Direct Demonstration of Insulin-induced GLUT4 Translocation to the ...

Vol. 268, No. 19, Issue of July 5, pp. 14523-14526, 1993 Printed in U.S.A.

THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1993 by The American Society for Biochemistry and Molecular Biology,, Inc

Direct Demonstration of Insulin-induced GLUT4 Translocation to the Surface of Intact Cells by Insertion of a c-myc Epitope into an Exofacial GLUT4 Domain* (Received for publication, January 19, 1993, and in revised form, March 29, 1993)

Fumihiko Kanai, Yasuhiko NishiokaS, Hideki Hayashi, Seika Kamohara, Mikio Todaka, and Yousuke EbinaQ From the DeDartment of Enzyme Genetics, Institute for Enzyme Research, th.e University of Tokushima, 3-18-15Kuramoto-cho, TokGhima 770, Japan

Stimulation of glucose transport is the main physiological effect of insulin in target tissues. This effect is linked to translocation of the GLUT4 glucose transporter from an intracellular pool to the cellsurface. To elucidate the molecular mechanisms involved in this effect, we developed a simple direct sensitive method to detect GLUT4 immunologically on the cell surface. cDNA containing GLUT4 inserted by a c-myc epitope in the first ectodomain (GLUT4myc) was constructed without disrupting the functions of GLUT4 and was expressed in 3T3-Ll and Chinese hamster ovary fibroblast cells. In response to insulin, the GLUT4myc expressed in 3T3-Ll adipocytes was translocated to the cell surface from the intracellular pool, as shown by assays of exofacial antibody binding against the myc epitope and of the uptake of 2-deoxyglucose. Insulin, guanosine 5’-0-(3-thiotriphosphate),guanylyl imidodiphosphate, NaF, and phorbol 12-myristate 13-acetate also induced the translocation of GLUT4myc in Chinese hamster ovary cells coexpressing the human insulin receptor.

The majority of diabetes mellitus, which is non-insulindependent diabetes mellitus, is caused mainly by an insensitivity of target cells to insulin, and its incidence appears to be increasing. An understanding of the insensitivity in these patients requires an understanding of the molecular mechanisms of insulin-stimulated glucose transport. The transport in target tissues is due mostly to translocation of GLUT4 from an intracellular siteto theplasma membrane (1-5). The molecular mechanism by which insulin induces this translocation of GLUT4 to theplasma membrane is not well understood. One major factor that has limited investigations on the molecular mechanism of insulin-stimulated GLUT4 translocation has been the absence of a direct sensitive method for

* This work was supported by research grants from the Ministry of Education, Science, and Culture of Japan, by a grant for diabetes research from the Ministry of Health and Welfare, by the Foundation for Health Science (Japan), and by a research grant for diabetes research from Otsuka Pharmaceutical Co., Ltd. (to Y. E.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18U.S.C. Section 1734 solelyto indicate this fact. $ Present address: 3rd Dept. of Internal Medicine, School of Medicine, the University of Tokushima, 3-18-15 Kuramoto-cho, Tokushima 770, Japan. 5 To whom correspondence should be addressed. Tel.:0886-313111 (ext. 2540); Fax: 0886-33-1845.

detecting GLUT4 translocation. Available methods include Western blot analysis of subcellular membrane fractions, binding assay of cytochalasin B, and analyses by immunofluorescence or immunoelectron microscopy.However, these techniques do not enable sufficient detection of GLUT4 translocation specifically and quantitatively. An ideal method would be direct immunological detection of GLUT4 on the cell surface with an anti-ectodomain antibody specific for GLUT4; heretofore, no such antibody against GLUT4 has yet been obtained. cDNAs encoding five glucose transporters, named GLUT1 to GLUT5, have been cloned and characterized (6-9). These glucose transporter isoforms share highly conserved structures with 12 transmembrane domains and one large N glycosylation chain (first externalloop) between the first and second transmembrane regions. Here we constructed on the cDNA the chimeric GLUT4 (GLUT4myc) by inserting 14 amino acids of the c-myc epitope (10) in the first ectodomain without destroying the functions of GLUT4. GLUT4myc was expressed in the cells, and the insulin-stimulated translocation of GLUT4myc was detected directly on the cell surface by the binding of anti-myc antibody (10). EXPERIMENTALPROCEDURES

