insulin receptors of chicken liver and brain : characterization of α and β subunit properties

Eur. J. Biochem. 158,125-132 (1986) 0 FEBS 1986

Insulin receptors of chicken liver and brain Characterization of a and /I subunit properties

Jean SIMON and Derek LEROITH Diabetes Branch, National Institute of Arthritis, Diabetes, and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland

(Received December 2, 1985/April8, 1986) - EJB 85 1301

Receptors on membranes of chicken liver and brain bound porcine '251-insulin in a specific and temperature- dependent manner. Competition with unlabeled insulin derivatives exhibited typical insulin potency ratios, i.e. chicken > porcine insulin > human proinsulin (2.1/1/0.02). Apparent binding affinity was higher in brain with a 50% inhibition of tracer binding of 1.3 f 0.2 nM porcine insulin as compared to 2.8 k 0.3 nM in liver. The apparent molecular mass of the '2sI-insulin cross-linked a subunit of the insulin receptor was 139 & 2 kDa for chicken liver and 127 2 kDa for chicken brain. These molecular masses were similar to those of rat liver and brain insulin receptors. Neuraminidase treatment of the cross-linked insulin receptor increased the mobility of the a subunit from liver but did not affect that from brain, suggesting a difference in the glycosylation of the chicken brain a subunit as previously described in the rat. Despite this change, both receptors could be purified on wheat germ agglutinin (WGA) chromatography after Triton solubilization. In the presence of CTP and vanadate (phosphatase inhibitors) insulin-stimulatable tyrosine-specific phosphorylation of exogenous substrates was demonstrated with chicken liver and brain receptors. The reaction was dependent on Mg2+ and Mn2+. As noted with other insulin receptors, the best artificial substrate for phosphorylation was poly(Glu,Tyr), : 1. In both chicken liver and brain the smallest effective insulin dose as well as maximal stimulation of phosphorylation of the substrate was similar to that seen with rat liver, and in all three tissues chicken insulin was more potent than porcine insulin. In chicken liver an active ATP hydrolytic activity copurified with the insulin receptors during WGA chromatography. Further purification using S-300 Sephacryl filtration or affinity (insulin-biotin-avidin) chromatography could dissociate the phosphorylation and the hydrolytic activities. Gel electrophoresis, under reducing conditions revealed fl subunits with apparent M, of 97 - 99 kDa in chicken liver and brain, which were phosphorylated in the presence of insulin. Similar apparent molecular masses have been described for the p subunit of rat liver receptors. These studies suggest that both chicken brain and liver insulin receptors exhibit coupling of a and subunits with fully active tyrosine kinase and that the structural difference of the brain insulin receptor is widespread and phylogenetically old.

Insulin exerts its biological functions by binding to specific cell-surface receptors, followed by the initiation of a number of cellular processes. The insulin receptor is a transmembrane glycoprotein composed of two tl subunits and two p subunits linked by disulphide bonds. Recent studies using recombinant DNA technology have demonstrated that the human insulin receptor is derived from a single-chain precursor [l , 21. The tl subunit, the major insulin-binding site, contains a number of potential glycosylation sites and is probably exclusively in the extracellular domain. The fl subunit, in addition to an extracellular domain that links it to the a subunit, has a 23-amino-acid-long hydrophobic transmembrane segment, and a cytoplasmic segment. The fl subunit demonstrates insulin-stimulatable tyrosine-specific autophosphorylation as well as insulin-stimulatable tyrosine phosphorylation of exogenous substrates [3 - 81.

Correspondence to D. LeRoith, Diabetes Branch, National In- stitute of Arthritis, Diabetes and Digestive and Kidney Diseases, NIH-Bldg. 10, Room 88-243, 9000 Rockville Pike, Bethesda, Mary- land, USA 20892

Abbreviations. KRP, Krebs-Ringer phosphate buffer; BSA, bovine serum albumin; SDS, sodium dodecyl sulfate; WGA, wheat germ agglutinin; PAGE, polyacrylamide gel electrophoresis.

Insulin receptors are widespread in various tissues of most animal species studied [9, lo]. There is strong conservation of insulin receptor binding characteristics (pH and temperature dependence as well as affinity and analogue specificity) throughout the vertebrates [lo]. Similar to most other mammalian tissues, the brain has specific insulin receptors [ll - 131 that appear to be widely distributed and present on neuronal elements, glial cells and capillaries [14- 171. As compared to other receptors, rat brain receptors demonstrate certain peculiarities showing differences in specificity, absence of negative cooperativity and the presence of up-regulation [18, 191. In addition, the rat brain a subunit of the insulin receptor appears to migrate faster on gel electrophoresis, when compared to that from liver and adipocytes, suggesting a lower M, [20 - 231. In addition, neuraminidase accelerates the migration of the a subunit from non-neural insulin receptors from the rat but has little effect on the mobility of the a subunit from brain receptors, suggesting that the difference between the two receptors may be due to an alteration in the carbohydrate residues of the brain receptor [21, 221. Despite these differences in structure, the insulin receptor of rat brain demonstrates autophosphorylation as well as insulin- stimulatable tyrosine-specific phosphorylation of exogenous substrates [23, 241.

