The Effect of Insulin, Glucagon and Prolactin on Lipid Synthesis and Related Metabolic Activity in Migratory and Non-migratory Finches.

-1. The effects of insulin, glucagon and prolactin on lipid synthesis were investigated in house sparrows (Passer domesticus) and white-crowned sparrows (Zonotrichia leucophrys gambelii). 2. The incorporation of glucose-U-C 14 and acetate-l-C 14 into abdominal fat pad lipid was measured in in vitro and in vivo experiments. 3. None of these hormones accelerated lipid synthesis, even though lipogenesis is sensitive to changes in carbohydrate metabolism (i.e. glucose uptake) in this tissue. 4. Insulin, a potent stimulator of lipid synthesis in mammalian adipose tissue, had no effect on the metabolism of finch fat pads. 5. Glucagon and prolactin inhibited the synthesis of fatty acids from acetate. Only glucagon has the necessary potency to be of physiological significance. 6. It is hypothesized that a relatively high glucagon output inhibits lipid synthesis during non-migratory periods. During migratory periods the output of glucagon falls, allowing lipid synthesis to proceed at a more rapid rate. INTRODUCTION MIGHATORY birds manifest at least three different physiological responses under the influences of long spring daylengths: (1) gonadal recrudescence, (2) migratory behavior, and (3) lipid deposition. The first of these, gonadal recrudescence, is caused by increased secretion of gonadotrophic hormones by the anterior pituitary (for review, see Marshall, 1961). Migratory behavior and lipid deposition are independent of the gonadal response (Lofts & Marshall, 1961 ; Morton & Mewaldt, 1962) and of one another (King & Farner, 1963). The immediate physiological changes which lead to these latter responses are unknown. Premigratory accumulation of fat occurs very rapidly, often within a period of 5-10 days, and may amount to 25 per cent or more of body weight (for details, see Farner, 1960; Odum, 1960; Wolfson, 1960). Two hypotheses have been proposed whereby long daylengths might cause increased lipid synthesis. First, * Present address: Department of Biological Chemistry, Harvard University School of ]Medicine, Boston 15, Massachusetts. 2 ALAN G. GOODRIDGE hyperphagia might be induced by stimulation of a "hunger center" or inhibition of a "satiety center" in the hypothalamus (Brobeck, 1955 ; Soulairac, 1958). Food eaten in excess of daily requirements would be stored as fat and glycogen without a qualitative change in intermediary metabolism. Farner et al. (1961) have shown, however, that liver and muscle glycogen decrease during photostimulated lipid deposition. This observation implies a qualitative change in intermediary metabolism favoring lipogenesis. Therefore, an alternative hypothesis is that the photostimulated increase in the rate of lipid synthesis occurs as a result of a hormone-mediated change in intermediary metabolism. Increased utilization of substrate for lipid synthesis would stimulate the appetite center in the hypothalamus (Mayer, 1955) and hyperphagia would follow. The purposes of this investigation were (1) to determine the effect of hormones on lipid synthesis in adipose tissue and liver of passerine birds and, if possible, to relate these effects to photostimulated lipid deposition, and (2) to compare the hormone effects observed in mammals (for reviews, see Langdon, 1960 ; Winegrad, 1962) to those observed in passerine birds. Insulin, glucagon and prolactin were the principal hormones tested in this investigation. Insulin was thoroughly tested because it accelerates lipid synthesis in mammals (Langdon, 1960). Glucagon was tested because a preliminary experiment showed that it was a powerful inhibitor of lipid synthesis. Lipogenesis in the isolated rat epididymal fat pad also is inhibited by glucagon (Orth et al., 1960). Prolactin was tested because it accelerates lipid synthesis in rat adipose tissue (Winegrad et al., 1959) and because it causes fat deposition in white-crowned sparrows (Meier, personal communication). MATERIALS AND METHODS The house sparrow (Passer domesticus), a non-migratory bird, was chosen for these studies because of its year-round abundance. Data from four experiments on the white-crowned sparrow (Zonotrichia leucophrys gambelii) and one experiment on the slate-colored junco (Junco hyemalis hyemalis), both migratory species, have been used for comparative purposes. One experiment was performed on SpragueDawley rats (100-180 g) and two on Holtzman rats (350-400 g). Care of animals House sparrows were captured in the vicinity of Ann Arbor, Michigan. They were maintained in 2 × 2 x 2 ft or 2 x 4 × 6 ft flight cages on a 14 hr daily photoperiod (except natural photoperiod from 24 July to 2 September 1963") at room temperature. The diet consisted of millet, cracked corn, chick-starter mash, grit, cuttlebone and mineralized salt, given ad