Vol. 168, No. 3, 1990
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
May 16, 1990
Pages
1201-1210
Insulin-like Growth Factor-IVMannose 6-Phosphate Receptor Is Incapable of Activating GTP-binding Proteins in Response to Mannose 6Phosphate, But Capable in Response to Insulin-like Growth Factor-II”” Tak,ashi
Okamoto#, lkuo Nishimoto#§l, Yoshitake Yoshihiro Ohkuni##, and Etsuro Ogata#
Murayama#,
#Life Science Laboratory Fourth Department of Internal Medicine University of Tokyo School of Medicine Tokyo 112, Japan SDepartment of Medicine Clinical Pharmacology Stanford University School of Medicine Stanford, CA 94305 Received March 28, 1990
We previously reported that insulin-like growth factor-II (IGF-II) stimulates both calcium influx and DNA synthesis by acting on the cell surface IGF-II receptor (IGF-IIR) in a manner sensitive to pertussis toxin, and recently demonstrated that IGF-II binding to the IGF-IIR gives rise to functional changes of purified Gi-2, a GTP-binding protein (G protein) in
**This work was supported by grants from the Ministry of Education, Science and Culture of Japan, the Ichiro Kanehara Foundation, and the Naito Foundation. vo
whom correspondence
should be addressed.
The abbreviations used are: G protein, guanine nucleotide-binding regulatory protein; IGF, insulin-like growth factor; man6P, mannose 6-phosphate; IGFIIR, IGF-II receptor; CI-MPR, cation-independent mannose 6-phosphate receptor; IAP, islet-activating protein, Gi-like protein, pertussis toxin-sensitive G protein with a 40-k-41-kDa a subunit; IGF-IIR/Gi-2 vesicles, vesicles reconstituted with the IGF-IIR and Gi-2; GTPyS, guanosine-S-o-(3-thiotriphosphate); CHAPS, 3- { (3-cholamidopropyl) dimethyl ammonio} - lpropame sulfonic acid; EDTA, ethylene diamine tetraacetic acid; Hepes, 4-(2hydroxyethyl)- 1-piperazineethane sulfonic acid; ECso, half maximum concentration. 0006-291X/90 1201
$1.50
Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
Vol.
168,
No.
3, 1990
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
phospholipid vesicles as well as in broken cell membranes. On the other hand, a variety of evidence indicates that the IGF-IIR binds mannose 6phosphate (man6P) with high affinity probably at a receptor extracellular region different from the IGF-II-binding site. In the present study, we examined whether man6P stimulation of the IGF-IIR evokes the activation of Gi-2 in phospholipid vesicles and in native cell membranes. In vesicles reconstituted with purified rat IGF-IIR and bovine Gi-2, man6P did not stimulate GDP dissociation from Gi-2 even in concentrations up to 10 mM, while IGF-II dosedependently facilitated GDP release from Gi-2 with an EC50 of 6 nM. The stimulatory effect of IGF-II was not observed in vesicles reconstituted with Gi2 alone. In addition, also in a native environment of cell membranes, man6P did not affect an endogenous 40-kDa protein or exogenously added purified Gi2 as assessed with reduction of the pertussis toxin-catalyzed ADP-ribosylation. These results indicate that the IGF-IIR does not activate Gi-like proteins upon man6P binding in phospholipid vesicles and in native cellular membranes, whereas the receptor activates Gi-like proteins upon IGF-II binding in both environments. Thus, we postulate that the IGF-IIR dissimilarly responds to the two structurally unrelated ligands, IGF-II and man6P, in the linkage function 01990Academic Press,Inc. with G proteins. Insulin-like growth factor (IGF )-II is a polypeptide structurally related to IGF-I and insulin (1). These related growth factors or hormones exert a wide spectra of biological actions on cellular proliferation and metabolism both in vitro and in vivo by binding the cell surface receptors specific for each of them (2-3). In spite of their similarities, the IGF-IIR has strikingly different structural and biochemical properties from the IGF-I receptor and the insulin receptor which are very close in both properties. The latter two receptors are comprised of disulfide-linked subunits of Mt=l15-135kDa and 90-95kDa, which retain, respectively, an agonist-binding site and intrinsic protein tyrosine kinase activity that is inferred to be involved in transmembrane signaling of their ligands (4). The IGF-IIR is quite distinct from them, and its function remains still uncertain. It is composed of a single protein of Mr=220-250-kDa without kinase activity (1). The cross-reactivity of IGF-II to the IGF-I and insulin receptors, and the established role of these receptors in signal transduction precluded us from discretely analyzing the transmembrane signaling function of the IGF-IIR. Recently, the molecular cloning of the IGF-IIR complementary DNAs (5-6) confounded the functions of the IGF-IlR. The primary sequence of the human IGF-IIR has turned out to be 80% and 99% homologous, respectively, to those of the bovine and human CI-MPR (5, 7-S), a man6P receptor implicated in the sorting of lysosomal enzymes (9). Subsequent research (6, 10-13) indicated that the IGF-IIR and CI1202
Vol.
