α-Subunit of Farnesyltransferase Is Phosphorylatedin Vivo:Effect of Protein Phosphatase-1 on...

JOBNAME: BBRC 222#2 PAGE: 1 SESS: 11 OUTPUT: Wed May 29 10:25:09 1996/xypage/worksmart/tsp000/71401a/21

a-Subunit of Farnesyltransferase Is Phosphorylated in Vivo: Effect ofProtein Phosphatase-1 on Enzymatic Activity

Amit Kumar, Maureen H. Beresini,* Punita Dhawan, and Kamal D. Mehta1

Departments of Biochemistry & Molecular Biology & Medicine, University of Arkansas for Medical Sciences, Collegeof Medicine, 4301 West Markham, Little Rock, Arkansas 72205; the *Department of BioAnalytical Technology,

Genentech, Inc., South San Francisco, California 94080

Received April 3, 1996

Farnesyltransferase is a heterodimer consisting of a 49 kDaa-subunit and a 46 kDab-subunit. In this report,we demonstrate that the endogenous heterodimeric farnesyltransferase protein is phosphorylated at thea-subunitin vivo and phosphorylation plays a role in the regulation of farnesyltransferase activity.In vivo 32P-labeling ofPC-12 cells followed by immunoprecipitation with specific anti rata-subunit IgG showed a labeleda-subunitprotein band at an expected molecular mass of 49 kDa. Treatment of PC-12 cells with protein phosphataseinhibitor, Calyculin A, resulted in a decrease in FTase activity, and phophoserine/phosphothreonine-specificprotein phosphatase-1 treatment of PC-12 and GM37 cell extracts resulted in 100% and 375% increase infarnesyltransferase activity, respectively, compared to untreated extracts.© 1996 Academic Press, Inc.

Prenylation ranks among the most common lipid modifications of proteins, and it is estimatedthat 2% (by mass) of all cellular proteins may undergo such modification (1). Although the list ofdocumented prenylated proteins has grown rapidly since 1990, studies employing metabolic la-beling techniques and two-dimensional gel electrophoresis suggest that many more prenylatedpolypeptides remain to be identified (2). Some of the known substrates include all four p21ras

proteins (H-Ras, K-RasA, K-RasB, and N-Ras), nuclear lamins A and B, transducing-subunit,rhodopsin kinase, cGMP phosphodiesterase a-subunit, Ras related Rho and Rac proteins andg-subunits of heterotrimeric G proteins (3). Prenylation at the carboxyl-terminal residues of pro-teins terminating with a CAAX sequence (where C is cysteine, A is an aliphatic acid, X is anyamino acid) involves attachment of either farnesyl (C15) or geranylgeranyl (C20) isoprenoids (3–5).The enzymes which catalyze specific prenylation of the CAAX sequence are called farnesyltrans-ferase (FTase) and geranylgeranyltransferase type I (GGTase 1) (6–8). FTase prefers amino acid(X) to be either serine or methionine for farnesylation, while the presence of leucine at that positionis preferred by GGTase I (3,7,9).The Ras proteins are the best characterized substrates for FTasein vivo (3). Attachment of Ras

proteins to the inner surface of the plasma membrane requires the farnesyl group, and theirtransformation potential is abolished either by mutations in the CAAX consensus sequence, or inthe presence of FTase peptidomimetic inhibitors (10,11).The FTase is composed of two nonidentical subunits,a andb, with apparent molecular masses

of approximately 49 kDa and 46 kDa, respectively (6). cDNA cloning of thea-andb-subunitsrevealed their homology to the yeast genes Ram2 and Dpr1/Ram 1, respectively, which encode forthe subunits of yeast FTase (12,13). Cross-linking studies suggest that the CAAX peptide andfanesyl pyrophosphate (FPP) both bind to theb-subunit (14,15). Thea-subunit is required tostabilize theb-subunit, and plays a catalytic role in the farnesyl transfer reaction (16). Thea-sub-unit found in FTase is shared by GGTase I with a uniqueb-subunit of molecular mass 43 kDa (17).The brain and the adrenal gland both have higher levels of FTase activity than do other tissues

(12). Therefore, rat adrenal medulla pheochromocytoma (PC-12) cells were utilized for investi-

1 To whom correspondence and reprints requests should be addressed. Fax: 501-686-8169.

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS222,445–452 (1996)ARTICLE NO. 0764

4450006-291X/96 $18.00Copyright © 1996 by Academic Press, Inc.All rights of reproduction in any form reserved.


gating post-translational modification of thea-subunit. We demonstrate that thea-subunit of FTaseis a phosphoprotein, and phosphorylation is involved in the regulation of the FTase activity inPC-12 cells.