Construction of c-myc Epitope-inserted GLUT4myc cDNA Expression Vectors-GLUT4 cDNA from the plasmid pIRGT ( 5 ) was subcloned into the pSRa vector (SRaGLUT4) ( l l ) , and a unique SmaI site was introduced into the nucleotide sequence coding for GLUT4 Proffi-Gly6’by changing CCTGGG to CCCGGG by polymerase chain reaction mutagenesis. SRaGLUT4myc was constructed by inserting 14 amino acids of the human c-myc epitope (GCAGAGGAGCAAAAGCTTATTTCTGAAGACGACTTGCTTGCTTAAG)(lO) at the SmaI site. This resulted in a fusion gene encoding the protein sequence P(AEEQKL1SEEDLLK)G. The GLUT4myc fragment was subcloned into the retroviral mammalian expression vector pDOL (12). Generation and Differentiation of 3T3-Ll-GLUT4myc Cells and Subcellular Fractionation and Zmmunoblot Analysis”ST3-Ll cells that stably express GLUT4myc (3T3-Ll-GLUT4myc) were established by infection of recombinant retrovirus as described (13). The viral supernatant was prepared by introduction of pDOL-GLUT4myc DNA into $2 cells by the calcium phosphate precipitation method. 3T3-Ll fibroblasts were infected with the virus and were selected by G418. Several clones were assayed for the stable expression of GLUT4myc. Induction of differentiation into adipocytes was performed ar described (13). Subcellular fractions from 3T3-Ll and3T3L1-GLUT4myc adipocytes in the absence and presence of 100 n M insulin were analyzed by 10% SDS-polyacrylamide gel electrophoresis, transferred to nitrocellulose, and probed with anti-insulin-regulatable glucose transporter antibody (mAb 1F8) (Genzyme) (4, 5) or anti-myc antibody (mAb 9E10) (Oncogene Science, Inc.) (10).

14523

14524

Exofacial Detection of Insulin-stimulated GLUT4 Translocation

Transfection of CHO' Cells and Immunoprecipitation of Lubeled GLUT4 or GLUT4myc"CHO cells were transfected with SRaGLUT4myc or SRaGLUT4 andpSV2neo by calcium phosphate precipitation. Cell clones resistant to G418 were assayed for the expression of GLUT4myc or GLUT4. CHO-GLUT4myc33 cells were transfected with SRaIR (wild type) (14, 15) or SRdRMet(kinasedeficient insulin receptor) (16) and pSV2hph. Clones expressing wildtype or mutant receptors obtained by hygromycin B selection were named CHO-GLUT4mycIR and CHO-GLUT4mycIRMet, respectively. Cells were metabolically labeled for 4 h using 0.15 mCi/ml Tran36S-label (ICN) and lysed with lysis buffer (50 mM Hepes, pH 7.5, 150 mM NaCI, 1%Triton X-100, 20 p~ phenylmethylsulfonyl fluoride, 1 mg/ml bacitracin, 5 mM EDTA,5 mM EGTA, 20 mM sodium pyrophosphate, 1 mM orthovanadate, and 20 mM NaF). Cell lysates were immunoprecipitated with mAb 1F8 or mAb 9E10, resolved by 10% SDS-polyacrylamide gel electrophoresis, andthen fluorographed. Cell-surface Anti-myc Antibody Binding Assay-3TB-Ll and 3T3L1-GLUT4myc adipocytes in 24-well plates were incubated in 1 ml of KRH buffer (136 mM NaCI, 4.7 mM KCI, 1.25 mM MgSO,, 1.25 mM Hepes, and 2 mg/ml bovine serum albumin) for 30 min C ata37 mM C 1 z Tand then in 200 pl of KRH buffer with 100 nM insulin for 30 min at 37 "C. After fixation with 2% paraformaldehyde, cells were washed with PBS and incubated with 5% skim milk/PBS for 1 h. They were then incubated with 200 pl of mAb 9E10 (1:lOO dilution) for 2 h, washed with 5% skim milk/PBS, and incubated for 1 h with 200 pl of 'Z51-labeiedgoat anti-mouse IgG (specific activity, 2-15 pCi/ pg; 1:200 dilution; ICN) at room temperature. The wells were then washed with 5% skim milk/PBS twice, with PBS twice, and with 0.05% Tween 2O/PBS three times. Bound '251-labeledgoat anti-mouse IgG was solubilized with 1%SDS, and radioactivity was determined in a y-counter. CHO cell lines were grown in 24-well plates, and anti-myc antibody binding was determined. Subconfluent cells were incubated in KRH buffer for 30 min at 37 "C and then in KRH buffer with various concentrations of insulin for 30 min at 37 "C.They were then washed with ice-cold KRH buffer and incubated with mAb 9E10 for 2 h on ice. The wells were washed with ice-cold KRH buffer and incubated for 1 h with '251-labeled goat anti-mouse IgG on ice. After washing with ice-cold KRH buffer four times, bound '251-labeledgoat antimouse IgG was solubilized, and itsradioactivity was determined. Anti-myc Antibody Binding Assay of Streptolysin 0-permeabilized CHO Cells-CHO-GLUT4mycIR cells in 24-well plates were washed with GK buffer (20 mM Hepes, 125 mM potassium glutamate, 5 mM EGTA, 5 mM MgC12,15 mM KCI, 5 mM NaC1, and 1 mg/ml bovine serum albumin) and incubated in GK buffer containing 0.05 unit/ml streptolysin 0 (Nissui Pharmaceutical Co., Ltd.) for 10 min a t 37 "C (17-19). They were then incubated with 100 nM insulin,10 p M GTPyS, 10 p M GMP-PNP, or 10 p M GDPbS in GK buffer for 20 min at 37 "C (19,20), washed with ice-cold GK buffer, and subjected to anti-myc binding assay. 2-Deonyglucose Uptake Measurements-2-Deoxy-~-glucoseuptake measurements were performed as previously described (15, 16). Cells in 24-well plates were treated with 100 nM insulin for 30 min and incubated with 0.1 mM 2-deoxy-~-[1,2-~H]glucose (DuPont-New England Nuclear) for 10 min at 37 "C. Cells were washed and solubilized, and theradioactivity was measured.

