Supplement to Nutrition Vol. 12, No. 1, 1996

The Gut-Brain Brain-Gut Axis in Anorexia: Toward an Understanding

of Food Intake Regulation

MICHAEL M. MEGUID, MD, PHD, ZHONG-JIN YANG, MD, AND JOHN R. GLEASON, PHD

From the Surgical Metabolism and Nutrition Laboratory, Neuroscience Program, Department of Surgery, University Hospital, SUNY Health Science Center; and the Statistics Program, College of Arts and Sciences,

Syracuse University, Syracuse, New York, USA

ABSTRACT

Our long-term objectives continue to be elucidation of the mechanisms that control spontaneous food intake (SFI), so that we may utilize this information in seeking ways to ameliorate abnormalities of SFI that occur in nutritionally ill humans. To this end, we have developed and used an Automated Computerized Rat Eater Meter (ACREM), which allows detailed determinations of food intake and feeding patterns under a wide variety of experimental conditions. Because food intake is the product of meal number and meal size, these indexes were studied in a variety of experimental situations: normal male Fischer rats, genetically obese Zucker rats, cancer-bearing rats, and an inflammatory bowel rat model. In each model, a reduction in food intake was accomplished; usually by a selective reduction in meal number and, occasionally, meal size; often in both. The independent regulation of meal number and meal size strongly suggests the existence of focal neuronal areas in the hypothalamic food regulatory areas of the brain, which independently control these feeding indexes. To these feeding pattern studies were added in vivo focal hypothalamic microdialysis to correlate changes in meal size and number with changes in the basic neurotransimtters, dopamine and serotonin. To further gain an understanding of anorexia and food intake regulation in these models as it relates to the brain and gut interaction, we used metabolic stimulants, anatomic ablation, and electrophysiological studies, cytokines, selective neurotransmitter agonists, and antagonists peripherally in the gut and cenu~lly in the brain. An integrated view of the gut-brain brain-gut control of food intake has emerged as a working and testable model system. The system includes oronasal pregastric factors, which stimulate an increase in LHA-dopamine facilitating gastric compliance via efferent vagal fibers; postabsorptive factors, including nutrients and hepatoportal receptors via afferent vagal fibers that inhibit further LHA-dopamine, thereby regulating meal size. The same postabsorptive factors simultaneously decrease VMH-dopamine, thereby determining postprandial intermeal duration, because food intake is resumed when VMH- dopamine normalizes--thus regulating meal number. Changes in plasma amino acids, the precursors for neurotransmitters, also affect brain availability for neurotransmitters. This in particular applies to tryptophan, the precursor of serotonin in the VMH, which induces a decrease in meal number and cytokines, which facilitate activity of both dopamine and serotonin.

Key words: food intake, meal size, meal number, lateral hypothalamus (LHA), ventromedial hypothalamus (VMH), neurotransmitters, parenteral nutrients, vagal afferents, eater meter, anorexia

INTRODUCTION

The basic tenant of our laboratory's research is that food intake is the product of meal number and meal size (FI = MN x MZ). Using primarily the Fischer 344 rat as an experimental model, we first developed an automated computerized rat eater meter (ACREM m) to measure food intake, meal number, meal size, meal duration, and intermeal intervals. The ACREM (Fig. 1) consists of commercially available metabolic cages, with the supplied feeding cup replaced by an electronic scale-balance and two

photoelectric cells centered above the food dish. Access to food is via a feeding tunnel monitored by photocells, and food consumption is measured by an electronic scale that weighs the amount eaten during each period of access to food. A remote computerized data-collection device integrates feeding activity, as measured by the electronic scale and the photocells during real time. The signals so generated are processed and interpreted by a computer and recorded. The ACREM continuously measures food intake and the other feeding indexes consisting of meal size,

Correspondence to: Michael M. Meguid, MD0 PhD, Department of Surgery, University Hospital, 750 East Adams Street, Syracuse, NY 13210, USA.

Nutrition 12:$57-$62, 1996 ©Elsevier Science Inc. 1996 Printed in the USA. All rights reserved.

