Increased in-vivo insulin sensitivity but normal liver insulin receptor kinase activity in dwarf chickens I. Gu\l=e'\ritault, J. P. M\l=e'\rat

Simon, B. Chevalier, M. Derouet, M. Tixier-Boichard and

INRA, Station de Recherches Avicoles, Nouzilly, 37380 Monnaie, France *INRA, Laboratoire de Génétique Factorielle, 78350 Jouy-en-Josas, France

(Requests for offprints should be addressed to J. Simon) revised manuscript received

13 November 1989

ABSTRACT

The effects of the recessive and sex-linked dw gene on insulin sensitivity and liver insulin receptors were compared in normal (Dw-dw) and dwarf (dw-dw) brother or half-brother chickens. At 3\m=.\5 weeks of age, following an overnight fast, exogenous insulin (0\p=n-\6\m=.\9 nmol/kg body weight) was slightly but significantly more hypoglycaemic in dwarf chickens. At 4 weeks of age, following an oral glucose load (2 g/kg), glucose tolerance was the same in both genotypes, whereas plasma insulin levels were greatly decreased in dwarf chickens. At 5 weeks of age, plasma concentrations of glucose and insulin were the same in both genotypes in the fasting state and decreased in the fed state in dwarf chickens. In liver membranes prepared from fasted chickens, insulin binding was increased in dwarf chickens, while the affinity of insulin receptors and the insulin-degrading activity of the membranes were the same in both genotypes. Following solubilization with Triton X-100, liver receptors were

tive of the genotype and the nutritional state. Therefore, the slight increase in insulin sensitivity observed in vivo in dwarf chickens is accounted for, at least partly, by a slight increase in liver insulin receptor number, but not by a change in the kinase activity of liver insulin receptors. In addition, post-insulin receptor kinase events and/or GH-dependent counter\x=req-\ regulatory mechanisms may superimpose and increase the insulin sensitivity of dwarf chickens. Journal ofEndocrinology (1990) 126, 67\p=n-\74

INTRODUCTION

or

The recessive and sex-linked dwarf (dw) gene in the chicken (Hutt, 1959; Mérat, 1969) causes a decrease in adult body weight and size of the long bones of the leg. Dwarf chickens also show decreased food efficiency and basal energy metabolism and, in young animals, increased hepatic lipogenesis and body fat content (Guillaume, 1969, 1976; Ricard, 1970; Ricard & Cochez, 1972; Touchburn, Guillaume, Leclercq &

Blum, 1980).

a preliminary study the effect of the dw gene on plasma glucose-insulin relationship was examined (J. Simon, unpublished data). Increases in plasma concentrations of insulin following oral glucose load

In

the

successively purified on lentil then wheat germ lectins. Autophosphorylation of the \g=b\-subunitdid not differ between either the genotype or the nutritional (fed or fasted) state. In the basal state (in the absence of insulin) the tyrosine kinase activity of the receptor towards artificial substrate poly(Glu,Tyr)4:1 was significantly decreased in dwarf chickens by fasting. However, the change in tyrosine kinase activity of the receptor in response to insulin was similar, irrespec-

refeeding were depressed in dw chickens (J. Simon, unpublished data). However, the control of plasma glucose appeared to be normal which suggests that

dwarf chickens are more sensitive to insulin. The present studies were therefore conducted to measure sensitivity to exogenous insulin, to examine glucose tolerance and the plasma glucose-insulin relationship further and to look for possible differences in number, affinity or intrinsic kinase activity of liver insulin receptors. These studies were focused on liver recep¬ tors since the liver is the main, if not the only, site for lipogenesis in birds, and chicken insulin receptors in the liver but not in the brain or muscle show changes in number and intrinsic tyrosine kinase activity in response to

prolonged fasting (Simon, Rosebrough,

McMurtry et al. 1986; Adamo, Simon, Rosebrough et al. 1987). Measurements were made either follow¬ ing overnight fasting, conditions which were also used

for the measurement of glucose tolerance, or in the fed state, using brother or half-brother dwarf (dw-dw) or normal (Dw-dw) chickens. MATERIALS AND METHODS

