4. This activation remained persistent for the full 2-hour duration of the experiment. Furthermore, at a higher dose of insulin, IKKβα/β was stimulated even more strongly. These findings were consistent with a recent report, in which insulin was found to induce the activation of IKKβ in cultured 3T3-L1 adipocytes [10].

Figure 5.1. Insulin activates inflammatory signaling in cultured hypothalamic neurons. A) Treatment of cultured hypothalamic neurons with insulin (10 nM) or PBS (0 nM) promotes the phosphorylation of proximal and downstream insulin signaling proteins, IRS-1/2 and FoxO1, respectively. Protein levels of IRS-1 and FoxO1 were unaltered by insulin treatment, but varied slightly depending on the initial time of treatment. Actin was used as a loading control. B) Insulin induced the activation of IKKα/β but did not affect the levels of total IKKβ during the course of the experiment. IKKβ/α activation was sustained over the course of the experiment and the effects were dose-dependent.

In order to test the hypothesis that insulin exerts its catabolic effects by inducing hypothalamic inflammation, we directly injected insulin into the 3rd ventricle and examined the expression of inflammatory cytokines. In this study, male Long-Evans rats were outfitted with stainless steel cannulas that targeted the third cerebral ventricle, as previously described [11].

Briefly, rats were anesthetized with ketamine/xylazine (60 mg/kg and 10 mg/kg, respectively)

along with buprenorphine (0.05 mg/kg) and the scalp was shaved. The surgical site was disinfected with alternating applications of betadine and ethanol. After securing rats in a stereotaxic device (Harvard Apparatus, Massachusetts), an incision was then made along the scalp, in order to expose bregma. After locating the coordinates for cannula insertion, a window into the skull was drilled -1 mm caudal from bregma and -1 mm lateral from bregma (right side).

Next, 4 screws were secured into the skull and a stainless steel cannula (PlasticsOne) was inserted ventrally 7.5 mm at a 10o inward angle in order to target the third ventricle. Dental acrylic was then applied to secure the cannula in place and was allowed to harden for several minutes prior to sealing the cannula with an obturator. Rats were then allowed to recover for 2 weeks.

Correct cannula placement was initially tested using the angiotensin II test, in which 10 ng of angiotensin II was administered to each rat [11; 12]. Rats that drank at less than 5 mL water within 5 minutes of injection were excluded from the study. In order to ensure that cannulas remained in-place in all rats following the recovery period, cannula placement was also verified at the end of the study by visual inspection of the cannula track. Rats were excluded if the cannula track did not intersect with the ventricle, or if there was noticeable clotting or irritation around the base of the ventricle. Based on these tests, 18% of the original group was excluded from the final analysis. Rats were provided an additional 1-week period of recovery prior to the start of the experiment. All rats were then repeatedly and individually handled for ~2-5 minutes each day, in order to acclimate them to being held during injections of the third-ventricle (i3vt), as previously described [13].

On the day prior to the study, rats were fasted overnight, in order to avoid any possible effects of food intake on hypothalamic inflammation. On the day of the study, rats were administered a 2-μL i3vt injection of insulin (10 mU) or vehicle (artificial CSF, aCSF). This dose of insulin approximates postprandial levels of insulin in the CSF when distributed (1 ng/mL) but

initially provides a higher concentration of insulin at the hypothalamic ventricular wall, enabling insulin to penetrate the neuropil more effectively [14]. As a positive control for the induction of hypothalamic inflammation, LPS (20 μg/2 μL aCSF) was administered i3vt. Brains were collected and mediobasal hypothalami were microdissected after 1.5 or 4 hours post-treatment using a Brain Block™. Samples were immediately stored in tubes filled with 2 mL of RNA later and were maintained at 4oC for one day prior to RNA isolation (RNAqueous Micro, Ambion) and cDNA preparation (iScript, BioRad). Assays for qPCR are listed in Table 2, of Appendix I.

When assessing M1 inflammatory markers via qPCR, including IL-6, TNFα, and IL-1β, we found that insulin did not reduce inflammation and the data were indistinguishable from those in the aCSF-treated group at both 1.5 hours (Fig 5.2) and 4 hours post-treatment. As a positive control for our methods, rats administered i3vt-LPS had significantly-elevated mRNA expression of IL-6 and IL-1β at the 4 hour time point (Fig 5.3A,C). To address the possibility that insulin may increase anti-inflammatory cytokine expression, we assessed IL-10 mRNA levels in the same samples and also found no difference in expression relative to the aCSF group (Fig 5.2D).

Figure 5.2. Inflammatory cytokine mRNA expression at 1.5 hours after the administration of insulin, artificial CSF (aCSF) or LPS (20 μg) into the third ventricle. No significant differences were observed in the expression of IL-6 (A), TNFα (B), IL-1β (C), and IL-10 (D) at this time point. n=6, mean ± SEM, P<0.05

Figure 5.3. Inflammatory cytokine expression at 4 hours after the administration of insulin, artificial CSF (aCSF) or LPS (20 μg) into the third ventricle. IL-6 (A), TNFα (B), IL-1β (C), and IL- 10 (D). n=6, mean ± SEM, P<0.05,

Because insulin did not increase the expression hypothalamic inflammatory cytokines under normal conditions, we further asked why insulin increases inflammation in vitro but not in vivo. Despite injecting insulin i3vt at a broad array of different doses and time points, we were unable to detect an increase in hypothalamic cytokine expression or in the activation of IKKβ signaling in the hypothalamus (data not shown). Consistent with this, several reports have described insulin’s pronounced anti-inflammatory effects in the context of inflammatory conditions [15–19].

5.2 Anti-Inflammatory Effects of Insulin

Systemic insulin treatment reduces mortality and reduces the levels of inflammatory cytokines detected in the blood, liver, and hypothalamus in a rat model of LPS-induced sepsis

[15; 17]. Importantly, the effects of insulin on LPS-induced sepsis and mortality can be entirely reproduced by administering insulin into the third ventricle [18]. Further, in a mouse model of

Alzheimer’s disease, insulin administered intranasally significantly reduces the proportion of reactive microglia [16]. Insulin is thought to reduce inflammation under these conditions via direct and indirect mechanisms.

Initial studies found that insulin can directly inhibit inflammation in human macrophages challenged with endotoxemia in vitro [20; 21]. More recently, insulin was hypothesized to reduce endotoxemia by promoting anti-inflammatory tone through the parasympathetic and/or sympathetic nervous system [22; 23], although this possibility remains to be directly tested [15;

18]. For example, when the cholinergic activity of peripheral nerves is stimulated during an inflammatory challenge with LPS, TNF production by the spleen is inhibited [24]; these anti- inflammatory effects are completely lost in animals with splenic denervation [22; 24], demonstrating the importance of the peripheral nervous system in influencing the inflammatory response.

As previously mentioned, obesity is often accompanied by hypothalamic inflammation

[25–29], and there is sufficient evidence supporting the conclusion that this chronic low-grade inflammation directly contributes to weight gain [25; 30]. While insulin could potentially minimize inflammation in this context, repeated administration of systemic insulin tends to promote weight gain [31; 32] and hypoglycemia [33], making this treatment route undesirable for the prevention of obesity. In contrast, central administration of insulin provides clinically meaningful weight- lowering benefits while avoiding hypoglycemia [34–36]. Because insulin can exert its effects on systemic inflammation even when administered solely into the third ventricle [18], we hypothesized that central insulin can also reduce the chronic low-grade hypothalamic inflammation observed in obesity. In so doing, we expect that insulin will minimize the risk of cardiovascular disease and type II diabetes, while also promoting weight loss. Ultimately, insulin could be administered intranasally prior to eating foods associated with hypothalamic inflammation in order to prevent these risk factors over time. Intranasal insulin is a well- established brain-delivery technique that reduces body fat [34], improves cognitive function [37;

38], and prevents cognitive decline in Alzheimer’s disease [39; 40]. Importantly, intranasal insulin treatment is a safe and effective strategy, and has no known negative side-effects to- date [34; 41].

In order to test the ability of insulin to act via the brain to reduce diet-induced inflammation, we chose to feed rats a HFD. A model of HFD-induced inflammation is ideal for testing our hypothesis [42], because it was recently found that mice and rats maintained on a high-fat (60% kcal from fat, Research Diets, D12450B), high-sucrose diet had significantly increased hypothalamic inflammation between 1-3 days of feeding [42]. Not only was cytokine expression increased, but microglia accumulated in the hypothalamus even within this short time frame [42]. During the progression of obesity over the course of 8 months, microglia continued to accumulate in the hypothalamus of HFD-fed rats to a greater extent than in chow-

fed rats. Subsequently, another lab reported similar findings when feeding the same 60% HFD to mice [43].

We therefore chose to utilize this model of diet-induced inflammation to determine the anti-inflammatory efficacy of central insulin within the hypothalamus. In order to completely reproduce the experiments originally reported by Thaler et al. [42], male Long-Evans rats (250 g) were maintained on standard lab chow diet (Envigo #7012) for 2 weeks prior to the study. On the first day of the study, half of the rats were provided ad-libitum access to a 60% HFD

(Research Diets, D12450B) prior to the onset of dark, while the other half were provided a fresh hopper of chow diet. After three days, all rats were sacrificed prior to the onset of dark and mediobasal hypothalami were removed, as described [42].

Although HFD-fed rats gained significantly more weight than chow-fed rats (Fig 5.4A), we were surprised to find no significant increases in the amount of hypothalamic inflammation when examining the mRNA expression of inflammatory markers after rats consumed HFD for 24 hours (Fig 5.4B), nor after 48 hours (Fig 5.4C). These discrepancies from the previously- mentioned studies were surprising, but perhaps not entirely unexpected; other reports using similar diets and experimental conditions have found no such acute effects of HFDs on hypothalamic inflammation [25; 44]. Regardless of these opposite findings in rodent models across different labs, it is still clear that obesity is an inflammatory condition when observed over the long-term in both humans and rodents [25; 26; 28; 42].

Figure 5.4. Acute HFD feeding does not significantly increase hypothalamic inflammation. A) Daily body weight gain was significantly higher in male rats fed a HFD. No significant differences were observed in the mRNA expression of any of the candidate inflammatory markers in the mediobasal hypothalamus of rats after 24 h of HFD-feeding (B) or after 48 h of HFD-feeding (C). Mean ± SEM, P<.05, n=6-7.

While it would be ideal to use a rodent model of chronic obesity in order to induce reliable hypothalamic inflammation, there are several reports suggesting that central insulin is ineffective once inflammation is established [13; 25; 45]. Therefore, in order to test whether central insulin can prevent diet-induced inflammation over this long time period, we would need to infuse insulin into the brain for several months. Thus, an alternative model of inflammation would be desirable in order to avoid the time and expenses involved in a 4-8-week HFD-feeding study with osmotic minipumps.

In order to directly test the ability of insulin to inhibit hypothalamic inflammation, we chose to challenge rats with LPS, which is known to be chronically elevated in obese individuals

[46–48], especially after HFD intake [47; 49] The relative contribution of LPS to chronic low- grade inflammation and weight gain has been demonstrated in mice, using osmotic minipumps.

In these studies, LPS was infused at levels observed during obesity for a period of 4 weeks;

LPS treatment led to significantly increased weight gain and adiposity [30]. Importantly, LPS reliably induces inflammation in rodent models within a short time frame, making it ideal for studying the ability of central insulin to blunt the effects of hypothalamic inflammation [15; 17].

We first verified the ability of LPS to acutely induce hypothalamic inflammation.

Previous studies using LPS to induce hypothalamic inflammation injected an ip bolus of LPS 7-

24 hours prior to sacrifice [15; 17]. We chose to terminate our study 4 hours after LPS administration, because centrally-administered insulin was found to prevent LPS-induced systemic inflammation within this time period [18]. Male Long-Evans rats (~250 g) were administered LPS (20 mg/kg or 5 mg/kg) or vehicle (saline) and hypothalami were collected 4 hours later. In addition, rats were anesthetized with isoflurane for 15 minutes prior to LPS injection in order to account for the possibility that anesthesia may inhibit inflammation [50; 51].

Isoflurane was specifically selected because it has the smallest effect on LPS-induced inflammation based on previous reports [50]. Hypothalamic expression of inflammatory cytokine mRNA levels revealed that the 20 mg/kg dose of LPS significantly induced inflammation in the hypothalamus within the 4-hour time frame (Fig 5.5).

Figure 5.5. LPS increases hypothalamic inflammation within 4 h of treatment.

LPS administered ip at a dose 5 mg/kg (LPS 5) or 20 mg/kg (LPS 20) led to a significant increase in the hypothalamic mRNA expression of IL1β (A) and TNFα (B), only at the 20 mg/kg dose. Mean ± SEM, P<.05, n=3.

Next, we tested the hypothesis that central insulin can prevent LPS-induced hypothalamic inflammation. Anti-inflammatory molecules are generally administered at least 30 minutes prior to the inflammatory event [10; 52; 53]. A pre-treatment time period of 40 minutes was allotted, because insulin has been shown to exert signaling and behavioral effects in the hypothalamus within this time frame, and can exert effects that persist for at least 4 to 24 hours

[15; 35; 54; 55]. On the day of the study, male Long-Evans rats (250 g, Envigo, Indiana) were anesthetized with isoflurane and administered buprenorphine (0.05 mg/kg). After shaving and disinfecting the surgical site in the same manner as described above, a small hole was drilled into the skull with the coordinates of -1 mm lateral to the midline (right side), -1.5 mm posterior to bregma in order to target the lateral ventricle, as previously described [56]. Next, a needle was directly inserted 4 mm ventrally from the dural surface. A 10-μL bolus of insulin (10 μg/rat) or aCSF was then administered into the lateral ventricle. This dose has been repeatedly used to induce food intake in rats [14; 57; 58]. After 2 minutes, the needle was withdrawn and pressure was applied to the injection site for 1 minute to prevent backflow. Gelfoam® (Pfizer,

New York) was applied in order to plug the hole in the skull, and surgical staples were applied.

Rats were then removed from isoflurane anesthesia and were returned to their home cage. 40 minutes after the injection of insulin, rats were administered ip LPS (10 mg/kg) or vehicle

(saline). This dose of LPS was expected to induce substantial inflammation while allowing the effects of insulin to be seen more clearly. After 4 hours, mediobasal hypothalami and hippocampi were microdissected and stored in RNA later. RNA was later extracted after this time and cytokine mRNA expression was assessed via qPCR, as described above.

Because insulin administered into the lateral ventricle has been repeatedly found to activate hypothalamic signaling [59–61], it is likely that the insulin injected in this experiment effectively reached the hypothalamus, as its bulk flow moved towards the third ventricle and subsequently to the fourth ventricle [60]. However, the mRNA expression of inflammatory

cytokines did not appear to be affected by insulin (Fig 5.6,A-E), raising the possibility that the administered insulin did not reach the third ventricle [62; 63]. As a control, we also examined inflammation of the hippocampus, which is adjacent to the insulin-injection site and thus had full exposure to insulin following injection into the lateral ventricle. Further, the hippocampus also has abundant insulin receptor expression [64; 65]. Similarly, insulin did not appear to affect inflammation in the hippocampus of LPS-injected rats (Fig 5.6D-F), confirming that even when insulin was in direct contact with the tissue, it still did not attenuate LPS-induced inflammation.

Figure 5.6. Central insulin is unlikely to attenuate inflammation of the hypothalamus or hippocampus. LPS (10 mg/kg) significantly increased inflammation in the hypothalamus (A- C) and hippocampus (D- F) of male rats 4 h following ip injection. The dot plots are reveal strong variability of inflammation between animals. Insulin did not affect the mRNA expression of the inflammatory cytokines: IL-6 (A,D), TNFα (B,E), IL-1β (C,F) in both brain regions. Mean ± SEM, P<0.05, n=3

Because our insulin was functional (Figs 5.1 and Paper II, Fig 5E), these findings imply that insulin had no effect on LPS-induced hypothalamic inflammation at the given dose. This is surprising when considering insulin’s anti-inflammatory effects in many reports [15–18].

Unfortunately, previous studies did not report the effects of centrally-administered insulin on

LPS-induced inflammation in the brain [18]. However, we initially expected that central insulin would have reduced inflammation in that study, because central inflammation and peripheral inflammation are usually directly proportional [15; 28; 66; 67]. It is also possible that the dose of

LPS was too high for insulin to overcome. This is unlikely, considering that previous studies have used even higher doses of LPS than our current study, but yet they saw that insulin was effective at providing anti-inflammatory effects throughout the body and brain [15; 17; 18].

It is also important to keep in mind that due to the limited sample sizes, the power of the final study was limited. It also appears as though one of the rats treated with insulin had a very limited response to LPS, although this could be indicative of an anti-inflammatory effect of insulin. We cannot come to a definitive conclusion based on these studies alone, but it appears that there are unknown factors influencing the interaction between insulin and inflammation interaction, and these were presumably at work in our experiment.

Our study differed in several ways from previous publications [18], which may explain why no effect was observed with central insulin administration under the given conditions. A former study infused insulin into the third cerebral ventricle, which more directly targets the hypothalamus [18]. We injected insulin into the lateral ventricle in order to minimize disturbance of the hypothalamus, as well as to more accurately simulate the concentration of hypothalamic insulin that would be observed during intranasal insulin administration. Additionally, the former study infused a continuous, low concentration of insulin (0.02 mU/h) by infusing insulin via osmotic minipumps for 4 hours during LPS-induced sepsis [18], while our study used a single bolus of insulin at a larger dose. From a therapeutic standpoint, chronic infusion into the CNS is

not feasible for humans, due to the risks involved with icv surgery. However, intranasal insulin can be safely administered in a bolus dose only 2-4 times daily, so as to avoid developing nasal irritation [16; 34; 37; 39]. In this way, our test of a single dose of central insulin is more clinically relevant [34; 37].

A remaining question is whether a different dose of central insulin would be capable of minimizing LPS-induced inflammation, or whether a bolus dose of insulin will be ineffective at any dose used. Considering that higher doses of insulin are reported to activate sympathetic responses of increased heart rate and blood pressure [68], there is a possibility that higher doses may exert different effects on inflammation. While we cannot reach a definitive conclusion based on the data collected thus far, it appears that while insulin can be efficacious at minimizing inflammation during endotoxemia and Alzheimer’s disease, its use as a therapeutic for preventing inflammation during obesity is less-likely.

REFERENCES

[1] Thaler JP, Choi SJ, Schwartz MW, Wisse BE. Hypothalamic inflammation and energy homeostasis: resolving the paradox. Front. Neuroendocrinol. 2010; 31 (1): 79–84.

[2] Romanatto T, Cesquini M, Amaral ME, Roman EA, Moraes JC, Torsoni MA, et al. TNF- alpha acts in the hypothalamus inhibiting food intake and increasing the respiratory quotient--effects on leptin and insulin signaling pathways. Peptides 2007; 28 (5): 1050–8.

[3] Fantino M, Wieteska L. Evidence for a direct central anorectic effect of tumor-necrosis- factor-alpha in the rat. Physiol. Behav. 1993; 53 (3): 477–83.

[4] Plata-Salamán CR, Sonti G, Borkoski JP, Wilson CD, French-Mullen JM 04189@brahmsudeledu. Anorexia induced by chronic central administration of cytokines at estimated pathophysiological concentrations. Physiol. Behav. 1996; 60 (3): 867–75.

[5] Oliff A, Defeo-Jones D, Boyer M, Martinez D, Kiefer D, Vuocolo G, et al. Tumors secreting human TNF/cachectin induce cachexia in mice. Cell 1987; 50 (4): 555–63.

[6] Michie HR, Sherman ML, Spriggs DR, Rounds J, Christie M, Wilmore DW. Chronic TNF infusion causes anorexia but not accelerated nitrogen loss. Ann. Surg. 1989; 209 (1): 19– 24.