A 3T3-Ll adipocyte

adipocyte Antibody

Insulin Fraction

'

%

'

Anti-GLUT4 Anti-rnyc

-

+ PM

-

+

LDM

0

x

3T3-Ll-GLUT4 rnyc

-

+

PM

-

+

LDM

-

+

PM

-

+

LDM

-insulin

5000

n C

a

o

3T3-Ll adipocyle

3T3-Ll-GLUT4myc adipocyte

FIG.1. Translocation of expressed GLUT4myc in 3T3-Ll adipocytes.A, immunoblot of subcellular fractions from 3T3-LI and 3T3-Ll-GLUT4myc adipocytes with anti-GLUT4 monoclonal antibody (mAb 1F8) or anti-myc monoclonal antibody (mAb 9E10). PM, plasma membranes; LDM, low density microsomes. B, anti-myc antibody binding to 3T3-Ll and 3T3-Ll-GLUT4myc adipocytes. Values represent means f S.E. of four determinations. C, 2-deoxyglucose uptake by 3T3-Ll and 3T3-Ll-GLUT4myc adipocytes. 2-Deoxyglucose uptake assay was performed as described (15, 16). Values represent means & S.E. of six determinations. The asterisk shows significant difference (p < 0.01) from insulin-stimulated 2-deoxyglucose uptake by 3T3-Ll adipocytes.