0899-9007/96/$15.00 SSD10899-9007(95)00083-6

S58 GUT-BRAIN BRAIN-GUT AXIS

Water Bottle .

-- -- ~ Photocells

Collector "~ - ~" Collector I ~ - ' - ' ~ - - To Computer . . . . ~ --- s,l~. ~ l ~ . U r i n e Collector Aluminum Box

FIG. 1. The Automated Computerized Rat Eater Meter (ACREM). The rat accesses food via the feeding tunnel, which is monitored by photocells. Food intake is measured by an electronic scale. A remote data collection device integrates eating related activity with real time. The data are then processed by a computer and recorded. Reproduced from Meguid, Kawashima, Campos, et al. 1, with permission.

number of meals, meal duration, meal sniffs, and eating activity for the 12-h light/dark cycle and the 24-h period.

EXPERIMENTAL ANOREXIA MODELS AS RELATED TO FOOD INTAKE. MEAL NUMBER, AND SIZE

In the normal male Fischer rat, approximately 80% of the food intake occurs during the dark phase, whereas the rest occurs during the light phase. Similarly, the meal number is significantly greater during the dark phase than during the light phase, whereas meal size is the same. In contrast, in the morbidly obese Zucker rat, which consumes almost double the quantity of food per 24 h compared to the Fischer rat, the amount of food ingested, the meal numbers, and meal sizes are the same in the dark and light cycles. ~.2

In the tumor-beating male Fischer rat, food intake remains stable until around Day 18 when food intake decreases as anorexia develops in the cancer-bearing rats. From Day 10 onward to Day 18, meal number gradually decreases in a linear fashion. Initially, a compensatory increase in meal size occurs that fails to compensate by Day 18, and, consequently, there is a sharp decrease in meal size with a resulting decrease in food intake. 3

When using the indomethac in i le i t is rat model , 4 the subcutaneous injection of indomethacin leads to a decrease in food intake brought about by a reduction in both meal number and meal size. This decrease lasts for about 4 days, following which ileal mucosa begins to heal and food intake gradually normalizes.

In the last two experimental models, namely, cancer and indomethacin i leit is , key factors may be cytokines . When interleukin-1 ( IL-I) is continuously injected peripherally in physiological concentrations for a period of 4 days into a normal Fischer rat, food intake rapidly decreases, reaching its nadir by Day 2. This occurs by a rapid reduction in meal number followed after a 24-h delay by a reduction in meal-size. When IL-1 infusion is stopped, there is a gradual normalization of food intake, led by a normalization of meal number and then subsequently followed by a normalization of meal size. ~ When subphysiological doses of a combination of TNF-~t and IL-1 are infused into normal rats, a synergist ic effect is seen, underscor ing the synergy of the cytokines. 6 Although the innervated l iver modulates eating activity via the vagus, cytokines act independently of hepatic afferent fibers as vagotomy does not alter this response. 7

Another model for inducing anorexia is the infusion of nutrients. When total parenteral nutrition (TPN) is infused peripherally (glucose:fat:amino acids = 50:30:20), there is a rapid decrease in spontaneous food intake, s The reduction in food

intake persists for the duration of the nutrient infusion. The reduction is a graded response corresponding to the caloric density of the solution infused. When the parenterai nutrient infusion is stopped, there is a gradual increase of food intake, with normalization occurring about 3 days later. 9-H A similar effect can be seen when simple nutrients, such as triacetin, are used. 12 There is no difference in food intake reduction whether the solution is given intravenously or intragastrically. 13 The likely cause for the reduct ion in food intake is the p rov i s ion of exogenous calories, t4 However, recently TNF has been isolated in the blood of rats receiving TPN, which may be a contributory factor to the anorexia seen. ,5

Food intake is the product of meal number and meal size. In comparison to normal controls, in rats in whom the liver has been totally denervated, food intake is the same, but meal size is halved. To maintain normal food intake, a compensatory increase in meal number occurs. 16 A similar phenomenon occurs after olfactory bulbectomy. 17 Total food intake of bulbectomized and control rats is the same, but in the bulbectomized rats, meal size is halved and a compensatory doubling of meal number occurs in order to maintain food intake (Fig. 2).