126

To determine whether the structural differences between brain and non-neural insulin receptors is restricted to mammals we studied the chicken as representing the Aves order. In this study we show that the structural differences in the a subunit of the insulin receptor between brain and liver tissues is also present in chickens. Furthermore, following the elimination of an ATP hydrolytic activity that copurified with chicken liver insulin receptors on wheat germ agglutinin (WGA) chromatography, we demonstrate that both brain and liver receptors were capable of autophosphorylation as well as insulin-stimulatable phosphorylation of exogenous substrates.

EXPERIMENTAL PROCEDURES

Materials Monoiodinated A14 porcine z51-insulin, purified by high-

performance liquid chromatography (radioreceptor grade, specific activity 250- 370 mCi/mg) and [y-32P]ATP(1000 - 3000 Ci/mmol) were purchased from New England Nuclear (Boston, MA). Porcine insulin was purchased from Elanco (Indianapolis, IN). Chicken insulin (from Littron Laborato- ries, lot 109) was obtained through the research resources program of the NlADDK, NIH (Bethesda, MD). Synthetic human proinsulin (A-18-4U6-253) was a gift from Dr Bruce Frank, Eli Lilly and Co. (Indianapolis, IN). Guinea-pig anti- porcine insulin antibody (batch 625) was purchased from the Department of Pharmacology, Indiana University (In- dianapolis, IN). Bovine serum albumin (fraction V, HPLC insulin-free) was purchased from Armour Pharmaceutical Co. (Kankakee, IL, USA). Wheat germ agglutinin (glycamino- silex) was purchased from Miles Yeda Ltd (Rehovot, Israel). Neuraminidase (from Clostridium perfringens, type N2133), casein (C4765) and synthetic tyrosine-containing polypeptides were purchased from Sigma Chemical Co. (St Louis, MO). All other chemicals used were reagent grade.

Preparation of membranes

Livers and brains were removed from 19-h-fasted 5-week- old male chickens (Heavy breed, Cornish Cross), placed on solid COz and stored at -70°C. Crude membranes were prepared by differential centrifugation [l 11. After homo- genization on ice in 1 mM sodium bicarbonate solution containing 2 mM phenylmethylsulfonyl fluoride, 10 pg/ml leupeptin and 1 TIU/ml aprotinin as protease inhibitors, the homogenates were centrifuged at 600 x g for 30 min at 4°C and the supernatants obtained following this initial centrifugation were centrifuged again at 20000 x g for 30 min. The pellets obtained were then washed and centrifuged again at 20000 x g for 30 min in the same buffer. The pellets were then resuspended in Krebs-Ringer phosphate (KRP) without calcium and stored at - 70°C before utilization. For control experiments male Sprague-Dawley rats (200 g) were used. Membranes from brain and liver were prepared similarly.

Binding ojinsulin to membranes

Insulin binding was measured in a calcium-free KRP buffer pH 7.8 containing 1% BSA and 1 mg/ml bacitracin using 0.03 nM 1z51-insulin as tracer and membranes at a protein concentration in the range 0.3 -0.5 mg/ml[25]. The reaction was stopped by centrifugation in a Beckman microfuge at 4°C for 10 min. The surface of the pellets was washed with

chilled 0.5 M sucrose and recentrifuged for 5 min. The non- specific binding was represented by the radioactivity bound in the presence of 10 pM unlabeled porcine insulin and was found to be 37-29% of the total binding for chicken liver and 25-37% for chicken brain. Degradation of the tracer was estimated as non-precipitable radioactive material in 5% trichloroacetic acid and was found to be 1.1 - 1.5% in the absence of membranes, and 3.0-3.6% and 3.7-4.1% in the presence of chicken liver and brain membranes respectively when tested for 4- 5 h at room temperature.

Cross-linking of porcine 1251-insulin to insulin receptors

Binding of porcine 1251-insulin (3 nM) to membranes (2 mg protein in 1.5 ml) was performed for 4 h at room temperature in KRP buffer, 1% BSA and 1 mg/ml bacitracin at pH 7.8. The reaction was stopped by centrifugation at 12000 x g for 10 rnin at 4°C. The pellet was washed twice in chilled KRP in the absence of BSA. The final pellet was suspended in 1.5 ml KRP without BSA pH 7.5 and cross- linking of the bound '251-insulin to the insulin receptors was performed by the addition of 15 pI 10 mM disuccinimidyl suberate for 15 min on ice [26]. The reaction was stopped by the addition of 300 p1 100 mM Tris, 10 mM EDTA, pH 7.4. The pellet was collected by centrifugation at 12000 x g for 10 rnin and solubilized on ice for 1 h in 50 mM Hepes, 10 mM magnesium sulfate, (pH 7.6) and 1 % Triton X-100 using 1 mM phenylmethylsulfonylfluoride as a protease inhibitor. The reaction mixture was centrifuged at 12000 x g for 30 rnin and the solubilized material was then immunoprecipitated using a guinea-pig anti-insulin antibody with staphylococcal protein A (pansorbin) adsorption [27]. After three washings in 50 mM Hepes, 0.1% Triton X-100 pH 7.6, the immuno- precipitate was boiled for 5 - 10 min in sample gel buffer using 5% mercaptoethanol as a reducing agent and the samples were analyzed by 7.5% PAGE in 0.1% SDS [28].

