Correlation of changes in brain indoleamine metabolism with onset of anorexia in rats

Correlation of Changes in Brain lndoleamine Metabolism With Onset of Anorexia in Rats

Maarten von Meyenfeldt, MD, Cincinnati, Ohio William T. Chance, PhD, Cincinnati, Ohio Josef E. Fischer, MD, Cincinnati, Ohio

Anorexia is a common feature of neoplastic disease [I] that contributes to the depletion of lean body tissue of the host (cachexia). Since both chemotherapy and radiation further compromise the maintenance of proper nutrition, use of these therapeutic measures is limited in the cachectic patient [2]. In addition, the prognosis for recovery after surgical resection of the tumor in such a patient is compromised. Although several variables such as physical obstruction of oral intake, malabsorption of nutrients, nausea and changes in taste thresholds undoubtedly influence caloric intake [2,3], the major abnormality in the tumor-bearing individual is decreased desire for food in the presence of increased demand [I 1. Thus, the additional metabolic burden imposed on the host by the tumor is not met by increased intake. This paradoxical situation suggests that cancer anorexia results from a dysfunction of brain mechanisms that normally mediate hunger and satiety. Although many hypotheses of central nervous system mechanisms failing to respond appropriately to hunger cues or responding to spurious satiety signals generated by the tumor have been suggested [I], few changes in central nervous system biochemistry that are not secondary to malnutrition have been observed. Since cancer anorexia can be demonstrated in laboratory rats [4], the effect of a tumor burden on central nervous system biochemistry may be easily investigated. Considerable evidence has accumulated from a variety of experiments [5,6] suggesting that brain serotonin activity varies inversely with food intake. Thus, increased brain serotonin has been associated with anorexia, while decreased levels of this putative neurotransmitter have been correlated with hyperphagia. Initial investigations from this laboratory [ 7,8] have demonstrated elevated brain From the Department of Surgery, University of Cincinnati Medical Center, Cincinnati, Ohio. Supported by Grant CA25786 from the U.S. Public Health Service, National Cancer Institute, National Institutes of Health, Bethesda, Maryland. Requests for reprints should be addressed to Josef E. Fischer, MD, Department of Surgery. 231 Bethesda Avenue. Cincinnati. Ohio 45267. Presented at the 22nd Annual Meeting of the Society for Surgery of the Alimentary Tract, New York, New York, May 19-20, 1981.

Volume 143, January 1982

levels of both the serotonin precursor, tryptophan, and metabolite, 5-hydroxyindoleacetic acid in immature anorectic female rats bearing intramuscular Walker 256 carcinosarcomas. This change in brain indole activity apparently resulted from decreased plasma levels of albumin and elevated levels of nonesterified free fatty acids in tumor-bearing rats [7]. Thus, although plasma total tryptophan was decreased, the free (unbound) fraction was significantly increased in tumor-bearing rats, resulting in increased transport of tryptophan across the bloodbrain barrier [9] and elevated turnover of serotonin. Therefore, the purposes of the experiments described herein are (1) to investigate the generality of this phenomenon by assaying central nervous system indoles in another anorexia-producing tumor lice, (2) to inveitigate the correlational relation of elevated central nervous system indole activity to the onset and development of anorexia, and (3) to investigate regional central nervous system changes in indole activity associated with cancer anorexia. Material and Methods Generality of elevated serotonin in cancer anorexia: To allow investigation of changes in central nervous system serotonin activity associated with anorexia produced by a different tumor line, the methylcholanthrene-induced sarcoma was obtained in situ from the National Cancer Institute (Bethesda, Maryland) and transplanted as a cell suspension. Methylcholanthrene cells were injected (2 X 10” cells, 0.2 cm3) subcutaneously in 16 adult (200 to 250 g) male Fisher 344 rats (Charles River, Wilmington, Massachusetts). To allow separation of the biochemical effects of malnutrition from those that are tumor-specific, saline-injected (0.2 cm3) pair-fed as well as freely feeding control groups were employed. Thus, while the freely feeding group had access to food as desired, each rat of the pair-fed group received only that amount of food consumed by its weight-matched tumor-bearing pair during the previous day. All rats were provided with water as desired. Food intake and body weight were monitored for 33 days after tumor inoculation, at which time all rats were killed by decapitation for biochemical analyses. Blood was collected in heparinized tubes, with the brains rapidly removed and frozen on dry ice. All tumors were excised and weighed.

