Farnesyl Transferase Inhibitors Block the Farnesylation of CENP-E and CENP-F and Alter the Association of CENP-E with the Microtubules

Human tumor cell lines that are sensitive to the effects of farnesyl transferase inhibitors accumulate in G2 → M (except for cells with an activated Ha-ras that accumulate in G1). A search for CAAX box proteins from Swiss-Prot revealed more than 300 peptides. Of these, the centromeric proteins CENP-E and CENP-F are preferentially expressed during mitosis and are implicated as mediators of the G2 → M checkpoint. Experiments performed here show that peptides from the COOH-terminal CAAX box of CENP-E and CENP-F are substrates for farnesyl transferase but not geranylgeranyl transferase-I. Although both proteins are prenylated in the human tumor cell line DLD-1, their prenylation is completely inhibited by the farnesyl transferase inhibitor, SCH 66336. Immunohistochemical data with the lung carcinoma cell line, A549, showed that preventing the farnesylation of CENP-E and CENP-F by treatment with the farnesyl transferase inhibitor SCH 66336 does not affect their localization to the kinetochores. However, the presence of farnesyl transferase inhibitors alters the association between CENP-E and the microtubules. Our results imply that the inhibition of CENP-E farnesylation results in the alteration of the microtubule-centromere interaction during mitosis and results in the accumulation of cells prior to metaphase. Human tumor cell lines that are sensitive to the effects of farnesyl transferase inhibitors accumulate in G2 → M (except for cells with an activated Ha-ras that accumulate in G1). A search for CAAX box proteins from Swiss-Prot revealed more than 300 peptides. Of these, the centromeric proteins CENP-E and CENP-F are preferentially expressed during mitosis and are implicated as mediators of the G2 → M checkpoint. Experiments performed here show that peptides from the COOH-terminal CAAX box of CENP-E and CENP-F are substrates for farnesyl transferase but not geranylgeranyl transferase-I. Although both proteins are prenylated in the human tumor cell line DLD-1, their prenylation is completely inhibited by the farnesyl transferase inhibitor, SCH 66336. Immunohistochemical data with the lung carcinoma cell line, A549, showed that preventing the farnesylation of CENP-E and CENP-F by treatment with the farnesyl transferase inhibitor SCH 66336 does not affect their localization to the kinetochores. However, the presence of farnesyl transferase inhibitors alters the association between CENP-E and the microtubules. Our results imply that the inhibition of CENP-E farnesylation results in the alteration of the microtubule-centromere interaction during mitosis and results in the accumulation of cells prior to metaphase. farnesyl protein transferase inhibitor farnesyl protein transferase geranylgeranyl protein transferase type I human tumor cell line phosphate-buffered saline microtubule-associated protein Farnesyl transferase inhibitors (FTIs)1 revert the anchorage-independent growth of several diverse tumors with and without ras mutations (2Nagasu T. Yoshimatsu K. Rowell C. Lewis M.D. Garcia A.M. Cancer Res. 1995; 55: 5310-5314PubMed Google Scholar,3Sepp-Lorenzino L. Ma Z. Rands E. Kohl N.E. Gibbs J.B. Oliff A. Rosen N. Cancer Res. 1995; 55: 5302-5309PubMed Google Scholar). 2H. R. Ashar, L. James, K. Gray, D. Carr, M. McGuirk, E. Maxwell, L. Armstrong, R. J. Doll, A. G. Taveras, W. R. Bishop, and P. Kirschmeier, submitted for publication. The compounds have limited effects on the growth of normal cells and are now being actively pursued for anti-cancer therapy (4Adjei A.A. Ehrlichman C. Davis J.N. Cutler J.L. Sloan J.A. Marks R.S. Hanson L.J. Atherton P. Bishop W.R. Kirschmeier P. Kauffmann S.H. Cancer Res. 2000; 60: 1871-1877PubMed Google Scholar, 5Hill B.T. Perrin D. Kruczynski A. Crit. Rev. Oncol. Hematol. 