168, No. 3, 1990
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
MPR each have a capacity to bind IGF-II and man6P with similar binding profiles between the two receptors, indicating that these receptors are probably identical. Of further interest, it was suggested that IGF-II and man6P bind to separate sites on the receptors, since neither man6P or IGF-II had an inhibitory effect on the binding of the other ligand to the IGF-IIR or CI-MPR (10, 13). This suggests the assumption that IGF-II and man6P bind to different sites of the IGF-IIR/CI-MPR and the binding of each ligand may have a respective biological significance of its own. Thus, it should be elucidated whether IGF-II and man6P each propagate different signals which correspond to their bindings to the IGF-IIR/CI-MPR. Prior work in our laboratory (14) has shown that IGF-II increases intracellular calcium concentrations, stimulates 45Ca influxes, activates a calcium-permeable cation channel and promotes DNA synthesis by acting on the IGF-IIR in mouse BALB/c-3T3 fibroblasts. Although it has been reported that IGF-II may exert its function by binding to the IGF-I receptor or the insulin receptor in several cells (1518), other laboratories have also determined that the IGF-IIR is able to mediate some biological actions of IGFII (19-23). In addition, we found that pertussis toxin potently inhibits the actions of IGF-II on Ca2+ mobilization and DNA synthesis, and the toxininduced ADP-ribosylation of a 40-kDa membrane protein coincided well with the inhibitory effects of pertussis toxin in BALB/c-3T3 cells (24). In an environment of BALB/c-3T3 cell membranes, a non-hydrolyzable GTP analogue, GTPyS reduces the affinity of IGF-II binding (24). These results suggested the hypothesis that the IGF-IIR may be linked directly with a Gi-like protein. Recently, we have demonstrated that in phospholipid vesicles, the purified rat IGF-IIR is coupled directly to porcine Gi-2 which has a 40-kDa a subunit, in response to IGF-II binding (25). G proteins are membrane-bound proteins comprised of a and Pr subunits which regulate specific ligand-dependent transmembrane signals by modulating the activity of membrane effecters (26). It is therefore reasonable that the IGF-IIR which lacks protein kinase activity develops transmembrane signals through the activation of G proteins. Since the CI-MPR plays a key role in the sorting of lysosomal enzymes upon man6P binding (9) and is a molecule almost identical to the IGF-IIR (5-Q it would be of significant interest to examine the effect of man6P on Gi-2 via the IGF-IIR. We conducted the present study to clarify whether man6P binding to the IGF-IIR produces the activation of G proteins in phospholipid vesicles and native cell membranes. The results indicate that man6P does not induce the interaction of the IGF-IIR with G proteins in phospholipid vesicles or in cell membranes. Thus, the IGF-IIR evokes a different response after binding man6P from the response after binding IGF-II. 1203
Vol. 168, No. 3, 1990
Experimental
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
Procedures
Cell culture: Culture of mouse BALB/c-3T3 fibroblasts (clone A31) and human erythroleukemia K562 cells was described previously (25). For pertussis toxin treatments in intact cells, cells were incubated with pertussis toxin (100 rig/ml) for 4 hours in their normal culture media at 37°C. Membrane nrenarations and ADP-ribosvlation: Crude membranes were prepared at 0°C and ADP-ribosylated by preactivated pertussis toxin as described (25). Reconstitution of Purified G nroteins with Membranes from Pertussis toxin-treated Cells: One hundred nanograms of Gi-2 were incubated with 50 pl of buffer B supplemented with IGF-II in the presence of 100 pg membrane preparations from pertussis toxin-treated cells at 35°C for 5 min, and then ADP-ribosylated by adding 50 p.1 of buffer C as described (25). The total concentration of CHAPS contained in the membrane preparations and G proteins was rendered less than 0.07%. Purification of IGF-IIR from Rat Placenta: Rat IGF-IIR with molecular mass of 250~kDa was purified from 17-day gravid rat placenta to homogeneity (Fig. 1A) as described previously (25). Reconstitution of Purified IGF-IIR and Gi-2 in Phosnholinid Vesicles: The purified IGF-IIR and bovine Gi-2 were mixed with azolectin (1 mg/ml), and reconstituted by gel filtration through a Sephadex G-50 column as described (25). Measurement of GDP Release from G proteins: Gt-2 was given prior treatment with 7.5 pM r3H]GDP in the presence of 10 mM Mg2+ at room temperature for 8 hours. The labeled Gi-2 was chromatographed through a Sephadex G-50 column with an eluant of 20 mM Hepes/NaOH (pH 7.5), 1 mM EDTA, 1 mM dithiothreitol, 100 mM NaCl and 0.7% CHAPS. Twenty to twenty five nanomolars of labeled G protein preparations were incubated with agonists in 20 mM Hepes/NaOH (pH 7.4), 1 mM EDTA, 1.02 mM MgCh, and 1 pM GTP at 37°C. Incubations were terminated by adding ten volumes of the ice-cold stopping buffer (100 mM Tris/HCl (pH SO), 25 mM MgC12, and 100 mM NaCl). Materials: Bovine brain Gi-1, Gi-2 and GO were generously provided by Dr. Toshiaki Katada (Tokyo Institute of Technology, Yokohama, Japan). We employed rat IGF-II (MSA 111-2) donated by Dr. S. P. Nissley (National Cancer Institute, Bethesda, MD) and recombinant human IGF-II purchased from Amgen. Either IGF-II yielded similar results. [3H]GDP and [32P]NAD were from Du-Pont New England Nuclear. Man6P and pertussis toxin were obtained from Sigma and Funakoshi (Japan), respectively. 1204
Vol.
BIOCHEMICAL
168, No. 3, 1990
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
Results Effect of Man6P and IGF-II on Gi-2 in Reconstituted Vesicles: In order to clarify whether man6P binding brings about the coupling of the IGF-IIR with Gi-2, we tested the effect of man6P in vesicles reconstituted with the purified IGF-IIR and Gi-2. As described previously (25), the IGF-IIR and Gi-2 were each purified to homogeneity and inserted into phospholipid vesicles. When the vesicles were stimulated with 100 nM of IGF-II, the GDP dissociation rate of Gi-2 was increased approximately twofold (Fig. 1B). The stimulatory effect of a 5-min IGF-II incubation on GDP release from Gi-2 was dose-dependently observed and had an EC50 of approximately 6 nM for IGF-II in the IGFIIR/Gi-2 vesicles. The stimulator-y effect of IGF-II could not be observed in the vesicles reconstituted with Gi-2 alone (not shown). In contrast, man6P had no significant effect on GDP release from Gi-2 in concentrations up to 10 mM in the IGF-IIR/Gi-2 vesicles. In either experimental condition, i.e., pH range between 6.5 and 8.0 or Mg2+ concentrations between =O and 10 mM, we observed no effect of man6P on GDP release from Gi-2 in the IGF-IIR/Gi-2 vesicles (not shown). We have already reported that IGF-II stimulates GTP@ binding to Gi-2 in IGF-IIR/Gi-2 vesicles (25). However, since GDP release is thought to be a more critical step on which receptor stimulation directly acts
0
200 -
z
60
4 3
60
al s 40
69
L E i
-
% 0 " [IGF-II]
92.5 45 -1
2
20
0
[manGP]
ji_::: 0 O-
1
10
100
1000
WV
0.01
0.1
1
10
@Ml
Fig. 1: (A) IGF-II receptor preparation purified from rat placenta. The IGFIIR was purified to homogeneity by means of three steps of affinity chromatographies. The receptor was electrophoresed in a 7.5% SDS-gel and stained with silver (lane 2). Lane 1 indicates Laemmli sample buffer alone. (B) Effect of man6P and IGF-II on GDP release from Gi-2 in IGF-IIR/Gi-2 vesicles. GDP release from Gi-2 was measured in a 5 min incubation of the IGF-IIR/Gi-2 vesicles with increasing concentrations of mau6P (open circle) or IGF-II (closed circle). GDP dissociation represents a fraction after subtracting GDP-bound Gi-2 from the total Gi-2, and is indicated in the figure as a rate per 5 min. Values represent means of three experiments. S. E. was less than 5% of each value. 1205
BIOCHEMICAL
Vol. 168, No. 3, 1990
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