133

von Meyenfeldt et al

I

o-_o

TB ~16

17-_-o

FF n=6

I-

I

I

I

I

,

I

I

1

I

17

19

21

23

2;

27

29

31

33

M’S

POST TUMOR

IMFIANT

Developmental

changes in anorexia and serotonin: [7,8], injection of 5 X lo4 Walker 256 cells into the thigh muscle of immature female rats consistently elicited significant anorexia 6 days later and spontaneous death by day 10. Therefore, in the present experiment, groups of tumor-bearing (n = 8), pair-fed (n = 8) and freely feeding (n = 8) rats were killed 3,5,7 and 9 days after tumor inoculation. Thus, this experiment should allow assessment of changes in central nervous system indole activity before the onset of anorexia as well as during later stages of anorectic severity. The Walker 256 tumor was obtained as frozen ascites fluid from EG & G Mason Research (Worcester, Massachusetts), with tumor stock maintained by harvesting ascites fluid 5 days after the intraperitoneal injection of 1 X lo6 viable tumor cells. In this experiment, 5 X lo4 Walker 256 cells were injected (intramuscularly, 0.1 cm3) in 32 (60 to 80 g) female Sprague-Dawley rats (Harlan Laboratories, Madison, Wisconsin). Normal saline solution was administered (intramuscularly, 0.1 cm3) to equal numbers of pair-fed and freely feeding control rats. Food intake and body weight were monitored daily, with eight rata from each group killed 3,5,7 and 9 days after inoculation for biochemical analyses. Regional changes in brain serotonin associated with cancer anorexia: In order to determine which areas of the brain exhibit the greatest changes in indole activity assoIn previous experiments

TABLE I

Figure 1. Dally ad Ilb food Intake (mean f standard error of the mean) of tumor-bearing (7B) and freely feed@ (FF) rats after the Injection of 2 X 10’ methylcholanthrene-lnduced sarcoma cells or contmi lnjectkm of normal saline solutlon. BW = body welght.

ciated with cancer anorexia, we next investigated regional brain levels of tryptophan, serotonin and 5-hydroxyindoleacetic acid in tumor-bearing (n = 8), pair-fed (n = 8) and freely feeding (n = 8) rats. The tumors were induced as in the previous experiment, with all rats killed 9 days after the injection of 5 X lo4 viable Walker 256 cells. Blood was collected and the brains were rapidly dissected on ice, before freezing in liquid nitrogen, into the following areas: cortex, corpus striatum, hippocampus, hypothalamus, mesencephalon, pons-medulla, cerebellum and diencephalon-remaining telencephalon. These regional brain dissections were conducted similarly to reported methods [IO] with the following exceptions: (1) mesencephalon and diencephalon were differentiated by a vertical cut just anterior to the superior colliculus, and (2) a smaller hypothalamic area (mean weight 26.5 f 1.2 mg) was dissected, defined as tissue taken to a depth of approximately 2.5 mm, bounded anteriorly by the optic chiasma, posteriorly by the mamillary area and laterally by the choroid fissure. Biochemical analyses: Plasma albumin concentration was determined by the calorimetric method of Doumas et al [Ill. Plasma free and total tryptophan were assayed fluorometrically as described by Bloxam and Warren [12], with free tryptophan determined in 50 ~1 of an ultrafiltrate prepared by centrifugation (100 g, 25 minutes, 2O’C) of 1 ml of plasma (pH = 7.4) in a CF 50 Diaflo membrane cone (Amicon, Lexington, Massachusetts). Brain serotonin and

Biochemical Changes (mean f standard error of the mean) 33 Days After the Induction of Methylcholanthrene Sarcoma (Tumor-Bearing) or Various Control Manipulations (Pair-Fed and Freely Fed)

Plasma Albumin (g/100 ml) Total tryptophan (nmol/ml) Free tryptophan (nmol/ml) Brain Tryptophan &g/g) Serotonin @g/g) 5-HIAA (ng/g)

Tumor-Searing

Pair-Fed

Freely Fed

3.66 f 0.21 77.6 f 4.6 19.5 f 1.4

4.52 f 0.06” 100.1 f 7.9’ 14.6 f 0.5’

4.36 f 0.06’ 113.5 f 5.7’ 17.1 f 2.0

3.54 f 0.21 775 f 16 610 f 16

2.61 f 0.12+ 663 f 23+ 454 f 9+

2.41 f 0.15+ 646 f 19+ 416 f 6+

p <0.05 and + p
134

The American Journal of Surgery

Brain lndoleamine Metabolism and Anorexia

Figure 2. Chanps &I wmulative body weight (mean f sfandard error of the mean) in tumor-bear&f ( 7B) and pair-fed (PF) rats after the /nject/on of 2 X IOn methylchofanthrene-fnduced s8rcoma cells or normal

4-

: ft

I I

I 8

1

1

I

I

I

I

I

15

I8

21

22

23

24

25

DAYS POST

saline solution.