In summary, in the above studies, the normal Fischer rat eats the ma jo r i ty of its food dur ing the dark cyc le . One can individually measure food intake, meal number, and meal size, and under physiological conditions it appears that meal number and meal s ize are r ec ip roca l ly changed . Howeve r , under pathological conditions, depending on the stimulus, a decrease in food intake may be accompl i shed via a decrease in meal number, Is meal size, m]7 or both to varying degrees, s Meal number appears to have the greatest variable and appears to be expressed in the dark cycle. These data show that meal number and meal size are regulated independently, suggesting that there

3

2

2

1

!

6 7 8 9 10

6 7 8 9 10

2.5

ZO

1.5

LO

0.5

0.0 6 7 8 9 10

FIG. 2. Observation was made after 5-day recovery from olfactory bulbectomy. Although meal size was significantly reduced in bulbectomized rats, normal food intake is maintained by a compensatory increase in meal number. Horizontal axis represents days.

GUT-BRAIN BRAIN-GUT AXIS $59

may be separate hypothalamic control areas for meal number and meal size. Based on further experimental data to be summarized below, we speculate that these regulatory sites for the control of meal number and size could be the ventromedial hypothalamus (VMH) and lateral hypothalamic (LHA) areas, respectively.

THE VAGUS AND ITS ROLE IN THE GUT-BRAIN, BRAIN-GUT AXIS

For nutrients to induce a reduction in food intake, it has been postulated that there are receptors in the hepatoportal area that sense nutrients2 9 These sensors send afferent impulses to the hypothalamus via the afferent fibers of the vagus, which terminate in the LHA, whereas sympathetic control is focused in the VMH. These two foci then influence the efferent loop to reduce food intake. Several electrophysiologicai studies have been performed in collaboration with Professor Akira Niijima (Department of Physiology, Niigata University, Japan) to provide lines of evidence in support of this general postulate. 2° In general, the degree and duration of depression of food intake correlates directly to the concentration of the nutrients in the solutions. Also, when individual amino acids were infused into the portal vein, a group of eight amino acids (alanine, arginine, histidine, leucine, lysine, serine, tryptophan, and valine) stimulated an increase in vagal afferent signals, whereas a group of seven amino acids (cysteine, glycine, isoleucine, methionine, phenylalanine, proline, and threonine) caused a decrease in vagal afferent activity. 2] This selective property of sensors in the portal system may provide information for the brain to regulate different kinds of food intake.

ROLE OF LHA AND VMH IN GUT-BRAIN AXIS: MICRODIALYSIS STUDIES

Using a microdialysis probe technique in the awake unrestricted Fischer rat, changes in LHA dopamine response during and after infusion of TPN, or its macroconstituents of glucose, fat, or amino acids, was measured. 22 Infusion of each of these nutrients led to a significant rise in LHA dopamine response, suggesting that dopamine is associated with a satiety signal. The decrease in dopamine response upon termination of the infusions varied between glucose and lipid.

When rats were fed an isocaloric diet either orally or intragastrically, a rise in LHA dopamine occurred only with oral feeding. ~ Thus, an oral nasal component of pregastric stimulation leads to a rise in LHA dopamine. When rats had an olfactory bulbectomy, meal size became half that of controls, and, concomitantly, LHA dopamine rose by one half that of controls (Fig. 3). Furthermore, when half the meal was given with an intact olfactory bulb, there was a corresponding decrease in LHA dopamine. Thus, a strong direct correlation exists between meal size and LHA dopamine rise. 24

- - 140

.~ 120

]

9 40

i

m ~

CO~O# J i , J Ouftm¢lon~4J~J nd ~ ell ~ t i t

'14

'!2

,lO g

n '-..

6 6

FIG. 3. Meal size is significantly lower in food-deprived olfactory bulbectomized rats. The elevation of LHA-dopamine release is also significantly lower in bulbectomized rats than in controls.

MEAL

~ o r o - B a s a ]

~,malJ

FIG. 4. Schema of dopaminergic feedback loop in regulation of eating. When eating starts, oronasal pregastric stimulation increases LHA- dopamine, which, in turn, stimulates the dorsal motor nucleus of the vagus (DMNX) and facilitates gastric compliance to receive food. The nutrients absorbed from the gastrointestinal tract via the liver act as inhibitory postgastric effects on the LHA to decrease dopamine release so as to control meal size.

