Vol. 129, No. 3, 1985 BIOCHEMICALANDBIOPHYSICALRESEARCH COMMUNICATIONS

June28.1985 Pages 739-745

REMOVAL OF SIALIC ACIDS FROM THE PURIFIED INSULIN RECEPTOR RESULTS IN ENHANCED INSULIN-BINDING AND KINASE ACTIVITIES

Yoke Fujita-Yamaguchi, Yuichi Sato, and Satish Kathuria

Department of Molecular Genetics, Beckman Research Institute of the City of Hope, 1450 East Duarte Road

Duarte, California 91010

Received fipril 9, 1985

SUMMARY: Neuraminidase treatment of the purified insulin receptor resulted in an increase in both insulin-bindinq and kinase activities. Neuraminidase- treated 0: and B subunits moved further than native subunits on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) under reducing conditions. The enhancement of insulin-binding and kinase activities and increased mobility of the subunits on SDS-PAGE were not observed when the receptor was treated with neuraminidase in the presence of neuraminidase inhibitor. These results suggest that terminal sialic acid residues have a significant role in insulin-binding and kinase activities. The involvement of sialic acid residues in the activi- ties of the receptor has not been detected by previous studies. QI 1985 Academic Press, Inc.

An intact insulin receptor is composed of a and p subunits with molecular

weight of 125,000 and 90,000, respectively (1). Both subunits are glycoproteins

which appear to have N-asparagine linked oligosaccharide chains (2-4).

Early studies by Cuatrecasas and Illiano suggested that sialic acids at the

cell surface may be involved in the process of signal-transduction from insulin-

binding to glucose transport and lipolysis, but is not involved in the recogni-

tion of insulin by the receptor (5). Since then, the role of terminal

carbohydrate residues in insulin-binding activity has been studied by several

groups (6-8). In general, removal of sialic acids from solubilized or

membrane-bound insulin receptors by neuraminidase has not been shown to

have any effect on insulin-binding activity, while recent studies on the

biosynthesis of the insulin receptor in cultured cells have revealed that

the oligosaccharide chains of the receptor are required for processing of

ABBREVIATIONS SDS-PAGE: Sodium dodecyl sulfate-polyacrylamide gel electrophoresis, P,3-Dehydro NANA: 2,3-Dehydro-neuraminic acid, DTT: Dithiothreitol

0006-291X/85 $I..50

739 Copyright CC, 19X.i b.v Academic Press, Inc,.

,411 rtghts of reprodtrctron tn att.v form rrsrrvcd.

Vol. 129, No. 3, 1985 8lOCHEMlCAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

the receptor-precursor which leads to the formation of functional insulin

receptors (9,lO).

We have previously purified and characterized the human insulin recep-

tor with high insulin-binding and kinase activities (11-14). Amino acid

analysis of the purified subunits showed significant amounts of N-acetylgluco-

samine, indicating the presence of N-linked carbohydrate chains in both receptor

subunits (14). In the course of our studies on structure-function relationship

of the insulin receptor, we have investigated the role of terminal carbohydrate

residues using our purified receptor.

MATERIALS AND METHODS: Neuraminidase Type X from Clostridium perfringens (150 units/mg protein) and @-galactosidase grade IX from E. coli (630 units/mg protein) were purchased from Sigma. Crystalline porcine insulin was kindly supplied by Eli Lilly. tion site of pp60src

A synthetic peptide resembling the tyrosine phosphoryla- (Arg-Arg-Leu~Ile-G~~~Asp-Ala-Glu-Tyr-Ala-Ala-A~~-Gly) was

purchased from Peninsula Laboratories. I-Labeled insulin and [r- PI-ATP were from New England Nuclear. 2,3-Dehydro-neuraminic acid was generated by saponifying its methyl ester which was generously supplied by Dr. S.W. Tanenbaum (College of Environmental Science and Forestry, State University of New York, Syracuse, NY) (15). Molecular weight markers (myosin, B-galactosidase, phosphorylase b, bovine albumin and ovalbumin) were obtained from Bio-Rad. All other chemicals used were reagent grade.

Insulin receptor was purified 2400-fold with a yield of 409: from human pla- cental membranes by sequential affinity chromatography on wheat germ agglutinin- and insulin-Sepharose columns as described previously (11).