ditions as for GLUT4myc translocation observed by immunoblot analysis, insulin treatment increased the number of binding sites of anti-myc antibody on the cell surface of 3T3L1-GLUT4myc adipocytes (Fig. 1B) and also increased 2deoxyglucose uptake (Fig. IC). Glucose uptake by 3T3-LlGLUT4myc adipocytes exceededthat of 3T3-Ll adipocytes, thus myc epitope insertion in the firstexternal loop of GLUT4 did not destroy the functions of translocation of GLUT4 and glucose uptake in response to insulin. We reported that insulin stimulates glucose uptake by CHO cells stably overexpressing human insulin receptors (15).3T3L1 adipocytes stably expressingGLUT4myc are the more relevant model of insulin-stimulated GLUT4 translocation. RESULTSANDDISCUSSION However, CHO cells more easily express other cDNAs than In an attempt to detect GLUT4 translocated to the cell 3T3-Ll cells and were used to analyze the additional molecsurface directly, we inserted synthetic oligonucleotides of the ular mechanisms of GLUT4 translocation. In this work, we c-myc epitope (14 amino acids) into the first external loop of examined the translocation of GLUT4myc in CHOcells. GLUT4 cDNA. Expression and insulin-stimulated translo- These cells were first transfected with the expression vector cation of the myc epitope-tagged GLUT4 (GLUT4myc) in of GLUTlimyc, and several clones were selected by their 3T3-Ll adipocytes using an anti-c-myc antibody (mAb 9E10) surface binding of anti-myc antibody in the absence of insulin. are shown in Fig. 1A. Immunoblot analysis of subcellular We also established CHO clones coexpressing normalor tymembrane fractions of 3T3-Ll adipocytes stably expressing rosine kinase-deficient human insulin receptors (16)with GLUT4myc showed translocation of the protein from low GLUT4myc. The expression of GLUT4myc in these CHO density microsomes to theplasma membrane, like endogenous clones is shown in Fig. 2A. There are reports that GLUT4 GLUT4 in response to insulin. Under almost the same con- expressed exogenouslyafter transfection with GLUT4 cDNA ' The abbreviations used are: CHO, Chinese hamster ovary; remains mainly in the intracellular pool of CHO and various GTPrS, guanosine 5'-0-(3-thiotriphosphate);GDPPS, guanosine 5'- cell lines and that insulin-stimulated translocation of the 0-(2-thiodiphosphate); GMP-PNP, guanylyl imidodiphosphate; PBS, expressed GLUT4 cannot be detected by measurement of glucose uptake, by immunoblot analysis of subcellular fracphosphate-buffered saline; mAb, monoclonal antibody.

Exofacial Detection

of Insulin-stimulated GLUT4 Translocation

14525

A Antibody

Anti-GLUT4

Anti-myc

8

5000

PMA

-E, W

4000

-

3000

-

B

s

2000-

Insulin

CHO-GLUT4myclR

rn

/

v

NaF

CHO-GLUT4myc A

6 C

a

4000 U

0

4

CHO-GLUT4myclRMet 0

0.1

1

10

100

Insulin concentration (nM)

p

I

2000

n" GTPYS

GMP-PNP

GDPpS

Insulin

FIG. 3. A, translocation of GLUT4myc in CHO cells by phorbol 12-myristate13-acetate and NaF. CHO-GLUT4mycIR cells were incubated with 100 nM phorbol 12-myristate 13-acetate ( P M A ) ,with 5 mM NaF and 10 pM AlC13 (26-28), or with 100 nM insulin for 30 min a t 37 "C and then subjected to anti-myc antibody binding assay. Values represent means -+ S.E. of six determinations. Asterisks show significant difference ( p < 0.01) from non-insulin-stimulated values. B, effects of insulin, GTP-yS, GMP-PNP, and GDPBS on translocation of GLUT4myc in streptolysin 0-permeabilized CHO cells. Antimyc binding assay was performed as described under "Experimental Time (min) Procedures." Nonspecific binding (permeabilized CHO cells, 2862 -+ FIG.2. Translocation of expressed GLUTImyc in CHO 124 cpm) was subtracted from total binding. Values represent means cells. A, expression of GLUT4myc and GLUT4 inCHO cells. Lysates f S.E. of four determinations. Asterisks show significant difference of 35S-labeledCHO (lanes I and 71, CHO-GLUT4 (lanes 2 and 8), ( p < 0.01) from non-insulin-stimulated values. CHO-GLUT4myc33 (lanes 3 and 9 ) , CHO-GLUT4myc29 (lanes 4 and IO),CHO-GLUT4mycIR (wild type) (lanes5 and I 1 ), and CHO- ity, with a half-maximal response at ~ 0 . nM 1 and maximum GLUT4mycIR'" (lanes 6 and 12) cells were immunoprecipitated stimulation of -3.6-fold the control level. In contrast, cells with mAb 1F8 (anti-GLUT4) or mAb 9E10 (anti-myc). The arrowhead indicates GLUT4myc or GLUT4. B, anti-myc antibodybinding expressing the kinase-deficient insulin receptor mutant to CHO cells. Shown is the dose response of insulin-stimulated (-105/cell) showed a lesser response to insulin than did the translocation of GLUT4myc in CHO cell lines. Values represent parental CHO-GLUT4myc cells, witha half-maximal remeans f S.E. of four determinations. Nonspecific binding (CHO sponse at -10 nM. Thus, receptor kinase activity plays a key cells, 782 122 cpm) was subtracted from total binding for each cell role in theprocess of GLUT4 translocation. The time course line. C, time course of insulin-dependent translocation of GLUT4myc. of insulin-stimulated translocation is illustrated in Fig. 2C. CHO-GLUT4mycIR cells were treated for 0, 2, 5, 10, 30, and 60 min with 100 nM insulin and thensubjected to anti-myc antibodybinding The translocation of GLUT4myc in CHO cells began within 1 min and reached a maximum within 5 min. These values assay. Results are representative of three experiments.