LHA DOPAMINE, GASTROINTESTINAL MOTILITY AND THE BRAIN-GUT AXIS

The LHA is connected to the dorsal motor nucleus of the vagus. It would seem that an increase in LHA dopamine leads to an increase in gastric compliance and a decrease in contraction. On eating, the oral nasal pregastric stimulant is associated with an increase in LHA dopamine and an increase in gastric compfiance. Thus, LHA dopamine impacts meal size via modulation of gastric compliance. When LHA dopamine release was measured in rats after total liver denervation in response to TPN, TPN + food, and food alone, 25 a significant increase in LHA dopamine occurred in rats receiving TPN and in rats receiving TPN + food. No difference in the LHA dopamine response occurred between control and total liver denervated rats when only food was given. These data support the concepts of (1) oral nasal stimulation in LHA dopamine release, and (2) hepatic inhibitory role in LHA doparaine release.

From these data, the dopaminergic feedback loop in the regulation of eating can be constructed as shown in Fig. 4. This shows that when a meal is consumed, oronasal pregastric stimulation leads to an increase in LHA dopamine, which stimulates the dorsal motor nucleus of the vagus and facilitates gastric compliance to receive a meal. The nutrients absorbed by the liver act as inhibitory postgastric effects on LHA to decrease dopamine secretion so as to control meal size. The operation of this feedback loop has been demonstrated using lean and obese Zucker rats 26 as well as by implanting harvested fetal cells known to secrete dopamine into bilateral LHA of normal rats. 27

From these experiments, we can conclude that the mechanistic insight based on LHA dopamine changes in rat are that: (1) oronasal stimulation have a positive stimulatory effect on LHA; (2) chronic increase in dopamine in bilateral LHA is associated with increase in daily food intake and, thus, an increase in body weight; and (3) the liver has a negative stimulatory LHA dopaminergic effect.

THE LHA/VMH RECIPROCAL RELATIONSHIP AND THE ROLE OF VMH DOPAMINE

Data from the literature show that there is a reciprocity of LHA and VMH in food intake regulation as documented by Oomura et al. 2s VMH lesions lead to hypophagia associated with dopaminergic system. 29 A role has been suggested for dopamine in

S60 GUT-BRAIN BRAIN-GUT AXIS

GI

orol~assl stop ~ BATING ~ u n

.

~ Imlm~tlon

I,a,,od~**u~,-I

T dmlecyztoklntn

modifies G1

mot~ty

FIG. 5. Schema of possible role of hypothalamic dopaminergic system in regulation of food intake. Eating induces a reciprocal change in hypothalamic dopamine levels. LHA-dopamine has a positive stimulatory affect on food intake by modulating meal size via changing gastric compliance, whereas LHA- and VMH-dopamine inhibit each other's activity. VMH-dopamine modulates meal number via altering intermeal intervals.

BLOOD II CNS BLOOD BRAIN

BARRIER

~PlasmaTRP

Plasma LNAA S (BCAA+Phe+Tyr+Met)

Hypothalamus

ANOREXIA

FIG. 6. Schema of serotonergic hypothesis of cancer anorexia. Tumor growth induces a high plasma tryptophan (TRP) level, which competes with plasma large neutral amino acids (LNAA) to cross the blood-brain barrier. Increased TRP in the cerebrospinal fluid (CSF) accelerates the synthesis of serotonin (5-HT) in the hypothalamus. The increased hypothalamic 5-HT level causes anorexia.

the VMH in the control of food intake, 3° and the VMH inhibits I2-1A activity. 31 An increase in serotonin in LHA and a decrease in serotonin in VMH is associated with regulation of food intake. 32 Eating increases LHA dopamine and correlates to meal size. ~