EXPERIMENTAL PROCEDURES

Materials.The anti-phosphotyrosine (PY-20) and protein G-plus agarose were obtained from Santa Cruz Biotechnology(CA). A polyclonal rabbit anti-peptide antibobody raised against rat FTasea-subunit (amino acids 357–375) was used inthis study. The normal rabbit IgG was purchased from Harlan Bioproducts (IN) and Caltag Laboratories (CA). Phospho-serine/phosphothreonine-specific protein phosphatase-1 (PP-1) was purchased from Upstate Biotechnology Inc (NY).Calyculin A (CL-A) was obtained from Boehringer Mannheim (IN).Cell lines and culture conditions.PC-12 cells (a gift from Dr. S. Chang, Dept. of Anatomy, U.A.M.S.) were grown in

RPMI containing 10% horse serum and 5% Fetal bovine serum (FBS) in a 5% CO2 incubator. Human fibroblast GM37 cellswere grown in MEM with 10% FBS respectively.Immunoblot analysis.Samples were subjected to electrophoresis on a 10% SDS-PAGE gel and transferred to nitrocel-

lulose filters as described earlier (18,19). The membrane was incubated with 1:1500 dilution of rabbit anti a-subunit IgGat room temperature for 1 hr. After washing, affinity-purified goat anti rabbit IgG-horseradish peroxidase (1:3500) wasadded followed by detection of bands using the ECL reagent and hyperfilm.Immunodepletion experiments.Cells were harvested and disrupted in lysis buffer (50 mM Tris-HCl, pH 7.5, 50mM

ZnCl2, 3 mM MgCl2, 20 mM KCl, 1 mM DTT, 0.4% (w/v) octyl-b-glucopyranoside and mixtures of protease inhibitors)by aspirating through 21 gauge needle. The cell extract was centrifuged at 100,000 × g for 1 hr at 4°C. The supernatant (1mg of protein) was immunoprecipitated with increasing amounts of anti a-subunit IgG. Immunoprecipitation and immu-noblotting were done as described above.[32P]Orthophosphate labeling and immunoprecipitation.Subconfluent cultures of PC-12 cells were transferred to phos-

phate free medium containing 15% dialyzed FBS for 1 hr and labeled with32P-labeled inorganic phosphate (ICN) at 1mCi/3ml of medium. After 3 hrs, the cultures were washed two times with ice cold phosphate buffered saline and lysed in RIPAbuffer for 20 min at 4°C (19). Lysates were cleared by centrifugation for 10 min at 4°C. Supernatant (2 mg of protein from32P-labeled cells or 1 mg of protein from unlabeled cells) were incubated with different IgGs for 2 hrs at 4°C. Protein-Gplus agarose (20ml) was added and further incubated for 2 hrs with shaking. Precipitates were washed 6 times with lysisbuffer followed by once with 10mM Tris-HC1/0.1% NP-40. The agarose pellets were then extracted with SDS samplebuffer and were subjected to 10% SDS-PAGE (16 cm × 20 cm). Proteins between 30 and 87 kDa were transferred to anitrocellulose membrane and detected by autoradiography.Assay of farnesyltransferase activity.PC-12 cells were harvested, pooled and disrupted by repeated aspiration as de-