are infair agreement withdata on the translocation of endogenous GLUT4 in adipocytes (23). tions, or by immunofluorescence microscopy (21, 22). We It would appear that in CHO-GLUT4mycIR cells, only a also found that exogenously overexpressed GLUT4and GLUT4myc were mainly sequestered intracellularly and that small percentage of intracellular GLUT4myc is translocated insulin-stimulated translocation of GLUT4 and GLUT4myc to the cell surface in response to insulin and that themethod to the cell surface could not be detected using the methods ofmyc antibody binding is more sensitive than heretofore described above (data notshown). However, by surface bind- available methods. As shown in Fig. 2A, the CHO-GLUT4myc29 clone (lanes ing assay using anti-mycantibody,some GLUT4myc was detected onthe plasma membraneof CHO cells in theabsence 4 and 10) expressed more proteins than did the CHOof insulin, andthe insulin-stimulated translocation of GLUT4myc33 clone (lanes 3 and 9). The exofacial binding sites of anti-myc antibodyand thebasal glucose uptake of the GLUT4myc was clearly observed (Fig. 2, B and C). The kinetics of translocation and theinsulin sensitivity of CHO-GLUT4myc29 clone in theabsence of insulin exceeded shown). Thus, these cell lines were determined. The insulin dependence of those of CHO-GLUT4myc33 (datanot anti-myc antibody-binding sites is illustrated in Fig. 2B. In- GLUT4myc functions in glucose uptake in CHO cells as well cubation of insulin with CHO-GLUT4myccells led to a dose- as in 3T3-Ll adipocytes. However, 3T3-Ll-GLUT4myc adipocytes seem to possess a markedly insulin-sensitive system dependent increase in antibody-binding sites, with a halffor translocation of GLUT4myc or a system for enhancing maximal response a t =5 nM and maximal stimulation of -2.5fold the control level. Insulin-like growth factor I induced a the intrinsic catalytic activityof cell-surface GLUT4. Using this surface binding assay, we detected the translosimilar magnitude, but platelet-derived growth factor and epidermal growth factor had no effect on translocation of cation of GLUT4myc by GTP-yS, GMP-PNP, NaF, and phorGLUT4myc in CHOcells (data notshown). The introduction bo1 12-myristate 13-acetate inCHO-GLUT4mycIR cells (Fig. of =lo6 wild-type human insulin receptors/cell inCHO3, A and B), as was observed in adipocytes (19, 20, 24, 25). GLUT4myc cells resulted in an ~50-fold increase in sensitiv- These data suggest the involvement of GTP-bindingpro-

14526

Exofacial Detection

of Insulin-stimulated GLUT4 Translocation

tein(s) and protein kinase C in GLUT4 translocation. CHO-GLUT4mycIR cells therefore seem to possess the basic machinery required for translocation of GLUT4myc, but an additional factor(s)is required for the translocation of GLUT4myc in 3T3-Ll-GLUT4myc adipocytes. CHOGLUT4mycIR cells should prove useful for the study of the molecular mechanisms of GLUT4 translocation as induced by insulin. Acknowledgments-We thank M. Mueckler for GLUT4 cDNA; R. C. Mulligan for pDOL; Y. Takebe for pSRa; R. A. Roth for the antiinsulin receptor antibody; P. F. Pilch for mAb 1F8 K. Shimotohno for $2 cells; Y. Takai, K. Kaibuchi, and Y. Oka for helpful discussions; and M. Ohara and E. Ichihara for reading the manuscript. REFERENCES 1. Simpson, I. A., and Cushman, S. W. (1986) Annu. Reu. Biochem. 55,10591089 2. Cushman, S. W., and Wardzala, L. J. (1980) J. Biol.Chem. 2 5 5 , 47584162 3. Suzuki, K., and Kono, T. (1980) Proc. Natl. Acad. Sci. U. S. A. 7 7 , 25422545 4. James, D.E., Brown, R., Navarro, J., and Pilch, P. F. (1988) Nature 3 3 3 , 183-185 5. James, D. E., Strube, M., andMueckler, M. (1989) Nature 338,83-87 6. Mueckler, M. (1990) Diubetes 39.6-11 7. Carruthers, A. (1990) Physiol. Reu. 7 0 , 1135-1176 8. Pessin, J. E., and Bell, G. I. (1992) Annu. Reu. Physiol. 5 4 , 911-930 9. Czech, M. P., Clancy, B. M., Pessino, A,, Woon, C.-W., and Harrison, S. A. (1992) Trends Biochem. Sci. 17,197-201