From these data we can postulate that dopamine in the VMH should play a contributory role in the feedback mechanism to control food intake, specifically, meal number. To test this hypothesis, dopamine microdialysis studies were performed which revealed that the concentrations of dopamine in the VMH rapidly decreased after eating and remained depressed for at least 120 min after eating had completed. 33 When one half the meal size was offered, a rapid decrease in dopamine occurred, which persisted for 60 min. This suggests a direct relationship between the decrease in VMH dopamine and the amount of food ingested. Similarly, when food was given orally or intragastrically, a decrease in VMH dopamine occurred with oral feeding, whereas none occurred with intragastric feeding. Thus, the oronasal pregastric stimulus decrease VMH dopamine. In summary, as shown in Fig. 5, the LHA via the dorsomotor nucleus of the vagus affected meal size, whereas the VMH influenced meal number-- both mediated via dopamine.

BRAIN-GUT AXIS: DOES SEROTONIN PLAY A ROLE?

Figure 6 shows the serotonergic hypothesis of cancer anorexia, as summarized by Cangiano and Rossi Fanelli. 34 This hypothesis is supported by a series of experiments performed in our laboratory in collaboration with Professor Filippo Rossi Fanelli's laboratory (Department of Medicine, University "La Sapienza," Rome, Italy). First is the finding in tumor-bearing Fischer rats of a close negative correlation between free tryptophan and food intake. 35 The higher the plasma free tryptophan levels the greater the degree of anorexia. In contrast, there is a direct positive relationship between free tryptophan and brain serotonin concentration in tumor-bearing rats and a direct correlation between the ratio of free tryptophan to large neutral amino acids and serotonin in the brain of rats. 35 More recently, we have established a correlation between food intake and CSF IL-I in anorectic tumor-bearing rats. 36 Data also exist in the literature, which shows that the peripheral infusion of IL-I is closely linked to brain 5-hydroxytryptophan. 37 Of interest, IL-1 leads to an increase in tryptophan competitive crossing of the blood-brain

barrier. 3s This leads to an increased serotonin synthesis in the brain. 39 IL-1 crosses the blood-brain barrier directly 4° and enters the VMH. 4~ Interleukin-I increases serotonin release in the brain, 42 and IL-1 modulates neuronal activity in the VMH via calcium channel. 43 Thus, a cytokine serotonergic interaction can be hypothesized in the multifactoriai etiology for cancer anorexia.

THE BRAIN--GUT AXIS IN CANCER ANOREXIA: ROLE OF LHA AND VMA

Postulating that the regulatory sites for the control of meal number and size could be the VMH and LHA, respectively, we investigated the role of VMH and LHA in mediating cancer anorexia in tumor-bearing (TB) and nontumor-bearing control rats by observing response of food intake, meal number, and size to temporary inhibition of VMH or LHA activity (Laviano et al., Cancer Research-l, submitted). When anorexia developed in tumor-bearing rats, food intake decreased initially via a selective reduction of meal number, with no change in meal size. Then both TB and nontumor-bearing rats had stereotaxically located bilateral intra-VMH (Exp. #1) or bilateral intra-LHA (Exp. #2) injections of colchicine (CX) or saline. After inhibition of VMH activity, in Experiment #1, food intake increased in both nontumor-bearing and TB rats by 60% via a selective increase of meal number for a period of 5 days, almost restoring food intake to normal in TB-CX rats. In contrast, in TB and control rats in Experiment #2, inhibition of LHA activity acutely reduced food intake to almost 0, via a selective reduction of meal size, resulting in TB rats starving to death. These data further confirm our postulate that meal size and number are controlled separately and independently, and that: (1) meal number and size are probably modulated by VMH and LHA, respectively; and (2) because cancer anorexia results from the reduction of meal number, impairment of VMH activity during tumor growth may constitute an early step leading to reduced food intake.