[7] Lord GM. Leptin as a proinflammatory cytokine. Contrib. Nephrol. 2006; 151: 151–64.

[8] Zhang F, Basinski MB, Beals JM, Briggs SL, Churgay LM, Clawson DK, et al. Crystal structure of the obese protein Ieptin-E100. Nature 1997; 387 (6629): 206–209.

[9] Jang P-G, Namkoong C, Kang GM, Hur M-W, Kim S-W, Kim GH, et al. NF-kappaB activation in hypothalamic pro-opiomelanocortin neurons is essential in illness- and leptin- induced anorexia. J. Biol. Chem. 2010; 285 (13): 9706–15.

[10] Lancaster GI, Skiba B, Yang C, Nicholls HT, Langley KG, Chan MHS, et al. IκB kinase β (IKKβ) does not mediate feedback inhibition of the insulin signalling cascade. Biochem. J. 2012; 442 (3): 723–32.

[11] Begg DP, May AA, Mul JD, Liu M, D’Alessio DA, Seeley RJ, et al. Insulin Detemir Is Transported From Blood to Cerebrospinal Fluid and Has Prolonged Central Anorectic Action Relative to NPH Insulin. Diabetes 2015; 64 (7): 2457–66.

[12] Epstein AN, Fitzsimons JT, Rolls BJ. Drinking induced by injection of angiotensin into the brain of the rat. J. Physiol. 1970; 210 (2): 457–474.

[13] Clegg DJ, Gotoh K, Kemp C, Wortman MD, Benoit SC, Brown LM, et al. Consumption of a high-fat diet induces central insulin resistance independent of adiposity. Physiol. Behav. 2011; 103 (1): 10–6.

[14] Begg DP, Woods SC. The central insulin system and energy balance. Handb.Exp.Pharmacol. 2012; 209 (0171–2004 (Print)): 111–129.

[15] Duan K, Yu W, Lin Z, Tan S, Bai X, Gao T, et al. Insulin ameliorating endotoxaemia- induced muscle wasting is associated with the alteration of hypothalamic neuropeptides and inflammation in rats. Clin. Endocrinol. (Oxf). 2015; 82 (5): 695–703.

[16] Chen Y, Zhao Y, Dai C-L, Liang Z, Run X, Iqbal K, et al. Intranasal insulin restores insulin signaling, increases synaptic proteins, and reduces Aβ level and microglia activation in the brains of 3xTg-AD mice. Exp. Neurol. 2014; 261: 610–9.

[17] Chen Q, Yu W, Shi J, Shen J, Gao T, Zhang J, et al. Insulin alleviates the inflammatory response and oxidative stress injury in cerebral tissues in septic rats. J. Inflamm. (Lond). 2014; 11 (1): 18.

[18] Milsom T. Central insulin signaling as a regulator of systemic inflammation. 2014;

[19] Li H, Liu B, Huang J, Chen H, Guo X, Yuan Z. Insulin inhibits lipopolysaccharide-induced nitric oxide synthase expression in rat primary astrocytes. Brain Res. 2013; 1506: 1–11.

[20] Leffler M, Hrach T, Stuerzl M, Horch RE, Herndon DN, Jeschke MG. Insulin Attenuates Apoptosis and Exerts Anti-Inflammatory Effects in Endotoxemic Human Macrophages. J. Surg. Res. 2007; 143 (2): 398–406.

[21] Das UN. Is insulin an antiinflammatory molecule? Nutrition 2001; 17 (5): 409–413.

[22] Tracey KJ. Reflex control of immunity. Nat. Rev. Immunol. 2009; 9 (6): 418–428.

[23] Steinman L. Elaborate interactions between the immune and nervous systems. Nat. Immunol. 2004; 5 (6): 575–581.

[24] Rosas-Ballina M, Ochani M, Parrish WR, Ochani K, Harris YT, Huston JM, et al. Splenic nerve is required for cholinergic antiinflammatory pathway control of TNF in endotoxemia. Proc. Natl. Acad. Sci. U. S. A. 2008; 105 (31): 11008–13.

[25] Zhang X, Zhang G, Zhang H, Karin M, Bai H, Cai D. Hypothalamic IKKbeta/NF-kappaB and ER stress link overnutrition to energy imbalance and obesity. Cell 2008; 135 (1): 61– 73.

[26] Xu H, Barnes GT, Yang Q, Tan G, Yang D, Chou CJ, et al. Chronic inflammation in fat plays a crucial role in the development of obesity-related insulin resistance. J.Clin.Invest 2003; 112 (0021–9738 (Print)): 1821–1830.

[27] Shoelson SE, Herrero L, Naaz A. Obesity, inflammation, and insulin resistance. Gastroenterology 2007; 132 (0016–5085 (Print)): 2169–2180.

[28] Morselli E, Fuente-Martin E, Finan B, Kim M, Frank A, Garcia-Caceres C, et al. Hypothalamic PGC-1α protects against high-fat diet exposure by regulating ERα. Cell Rep. 2014; 9 (2): 633–45.

[29] Purkayastha S, Zhang G, Cai D. Uncoupling the mechanisms of obesity and hypertension by targeting hypothalamic IKK-β and NF-κB. Nat. Med. 2011; 17 (7): 883–7.

[30] Cani PD, Amar J, Iglesias MA, Poggi M, Knauf C, Bastelica D, et al. Metabolic endotoxemia initiates obesity and insulin resistance. Diabetes 2007; 56 (7): 1761–72.

[31] Raslová K, Tamer SC, Clauson P, Karl D. Insulin detemir results in less weight gain than NPH insulin when used in basal-bolus therapy for type 2 diabetes mellitus, and this advantage increases with baseline body mass index. Clin. Drug Investig. 2007; 27 (4): 279–85.

[32] Zachariah S, Sheldon B, Shojaee-Moradie F, Jackson NC, Backhouse K, Johnsen S, et al. Insulin detemir reduces weight gain as a result of reduced food intake in patients with type 1 diabetes. Diabetes Care 2011; 34 (7): 1487–91.

[33] The NICE-SUGAR Study. Intensive versus Conventional Glucose Control in Critically Ill Patients. N. Engl. J. Med. 2009; 360 (13): 1283–1297.

[34] Hallschmid M, Benedict C, Schultes B, Fehm HL, Born J, Kern W. Intranasal insulin reduces body fat in men but not in women. Diabetes 2004; 53 (11): 3024–3029.

[35] Clegg DJ, Brown LM, Woods SC, Benoit SC. Gonadal hormones determine sensitivity to central leptin and insulin. Diabetes 2006; 55 (4): 978–987.

[36] Air EL, Benoit SC, Blake Smith KA, Clegg DJ, Woods SC. Acute third ventricular administration of insulin decreases food intake in two paradigms. Pharmacol Biochem. 2002; 72 (0091–3057 (Print)): 423–429.

[37] Hallschmid M, Benedict C, Schultes B, Born J, Kern W. Obese men respond to cognitive but not to catabolic brain insulin signaling. Int. J. Obes. (Lond). 2008; 32 (2): 275–82.

[38] Benedict C, Kern W, Schultes B, Born J, Hallschmid M. Differential sensitivity of men and women to anorexigenic and memory-improving effects of intranasal insulin. J Clin.Endocrinol.Metab 2008; 93 (0021–972X (Print)): 1339–1344.

[39] Maimaiti S, Anderson KL, DeMoll C, Brewer LD, Rauh BA, Gant JC, et al. Intranasal Insulin Improves Age-Related Cognitive Deficits and Reverses Electrophysiological Correlates of Brain Aging. J. Gerontol. A. Biol. Sci. Med. Sci. 2016; 71 (1): 30–9.

[40] Craft S, Baker LD, Montine TJ, Minoshima S, Watson GS, Claxton A, et al. Intranasal insulin therapy for Alzheimer disease and amnestic mild cognitive impairment: a pilot clinical trial. Arch. Neurol. 2012; 69 (1): 29–38.

[41] Stockhorst U, de FD, Steingrueber HJ, Scherbaum WA. Insulin and the CNS: effects on food intake, memory, and endocrine parameters and the role of intranasal insulin administration in humans. Physiol Behav. 2004; 83 (0031–9384 (Print)): 47–54.

[42] Thaler JP, Yi C-X, Schur EA, Guyenet SJ, Hwang BH, Dietrich MO, et al. Obesity is associated with hypothalamic injury in rodents and humans. J. Clin. Invest. 2012; 122 (1): 153–62.

[43] Waise TMZ, Toshinai K, Naznin F, NamKoong C, Md Moin AS, Sakoda H, et al. One-day high-fat diet induces inflammation in the nodose ganglion and hypothalamus of mice. Biochem. Biophys. Res. Commun. 2015; 464 (4): 1157–62.

[44] Lee YS, Li P, Huh JY, Hwang IJ, Lu M, Kim JI, et al. Inflammation is necessary for long- term but not short-term high-fat diet-induced insulin resistance. Diabetes 2011; 60 (10): 2474–83.

[45] Holland WL, Bikman BT, Wang L-P, Yuguang G, Sargent KM, Bulchand S, et al. Lipid- induced insulin resistance mediated by the proinflammatory receptor TLR4 requires saturated fatty acid-induced ceramide biosynthesis in mice. J. Clin. Invest. 2011; 121 (5): 1858–70.

[46] Lassenius MI, Pietiläinen KH, Kaartinen K, Pussinen PJ, Syrjänen J, Forsblom C, et al.

Bacterial endotoxin activity in human serum is associated with dyslipidemia, insulin resistance, obesity, and chronic inflammation. Diabetes Care 2011; 34 (8): 1809–15.

[47] Clemente-Postigo M, Queipo-Ortuño MI, Murri M, Boto-Ordoñez M, Perez-Martinez P, Andres-Lacueva C, et al. Endotoxin increase after fat overload is related to postprandial hypertriglyceridemia in morbidly obese patients. J. Lipid Res. 2012; 53 (5): 973–8.

[48] Hawkesworth S, Moore SE, Fulford AJC, Barclay GR, Darboe AA, Mark H, et al. Evidence for metabolic endotoxemia in obese and diabetic Gambian women. Nutr. Diabetes 2013; 3 (8): e83.

[49] Erridge C, Attina T, Spickett CM, Webb DJ. A high-fat meal induces low-grade endotoxemia: evidence of a novel mechanism of postprandial inflammation. Am. J. Clin. Nutr. 2007; 86 (5): 1286–92.

[50] Suliburk JW, Helmer KS, Gonzalez EA, Robinson EK, Mercer DW. Ketamine attenuates liver injury attributed to endotoxemia: Role of cyclooxygenase-2. Surgery 2005; 138 (2): 134–140.

[51] Tanaka T, Kai S, Matsuyama T, Adachi T, Fukuda K, Hirota K. General anesthetics inhibit LPS-induced IL-1β expression in glial cells. PLoS One 2013; 8 (12): e82930.

[52] Aljada A, Ghanim H, Saadeh R, Dandona P. Insulin inhibits NFkappaB and MCP-1 expression in human aortic endothelial cells. J. Clin. Endocrinol. Metab. 2001; 86 (1): 450–3.

[53] Uitdehaag JCM, Verkaar F, Alwan H, de Man J, Buijsman RC, Zaman GJR. A guide to picking the most selective kinase inhibitor tool compounds for pharmacological validation of drug targets. Br. J. Pharmacol. 2012; 166 (3): 858–76.

[54] Niswender KD, Morrison CD, Clegg DJ, Olson R, Baskin DG, Myers Jr. MG, et al. Insulin activation of phosphatidylinositol 3-kinase in the hypothalamic arcuate nucleus: a key mediator of insulin-induced anorexia. Diabetes 2003; 52 (0012–1797 (Print)): 227–231.

[55] Brown LM, Clegg DJ, Benoit SC, Woods SC. Intraventricular insulin and leptin reduce food intake and body weight in C57BL/6J mice. Physiol Behav. 2006; 89 (0031–9384 (Print)): 687–691.

[56] DeVos SL, Miller T. Direct Intraventricular Delivery of Drugs to the Rodent Central Nervous System. J. Vis. Exp. 2013; (75): e50326–e50326.

[57] D’Alessio DA, Kieffer TJ, Taborsky GJ, Havel PJ. Activation of the parasympathetic nervous system is necessary for normal meal-induced insulin secretion in rhesus macaques. J. Clin. Endocrinol. Metab. 2001; 86 (3): 1253–9.

[58] Holt SH, Brand Miller JC, Petocz P. Interrelationships among postprandial satiety, glucose and insulin responses and changes in subsequent food intake. Eur. J. Clin. Nutr. 1996; 50 (12): 788–97.

[59] Carvalheira JBC, Ribeiro EB, Araújo EP, Guimarães RB, Telles MM, Torsoni M, et al. Selective impairment of insulin signalling in the hypothalamus of obese Zucker rats. Diabetologia 2003; 46 (12): 1629–40.

[60] Baskin DG, Woods SC, West DB, van HM, Posner BI, Dorsa DM, et al.

Immunocytochemical detection of insulin in rat hypothalamus and its possible uptake from cerebrospinal fluid. Endocrinology 1983; 113 (0013–7227 (Print)): 1818–1825.

[61] Widdowson PS, Upton R, Buckingham R, Arch J, Williams G. Inhibition of food response to intracerebroventricular injection of leptin is attenuated in rats with diet-induced obesity. Diabetes 1997; 46 (0012–1797 (Print)): 1782–1785.

[62] Woods SC, Begg DP. Food for Thought: Revisiting the Complexity of Food Intake. Cell Metab. 2015; 22 (3): 348–51.

[63] Woods SC, Langhans W. Inconsistencies in the assessment of food intake. Am. J. Physiol. Endocrinol. Metab. 2012; 303 (12): E1408-18.

[64] Marks JL, Porte Jr. D, Stahl WL, Baskin DG. Localization of insulin receptor mRNA in rat brain by in situ hybridization. Endocrinology 1990; 127 (0013–7227 (Print)): 3234–3236.

[65] Baskin DG, Porte Jr. D, Guest K, Dorsa DM. Regional concentrations of insulin in the rat brain. Endocrinology 1983; 112 (0013–7227 (Print)): 898–903.

[66] Milanski M, Arruda AP, Coope A, Ignacio-Souza LM, Nunez CE, Roman EA, et al. Inhibition of hypothalamic inflammation reverses diet-induced insulin resistance in the liver. Diabetes 2012; 61 (6): 1455–62.

[67] Valdearcos M, Robblee MM, Benjamin DI, Nomura DK, Xu AW, Koliwad SK. Microglia Dictate the Impact of Saturated Fat Consumption on Hypothalamic Inflammation and Neuronal Function. Cell Rep. 2014; 9 (6): 2124–2139.

[68] Ward KR, Bardgett JF, Wolfgang L, Stocker SD. Sympathetic response to insulin is mediated by melanocortin 3/4 receptors in the hypothalamic paraventricular nucleus. Hypertension 2011; 57 (1524–4563 (Electronic)): 435–441.

Chapter 6:

Summary and General Discussion

Insulin is well-established as an important effector of energy balance. It is secreted by the pancreas in direct proportion to adiposity and acts within the hypothalamus to reduce food intake and body weight. For this reason, it has been classified as an adiposity signal, which informs the brain about the abundance of stored energy. Because little-to-no insulin is produced by the brain, the majority of insulin’s effects in the CNS are dependent on the transport of insulin from the blood into the brain. However, the mechanisms influencing the transport of insulin through the BBB are still poorly understood.

6.1 CCK Increases Insulin Transport into the CNS

The satiation signal, CCK, is secreted from the small intestine in response to ingested nutrients and informs the brain about the relative amount and content of ingested calories. It was recently demonstrated that CCK increases the transport of leptin into the central nervous system (CNS) as measured in the CSF, suggesting that the effects of CCK may be partly mediated by increasing transport of adiposity signals through the BBB. In Chapter 2 (and Paper

II) we therefore tested the hypothesis that CCK increases the transport of insulin into the CNS.

Consistent with the hypothesis, we found that CCK-1 receptors and insulin receptors are co- expressed in endothelial cells of the BBB and that CCK increased the transport of insulin into the CNS. These findings are interesting, because they provide evidence in support of a novel mechanism by which CCK effects satiation. We expect that this interaction occurs through the process of CCK binding to CCK-1R expressed at the BBB in order to influence the transport of insulin through the BBB. Among the many possible mechanisms by which CCK increases insulin transport, this could involve 1) increased affinity of insulin receptors for binding insulin, 2) increased speed of vesicular trafficking through the endothelial cell, or 3) increased speed (or recycling) of insulin receptors back to the luminal side of the brain capillary, in order to bind and transport more insulin. Although insulin transport is still one of the most well-characterized

transport mechanisms into the CNS, such mechanisms still remain poorly understood and require further investigation.

6.2 Estrogen and Insulin Transport into the CNS

Similarly to CCK, E2 enhances the effects of catabolic signals in the brain and increases systemic insulin sensitivity. Although both insulin and E2, when given alone, reduce food intake and body weight via the hypothalamus, E2 surprisingly renders the hypothalamus insensitive to the catabolic effects of exogenous insulin. We therefore tested the hypothesis that E2 increases the levels of endogenous insulin in the CNS, helping create a maximal response, and thereby reduces the apparent effect of additional exogenous insulin. In Chapter 3 (and Paper

II), after determining that receptors for both E2 and insulin are co-expressed in BBB-endothelial cells, we administered E2 (acute or chronic) and assessed insulin transport into the CNS of ovariectomized (OVX) female rats and male rats fed chow or HFD. Although E2 was effective at reducing food intake, body weight, and improving insulin sensitivity, it had no effect on insulin transport into the CNS.

Although E2 does not affect the transport of insulin through the BBB, it is known to maintain BBB integrity in the face of inflammatory challenges and diseases [1–3] and thus may also promote barrier integrity during the metabolic syndrome. During the progression of the metabolic syndrome and obesity, brain microvessel structure can gradually become altered or damaged, and this can contribute to the co-morbidities of obesity [4; 5]. Under these conditions, the expression of tight junction proteins are reduced [6], which compromises BBB integrity and increases BBB permeability [5; 7]. Although BBB permeability is increased after the progression of obesity [5], this surprisingly does not correspond with increased entry of insulin or leptin into the CNS [8; 9]. Rather, the transport systems of the BBB appear to be impaired, leading to reduced transport of these signals into the CNS [8; 9].

In most cases, the BBB is quite effective at maintaining its integrity in the face of low- grade inflammation and metabolic stress, but as these challenges persist over time, this can result in compromised barrier integrity [5]. In support of this view, a recent longitudinal study found that women with greater mid-life adiposity had a higher rate of vascular disorders that affected BBB permeability 24 years later [10]. Having the metabolic syndrome earlier in life also predisposes to a risk of developing Alzheimer’s disease and other diseases, which can result from chronically impaired BBB function [11; 12]. These diseases become manifest over long periods of time, and may only initially reveal themselves as symptoms of progressive “brain fog” or fatigue [13]. Although E2 seems like a strong candidate for promoting BBB integrity, it is not always beneficial in every circumstance. For example, when reproductive senescent female

Sprague Dawley rats were provided supraphysiological levels of E2, this resulted in increased generalized BBB-permeability [1]. Therefore, the dose and the timing of E2 treatments should be taken into careful consideration when attempting to promote or restore BBB integrity.