Neuraminidase digestion of cross-linked receptors [22 J After cross-linking of 1251-insulin to insulin receptors from

liver and brain membranes the pellets were resuspended in 1 ml solution containing 5 mM 2(N-morpholino)ethane- sulfonic acid, 1 mM calcium chloride, pH 5.05, in the presence or absence of 2.5 units neuraminidase and incubated for 30 rnin at 37°C. Pellets were obtained following centrifugation at 12000 x g for 15 rnin and washed once in 50 mM Hepes, 10 mM magnesium sulfate before solubilization as described above. The effect of neuraminidase on the cross-linked a subunit was analyzed in each case by PAGE.

Insulin receptor autophosphorylation and phosphorylation of exogenous substrates

Membranes (5 - 6 mg protein/ml in KRP buffer) were solubilized using 1% Triton in the presence of 1 mM phenylmethylsulfonyl fluoride on ice for 1 h. The mixture was centrifuged at 150000 x g for 45 min at 4°C and passed over a 2-ml WGA-agarose column three times at 4 'C. The material that adsorbed to the column was washed with 40 ml 150 mM sodium chloride, 0.1% Triton X-100, 50 mM Hepes buffer, pH 7.6, and then eluted in 1-ml fractions using 0.3 M N-acetylglucosamine in the same buffer [29]. Protein concentration of the eluates was assessed with the Bio-Rad assay using BSA diluted in 50 mM Hepes pH 7.6 containing 150 mM sodium chloride, 0.1% Triton X-100 and 0.3 M

127

HOURS I! L\ I I I I I I I I

0 1 2 3 4 5 6 7 8 HOURS

22-24OC

Fig. 1. Time course of ussociation ofporcine '2sI-insulin with chicken liver and bruin membranes at room temperature. These data are prescnted as percentages of specific bound over total, corrected for 0.5 mg proteinlml. Degradation and non-specific binding were low and are detailed in Experimental Procedures. In the insert the effect of temperature on the binding of tracer insulin to chicken brain membranes is shown

N-acetylglucosamine as protein standard [30] (Bio-Rad Assay, Richmond, CA). The concentration of insulin receptors eluted from the wheat germ agglutinin column was measured by incubation with 1251-insulin for 3 h at room temperature pH 7.8 using 1% BSA and 1 mg/ml bacitracin [29]. Further purification of the WGA-purified chicken liver insulin receptors was performed by S-300 Sephacryl (Pharmacia) filtration or affinity chromatography as originally described [31] using a 2.5-ml insulin-biotin-avidin column (the generous gift of Dr Frances M. Finn, University of Pittsburgh). Phos- phorylation of exogenous substrates poly(Glu, Tyr) (at 0.4 mg in 180 pl) and the p subunits of the insulin receptor was performed at room temperature under conditions previously described. Preincubation of insulin at 1/7 dilution in KRP buffer 0.1 % BSA was performed for 30 min at room temperature prior to the addition of the labeled [32P]ATP mixture [32, 331. Hydrolysis of ATP was measured using the molybdate technique [32, 341.

RESULTS

Insulin binding to chicken liver and brain membranes

Binding of porcine '251-insulin to chicken membranes was directly dependent on protein concentration up to 1.9 mg/ml (data not shown) as well as being time-dependent with a similar time course for brain and liver at 22°C (Fig. 1). Tem- perature dependence of binding was studied in brain and demonstrated higher binding at low temperatures (Fig. 1 insert). In brain 50% inhibition of binding of the tracer was achieved at 1.3 0.17 nM (mean f SEM, n = 5). In liver the competition was less sensitive and the 50% inhibition required 2.8 f 0.3 nM (Fig. 2). Scatchard-type analysis of these data was curvilinear upward in both tissues (data not shown). The fact that the half-displacement concentration is lower for the brain receptor while the tracer binding is the same in both tissues may suggest that the binding capacity is somewhat higher in liver. Chicken and porcine insulins inhibited the

T

INSULIN InM)

Fig. 2. Competition inhibition of binding of porcine insulin to chicken h e r and brain membranes. Porcine L251-insulin (0.03 nM) was incubated with membranes for 4- 5 h at room temperature in the absence and presence of increasing concentrations of unlabeled porcine insulin. The results are presented as means f SEM (a total of five experiments were performed for both liver and brain membranes)

binding of tracer-labeled porcine insulin with relative potencies of 2: 1 in both brain and liver, and human proinsulin exhibited a 0.02 potency ratio when compared to porcine insulin in brain membranes (Fig. 3).