2000; 33: 7-23Crossref PubMed Scopus (51) Google Scholar, 6Zujewski J. Horak I.D. Bol C.J. Woestenborghs R. Bowden C. End D.W. Piotrovsky V.K. Chiao J. Belly R.T. Todd A. Kopp W.C. Kohler D.R. Chow C. Noone M. Hakim F.T. Larkin G. Gress R.E. Nusenblatt R.B. Kremer A.B. Cowan K.H. J. Clin. Oncol. 2000; 18: 927-941Crossref PubMed Google Scholar, 7End D.W. Invest. New Drugs. 1999; 17: 241-258Crossref PubMed Scopus (100) Google Scholar, 8Wright J. Blatner G.L. Cheson B.D. Oncology. 1999; 13 (, 1530, 1533): 1527PubMed Google Scholar, 9Liu M. Bryant M.S. Chen J. Lee S. Yaremko B. Lipari P. Malkowski M. Ferrari E. Nielsen L. Prioli N. Dell J. Sinha D. Syed J. Korfmacher W.A. Nomeir A.A. Lin C-C. Wang L. Taveras A.G. Doll R.J. Ngoroge F.G. Mallams A.K. Remiszewski S. Catino J.J. Girijavallabhan V.M. Kirschmeier P. Bishop W.B. Cancer Res. 1998; 58: 4947-4956PubMed Google Scholar). Although FTIs effectively suppress the anchorage-independent growth of hTCLs with Ha-ras mutations, hTCLs with K-ras or N-ras mutations show a variable response (2Nagasu T. Yoshimatsu K. Rowell C. Lewis M.D. Garcia A.M. Cancer Res. 1995; 55: 5310-5314PubMed Google Scholar, 3Sepp-Lorenzino L. Ma Z. Rands E. Kohl N.E. Gibbs J.B. Oliff A. Rosen N. Cancer Res. 1995; 55: 5302-5309PubMed Google Scholar, 10Kohl N.E. Mosser S.D. deSolms S.J. Giuliani E.A. Pompliano D.L. Graham S.L. Smith R.L. Scolnick E.M. Oliff A. Gibbs J.B. Science. 1993; 260: 1934-1937Crossref PubMed Scopus (619) Google Scholar, 11James 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. Marsters Jr., J.D. Science. 1993; 260: 1042-1937Crossref Scopus (607) Google Scholar, 12Garcia A.M. Rowell C. Ackerman K. Kowalczyk J.J. Lewis M.D. J. Biol. Chem. 1993; 268: 18415-18418Abstract Full Text PDF PubMed Google Scholar, 13Cox A.D. Garcia A.M. Weskwick J.K. Kowalczyk J.J. Lewis M.D. Brenner D.A. Der C.J. J. Biol. Chem. 1994; 269: 19203-19206Abstract Full Text PDF PubMed Google Scholar). This can be explained partially because FTIs are unable to prevent the membrane association of K-Ras or N-Ras. These two Ras isoforms are alternatively prenylated by the enzyme geranylgeranyl protein transferase (GGPT-I) in the presence of FTIs (14James G.L. Goldstein J.L. Brown M.S. J. Biol. Chem. 1995; 270: 6221-6226Abstract Full Text Full Text PDF PubMed Scopus (295) Google Scholar, 15Whyte D.B. Kirschmeier P. Hockenberry T.N. Nunez-Oliva I. James L. Catino J.J. Bishop W.R. Pai J. J. Biol. Chem. 1997; 272: 14459-14464Abstract Full Text Full Text PDF PubMed Scopus (728) Google Scholar, 16Njoroge F.G. Vibulbhan B. Rane D.F. Bishop W.R. Petrin J. Patton R. Bryant M.S. Chen K.J. Nomeir A.A. Lin C.C. Liu M. King I. Chen J. Lee S. Yaremko B. Dell J. Lipari P. Malkowski M. Li Z. Catino J. Doll R.J. Girijavallabhan V. Ganguly A.K. J. Med. Chem. 1997; 40: 4290-4301Crossref PubMed Scopus (54) Google Scholar, 17Rowell C.A. Kowalczyk J.J. Lewis M.D. Garcia A.M. J. Biol. Chem. 1997; 272: 14093-14097Abstract Full Text Full Text PDF PubMed Scopus (330) Google Scholar, 18Zhang F.L. Kirschmeier P. Carr D. James L. Bond R.W. Wang L. Patton R. Windsor W.T. Syto R. Zhang R. Bishop W.R. J. Biol. Chem. 1997; 272: 10232-10239Abstract Full Text Full Text PDF PubMed Scopus (201) Google Scholar). Thus, the antitumor effects of FTIs in sensitive hTCLs with K-ras or N-ras mutations may be independent of effects on K-ras or N-ras function. A critical hint to additional biological effects of FTIs was obtained from cell cycle studies showing that most cell lines that are sensitive to FTI treatment accumulate in the G2 → M phase of the cell cycle.2 A protein that is farnesylated and contributes to cell cycle progression would be an appropriate candidate for mediating this biological effect. Since all farnesylated proteins end in a CAAX