scribed earlier (12). The supernatants were assayed for FTase activity by measuring the amount of of [3H]farnesyltransferred from [3H]FPP to a biotinylated peptide with the sequence, KTSCVIM, corresponding to the carboxy-terminalregion of p21H-ras protein as described previously (20). This assay has two discrete steps: a 60-minute enzyme reactionwhich occurs in solution and a peptide capture onto a solid phase. In the reaction step, each assay well contained in a finalvolume of 100ml the indicated quantity of cytosolic extract in 30ml lysis buffer plus 125 mM Tris-HCl, 100mM ZnCl2,50 mM MgCl2, 1mM DTT, 0.2% (w/v) octyl-b-glucopyranoside, 0.9mM biotinylated-peptide, and 4.65 pmol of [3H]FPP.After incubation for 1 hr at 37°C, the reaction was stopped by the addition of 50ml of 0.25 M sodium phosphate, pH 4.0.The biotynylated peptide was then captured on a streptavidin-coated surface during a 20 min incubation. After washing thewells, the radioactivity associated with the wells was measured to allow the quantitation of the peptide-bound [3H]farnesylgroups. Activity is represented in terms of pmol of [3H]farnesyl transferred from [3H]FPP to peptide in 60 min. Each valueis the average of duplicate determination.Dephosphorylation reaction.The dephosphorylation reaction was carried out by incubating 100mg of the cytosolic

extract with different amounts of PP-1 (0.4 and 2 units) at 30°C for 30 min in 150ml of reaction buffer as suggested bythe supplier. FTase activity in PP-1 treated and untreated samples was measured as described above.Treatment of PC-12 cells with CL-A.The PC-12 cells were treated with protein phosphatase inhibitor CL-A (2 and 10

nM) for a period of 60 min. The cells were harvested and FTase activity was measured as described above.

RESULTS

Immunoblot analysis of rat PC-12 cell extract using an anti rata-subunit IgG showed a cross-reactive protein of apparent molecular mass 49kDa (Fig. 1), and the corresponding protein bandwas not detected in cell extracts of human fibroblast GM 37 and human hepatoma HepG2 cells.Extracts from both human cell lines cross-reacted with a band of expected size using human antia-subunit IgG (data not shown). The lack of cross-reactivity of the anti rata-subunit IgG to humancell extracts is to be expected as the human and rat carboxyl-terminal sequences of thea-subunitdiffer considerably from each other in the region selected for producing antibody.

Vol. 222, No. 2, 1996 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

446


To establish the specificity of the antia-subunit antibody, immunodepletion studies were carriedout. Fig. 2 (A&B) shows that this antibody caused a significant decrease in the FTase activity ofthe supernatant fraction from PC-12 cell lysates in a dose-dependent manner. More than 90% of themeasurable FTase activity could be removed from the supernatant at 2mg of an anti rata-subunitIgG/mg of extract. Fig. 2C shows the immunoblot analysis of the depleted lysates and immuno-precipitates obtained using increasing amounts of antibody. The intensity of the 49 kDa proteinband diminished from the supernatant liquids, with a corresponding band appearing in the pre-cipitates, in a manner dependent on antibody dosage. The intensity of the 49 kDa protein bandremaining in the supernatant correlated well with the residual FTase activity in the supernatant.Control experiments using a normal rabbit IgG (up to 2mg) did not result in any detectabledepletion of the 49 kDa protein (Fig. 2C) or a decrease in the FTase activity (Fig. 2, A&B) fromthe supernatant. These results show that the anti-peptide antibody utilized in our study is specifi-cally reacting with thea-subunit of rat FTase.The phosphorylation status of thea-subunit was directly evaluated by [32P]orthophosphate