Cross-linking of I-insulin to insulin receptors and the ejject of neuraminidase digestion

Receptors from chicken liver and brain membranes were cross-linked with 1251-insulin, solubilized, immunoprecipitated using an anti-insulin antibody and then subjected to PAGE under reducing conditions (Fig. 4). Following autoradiography, a major band was seen in both tissues. This band migrated faster for brain receptors than for liver receptors (Fig. 4). The mean values of the apparent M , of these bands

128

F INSUUN lnYl INSULIN hYl F

Fig. 3. Analogue specificity for the inhibition of the binding of porcine '2sI-insulin to chicken liver and brain membranes. Porcine and chicken insulins as well as human biosynthetic proinsulin were compared for their ability to inhibit the binding of porcine 12SI-insulin in chicken brain membranes (right panel) for 4- 5 h at room temperature as described under Experimental Procedures. Porcine and chicken insulin were studied for their ability to inhibit specific binding of tracer in chicken liver membranes (left panel). The dose of insulin or proinsulin which caused 50% inhibition of tracer binding is represented by the dashed line

Table 1. Molecular mass of the insulin receptor a subunit in chicken and rat liver and brain membranes afer cross-linking porcine lZ5 I-insulin and SDS-PAGE Mean f SEM for six gels for rat and seven for chicken tissues. In each experiment, rat and chicken tissues were electrophoresed on the same gel

Tissue Chicken Rat

kDa

Liver 139+2 142 & 2 Brain 127 2 130&2

Fig. 4. Cross-linking ofporcine '251-insulin to the a subunit of the insulin receptor from chicken liver and brain membranes. Crude membranes of chicken liver and brain were incubated with porcine '2sI-insulin and cross-linked using disuccinimidyl suberate. The cross-linked materials were solubilized and immunoprecipitated using guinea-pig anti-insulin serum. The immunoprecipitated material was run on polyacrylamide gels under reducing conditions. The effect of neuraminidase was tested by incubating equivalent amounts of cross- linked material with or without neuraminidase prior to solubilization and immunoprecipitation. From the left to the right, lane 1 represents chicken liver membranes without neuraminidase and lane 2 with neuraminidase after 16 h autoradiography. Lane 3 represents chicken brain membranes without neuraminidase and lane 4 with neuraminidase after 4 days autoradiography. Molecular mass markers were run in parallel on the same gel

is presented in Table 3 and were very similar to those found in control rat tissues. Both chicken and rat brain a subunits were about 10 kDa smaller than liver a subunits (Table 1). Neuraminidase treatment increased the mobility of the a subunit of the chicken liver insulin receptor by about 5 kDa but did not affect the mobility of the a subunit of the chicken brain insulin receptor (Fig. 4).

Phosphorylation of exogenous, arti jkial substrates

Solubilized membranes were purified on WGA-agarose columns. This treatment concentrated the chicken liver and brain insulin receptors and was associated with an increase in the affinity for insulin (data not shown) as has previously been described for other tissues [29]. WGA-purified insulin receptors from chicken liver, in the presence of 20 mM Mgz+ alone failed to phosphorylate the artificial substrate poly- (Glu,Tyr)4: eitheras a function of time or insulin concentration. This is in contrast to findings with the rat liver receptors (data not shown). This difference could not be accounted for by any changes in binding of insulin to the solubilized receptors since under these conditions the time course of associ- ation of insulin with receptor occurs at the same rate in both species (data not shown). When studied in the presence of 20 mM MgZ ', 1 mM CTP and 1 mM sodium orthovanadate (a phosphotyrosylprotein phosphatase inhibitor [35]), phosphorylation of artificial substrate by chicken liver receptors was demonstrable and was insulin-stirnulatable (data not shown). These responses were further increased by addition of 3 mM Mn". Brain insulin receptors similarly demonstrated maximum phosphorylation of exogenous substrate in the presence of Mgz+, Mn2+, CTP and vanadate with a lesser response in the absence of Mgz+. MgZ+, Mn2+ and CTP without vanadate resulted in maximum phosphorylation (data not shown).

As demonstrated previously [24, 331 in other systems, the artificial polymer poly(Glu,Tyr)4 : appeared to be a preferen- tial substrate for both chicken liver and brain insulin receptors (data not shown). Casein was phosphorylated by chicken liver

129

300-

cINSI1O-'MI 0 rn- 25 - _____--------_

0 /---

,A 200- ,d' 2 0 -

I + INS(lO-'MI ,,*'&

,L , ,

A I'

Fig. 6. Insulin dose-response and specijicity of insulin analogues for the phosphorylation of artijicial substrate poly(G1u. f i r ) 4 : , in WGA-purified insulin receptors. Insulin receptors from chicken liver (A, 55 pg protein/ml; tracer binding capacity 59%). chicken brain (B, 41 pg protein/ ml; tracer binding capacity 21 %) and rat liver (C, 27 pg protein/ml; tracer binding capacity 62%) were first preincubated for 30 rnin at room temperature in the presence of chicken or porcine insulin at various concentrations. Phosphorylation was initiated using the combination of Mg2+, Mn2+, CTP and vanadate and allowed to continue at room temperature for 5 min. At each insulin concentration, data are plotted as a percentage of the maximal increase above basal activity (measured in the absence of insulin). Maximal stimulation of the phosphorylation was typically achieved at 0.1 pM insulin. x basal and maximally stimulated activities (cpm) were: 6.1 and 20.0 in chickcn liver, 9.6 and 44.5 in chicken brain and 8.8 and 47 in rat liver