(27), this result provides compelling evidence that the IGF-IIR activates Gi-2 in response to IGF-II binding in a manner quite similar to conventional Gcoupled receptor. Furthermore, it indicates that man6P binding to the IGF-IIR did not give rise to the activation of Gi-2. Effect of Man6P or IGF-II on 40-kDa Gi-like Proteins in Native Membranes: To examine whether man6P stimulation of the IGF-IIR does not induce a change of Gi-like proteins in a native environment of cell membranes, we tested the effect of man6P in human erythroleukemia K562 cell membranes. Since the K562 cells exhibit the typical IGF-IIR, but no IGF-I receptor (2021), the effect of IGF-II is considered as mediated by either IGF-IIR or insulin receptor. When K562 cell membranes were incubated for 10 min with man6P in concentrations up to 10 mM, we observed no significant change of the pertussis toxin-induced ADP-ribosylation of the endogenous 40-kDa protein in the K562 cell membranes (Fig. 2A). In addition, when the membranes were A
C B
68
-
4s
-
30
-
0
0.1 I man6Pl (mfl)
I
IO
68
-
45
-
30
-
IGF-II IOnM
68
-
4s
-
30
-
0 0’
5’
IO’
15’
30’
60’
01
I
I man6PI CrnM)
Fig. 2: Effect of man6P on the pertussis toxin-catalyzed ADP-ribosylation activity of 40-kDa Gi-like proteins in K562 cell membranes. (A) Dose response relation for the effect of man6P. Membranes were incubated with increasing concentrations of man6P for 5 min and then ADP-ribosylated with pertussis toxin as described under “Experimental Procedures”. The right lane indicates the effect of 10 nM IGF-II in the same conditions as man6P incubation (B) Time course of man6P effect on the 40-kDa pertussis toxin substrate protein. Membranes were incubated with 10 mM man6P for indicated periods and ADP-ribosylated with pertussis toxin. (C) Effect of man6P on exogenously added Gi-2 in membranes from pertussis toxin-treated KS62 cells. The toxin-pretreated membranes supplemented with purified Gi-2 were incubated with increasing concentrations of man6P for 5 min, and then ADP-ribosylated. The right lane shows the effect of 10 nM IGF-II on Gi-2 observed in the same conditions. 1206
IO
IGF-II’ IOnM
Vol. 168, No. 3, 1990
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
incubated with 10 mM man6P, no significant change in the pertussis toxin substrate activity of the endogenous 40-kDa protein was observed over the course of a 60 min incubation (Fig. 2B). The inability of man6P to act on the 40-kDa pertussis toxin substrate protein was also the case with K562 cells membranes supplemented with purified Gi-2. As shown in Fig. 2C, man6P elicited no change in the pertussis toxin substrate activity from Gi-2 exogenously added to the pertussis toxin-treated KS62 cell membranes. Although 10 mM man6P sometimes affected the ADP-ribosylation of Gi-2, this was not reproducible. Even when Gi-1 or Go was incubated with man6P in the presence of toxin-pretreated K562 cell membranes, we could observe no change in the toxin substrate activity of either G proteins (not shown). In contrast, 10 nM IGF-II reduced the toxin sensitivity of the 40-kDa protein by approximately l/S in either experiments from K562 cell membranes or from K562 cell membranes supplemented with purified Gi-2. The effect of IGF-II could not be reproduced by IGF-I or insulin which had no effect in these cell membranes (data not shown). Essentially similar results were obtained by using mouse BALB/c-3T3 cell membranes where the toxin sensitivity of endogenous and exogenous 40-kDa Gi-like proteins was not affected by man6P, but reduced by IGF-II (not shown). Discussion We documented herein that the purified rat IGF-IIR could respond dissimilarly to the two distinct high affinity ligands, IGF-II and man6P. It is well known that these two structurally unrelated ligands bind to the different site on the IGF-IIR with high affinity, and that they evokes different cellular events; IGF-II binding induces metabolic/mitogenic stimulations and man6P binding activates sorting/endocytosis processes. When the IGF-IIR was stimulated with 100 nM of IGF-II, the GDP release rate of Gi-2 was augmented approximately twofold in phospholipid vesicles reconstituted with the IGF-IIR and Gi-2. Furthermore, the action of IGF-II was observed in a dose-dependent manner for IGF-II with an ECso of 6 nM and only in the presence of the IGFIIR. We have described that IGF-II stimulates GTPyS binding to Gi-2 in the IGF-IIR/Gi-2 vesicles (25). The IGF-II potency and the intensity of the IGF-II action in the stimulation of the GDP release rate of Gi-2 correlated well with the IGF-II potency and intensity in the stimulation of the GTP+ySbinding rate of the G protein. Moreover, the extent of IGF-IIR-induced increase in guanine nucleotide metabolism on Gi-2 was comparable to that reported in conventional Gi-coupled receptors, muscarinic receptors (28) and az-adrenergic receptors (29). Therefore, the present study indicates that the IGF-IIR activates Gi-2 in response to IGF-II binding as significantly and as similarly as physiological Gicoupled receptors. 1207
Vol.
168, No. 3, 1990
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
In marked contrast, man6P stimulated no GDP release from Gi-2 in concentrations up to 10 mM in the IGF-IIR/Gi-2 vesicles. MacDonald et al. (6) have reported that the purified rat IGF-IIR binds man6P with a high affinity. Hence, this result indicates that the purified IGF-IIR does not activate Gi-2 upon man6P binding. In addition, it has been demonstrated that man6P and IGF-II each bind to the separate sites of the rat IGF-IIR (30). It is therefore suggested that the ligand-dependent stimulations of the different extracellular regions of the IGF-IlR lead to the qualitatively different intracellular reactions of the receptor. Such dual functions of the IGF-IIR have been preliminarily reported by Rogers and Hammerman (31). They described that man6P does not stimulate polyphosphoinositide breakdown in canine renal cells, despite the fact that IGF-II does so. However, it remained unclear whether IGF-II and man6P acted on the same receptor protein in their study. The present study indicates that the single IGF-IIR has a multifunctional property, not only in the aspect of ligand binding but also the receptor signaling. We next examined whether the multifunctional response of the IGF-IIR is observed even in an environment of cell membranes which more approximates a physiological condition. The result also indicated that man6P had no effect on either the endogenous 40-kDa protein or Gi-2 in concentrations up to 10 mM in human K562 cell membranes. It has been reported that in both these cell membranes, IGF-II potently attenuates the pertussis toxin-catalyzed ADP-ribosylation of the 40-kDa Gi-like proteins, mainly through the IGF-IIR (25). Furthermore, such a modification of Gi-like proteins is likely to represent a subunit dissociation of them (25), an essential aspect of G protein activation (32). Therefore, we assumed that a native environment of cell membranes maintains the multifunctional property of the IGF-IIR observed in reconstituted vesicles; the IGF-II binding leads to activation of Gi-2, while the man6P binding does not do so. The other possible explanation for the lack of effects of man6P in these cell membranes is that man6P does not bind to the human IGF-IIR. For example, Canfield and Komfeld (33) have reported that the chicken CI-MPR, a possible homologue of the chicken IGF-IIR, lacks the high affinity binding site for IGF-II, while the receptor retains that for man6P. However, the human IGF-IIR exhibits the high affinity binding for man6P (34). Furthermore, K562 cell membranes demonstrate the high affinity binding of man6P as assessed with manBPinhibitable P-glucuronidase binding, which reflects specific man6P binding to the IGF-IIR (unpublished observation: Murayama, Y. and Nishimoto, I.). It is therefore likely that in the native cell membranes man6P binds to the IGF-IIR, but does not produce the possible activation of Gi-like proteins. Given the assumption that the IGF-IIR is identical to the CI-MPR, the binding of man6P to the IGF-IIR may evoke the sorting or endocytosis of man6Pcontaining proteins. It has been reported recently that the sorting and 1208
Vol.