6.3 Interactions of E2 With Insulin Signaling in the Hypothalamus

Our findings that E2 does not increase the levels of CSF insulin prompted the hypothesis that perhaps E2 increases insulin signaling in the hypothalamus. In Chapter 3 we found that although E2 did not increase the activation of insulin signaling molecules in the hypothalamus of rats maintained on a HFD, it did reduce the activation of IKKβ, an inflammatory signaling molecule, which is known to contribute to the metabolic syndrome [14]. Along these lines, it was recently found that female, but not male mice, are protected against hypothalamic inflammation and impaired cardiovascular function after chronic HFD feeding. ERα is reduced in the hypothalamus of males, but not females, leading to the further hypothesis that the protection of hypothalamic ERα during chronic HFD feeding is important in the protection of cardiovascular function. We therefore sought to determine the mechanism by which E2 protects hypothalamic ERα expression. Building on our initial findings, we observed that IKKβ is partly

responsible for the downregulation of hypothalamic ERα. The mechanism by which this regulation occurs remains to be delineated.

Based on these findings, it appears that palmitate may activate degradative processes that reduce ERα expression via pathways other than IKKβ. These intriguing possibilities remain to be tested. E2 may act through a variety of pathways to protect the expression of Pgc1α and

ERα, such as by reducing the levels of palmitate and other saturated fatty acids by directly increasing their oxidation in the mitochondria [15]. It was recently reported that E2 prevents the hypothalamic accumulation of pro-inflammatory ceramides [16], a class of membrane lipids derived from palmitate that are known to activate the IKKβ/NF-κB inflammatory signaling pathway [17] and to increase body weight via the hypothalamus [18]. The inhibition of IKKβ signaling also prevents hypothalamic ceramide accumulation during HFD-induced obesity [17].

Pharmacological inhibition of IKKβ in mice prevents ceramide accumulation in the hypothalamus during maintenance on a HFD [17]. Lastly, E2 can attenuate ER stress [19], which can be evoked during HFD feeding [14]. It is likely that E2 affects a host of these processes in order to exert its anti-inflammatory effects.

However, if IKKβ could be selectively targeted, it would likely be beneficial. In the case of obesity predisposing to developing a risk of cardiovascular disease, a meaningful reduction in hypothalamic inflammation and body weight would be expected to reduce the risk of cardiovascular disease [20; 21]. E2-related treatments that may prove efficacious in men are stable E2 conjugates that promote targeted delivery of E2 primarily to the desired regions of the hypothalamus and other tissues [22] or by utilizing pathway-selective estrogen receptor agonists, which inhibit inflammatory pathways without affecting other pathways that promote the secondary sex characteristics of females [23]. Such treatments would also avoid the risk factors of currently available estrogen replacement therapies, including reproductive endocrine toxicity and carcinogenesis [22].

6.4 Insulin and Hypothalamic Inflammation

After finding that insulin increases inflammation in cultured hypothalamic neurons, we became intrigued at the possibility that insulin may mediate its catabolic effects via inflammatory signaling in the hypothalamus. However, we found that at doses that reduce food intake, insulin does not exert inflammatory effects in the hypothalamus. Instead, we found evidence that insulin shares similar anti-inflammatory properties with E2. Insulin ameliorates sepsis, hypothalamic inflammation, and mortality following treatment with endotoxin and it also attenuates microglial activation in Alzheimer’s disease [24–27]. We considered whether insulin could be effective in preventing diet-induced inflammation in a similar way. However, in order to serve as a viable treatment, insulin cannot be repeatedly administered systemically, due to issues of weight gain and hypoglycemia [28; 29]. Central administration of insulin exerts a net catabolic effect without these undesirable side-effects, and increasing the levels of insulin in the

CNS via intranasal administration has proven to be effective at reducing adiposity in clinical studies [30].

In our preliminary studies into this area, we found that when insulin is administered via the lateral ventricle, it is ineffective at attenuating a hypothalamic inflammatory challenge.

Although this experiment was underpowered due to a low sample size, we nonetheless are able to make the tentative conclusion that insulin is unlikely to be helpful in reducing diet-induced hypothalamic inflammation. Future studies should utilize approaches for treatments that are applicable for humans, such as a single bolus dose of central insulin or repeated administration into the CNS no more than the recommended limit for humans. It is still possible that a lower dose of insulin could provide an anti-inflammatory effect, so this possibility cannot be ruled out.

In summary, insulin and estrogen exert their complex effects at multiple interaction sites that influence food intake and body weight, as well as inflammation. Sorting out these actions will likely to be efficacious in developing therapies.

6.5 Conclusions

Collectively, the studies in this dissertation demonstrate that insulin transport through the

BBB can be regulated in unique ways; and consistent with this, we have made several important findings that could contribute to novel antiobesity therapeutics and/or drug delivery techniques through the BBB. Further, hypothalamic inflammation is emerging as an important therapeutic target, and we found it to be partly responsible for a reduction in ERα signaling. Lastly, although insulin is a promising anti-inflammatory molecule, it does not appear to be well-suited for preventing diet-induced inflammation.

With the dire projections of the obesity epidemic in coming years, it is imperative that we meet the ever-growing demand for safe and effective therapeutics. The BBB is an ideal target for this purpose, due to its important role in transporting endogenous catabolic signals into the

CNS. Identifying the mechanisms by which the transport of catabolic molecules is influenced could provide great insights into the reduction of hypothalamic inflammation and the amelioration of obesity. In addition, it is hopeful that dieting and lifestyle strategies will also be improved, in order identify personalized therapies that are most effective for the long-term adherence for each individual [31–33]. Even if drugs are dramatically improved, they should be used in tandem with personalized dieting and weight loss strategies in order to ensure that weight loss is sustainable and that medications do not need to be continually consumed for the lifetime of the individual [34; 35]. It is unlikely that we will find any miraculous quick-fix strategies for obesity in the near future that require no personal effort on the part of the recipient.

In conclusion, the studies herein provide substantial contributions to our understanding of the ways by which hormones influence energy balance by impacting hypothalamic inflammation and insulin transport through the BBB. Our findings further support hypothesis

that increasing the transport of insulin and other catabolic molecules through the BBB will be beneficial in ameliorating obesity and its co-morbidities.

REFERENCES

[1] Bake S, Sohrabji F. 17beta-estradiol differentially regulates blood-brain barrier permeability in young and aging female rats. Endocrinology 2004; 145 (12): 5471–5.

[2] Liu R, Wen Y, Perez E, Wang X, Day AL, Simpkins JW, et al. 17β-Estradiol attenuates blood-brain barrier disruption induced by cerebral ischemia-reperfusion injury in female rats. Brain Res. 2005; 1060 (1–2): 55–61.

[3] Maggioli E, McArthur S, Mauro C, Kieswich J, Kusters DHM, Reutelingsperger CPM, et al. Estrogen protects the blood–brain barrier from inflammation-induced disruption and increased lymphocyte trafficking. Brain. Behav. Immun. 2016; 51: 212–222.

[4] Hsuchou H, Kastin AJ, Mishra PK, Pan W. C-reactive protein increases BBB permeability: implications for obesity and neuroinflammation. Cell. Physiol. Biochem. 2012; 30 (5): 1109–19.

[5] Hsu TM, Kanoski SE. Blood-brain barrier disruption: mechanistic links between Western diet consumption and dementia. Front. Aging Neurosci. 2014; 6: 88.

[6] Freeman LR, Granholm A-CE. Vascular changes in rat hippocampus following a high saturated fat and cholesterol diet. J. Cereb. Blood Flow Metab. 2012; 32 (4): 643–53.

[7] Kanoski SE, Zhang Y, Zheng W, Davidson TL. The effects of a high-energy diet on hippocampal function and blood-brain barrier integrity in the rat. J. Alzheimers. Dis. 2010; 21 (1): 207–19.

[8] Begg DP, Mul JD, Liu M, Reedy BM, D’Alessio DA, Seeley RJ, et al. Reversal of diet- induced obesity increases insulin transport into cerebrospinal fluid and restores sensitivity to the anorexic action of central insulin in male rats. Endocrinology 2013; 154 (3): 1047– 54.

[9] Banks WA, DiPalma CR, Farrell CL. Impaired transport of leptin across the blood-brain barrier in obesity. Peptides 1999; 20 (11): 1341–1345.

[10] Gustafson DR, Karlsson C, Skoog I, Rosengren L, Lissner L, Blennow K. Mid-life adiposity factors relate to blood–brain barrier integrity in late life. 2007; 262 (6):

[11] Cholerton B, Baker LD, Craft S. Insulin resistance and pathological brain ageing. Diabet. Med. 2011; 28 (12): 1463–75.

[12] Berzin TM, Zipser BD, Rafii MS, Kuo-Leblanc V, Yancopoulos GD, Glass DJ, et al. Agrin and microvascular damage in Alzheimer’s disease. Neurobiol.Aging 2000; 21 (0197– 4580 (Print)): 349–355.

[13] Theoharides TC, Stewart JM, Hatziagelaki E. Brain “fog,” inflammation and obesity: Key aspects of neuropsychiatric disorders improved by luteolin. Front. Neurosci. 2015; 9 (JUN): 225.

[14] Zhang X, Zhang G, Zhang H, Karin M, Bai H, Cai D. Hypothalamic IKKbeta/NF-kappaB and ER stress link overnutrition to energy imbalance and obesity. Cell 2008; 135 (1): 61– 73.

[15] McFadden JW, Aja S, Li Q, Bandaru VVR, Kim E-K, Haughey NJ, et al. Increasing fatty

acid oxidation remodels the hypothalamic neurometabolome to mitigate stress and inflammation. PLoS One 2014; 9 (12): e115642.

[16] Morselli E, Fuente-Martin E, Finan B, Kim M, Frank A, Garcia-Caceres C, et al. Hypothalamic PGC-1α protects against high-fat diet exposure by regulating ERα. Cell Rep. 2014; 9 (2): 633–45.

[17] Holland WL, Bikman BT, Wang L-P, Yuguang G, Sargent KM, Bulchand S, et al. Lipid- induced insulin resistance mediated by the proinflammatory receptor TLR4 requires saturated fatty acid-induced ceramide biosynthesis in mice. J. Clin. Invest. 2011; 121 (5): 1858–70.

[18] Contreras C, González-García I, Martínez-Sánchez N, Seoane-Collazo P, Jacas J, Morgan DA, et al. Central ceramide-induced hypothalamic lipotoxicity and ER stress regulate energy balance. Cell Rep. 2014; 9 (1): 366–77.

[19] Andruska ND, Zheng X, Yang X, Mao C, Cherian MM, Mahapatra L, et al. Estrogen receptor α inhibitor activates the unfolded protein response, blocks protein synthesis, and induces tumor regression. Proc. Natl. Acad. Sci. U. S. A. 2015; 112 (15): 4737–42.

[20] Eckel RH, Jakicic JM, Ard JD, de Jesus JM, Houston Miller N, Hubbard VS, et al. 2013 AHA/ACC guideline on lifestyle management to reduce cardiovascular risk: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation 2014; 129 (25 Suppl 2): S76-99.

[21] Purkayastha S, Zhang G, Cai D. Uncoupling the mechanisms of obesity and hypertension by targeting hypothalamic IKK-β and NF-κB. Nat. Med. 2011; 17 (7): 883–7.

[22] Finan B, Yang B, Ottaway N, Stemmer K, Müller TD, Yi C-X, et al. Targeted estrogen delivery reverses the metabolic syndrome. Nat. Med. 2012; 18 (12): 1847–56.

[23] Chadwick CC, Chippari S, Matelan E, Borges-Marcucci L, Eckert AM, Keith JC, et al. Identification of pathway-selective estrogen receptor ligands that inhibit NF- B transcriptional activity. Proc. Natl. Acad. Sci. 2005; 102 (7): 2543–2548.

[24] Chen Q, Yu W, Shi J, Shen J, Gao T, Zhang J, et al. Insulin alleviates the inflammatory response and oxidative stress injury in cerebral tissues in septic rats. J. Inflamm. (Lond). 2014; 11 (1): 18.

[25] Duan K, Yu W, Lin Z, Tan S, Bai X, Gao T, et al. Insulin ameliorating endotoxaemia- induced muscle wasting is associated with the alteration of hypothalamic neuropeptides and inflammation in rats. Clin. Endocrinol. (Oxf). 2015; 82 (5): 695–703.

[26] Milsom T. Central insulin signaling as a regulator of systemic inflammation. 2014;

[27] Chen Y, Zhao Y, Dai C-L, Liang Z, Run X, Iqbal K, et al. Intranasal insulin restores insulin signaling, increases synaptic proteins, and reduces Aβ level and microglia activation in the brains of 3xTg-AD mice. Exp. Neurol. 2014; 261: 610–9.

[28] The NICE-SUGAR Study. Intensive versus Conventional Glucose Control in Critically Ill Patients. N. Engl. J. Med. 2009; 360 (13): 1283–1297.

[29] Carver C. Insulin treatment and the problem of weight gain in type 2 diabetes. Diabetes Educ. 2006; 32 (6): 910–7.

[30] Hallschmid M, Benedict C, Schultes B, Fehm HL, Born J, Kern W. Intranasal insulin reduces body fat in men but not in women. Diabetes 2004; 53 (11): 3024–3029.

[31] Zeevi D, Korem T, Zmora N, Israeli D, Rothschild D, Weinberger A, et al. Personalized Nutrition by Prediction of Glycemic Responses. Cell 2015; 163 (5): 1079–94.

[32] Makris A, Foster GD. Dietary approaches to the treatment of obesity. Psychiatr. Clin. North Am. 2011; 34 (4): 813–27.

[33] Levy P. The current unmet need in type 2 diabetes mellitus: addressing glycemia and cardiovascular disease. Postgrad. Med. 2009; 121 (3 Suppl 1): 7–12.

[34] FDA. Postmarket Drug Safety Information for Patients and Providers - Orlistat (marketed as Alli and Xenical) Information. 2015;

[35] Chaput JP, Tremblay A. Current and novel approaches to the drug therapy of obesity. Eur.J.Clin.Pharmacol. 2006; 62 (0031–6970 (Print)): 793–803.

APPENDIX I

Table 1. Antibodies used for Western Blotting and/or Immunohistochemistry

Protein i.d. Item i.d. Company Actin-5γ MAB1501 Millipore

CCKAR sc-16172 Santa Cruz

pCREB 4377 Cell Signaling CREB 4034 Cell Signaling

ERα sc542 Santa Cruz

pFoxO1 9464 Cell Signaling FoxO1 2880 Cell Signaling

pIR/IGFR 3021 Cell Signaling

IR-β subunit sc-711 Santa Cruz

pIRS-1/2 sc-17195 Santa Cruz IRS-2 Cell Signaling

p-IKKβ 2697 Cell Signaling IKKβ 8943 Cell Signaling

pJNK 4668 Cell Signaling JNK 9252 Cell Signaling

p-NF-κB, p65 3033 Cell Signaling NF-κB, p65 3034 Cell Signaling

Table 2. TaqMan qPCR Assays used for cell culture and hypothalamic inflammation TaqMan assay

qPCR Target i.d. Actb Rn00667869_m1

Aif1 Rn00574125_g1 ERα Mm01208835_m1 Emr1 Rn01527631_m1 Gfap Rn01253033_m1 IKKβ Rn00584379_m1 IL-1β Mm00434228_m1 IL-6 Mm00446190_m1 IL-10 Rn99999012_m1 PGC-1α Mm01208835_m1 TNF-α Mm00443260_m1

APPENDIX II:

Paper I: Cholecystokinin increases insulin transport into the brain Physiology & Behavior 165 (2016) 392–397

Contents lists available at ScienceDirect

Physiology & Behavior

journal homepage: www.elsevier.com/locate/phb

Brief communication CCK increases the transport of insulin into the brain

Aaron A. May a,MinLiua, Stephen C. Woods b,⁎, Denovan P. Begg c a University of Cincinnati College of Medicine, Department of Pathology and Molecular Medicine, Metabolic Diseases Institute, OH, USA b University of Cincinnati College of Medicine, Department of Psychiatry and Behavioral Neuroscience, Metabolic Diseases Institute, OH, USA c University of New South Wales, School of Psychology, Sydney, NSW 2052, Australia

HIGHLIGHTS

• Cholecystokinin (CCK) increases the transport of insulin into the CNS of rats. • CCK-1 receptors are expressed by blood-brain barrier (BBB)-endothelial cells. • CCK may promote satiation by enhancing insulin transport through the BBB.

article info abstract

Article history: Food intake occurs in bouts or meals, and numerous meal-generated signals have been identified that act to limit Received 14 July 2016 the size of ongoing meals. Hormones such as cholecystokinin (CCK) are secreted from the intestine as ingested Received in revised form 24 August 2016 food is being processed, and in addition to aiding the digestive process, they provide a signal to the brain that con- Accepted 24 August 2016 tributes to satiation, limiting the size of the meal. The potency of CCK to elicit satiation is enhanced by elevated Available online 26 August 2016 levels of adiposity signals such as insulin. In the present experiments we asked whether CCK and insulin interact at the level of the blood-brain barrier (BBB). We first isolated rat brain capillary endothelial cells that comprise Keywords: Blood-brain barrier the BBB and found that they express the mRNA for both the CCK1R and the insulin receptor, providing a basis Cholecystokinin receptor for a possible interaction. We then administered insulin intraperitoneally to another group of rats and 15 min CNS insulin transport later administered CCK-8 intraperitoneally to half of those rats. After another 15 min, CSF and blood samples Satiation were obtained and assayed for immunoreactive insulin. Plasma insulin was comparably elevated above baseline Endocrinology in both the CCK-8 and control groups, indicating that the CCK had no effect on circulating insulin levels given CCK-8 these parameters. In contrast, rats administered CCK had CSF-insulin levels that were more than twice as high as those of control rats. We conclude that circulating CCK greatly facilitates the transport of insulin into the brain, likely by acting directly at the BBB. These findings imply that in circumstances in which the plasma levels of both CCK and insulin are elevated, such as during and soon after meals, satiation is likely to be due, in part, to this newly-discovered synergy between CCK and insulin. © 2016 Elsevier Inc. All rights reserved.