labeling of PC-12 cells, followed by immunoprecipitation with the above antibody from32P-labeled and unlabeled PC-12 cells. As detailed under “Experimental Procedures”, PC-12 cellsgrown normally, or after incubation with32P, were lysed in RIPA buffer and immunoprecipitatedwith an anti rata-subunit IgG or a normal rabbit IgG, as well as antibody raised against a peptideselected from the carboxy-terminal sequence of the human homologue of thea-subunit. Fig. 3shows that the rata-subunit antibody immunoprecipitated a 49 kDa radiolabeled band (Panel A,lane 9) which corresponded in electrophoretic mobility to a immunoprecipitated protein band fromunlabeled PC-12 cells (Panel B, lane 4) and from total cell lysate (Panel B, lane 1). Immunoblottingof the filter shown in Panel A confirmed that the 49 kDa32P-labeled band in Panel A (lane 9) alsocross-reacted with the rat a-subunit antibody (Panel B, lane 9). As expected, the amount of thea-subunit was significantly decreased in the immunodepleted lysates (Panel C). No immunopre-cipitable band was observed from PC-12 cell extracts under identical conditions using either an antihuman a-subunit IgG or a normal rabbit IgG. We conclude from these results that the rata-subunitof FTase is labeled with32P in PC-12 cells.To test the possibility that any tyrosine residues are phosphorylated in thea-subunit, cellular

extracts were immunoprecipitated with an anti rata-subunit IgG, and evaluated for cross-reactivitywith an anti-phosphotyrosine (PY-20) by immunoblotting Rata-subunit protein band did notcross-react with this antibody (results not shown).

FIG. 1. Specificity of rata-subunit antibody by immunoblotting. Immunoblot analysis of soluble extracts of various celllines probed with the rat antia-subunit IgG. 30mg of protein/lane was subjected to electrophoresis on a 10% SDS-PAGEand transferred to nitrocellulose filters. The filters were incubated for 1 hr at room temperature with 1mg/ml of rabbit antia-subunit IgG, followed by incubation with HRP conjugated goat anti-rabbit IgG. Blots were developed with ECL reagent.


447


FIG. 2. Immunodepletion of FTase activity from PC-12 cell lysates using an anti rata-subunit IgG. Panel A: Mea-surement of FTase activity in immunodepleted extracts of PC-12 cells. For each experiment, PC-12 cell lysate at 2 mg/mlwas immunodepleted with either a non-specific IgG or an anti-a-subunit IgG at various indicated quantities. The super-natants were assayed for FTase activity. The values shown are the averages of duplicate assays from the same lysates.l—l, total cell extract;V—V, 0.1mg; x—x, 0.4mg; v—v, 1 mg; *—*, 2 mg of an anti rata-subunit IgG; +—+, 1mg; V—V, 2 mg normal rabbit IgG (Harlan Bioproducts). Panel B: Relative FTase activity in immunodepleted extractsusing increasing doses of an anti-a-subunit IgG and a normal rabbit IgG. Panel C: Immunoblot analysis after immuno-depletion of lysates. After immunodepletion experiments, 10% of the immunodepleted lysates or 50% of the correspondingimmunoprecipitated pellets per sample were used for immunoblot analysis. The 55 kDa band corresponds to the heavy chainof IgG.


448


In order to understand the significance of phosphorylation in the farnesyl transfer reaction,PC-12 cells were treated with two different concentrations of protein phosphatases inhibitor CL-A,and FTase activity was measured in the untreated and treated cell extracts. A dose-dependentdecrease of 25% and 45% in the FTase activity of PC-12 cells was observed on treatment with 2and 10 nM of CL-A respectively (Fig. 4A). In consistent with the above observation, treatment ofPC-12 cell extract with phosphoserine/phosphothreonine specific PP-1 (2 units) resulted in an100% increase in FTase activity as compared to untreated extracts or extracts treated with 0.4 unitsof PP-1 (Fig. 4B). Similar PP-1 treatment of human fibroblast GM37 cell extract resulted in an375% increase in the FTase activity (Fig. 4C).

DISCUSSION

Immunoprecipitation of labeleda-subunit from32P metabolically labeled PC-12 cell extractsprovides direct evidence that the endogenousa-subunit of heterodimeric FTase is phosphorylatedin vivo.