10. Evan, G. I., Lewis, G. K., Ramsay, G., and Bishop, J. M. (1985) Mol. Cell. Biol. 5,3610-3616 11. Takehe, Y., Seiki, M., Fujisawa, J., Hoy, P., Yokota, K., Arai, K., Yoshida, M., and Arai, N. (1988) Mol. Cell. Biol. 8, 466-472 12. Korman, A. J., Frantz, J. D., Strominger, J. L., and Mulligan, R. C. (1987) Proc. Natl. Acad. Sci. U. S. A. 8 4 , 2150-2154 13. Gould, G. W., Derechin, V., James, D. E., Tordjman, K., Ahern, S., Gibbs, E. M., Lienhard, G. E., and Mueckler, M. (1989) J. Biol. Chem. 2 6 4 , 2180-2184 14. Ebina, Y., Ellis, L., Jarnagin, K., Edery, M., Graf, L., Clauser, E., Ou, J.H., Masiartz, F., Kan, Y. W., Goldfine, I. D., Roth, R. A,, and Rutter, W. J. (1985) Cell 40,747-758 15. Ebina, Y., Edery, M., Ellis, L., Standring, D., Beaudoin, J., Roth, R. A,, and Rutter, W. J. (1985) Proc. Natl. Acad. Sci. U. S. A. 82,8014-8018 16. Ebina, Y., Araki, E., Taira, M., Shimada, F., Mori, M., Craik, C. S., Siddle, K., Pierce, S. B., Roth, R. A., and Rutter, W.J. (1987) Proc. Natl. Acad. Sci. U. S. A. 84,704-708 17. Ahnert-Hilger, G., Mach, W., Fohr, K. J., and Grantzl, M. (1989) Methods Cell Biol. 3 1 , 63-90 18. Miller, S. G., and Moore, H.-P. H. (1991) J . Cell Biol. 1 1 2 , 39-54 19. Robinson, L. J., Pang, S., Harris, D. S., Heuser, J., and James,D. E. (1992) J. Cell Biol. 1 1 7 , 1181-1196 20. Baldini, G., Hohman, R., Charron, M. J., and Lodish, H. F. (1991) J. Biol. Chem. 266,4037-4040 21. Haney, P. M., Slot, J. W., Piper, R. C., James, D. E., and Mueckler, M. (1991) J . Cell Biol. 114,689-699 22. Shibasaki Y., Asano, T., Lin, J.-L., Tsukuda, K., Katagiri, H., Ishihara, H., Yaiaki, Y., and Oka, Y.(1992) Biochem. J. 281,829-834 23. Karnieli, E., Zarnowski, M. J., Hissin, P. J., Simpson, I. A,, Salans, L. B., and Cushman, S. W. (1981) J. Biol. Chem. 256,4772-4777 24. Saltis, J., Habberfield, A. D., Egan, J. J., Londos, C., Simpson, I. A., and Cushman, S. W. (1991) J. Biol. Chem. 266,261-267 25. Vogt, B., Mushack, J., Seffer, E., and Haring, H. U. (1991) Biochem. J. 275,597-600 26. Gilman, A. G. (1987) Annu. Rev. Biochem. 56,615-649 27. Bokoch, G. M., and Gilman, A. G. (1984) Cell 39,301-308 28. Blackmore, P. F., Bocckino, S. B., Waynick, L. E., and Exton, J. H. (1985) J. Biol. Chem. 2 6 0 , 14477-14483