We tested the effects of intra-VMH injection of mianserin (MIA), dopamine, or IL-I receptor antagonist (IL-lra) on food intake, meal number, and size in anorectic MCA sarcoma bearing (TB) rats to determine their role in the pathogenesis of tumor- related anorexia (Laviano et al., unpublished observations). Nontumor-bearing rats received a stereotaxically located intra- VMH microinjection of an equal volume of the vehicle. Rats had

GUT-BRAIN BRAIN--GUT AXIS $61

ORGAN BLOOD m HYPOTHALAMUS BLOOD SP.A~

-"- ANOREXIA

t ~FOOD INTAKE

FIG. 7. Schema representing the conceptual relationship of IL-I, dopamine, and serotonin in centrally mediated early cancer anorexia. Tumor growth interferes with host immune system. Stimulating cytokines are released. Increased plasma IL-I elevates tryptophan (TRP) level in plasma. Both increased plasma IL-I and TRP cross the blood brain barrier to increase TRP concentration in cerebrospinal fluid (CSF). IL-1 directly acts on the hypothalamus, combined with high CSF TRP, enhancing serotonin (5-HT) synthesis. Increased ratio of 5-HT/dopamine in the VMH and decreased ratio of dopamine/5-HT in LHA cause anorexia.

their food intake, meal size, and number recorded during the study. Food intake in tumor-bearing rats improved following intra-VMH injection of MIA via a selective increase in meal number. In TB rats in jec ted with dopamine , meal number significantly increased after surgery, but was offset by a sharp dec l ine in meal size, so that food intake did not change. Following intra-VMH injection of either I L - 1 or the vehicle, nontumor-bearing rats temporarily decrease their food intake due to a predominant reduction of meal size. In TB-IL-Ira rats, food intake improved primarily via an increase of meal number 44 From these data, we deduce that in the TB rat intra-VMH serotonin, IL- l, and to a lesser extent dopamine, play a role in development of cancer anorexia.

Food intake, meal number, meal size in TB and nontumor- bearing rats was observed after blocking impaired VMH-LHA activity from the effects of cancer. Early anorexia in tumor-

bearing rats leads to a decrease in food intake caused by a decrease in meal number with no change in meal size. Progressive anorexia leads to an additional decrease in meal size. When bilateral intra-VMH or LHA colchicine is given, VMH block leads to 60% increase in food intake in nontumor-bearing rats and TB rats via an increase in meal number, restoring food intake to normal in TB cachectic rats. VMH did not influence meal size. LHA block leads to acute decrease of food intake to almost zero via a reduction in meal size in TB rats starved to death as there is no compensatory increase in meal number. The data suggest that meal number and meal size are controlled by VMH and LHA, respectively. VMH (meal number) has the primary role over LHA (meal size), and cancer anorexia results from a decrease in meal number.

Based on these preliminary data, VMH impairment during tumor growth appears to be an early step that leads to a decrease in food intake. Thus, as summarized in Fig. 7, the regulation of food intake and the anorexia associated with cancer is, as would be anticipated, quite complex. We hypothesize that it primarily involves the LHA and the VMH and possibly other brain food regulatory areas yet to be investigated. In addition to cytokines represented by IL-I , tumor necrosis factor and IL-8, induce anorexia, which appears to be the net effect of the ratio of dopamine to serotonin in the LHA and VMH. This complex postulate is, however, testable and is the thrust of future studies in this laboratory, in conjunction with our close collaborators and colleagues in Italy and Japan. Ultimately, the goal of our joint research efforts is to obtain a bet ter unders tanding of the mechanisms that control food intake in order to develop a rational treatment for obesity and the anorexia associated with acute and chronic disease.

ACKNOWLEDGEMENTS

Work presented was supported in part by Grant DK43796 from the National Institutes of Diabetes and Digestive and Kidney Diseases, NIH; by Grant 85A95 from the American Institute for Cancer Research; Whitehall Foundation Grant #J90-39; a grant from the Hendricks Fund for Medical Research; and Grant CRG 940731 from the NATO Internat ional Sc ien t i f ic Exchange Programmes.

A heartfelt thank you to all our collaborators who are too many to list as coauthors of this summary but whose names appear listed in the bibliography. Without their collaborative efforts our progress would have been slower and not as interesting.