168, No. 3, 1990
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
endocytosis functions of the CI-MPR/IGF-IIR may involve a G proteinmediated process (35). On the other hand, the present study indicates that man6P-induced stimulation of the IGF-IIR does not produce the activation of Gi-like proteins in phospholipid vesicles or in native cell membranes. This suggests that the signal for sorting may not accompany the activation of Gilike proteins. Furthermore, we describe here that the single IGF-IIR responds differently to man6P and IGF-II that bind to separate binding regions of the receptor. Studies need to be done to elucidate the molecular mechanism underlying the unique dual function of the IGF-IIR in response to the two distinct ligands, IGF-II and man6P. Acknowledgments We are especially indebted to Dr. Toshiaki Katada for purified G proteins and helpful discussion. We also thank Dr. S. Peter Nissley for his gift of rat IGF-II; Drs. William S. Sly of St. Louis University, Michio Ui of University of Tokyo, Taroh Iiri of Tokyo Institute of Technology, Shintaro Iwashita of Mitsubishi-Kasei Institute of Life Science, Yumi and Yoshiomi Tamai of the Society For Educational Aid To Traffic Orphans for indispensable support during the course of this study. Technical assistance of Tomomi Konishi is gratefully acknowledged. References 1. Rechler, M. M. and Nissley, S. P. (1985) Annu. Rev. Physiol. 47,425-442 2. Kasuga, M., Van Obberghen, E., Nissley, S. P. and Rechler, M. M. (1981) J. Biol. Chem. 256,5305-5308 3. Massague, J. and Czech, M. P. (1982) J. Biol. Chem. 257,5038-5045 4. Morgan, D. 0. and Roth, R. A. (1987) Proc. Natl. Acad. Sci. U.S.A. 84, 41-45 5. Morgan, D. O., Edman, J. E., Standring, D. N., Fried, V. A., Smith, M. C., Roth, R. A. and Rutter, W. J. (1987) Nature 329, 301-307 6. MacDonald, R. G., Pfeffer, S. R., Coussens, L., Tepper, M. A., Brocklebank, C. M., Mole, J. E., Anderson, J. K., Chen, E., Czech, M. P. and Ullrich, A. (1988) Science 239, 1134-1137 7. Lobel, P., Dahms, N. M. and Kornfeld, S. (1988) J. Biol. Chem. 263,25632570 8. Oshirna, A., Nolan, C. M., Kyle, J. W., Grubb, J. H. and Sly, W. S. (1988) J. Biol. Chem. 263,2553-2562 9. von Figura, K. and Hasilik, A. (1986) AMU. Rev. Biochem. 55, 167-193 10. Roth, R. A., Stover, C., Hari, J., Morgan, D. O., Smith, M. C., Sara, V. and Fried, V. A. (1987) Biochem. Biophys. Res. Commun. 149,600-606 11. Kiess, W., Blickenstaff, G. D., Sklar, M. M., Thomas, C. L., Nissley, S. P. and Sahagian, G. G. (1988) J. Biol. Chem. 263,9339-9344 1209
Vol. 166, No. 3, 1990
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
12. Brat&e, T., Causin, C., Waheed, A., Junghans, U., Hasilik, A., Maly, P., Humbel, R. E. and von Figura, K. (1988) Biochem. Biophys. Res. Commun. 150,1287-1293 13. Tong, P., Tollefsen, S. E. and Kornfeld, S. (1989) J. Biol. Chem. 263, 25852588 14. Nishimoto, I. and Kojima, I. (1989) N. I. P. S. 4, 94-97 15. Yu, K-T. and Czech, M. P. (1984) J. Biol. Chem. 259,3090-3095 16. Conover, C. A., Rosenfeld, R. G. and Hintz, R. L. (1987) J. Cell. Physiol. 133, 560-566 17. King, G. L., Kahn, C. R., Rechler, M. M. and Nissley, A. P. (1980) J. Clin. Invest. 66, 130-140 18. Mottola, C. and Czech, M. P. (1984) J. Biol. Chem. 259, 12705-12713 19. Hari, J., Pierce, S. B., Morgan, D. O., Sara, V., Smith, M. C. and Roth, R. A. (1987) EMBO J. 6,3367-3371 20. Tally, M., Li, C. H. and Hall, K. (1987) Biochem. Biophys. Res. Commun. 148, 811-816 21. Blanchard, M. M., Barenton, B., Sullivan, A., Foster, B., Guyda, H. J. and Posner, B. I. (1988) Mol. Cell. Endocrinol. 56, 235-244 22. Simizu, M., Webster, C., Morgan, D. O., Blau, H. M. and Roth, R. A. (1986) Am. J. Physiol. 251, E611-E615 23. Rogers, S. A. and Hammerman, M. R. (1988) Proc. Natl. Acad. Sci. U. S. A. 85,4037-4041 24. Nishimoto, I., Hata, Y., Ogata, E. and Kojima, I. (1987) J. Biol. Chem. 262, 12120-12126 25. Nishimoto, I., Murayama, Y ., Katada, T., Ui, M. and Ogata, E. (1989) J. Biol. Chem. 264,14029-14038 26. Gihnan, A. G. (1987) Amru. Rev. Biochem. 56,615-650 27. Ferguson, K. M., Higashijima, T., Smigel, M. D. and Gihnan, A. G. (1986) J. Biol. Chem. 261,7393-7399 28. Cerione, R. A., Regan, J. W., Nakata, H., Codina, J., Benovic, J. L., Gierschik, P., Somers, R. L., Spiegel, A. M., Birnbaumer, L., Lefkowitz, R. J., and Caron, M. G. (1986) J. Biol. Chem. 261, 3901-3909 29. Kurose, H., Katada, T., Haga, T., Haga, K., Ichiyama, A., and Ui, M. (1986) J. Biol. Chem. 261,6423-6428 30. Roth, R. A. (1988) Science 239,1269-1271 31. Rogers, A. A., and Hammerman, M. R. (1989) J. Biol. Chem. 264,42734276 32. Neer, E. J. and Clapham, D. E, (1988) Nature 333, 129-133 33. Canfield, W. M. and Kornfeld, S. (1989) J. Biol. Chem. 264, 7100-7103 34. Nolan, C. M., Kyle, J. W., Watanabe, H. and Sly, W. S. (1990) Cell Regulation 1, 197-213 35. Goda, Y. and Pfeffer, S. R. (1988) Cell 55, 309-320 1210
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
May 16, 1990
Pages
1201-1210
Insulin-like Growth Factor-IVMannose 6-Phosphate Receptor Is Incapable of Activating GTP-binding Proteins in Response to Mannose 6Phosphate, But Capable in Response to Insulin-like Growth Factor-II”” Tak,ashi
Okamoto#, lkuo Nishimoto#§l, Yoshitake Yoshihiro Ohkuni##, and Etsuro Ogata#
Murayama#,
#Life Science Laboratory Fourth Department of Internal Medicine University of Tokyo School of Medicine Tokyo 112, Japan SDepartment of Medicine Clinical Pharmacology Stanford University School of Medicine Stanford, CA 94305 Received March 28, 1990
We previously reported that insulin-like growth factor-II (IGF-II) stimulates both calcium influx and DNA synthesis by acting on the cell surface IGF-II receptor (IGF-IIR) in a manner sensitive to pertussis toxin, and recently demonstrated that IGF-II binding to the IGF-IIR gives rise to functional changes of purified Gi-2, a GTP-binding protein (G protein) in
**This work was supported by grants from the Ministry of Education, Science and Culture of Japan, the Ichiro Kanehara Foundation, and the Naito Foundation. vo
whom correspondence
should be addressed.