D 1 5 -

B x l o -

receptors at a high level in the basal state but exhibited no further increase in response to insulin (data not shown). Using poly(Glu,Tyr),: ,, the time course of the phosphorylation reached a plateau at about 6 min for both basal and insulin- stimulated activities with chicken liver insulin receptors (Fig. 5), whereas it was linear for at least 1 h and achieved much higher levels for basal and insulin-stimulated activities with chicken brain receptors (Fig. 5). These data for chicken brain are similar to those obtained with rat liver insulin receptors (data not shown). In all three tissues, maximal stimulation was achieved at about 0.1 pM porcine insulin (Fig. 6). When the dose response for insulin-stimulatable phosphorylation was measured at 5 min, the smallest effective dose was found to be 0.1 nM porcine insulin and the half-maximal

150 - E

I I 8: B

I P ' - 100

5 0 -

E

effective dose was 1.2 - 2 nM for both chicken liver and brain insulin receptors (Fig. 6A and B). These concentrations were quite similar to those obtained with rat liver receptors (Fig. 6C). Chicken insulin was more potent than porcine insulin in stimulating the phosphorylation of the artificial substrate with a potency ratio of 1.3 in chicken liver, 1.8 in chicken brain and 2.6 in rat liver (Fig. 6A, B, C).

ATP-hydrolytic activity

The possibility that an ATP-hydrolytic activity associated with chicken liver insulin receptors accounted for the low activity and the early plateau of the reaction observed in this tissue was examined. WGA-purified insulin receptors from

130

chicken liver exhibited a very potent ATP-hydrolytic activity (see Fig. 8) whereas WGA-purified receptors from chicken brain or rat liver, at about the same protein concentrations, were completely free of 'ATPase-like' activity (data not shown). Attempts to inhibit hydrolytic activity using 2 mM ouabain were unsuccessful. Furthermore, extensive washing of the receptors while bound to the WGA column using 20 mM 3 - [(3 -cholamidopropyl)dimethylammonio] - 1 -pro- panesulfonate was unable to dissociate the hydrolytic activity from the receptor. Therefore, the WGA eluate of chicken liver insulin receptors was further purified separately by both S-300 Sephacryl filtration and affinity chromatography using a 2.5-ml insulin-biotin-avidin column [31]. Insulin receptors eluting from the S-300 Sephacryl column were divided into

VO CATALASE

two separate pooled fractions I and I1 (Fig. 7) and each pool was assessed for its phosphorylation and hydrolytic activities (Fig. 8). When both pools were compared at the same binding capacity, the time course of phosphorylation of exogenous substrate poly(Glu,Tyr)4: was linear for at least 45 rnin (Fig. 8A) and the ATP hydrolysis was minimal in fraction I (Fig. SB). In contrast, in fraction 11, the phosphorylation again reached its maximum between 6 min and 15 min and exhibited low activity (Fig. SC). When ATP hydrolysis by fraction I1 receptors and WGA eluate were compared, the

0.6 rnl FRACTIONS

Fig. 7. Gel filtration of WGA-purified insulin receptors from chicken liver on S-300 Sephacryl column (50 x 0.9 c m ) . In this experiment about 1.7 ml WGA-purified receptors (285 pg protein/ml) were applied to the column. Elution was performed in 0.1 % Triton, 50 mM Hepes, 150 mM NaCl buffer pH 7.6; 0.5-ml fractions were collected. Aliquots (0.025 ml) were assayed for 1251-insulin binding to determine the elution profile of insulin receptors. The column was calibrated with blue dextrdn (Vo) and catalase. The peak of insulin receptors was separated into two pooled fractions (I and 11) as shown

Fig. 9. Autophosphorylution of the /3 subunit of the insulin receptors. WGA-purified receptors from chicken liver (92 pg protein/ml) and chicken brain (36 pg/ml) were preincubated for 30 rnin at room temperature in the absence or the presence of 0.1 pM porcine insulin, without added artificial substrate. 32P-labeled ATP mixture containing Mgz+, MnZ*, CTP and vanadate was added. After 5 min the reaction was stopped and the samples were electrophoresed on SDS/polydcryhnide gels under reducing conditions as previously described [4]. Gels have been autoradiographed for 17 h. Migration of molecular mass markers is indicated

INS. BIOT. AVID. S-300 FRACTION I S-300 FRACTION II PURIFIED RECEPTORS

3 0 C

9 8'

+INSIlO-'MI ,,*

/'