REFERENCES

I. Meguid MM, Kawashima Y, Campos ACL, Gelling E Hill TW, Chen T-Y, Hitch DC, Mueller WJ. Automated computerized rat eater meter: Description and application. Physiol Behav 1990;48:759

2. Yang Z-J, Meguid MM. LHA-dopaminergic activity in obese and lean Zucker rats. NeuroReport 1995;6:1191

3. Meguid MM, Muscaritoli M, Beverly JL, Zang Z-J, Cangiano C, Rossi Fanelli E The early cancer anorexia paradigm: Changes in plasma free tryptophan and feeding indexes. JPEN 1992; 16:56S

4. Opara E, Veerabagu M, Yang Z-J, Koseki M, Meguid MM. The enigma of anorexia in inflammatory bowel disease. Abstract presented at the 35th annual meeting of the American College of Nutrition, October 7-9, 1994

5. Debonis D, Meguid MM, Laviano A, Yang Z-J, Gleason JR. Temporal changes in 24-hr meal number and meal size relationship in response to rHu IL-l. NeuroReport 1995;6:1752

6. Yang Z-J, Koseki M, Meguid MM, Gleason JR, Debonis D. Synergistic effects of TNF and intedeukin-1 in inducing anorexia in rats. Am J Physiol 1994;267:R1056

7. Laviano A, Yang Z-J, Meguid MM, Koseki M, Beverly JL. Hepatic vagus does not mediate 1L-I induced anorexia. NeuroReport 1995;6:1394

8. Meguid MM, Chen T-Y, Yang Z-J, Campos AC, Hitch DC, Gleason JR. Effects of continuous graded TPN on feeding indices and metabolic concomitants in rats. Am J Physiol 1991 ;260:E126

9. Beverly JL, Meguid MM, Yang Z-J, Yue M-X, Fetterman BL. Metabolic influences on satiety in rats receiving parenteral nutrition. Am J Physiol 1994;266:R381

10. Beverly JL, Yang Z-J, Meguid MM. Hepatic vagotomy effects on metabolic challenges during parenteral nutrition in rats. Am J Physiol 1994;266:R646

11. Beverly JL, Yang Z-J, Meguid MM. Factors influencing compensatory feeding during parenteral nutrition in rats. Am J Physiol 1994;266:R1928

12, Bodoky GM, Meguid MM, Yang Z-J, Laviano A. Effects of different types of isocaloric parenteral nutrients on food intake and metabolic concomitants. Physiol Behav 1995;58:75

13. Giner M, Meguid MM. Effect of intravenous and intragastric nutrients on food intake in rats. J Surg Res 1991;51:259

14. Meguid RA, Beverly JL, Meguid MM. Surfeit calories during total parenteral nutrition influences food intake and carcass adiposity in rats. Physiol Behav 1995;57:265

15. Opara El, Meguid MM, Yang Z-J, Chai J-K, Veerabagu M. Tumor

$62 GUT-BRAIN BRAIN-GUT AXIS

necrosis factqr-ct and total parenteral nutrition-induced anorexia. Surgery 1995;118:756

16. Ratto C, Gleason JR, Yang Z-J, Bellantone R, Crtaeitti F, Meguid MM. Change in meal size, number and duration after neural isolation of liver and with TPN. Physiol Behav 1991 ;50:607

17. Meguid MM, Gleason JR, Yang Z-J. Olfactory bulbectomy in rats modulates feeding pattern but not total food intake. Physiol Behav 1993;54:471

18. Kurzer M, Janiszewski J, Meguid MM. Amino acid profiles in tumor bearing and nontumor bearing malnourished rats. Cancer 1988;62:1492

19. Russek M. Hypothesis on the participation of hepatic glucoreceptors in the control of intake of food. Nature (London) 1963;200:176

20. Niijima A, Meguid MM. Parenteral nutrients in rat suppresses hepatic vagal afferent signals from portal vein to hypothalamus. Surgery 1994;116:294

21. Niijima A, Meguid MM. An electrophysiological study on amino acid sensors in the hepato-portal system in the rat. Obesity Res 1995;3:741S

22. Meguid MM, Yang Z-J, Montante A. Lateral hypothalamic dopaminergic neuron activity in response to TPN. Surgery 1993; 114:400

23. Yang Z-J, Koseki M, Meguid MM, Laviano A. Eating-related increase in dopamine concentration in the LHA with oronasal stimulation. Am J Physiol (in press).