I

I

27

28

1 29

I

I

I

I

30

31

32

33

IMPLANT

fed rats from the onset of anorexia (Figure 2) emphasizes that the daily food consumption by the tumor-bearing group was not sufficient to support normal growth. The degree of cachexia in this tumor model is indicated by body weight differences on the day the rats were killed, Thus, although the mean body weight of tumor-bearing rats was 250 f 3 g, the mean tumor weight was 52 f 4 g, yielding a net body weight of tumor-bearing rats (198 f 5 g) that was less than pair-fed (215 f 7 g) or freely feeding (250 f 4 g) groups. Biochemical analyses (Table I) revealed similar changes in serotonin metabolism as observed in the Walker 256 model. Thus, plasma albumin and total tryptophan were significantly decreased, while plasma free tryptophan was significantly elevated in tumor-bearing rats. Braip levels of tryptophan, se-

5-hydroxyindoleacetic acid were also determined fluorometrically after acid-butanol extraction, according‘to the method of Curzon and Green [13], with serotonin and 5hydroxyindoleacetic acid complexed ,with OPT (Sigma Chemical, St. Louis) and read on an Aminco-Bowman fluorometer. Brain tryptophan was extracted in the serotonin fraction and assayed fluorometrically according to the procedures utilized for plasma tryptophan. Results Palpable tumors were observed 9 to 13 days after the injection of 2 X lo6 methylcholanthrene sarcoma cells. Food intake by the tumor-bearing rats was significantly (p <0.05) reduced by day 23, with anorexia becoming more severe as the tumor grew (Figure 1). The absence of body weight gain in pair-

TABLE II

TUMOR

1 26

Levels (mean f standard error of the mean) of Brain Tryptophan, Serotonin and 5-Hydroxylndoleacetic Acid (5-HIAA) in Tumor-Bearing (TB), Pair-Fed (PF) and Freely Fed (FF) Rats 3,5, 7 and 9 Days Afler lnjectlon of Walker 256 Carc!nosarcoma Cells Dav Animals Were Killed

Tryptophan(cLg/g) TB PF FF Serotonin (ng/g) TB PF FF 5-HIAA (nglg) TB PF FF

3.88 zk 0.21 4.41 f 0.14 3.83 zk 0.15

4.56 z!c0.20 4.69 f 0.37 3.81 f 0.26

644 f 12 682 f 21 649 f 12

707 f 650 f 629 f

628 f 631 f 589 f

694 f 21 649 f 21 587 f 18’

15 17 18

14 19+ 16’

4.73 Ik 0.20 3.49 f 0.17’ 3.79 f 0.14’

4.79 f 0.25 3:47 f 0.17’ 3.80 f 0.21’

604 f 20 623 f 15+ 627 4 14+

732 f 19 641 f 16’ 635 f 8’

773 f 586 f 590 f

793 f 17 592 f 9’ 588 f 15‘

17 18’ 12’

* p
Volume 143,

January 1982

135

von Meyenfeldt et al

0 T8 Off

14-

13-

'p
YS

b

‘*oc005vs

8

12-

$

II-

E B 0

IO-

TB TE

8 :

Q-

Shoturn

$ B t

8-

1600 -

I

0

I2

I

I 3

I

4

I 5

I 6

I 7

I 6

I

9

$ e

1200 -

k I;, 800

DAY6 POST TUMOR

-

IMPLANT

Flgure 3. Daily ad lib food intake (mean f standard error of the mean) of tumor-bearing ( TB) and free/y feedtng (FF) rats after the Injection of 5 X lo4 Walker 256 carcinosarcoma cells or normal saline solution. For each data point, n = 6. BW = body weight.

rotonin and 5-hydroxyindoleacetic acid were significantly (p
136

Htppocom e”C

Striotum

&as/hledCCrak”um

cwlcx

Gnler

Dwx

Figure 4. Levels of tryptophan, serotontn (5-HT) and 5-hydroxyktdoleacettc acid (5-HIAA) (mean f standard error of the mean) In various brain regfons of tumor-bearing ( TB), pair-fed ( W) and fieety feeding ( FF) rats. lhe rats were killed 9 days after the inject/on of 5 X lo4 Walker 256 carcinosarcoma cells. For each region Investigated in each group, n = 6.

bearing group than in the freely feeding rats, while serotonin levels were greater in the tumor-bearing group than in the pair-fed and freely feeding groups on days 5 (p <0.05), 7 (p CO.05) and 9 (p
The American

Journal

Of Surgery

Brain

TABLE III

lndoleamine

Metabolism

and Anorexia

Brain Regional Ratio (mean f standard error of the mean) of !%Hydroxyindoleacetic Acid to Serotonin in Tumor-Bearing, Pair-Fed and Freely Fed Rats 9 Days After the Injection of Walker 256 Carcinosarcoma Cells

-

_-

Tumor-Bearing

-..