1. Introduction information is complex and provides ample opportunity for synergy, interference or other interactions among the various signals. Energy intake is orchestrated by numerous signals including those Satiation signals arise in the gastrointestinal tract and are secreted that provide information about the quantity and composition of acutely in proportion to the quantity and quality of food being con- ingested food (satiation signals), those that report the amount of stored sumed; they signal to the brain neurally and hormonally and thereby energy (adiposity signals), and nutrients, themselves, that are present contribute to meal termination as ingested calories accumulate. Several in the circulation and extracellular ﬂuid. All of these signals, in turn, con- gastrointestinal peptides/hormones are included in this category in- verge on circuits in the brain where they are integrated with cognitive, cluding cholecystokinin (CCK), glucagon-like peptide-1 (GLP-1), mem- social, hedonic and other situational factors. The calculus of all of this bers of the bombesin family of peptides, peptide YY (PYY), amylin, apolipoprotein A-IV and many others [1–4]. The exogenous administration of any of these just prior to a meal thus provides a false signal to the brain, implying that more calories have been eaten than actually have been consumed, with the result that a smaller-than-normal meal is con- ⁎ Corresponding author at: Department of Psychiatry, University of Cincinnati, 2170 East Galbraith Road, Cincinnati, OH 45237, USA. sumed. CCK is the most-investigated satiation signal, and its isoforms, E-mail address: [email protected] (S.C. Woods). such as CCK-8, reduce meal size dose-dependently in experimental

http://dx.doi.org/10.1016/j.physbeh.2016.08.025 0031-9384/© 2016 Elsevier Inc. All rights reserved. A.A. May et al. / Physiology & Behavior 165 (2016) 392–397 393 animals [5–7]. Humans administered CCK-8 report feeling fuller or 2.2. Isolation of rat brain microvessels more satiated, and they eat a smaller meal than occurs in the control condition, without feeling ill [8,9]. The physiological relevance of CCK The brain capillary endothelial cells that comprise the BBB express during consumption of a normal meal is demonstrated by the observa- insulin receptors [13,33,34]. Because insulin is too large to passively dif- tion that animals eat larger-than-normal meals when administered an- fuse through the BBB, the generally-accepted model of insulin transport tagonists of the CCK-1 receptor, implying that endogenous CCK involves insulin binding its receptors at the BBB [30,33,35], with the normally contributes to limiting meal size [10,11]. bound insulin receptors then being endocytosed within the endothelial Adiposity signals, including insulin and leptin, are secreted in direct cells [35]. Following receptor-mediated transport to the abluminal side proportion to the amount of fat stored in the body [12,13]. Each of these (brain-side) of the capillary, insulin is then released intact into the CNS hormones is transported from the blood into the brain, and the admin- [35–37]. To determine the feasibility of an interaction of CCK and insulin istration of either directly into the brain causes animals and humans to at the level of the BBB, we first asked whether brain capillary cells ex- eat less; if insulin or leptin is administered chronically into the brain, press CCK-1 receptors. body weight is also reduced [14–16]. The physiological relevance of in- Brain microvessels were isolated as previously described [33].Rats sulin and leptin in this regard is demonstrated when their respective were quickly and deeply anesthetized with isoflurane and were signals in the brain are blocked by pharmacological or genetic means. sacrificed after rapid collection of the whole brain. After allowing the In both instances, individuals become hyperphagic and carry more fat brain to quickly cool to 4 °C in chilled M199 buffer (Gibco, #11150) on than controls [17–19]. ice for ~10–15 min, meninges were removed. Forebrains were isolated Satiation signals and adiposity signals are catabolic, eliciting a net and individually homogenized with 10 strokes in a Dounce homogeniz- decrease of bodily energy reserves through combinations of reduced en- er in ~4 mL M199 buffer. Each homogenate was then resuspended in a ergy intake and increased energy expenditure, favoring weight loss [20]. solution containing a final concentration of 20% dextran (40,000 M.W., An important question is whether or how these various catabolic signals Sigma, #31390). Samples were centrifuged at 3000 ×g for 10 min, interact to influence food intake. CCK secreted from the duodenum in yielding a pure pellet of microvessels at the bottom of the tube. To response to ingested food acts locally in the wall of the intestine to stim- isolate microvessels with capillary diameters within a size range of ulate CCK-1 receptors (CCK-1R) expressed on branches of the vagus 20–100 μm, pelleted vessels were isolated, resuspended in M199 buffer nerve that project to the nucleus of the solitary (NST) tract in the hind- with 1% BSA, and filtered through a 100-μm nylon mesh (Small Parts) brain [21–24]. The vagal CCK signal then activates a complex circuit that with flow-through then being collected in 20-μm nylon mesh forwards the satiating signal to several brain areas including the hypo- (Millipore). Microvessels that were retained in the filter were then thalamus. Severing the vagus nerve proximal to the duodenum, or se- rinsed several times with 1% BSA and briefly vortexed in M199 buffer lectively cutting vagal afferent fibers entering the hindbrain, greatly until detached. After centrifuging the microvessels for 10 min at reduces the ability of exogenous CCK to reduce meal size [24,25]. 1000 ×g and isolation of the pellets, qPCR and Western Blotting or We and others have found that when insulin is elevated in the immunohistochemistry were conducted. brain of animals prior to a meal, the ability of exogenous CCK to reduce meal size is dose-dependently increased [26,27]. Similarly, 2.3. Analysis of brain microvessels an increased leptin signal also renders animals more sensitive to satiation signals [28,29]. The implication is that there is normally a The relative purity of microvessel isolates was first verified with cooperative catabolic action caused by combinations of satiation qPCR (TaqMan) using a StepOne™ Plus device (Thermo Scientific), and adiposity signals. Thus, when an individual has gained weight/ by screening isolates for markers of microvessel mRNA and assessing fat, this consequently leads to increased levels of circulating insulin, the extent of contamination from glia and neurons, as described [33]. leptin and other adiposity signals that then enter and stimulate the β-actin was used as the reference housekeeping gene. Only samples brain. Homeostatic models of body weight regulation would suggest that were significantly enriched to a comparable extent with previous- that the individual should eat smaller meals in order to reduce ly-defined cutoffs [33] were used for Western blotting (Fig. 1). Relative weight to normal. Consistent with this, the elevated adiposity signal purity of the brain microvessel isolates was further assessed by visual in the brain renders the individual more sensitive to meal-generated inspection and by immunohistochemistry with a rat-BBB antibody, as satiation signals with the consequence that they do in fact consume described [33]. Briefly, aliquots of the fresh microvessel samples were smaller meals [12,30]. In theory, adiposity signals could interact directly spread onto glass slides. Samples were fixed in 4% paraformal- with satiation signals at any point along the circuit from the vagal dehyde/PBS for 10 min and then washed 5 times in PBS (30 min). afferent nerves to the NST to other brain areas. After permeabilization with 0.1% Triton in PBS (30 min), rinsing 3 While all of these examples imply that the interactions among dif- times in PBS (30 min), and blocking in 2% serum (30 min, 26 °C), ferent classes of catabolic signals occurs mainly in the brain, there are microvessels were incubated in primary antibody for 24 h at 4 °C. Anti- other, not mutually exclusive, possibilities as well. Cano et al. reported bodies used were targeted to the CCK-1R (Santa Cruz, sc-16172), insu- that exogenous CCK increases the transport of leptin into the CNS [31, lin-receptor- β (IR-β) subunit (Santa Cruz, sc-711) or the rat-BBB 32]. In the present series of experiments, we demonstrate that CCK antibody (abcam, #24764). After 5 washes with PBS, secondary fluores- also increases the transport of insulin into the CNS and we investigate cent antibodies (Cy-3, AlexaFluor® 594 or AlexaFluor® 488, Molecular the mechanism mediating this effect. Probes, Carlsbad, CA) were incubated with microvessel samples for 1 h in 1% BSA solution. Slides were then rinsed 5× in PBS and tamped to remove excess liquid prior to mounting with SlowFade® Gold re- 2. Materials and methods agent (Molecular Probes). After curing for 24 h at 26 °C, microvessels were examined with fluorescence microscopy (Olympus) (Fig. 1C–E). 2.1. Animals For verification of specific binding in immunohistochemistry, we processed microvessels side-by-side with the other samples, except that Adult male Long-Evans rats (12–14 wk. of age; Envigo, Indianapolis, no primary antibodies were added to the incubation solution. IN) were housed individually in tub cages and maintained on a 12/12-h Microvessels were then incubated with secondary antibodies and light/dark cycle. Animals were provided ad-libitum access to water and were visualized, as described (Fig. 1H–I). pelleted chow (LabDiet® # TD7012), except where otherwise specified. Microvessel lysates were analyzed via Western blotting (Fig. 1E–G) All protocols were approved by the University of Cincinnati Institutional using previously-validated antibodies for CCK-1R (Santa Cruz, Animal Care and Use Committee. sc-16172) and LS Biosciences (LS-C177096). The insulin-receptor- β 394 A.A. May et al. / Physiology & Behavior 165 (2016) 392–397

Fig. 1. CCK-1R is expressed in rat brain microvessels. CCK-1R immunoreactivity was detected in freshly-isolated brain microvessels via ﬂuorescence microscopy (A-B, orange/Alexa Fluor® 594, 40× magniﬁcation), imaged with phase contrast (A) and without phase contrast (B). Nuclei were stained with DAPI (blue). No immunoreactivity was observed in control IHC experiments that omitted primary antibody (C-D); images were taken with phase contrast (C) and without phase contrast (D), for comparison. Western blotting was performed with protein lysates from brain microvessel isolates (E) and choroid plexuses (F) using the Santa Cruz CCK-1R antibody (47–50 kDa) and using the LS Biosciences (~90 kDa) antibody in separate blots of brain microvessel isolates (G) and choroid plexuses (H). Insulin receptor protein expression was detected in brain microvessels (I) and vessels were co- immunoblotted with a rat brain microvessel antibody (J, alone and K, together). CCK-1R mRNA expression (Cck1r) was also detected in rat brain microvessel isolates (L) at levels comparable to that of the full leptin receptor transcript (Lepr). Other genes examined included the GLP-1 receptor (Glp1r) and insulin receptor (Insr). n = 6, ± SEM.

(IR-β) subunit antibody (Santa Cruz, sc-711) was validated in a previ- anesthetized with isoflurane (Isothesia, Henry Schein, Dublin, Ohio) ous publication [33]. After running SDS-PAGE with protein lysates pre- and positioned into a stereotaxic instrument, with the head maximally pared using T-PER extraction buffer (Pierce, #78510) and transferring ventroflexed. A modified 25-G needle, with a tip prepared at a 30° to an Immobilon PVDF membrane (Millipore), Western Blots were incu- angle [38], was connected to microrenethane tubing filled with sterile bated with primary antibody for 16–24 h at 4 °C. After rinsing and incu- saline and secured in an electrode holder. The scalp and neck were bating in horseradish peroxidase-conjugated goat anti-rabbit IgG shaved and sterilized. An incision was made to expose the atlanto-oc- antibody for 1 h at room temperature (1:5000–1:10,000, Dako cipital membrane. Exactly 30 min after the initial insulin injection, the #P0448) and rinsing again, membranes were incubated in Immobilon® needle was inserted into the cisternum magnum and CSF flow was ini- HRP substrate (Millipore Cat. #WBKLS) 5 min at room temperature and tiated by applying slight negative pressure with a 1-mL syringe attached blots were visualized with photolithography film (Denville Scientific, to the microrenathane tubing. CSF was then collected (approximately #E3018). Interestingly, the Santa Cruz CCK-1R antibody detected 200 μL CSF per rat) into a chilled 1.5-mL microcentrifuge tube within a bands of the correct size only for freshly-prepared blots and these 2-min period. Rats were then sacrificed, and blood was immediately bands were not apparent after gentle stripping and re-probing. collected in chilled EDTA-coated tubes. Prior to analysis, the CSF samples were screened for blood contamination via spectrophotometric in- 2.4. Assessing insulin transport into the CNS spection. Only pure samples with b0.001% blood contamination were analyzed. Insulin levels were then measured in plasma and pure CSF Rats were fasted overnight for 16 h. The next day, they were admin- samples using a previously-validated ultrasensitive rat insulin ELISA istered intraperitoneal (ip) insulin (0.3 U/kg, NPH, Novo Nordisk) or ve- kit (Crystal Chem, #90060, Downers Grove, IL) [13,39]. In order to de- hicle (saline) 30 min prior to cerebrospinal fluid (CSF) collection. termine the absorbance of the human insulin used for this study, we Precisely 15 min after the insulin or vehicle injection, half of the rats prepared a standard curve with insulin NPH; this was compared to the in each group received ip CCK-8 (10 μg/rat; Sigma) and the other half re- absorbance of the reference rat insulin standard, as described [13].As ceived sterile saline. 10 min after the second injection, rats were previously found, the absorbance of insulin NPH is typically 80% that A.A. May et al. / Physiology & Behavior 165 (2016) 392–397 395 of rat insulin with the low-range assay. An exact correction was not made in this instance, because we did not have basal CSF insulin readings from the same cohort of rats. If a correction were made, it would increase the reported concentrations of insulin in the plasma and CSF, and would only strengthen the observed effect in this instance [13].

2.5. Statistical analysis

All statistical analyses were conducted with GraphPad Prism 5 software using a Student's t-test for comparisons between groups. Signiﬁ- cance was pre-deﬁned as P ≤ 0.05.