The purified insulin receptor (-0.1 pg) was treated with neuraminidase (0.5 - 10 pg/ml) and/or B-galactosidase (100 and 1000 pg/ml) in 20 ~1 of 50 mM Tris-HCl buffer, pti 7.4, containing 0.1% Triton X-100 at 37°C for 1 h. After the reaction, an aliquot (6 ~1) was assayed for insulin-binding activity as pre- viously described (11). Insulin receptor incubated under the same conditions without the enzyme was also assayed as a control. Specific insulin-binding activity of the enzyme-treated receptor was expressed as percent of that of the control.

In order to examine the effect of neuraminidase treatment on kinase acti- vity in the absence and presence of insulin, an exogeneous substrate was used.

for 8 h f ' The puri ied receptor (-0.1 pg) was treated with neuraminidase (5 pg/ml) at 4" C

and incubated with 1 pM of insulin at 4°C overnight in 20 ~1 of 50 mM Tris-HCl buffer, pH 7.4, containing 0.12 Triton X-100. Added to this solution was 5 ~1 of 50 mM Tris-HCl buffer, pH 7.4, containing 12 mM MnC12, 90 mM MgCl2 la;dazdi:go: ~:e,:r~;o'e~~t~dr_,,,tide. The phosphorylation reaction was initiated

PI-ATP (5-10 &i/nmol) and carried out for 40 min at room temperature. The reaction was terminated by the addition of 50 ,,l of 5x trichloroacetic acid and 20 ~1 of bovine serum albumin (10 mg/ml). After incu- bating this solution at 0°C for 30 min, the proteins were precipitated by centrifugation for 5 min. A 35 ~1 aliquot of the supernatant was spotted on a piece of phosphocellulose paper (Whatman, P81). The paper was extensively

1 We have used two different conditions for neuraminidase treatment in this study, since the kinase activity is found to be very unstable at 37'C. As far as the binding activity is concerned, effects of the enzyme treatments under both conditions (37“C, 1 h, and 4"C, 8-14 h) were the same.

740

Vol. 129,No. 3, 1985 BlOCHEMlCALAND8lOPHYSlCALRESEARCHCOMMUNlCATlONS

washed in 75 mM phosphoric acid and placed in a vial containing Aquasol (New England Nuclear). Duplicate of papers were prepared from each reaction mixture and 32P incorporated into the peptide was quantitated in a liquid scintillation counter. The kinase activity of neuraminidase-treated receptor was expressed as percent of that of the receptor without the enzyme-treatment.

RESULTS

The effect of neuraminidase treatment on insulin-binding activity of the

purified receptor was examined. As summarized in Table 1, insulin-binding acti-

vity was significantly enhanced after the receptor was treated with neuramini-

dase (2-10 ug/ml) at 37°C for 1 h. Neither 5-galactosidase treatment of the

intact receptor nor that of neuraminidase treated-receptor affected its insulin-

binding activity (data not shown).

In order to determine the effect of neuraminidase treatment on kinase acti-

vity, the enzyme-treated receptor was assayed by phosphorylation of the

src-related peptide (Table 2). The peptide assay revealed that neuraminidase-

treated receptor exhibits higher kinase activity than the receptor without the

enzyme treatment in the absence and presence of insulin (Table 2). These

results suggest that neuraminidase treatment of the purified insulin receptor

enhances both insulin-binding and kinase activities.

SDS-PAGE analysis of the receptor after neuraminidase treatment revealed

that the mobility of the enzyme-treated a and 5 subunits was increased as pre-

viously observed by others (16-17) (Fig. 1, lane l-4).

TABLE 1 The effect of neuraminidase treatment on insulin-binding activity of the

purified insulin receptor

Neuraminidase

w/ml

Insulin-binding activity(a) %

Average + SD No. of Experiments

0 100

1 126.5 + 20.0 4

2 143.0 2

5 150.8 + 14.7 9

10 168.4 + 18.6 5

(a)Insulin-binding activity was measured at a concentration of 0.8ng of 1251-insulin/ml as previously described (11).

741

Vol. 129, No. 3, 1985 8lOCHEMlCALANDBlOPHYSlCALRESEARCHCOMMUNlCATlONS

TABLE 2

The effect of neuraminidase on kinase activity of the purified insulin receptor in the absence and presence of insulin

Neuraminidase Insulin

a/ml !JM

Kinase activity(a)

Average + SD No. of Experiments

0 0 100

(a) Kinase activity was measured using the synthetic peptide resembling the autophosphoyrlation site of pp60SrC as described in Methods.