0.733 0.558 0.585 0.396 0.663 0.866 0,484 0.765

Diencephalon Hypothalamus Mesencephalon Cerebellum Cortex Corpus striatum Pons-medulla Hippocampus l

p
t p <0.05

versus tumor-bearing

f f f f f f f f

0.023 0.029 0.048 0.029 0.043 0.035 0.061 0.047

0.661 0.419 0.487 0.316 0.621 0.866 0.381 0.740

f f f f f f f f

0.017’ 0.044’ 0.024 0.013’ 0.023 0.040 0.024 0.036

Freely 0.647 0.422 0.470 0.311 0.600 0.801 0.361 0.693

f f f f f f f f

Fed 0.018+ 0.030f 0.013’ 0.027 0.017 0.039 0.035 0.057

group.

(p
Volume 143, January 1992

Pair-Fed

serotonin activity associated with cancer anorexia appears to be a general phenomenon, being readily demonstrated within at least two anorexia-producing tumor lines. As demonstrated in the second experiment, the changes in central nervous system indole activity occur immediately before the onset of anorexia. Thus, rats bearing Walker 256 tumors exhibited increased brain serotonin and 5-hydroxyindoleacetic acid levels on day 5, while significant anorexia was not observed until day 6. Therefore, these data again support the hypothesis that the elevations in brain indole activity in tumor-bearing rats is not merely the result of reduced food intake. This early change in central nervous system indole activity is the first demonstration that biochemical alterations in putative neurotransmitters known to influence food intake precede the development of cancer anorexia. The analysis of regional brain changes in serotonin activity also demonstrated definite differences in tumor-bearing and control groups. Although elevations in brain tryptophan were widespread, significant increases in serotonin and 5-hydroxyindoleacetic acid were more localized. Thus, the diencephalon exhibited the most consistent increases in indole activity. The ratio of 5-hydroxyindoleacetic acid to serotonin was also significantly increased in the hypothalamus of tumor-bearing rats, however, suggesting greater turnover of the indoleamine within this region of the brain. Although previous investigations have suggested that the hypothalamus is normal in tumor-bearing rats [I], these conclusions were based on lesions of the ventromedial [15] or lateral [16] hypothalamic areas and the intact feeding response of tumor-bearing rats to exogenously administered insulin (I 71. Thus, these experiments do not represent the precision obtained by biochemical assay of brain regions. Perhaps a more logical interpretation of central nervous system regional control of feeding behavior is the consideration of brain neuronal circuits that interconnect the areas rather than dealing with discrete nuclei alone. Thus, several areas of the limbic system including the amygdala,

137

von Meyenfeldt et al

olfactory cortical areas, septal area and hippocampus have been reported to influence food intake [18]. Tumor-bearing rats exhibited significant increases in 5-hydroxyindoleacetic acid in regiqns -that included each of these areas. Furthermore, the hypothalamus has extensive neural connections with each of these brain areas. Therefore, it is possible that biochemical changes anywhere along the limbic circuit could influence feeding behavior in tumorbearing rats. Although these data demonstrate correlational changes in brain serotonin associated with cancer anorexia, they do not prove that increased indoles are the cause of’it. Recent data employing acute intraventricular injection of the a&agonist of serotonin synthesis, parachlorophenylalanine, suggest that the onset of anorexia could be delayed for 2 days in the Walker 256 model [19]. Therefore, increased brain serotonin activity may contribute to the anorectic response early in i& development. Further investigations are required to establish the role of serotonin in the later, more severe phases of anorexia. Should pharmacologic interventions with serotonin receptor antagonists or reduction of the transport of tryptophan into the brain prove successful in alleviating the anorectic response, a significant contribution to the overall therapy of cancer patients with anorexia and cachexia could be achieved.