Fig. 2. Acute CCK treatment increases insulin transport into the CNS. A) No difference in 3. Results plasma insulin was observed between groups treated with CCK (Ins + CCK) or saline (Ins), following an ip injection of 0.3 U/kg insulin in all rats. For reference, in comparably prepared rats injected with saline (0 U/kg insulin), the basal concentration 3.1. CCK-1R is expressed at the BBB of insulin in the plasma is 1.11 ± 0.27 ng/mL (not depicted). B) Rats treated with CCK had a significantly higher concentration of insulin in the CSF in comparison to rats CCK-1R protein expression was detected in microvessels via immu- administered saline. For comparison, in rats injected with saline (0 U/kg insulin), the nohistochemistry (Fig. 1A,B, 40× magnification). Fluorescence was spe- basal concentration of CSF insulin is 0.53 ± 0.07 ng/mL n = 6, P b 0.05, ± SEM. cific to the binding of primary antibody, since samples processed without primary antibody were devoid of nonspecific binding (Fig. 1C,D). Western blotting with the CCK-1R antibody revealed solitary 4. Discussion bands of the expected size for the truncated type 1 receptor (~47 kDa) fi [40–42] in protein lysates from brain microvessels (Fig. 1E) and from Collectively, our ndings indicate that the satiation signal, CCK-8, in- the choroid plexus (Fig. 1F), indicating that the immunoreactivity ob- creases the transport of insulin into the CNS and that these effects are served with fluorescence microscopy in Fig. 1A,B is solely from the trun- likely mediated by CCK-1Rs that are expressed throughout the BBB cated form of CCK-1R. For comparison, microvessel protein lysates were and blood-CSF barrier. These observations are in agreement with previ- blotted with the LS Biosciences CCK-1R antibody (which is solely vali- ous reports that CCK-8 increases the appearance of the adiposity signal, dated for Western blotting, and not immunohistochemistry). In this leptin, into the CSF of rats [31,32]. These intriguing observations suggest case, we detected a band at the ~90 kDa range in brain microvessels that CCK may also promote satiation by increasing the transport of other (Fig. 1G) and choroid plexus samples (Fig. 1H), which is the expected catabolic signals into the CNS (although CCK itself does not cross the size for the full-length CCK-1R [40–42]. Due to this discrepancy between BBB (43)) in addition to sending signals to the CNS via its action in the antibodies, we verified expression of Cck1r mRNA in freshly-isolated rat peripheral nervous system. fi brain microvessels via qPCR and found that its expression was compara- One implication of these ndings is that this cooperative process ble to that of the leptin receptor (Fig. 1L). The finding that CCK-1R is would be active during meals when endogenous CCK and insulin are expressed at the BBB is important because CCK-1R is the receptor in- both elevated; it would be anticipated to be greatly enhanced in animals volved in the satiating effects of CCK-8 (21−23). As we and others with higher and more prolonged elevations of GI hormones, such as can have seen before [33,34], insulin receptor protein was also expressed occur following successful bariatric surgeries [44]. In turn, increasing throughout microvessels (Fig. 1I) in a pattern that was comparable to the levels of these hormones is anticipated to lead to a pronounced sup- that of truncated CCK-1R; it appears that both CCK-1R and insulin re- pression of food intake and body weight, by increasing the levels of ad- ceptors are expressed in every endothelial cell. A rat-BBB-specificanti- iposity signals in the brain [45]. body was also used as validation of microvessel specificity in these Insulin was increased in the CSF within 15 min following ip CCK-8 studies (Fig. 1J,K) and the relative purity of microvessel isolates was ver- administration. If CCK exerts its effects directly at the BBB as proposed ified via qPCR using our previously-published approach [33].Thesefind- [31,32], then it must diffuse from the peritoneal cavity into the blood- ings provide an anatomical foundation for an interaction of CCK-8 with stream, circulate and bind to CCK-1R on endothelial cells of the BBB, insulin transport at the BBB. and consequently increase insulin's appearance in the CSF within a very short time-frame (as CCK-8 does not cross the BBB). This process involves the transport of insulin across the BBB and its subsequent re- 3.2. Acute CCK treatment increases insulin transport into the CNS lease into the brain interstitial fluid [36]. Each molecule of insulin may then bind to brain insulin receptors and/or may be degraded. The bulk Fasted rats receiving ip insulin 30 min earlier had comparable plas- flow of unbound insulin will enter the CSF, where it then is transported ma insulin whether they subsequently received ip CCK-8 or not, such back into the bloodstream [36]. Given the many processes that need to that CCK had no effect on circulating insulin (Fig. 2A). Nonetheless, ip occur in this short time frame, it is unlikely that significant increases in CCK resulted in CSF insulin levels being more than twice as high as oc- insulin receptor expression could have accounted for the increased CSF curred in rats given insulin but no CCK (Fig. 2B). A separate experiment insulin concentration. Instead, CCK may act to increase the rate of insu- was conducted to establish baseline insulin levels in a group of rats lin transport through the BBB by increasing the saturation threshold of injected ip with 0 U/kg insulin (saline, n = 6). Because this experiment the insulin receptors already present at the outer cell membranes of en- was conducted on a different day, these data are not incorporated into dothelial cells comprising the BBB. This could either involve increasing the graph. As listed in the description for Fig. 2, insulin levels in the plas- the sensitivity of insulin receptors or increasing the insulin receptor- ma were 1.11 ± 0.27 ng/mL and baseline concentrations of insulin in recycling rate within endothelial cells as they are transported from the the CSF were comparable to those of the group injected with 0.3 U/kg abluminal side to the luminal side of BBB-endothelial cells, prior to insulin. These values are comparable to previously-published results binding another insulin molecule [30]. Future experiments will have from our lab [13]. Taken together, these results indicate that the low to determine if this is the case. dose of ip insulin did not significantly increase CSF insulin, whereas An alternative explanation for our results is that CCK somehow pre- the administration of CCK significantly increased insulin transport vents the degradation of insulin (and presumably leptin, as well) within under these conditions. the brain. Since CCK does not cross the BBB [43] and is unlikely to affect 396 A.A. May et al. / Physiology & Behavior 165 (2016) 392–397 the activity of the insulin-degrading enzyme (IDE) therein, CCK would [16] R. Pal, A. Sahu, Leptin signaling in the hypothalamus during chronic central leptin infusion, Endocrinology [Internet] 144 (2003) 3789–3798 Available from: http:// have to substantially reduce the activity of IDE expressed at the BBB in press.endocrine.org/doi/pdf/10.1210/en.2002-0148. order for such an interaction to take place [46]. Further, plasma insulin [17] J.C. Bruning, D. Gautam, D.J. Burks, J. Gillette, M. Schubert, P.C. Orban, et al., Role of levels were unaffected by CCK treatment, indicating that CCK does not brain insulin receptor in control of body weight and reproduction, Science (80-.). [Internet] 289 (2000) 2122–2125 Available from: http://www.ncbi.nlm.nih.gov/ alter IDE activity acutely in other tissues that are important for insulin pubmed/11000114. degradation via IDE, such as the liver and kidney. Alternatively, CCK [18] C.A. Grillo, K.L. Tamashiro, G.G. Piroli, S. Melhorn, J.T. Gass, R.J. Newsom, et al., Len- might send signals to the CNS that acutely suppress IDE activity, but tivirus-mediated downregulation of hypothalamic insulin receptor expression, – such a mechanism has yet to be reported in the literature. Physiol. Behav. [Internet] 92 (2007 [cited 2016 Mar 28]) 691 701 Available from: http://www.sciencedirect.com/science/article/pii/S003193840700193X. In conclusion, our findings support a model in which CCK increases [19] N. Ottaway, P. Mahbod, B. Rivero, L.A. Norman, A. Gertler, D.A. D'Alessio, et al., Diet- the appearance of insulin into the CNS by acting on CCK-1Rs expressed induced obese mice retain endogenous leptin action, Cell Metab. [Internet] 21 (2015 – in the BBB, although the exact mechanism(s) remains to be demonstrat- [cited 2016 May 25]) 877 882 Available from: http://www.sciencedirect.com/sci- fi ence/article/pii/S1550413115001709. ed. These ndings point to a direct role of CCK at the BBB in the media- [20] S.C. Woods, R.J. Seeley, Adiposity signals and the control of energy homeostasis, Nu- tion of satiation. Whether other GI hormones can similarly exert these trition [Internet] 16 (2000) 894–902 Available from: http://www.ncbi.nlm.nih.gov/ effects at the BBB is not known [45,47]. Further studies are warranted pubmed/11054594. [21] R.L. Corwin, J. Gibbs, G.P. Smith, Increased food intake after type A but not type to determine the mechanism by which the transport of insulin and lep- B cholecystokinin receptor blockade, Physiol. Behav. [Internet] 50 (1991 [cited tin are increased following CCK-8 administration. 2016 Apr 26]) 255–258 Available from: http://www.ncbi.nlm.nih.gov/pubmed/ 1946726. [22] T.H. Moran, P.J. Ameglio, G.J. Schwartz, P.R. McHugh, Blockade of type A, not type B, CCK receptors attenuates satiety actions of exogenous and endogenous CCK, Am. J. Acknowledgements Physiol. [Internet] 262 (1992 [cited 2016 May 25]) R46–R50 Available from: http://www.ncbi.nlm.nih.gov/pubmed/1733339. This work was funded by the National Institute of Diabetes and Di- [23] T.H. Moran, P.J. Ameglio, H.J. Peyton, G.J. Schwartz, P.R. McHugh, Blockade of type A, but not type B, CCK receptors postpones satiety in rhesus monkeys, Am. J. Physiol. gestive and Kidney Diseases (NIDDK) awards DK017844, DK92779, [Internet] 265 (1993 [cited 2016 May 25]) R620–R624 Available from: http:// DK95440, and DK059803. www.ncbi.nlm.nih.gov/pubmed/8214156. [24] T.H. Moran, A.R. Baldessarini, C.F. Salorio, T. Lowery, G.J. Schwartz, Vagal afferent and efferent contributions to the inhibition of food intake by cholecystokinin, Am. J. Physiol [Internet] 272 (1997) R1245–R1251 Available from: http://www.ncbi.nlm. References nih.gov/pubmed/9140026. [25] G.J. Schwartz, T.H. Moran, CCK elicits and modulates vagal afferent activity arising [1] S.C. Woods, R.J. Seeley, D. Porte Jr., M.W. Schwartz, Signals that regulate food intake from gastric and duodenal sites, Ann. N.Y. Acad. Sci. [Internet] 713 (1994) and energy homeostasis, Science (80-.) 280 (1998) 1378–1383. 121–128 Available from: http://www.ncbi.nlm.nih.gov/pubmed/8185153. [2] P. Tso, W. Sun, M. Liu, Gastrointestinal satiety signals IV. Apolipoprotein A-IV, Am. J. [26] D.P. Figlewicz, A.J. Sipols, R.J. Seeley, M. Chavez, S.C. Woods, D. Porte Jr., Intraventric- Physiol Gastrointest. Liver Physiol. 286 (2004) G885–G890. ular insulin enhances the meal-suppressive efficacy of intraventricular cholecystoki- [3] D.L. Williams, D.G. Baskin, M.W. Schwartz, Evidence that intestinal glucagon-like nin octapeptide in the baboon, Behav. Neurosci. [Internet] 109 (1995) 567–569 peptide-1 plays a physiological role in satiety, Endocrinology 150 (2009) Available from: http://www.ncbi.nlm.nih.gov/pubmed/7662167. 1680–1687. [27] C.A. Riedy, M. Chavez, D.P. Figlewicz, S.C. Woods, Central insulin enhances sensitiv- [4] P. Tso, M. Liu, Ingested fat and satiety, Physiol. Behav. 81 (2004) 275–287. ity to cholecystokinin, Physiol. Behav. 58 (1995) 755–760. [5] J. Gibbs, R.C. Young, G.P. Smith, Cholecystokinin decreases food intake in rats, J. [28] D.L. Williams, D.G. Baskin, M.W. Schwartz, Hindbrain leptin receptor stimulation en- Comp. Physiol. Psychol. [Internet] 84 (1973 [cited 2016 Jan 30]) 488–495 Available hances the anorexic response to cholecystokinin, Am. J. Physiol. Regul. Integr. Comp from: http://www.ncbi.nlm.nih.gov/pubmed/4745816. Physiol [Internet] 297 (2009) R1238– R1246 Available from: http://www.ncbi.nlm. [6] J. Antin, J. Gibbs, J. Holt, R.C. Young, G.P. Smith, Cholecystokinin elicits the complete nih.gov/pubmed/19726710. behavioral sequence of satiety in rats, J. Comp. Physiol. Psychol. [Internet] 89 (1975 [29] M. Emond, G.J. Schwartz, E.E. Ladenheim, T.H. Moran, Central leptin modulates be- [cited 2016 May 25]) 784–790 Available from: http://www.ncbi.nlm.nih.gov/ havioral and neural responsivity to CCK, Am. J. Physiol [Internet] 276 (1999) pubmed/1176672. R1545–R1549 Available from: http://www.ncbi.nlm.nih.gov/pubmed/10233050. [7] G. Yoshimichi, C.C. Lo, K.L. Tamashiro, L. Ma, D.M. Lee, D.P. Begg, et al., Effect of pe- [30] D.P. Begg, S.C. Woods, The central insulin system and energy balance, ripheral administration of cholecystokinin on food intake in apolipoprotein AIV Handb.Exp.Pharmacol. [Internet] 209 (2012) 111–129 Available from: http:// knockout mice, Am. J. Physiol. Gastrointest. Liver Physiol. [Internet] 302 (2012) www.ncbi.nlm.nih.gov/pubmed/22249812. G1336–G1342 Available from: http://ajpgi.physiology.org/content/ajpgi/early/ [31] V. Cano, B. Merino, L. Ezquerra, B. Somoza, M. Ruiz-Gayo, A cholecystokinin-1 recep- 2012/03/26/ajpgi.00325.2010.full.pdf. tor agonist (CCK-8) mediates increased permeability of brain barriers to leptin, Br. J. [8] G. Stacher, H. Bauer, H. Steinringer, Cholecystokinin decreases appetite and activa- Pharmacol. [Internet] 154 (2008 [cited 2016 Feb 1]) 1009–1015 Available from: tion evoked by stimuli arising from the preparation of a meal in man, Physiol. http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2451056&tool= Behav. [Internet] 23 (1979 [cited 2016 May 25]) 325–331 Available from: http:// pmcentrez&rendertype=abstract. www.sciencedirect.com/science/article/pii/0031938479903743. [32] V. Cano, L. Ezquerra, M.P. Ramos, M. Ruiz-Gayo, Regulation of leptin distribution be- [9] X. Pi-Sunyer, H.R. Kissileff, J. Thornton, G.P. Smith, C-terminal octapeptide of chole- tween plasma and cerebrospinal fluid by cholecystokinin receptors, Br. J. Pharmacol. cystokinin decreases food intake in obese men, Physiol. Behav. [Internet] 29 (1982 [Internet] 140 (2003 [cited 2016 Feb 1]) 647–652 Available from: http://www. [cited 2016 Apr 28]) 627–630 Available from: http://www.ncbi.nlm.nih.gov/ pubmedcentral.nih.gov/articlerender.fcgi?artid=1574067&tool= pubmed/6294699. pmcentrez&rendertype=abstract. [10] D.P. Yox, L. Brenner, R.C. Ritter, CCK-receptor antagonists attenuate suppression of [33] A.A. May, N.D. Bedel, L. Shen, S.C. Woods, M. Liu, Estrogen and insulin trans- sham feeding by intestinal nutrients, Am. J. Physiol. [Internet] 262 (1992 [cited port through the blood-brain barrier, Physiol. Behav. [Internet] 163 (2016) 2016 May 25]) R554–R561 Available from: http://www.ncbi.nlm.nih.gov/pubmed/ 312–321. 1373576. [34] M.W. Schwartz, D.F. Figlewicz, S.E. Kahn, D.G. Baskin, M.R.C. Greenwood, D. Porte, [11] L. Brenner, R.C. Ritter, Peptide cholesystokinin receptor antagonist increases food in- Insulin binding to brain capillaries is reduced in genetically obese, hyperinsulinemic take in rats, Appetite [Internet] 24 (1995 [cited 2016 May 25]) 1–9 Available from: Zucker rats, Peptides [Internet] 11 (1990 [cited 2013 Nov 19]) 467–472 Available http://www.ncbi.nlm.nih.gov/pubmed/7537951. from: http://www.sciencedirect.com/science/article/pii/0196978190900446. [12] M.W. Schwartz, S.C. Woods, D. Porte Jr., R.J. Seeley, D.G. Baskin, Central nervous sys- [35] K.R. Duffy, W.M. Pardridge, Blood-brain barrier transcytosis of insulin in developing tem control of food intake, Nature 404 (6778) (2000 Apr 6) 661–671 [Internet]. rabbits [Internet], 420 (1987 [cited 2013 Oct 22]) 32–38 Available from: http:// 2000;404:661–671. Available from: http://www.nature.com/nature/journal/v404/ www.sciencedirect.com/science/article/pii/0006899387902368. n6778/full/404661a0.html. [36] M.W. Schwartz, R.N. Bergman, S.E. Kahn, G.J. Taborsky, L.D. Fisher, A.J. Sipols, et al., [13] D.P. Begg, A.A. May, J.D. Mul, M. Liu, D.A. D'Alessio, R.J. Seeley, et al., Insulin detemir Evidence for entry of plasma insulin into cerebrospinal fluid through an intermedi- is transported from blood to cerebrospinal fluid and has prolonged central anorectic ate compartment in dogs. Quantitative aspects and implications for transport, J. Clin. action relative to NPH insulin, Diabetes [Internet] 64 (2015 [cited 2015 Dec 2]) Invest. [Internet] 88 (1991 [cited 2016 Apr 6]) 1272–1281 Available from: http:// 2457–2466 Available from: http://www.ncbi.nlm.nih.gov/pubmed/25667307. www.pubmedcentral.nih.gov/articlerender.fcgi?artid=295596&tool= [14] S.C. Woods, E.C. Lotter, L.D. McKay, D. Porte, D. Porte Jr., Chronic intracerebroventric- pmcentrez&rendertype=abstract. ular infusion of insulin reduces food intake and body weight of baboons, Nature [In- [37] G.D. Baura, D.M. Foster, D. Porte Jr., S.E. Kahn, R.N. Bergman, C. Cobelli, et al., Satura- ternet] 282 (1979 [cited 2016 Jul 27]) 503–505 Available from: http://www.nature. ble transport of insulin from plasma into the central nervous system of dogs in vivo. com/nature/journal/v282/n5738/abs/282503a0.html. A mechanism for regulated insulin delivery to the brain, J. Clin. Invest. 92 (1993) [15] M. Chavez, K. Kaiyala, L.J. Madden, M.W. Schwartz, S.C. Woods, Intraventricular in- 1824–1830. sulin and the level of maintained body weight in rats, Behav. Neurosci [Internet] [38] M. Liu, L. Shen, D.P. Begg, D.A. D'alessio, S.C. Woods, Insulin increases central apoli- 109 (1995 [cited 2014 Jan 25]) 528–531 Available from: http://psycnet.apa.org/ poprotein E levels as revealed by an improved technique for collection of cerebro- journals/bne/109/3/528.html. spinal fluid from rats, J. Neurosci. Methods [Internet] 209 (2012 [cited 2014 Jun A.A. May et al. / Physiology & Behavior 165 (2016) 392–397 397

9]) 106–112 Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi? [43] B.A. Baldwin, R.F. Parrott, I.S. Ebenezer, Food for thought: a critique on the hypoth- artid=3402581&tool=pmcentrez&rendertype=abstract. esis that endogenous cholecystokinin acts as a physiological satiety factor, Prog. [39] L. Shen, M. Haas, D.Q.-H. Wang, A. May, C.C. Lo, S. Obici, et al., Ginsenoside Rb1 in- Neurobiol. [Internet] 55 (1998 [cited 2016 Feb 2]) 477–507 Available from: http:// creases insulin sensitivity by activating AMP-activated protein kinase in male rats, www.ncbi.nlm.nih.gov/pubmed/9670215. Physiol. Rep. [Internet] 3 (2015 [cited 2016 Jan 28]) Available from: http://www. [44] C.N. Ochner, C. Gibson, M. Shanik, V. Goel, A. Geliebter, Changes in neurohormonal pubmedcentral.nih.gov/articlerender.fcgi?artid=4600387&tool= gut peptides following bariatric surgery, Int. J. Obes. (Lond). [Internet] 35 (2011 pmcentrez&rendertype=abstract. [cited 2016 May 31]) 153–166 Available from: http://www.pubmedcentral.nih. [40] S. Poirot, C. Escrieut, M. Dufresne, J. Martinez, M. Bouisson, N. Vaysse, et al., gov/articlerender.fcgi?artid=3632050&tool=pmcentrez&rendertype=abstract. Photoafﬁnity labeling of rat pancreatic cholecystokinin type A receptor antagonist [45] S.C. Woods, Gastrointestinal satiety signals I. An overview of gastrointestinal signals binding sites demonstrates the presence of a truncated cholecystokinin type A re- that inﬂuence food intake, Am. J. Physiol.Gastrointest. Liver Physiol. [Internet] 286 ceptor, Mol. Pharmacol. [Internet] 45 (1994 [cited 2016 May 31]) 599–607 Available (2004) G7–13 Available from: http://www.ncbi.nlm.nih.gov/pubmed/14665437. from: http://molpharm.aspetjournals.org/content/45/4/599. [46] W. Gao, P.B. Eisenhauer, K. Conn, J.A. Lynch, J.M. Wells, M.D. Ullman, et al., Insulin [41] B. Schjoldager, X. Molero, L.J. Miller, Gallbladder CCK receptors: species differences degrading enzyme is expressed in the human cerebrovascular endothelium and in in glycosylation of similar protein cores, Regul. Pept. [Internet] 28 (1990 [cited cultured human cerebrovascular endothelial cells, Neurosci. Lett. [Internet] 371 2016 May 31]) 265–272 Available from: http://www.sciencedirect.com/science/ar- (2004 [cited 2016 Feb 26]) 6–11 Available from: http://www.sciencedirect.com/sci- ticle/pii/016701159090024Q. ence/article/pii/S0304394004008766. [42] A. Deweerth, J.R. Pisegna, K. Huppi, S.A. Wank, Molecular cloning, functional expres- [47] S.C. Woods, The control of food intake: behavioral versus molecular perspectives, sion and chromosomal localization of the human cholecystokinin type a receptor, Cell Metab. [Internet] 9 (2009) 489–498 Available from: http://www.ncbi.nlm.nih. Biochem. Biophys. Res. Commun. [Internet] 194 (1993 [cited 2016 May 31]) gov/pubmed/19490904. 811–818 Available from: http://www.scopus.com/inward/record.url?eid=2-s2.0- 0027234316&partnerID=tZOtx3y1.

APPENDIX III:

Paper III: Estrogen and insulin transport through the blood-brain barrier Physiology & Behavior 163 (2016) 312–321

Contents lists available at ScienceDirect

Physiology & Behavior

journal homepage: www.elsevier.com/locate/phb

Estrogen and insulin transport through the blood-brain barrier

Aaron A. May a, Nicholas D. Bedel a,LingShena, Stephen C. Woods b, Min Liu a,⁎ a Department of Pathology and Molecular Medicine, Metabolic Diseases Institute, University of Cincinnati College of Medicine, OH, USA b Department of Psychiatry and Behavioral Neuroscience, Metabolic Diseases Institute, University of Cincinnati College of Medicine, OH, USA

HIGHLIGHTS

• Acute estrogen (E2) does not alter insulin transport into the CNS in rats. • Chronic E2 can both prevent and reverse diet-induced obesity (DIO) in rats. • Chronic E2 does not increase insulin transport into the CNS. • Food restriction reduces DIO and improves central insulin transport.

article info abstract

Article history: Obesity is associated with insulin resistance and reduced transport of insulin through the blood-brain barrier Received 10 February 2016 (BBB). Reversal of high-fat diet-induced obesity (HFD-DIO) by dietary intervention improves the transport of in- Received in revised form 11 May 2016 sulin through the BBB and the sensitivity of insulin in the brain. Although both insulin and estrogen (E2), when Accepted 12 May 2016 given alone, reduce food intake and body weight via the brain, E2 actually renders the brain relatively insensitive Available online 13 May 2016 to insulin's catabolic action. The objective of these studies was to determine if E2 inﬂuences the ability of insulin to be transported into the brain, since the receptors for both E2 and insulin are found in BBB endothelial cells. E2 Keywords: Blood-brain barrier (acute or chronic) was systemically administered to ovariectomized (OVX) female rats and male rats fed a chow Metabolic syndrome or a high-fat diet. Food intake, body weight and other metabolic parameters were assessed along with insulin 17β-Estradiol entry into the cerebrospinal ﬂuid (CSF). Acute E2 treatment in OVX female and male rats reduced body weight CNS insulin transport and food intake, and chronic E2 treatment prevented or partially reversed high-fat diet-induced obesity. Howev- Insulin resistance er, none of these conditions increased insulin transport into the CNS; rather, chronic E2 treatment was associated Diet-induced obesity less-effective insulin transport into the CNS relative to weight-matched controls. Thus, the reduction of brain insulin sensitivity by E2 is unlikely to be mediated by increasing the amount of insulin entering the CNS. © 2016 Elsevier Inc. All rights reserved.