The abbreviations used are: G protein, guanine nucleotide-binding regulatory protein; IGF, insulin-like growth factor; man6P, mannose 6-phosphate; IGFIIR, IGF-II receptor; CI-MPR, cation-independent mannose 6-phosphate receptor; IAP, islet-activating protein, Gi-like protein, pertussis toxin-sensitive G protein with a 40-k-41-kDa a subunit; IGF-IIR/Gi-2 vesicles, vesicles reconstituted with the IGF-IIR and Gi-2; GTPyS, guanosine-S-o-(3-thiotriphosphate); CHAPS, 3- { (3-cholamidopropyl) dimethyl ammonio} - lpropame sulfonic acid; EDTA, ethylene diamine tetraacetic acid; Hepes, 4-(2hydroxyethyl)- 1-piperazineethane sulfonic acid; ECso, half maximum concentration. 0006-291X/90 1201
$1.50
Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
Vol.
168,
No.
3, 1990
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
phospholipid vesicles as well as in broken cell membranes. On the other hand, a variety of evidence indicates that the IGF-IIR binds mannose 6phosphate (man6P) with high affinity probably at a receptor extracellular region different from the IGF-II-binding site. In the present study, we examined whether man6P stimulation of the IGF-IIR evokes the activation of Gi-2 in phospholipid vesicles and in native cell membranes. In vesicles reconstituted with purified rat IGF-IIR and bovine Gi-2, man6P did not stimulate GDP dissociation from Gi-2 even in concentrations up to 10 mM, while IGF-II dosedependently facilitated GDP release from Gi-2 with an EC50 of 6 nM. The stimulatory effect of IGF-II was not observed in vesicles reconstituted with Gi2 alone. In addition, also in a native environment of cell membranes, man6P did not affect an endogenous 40-kDa protein or exogenously added purified Gi2 as assessed with reduction of the pertussis toxin-catalyzed ADP-ribosylation. These results indicate that the IGF-IIR does not activate Gi-like proteins upon man6P binding in phospholipid vesicles and in native cellular membranes, whereas the receptor activates Gi-like proteins upon IGF-II binding in both environments. Thus, we postulate that the IGF-IIR dissimilarly responds to the two structurally unrelated ligands, IGF-II and man6P, in the linkage function 01990Academic Press,Inc. with G proteins. Insulin-like growth factor (IGF )-II is a polypeptide structurally related to IGF-I and insulin (1). These related growth factors or hormones exert a wide spectra of biological actions on cellular proliferation and metabolism both in vitro and in vivo by binding the cell surface receptors specific for each of them (2-3). In spite of their similarities, the IGF-IIR has strikingly different structural and biochemical properties from the IGF-I receptor and the insulin receptor which are very close in both properties. The latter two receptors are comprised of disulfide-linked subunits of Mt=l15-135kDa and 90-95kDa, which retain, respectively, an agonist-binding site and intrinsic protein tyrosine kinase activity that is inferred to be involved in transmembrane signaling of their ligands (4). The IGF-IIR is quite distinct from them, and its function remains still uncertain. It is composed of a single protein of Mr=220-250-kDa without kinase activity (1). The cross-reactivity of IGF-II to the IGF-I and insulin receptors, and the established role of these receptors in signal transduction precluded us from discretely analyzing the transmembrane signaling function of the IGF-IIR. Recently, the molecular cloning of the IGF-IIR complementary DNAs (5-6) confounded the functions of the IGF-IlR. The primary sequence of the human IGF-IIR has turned out to be 80% and 99% homologous, respectively, to those of the bovine and human CI-MPR (5, 7-S), a man6P receptor implicated in the sorting of lysosomal enzymes (9). Subsequent research (6, 10-13) indicated that the IGF-IIR and CI1202
Vol.