24. Meguid MM, Yang Z-J, Koseki M. Eating induced rise in LHA- dopamine correlates with meal size in normal and bulbectomized rats. Brain Res Bull 1995;36:487

25. Yang Z-J, Oler A, Meguid MM, Gleason JR. Liver plays an inhibitory role on hypothalamic dopamine release during eating and TPN. Surg Forum 1994;45:166

26. Yang Z-J, Meguid MM. LHA-dopaminergic activity in obese and lean Zucker rats. NeuroReport 1995;6:1191

27. Yang Z-J, Meguid MM, Koseki M, Oler A, Chong C, Boyd J. Increase in food intake and body weight gain after bilateral hypothalamic fetal-dopaminergic cell transplantation, NeuroReport (in press).

28. Oomura Y, Ooyama H, Yamamoto T, Naka F. Reciprocal relationship of the lateral and ventromedial hypothalamus in the regulation of food intake. Physiol Behav 1967;2:97

29. Najam M. Involvement of dopaminergic systems in the ventromedial hypothalamic hyperphagia. Brain Res Bull 1988;21:571

30. Robert J J, Oroseo M, Rouch C, et al. Opposite dopaminergic activity in lateral and median hypothalamic nuclei in relation to the feeding effect of D-Ser2-Leu-Enk-Thr6 (DSLET). Brain Res 1990;510:7

31. VanAtta L, Sutin J. The response of single lateral hypothalamic neurons to ventromedial nuclei and limbic stimulation. Physiol Behav 1971 ;6:523

32. Hoebel BG, Hemandez L, Schwartz DH, et al. Microdialysis studies of brain norepinephrine, serotonin, and dopamine release during ingestive behavior: Theoretical and clinical implications. Ann NY Acad Sci 1989;575:171

33. Meguid MM, Yang Z-J, Laviano A, Oler A. Post-meal satiety is associated with eating-induced persistently low ventromedial hypothalamic dopamine in rats. NeuroReport, March 1996

34. Cangiano C, Rossi Fanelli E Increased availability of tryptophan in brain as a common pathogenic mechanism for anorexia associated with different diseases. Nutrition 1991 ;7:364

35. Muscaritoli M, Meguid MM, Beverly JL, Yang Z-J, Cangiano C, Rossi Fanelli E Mechanism of early tumor anorexia. J Surg Res, 1996 (in press)

36. Opara El, Laviano A, Meguid MM. Correlation between food intake and CSF-ILI in anorectic tumor-bearing rats. NeuroReport 1995;6:750

37. Mohankumar PS, Thyagarajan S, Quadri SK. Interleukin-ll~ increases 5-hydroxyindoleacetic acid release in the hypothalamus in vivo. Brain Res Bull 1993;31:745

38. Takao Y, Kamisaki Y, Itoh T. Beta-adrenergic regulation of amine precursor amino acid transport across the blood brain barrier. Eur J Pharmacol 1992;215:245

39. Dunn AJ. Endotoxin-induced activation of cerebral catechlamine and serotonin metabolism: Comparison with interleukin-l. J Pharm Exp Therap 1992;261:964

40. Banks WA, Ortiz L, Plotkin SR, Kastin AJ. Human interleukin-1, murine IL- 1, and murine IL- 1 ~ are transported from blood to brain in the mouse by a shared saturable mechanism. J Pharmacol Exp Therap 1991 ;259:988

41. Benveniste EN. Inflammatory cytokines within the central nervous system: Sources, function, and mechanism of action. Am J Physiol 1992;263:C1

42. Shintani F, Kanba S, Nakaki T, et al. Intedeukin-ll3 augments release of norepinephrine, dopamine, and serotonin in the rat anterior hypothalmus. J Neurosci 1993; 13:3574

43. Plata-Salaman CR, Ffrench-Mullen JMH. Interleukin-ll3 depresses calcium currents in CA1 hippocampa] neurons at pathophysiological concentrations. Brain Res Bull 1992;29:221

44. Laviano A, Meguid MM, Renvyle T, Opara El, Yang Z-J, Rossi Fanelli E IL-I acts directly on hypothalamic VMH to mediate cancer anorexia. Surg Forum 1995;46:491