1. Introduction compromising the function of insulin receptors locally within the mediobasal hypothalamus [3,8], results in hyperphagia and obesity. In The International Diabetes Federation has estimated that 25% of contrast, mice with a knockout of insulin receptors in all tissues other adults worldwide currently have the metabolic syndrome [1]. The met- than the brain do not have altered body weight, despite the presence abolic syndrome is frequently associated with insulin resistance, a syn- of pronounced hyperglycemia [9]. drome in which the body's responsiveness to the pancreatic hormone, In order to enter the brain, insulin must cross the blood-brain barrier insulin, is reduced. Insulin's function in the CNS is of particular impor- via a selective receptor-mediated transport process. Insulin transport tance to energy homeostasis, since it suppresses hepatic glucose pro- into the CNS of rats is impaired in high-fat diet-induced obesity (HFD- duction [2,3], promotes glucose uptake [3] and reduces food intake via DIO), but can be fully restored by dietary intervention with a low-fat the hypothalamus and other brain regions [4–6]. Intranasal insulin, diet (LFD) [10]. This observation has also been implicated in obese which directly enters the CNS, produces clinically-meaningful weight humans; the proportion of insulin in the brain relative to the blood is loss in lean men, but not in obese men with insulin resistance [6,7]. lower in individuals with greater visceral adiposity and insulin resis- Insulin must enter the brain in order to reduce food intake and body tance [11,12]. Therefore, insulin transport through the BBB and its sub- weight. Genetic deletion of insulin receptors in the whole brain [5],or sequent action in the brain are both considered important for maintaining energy balance [13]. Therapeutics that increase the sensitivity of insulin or other catabolic ⁎ Corresponding author at: Department of Pathology, University of Cincinnati, 2180 East Galbraith Road, Cincinnati, OH 45237, USA. signals in the brain could be used to ameliorate the metabolic syndrome E-mail address: [email protected] (M. Liu). and avoid the associated risks of type-II diabetes, cardiovascular disease,

http://dx.doi.org/10.1016/j.physbeh.2016.05.019 0031-9384/© 2016 Elsevier Inc. All rights reserved. A.A. May et al. / Physiology & Behavior 163 (2016) 312–321 313 and other co-morbidities [14,15]. Estradiol (E2) is a strong therapeutic 0200 h). Rats were maintained on standard rat chow (Purina, St. Louis, candidate in this regard, since it promotes systemic insulin sensitivity MO) or were provided a 40% high-fat diet (HFD), as indicated, which [16,17] and enhances the strength of other signals that reduce food in- contains butter fat as the primary lipid (Research Diets, #D03082705). take and body weight, including leptin [18], cholecystokinin [19,20], Animals were provided ad libitum access to food and water, except and apolipoprotein-A-IV [21]. In clinical and rodent models of surgical where indicated. Following two weeks of acclimation in the facility, fe- menopause, E2 has also been found to improve insulin sensitivity in males were ovariectomized [37] and allowed to recover for two addi- humans and rodents, as assessed using a hyperinsulinemic-euglycemic tional weeks prior to the study. All protocols were followed in clamp [22–24] and prevents visceral adiposity [25,26]. Furthermore, a accordance to the guidelines approved by the University of Cincinnati recently-developed conjugate of E2 and glucagon-like peptide-1 (GLP- Institutional Animal Care and Use Committee (IACUC). 1) has been demonstrated to reverse the metabolic syndrome in both Humans with a body mass index (BMI) of greater than or equal to 30 male and ovariectomized (OVX) female mice, while avoiding the unde- are classified as obese by the World Health Organization. However, sirable side-effects of estrogen treatment, such as oncogenicity and re- there is no standard definition of “obesity” in rodents [38],sincethepro- productive endocrine toxicity [27,28]. Considering these benefits, it is gression of obesity is highly dependent on the strain being studied [39]. surprising that E2 reduces the apparent sensitivity of insulin that has Our lab has previously characterized diet-induced obesity in Long- been administered into the brain of rats, despite increasing the sensitiv- Evans rats using the same HFD that was selected for the current study, ityofleptininthebrainunderthesameconditions[18].Consistentwith which defined “obesity” as a 10% or greater difference in body weight this, human studies have also confirmed that relative to females, males relative to controls fed a LFD [40]. Therefore, we continued treatments are more sensitive to insulin as an anorectic signal when administered with E2 and vehicle as long as necessary, in order to achieve this weight so as to act within the brain [6,29]. This raises a paradox: why does E2 difference, which took approximately 5 weeks. Similar to human obesi- improve peripheral insulin sensitivity in humans and rodents, but act ty, our weight-based definition of obesity approximates the BMI and oppositely in the brain [16,30]? does not account for the relative proportion of fat mass, lean mass, or Although E2 acts through several signaling pathways that also play metabolic parameters. an active role in insulin signaling (e.g., PI3K/Akt and Erk), it is unlikely that E2 out-competes insulin for activation of these common pathways. 2.2. Rat brain microvessel isolation and analysis Indeed, E2 cooperates and/or synergizes with leptin and several satiation signals to reduce food intake and body weight via shared signaling Brain microvessels (comprising the BBB) were isolated from rat pathways (e.g., PI3K/Akt) in the hypothalamus and nucleus tractus brains using an approach based on existing protocols [41,42].Afterrap- solitarius (NTS) [31]. Additionally, E2 does not reduce the expression idly anesthetizing the rats with isoflurane, brains were harvested as pre- or sensitivity of insulin receptors in the brain of rodents, making it un- viously described [37] and immediately cooled to 4 °C in M199 medium likely to reduce insulin signaling [18,32,33]. on ice for 15 min. Meninges were removed and the forebrains were then Insulin acts in a dose-dependent manner in the CNS; the greater the gently homogenized with a Dounce homogenizer (Bellco, Vineland, NJ). administered dose of insulin directly into the CNS, the larger the sup- Microvessels were pelleted in a 20% Dextran solution (Sigma) via cen- pression of food intake and body weight, up to the point of receptor sat- trifugation at 4 °C (15 min at 2500 ×g). After careful removal of super- uration [34]. E2 may therefore interact with the transport of insulin into natant, microvessel pellets were resuspended in chilled M199 medium the brain. An E2-elicited enhancement of insulin transport in females and filtered through 100-micron nylon mesh (Small Parts, Inc. #7050- might increase brain insulin to levels that saturate brain insulin recep- 1220-000-20). The flow-through was collected and then passed tors, rendering females relatively less sensitive to centrally-adminis- through 20-micron nylon mesh (Millipore). Microvessels were re- tered exogenous insulin. Based on this possibility, we hypothesized moved from the filter via brief vortexing in 1% bovine serum albumin that E2 exerts its effects on food intake and insulin sensitivity, in part, (BSA) in M199 medium in a new conical tube. Remaining microvessels by increasing the transport of insulin into the brain. To test this hypoth- were pelleted a final time (10 min at 1000 ×g, 4 °C), collected, and an- esis, we determined the effect of estrogen on insulin transport into the alyzed via qPCR or Western blotting/immunohistochemistry. brain following acute or chronic E2-treatment of OVX and male rats The relative purity of microvessel isolates was first verified via after verifying the expression of estrogen receptor-α at the BBB. TaqMan qPCR using a StepOne™ Plus device (Life Technologies), by Males were included in our studies, because the male brain appears screening isolates for markers of microvessel mRNA and assessing con- to be relatively more sensitive to insulin than OVX females [18];al- tamination from glial and neuronal mRNA (Fig. 1A) [41]. β-Actin was though OVX reduces estrogen production and increases the apparent used as the reference housekeeping gene to determine the relative sensitivity of the brain to insulin, they remain less sensitive to the gene expression of each microvessel and brain sample. Relative purity same dose of insulin than males and an effect can only be observed at was also assessed by visual inspection, as well as by confirming protein higher doses [18]. This is consistent with the observation that post- expression of the BBB-marker, p-glycoprotein using immunohisto- menopausal women remain insensitive to the anorectic effects of intra- chemistry. A portion of the remaining microvessel samples was then di- nasal insulin, which is comparable to the effects seen in pre-menopausal rectly spread onto glass slides for immunohistochemical analysis. After women [29]. Males are very responsive to the effects of E2, which re- fixing with 4% paraformaldehyde/PBS for 10 min, the microvessels duces the apparent sensitivity of insulin in the brain, such that they were washed 5 times, permeabilized in 0.1% Triton in PBS, rinsed 3 have the same response as females [18]. Therefore, if E2 is affecting in- times in PBS, and blocked in 2% serum. Microvessels were then incubat- sulin transport into the CNS, this effect would likely be easier to detect ed in primary antibody for 24 h at 4 °C. After rinsing 5 times with PBS, in males than in OVX rats. In all studies, insulin transport into the CNS secondary fluorescent antibodies (Cy-3 and Alexa-488, Molecular was assessed in vivo using our cerebrospinal fluid (CSF) collection tech- Probes, Carlsbad, CA) were added for 1 h in 1% BSA solution. This process nique [10,35,36]. was repeated with a different primary and secondary antibody for co- staining experiments. After rinsing 5× in PBS, slides were mounted 2. Materials and methods with SlowFade® Gold reagent (Molecular Probes). 24 h after curing at 26 °C, microvessels were visualized with fluorescence microscopy 2.1. Animals, diets, and determination of obesity (Olympus) (Fig. 1C–E). For antibody verification, microvessel lysates were analyzed via Western blotting (Fig. 1F) with previously-validated Adult female and male Long-Evans rats (14 weeks of age; Envigo, In- antibodies for estrogen receptor-α (ERα) (Millipore) and the insulin- dianapolis, IN) were housed in individual tub cages in a temperature- receptor-β (IR-β) subunit (Santa Cruz, sc-711). As a negative control controlled vivarium with a 12/12-hour light/dark cycle (lights on at for microvessel immunohistochemistry, we processed microvessels 314 A.A. May et al. / Physiology & Behavior 163 (2016) 312–321

Fig. 1. The 67-kDa isoform of ERα is expressed in rat brain microvessels. The purity of microvessel isolation was confirmed via qPCR (A); P-glycoprotein (P-gp); glucose-transporter-1 (GLUT1); S100 calcium-binding protein B (S100b); synaptophysin (Syp); Insulin receptor (INSR). ERα mRNA (Esr1) was also detected in brain microvessels (B) and its expression is comparable to that of the leptin receptor (Lepr); this graph is presented as a relative log∏ scale and the raw ΔCT for Esr1 in this experiment is 12.41 ± 0.30, relative to Actin. Estrogen receptor-α (ERα) immunoreactivity was detected on freshly isolated brain microvessels (C, red/Cy3) (10× magnification). Insulin receptor-β subunit (IR-β) (D, green/Alexa-488) was expressed in brain microvessels and co-localized with ERα (E, yellow). Nuclear staining with DAPI is observed in blue. The specificity of immunoreactivity against ERα and IR-β (IR) was verified via Western blotting (F). IR-β was also co-stained along with the BBB-specific marker, P-glycoprotein (P-gp) (G) (40× magnification). As a negative control, microvessels were co-stained for Cy3 and Alexa-488 fluorescent antibodies after being incubated without primary antibodies (H–I). Fluorescent staining is negligible in these samples. n = 4, *P b 0.05, and **P b 0.005. +/−SEM.

side-by-side with other samples, except that no primary antibodies were microrenethane tubing filled with sterile saline. The rat's scalp and added to the incubation solution. Microvessels were then incubated with neck were shaved, the site was sterilized, and an incision was made to secondary antibodies and were visualized, as described (Fig. 1H–I). expose the atlanto-occipital membrane. A needle was then carefully inserted into the cisternum magnum to collect CSF beginning exactly 2.3. Cerebrospinal fluid collection 30 min following insulin injection. The 30-min time-point of CSF collection has been demonstrated in previous studies to be ideal for measur- Cerebrospinal fluid (CSF) was collected from the cisterna magna of ing the appearance of insulin in the CSF in both chow-fed and HFD-fed rats as previously described [10,35].Briefly, rats were fasted overnight rats [10,36]. Using a 1-mL syringe, slight negative pressure was applied prior to the study, as indicated. Rats were then administered intraperi- to the tubing attached to the needle for 1–2 s to initiate the flow of CSF toneal (ip) injections of insulin during the middle of the light cycle into a chilled 1.5-mL microcentrifuge tube. Approximately 200 μLofCSF [10,36] to maintain consistent levels of plasma insulin among groups, was collected in 2–3 min and was immediately frozen on dry ice. Fol- and CSF was collected 30 min later. After the ip injection of insulin, lowing CSF collection and decapitation, blood was immediately collect- each rat was placed back in its home cage for 15 min prior to being anes- ed in chilled EDTA-coated tubes and plasma was isolated. Potential thetized with ketamine/xylazine (55.6 mg/kg and 8.8 mg/kg, respec- blood contamination of CSF samples was measured via spectrophotom- tively). Rats were then placed into a stereotaxic instrument with the etry, as described [43], by comparison to frozen, diluted rat-blood stan- head maximally ventroflexed. A 25-G needle, with a tip prepared at a dards. Samples with N 0.001% blood contamination were excluded from 30° angle [35], was inserted into an electrode holder and connected to analysis. A.A. May et al. / Physiology & Behavior 163 (2016) 312–321 315

CSF and plasma samples were analyzed with an ultrasensitive rat in- adipose tissue depots. Epididymal, mesenteric, retroperitoneal, and insulin ELISA kit (Crystal Chem, Downers Grove, IL, #90060). Since the guinal fat pads were weighed prior to freezing. low-range assay reliably detects insulin concentrations between 0.1 and 6.4 ng/mL, we loaded 20 μL of each CSF sample per well, according 2.7. Analysis of insulin transport to the kit instructions [36]. The resulting absorbance readings fell within the middle range of the standard curve. Values were then divided by 4 Insulin transport was assessed in terms of the absolute CSF insulin to adjust for the extra volume used, relative to the 5-μL volumes of the concentration, as well as the ratio of CSF/plasma insulin. This ratio pro- insulin standards. vides an adjusted measure of insulin in the CSF and accounts for the possibility that higher/lower insulin concentrations in the bloodstream 2.4. Acute E2 treatment in ovariectomized rats after ip injection could lead to a higher/lower concentration of insulin in the CSF at a dose of insulin that does not saturate insulin receptors A once-daily injection of estradiol-3-benzoate (E2; 10 μg/kg, Sigma) at the BBB [10]. or vehicle (20 μL, sesame oil, Sigma) was administered subcutaneously Similarly, we performed a correlational analysis of insulin transport (sc) to OVX rats for 2 days, a treatment regimen previously found to among groups by conducting a linear regression of plasma vs. CSF insu- exert catabolic effects in OVX rats [22]. During the 2-day treatment pe- lin. This approach determines whether higher/lower plasma insulin riod, food was removed to eliminate the variable of differential food in- concentrations are associated with higher/lower CSF concentrations, intake between groups. Using this protocol, insulin was injected 30 min dicative of effective transport. Previous studies that utilized Pearson prior to CSF collection after 2 days of E2 or vehicle administration. analysis to assess insulin transport into the CNS following a 1 U/kg- In a parallel study, brain microvessels were isolated from a separate dose of insulin (which targets postprandial levels of insulin in the ﬁ cohort of OVX rats and examined via qPCR for the expression of the in- blood) found that rats maintained on a low-fat diet had a signi cant cor- ﬁ sulin receptor, as described above. Preparation of these OVX rats mir- relation between CSF and plasma insulin levels, whereas no signi cant rored the original cohort, except that insulin was not administered correlation was observed in rats maintained on a HFD [10].Inour prior to brain-harvesting.

2.5. Acute E2 treatment in male rats

Because the effect of E2 on insulin transport may be blunted in OVX rats relative to males (as explained above) [18,44], we proceeded to use males in all subsequent studies. A single sc injection of E2 (10 μg/kg) or vehicle (20 μL) was administered to 14-week-old male rats 24 h prior to ip insulin and CSF collection. Both groups were pair-fed, by providing a ﬁxed, limited quantity of chow at the onset of dark, in order to normal- ize food intake across all rats. The amount of food was pre-determined based on pilot studies involving the injection of E2 or vehicle in ad- libitum-fed rats. Food was completely consumed at least 4 h prior to CSF collection.

2.6. Chronic treatment with E2

Male rats were treated with cyclic sc injections of E2 (10 μg/kg) or vehicle (20 μL) every 4th day for a period of 5 weeks, in order to achieve plasma levels of E2 that target the physiological range of the female es- trous cycle [21,22,37]. Both control rats and pair-fed rats received vehicle injections. In the ﬁrst cohort, male rats were maintained on HFD for 10 weeks to induce HFD-DIO prior to treatment with E2 or vehicle. Due to the anticipated reduction of body weight in rats receiving E2, a pair-fed group given vehicle was included to match the body weight of E2-treated rats. Thus, HFD-DIO rats were randomly divided into three groups: E2, vehicle, and pair-fed (PF). All groups were then treated with cyclical E2 while remaining on the HFD for 5 additional weeks prior to CSF collection, to promote at least a 10% lower body weight than controls [40]. Two d prior to CSF collection, rats were treated with a ﬁnal sc injection of E2 or vehicle. Insulin was injected ip at a dose of 1 U/kg, 30 min prior to CSF collection. A separate cohort of male rats was prepared in a manner that paralleled the previous study, except that treatment with E2 or vehicle was started at the onset of HFD-feeding in order to prevent HFD-DIO.

Again, both control rats and pair-fed rats received vehicle injections. Fig. 2. Acute treatment of E2 in chow-fed OVX rats does not affect insulin transport After 5 weeks of treatments and HFD-feeding, CSF was collected. through the BBB. Similar plasma levels of insulin were detected between control (Con) Again, a relative prevention of obesity in these animals was pre-defined and E2-treated (E2) OVX rats, 30-min after a 2 U/kg injection (A). There was no as having at least a 10% lower body weight than controls [40]. Fasting difference in CSF insulin (B) or in the ratio between insulin levels in the CSF and plasma (C). Brain microvessel mRNA from a different group of identically-treated OVX rats was glucose and body composition analysis were assessed 10 d prior to ip in- analyzed via qPCR for the relative expression of the insulin receptor (Insr) (D). The sulin and CSF collection, with E2 being injected 2 days prior to the test. acute E2-treatment regimen led to a significant reduction in body weight relative to Following CSF collection, whole fat pads were removed from white controls (E). n = 7–8, *P b 0.05, and +/−SEM. 316 A.A. May et al. / Physiology & Behavior 163 (2016) 312–321 current studies (Figs. 4 and 5), the plasma concentration of insulin was and water mass were measured and body weight was also assessed at increased to a level that was below the saturation point of insulin trans- this time. In order to determine the body fat percentage, we divided port through the BBB, as assessed in time-course studies [10,36]. Varia- each rat's fat mass by its body weight. The same calculation was used tions of the plasma insulin concentration in each rat therefore will for the lean mass and water mass. No differences in water mass or per- correspond with slightly different rates of insulin transport into the cent water mass were noted, so are not reported here. CNS of each rat [36]. For example, if E2 does not affect insulin transport through the BBB, the CSF concentration of insulin will be relatively con- 2.10. Analysis of insulin degrading enzyme (IDE) expression in the hypo- stant, regardless of the concentration of insulin in the blood. However, if thalamus and cortex E2 improves insulin transport on a HFD, then rats with higher plasma insulin would be expected to have higher CSF insulin levels, whereas Following CSF collection in rats that underwent chronic E2 treat- controls would not. This relationship was assessed by examining the ment, brains were collected and immediately flash-frozen. After dissec- strength and significance of the Pearson coefficients among groups. tion of the hypothalamus and the cortex, protein lysates were prepared In the first two CSF-transport studies examining acute E2 treatment and protein was quantified. After running SDS-PAGE and transferring to (Figs. 2 and 3), a high dose of insulin (2 U/kg) was used in order to sat- an Immobilon PVDF membrane (Millipore), Western Blots were incu- urate transporters at the BBB and to thereby maximize insulin transport bated with primary antibody for insulin degrading enzyme for 14 h at [36]. In this case, variations of blood insulin do not vary in proportion to 4 °C (1:1000, Abcam, San Francisco, CA). Blots were rinsed and incubat- CSF transport, preventing the use of regression analysis [36]. Therefore, ed in horseradish peroxidase-conjugated goat anti-rabbit IgG antibody Pearson analyses were only conducted in rats injected with 1 U/kg which was applied for 1 h at room temperature (1:10,000, Dako insulin. #P0448). After rinsing, Immobilon® HRP substrate (Millipore Cat. # WBKLS) was applied for 5 min at room temperature and blots were vi- 2.8. Glucose measurements sualized on photolithography film (Denville Scientific, #E3018), scanned, and quantified using ImageJ software (NIH). After gently strip- After an overnight fast, rats were anesthetized brieflywith ping blots for 15 min at 37 °C, membranes were rinsed and re-blotted isoflurane (Isothesia™, Butler Schein) and tail tips were clipped for with a loading control, the actin 5γ subunit (mouse IgG, Millipore) blood measurement. After 2 h, 300–400 μL blood was withdrawn from using similar procedures. the tail of each rat, following removal of any scabs, and fasting blood glucose was measured in duplicate using a glucometer (Accu-Chek, 2.11. Statistical analysis Roche, Indianapolis, IN). For the insulin tolerance tests, glucose was measured from plasma samples collected 30 min following ip insulin in- All statistics were analyzed using GraphPad Prism 5 using 2-way jection, using an Analox GM7 analyzer (Analox Instruments Limited, ANOVA with a Tukey post-hoc test for multiple comparisons, where ap- London). plicable. Pearson analysis was conducted along with multiple linear regression, as described in Section 2.7, to determine the strength and 2.9. Body composition analysis significance of the correlations between insulin concentrations in the plasma and CSF. In all studies, we pre-defined our cutoff for statistical For chronic E2-treatment studies, the rats' body compositions were significance at P ≤ 0.05. assessed with a rodent magnetic resonance imaging (MRI) machine (EchoMRI, EchoMedical Systems, Houston, TX). Fat mass, lean mass, 3. Results