168, No. 3, 1990
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
MPR each have a capacity to bind IGF-II and man6P with similar binding profiles between the two receptors, indicating that these receptors are probably identical. Of further interest, it was suggested that IGF-II and man6P bind to separate sites on the receptors, since neither man6P or IGF-II had an inhibitory effect on the binding of the other ligand to the IGF-IIR or CI-MPR (10, 13). This suggests the assumption that IGF-II and man6P bind to different sites of the IGF-IIR/CI-MPR and the binding of each ligand may have a respective biological significance of its own. Thus, it should be elucidated whether IGF-II and man6P each propagate different signals which correspond to their bindings to the IGF-IIR/CI-MPR. Prior work in our laboratory (14) has shown that IGF-II increases intracellular calcium concentrations, stimulates 45Ca influxes, activates a calcium-permeable cation channel and promotes DNA synthesis by acting on the IGF-IIR in mouse BALB/c-3T3 fibroblasts. Although it has been reported that IGF-II may exert its function by binding to the IGF-I receptor or the insulin receptor in several cells (1518), other laboratories have also determined that the IGF-IIR is able to mediate some biological actions of IGFII (19-23). In addition, we found that pertussis toxin potently inhibits the actions of IGF-II on Ca2+ mobilization and DNA synthesis, and the toxininduced ADP-ribosylation of a 40-kDa membrane protein coincided well with the inhibitory effects of pertussis toxin in BALB/c-3T3 cells (24). In an environment of BALB/c-3T3 cell membranes, a non-hydrolyzable GTP analogue, GTPyS reduces the affinity of IGF-II binding (24). These results suggested the hypothesis that the IGF-IIR may be linked directly with a Gi-like protein. Recently, we have demonstrated that in phospholipid vesicles, the purified rat IGF-IIR is coupled directly to porcine Gi-2 which has a 40-kDa a subunit, in response to IGF-II binding (25). G proteins are membrane-bound proteins comprised of a and Pr subunits which regulate specific ligand-dependent transmembrane signals by modulating the activity of membrane effecters (26). It is therefore reasonable that the IGF-IIR which lacks protein kinase activity develops transmembrane signals through the activation of G proteins. Since the CI-MPR plays a key role in the sorting of lysosomal enzymes upon man6P binding (9) and is a molecule almost identical to the IGF-IIR (5-Q it would be of significant interest to examine the effect of man6P on Gi-2 via the IGF-IIR. We conducted the present study to clarify whether man6P binding to the IGF-IIR produces the activation of G proteins in phospholipid vesicles and native cell membranes. The results indicate that man6P does not induce the interaction of the IGF-IIR with G proteins in phospholipid vesicles or in cell membranes. Thus, the IGF-IIR evokes a different response after binding man6P from the response after binding IGF-II. 1203
Vol. 168, No. 3, 1990
Experimental
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
Procedures
Cell culture: Culture of mouse BALB/c-3T3 fibroblasts (clone A31) and human erythroleukemia K562 cells was described previously (25). For pertussis toxin treatments in intact cells, cells were incubated with pertussis toxin (100 rig/ml) for 4 hours in their normal culture media at 37°C. Membrane nrenarations and ADP-ribosvlation: Crude membranes were prepared at 0°C and ADP-ribosylated by preactivated pertussis toxin as described (25). Reconstitution of Purified G nroteins with Membranes from Pertussis toxin-treated Cells: One hundred nanograms of Gi-2 were incubated with 50 pl of buffer B supplemented with IGF-II in the presence of 100 pg membrane preparations from pertussis toxin-treated cells at 35°C for 5 min, and then ADP-ribosylated by adding 50 p.1 of buffer C as described (25). The total concentration of CHAPS contained in the membrane preparations and G proteins was rendered less than 0.07%. Purification of IGF-IIR from Rat Placenta: Rat IGF-IIR with molecular mass of 250~kDa was purified from 17-day gravid rat placenta to homogeneity (Fig. 1A) as described previously (25). Reconstitution of Purified IGF-IIR and Gi-2 in Phosnholinid Vesicles: The purified IGF-IIR and bovine Gi-2 were mixed with azolectin (1 mg/ml), and reconstituted by gel filtration through a Sephadex G-50 column as described (25). Measurement of GDP Release from G proteins: Gt-2 was given prior treatment with 7.5 pM r3H]GDP in the presence of 10 mM Mg2+ at room temperature for 8 hours. The labeled Gi-2 was chromatographed through a Sephadex G-50 column with an eluant of 20 mM Hepes/NaOH (pH 7.5), 1 mM EDTA, 1 mM dithiothreitol, 100 mM NaCl and 0.7% CHAPS. Twenty to twenty five nanomolars of labeled G protein preparations were incubated with agonists in 20 mM Hepes/NaOH (pH 7.4), 1 mM EDTA, 1.02 mM MgCh, and 1 pM GTP at 37°C. Incubations were terminated by adding ten volumes of the ice-cold stopping buffer (100 mM Tris/HCl (pH SO), 25 mM MgC12, and 100 mM NaCl). Materials: Bovine brain Gi-1, Gi-2 and GO were generously provided by Dr. Toshiaki Katada (Tokyo Institute of Technology, Yokohama, Japan). We employed rat IGF-II (MSA 111-2) donated by Dr. S. P. Nissley (National Cancer Institute, Bethesda, MD) and recombinant human IGF-II purchased from Amgen. Either IGF-II yielded similar results. [3H]GDP and [32P]NAD were from Du-Pont New England Nuclear. Man6P and pertussis toxin were obtained from Sigma and Funakoshi (Japan), respectively. 1204
Vol.
BIOCHEMICAL
168, No. 3, 1990
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
Results Effect of Man6P and IGF-II on Gi-2 in Reconstituted Vesicles: In order to clarify whether man6P binding brings about the coupling of the IGF-IIR with Gi-2, we tested the effect of man6P in vesicles reconstituted with the purified IGF-IIR and Gi-2. As described previously (25), the IGF-IIR and Gi-2 were each purified to homogeneity and inserted into phospholipid vesicles. When the vesicles were stimulated with 100 nM of IGF-II, the GDP dissociation rate of Gi-2 was increased approximately twofold (Fig. 1B). The stimulatory effect of a 5-min IGF-II incubation on GDP release from Gi-2 was dose-dependently observed and had an EC50 of approximately 6 nM for IGF-II in the IGFIIR/Gi-2 vesicles. The stimulator-y effect of IGF-II could not be observed in the vesicles reconstituted with Gi-2 alone (not shown). In contrast, man6P had no significant effect on GDP release from Gi-2 in concentrations up to 10 mM in the IGF-IIR/Gi-2 vesicles. In either experimental condition, i.e., pH range between 6.5 and 8.0 or Mg2+ concentrations between =O and 10 mM, we observed no effect of man6P on GDP release from Gi-2 in the IGF-IIR/Gi-2 vesicles (not shown). We have already reported that IGF-II stimulates GTP@ binding to Gi-2 in IGF-IIR/Gi-2 vesicles (25). However, since GDP release is thought to be a more critical step on which receptor stimulation directly acts
0
200 -
z
60
4 3
60
al s 40
69
L E i
-
% 0 " [IGF-II]
92.5 45 -1
2
20
0
[manGP]
ji_::: 0 O-
1
10
100
1000
WV
0.01
0.1
1
10
@Ml
Fig. 1: (A) IGF-II receptor preparation purified from rat placenta. The IGFIIR was purified to homogeneity by means of three steps of affinity chromatographies. The receptor was electrophoresed in a 7.5% SDS-gel and stained with silver (lane 2). Lane 1 indicates Laemmli sample buffer alone. (B) Effect of man6P and IGF-II on GDP release from Gi-2 in IGF-IIR/Gi-2 vesicles. GDP release from Gi-2 was measured in a 5 min incubation of the IGF-IIR/Gi-2 vesicles with increasing concentrations of mau6P (open circle) or IGF-II (closed circle). GDP dissociation represents a fraction after subtracting GDP-bound Gi-2 from the total Gi-2, and is indicated in the figure as a rate per 5 min. Values represent means of three experiments. S. E. was less than 5% of each value. 1205
BIOCHEMICAL
Vol. 168, No. 3, 1990
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