FIG. 3. Immunoprecipitation of labeleda-subunit from32P-labeled PC-12 cells. Immunoprecipitates from32P meta-bolically labeled or unlabeled PC-12 cell extracts using either an anti rata-subunit IgG or a normal rabbit IgG werefractionated by SDS-PAGE, blotted onto nitrocellulose, and subjected to autoradiography. The filters were exposed toKodak XAR-5 film for 48 h at −70°C. Lane 1, total extract; Lanes 2 & 7, protein G-agarose alone; 3 & 8, normal rabbitIgG (Harlan Bioproducts); 4 & 9, anti rata-subunit IgG; 5 & 10, anti humana-subunit IgG; 6 & 11, normal rabbit IgG(Caltag Laboratories). The position and size (kDa) of the molecular weight size standards are indicated. Panel A: Gelcontaining unlabeled and32P-labeled immunoprecipitates followed by autoradiography. Panel B: Immunoblot analysis ofthe filter shown in panel A using an anti-a-subunit IgG. Panel C: Immunoblotting of immunodepleted supernatant using theanti-a-subunit IgG.


449


FIG.4.

FTaseactivity

aftertreatmentofPC-12cells

with

calyculin

A,andeffectofphosphoserine/phosphothreonine-specific

PP-1

treatmentofcytosolic

extractson

FTaseactivity.

PanelA:P

C-12cells

weretreatedwith

indicatedconcentrations

ofcalyculin

Afor2hrandFTaseactivity

was

measuredinboththeuntreatedandtreatedextracts.

V—

V,untreated

PC-12

cells;x—

x,2nMcalyculin

A;l

—l,10

nMcalyculin

A.PanelB:PC-12cellextractwas

subjectedtoPP-1

treatmentfollowed

bymeasurementofFTaseactivity.x—

x,PP-1

alone;

V—

V,cytosolic

extract;*—

*,0.4units

ofPP-1;v—

v,2units

ofPP-1.PanelC:FTaseactivity

was

measuredinPP-1

treatedGM37

cellextract.

l—

l,cytosolic

extract;x—

x,2

units

ofPP-1.


450


Although the significance of our finding requires additional study, phosphorylation of thea-sub-unit may be involved in modifying the proposed roles of this subunit in the FTase reaction. Earliermutational studies to probe the role of thea-subunit in the farnesylation reaction suggested that itis required to stabilize theb-subunit, and also plays a catalytic role in the farnesyl transfer reaction(16). Specific deletions in thea-subunit have defined regions of the proteins which are required forthe stabilization and catalytic functions. Deletions of 51 amino acids at the amino-terminus or 20amino acids from the carboxyl-terminus of thea-subunit produced an appreciable decrease inFTase activity without affecting the stability of either subunit. In addition, substitution of the aminoacid at position 164 produced ana-subunit that stabilized theb-subunit normally, however the rateof transfer of farnesyl group to p21raswas markedly reduced. The above results suggested that thestabilization and catalytic roles of the a-subunit are separable and are probably carried out bydifferent regions of the protein. The recent demonstration that the specific activity of the bacteriallyexpressed FTase enzyme is the same as that of the enzyme purified from eukaryotic tissues makesit likely that the phosphorylation is not directly affecting the catalytic role of thea-subunit (15).In view of the above results, it is possible that the phosphorylation may be modulating the role ofthe a-subunit in stabilization of theb-subunit. One possible scenario is that the phosphorylateda-subunit has a lower affinity for theb-subunit than the unphosphorylated form, and the phos-phorylateda-subunit by itself is less stable and is preferentially degraded. The increased FTaseactivity upon dephosphorylation may be due to the availability of more unphosphorylateda-sub-unit and an increase in the heterodimerization of the unphosphorylateda-subunit with theb-sub-unit. The analysis of the predicted amino acid sequence of thea-subunit revealed multiple con-sensus amino acid sequences for known protein kinases including mitogen activated kinase, caseinkinase I and cAMP/cGMP-dependent protein kinases (21). Studies are in progress to identify theamino acids phosphorylated in thea-subunit, and to determine the phosphorylation status of theb-subunit in PC-12 cells.The a-subunit found in FTase is shared by GGTase I with a differentb-subunit of molecular

mass 43 kDa (17). The role of thea-subunit in the geranylgeranyl transfer reaction, and the natureof its interaction with theb-subunit of GGTase I is not known. However, differential phosphor-ylation of thea-subunit modulating its affinity between theb-subunits of FTase and GGTase I orwith other unrelated protein (22) is a distinct possibility.