3.1. Veriﬁcation of estrogen receptor-α expression at the BBB

As depicted in Fig. 1A, the brain microvessel isolates were highly enriched for P-gp (p-glycoprotein) and Glut-1 (glucose transporter-1) mRNA, which are highly expressed in brain endothelial cells relative to whole brain homogenates [41]. Expression of the glial cell marker, S100b (S100 calcium-binding protein B), and the neuronal cell marker, Syp (synaptophysin) mRNA was significantly lower in the BBB-isolates relative to the whole-brain homogenates, indicating that the isolation process enriched brain microvessels while selecting against neurons and glia [41]. Insulin receptor mRNA expression was elevated 3-fold in microvessels relative to the whole-brain homogenate. This finding was also reflected at the protein level (data not shown). Since insulin cannot passively diffuse through the BBB [45] and is understood to rely on an insulin receptor-mediated transport process in brain endothelial cells [46], the enrichment of insulin receptor expression in microvessels provides further evidence that the insulin receptor has a direct role in insulin transport through the BBB. We also verified microvessel purity at the protein level via immunohistochemistry for p-glycoprotein (P-gp), which is considered a marker of the BBB [41], due to its high enrichment in brain microvessels relative to the rest of the brain [47,48] (Fig. 1G). Microvessels that were processed without the addition of primary antibodies were devoid of nonspecificbinding from the secondary fluorescent antibodies (Fig. 1H–I). Fig. 3. Acute treatment of E2 in chow-fed male rats does not affect the transport of insulin Since estrogen receptor-α (ERα) is the most pertinent E2 receptor through the BBB. No differences were observed in plasma insulin (A), CSF insulin (B), or involved in the regulation of food intake and body weight, we first de- the ratio of CSF to plasma insulin 30 min following ip insulin. A single sc injection of E2 α α in ad libitum chow-fed male rats acutely reduced food intake relative to vehicle-injected termined that ER is expressed at the BBB. ER was detected at the controls (D). n = 8, *P b 0.05, and +/−SEM. mRNA level in freshly-isolated microvessels, with comparable A.A. May et al. / Physiology & Behavior 163 (2016) 312–321 317 expression as the leptin receptor (Fig. 1B). Insulin receptor mRNA ex- 3.3. Effects of short-term treatment with E2 on BBB-insulin transport in pression was approximately 3-fold higher than that of ERα.ERα immu- male rats noreactivity was also detected in freshly-isolated brain microvessels from both male and OVX female rats and appeared to be ubiquitously Consistent with the effects observed in OVX rats, acute E2 treatment expressed in the microvessels (representative staining in Fig. 1C). We did not alter the appearance of insulin in the CSF of males (Fig. 3B–C). also detected abundant protein expression of insulin receptor (Fig. The efficacy of acute E2 treatment in males was verified in pilot studies, 1D), as expected [45,46], which co-localized with ERα (Fig. 1E). The which demonstrated that food intake was significantly reduced relative specificity of the ERα and IR-β antibodies was verified with Western to vehicle-treated controls (Fig. 3D). Thus, acute E2 treatment, while blotting of brain microvessel lysates (Fig. 1F). Single bands correspond- having its expected effect on food intake and body weight, did so with- ing to the expected sizes of IR-β and ERα were observed. These findings out altering the expression of insulin receptors in the BBB and without collectively established an anatomical basis for an interaction of E2 with eliciting a change of insulin transport into the CNS. insulin transport at the BBB.

3.4. E2 reverses HFD-DIO without improving insulin transport through the 3.2. Effects of short-term E2-treatment on BBB-insulin transport in OVX rats BBB

While E2-treatment of OVX rats signiﬁcantly enhanced weight loss We next addressed the possibility that E2 requires a longer treat- relative to control rats under these conditions (Fig. 2E), no difference ment period to exert detectable effects on insulin transport at the BBB. in the appearance of insulin in the CSF was observed 30 min after ip in- Previously, we found that male rats with HFD-DIO have impaired trans- sulin (Fig. 2B). The plasma-to-CSF insulin ratio also did not differ among port of insulin into the CSF relative to controls on a LFD, and that this im- groups (Fig. 2C). Microvessels from OVX rats treated with acute E2 or pairment was reversed by dietary intervention with LFD [10].Since vehicle injections also had comparable expression of insulin receptor systemic E2 both reverses HFD-DIO [27] and increases systemic insulin (Insr) mRNA (Fig. 2D). sensitivity in rodent models [17,49], we determined whether these

Fig. 4. E2 reverses HFD-DIO in male rats independently of the BBB-insulin-transport system. (A) 24-week-old HFD-DIO male rats were treated with E2 or vehicle (Con) with ad-lib access to HFD or were pair-fed (PF) to the body weight of E2-treated rats for 33 days. (B) Fat pad masses of the rats were normalized to the total body weight to calculate the percent weight of each pad; Epi: epididymal, Ing: inguinal, Mes: mesenteric, and Ret: retroperitoneal. Body fat percent (C) was relatively constant during the treatment period. Plasma glucose was measured 30 min following ip insulin (D). Plasma insulin levels were comparable among all groups after 30 min following ip insulin (E). CSF insulin levels were assessed (F) and the ratio of CSF to plasma insulin did not differ among groups (G). Pearson analysis of the concentrations of plasma insulin vs. CSF insulin did not reveal any significant correlations (H) and lines were not significantly different from one another (P = 0.884), indicating that plasma insulin levels were not associated with increased insulin transport in all animals. The y-intercepts of the lines were also not significantly different from one another (P = 0.705). Even when all data points were combined, the slope of the resulting line was still not significantly different from zero (P = 0.312). n = 6–7, *P b 0.05, and +/−SEM. 318 A.A. May et al. / Physiology & Behavior 163 (2016) 312–321 metabolic improvements are partly mediated by chronically improving While fasting glucose was not significantly different between E2- insulin transport through the BBB. treated and PF rats (Fig. 5D), plasma glucose levels were significantly re- As expected, cyclic E2 treatment led to a significant reduction in ducedinE2-treatedandPFrats30minfollowingipinsulin(Fig. 5E). In body weight (12%) relative to vehicle-treated controls by the end of contrast, control rats did not have significant changes in plasma glucose the 5-week treatment period (Fig. 4A). Pair-fed rats were effectively after insulin injection, confirming that these rats were insulin-resistant. matched to the body weight of E2-treated rats. Because visceral adipos- The comparable effects observed between E2-treated and PF rats indi- ity is associated with impaired insulin transport through the BBB [10], cate that in this paradigm, E2 mediates its effects on glucose homeosta- we also examined the influence of visceral adiposity per se. E2 and sis and insulin sensitivity via a reduction in body weight, but not via a pair-feeding reduced total body fat relative to controls (data not reduction in body adiposity. shown), but did not change the percentage of body fat (Fig. 4C) or of 30 min after ip insulin, no difference in the concentration of CSF in- lean mass (data not shown), indicating that the loss of body weight re- sulin was observed among groups (Fig. 5G). Plasma insulin levels were sulted from a proportional loss of fat and lean tissue. There was also no also not statistically different among groups (Fig. 5F), and the ratio of in- relative change in the percent fat pad mass of inguinal fat (subcutane- sulin between the CSF and plasma compartments also did not differ ous) or retroperitoneal and mesenteric fat (visceral), among all groups among groups, which accounts for small variations in plasma insulin (Fig. 4B). Interestingly, although PF rats had a comparable proportion among groups (Fig. 5H). of epididymal fat as controls, E2-treatment significantly increased the Linear regression analysis revealed a significant Pearson correlation proportion of epididymal fat relative to PF rats (Fig. 4B). In addition, for the PF group, but not for the E2-treated or control rats (Fig. 5I). all rats had comparable fasting blood glucose levels (data not shown) The best-fit line of the PF group was also significantly different from and there was no difference in the levels of plasma glucose 30 min the lines of the E2-treated and control rats, suggesting that pair-feeding after ip insulin injection despite the significant difference in body is associated with a modest protection of BBB-insulin transport against weight and epididymal fat (Fig. 4D). the impairment of HFD-DIO, despite no detectable differences in abso- CSF insulin levels were not significantly altered by E2 treatment rel- lute CSF insulin levels among groups. In contrast, E2-treated rats did ative to controls (Fig. 4E), despite comparable levels of insulin in the not differ from controls based on this analysis. blood among groups in this paradigm (Fig. 4F). An additional analysis, To further verify that BBB-insulin transport is not changed by E2 comparing the ratio of insulin in the CSF relative to the level of insulin treatment, we measured insulin-degrading enzyme (IDE) in the brain. in the blood, yielded no difference among treatments (Fig. 4G). Previous reports have highlighted the possibility that E2 may upregu- Because saline-injected groups were not included to control for po- late the expression of the IDE in the brain under certain conditions tential differences in the baseline insulin levels in the CSF, we conducted [50], which could increase the degradation of insulin and reduce insulin an additional correlational analysis to identify differences in insulin levels in the CSF. This could potentially counteract an elevated rate of transport among groups. Since there was variability in the levels of plas- BBB-insulin transport if E2 increases transport but also increases IDE ex- ma and CSF insulin 30 min after ip insulin, we correlated the concentra- pression. We therefore measured the expression of IDE in the hypothal- tion of plasma insulin with the concentration of CSF-insulin that was amus (Fig. 5I) and cortex (Fig. 5J) via Western blotting, and found that detected at the same time-point. No significant associations were ob- the expression of IDE was not altered by E2 treatment or pair-feeding, served with Pearson analysis; both the slopes of the lines and y-inter- relative to controls, further strengthening the conclusion that E2 does cepts were not different among groups, such that lines were not exert its catabolic or systemic metabolic actions by altering insulin indistinguishable from each other. Since groups were no different transport into the CNS. from each other, all points were combined into a single correlational analysis. The Pearson correlation of all groups combined resulted in a 4. Discussion slope that was still not significantly different from zero, indicating that the concentration of insulin in the CSF was independent of the concen- Our finding that ERα is abundantly expressed in rat brain tration of insulin in the blood. microvessels is consistent with previous reports examining ERα in rat [51] and human [52] cortical endothelial cells. Our studies are unique 3.5. E2 prevents HFD-DIO and protects systemic insulin sensitivity, but does in revealing strong expression of the primary, 67-kDa ERα isoform, not affect insulin transport through the BBB which is enriched in the ovaries, muscle and brain [51],andwhichhas a strong link with energy homeostasis [53]; and importantly, ERα was Although E2 was unable to reverse a pre-existing impairment of in- co-expressed in the same BBB cells as the insulin receptor. Overall, sulin transport through the BBB, it is still possible that E2 may influence these findings provided a basis for a potential intracellular interaction BBB-insulin transport by HFD-DIO in younger rats (14-week-old males between E2-ERα signaling and insulin transport at the BBB. treated with E2 or vehicle immediately at the onset of HFD-feeding for Acute E2 treatment did not affect insulin transport in OVX or male 5 weeks). E2 readily prevents HFD-DIO and improves insulin sensitivity rats, despite significantly reducing body weight and food intake, respec- in a similar paradigm [17,33]. tively, relative to controls (Fig. 2A–C and Fig. 3A–C). Similarly, under E2-treatment significantly prevented weight gain over the 5-week conditions of chronic E2 treatment, insulin transport was not increased HFD-feeding period (Fig. 5A). Fat mass and lean mass were significantly after the prevention or reversal of HFD-DIO (Fig. 4E–GandFig. 5F–H), reduced, but E2 did not alter the total percentage of body fat (Fig. 5C) or even though insulin sensitivity was significantly protected in males lean mass (not shown) relative to controls. However, PF rats did have a treated with E2 at the onset of HFD-feeding. Because systemic insulin significant reduction in the proportion of body fat (Fig. 5C), while the sensitivity is positively associated with insulin transport through the proportion of lean mass was increased in comparison to controls (not BBB [10,54], it is surprising that E2-treatment did not affect insulin shown). The proportion of visceral and subcutaneous fat pads did not transport into the CNS under these conditions. However, since systemic differ between E2-treated and control rats (Fig. 5B). However, PF rats insulin resistance was not reversed in HFD-DIO male rats, despite signif- had a significantly smaller percentage of fat in visceral fat pads com- icant loss of body weight over the 5-week period (Fig. 4A, F), it is less pared to E2-treated and control rats and also had a smaller percentage surprising that insulin transport was not improved in these animals. of fat in the inguinal fat depot (a subcutaneous fat pad) relative to E2- It is possible that the failure to detect a difference of insulin transport treated rats (Fig. 5B). Together, this indicates that E2-treatment reduced following chronic E2-treatment and HFD-feeding was due to differences body weight by evenly reducing the proportion of fat and lean mass, in the levels of insulin in the blood and CSF. Because control rats main- whereas the reduction of body weight by pair-feeding resulted in a tained on HFD are insulin resistant relative to E2 or PF groups (Fig. 5E), it modest shifting of fat mass to lean mass. is possible that controls started with slightly higher levels of CSF-insulin A.A. May et al. / Physiology & Behavior 163 (2016) 312–321 319

Fig. 5. E2 prevents HFD-DIO in male rats independently of the BBB-insulin-transport system. (A) 14-week-old HFD-DIO male rats were treated with E2 or vehicle with ad-lib access to HFD or were pair-fed to the body weight of E2-treated rats for 37 days. (B) Fat pad masses of the rats were expressed as percent body weight. Body fat percentage was significantly lower in PF, but not E2-treated rats (C). Fasting glucose levels were not different among groups (D) but 30 min after ip injection of 1 U/kg insulin, plasma glucose was significantly lower in E2-treated and PF rats compared to controls (E). Plasma insulin (F) and CSF insulin (G) were measured from samples collected 30-min following ip insulin (F). The ratio of CSF to plasma insulin did not differ among groups (H). CSF insulin levels were correlated with plasma insulin levels and the Pearson coefficient was calculated (I). The slope of the PF line was significantly different from zero (P = 0.0135, r2 = 0.7565), whereas there was no significant correlation in the E2 group (P = 0.4442, r2 = 0.1524) or controls (P = 0.3741, r2 = 0.1998). The expression of insulin-degrading enzyme (IDE) in the hypothalamus (J) and cortex (K) was also assessed from brains of the same rats tested in this study. n = 6–7, *P b;0.05,and+/−SEM. prior to insulin injection [10]. This baseline difference could potentially was not improved in these groups. Surprisingly, this suggests that in mask any differences of insulin transport into the CSF among groups, so rats of the same body weight, E2 may actually inhibit the transport of in- we used an alternative approach that is effective in assessing insulin sulin into the CNS when maintained on a HFD. It is also possible that E2's transport under these conditions [10]: linear regression analysis re- ability to prevent and reverse HFD-DIO and to protect peripheral insulin vealed that only PF rats had a significant Pearson correlation between sensitivity occurs independently of changes in BBB-insulin transport. the level of insulin in the blood and CSF following the prevention of These possibilities raise a number of questions that warrant future HFD-DIO (Fig. 5I), indicating that a portion of the administered insulin study. was transported into the CNS. In contrast, E2-treated and control rats In comparison with the cohort of rats that underwent a reversal of had no such correlation, implying that insulin transport into the CNS HFD-DIO, Pearson analysis revealed a nonsignificant correlation for all 320 A.A. May et al. / Physiology & Behavior 163 (2016) 312–321 groups (Fig. 4H). It is interesting that insulin transport into the CNS was [65]. However, data on this matter are limited by a lack of consistency associated with improved insulin transport only when HFD was not between individual CSF samples, such as not using a standardized consumed chronically prior to the start of treatments. fasting duration prior to CSF withdrawal, a factor known to affect the in- Because our approach measures unbound insulin in CSF samples, the sulin concentration in the plasma and CSF [66]. The menopausal status detection of insulin depends not only on the rate of insulin transport of female subjects should also be accounted for when analyzing such into the CNS, but also on its subsequent degradation in the CNS and data, to avoid differences in the levels of E2 in this group. Our data cir- the relative amount of insulin that is bound to insulin receptors on cumvent these issues by conducting the studies in a controlled manner brain cells. We did not observe a difference in the expression of insulin with a standardized diet. degrading enzyme in the brains of chronically E2-treated, PF, or control Taken together, our data demonstrate that E2 can improve energy rats, implying that differential insulin degradation in the CNS is unlikely balance and systemic insulin sensitivity independently of altering to be a factor. Additionally, obesity decreases the expression of insulin BBB-insulin transport. Further, our data indicate that food restriction receptors at the BBB, which is thought to reduce insulin transport into may protect BBB-insulin transport during maintenance on a HFD. the CNS [54]. E2-treatment does not affect insulin binding in the brains These results raise intriguing questions as to how E2 reduces insulin of OVX rats [32], making it is unlikely that insulin is transported differ- sensitivity in the brain and suggest the involvement of E2 in the modu- entially in E2-treated rats, and that the difference is masked by higher lation of neuronal or glial insulin signaling in the brain. Understanding binding of insulin to brain cells. Taken together, these observations this system could lead to the development of new therapeutic strategies strengthen our conclusion that E2 neither alters the transport of insulin in the prevention and treatment of obesity. into the CNS nor its degradation therein. Since systemic insulin sensitivity did not significantly differ between Acknowledgements PF and E2-treated rats, it is intriguing to ask what other factors might be playing a role in the increased BBB-insulin transport of PF rats. Since We would like to thank Dr. Deborah Clegg for her valuable input in total body fat and visceral fat are associated with impaired insulin trans- the preparation of this manuscript and to Joyce Sorrell and Dr. Alfor port into the CNS [10,11,18], we examined the relationship with adipos- Lewis for their helpful technical support. This work was funded by the ity to account for the differences noted in insulin transport between E2- National Institute of Diabetes and Digestive and Kidney Diseases treated and PF rats. Rats that underwent treatment for the reversal of (NIDDK) awards DK017844, DK92779, DK95440, and DK059803. HFD-DIO for 5 weeks (Fig. 4A) displayed clear changes in body weight, which corresponded with a lower proportion of epididymal fat in PF rats relative to E2-treated rats (although there was no change in the total References body fat percentage), but this did not influence BBB-insulin transport, [1] G.L. Alberti, P. Zimmet, J. Shaw, The IDF Consensus Worldwide Definition of the and this also did not appear to affect the metabolic syndrome to a sub- Metabolic Syndrome, IDF Communications, 2006 (http://www.idf.org/webdata/ stantial degree. docs/IDF_Meta_def_final.pdf. [accessed May 10, 2016]). In contrast, in the follow-up study of HFD-DIO-prevention, pair- [2] S. Obici, B.B. Zhang, G. Karkanias, L. Rossetti, Hypothalamic insulin signaling is – fi required for inhibition of glucose production, Nat. Med. 8 (12) (2002) 1376 1382, feeding led to a signi cantly lower percentage of total body fat and vis- http://dx.doi.org/10.1038/nm1202-798. ceral fat storage relative to both E2-treated and control rats. The reduc- [3] S. Obici, Z. Feng, G. Karkanias, D.G. Baskin, L. Rossetti, Decreasing hypothalamic insu- tion of adiposity exclusively in PF rats might potentially explain why lin receptors causes hyperphagia and insulin resistance in rats, Nat. Neurosci. 5 (6) (2002) 566–572, http://dx.doi.org/10.1038/nn0602-861. pair-feeding was associated with protected BBB-insulin transport. Fur- [4] S.C. Woods, E.C. Lotter, L.D. McKay, D. Porte Jr., Chronic intracerebroventricular ther studies are necessary to determine the mechanism responsible infusion of insulin reduces food intake and body weight of baboons, Nature 282 for this difference. It is surprising that E2-treatment did not reduce the (0028-0836 (Print)) (1979) 503–505. [5] J.C. Bruning, D. Gautam, D.J. Burks, J. Gillette, M. Schubert, P.C. Orban, et al., Role of proportion of visceral fat storage and did not lead to increased subcuta- brain insulin receptor in control of body weight and reproduction, Science 289 neous fat in male rats, despite many publications reporting that E2 pro- (0036-8075) (2000) 2122–2125. motes increased storage of subcutaneous fat in a “gynoid” distribution [6] M. Hallschmid, C. Benedict, B. Schultes, H.L. Fehm, J. Born, W. Kern, Intranasal insulin [18,55]. However, several reports examining the effects of E2-treatment reduces body fat in men but not in women, Diabetes 53 (0012-1797 (Print)) (2004) 3024–3029. in male and OVX rats and mice have reported similar findings to ours [7] M. Hallschmid, C. Benedict, B. Schultes, J. Born, W. Kern, Obese men respond to [56,57]. It is possible that testosterone interacts with E2 to alter the typ- cognitive but not to catabolic brain insulin signaling, Int. J. Obes. 32 (2) (2008) – ical fat distribution and/or that HFD alters the fat distribution normally 275 282, http://dx.doi.org/10.1038/sj.ijo.0803722 (2005). [8] C.A. Grillo, K.L. Tamashiro, G.G. Piroli, S. Melhorn, J.T. Gass, R.J. Newsom, et al., maintained by E2. Regardless of differences in fat distribution, our re- Lentivirus-mediated downregulation of hypothalamic insulin receptor expression, sults are consistent with previous findings that E2 increases adiposity Physiol. Behav. 92 (4) (2007) 691–701, http://dx.doi.org/10.1016/j.physbeh.2007. relative to weight-matched controls (either OVX or males) [58,59]. 05.043. [9] L. Koch, F.T. Wunderlich, J. Seibler, A.C. Könner, B. Hampel, S. Irlenbusch, et al., The majority of insulin in the CNS is thought to be derived from the Central insulin action regulates peripheral glucose and fat metabolism in mice, transport of blood-borne insulin through the BBB [45,60,61]. After en- J. Clin. Invest. 118 (6) (2008) 2132–2147, http://dx.doi.org/10.1172/JCI31073. tering the brain interstitial fluid compartment, insulin can bind its re- [10] D.P.Begg,J.D.Mul,M.Liu,B.M.Reedy,D.A.D’Alessio, R.J. Seeley, et al., Reversal of diet- fl fl induced obesity increases insulin transport into cerebrospinal uid and restores ceptors on brain cells. The bulk ow of unbound insulin then makes sensitivity to the anorexic action of central insulin in male rats, Endocrinology 154 its way into the CSF and is then transported back into the blood. Based (3) (2013) 1047–1054, http://dx.doi.org/10.1210/en.2012-1929. on this model, the most ideal assessment of insulin transport into the [11] W. Kern, C. Benedict, B. Schultes, F. Plohr, A. Moser, J. Born, et al., Low cerebrospinal fluid insulin levels in obese humans, Diabetologia 49 (11) (2006) 2790–2792, CNS would involve measuring the insulin concentration in brain inter- http://dx.doi.org/10.1007/s00125-006-0409-y. stitial fluid before and after ip-insulin administration. Unfortunately, [12] S.C. Woods, R.J. Seeley, Adiposity signals and the control of energy homeostasis, the technical demands of interstitial fluid collection limit its reliability. Nutrition 16 (10) (2000) 894–902 (2000 Oct.). CSF collection is a strong alternative [62,63], because a large volume [13] D.P. Begg, S.C. Woods, The central insulin system and energy balance, Handb. Exp. Pharmacol. 209 (0171-2004 (Print)) (2012) 111–129. can be collected rapidly at high purity and the concentration of insulin [14] A. Torquati, R. Lutfi, N. Abumrad, W.O. Richards, Is Roux-en-Y gastric bypass surgery in the CSF correlates strongly with the amount of insulin in the brain the most effective treatment for type 2 diabetes mellitus in morbidly obese – [36,64]. patients? J. Gastrointest. Surg. 9 (1091-255X (Print)) (2005) 1112 1116. [15] L. Shen, M. Haas, D.Q.-H. Wang, A. May, C.C. Lo, S. Obici, et al., Ginsenoside Rb1 If insulin transport in the CNS were increased by E2, then females or increases insulin sensitivity by activating AMP-activated protein kinase in male E2-treated subjects might be expected to have higher concentrations of rats, Physiol. Rep. 3 (9) (2015), http://dx.doi.org/10.14814/phy2.12543. insulin in the CNS than males or vehicle-treated subjects of the same age [16] L. Zhu, W.C. Brown, Q. Cai, A. Krust, P. Chambon, O.P. McGuinness, et al., Estrogen treatment after ovariectomy protects against fatty liver and may improve and body weight. One study examining this question found that healthy pathway-selective insulin resistance, Diabetes 62 (2) (2013) 424–434, http://dx. women and men do not differ in the concentration of insulin in the CSF doi.org/10.2337/db11-1718. A.A. May et al. / Physiology & Behavior 163 (2016) 312–321 321