(27), this result provides compelling evidence that the IGF-IIR activates Gi-2 in response to IGF-II binding in a manner quite similar to conventional Gcoupled receptor. Furthermore, it indicates that man6P binding to the IGF-IIR did not give rise to the activation of Gi-2. Effect of Man6P or IGF-II on 40-kDa Gi-like Proteins in Native Membranes: To examine whether man6P stimulation of the IGF-IIR does not induce a change of Gi-like proteins in a native environment of cell membranes, we tested the effect of man6P in human erythroleukemia K562 cell membranes. Since the K562 cells exhibit the typical IGF-IIR, but no IGF-I receptor (2021), the effect of IGF-II is considered as mediated by either IGF-IIR or insulin receptor. When K562 cell membranes were incubated for 10 min with man6P in concentrations up to 10 mM, we observed no significant change of the pertussis toxin-induced ADP-ribosylation of the endogenous 40-kDa protein in the K562 cell membranes (Fig. 2A). In addition, when the membranes were A
C B
68
-
4s
-
30
-
0
0.1 I man6Pl (mfl)
I
IO
68
-
45
-
30
-
IGF-II IOnM
68
-
4s
-
30
-
0 0’
5’
IO’
15’
30’
60’
01
I
I man6PI CrnM)
Fig. 2: Effect of man6P on the pertussis toxin-catalyzed ADP-ribosylation activity of 40-kDa Gi-like proteins in K562 cell membranes. (A) Dose response relation for the effect of man6P. Membranes were incubated with increasing concentrations of man6P for 5 min and then ADP-ribosylated with pertussis toxin as described under “Experimental Procedures”. The right lane indicates the effect of 10 nM IGF-II in the same conditions as man6P incubation (B) Time course of man6P effect on the 40-kDa pertussis toxin substrate protein. Membranes were incubated with 10 mM man6P for indicated periods and ADP-ribosylated with pertussis toxin. (C) Effect of man6P on exogenously added Gi-2 in membranes from pertussis toxin-treated KS62 cells. The toxin-pretreated membranes supplemented with purified Gi-2 were incubated with increasing concentrations of man6P for 5 min, and then ADP-ribosylated. The right lane shows the effect of 10 nM IGF-II on Gi-2 observed in the same conditions. 1206
IO
IGF-II’ IOnM
Vol. 168, No. 3, 1990
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
incubated with 10 mM man6P, no significant change in the pertussis toxin substrate activity of the endogenous 40-kDa protein was observed over the course of a 60 min incubation (Fig. 2B). The inability of man6P to act on the 40-kDa pertussis toxin substrate protein was also the case with K562 cells membranes supplemented with purified Gi-2. As shown in Fig. 2C, man6P elicited no change in the pertussis toxin substrate activity from Gi-2 exogenously added to the pertussis toxin-treated KS62 cell membranes. Although 10 mM man6P sometimes affected the ADP-ribosylation of Gi-2, this was not reproducible. Even when Gi-1 or Go was incubated with man6P in the presence of toxin-pretreated K562 cell membranes, we could observe no change in the toxin substrate activity of either G proteins (not shown). In contrast, 10 nM IGF-II reduced the toxin sensitivity of the 40-kDa protein by approximately l/S in either experiments from K562 cell membranes or from K562 cell membranes supplemented with purified Gi-2. The effect of IGF-II could not be reproduced by IGF-I or insulin which had no effect in these cell membranes (data not shown). Essentially similar results were obtained by using mouse BALB/c-3T3 cell membranes where the toxin sensitivity of endogenous and exogenous 40-kDa Gi-like proteins was not affected by man6P, but reduced by IGF-II (not shown). Discussion We documented herein that the purified rat IGF-IIR could respond dissimilarly to the two distinct high affinity ligands, IGF-II and man6P. It is well known that these two structurally unrelated ligands bind to the different site on the IGF-IIR with high affinity, and that they evokes different cellular events; IGF-II binding induces metabolic/mitogenic stimulations and man6P binding activates sorting/endocytosis processes. When the IGF-IIR was stimulated with 100 nM of IGF-II, the GDP release rate of Gi-2 was augmented approximately twofold in phospholipid vesicles reconstituted with the IGF-IIR and Gi-2. Furthermore, the action of IGF-II was observed in a dose-dependent manner for IGF-II with an ECso of 6 nM and only in the presence of the IGFIIR. We have described that IGF-II stimulates GTPyS binding to Gi-2 in the IGF-IIR/Gi-2 vesicles (25). The IGF-II potency and the intensity of the IGF-II action in the stimulation of the GDP release rate of Gi-2 correlated well with the IGF-II potency and intensity in the stimulation of the GTP+ySbinding rate of the G protein. Moreover, the extent of IGF-IIR-induced increase in guanine nucleotide metabolism on Gi-2 was comparable to that reported in conventional Gi-coupled receptors, muscarinic receptors (28) and az-adrenergic receptors (29). Therefore, the present study indicates that the IGF-IIR activates Gi-2 in response to IGF-II binding as significantly and as similarly as physiological Gicoupled receptors. 1207
Vol.
168, No. 3, 1990
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
In marked contrast, man6P stimulated no GDP release from Gi-2 in concentrations up to 10 mM in the IGF-IIR/Gi-2 vesicles. MacDonald et al. (6) have reported that the purified rat IGF-IIR binds man6P with a high affinity. Hence, this result indicates that the purified IGF-IIR does not activate Gi-2 upon man6P binding. In addition, it has been demonstrated that man6P and IGF-II each bind to the separate sites of the rat IGF-IIR (30). It is therefore suggested that the ligand-dependent stimulations of the different extracellular regions of the IGF-IlR lead to the qualitatively different intracellular reactions of the receptor. Such dual functions of the IGF-IIR have been preliminarily reported by Rogers and Hammerman (31). They described that man6P does not stimulate polyphosphoinositide breakdown in canine renal cells, despite the fact that IGF-II does so. However, it remained unclear whether IGF-II and man6P acted on the same receptor protein in their study. The present study indicates that the single IGF-IIR has a multifunctional property, not only in the aspect of ligand binding but also the receptor signaling. We next examined whether the multifunctional response of the IGF-IIR is observed even in an environment of cell membranes which more approximates a physiological condition. The result also indicated that man6P had no effect on either the endogenous 40-kDa protein or Gi-2 in concentrations up to 10 mM in human K562 cell membranes. It has been reported that in both these cell membranes, IGF-II potently attenuates the pertussis toxin-catalyzed ADP-ribosylation of the 40-kDa Gi-like proteins, mainly through the IGF-IIR (25). Furthermore, such a modification of Gi-like proteins is likely to represent a subunit dissociation of them (25), an essential aspect of G protein activation (32). Therefore, we assumed that a native environment of cell membranes maintains the multifunctional property of the IGF-IIR observed in reconstituted vesicles; the IGF-II binding leads to activation of Gi-2, while the man6P binding does not do so. The other possible explanation for the lack of effects of man6P in these cell membranes is that man6P does not bind to the human IGF-IIR. For example, Canfield and Komfeld (33) have reported that the chicken CI-MPR, a possible homologue of the chicken IGF-IIR, lacks the high affinity binding site for IGF-II, while the receptor retains that for man6P. However, the human IGF-IIR exhibits the high affinity binding for man6P (34). Furthermore, K562 cell membranes demonstrate the high affinity binding of man6P as assessed with manBPinhibitable P-glucuronidase binding, which reflects specific man6P binding to the IGF-IIR (unpublished observation: Murayama, Y. and Nishimoto, I.). It is therefore likely that in the native cell membranes man6P binds to the IGF-IIR, but does not produce the possible activation of Gi-like proteins. Given the assumption that the IGF-IIR is identical to the CI-MPR, the binding of man6P to the IGF-IIR may evoke the sorting or endocytosis of man6Pcontaining proteins. It has been reported recently that the sorting and 1208
Vol.