+ INSIlO-'MI +INSIlO-'MI ,' ,' o 151 ,/

9,'' 7 im ,/' p 10 J

0 st /"" j 5 " -INS

o , , , 0 1 U ' I

0 2 6 15 30 45 02 6 16 30 45 02 6 15 J) 45

70- Bo-

B

40-

20-

O C I I r r I 1 1 0 2 4 8 10 15 30

BlanklBuffed

Jo I,,,:, 0 2 4 6 10 15 , TIME, rnin

Fig. 8. Time course ofphosphorylation of urtficial substrate poly(Glu,Tyr)4: (upper part) and ATP hydrolysis (lower part) by WGA-purified insulin receptors from chicken liver further purijied by either gel filtration or affinity (insulin-biotin-avidin) chromatography. Receptors from Sephacryl filtration, fractions I and I1 (see Fig. 7; tracer binding capacity of 42%) or affinity-chromatographed receptors (tracer binding capacity of 8%) were preincubated for 30 min at room temperature in the absence or the presence of 0.1 pM porcine insulin. Phosphorylation was initiated using the combination of Mg2+, Mn", CTP and vanadate as described in Fig. 5. ATP hydrolysis was measured under the same conditions, in the absence of substrate, using phosphomolybdate precipitate as previously described [32, 341. Hydrolytic activity of the WGA receptors was also measured

131

hydrolytic activities were indistinguishably high (Fig. 8 D). When WGA-purified chicken liver insulin receptors were purified by affinity (insulin-biotin-avidin) chromatography, only a very small fraction of the receptors bound to the column. These receptors (at a binding capacity for the tracer of So/,) were able to phosphorylate the exogenous substrate poly(Glu,Tyr)4: I at a linear rate for at least 45 min both in the basal and insulin-stimulated conditions (Fig. 8 E) and exhibited no hydrolytic activity (Fig. 8F).

Cation dependence and substrate specificity were re-examined using chicken liver receptors which had been further purified by S-300 Sephacryl filtration. Mn2+ alone supported basal and insulin-stimulatable phosphorylation of the substrate p~ly(Glu,Tyr)~, at a low level (data not shown). Since Mn2 + alone was unable to support phosphorylation when WGA-purified receptors were used (see above), the possibility exists that the removal of the ATPase-like activity may have accounted for these differences. Both basal and insulin- stimulated activities were largely increased in the presence of Mgz+ alone and combination of both cations produced a further increase. These receptors were also unable to phosphorylate poly(Glu,Tyr)l : (data not shown).

Phosphorylation o j the

Phosphorylation of the /? subunit of the WGA-purified insulin receptors was carried out in the presence of Mg2+, Mn2 +, CTP and vanadate. Following gel electrophoresis and autoradiography there was no detectable phosphorylation of the /3 subunit in the basal state (Fig. 9). However, insulin at 0.1 pM stimulated the phosphorylation of a band with an apparent molecular mass of 101 kDa in control rat liver (data not shown) and 97-99 kDa in chicken liver and brain (Fig. 9). The 97-99 kDa band was recognised by human anti-insulin receptor antiserum (Bio; kindly provided by Dr Simeon Taylor, Bethesda, MD) strongly suggesting that it represents the #l subunit of the insulin receptor. In addition, in both chicken brain and liver, a band of an apparent molecular mass of 185 kDa also appeared to be phosphorylated in- dependently of insulin. A similar observation was seen with rat liver (data not shown). This band requires further in- vestigation. It may well correspond to that previously described in rat liver, where streptozotocin treatment appeared to decrease the phosphorylation of this band [36].

subunit of the insulin receptor

DISCUSSION Specific insulin receptors were present in the brain and

liver of young growing chickens. The characteristics of binding were similar to those previously reported for other chicken tissues [37 - 431. Thus binding was higher at low temperatures, and the potency of various insulins was similar to that described in other tissues and other species, i.e. chicken insulin > porcine insulin > proinsulin. Interestingly the apparent affinity of the insulin receptor was greater in brain tissue. Liver tissue, on the other hand, apparently had higher binding capacity than brain tissue.

The a subunit of chicken brain receptors migrated faster than that of chicken liver on SDS-PAGE under reducing conditions. The apparent molecular mass difference was about 10 kDa. The migration of the brain o! subunit was unaffected by treatment with neuraminidase whereas liver x subunit migrated faster following exposure to neuraminidase. These differences are similar to those found in rats 121,

brain and liver. However, both chick brain and liver insulin receptors bound to and were eluted from WGA-agarose columns using N-acetylglucosamine. These findings suggest that either WGA-agarose binds residues other than sialic acid or that the sialic acid residues in chick brain receptors are inaccessible or resistant to neuraminidase treatment.

Despite the apparent differences in M , and glycosylation of the a subunits from chick brain and liver, both receptors demonstrated the ability for insulin-stimulatable autophosphorylation of the j subunit in a manner similar to that described in other tissues and other species [3-8, 23-24, 32-33]. Insulin receptors of both chicken brain and liver membranes functioned as tyrosine kinases, phosphorylating the exogenous substrate p~ly(Glu,Tyr)~: 1. Both tissues demonstrated a dependence of phosphorylation on the divalent cations (Mg2+ and Mn2+ [44]) and demonstrated similar specificities, i.e. phosphorylation of exogenous substrate was greater for poly(Glu,Tyr)4: than poly(Gly,Tyr)l : which was also a characteristic for other insulin receptors [24, 331. In addition, chicken insulin stimulated tyrosine-specific phosphorylation of exogenous substrates with a potency almost twice that of porcine insulin, similar to the higher binding affinity and biological potency of chicken insulin [45,46].