libitum. Water, to which a small amount of ABDEC vitamins (Parke-Davis) had been added, was available at all times. A minimum of 5 days was allowed for the birds to adjust to the diet and photoperiod before using them for experiments. All rats were fed ad libitum on Ralston Purina rat chow. * The daylength was 14'5 hr on 24 July and decreased to 13 hr by 13 September. EFFECT OF I N S U L I N , GLUCAGON AND PROLACTIN ON L I P I D SYNTHESIS 3 The white-crowned sparrows were captured near Pullman, Washington, and shipped by air express to Ann Arbor. The slate-colored juncos were captured near Ann Arbor. The white-crowned sparrows and juncos were caged and maintained as noted above for the house sparrows except that they were kept on natural photoperiods. The experiment with juncos was performed in June after these birds had attained maximum premigratory obesity and had begun to lose their fat. The experiments with white-crowned sparrows were performed in October. All but a few birds had attained peak fall premigratory fatness and were in a static phase of obesity. In vitro procedures After decapitation the abdominal fat pad (ventral bilobed strip of mesenteric adipose tissue which runs caudad from the stomach and pancreas to the cloaca where it splits into two lobes which pass dorso-anteriorly along the body walls for about 1 cm), or epididymal fat pad when rats were used, was removed and cut into two pieces. One piece served as a control, the other as experimental. Each piece was weighed on a torsion balance to an accuracy of + 1 mg and placed in the incubation medium. Tissues were incubated for 3 hr at 40°C in Krebs Bicarbonate Buffer, pH 7.3-7.4 (Cohen, 1957), containing glucose at a concentration of 2 or 4 mg/ml. The gas phase was 95°/,o 02 and 5 °/,,o CO2. Acetate (sodium salt), when present, was at concentrations of 0-02 m M or 0.5 mM. Either glucose-UC 1~ or acetatel-C 14 (Nuclear Chicago Corporation) was added to the medium in tracer amounts in order to measure the rate of lipid synthesis. Hormones were dissolved directly in the buffer or added to the reaction vessels in 0-05 ml of saline. Total volume in the reaction vessels was 2 or 3 ml if the hormone was dissolved directly in the buffer and 2.05 or 3-05 ml if the hormone was added to reaction vessels in saline solution. Equal volumes of solvent without hormone ,sTere used in the control vessels. In one experiment bovine serum albumin at a concentration of 3 g/100 ml was added to the incubation medium to act as a free fatty acid acceptor (Reshef et al., 1958). The albumin (Nutritional Biochemical Corporation, Cohn Fraction V) was dissolved in bicarbonate buffer and dialyzed against the same buffer without albumin for 24 hr at 7°C with two changes. Free fatty acids (FFA)* released into the medium during the incubation were extracted by the method of Dole (1956). Aliquots of the heptane extract were evaporated to dryness and counted and titrated as indicated below. Non-incubated medium, containing albumin, was extracted similarly and served as a zero-time control. In vivo procedures After a blood sample was taken from the leg vein, the birds were injected intramuscularly with 0-1 cm 3 of a hormone or control solution and intraperitoneally * The following abbreviations are used in this report: CoA--coenzyme A; FFA--free fatty acid; glucose-U-Cl~--glucose uniformly labelled with carbonl4; NADPH--reduced nicotinamide adenine diphosphonucleotide phosphate. 4 ALAN G. Q;OODRIUGE with about 2 tzc of acetate-l-C H (10-20/,moles). Next, they were placed in darkened metabolism chambers for 2 hr. At the end of the incubation period a terminal blood sample was collected by decapitation. The abdominal fat pad and the liver were removed and analyzed as indicated below. Hormone preparations Insulin (Lilly, Iletin U-40), glucagon-free insulin (Lilly) and glucagon (Lilly, glucagon hydrochloride) were diluted with physiological saline before use. Ovine crystalline prolactin (Squibb, lot No. 53273-002R, 22"5 IU/mg) was dissolved directly in the buffer. Extraction of fatty acids Fat pads from the in vitro experiments were blotted carefully and placed in 10°j~ alcoholic KOH. Fat pads from in vivo experiments were placed directly into the alcoholic KOH. After saponification was complete (e.g. 1 hr at 100°C), the mixture was acidified with 3 N HC1 to a thymol blue endpoint and then extracted with 5, 10 or 20 ml of petroleum ether depending on the size of the tissue. The resultant extract of fatty acids was then washed with distilled water. One aliquot of this extract was dried on a planchet and the radioactivity counted (standard error: _+ 2 per cent) in a Nuclear Chicago gas-flow geiger counter equipped with a "micromil" window. All samples were corrected for self-absorption (Broda, 1960). A second aliquot was evaporated to dryness in a 10 ml Erlenmeyer flask. The residue was dissolved