168, No. 3, 1990
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
endocytosis functions of the CI-MPR/IGF-IIR may involve a G proteinmediated process (35). On the other hand, the present study indicates that man6P-induced stimulation of the IGF-IIR does not produce the activation of Gi-like proteins in phospholipid vesicles or in native cell membranes. This suggests that the signal for sorting may not accompany the activation of Gilike proteins. Furthermore, we describe here that the single IGF-IIR responds differently to man6P and IGF-II that bind to separate binding regions of the receptor. Studies need to be done to elucidate the molecular mechanism underlying the unique dual function of the IGF-IIR in response to the two distinct ligands, IGF-II and man6P. Acknowledgments We are especially indebted to Dr. Toshiaki Katada for purified G proteins and helpful discussion. We also thank Dr. S. Peter Nissley for his gift of rat IGF-II; Drs. William S. Sly of St. Louis University, Michio Ui of University of Tokyo, Taroh Iiri of Tokyo Institute of Technology, Shintaro Iwashita of Mitsubishi-Kasei Institute of Life Science, Yumi and Yoshiomi Tamai of the Society For Educational Aid To Traffic Orphans for indispensable support during the course of this study. Technical assistance of Tomomi Konishi is gratefully acknowledged. References 1. Rechler, M. M. and Nissley, S. P. (1985) Annu. Rev. Physiol. 47,425-442 2. Kasuga, M., Van Obberghen, E., Nissley, S. P. and Rechler, M. M. (1981) J. Biol. Chem. 256,5305-5308 3. Massague, J. and Czech, M. P. (1982) J. Biol. Chem. 257,5038-5045 4. Morgan, D. 0. and Roth, R. A. (1987) Proc. Natl. Acad. Sci. U.S.A. 84, 41-45 5. Morgan, D. O., Edman, J. E., Standring, D. N., Fried, V. A., Smith, M. C., Roth, R. A. and Rutter, W. J. (1987) Nature 329, 301-307 6. MacDonald, R. G., Pfeffer, S. R., Coussens, L., Tepper, M. A., Brocklebank, C. M., Mole, J. E., Anderson, J. K., Chen, E., Czech, M. P. and Ullrich, A. (1988) Science 239, 1134-1137 7. Lobel, P., Dahms, N. M. and Kornfeld, S. (1988) J. Biol. Chem. 263,25632570 8. Oshirna, A., Nolan, C. M., Kyle, J. W., Grubb, J. H. and Sly, W. S. (1988) J. Biol. Chem. 263,2553-2562 9. von Figura, K. and Hasilik, A. (1986) AMU. Rev. Biochem. 55, 167-193 10. Roth, R. A., Stover, C., Hari, J., Morgan, D. O., Smith, M. C., Sara, V. and Fried, V. A. (1987) Biochem. Biophys. Res. Commun. 149,600-606 11. Kiess, W., Blickenstaff, G. D., Sklar, M. M., Thomas, C. L., Nissley, S. P. and Sahagian, G. G. (1988) J. Biol. Chem. 263,9339-9344 1209
Vol. 166, No. 3, 1990
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
12. Brat&e, T., Causin, C., Waheed, A., Junghans, U., Hasilik, A., Maly, P., Humbel, R. E. and von Figura, K. (1988) Biochem. Biophys. Res. Commun. 150,1287-1293 13. Tong, P., Tollefsen, S. E. and Kornfeld, S. (1989) J. Biol. Chem. 263, 25852588 14. Nishimoto, I. and Kojima, I. (1989) N. I. P. S. 4, 94-97 15. Yu, K-T. and Czech, M. P. (1984) J. Biol. Chem. 259,3090-3095 16. Conover, C. A., Rosenfeld, R. G. and Hintz, R. L. (1987) J. Cell. Physiol. 133, 560-566 17. King, G. L., Kahn, C. R., Rechler, M. M. and Nissley, A. P. (1980) J. Clin. Invest. 66, 130-140 18. Mottola, C. and Czech, M. P. (1984) J. Biol. Chem. 259, 12705-12713 19. Hari, J., Pierce, S. B., Morgan, D. O., Sara, V., Smith, M. C. and Roth, R. A. (1987) EMBO J. 6,3367-3371 20. Tally, M., Li, C. H. and Hall, K. (1987) Biochem. Biophys. Res. Commun. 148, 811-816 21. Blanchard, M. M., Barenton, B., Sullivan, A., Foster, B., Guyda, H. J. and Posner, B. I. (1988) Mol. Cell. Endocrinol. 56, 235-244 22. Simizu, M., Webster, C., Morgan, D. O., Blau, H. M. and Roth, R. A. (1986) Am. J. Physiol. 251, E611-E615 23. Rogers, S. A. and Hammerman, M. R. (1988) Proc. Natl. Acad. Sci. U. S. A. 85,4037-4041 24. Nishimoto, I., Hata, Y., Ogata, E. and Kojima, I. (1987) J. Biol. Chem. 262, 12120-12126 25. Nishimoto, I., Murayama, Y ., Katada, T., Ui, M. and Ogata, E. (1989) J. Biol. Chem. 264,14029-14038 26. Gihnan, A. G. (1987) Amru. Rev. Biochem. 56,615-650 27. Ferguson, K. M., Higashijima, T., Smigel, M. D. and Gihnan, A. G. (1986) J. Biol. Chem. 261,7393-7399 28. Cerione, R. A., Regan, J. W., Nakata, H., Codina, J., Benovic, J. L., Gierschik, P., Somers, R. L., Spiegel, A. M., Birnbaumer, L., Lefkowitz, R. J., and Caron, M. G. (1986) J. Biol. Chem. 261, 3901-3909 29. Kurose, H., Katada, T., Haga, T., Haga, K., Ichiyama, A., and Ui, M. (1986) J. Biol. Chem. 261,6423-6428 30. Roth, R. A. (1988) Science 239,1269-1271 31. Rogers, A. A., and Hammerman, M. R. (1989) J. Biol. Chem. 264,42734276 32. Neer, E. J. and Clapham, D. E, (1988) Nature 333, 129-133 33. Canfield, W. M. and Kornfeld, S. (1989) J. Biol. Chem. 264, 7100-7103 34. Nolan, C. M., Kyle, J. W., Watanabe, H. and Sly, W. S. (1990) Cell Regulation 1, 197-213 35. Goda, Y. and Pfeffer, S. R. (1988) Cell 55, 309-320 1210