Summary The association of elevated brain serotonergic activity with cancer a&orexia was investigated in three experiments. Significant increases in whole brain tryptophan, serotonin and 5-hydroxyindoleacetic acid in anorectic rats bearing methylcholanthrene sarcomas demonstrated the general nature of this phenomenon. In the Walker 256 sarcoma model, significant increases in whole brain serotonin and EChydroxyindoleacetic acid immediately preceded the onset of anorexia, suggesting that these changes in serotonin activity were not caused by decreased food intake. Investigations of regional brain changes in serotonin activity of rats bearing Walker 256 tumors revealed widespread increaSes in tryptophan, while serotonin was increased in the diencephalon and cerebellum and 5-hydroxyindoleacetic acid was increased in the cortex, hippocampus, diencephalon, pons-medulla and cerebellum. Therefore, these experiments demonstrate that increased brain serotonin activity is a general phenomenon, being observed in two lines of anorexiaproducing tumors, that it precedes the onset of anorexia and that selective increases in serotonin activity occur within different regions of the brain.

References 1. Morrison SD. Control of food intake in cancer cachexia: a challenge and a tool. Physiol Behav 1976;17:705-14. 2. Dewys SD. Anorexia as a general effect of cancer. Cancer 1979;43:2013-9.

138

3. Costa G. Cachexia, the metabolic component of neoplastic diseases. Cancer Res 1977;37:2327-35. 4. Mider GB, Tesluk J, Morton JJ. Effects of Walker carcinoma 256 on food intake, bodyweight and nitrogen metabolism of growing rats. Acta Unio Int Contra Cancrum 1948;6:40920. 5. Breisch ST, Zemlan FP, Hoebel BG. Hyperphagia and obesity following serotonin depletion by intraventricular p-chlorophenylalanine. Science 1976;192:382-4. 6. Blundell JE. Is there a role for serotonin (5hydroxytryptamine) in feeding? Int J Obes 1977;1:15-42. 7. Krause R, James JH, Ziparo V, Fischer JE. Brain tryptophan and the neoplastic anorexia-cachexia syndrome. Cancer 1979;44:1003-8. 8. Krause R, James JH, Humphrey C, Fischer JE. Plasma and brain amino acids in Walker 256 carcinosarcoma-bearing rats. Cancer Res 1979;39:3065-9. 9. Perez-Cruet J, Chase TN, Murphy DL. Dietary regulation of brain tryptophan metabolism by plasma ratio of free tryptophan and neutral amino acids in humans. Nature 1974;248: 693-5. 10. Glowinski J, lversen LL. Regional studies of catecholamines in the rat brain. J Neurochem 1966; 13:655-69. 11. Doumas BT, Watson WA, Biggs HJ. Albumin standards and the measurement of serum albumin with bromcresol green. Clin Chim Acta 1971;31:87-96. 12. Bloxam DL, Warren WH. Error in the determination of ttyptophan by the method of Dencla and Dewey: a revised procedure. Anal Biochem 1974;60:621-5. 13. Curzon G, Green AR. Rapid method for the determination of 5-hydroxyhyptamine and 5hydroxyindoleacetic acid in small regions of rat brain. Br J Pharmacol Chemother 1970;37: 689-97. 14. Hefti F, Melamed E, Wurtman RJ. Partial lesions of the dopaminergic nigrostriatal system in rat brain: biochemical characterization. Brain Res 1980;195: 123-37. 15. Baillie P, Millar FK, Pratt AW. Food and water intakes and Walker tumor growth in rats with hypothalamic lesions. Am J Physiol 1965;209:293-300. 16. Morrison SD. Effect of growth of a tumor on the regulation of water intake. J Natl Cancer lnst 1968;41:1241-8. 17. Morrison SD. Control of food intake during growth of a Walker 256 carcinosarcoma. Cancer Res 1973;33:526-8. 18. Grossman SP. Neurophysiologic aspects: extrahypothalamic factors in the regulation of food intake. Adv Psychosom Med 1972;7:49-72. 19. von Meyenfeldt M, Chance WT, James JH, Fischer JE. Investigation of serotonergic influence on cancer-induced anorexia. Proc Sot Neurosci 1980;6:528.

Discussion Arthur H. Aufses, Jr. (New York, NY): Have you looked at the effect of Periactin on results in the brain? Norman B. Ackerman (New York, NY): I wonder whether it is possible to modify this effect in the hypothalamus with destruction of the ventromedial nucleus, using something like gold thioglucose in these animals. You might be able to titrate the anorexia with a constant increased stimulation of appetite. Maarten von Meyenfeldt (closing): No, we didn’t look at that specific drug. We tried to lower brain tryptophan levels by giving diets enriched with branch chain amino acids, thus increasing the competition for transport of tryptophan into the brain. We were able to delay the onset of anorexia in animals by 24 to 48 hours. We also gave intraventricular injections of parachlorophenylalanine, which is a serotonin depleting agent, and we were able to delay the onset of anorexia by 2 days in this model.

The American Journal of Surgery