experiments were conducted. Fertilized

brown-egg type breed were obtained from the following cross: (DwS/dws) males (dwsj ) females. Eggs from two different flocks (312 and 350 eggs for experiments 1 and 2 respectively) were collected. At hatching, birds were sexed and only males kept. They were either hétérozygote normal (DwS/dws) or homozygote dwarf (dws/dws). The recessive coloration s gene linked to dw made it poss¬ ible to identify dw chickens by the colour of their feathers. Pairs of normal Dw or dwarf dw brothers or half-brothers were selected. Chickens were kept up to the age of 2 weeks in a conventional floor pen and then transferred to individual wire-floored cages. A balanced and conven¬ tional pelleted diet was available ad libitum (about 3040 kcal metabolizable energy/kg and 22-1% pro¬ tein), and the birds were kept under a photoperiod of 14 h light per day. —

Experimental conditions The sensitivity to exogenous insulin was measured at 3-5 weeks of age following an overnight fast. Highly purified porcine insulin (Endopancrine; Organon, St Denis, France) was injected i.m. into the leg at doses of 0-6-9 nmol/kg in 0-5 ml/kg. A blood sample was taken from a wing vein 90 min after the injection of insulin (n seven or eight chickens/dose). This period corresponds to the hypoglycaemic plateau in broiler chickens (Simon & Leclercq, 1985). At 4 weeks of age, glucose tolerance was measured after an overnight (about 16 h) fast, by using an oral 2 g glucose load/kg body weight (50% (w/v) glucose solution). One blood sample per chicken was obtained by heart puncture at the times indicated (n seven or eight different chickens each time). Plasma was stored at —20 °C following centrifugation. After blood sampling at 5 weeks of age, fed or (overnight) fasted chickens were killed by decapi¬ tation (n eight chickens per nutritional state and genotype in each experiment). Livers were quickly excised, weighed, frozen and ground in liquid nitrogen =

=

=

70 °C.

and then stored at

-

Monoiodinated A,4 human l25I-labelled insulin (about

(about 3000 Ci/ 2000Ci/mmol) and [ -32 ] Amersham France (Les from were purchased mmol) Ulis, France). Monocomponent porcine insulin was purchased from Novo Industrie Pharmaceutique (Paris, France). Chicken insulin was purchased from Litron Laboratories (Rochester, NY, U.S.A.). Bovine serum

Animals and diets Two identical eggs from a

Materials

albumin (BSA; fraction V, radioimmunoassay

grade), bacitracin, phenyhnethylsulphonyl fluoride (PMSF), leupeptin, aprotinin and the synthetic tyrosine-containing polypeptide poly(Glu,Tyr 4:1 were purchased from Sigma Chemical Co. (St Louis, MO, U.S.A.). Immobilized wheat germ agglutinin (glycoamino-silex; WGA) and lentil haemagglutinin were purchased from Laboratoires Miles (Paris, France). All other chemicals used were reagent or electrophoresis grade. Methods

Plasma concentrations of glucose were determined with a glucose analyser (Model 2, Beckman Instruments, Palo Alto, CA, U.S.A.). Plasma concentrations of insulin were measured by radioimmunoassay using guinea-pig anti-porcine insulin serum (Ab 27-6, a gift from G. Rosselin, Hôpital Saint-Antoine, Paris, France) with chicken insulin as standard as described previously (Simon, Freychet & Rosselin, 1974). Crude liver membranes were prepared by differential centrifugation (Havrankova, Roth & Brownstein, 1978). Livers were homogenized on ice in sodium bicarbonate solution (1 mmol/1) containing 2 mmol PMSF/1, 10 µg leupeptin/ml and 1 trypsin inhibitory unit (TIU) aprotinin/ml as protease inhibitors. The homogenates were centrifuged at 600 g for 30 min at 4°C, and the supernatants obtained following this initial centrifugation were diluted in the same buffer and centrifuged again at 20 000 # for 30 min. The pellets obtained were then washed and centrifuged again at 20 000 g for 30 min in the same buffer (except for the absence of aprotinin). The pellets were then resuspended in Krebs-Ringer phosphate buffer (KRP; 118 mmol NaCl/1, 5 mmol KC1/1, 1-2 mmol MgS04.7H20/l, 1-2 mmol KH2P04/1 and 10 mmol Na2HPO /l) and stored at -70 °C before utilization. Insulin binding to liver membranes was measured in 015 ml KRP buffer 7-8), containing 1% (w/v) BSA and 1 mg bacitracin/ml using about 0017nmol monoiodinated A14 human insulin/1 and 0-3 mg membrane protein/ml (Lowry, Rosebrough, Farr & Randall, 1951). Tracer binding was inhibited by increasing concentrations of unlabelled monocom¬ ponent porcine insulin (0-2-3 µ / ). The reaction was stopped after 16-18 h of incubation at 4°C by