ACKNOWLEDGMENTS

We are thankful to Dr. S. Goldstein for providing us with the human fibroblast cell line. We are also thankful to Drs. AlanD. Elbein and Timothy Chambers for critical reading of the manuscript.

REFERENCES

1. Maltese, W. A., (1990)FASEB J.4, 3319–3328.2. Epstein, W. W., Lever, D., Leining, L. M., Bruenger, E., and Rilling, H. C., (1991)Proc. Natl. Acad. Sci. U.S.A.88,

9668–9670.3. Clarke, S., (1992)Annu. Rev. Biochem.61, 355–386.4. Casey, P., and Seabra, M. C., (1996)J. Biol. Chem.271,5289–5292.5. Schafer, W. R., and Rine, J., (1992)Ann. Rev. Genet.30, 209–237.6. Reiss, Y., Goldstein, J. L., Seabra, M. C., Casey, P. J., and Brown, M. S., (1990)Cell 62, 81–88.7. Reiss, Y., Stradley, S. J., Gierasch, L. M., Brown, M. S., and Goldstein, J. L., (1991)Proc. Natl. Acad. Sci. USA88,

732–736.8. Moomaw, J. F., and Casey, P. J., (1992)J. Biol. Chem.267,17438–17443.9. Moores, S. L., Schaber, M. D., Mosser, S. D., Rands, E., O’Hara, M. B., Garsky, V. M., Marshall, M. S., Pompliano,

D. L., and Gibbs, J. B., (1991)J. Biol. Chem.266,14603–14610.10. Hancock, J. F., Magee, A. I., Childs, J. E., and Marshall, C. J., (1989)Cell 57, 1167–1177.11. James, G. L., Goldstein, J. L., Brown, M. S., Rawson, T. E., Somers, T. C., McDowell, R. S., Crowley, C. W., Lucas,

B. K., Levinson, A. D., and Marsters, J. C., (1993)Science260,1937–1942.12. Chen, W. J., Andres, D. A., Goldstein, J. L., Russell, D. W., and Brown, M. S., (1991)Cell 66, 327–334.13. Chen, W. J., Andres, D. A., Goldstein, J. L., and Brown, M. S., (1991)Proc. Natl. Acad. Sci.88, 11368–11372.


451


14. Reiss, Y., Seabra, M. C., Armstrong, S. A., Slaughter, C. A., Goldstein, J. L., and Brown, M. S., (1991)J. Biol. Chem.266,10672–10677.

15. Omer, C. A., Kral, A. M., Diehl, R. E., Prendergast, G. C., Powers, S., Allen, C. M., Gibbs, J. B., and Kohl, N. E.,(1993)Biochemistry32, 5167–5176.

16. Andres, D. A., Goldstein, J. L., Ho, Y. K., and Brown, M. S., (1993)J. Biol. Chem.268,1383–1390.17. Zhang, F. L., Diehl, R. E., Kohl, N. E., Gibbs, J. B., Giros, B., Casey, P. J., and Omer, C. A., (1994)J. Biol. Chem.269,

3175–3181.18. Mehta, K. D., Leung, D., Lefebvre, L., and Smith, M., (1990)J. Biol. Chem.265,8802–8807.19. Mehta, K. D., Chen, W. J., Goldstein, J. L., and Brown, M. S., (1991)J. Biol. Chem.266,10406–10414.20. Beresini, M. H., Yeh, S. H., Chang, A. C., Yen, R., Galloway, A. L., Wong, W. L., and Marsters, J. C., (1995)FASEB

J. 9(6), A1315.21. Kennelly, P. J., and Krebs, E. G., (1991)J. Biol. Chem.266,15555–15558.22. Wang, T., Danielson, P. D., Li, B-y, Shah, P. C., Kim, S. D., and Donahoe, P. K., (1996)Science271,1120.


452

α-Subunit of Farnesyltransferase Is Phosphorylatedin Vivo:Effect of Protein Phosphatase-1 on...

Documents

Transcript of α-Subunit of Farnesyltransferase Is Phosphorylatedin Vivo:Effect of Protein Phosphatase-1 on...