(b) Insulin stimulated kinase activity an average 3.1-fold.

To determine whether or not neuraminidase itself is really responsible for

enhancement of the insulin-binding and kinase activities and increased mobility

of the subunits on SDS-PAGE, the neuraminidase inhibitor, 2,3-dehydro-neuraminic

123456789

i (II Mrx 10m3

- 200

- 116

* ',% 3-l i p= ^i I^ - 92

- 66

- 45

Neuraminidase (Pgl ml 1 012512550

23'7e;gy NANA 0 0 0 0 2 2 4 8 8

Fig. 1.. The effects of neuraminidase and neuraminidase plus inhibitor on the mobilltles of insulin receptor subunits on SDS-PAGE. The purified insulin receptor (-4l.5 ug) was incubated at 4°C for 14 hr with no addition (lanes 1,9) 1 ug/ml (lanes 2,5), 2 ug/ml (lanes 3,6) and 5 ug/ml (lanes 4,7,8) of neuramini- dase, in the absence (lanes 1,2,3,4) and presence (lanes 5,6,7,8,9) of 2.3-dehydro NANA (lanes 5,6; &M, lane 7; 4mM, lanes 8,9; 8mM). Each sample was analyzed by SDS-PAGE (18) under reducing conditions, and stained with silver.

742

Vol. 129, No. 3, 1985 BIOCHEMICALAND BIOPHYSICALRESEARCH COMMUNICATIONS

TABLE 3

The efflect of neuraminidase inhibitor on insulin binding (a) and kinase activity (b) of the purified insulin receptor during neuraminidase treatment

Neuraminidase (us/ml)

P,3-Dehydro neuraminic acid

(mM) 0 1 5

(a) 0 100 125 147 0.1 --- ;; 149 1.0 --- 125 2.0 95 87 113

(b) 0 100 174 8 103 106

(a)The binding assay was done in duplicate. The mean values are listed. The two individual values did not differ by more than 107 from the mean value.

(b)Kinase activity was measured using the Src-related peptide in the absence of insulin as described in Methods.

acid (2,3-dehydro NANA) was incubated together with the enzyme. The neuramini-

dase inhibitor prevented the effect of neuraminidase treatment on insulin-

binding and kinase activities (Table 3). An increase in the electrophoretic

mobilities of both a and B subunits was also prevented by the presence of

2,3-dehydro NANA (Fig. 1, 5-9). These studies strongly suggested that terminal

sialic acid residues have a significant role in both insulin-binding and kinase

activities of the insulin receptor, and excluded the possibility that proteoly-

tic contaminants from the neuraminidase preparation induced such changes.

DISCUSSION

The role of terminal carbohydrate residues in insulin binding has been

studied by several groups (5-8). In these studies, membrane-bound or

solubilized receptor preparations were treated with glycosidases. In summary,

glycosidase treatment did not affect the binding activity (5-7) or decreased the

activity (8). In contrast to these studies, we have found, using purified insu-

lin receptor, that neuraminidase treatment enhances insulin-binding activity.

Scatchard analyses (19) of the receptor with or without neuraminidase treatment

indicated that the enhanced insulin-binding activity was caused by affinity

743

Vol. 129, No. 3, 1985 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

changes (data not shown). In some previous studies (5,7,8), since no evidence

was shown to indicate that the glycosidases were active on the receptor, the

lack of an effect of neuraminidase treatment on insulin-binding activity could

be explained by incomplete digestion of the receptor. On the other hand, Kasuga

et al. showed the evidence for removal of sialic acid residues from the insulin

receptor, but they still did not observe any changes in insulin-binding acti-

vity (6). The reason for their results could be explained by a contamination of

the neuraminidase preparation with proteases which could cancel the effect of

neuraminidase treatment. The difference between our results and previous

reports is most likely to be attributed to the purity of the assay system. The

present study using purified receptor should also be distinguished from somewhat

similar studies using intact cells, since protein synthesis and degradation are

not involved in our assay system.

Effects of the sialic acid removal on tyrosine-specific protein kinase

activity have never been reported. Our present study using purified receptor

revealed that sialic acid removal enhances the kinase activity to a similar

extent as is observed with insulin-binding activity. Preliminary Lineweaver-

Burk plot (20) analysis on the kinase activity of the receptor with or without

neuraminidase treatment showed an increase in the Vmax for the peptide after the

treatment.