(pH

centrifugation in a Beckman microfuge at 4 °C for 3 min. After aspirating the supernatant, the surface of the pellets was washed without centrifugation, using 0-2 ml sucrose (0-25 mol/1). The pellets were then counted in a -counter. Non-specific binding was represented by the radioactivity bound in the presence of 2-3 pmol unlabelled insulin/ml and was found to be 12-21% of the total binding in experiment 1 and 15-23% in experiment 2. Degradation of tracer was estimated as non-precipitable radioactive material in 5% (w/v) trichloroacetic acid. Liver membranes (5-6 mg protein/ml in KRP buffer) were solubilized using 1% (v/v) Triton X-100 in the presence of 1 mmol PMSF/1 on ice for 60-90 min. The mixture was centrifuged at 150 000g (bottom of the tube) for 45 min at 4 °C. Solubilized receptors were partially purified using two types of lectin columns. The receptors were first adsorbed, washed and eluted from an 8 ml lentil column using 0-3 mol mannose/1, then concentrated by adsorption, further washing and elution from a 2 ml WGA column using 0-3 mol acetylglucosamine/1. These lentil receptors are ATPase-free (Simon et al. 1986) but represent only about 20% of the total receptors. The concentration of solubilized receptors was measured by incubation with l25I-labelled insulin for 3-4 h at room tempera¬ ture in 50 mmol Hepes/1-150 mmol sodium chloride/1 buffer at pH 7-8 using 1% BSA and 1 mg bacitracin/ ml. The reaction was stopped by precipitation and washing using polyethylene glycol as described by Hedo, Harrison & Roth (1981). Tyrosine kinase activity of the receptors was measured at room temperature as described pre¬ viously (Simon & LeRoith, 1986; Simon et al. 1986). Receptors were first incubated in the absence or presence of insulin for 30 min before the addition of the 32P-labelled ATP mixture which contained

(final concentration): 20 mmol Mg2+, 3 mmol Mn2+, 1 mmol cytidine 5'-triphosphate and 1 mmol sodium vanadate/1. For the phosphorylation of artificial sub¬ strate Poly(Glu,Tyr)4:l, 50 pmol unlabelled ATP/1

used and the reaction stopped during the linear phase at 15 min. For the autophosphorylation of the ß-subunit of the receptor, the concentration of unlabelled ATP was reduced to 25 pmol/l and the amount of [32P]ATP was increased twofold. The reaction was stopped after 6 min. Following direct application, the samples were run on sodium dodecyl was

sulphate (SDS)-polyacrylamide gel electrophoresis under reducing conditions. After autoradiography,

the radioactive bands were excised from the gels and the radioactivity was extracted and counted by liquid scintillation. Statistical analysis of the data was performed using r-test or multifactorial analysis of variance (Snedecor

&Cochran, 1956).

RESULTS

The general effects of the dw gene are shown in 5-week-old birds (Table 1). Body weight, tarsal length, food intake (data not shown), food efficiency (between 2-5 weeks of age) and body temperature were significantly (P< 0-05-0-01, Table 1) reduced in dwarf chickens. In the fed or fasted states liver weight (expressed as per cent of body weight) was not modi¬ fied (data not shown). The abdominal fat content (per cent of body weight) which is a good estimate of total body fat content (Delpech & Ricard, 1965) was increased in dwarf chickens (P