A very active ATPase-like activity copurified with the chicken liver insulin receptors during WGA chromatography and largely masked the phosphorylation activity of the receptor. This hydrolytic activity probably does not represent an intrinsic activity of the receptor molecule since it can be dissociated from the receptor with full recovery of the normal properties of the phosphorylation activity of the receptor.

Thus chicken brain and liver insulin receptors demonstrate the classic a and /? subunit structure similar to that previously described for chicken embryonic heart cells [47]. Furthermore, as described in the rat [20-231, the chicken brain insulin receptors have structural differences as compared to liver. This difference is already present in the chicken during embryonic development [48] and has also been found in various species such as lizard and guinea-pig [49, 501 suggesting that it is phylogenetically old and well conserved. Whether these structural differences are related to a specific function for insulin in neural tissue is unknown [51]. Interestingly, similar differences have recently been noted in cholecystokinin brain and pancreatic receptors 152 - 541. Despite the structural differences in chick brain and liver insulin receptors, autophosphorylation of the /? subunit and insulin-stimulatable tyrosine-specific phosphorylation of artificial exogenous substrates are functional in both tissues suggesting normal coupling of c( and fl subunits of the insulin receptors in both tissues. Though phosphorylation may be one of the earliest of insulin actions on the cell, its biological significance remains to be determined.

We wish to thank Jesse Roth, Yehiel Zick, Simeon 1. Taylor and Charles Bevins for helpful discussions. The generous gift of insulin- biotin-avidin agarose by Frances Finn is greatly appreciated. Advice regarding the experimental techniques from M. A. Lesniak, G . Grunberger, J. A. Hedo, C. Hart, D. Rouiller, R. Comi, J. P. Kinet, and V. Moncada are also acknowledged. V. Katz is thanked for typing of the manuscript. J. S . was on sabbatical leave from Institut National de la Recherche Agronomique, Station de Recherches Avicoles. Nouzilly, France.

REFERENCES 1 . Ebina, Y., Ellis, L., Jarnagin, K., Edery, M., Graf, L., Clauser,

E., Ou, J. H., Masiarz, F., Kan, Y. W., Goldfine, I. D., Roth, R. A. & Rutter. W. J. (1985) Cell40. 147-758. 221 and suggest a difference in sialic acid residues between

132

2. Ullrich, A., Bell, J. R., Chen, E. Y., Herrera, R., Petruzzelli, L. M., Dull, T. J., Gray, A,, Coussens, L., Liao, Y. C., Tsubokawa, M., Mason, A., Seeburg, P. H., Grunfeld, C., Rosen, 0. M. & Ramachandran, J. (1985) Nature (Lond.) 313,756-761.

3. Avruch, J., Nemenoff, R. A., Blackshear, P. J., Pierce, M. W. & Osathanondh, R. (1982) J. Biol. Chem. 257, 15162-15166.

4. Kasuga, M., Karlsson, F. A. & Kahn, C. R. (1982) Science (Wash.

5 . Kasuga, M., Zick, Y., Blithe, D. L., Crettaz, M. & Kahn, C. R. (1982) Nature (Lond.) 298,667-669.

6. Kasuga, M., Zick, Y., Blithe, D. L., Karlsson, F. A., Haring, H. U. & Kahn, C. R. (1982) J. Biol. Chem. 257,9891 -9894.

7. Petruzzelli, L. M., Ganguly, S., Smith, C. J., Cobb, M. H., Rubin, C. S. & Rosen, 0. M. (1982) Proc. Natl Acad. Sci. USA 79,

8. Van Obberghen, E. & Kowalski, A. (1982) FEBSLett. 143,179- 182.

9. Muggeo, M., Ginsberg, B. H., Roth, J., Neville, D. M. Jr, DeMeyts, P. & Kahn, C. R. (1979) Endocrinology 104, 1393- 1402.

10. Muggeo, M. P., Van Obberghen, E., Kahn, C. R., Ginsberg, B. H., DeMeyts, P., Emdin, S. 0. & Falkmer, S. (1979) Diabetes

11. Havrankova, J., Roth, J. & Brownstein, M. (1978) Nature

12. Havrankova, J., Roth, J. & Brownstein, M. J. (1979) J. Clin.

13. Pacold, S . T. & Blackard, W. G. (1979) Endocrinology 105,1452-

14. Boyd, F. T., Jr & Raizada, M. K. (1983) Am. J . Physiol. 245,

15. Clarke, D. W., Boyd, F. T., Jr, Kappy, M. S. & Raizada, M. K.

16. Pardridge, W. M., Eisenberg, J . & Yang, J. (1985) Diabetes 34,

17. Haskell, J. F., Meezan, E. & Pillion, D. J. (1985) Am. J . Physiol.

18. Gammeltoft, S . , Staun-Olsen, P., Ottesen, B. & Fahrenkrug, J. (1984) Peptides 5 , 937-944.

19. Van Schravendijk, C. F. H., Hooghe-Peters, E. L., De Meyts, P. & Pipeleers, D. G. (1984) Biochem. J. 220, 165-172.

20. Yip, C. C., Moule, M. L. & Yeung, C. W. T. (1980) Biochem. Biophys. Res. Commun. 96,1671 - 1678.

21. Heidenreich, K. A., Zahniser, N. R., Berhanu, P., Brandenburg, D. & Olefsky, J. M. (1983) J. Biol. Chem. 258,8527-8530.