[17] E. Riant, A. Waget, H. Cogo, J.-F. Arnal, R. Burcelin, P. Gourdy, Estrogens protect [41] S. Yousif, C. Marie-Claire, F. Roux, J.-M. Scherrmann, X. Declèves, Expression of drug against high-fat diet-induced insulin resistance and glucose intolerance in mice, En- transporters at the blood-brain barrier using an optimized isolated rat brain docrinology 150 (5) (2009) 2109–2117, http://dx.doi.org/10.1210/en.2008-0971. microvessel strategy, Brain Res. 1134 (1) (2007) 1–11, http://dx.doi.org/10.1016/j. [18] D.J. Clegg, L.M. Brown, S.C. Woods, S.C. Benoit, Gonadal hormones determine sensi- brainres.2006.11.089. tivity to central leptin and insulin, Diabetes 55 (0012-1797 (Print)) (2006) [42] B.E. Enerson, L.R. Drewes, The rat blood-brain barrier transcriptome, J. Cereb. Blood 978–987. Flow Metab. 26 (7) (2006) 959–973, http://dx.doi.org/10.1038/sj.jcbfm.9600249. [19] S. Thammacharoen, T.A. Lutz, N. Geary, L. Asarian, Hindbrain administration of estra- [43] J.S. You, V. Gelfanova, M.D. Knierman, F.A. Witzmann, M. Wang, J.E. Hale, The diol inhibits feeding and activates estrogen receptor-alpha-expressing cells in the impact of blood contamination on the proteome of cerebrospinal fluid, Proteomics nucleus tractus solitarius of ovariectomized rats, Endocrinology 149 (0013-7227 5 (1615-9853 (Print)) (2005) 290–296. (Print)) (2008) 1609–1617. [44] S.C. Woods, K. Gotoh, D.J. Clegg, Gender differences in the control of energy homeo- [20] L. Asarian, N. Geary, Cyclic estradiol treatment phasically potentiates endogenous stasis, Exp. Biol. Med. (Maywood) 228 (1535-3702 (Print)) (2003) 1175–1180. cholecystokinin’s satiating action in ovariectomized rats, Peptides 20 (0196-9781 [45] W.A. Banks, The source of cerebral insulin, Eur. J. Pharmacol. 490 (1-3) (2004) 5–12, (Print)) (1999) 445–450. http://dx.doi.org/10.1016/j.ejphar.2004.02.040. [21] L. Shen, D.Q. Wang, C.M. Lo, P. Tso, W.S. Davidson, S.C. Woods, et al., Estradiol in- [46] K.R. Duffy, W.M. Pardridge, Blood-brain barrier transcytosis of insulin in developing creases the anorectic effect of central apolipoprotein A-IV, Endocrinology 151 rabbits, 420 (1987). (1945-7170 (Electronic)) (2010) 3163–3168. [47] A.M. Karssen, O. Meijer, D. Pons, E.R. De Kloet, Localization of mRNA expression [22] N. Geary, L. Asarian, Cyclic estradiol treatment normalizes body weight and test of P-glycoprotein at the blood-brain barrier and in the hippocampus, Ann. N. Y. meal size in ovariectomized rats, Physiol. Behav. 67 (0031-9384 (Print)) (1999) Acad. Sci. 1032 (2004) 308–311, http://dx.doi.org/10.1196/annals.1314.048. 141–147. [48] A. Schinkel, P-Glycoprotein, a gatekeeper in the blood-brain barrier, Adv. Drug Deliv. [23] A. Alonso, H. González-Pardo, P. Garrido, N.M. Conejo, P. Llaneza, F. Díaz, et al., Acute Rev. 36 (2-3) (1999) 179–194. effects of 17 β-estradiol and genistein on insulin sensitivity and spatial memory in [49] M.L. Liu, X. Xu, W.Q. Rang, Y.J. Li, H.P. Song, Influence of ovariectomy and 17beta-es- aged ovariectomized female rats, Age (Dordr.) 32 (4) (2010) 421–434, http://dx. tradiol treatment on insulin sensitivity, lipid metabolism and post-ischemic cardiac doi.org/10.1007/s11357-010-9148-6. function, Int. J. Cardiol. 97 (0167-5273 (Print)) (2004) 485–493. [24] D.A. Sandoval, A.C. Ertl, M.A. Richardson, D.B. Tate, S.N. Davis, Estrogen Blunts Neu- [50]L.Zhao,J.Yao,Z.Mao,S.Chen,Y.Wang,R.D.Brinton,17β-Estradiol regulates roendocrine and Metabolic Responses to Hypoglycemia, Diabetes 52 (7) (2003) insulin-degrading enzyme expression via an ERβ/PI3-K pathway in hippocampus: 1749–1755, http://dx.doi.org/10.2337/diabetes.52.7.1749. relevance to Alzheimer’s prevention, Neurobiol. Aging 32 (11) (2011) 1949–1963, [25] R.E. Van Pelt, W.S. Gozansky, P. Wolfe, J.M. Kittelson, C.M. Jankowski, R.S. Schwartz, http://dx.doi.org/10.1016/j.neurobiolaging.2009.12.010. et al., Estrogen or raloxifene during postmenopausal weight loss: Adiposity and car- [51] C. Stirone, S.P. Duckles, D.N. Krause, Multiple forms of estrogen receptor-alpha in diometabolic outcomes, Obesity (Silver Spring) (2013), http://dx.doi.org/10.1002/ cerebral blood vessels: regulation by estrogen, Am. J. Physiol. Endocrinol. Metab. oby.20653. 284 (1) (2003) E184–E192, http://dx.doi.org/10.1152/ajpendo.00165.2002. [26] H. Shi, R.J. Seeley, D.J. Clegg, Sexual differences in the control of energy homeostasis, [52] A. Razmara, L. Sunday, C. Stirone, X.B. Wang, D.N. Krause, S.P. Duckles, et al., Front. Neuroendocrinol. 30 (1095-6808 (Electronic)) (2009) 396–404. Mitochondrial effects of estrogen are mediated by estrogen receptor alpha [27] B. Finan, B. Yang, N. Ottaway, K. Stemmer, T.D. Müller, C.-X. Yi, et al., Targeted in brain endothelial cells, J. Pharmacol. Exp. Ther. 325 (3) (2008) 782–790, http:// estrogen delivery reverses the metabolic syndrome, Nat. Med. 18 (12) (2012) dx.doi.org/10.1124/jpet.107.134072. 1847–1856, http://dx.doi.org/10.1038/nm.3009. [53] E. Morselli, E. Fuente-Martin, B. Finan, M. Kim, A. Frank, C. Garcia-Caceres, et al., Hy- [28] P. Xu, X. Cao, Y. He, L. Zhu, Y. Yang, K. Saito, et al., Estrogen receptor-α in pothalamic PGC-1α protects against high-fat diet exposure by regulating ERα,Cell medial amygdala neurons regulates body weight, J. Clin. Invest. 125 (7) (2015) Rep. 9 (2) (2014) 633–645, http://dx.doi.org/10.1016/j.celrep.2014.09.025. 2861–2876, http://dx.doi.org/10.1172/JCI80941. [54] M.W. Schwartz, D.F. Figlewicz, S.E. Kahn, D.G. Baskin, M.R.C. Greenwood, D. Porte, [29] R. Krug, C. Benedict, J. Born, M. Hallschmid, Comparable Sensitivity of Postmeno- Insulin binding to brain capillaries is reduced in genetically obese, hyperinsulinemic pausal and Young Women to the Effects of Intranasal Insulin on Food Intake Zucker rats, Peptides 11 (3) (1990) 467–472. and Working Memory, J. Clin. Endocrinol. Metab. 95 (12) (2010) E468–E472, [55] Y. Xu, T.P. Nedungadi, L. Zhu, N. Sobhani, B.G. Irani, K.E. Davis, et al., Distinct hypo- http://dx.doi.org/10.1210/jc.2010-0744. thalamic neurons mediate estrogenic effects on energy homeostasis and reproduc- [30] A. Alonso, R. Fernandez, M. Moreno, P. Ordonez, H. Gonzalez-Pardo, N.M. Conejo, tion, Cell Metab. 14 (4) (2011) 453–465, http://dx.doi.org/10.1016/j.cmet.2011.08. et al., Positive Effects of 17ss-Estradiol on Insulin Sensitivity in Aged Ovariectomized 009. Female Rats, J. Gerontol. Ser. A Biol. Med. Sci. 61 (5) (2006) 419–426, http://dx.doi. [56] V. Gloy, W. Langhans, J.J. Hillebrand, N. Geary, L. Asarian, Ovariectomy and org/10.1093/gerona/61.5.419. overeating palatable, energy-dense food increase subcutaneous adipose tissue [31] K.D. Niswender, C.D. Morrison, D.J. Clegg, R. Olson, D.G. Baskin, M.G. Myers Jr., et al., more than intra-abdominal adipose tissue in rats, Biol. Sex Differ. 2 (1) (2011) 6, Insulin activation of phosphatidylinositol 3-kinase in the hypothalamic arcuate nu- http://dx.doi.org/10.1186/2042-6410-2-6. cleus: a key mediator of insulin-induced anorexia, Diabetes 52 (0012-1797 (Print)) [57] P.A. Heine, J.A. Taylor, G.A. Iwamoto, D.B. Lubahn, P.S. Cooke, Increased adipose (2003) 227–231. tissue in male and female estrogen receptor-alpha knockout mice, Proc. Natl. [32] D.G. Baskin, E.P. Gierke, B.J. Wilcox, A.M. Matsumoto, M.W. Schwartz, Food intake Acad. Sci. U. S. A. 97 (0027-8424 (Print)) (2000) 12729–12734. and estradiol effects on insulin binding in brain and liver, Physiol. Behav. 53 [58] K.J. Nickelson, K.L. Stromsdorfer, R.T. Pickering, T.-W. Liu, L.C. Ortinau, A.F. Keating, (0031-9384 (Print)) (1993) 757–762. et al., A comparison of inflammatory and oxidative stress markers in adipose tissue [33] W. Pratchayasakul, N. Chattipakorn, S.C. Chattipakorn, Effects of estrogen in from weight-matched obese male and female mice, Exp. Diabetes Res. (2012), preventing neuronal insulin resistance in hippocampus of obese rats are different 859395, http://dx.doi.org/10.1155/2012/859395. between genders, Life Sci. 89 (19-20) (2011) 702–707, http://dx.doi.org/10.1016/ [59] K.E. Davis, M.D. Neinast, K. Sun, W.M. Skiles, J.D. Bills, J.A. Zehr, et al., The sexually j.lfs.2011.08.011. dimorphic role of adipose and adipocyte estrogen receptors in modulating adipose [34] M. Chavez, K. Kaiyala, L.J. Madden, M.W. Schwartz, S.C. Woods, Intraventricular In- tissue expansion, inflammation, and fibrosis, Mol. Metab. 2 (3) (2013) 227–242, sulin and the Level of Maintained Body Weight in Rats, Behav. Neurosci. 109 (3) http://dx.doi.org/10.1016/j.molmet.2013.05.006. (1995) 528–531. [60] G.D. Baura, D.M. Foster, D. Porte Jr., S.E. Kahn, R.N. Bergman, C. Cobelli, et al., Satura- [35] M. Liu, L. Shen, D.P. Begg, D.A. D’alessio, S.C. Woods, Insulin increases central apoli- ble transport of insulin from plasma into the central nervous system of dogs in vivo. poprotein E levels as revealed by an improved technique for collection of cerebro- A mechanism for regulated insulin delivery to the brain, J. Clin. Invest. 92 (0021- spinal fluid from rats, J. Neurosci. Methods 209 (1) (2012) 106–112, http://dx.doi. 9738 (Print)) (1993) 1824–1830. org/10.1016/j.jneumeth.2012.05.034. [61] M.W. Schwartz, R.N. Bergman, S.E. Kahn, G.J. Taborsky, L.D. Fisher, A.J. Sipols, et al., [36] D.P. Begg, A.A. May, J.D. Mul, M. Liu, D.A. D’Alessio, R.J. Seeley, et al., Insulin Detemir Evidence for entry of plasma insulin into cerebrospinal fluid through an intermedi- Is Transported From Blood to Cerebrospinal Fluid and Has Prolonged Central ate compartment in dogs. Quantitative aspects and implications for transport, J. Clin. Anorectic Action Relative to NPH Insulin, Diabetes 64 (7) (2015) 2457–2466, Invest. 88 (4) (1991) 1272–1281, http://dx.doi.org/10.1172/JCI115431. http://dx.doi.org/10.2337/db14-1364. [62] D.D. Shen, A.A. Artru, K.K. Adkison, Principles and applicability of CSF sampling for [37] L. Shen, Y. Liu, D.Q.H. Wang, P. Tso, S.C. Woods, M. Liu, Estradiol stimulates apolipo- the assessment of CNS drug delivery and pharmacodynamics, Adv. Drug Deliv. protein A-IV gene expression in the nucleus of the solitary tract through estrogen Rev. 56 (0169-409X (Print)) (2004) 1825–1857. receptor-α, Endocrinology 155 (10) (2014) 3882–3890, http://dx.doi.org/10.1210/ [63] E. Dolgodilina, S. Imobersteg, E. Laczko, T. Welt, F. Verrey, V. Makrides, Brain en.2014-1239. interstitial fluid glutamine homeostasis is controlled by blood-brain barrier [38] E.L.B. Novelli, Y.S. Diniz, C.M. Galhardi, G.M.X. Ebaid, H.G. Rodrigues, F. Mani, et al., SLC7A5/LAT1 amino acid transporter, J. Cereb. Blood Flow Metab. (2015), Anthropometrical parameters and markers of obesity in rats, Lab. Anim. 41 (1) http://dx.doi.org/10.1177/0271678X15609331. (2007) 111–119, http://dx.doi.org/10.1258/002367707779399518. [64] M.W. Schwartz, A. Sipols, S.E. Kahn, D.F. Lattemann, G.J.J. Taborsky, R.N. Bergman, [39] M.K. Montgomery, N.L. Hallahan, S.H. Brown, M. Liu, T.W. Mitchell, G.J. Cooney, et al., et al., Kinetics and specificity of insulin uptake from plasma into cerebrospinal Mouse strain-dependent variation in obesity and glucose homeostasis in response fluid, Am. J. Physiol. Endocrinol. Metab. 259 (3) (1990) E378–E383. to high-fat feeding, Diabetologia 56 (5) (2013) 1129–1139, http://dx.doi.org/10. [65] E. Bilić,E.Bilić,V.Kušec, K. Zibar, D. Čerimagić, P. Konjevoda, et al., Concentrations of 1007/s00125-013-2846-8. GH, IGF-1 and insulin in CSF of healthy people, Neurol. Croat. 60 (2) (2011) 73–81. [40] S.C. Woods, R.J. Seeley, P.A. Rushing, D. D’Alessio, P. Tso, A controlled high-fat diet [66] J.H. Strubbe, D. Porte Jr., S.C. Woods, Insulin responses and glucose levels in plasma induces an obese syndrome in rats, J. Nutr. 133 (0022-3166 SB - IM) (2003) and cerebrospinal fluid during fasting and refeeding in the rat, Physiol. Behav. 44 1081–1087. (0031-9384 (Print)) (1988) 205–208.