in 95°'0 ethanol and titrated with 0"02 N NaOH to a phenophthalein-thymol blue endpoint. Re-extraction of the acidified saponification mixture showed that 97 per cent or more of the fatty acids were removed during the first petroleum ether extraction. Liver pieces from the in vivo experiments were homogenized in 1 : 1 chloroform-methanol and centrifuged. The supernatant was evaporated to dryness and saponified (1 hr at 100°C) with 10°~ alcoholic KOH. The alkaline saponified mixture was extracted once with a large volume (50 ml or more) of petroleum ether to remove the non-saponifiable lipids. This extract was discarded. The saponified mixture was then acidified with 3 N HC1 to a thymol blue endpoint and extracted with 5 ml of petroleum ether. This extract, which contained the fatty acids, was washed once with distilled water. Aliquots of the extract were counted or titrated as noted previously. A zero-time control was used to compensate for the slight contamination of the isolated fatty acids by glucose-U-C~L A weighed piece of adipose tissue was rinsed in iced medium (containing all components including the radioactive label) and transferred to alcoholic K O H without blotting. Saponification and extraction were performed as indicated above. Counts in the fatty acid fraction isolated from this tissue were subtracted from those isolated from the incubated tissues. Using a similar technique contamination from acetate-l-C H was found to be less than EFFECT OF INSULIN, GLUCAGON AND PROLACTIN ON LIPID SYNTHESIS 5 0"1 per cent of the total counts recovered from the fatty acid fraction of the incubated tissues. It was disregarded in experiments employing acetate. Glucose determinations Glucose concentration was measured in aliquots of the incubated and unincubated medium in order to determine glucose uptake by the isolated tissues. The medium was diluted 1 : 100 with distilled water and glucose determinations performed with glucose oxidase (Teller, 1956) (Glucostat-Worthington Biochemical Corporation). For the in vivo experiments blood was collected from the leg vein into heparinized capillary tubes or over citrate after decapitation. The blood was deproteinized by Folin-Wu reagent (Folin & Wu, 1919), and glucose was determined on 0"5 or 1.0 ml of the supernatant using the anthrone reagent (Morris, 1948). Analysis of data In experiments utilizing paired abdominal fat pads in vitro the data were tested for significance by the Wilcoxon Matched Pairs Test (Siegel, 1956). The in vivo data were tested for significance by the Mann-Whitney Test (Siegel, 1956). RESULTS Effect of insulin in vitro Insulin has been tested for its effect on glucose uptake, fatty acid and triglyceride synthesis from C1Mabeled glucose, and fatty acid synthesis* from C14-1abeled acetate in isolated avian adipose tissue (Tables 1-11). Doses used varied from 1.7 mU to 0"5 U per ml of medium, encompassing and exceeding the range of effective doses for the rat epididymal fat pad (Doisy, 1963). The major results are as follows: 1. Insulin had no detectable effect upon fatty acid synthesis from labeled glucose in the abdominal fat pad (p > 0.05) (Table 1). 2. The synthesis of triglyceride likewise is not affected (p > 0.05) by insulin (Table 2). Since the synthesis of fatty acids is not influenced by insulin (Table 1), the lack of insulin effect on triglyceride synthesis indicates that the hormone does not affect glyceride-glycerol synthesis. 3. The effect of insulin on fatty acid synthesis from labeled glucose also was tested in furcular adipose tissue, a subcutaneous fat body located near the clavicle (Table 3). The incorporation of glucose carbon into fatty acids of both minced and intact furcular fat was equally unaffected by insulin (p > 0.05). The rate of incorporation of labeled glucose into fatty acids of the abdominal and furcular fat pads is very slow. The radioactivity isolated in the fatty acid fraction was considerably less than twice the background level. For this reason subsequent experiments were conducted using a small amount of acetate-l-C 14 as the label and non-labeled glucose as substrate. * The terms fatty acid synthesis and triglyceride synthesis as used in this paper mean the rate of incorporation of labeled glucose or labeled acetate into fatty acid or triglyceride. 6 A L A N G . G O O D R I D G E 4. Table 4 shows the effect of insulin on the incorporation of acetate into lipid. Insulin had no effect on fat pads taken from untreated birds (p > 0-05) (Table 4(a)). The insensitivity of bird fat pads to insulin resembles the greatly diminished ~]'ABLE 1 T H E EFFECT OF I N S U L I N ON INCORPORATION* OF GLUCOSE CARBON INTO LIPID IN THE ISOLATED ABDOMINAL FAT PAD OF THE HOUSE SPARROW Insulin 0.3 U/ml Exptl. as % With Insulin W'ithout Insulin control 78"9 130 61 107 84.5 127 68.0 60.9 112 49.6 43-6 114 246 310 79 26.5 23-1 115 10.3 5-8 178