The degree of enhancement in the receptor activities could be changed if

heterogeneity in terms of sialic acid contents exists in the purified receptor

preparations. In fact, our purified receptor subunits have been shown to have

p1 heterogeneity by O'Farrell's two-dimentional gel analysis (14), which is most

likely caused by heterogeneity in sialic acid contents. Terminal sialic acid

residues could have been removed from a certain population of the receptor mole-

cules during purification or even while the receptors are still in the membra-

nes. Such sialic acid-free receptors would exhibit higher binding and kinase

activities than an intact receptor. Thus, depending on the amount of sialic

acid free-receptors in the preparation, the enhancement of the receptor

activities after neuraminidase treatment is expected to be different. This

744

Vol. 129, No. 3, 1985 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

would be one of the reasons for rather high errors obtained in the present

experiments. Whether sialic acid residues are directly involved in insulin

binding sites and kinase active sites or in holding a certain conformation of

the receptor molecule is not known. However, it is of particular interest that

sialic acid-free receptor shows an increase in insulin-binding and kinase acti-

vities, which suggests that an intact insulin receptor is not fully active.

ACKNOWLEDGMENT

We thank Dr. S.W. Tanenbaum for a gift of 2,3-dehydro NANA methyl ester, and S. Hartmann for preparing placental membranes. We also thank M. Neermann for typing this manuscript. This work was supported by Grants from the National Institute of Arthritis, Diabetes, and Digestive and Kidney Diseases (AM29770 and AM34427).

1.

2.

3. 4.

2

7.

8.

9.

10.

11.

12.

13.

14. 15.

16.

17.

18. 19. 20.

REFERENCES

Czech, M.P., Massague, J., and Pilch, P.F. (1981) Trends Biockem. sci. 5, 222-225. Cuatrecasas, P., and Tell, G.P.E. (1973) Proc. mtZ. ACCZ~. sci. USA, 70, 485-489. Hedo, J.A., Harrison, L.C., and Roth, J. (1981) Biochemistry 20, 3385-3393. Hedo, J.A., Kasuga, M., Van Obberghen E., Roth, J., and Kahn,T.R. (1981) Proc. watt. Acad. SC?:. USA,78, 4791-4795. Cuatrecasas, P., and Illiano, G. (1971) c7. niol. fiem. 246, 4938-4946. Kasuga, M., Ezaki, O., Okanuma, Y., and Kosaka, K. (198rHom. Met&. qes. 12, 494-496. Clark, S., DeLuise, M., Larkins, R.G., Melick, R.A., and Harrison, L.C. (1978) Kockem. ,T. 174, 37-43. Caron, M., Picard, J., and Kern, P. (1978) Riockim. ~iophys. Acta, 512, 29-40. Hedo, J.A., Kahn, C.R., Hayashi, M., Yamada, K.M., and Kasuga, M. (1983) J. ,X02. aem. 258, 10020-10026. Ronnett, G.V., Knutson, V.P. Kohanski, R.A., Simpson, T.L., and Lane, M.D. (1984) .T. R?:OZ mem. 259, 4566-4575. Fujita-Yamaguchi, Y., Choi, S., Sakamoto, Y., and Itakura, K. (1983) ,T. viol. fiem. 258, 5045-5049. Kasuga, M., Fujita-Yamaguchi, Y., Blithe, D.L., and Kahn, C.R. (1983) Proc. VU~Z. Acad. ~ci. USA 80, 2137-2141. Kasuga, M., Fujita-Yamaguchi, Y., Blithe, D.L., White, M.F., and Kahn, C.R. (1983) J. Rio%. &em. 258, 10973-10980. Fujita-Yamaguchi, Y. (1984) 6. niot. &em.=, 1206-1211. Kumar, V., Kessler, J., Scott, M.E., Patwardhan, B.M., Tanenbaum, S.W. and Flashner, M. (1981) Carbohydrates Res. 94, 123-130. Jacobs, S., Hazum, E., and Cuatrecasas, P. (1980) Riockem. ~iophys. Res. corrimm. 94, 1066-1073. Heidenrezh, K.A., Zahniser, N.R., Berhanu, P., Brandenburg, D., and Olefsky, J.M. (1983) ,T. niot. &em. 258, 8527-8530. Laemmli, U.K. (1970) Nature (London) 227, 680-685. Scatchard, G. (1949) AWZ. N.Y. Acad. xi. s, 660-672. Lineweaver, H. and Burk, D. (1934) J. Amer. fiem. SOC. 3: 658-666.

745