22. Hendricks, S. A., Agardh, C.-D., Taylor, S. I. & Roth, J. (1984) J. Neurochem. 43,1302- 1309.

23. Gammeltoft, S., Fehlmann, M. & Van Obberghen, E. (1985) Biochimie (Paris) 67, 1147-1153.

24. Rees-Jones, R. W., Hendricks, S. A., Quarum, M. & Roth, J. (1984) J. Biol. Chem. 259, 3470-3414.

25. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. (1951) J . Biol. Chem. 193, 265-275.

26. Pilch, P. F. & Czech, M. P. (1980) J. Biol. Chem. 255, 1722- 1731.

DC) 215,185-187.

6792 - 6796.

28,175-181.

(Lond.) 272,827-829.

Invest. 64, 636-642.

1457.

C283 -C287.

(1984) J. Biol. Chem. 259,11672 - 11 675.

suppl. 1, 164A.

248, E115-El25.

27. Kasuga, M., White, M. F. & Kahn, C. R. (1985) Methods

28. Kasuga, M., Van Obberghen, E., Yamada, K. M. & Harrison, L.

29. Hedo, J. A., Harrison, I. C. & Roth, J. (1981) Biochemistry 20,

30. Bradford, M. M. (1976) Anal. Biochem. 72, 248-254. 31. Finn, F. M., Titus, G., Horstman, D. & Hofmann, K. (1984)

32. Zick, Y., Kasuga, M., Kahn, C. R. & Roth, J. (1983) J. Biol.

33. Zick, Y., Grunberger, C., Rees-Jones, R. W. & Comi, R. J. (1985)

34. Reimann, E. M. & Umfleet, R. A. (1978) Biochim. Biophys. Acla

35. Swarup, G., Cohen, S. & Garbers, D. L. (1982) Biochem. Biophys.

36. Okamoto, M., White, M. F., Maron, R., Takayama, S. & Kahn,

37. Ginsberg, B. H., Kahn, C. R. & Roth, J. (1977) Endocrinology

38. Simon, J., Freychet, P. & Rosselin, G. (1977) Diabetologia 13,

39. Kemmler, W., Renner, R., Zynamon, A. & Hepp, K. D. (1978)

40. Simon, J. (1979) Biochim. Biophys. Acta 585, 563-574. 41. Wheeler, F. B., Santora, A. C. I1 & Elsas, L. J. I1 (1980)

42. Krupp, M. & Lane, M. D. (1981) J, Biol. Chem. 256,1689- 1694. 43. Hendricks, S. A., De Pablo, F. & Roth, J. (1984) Endocrinology

44. Nemenoff, R. A., Kwok, Y. C., Shulman, G. I., Blackshear, P. J., Osathanondh, R. & Avruch, J. (1984) J. Biol. Chem. 259,

45. Simon, J., Freychet, P. & Rosselin, G. (1974) Endocrinology 95,

46. Simon, J., Freychet, P., Rosselin, G. & DeMeyts, P. (1977)

47. Endo, F. & Elsas, L. J. I1 (1984) Endocrinology 115, 1828 - 1837. 48. Bassas, L. L., DePablo, F., Lesniak, M. A. & Roth, J. (1985) XII

Congr. Int. Diabetes Fed. Madrid (Sept. 1985), Diabetes Res. Clin. Pract. (suppl. 1; Baba, S., ed.) p. S35, Elsevier, Amsterdam.

49. Shemer, J., Penhos, J . C. & LeRoith, D. (1986) Diabetologia, in the press.

50. Lowe, W. & LeRoith, D. (1986) Endocrinology, in the press. 51. Hendricks, S. A., Roth, J., Rishi, S. & Becker, K. L. (1983) in

Brain peptides (Krieger, D. T., Brownstein, M. & Martin, J., eds) pp. 903-939, Wiley, New York.

52. lnnis, R. B. & Snyder, S. H. (1980) Proc. Narl Acad. Sci. USA

53. Saito, A,, Goldfine, I. D. & Williams, J. A. (1981) J. Neurochem.

54. Sakamoto, C . , Williams, J . A. & Goldfine, I. D. (1984) Biochem.

Enzymol. 109, 609 - 621.

C. (1981) Diabetes 30, 354-357.

3385 - 3393.

Proc. Natl Acad. Sci. USA 81,7328 - 7332.

Chem. 258,75 - 80.

Eur. J . Biochem. 148,177- 182.

523,516-521.

Res. Commun. 107,1104-1109.

C. R. (1985) Clin. Res. 33, 573A (abstr.).

100,82-90.

219-228.

Biochim. Biophys. Acta 543, 349- 356.

Endocrinology 107,195 - 207.

115,1315-1323.

5058 - 5065.

1439 - 1449.

Endocrinology 100, 115- 121.

77,691 7 - 6921.

37,483-490.

Biophys. Res. Commun. 124,497 - 502.

insulin receptors of chicken liver and brain : characterization of α and β subunit properties

Documents