sequence (where C is a cysteine, A is an aliphatic amino acid, and X is a methionine, glutamine, serine, or threonine), a data base search was performed to identify CAAX box proteins that could potentially be important for the G2 → M transition or during mitosis. Of the hundreds of potentially farnesylated peptides that were identified from Swiss-Prot, two proteins, CENP-E and CENP-F/mitosin, were previously associated with mitosis (19–37). Specifically, a role for both CENP-E and CENP-F/mitosin prior to the metaphase-anaphase transition was previously suggested (20Yen T.J. Compton D.J. Wise D. Zinkowski R.P. Brinkley B.R. Earnshaw W.C. Cleveland D.W. EMBO J. 1991; 10: 1245-1254Crossref PubMed Scopus (298) Google Scholar, 21Liao H. Li G. Yen T.J. Science. 1994; 265: 394-398Crossref PubMed Scopus (118) Google Scholar, 22Wordeman L. Mitchison T. J. Cell Biol. 1995; 128: 95-105Crossref PubMed Scopus (327) Google Scholar, 23Lombillo V.A. Nislow C. Yen T.J. Gelfand V.I. McIntosh J.R. J. Cell Biol. 1995; 128: 107-115Crossref PubMed Scopus (196) Google Scholar, 24Thrower D.A. Jordan M.A. Schaar B.T. Yen T.J. Leslie W. EMBO J. 1995; 14: 918-926Crossref PubMed Scopus (57) Google Scholar, 25Brown K.B. Wood K.W. Cleveland D.W. J. Cell Sci. 1996; 109: 961-969PubMed Google Scholar, 26Yao X. Anderson K.L. Cleveland D.W. J. Cell Biol. 1997; 139: 435-447Crossref PubMed Scopus (189) Google Scholar, 27Cooke C. Schaar B. Yen T.J. Earnshaw W.C. Chromosoma. 1997; 106: 446-455Crossref PubMed Scopus (130) Google Scholar, 28Wood K.W. Sakowitz R. Goldstein L.S.B. Cleveland D.W. Cell. 1997; 91: 357-366Abstract Full Text Full Text PDF PubMed Scopus (335) Google Scholar,29Schaar B.T. Chan G.K.T. Maddox P. Salmon E.D. Yen T.J. J. Cell Biol. 1997; 139: 1373-1382Crossref PubMed Scopus (264) Google Scholar, 31Zecevic M. Catling A.D. Catling A.D. Eblen S.T. Renzi L. Hittle J.C. Yen T.J. Gorbsky G.J. Weber M.J. J. Cell Biol. 1998; 142: 1547-1558Crossref PubMed Scopus (194) Google Scholar, 32Chan G.K.T. Jablonski S.A. Sudakin V. Hittle J.C. Yen T.J. J. Cell Biol. 1999; 146: 91-954Crossref Scopus (312) Google Scholar, 33Rattner J.B. Rao A. Fritzler M.J. Valencia D.W. Yen T.J Cell Motil. Cytoskeleton. 1993; 26: 216-226Crossref Scopus (182) Google Scholar, 34Liao H. Winkfein R.J. Mack G. Rattner J.B. Yen T.J. J. Cell Biol. 1995; 130: 507-518Crossref PubMed Scopus (301) Google Scholar, 35Zhu X. Mancini M. Chang K. Liu C. Chen C. Shan B. Jones D. Yang-Feng T.L. Lee W. Mol. Cell. Biol. 1995; 15: 5017-5029Crossref PubMed Scopus (124) Google Scholar, 36Zhu X. Ding L. Pei G. J. Cell. Biochem. 1997; 66: 441-449Crossref PubMed Scopus (24) Google Scholar, 37Zhu X. Mol. Cell. Biol. 1999; 19: 1016-1024Crossref PubMed Scopus (28) Google Scholar). Both CENP-E and CENP-F/mitosin are localized on kinetochores, and CENP-E is attached to the kinetochores and microtubules (19Pennisi E. Science. 1998; 279: 477-478Crossref PubMed Scopus (33) Google Scholar, 20Yen T.J. Compton D.J. Wise D. Zinkowski R.P. Brinkley B.R. Earnshaw W.C. Cleveland D.W. EMBO J. 1991; 10: 1245-1254Crossref PubMed Scopus (298) Google Scholar, 29Schaar B.T. Chan G.K.T. Maddox P. Salmon E.D. Yen T.J. J. Cell Biol. 1997; 139: 1373-1382Crossref PubMed Scopus (264) Google Scholar, 30Chan G.K.T. Schaar B.T. Yen T.J. J. Cell Biol. 1998; 143: 49-63Crossref PubMed Scopus (225) Google Scholar, 31Zecevic M. Catling A.D. Catling A.D. Eblen S.T. Renzi L. Hittle J.C. Yen T.J. Gorbsky G.J. Weber M.J. J. Cell Biol. 1998; 142: 1547-1558Crossref PubMed Scopus (194) Google Scholar). Rieder and others have clearly shown the existence of a spindle checkpoint or a kinetochore attachment checkpoint prior to metaphase progression. This checkpoint appears to monitor the interaction of chromosomes with the spindle microtubules, and chromosomes that fail to attach to the spindle appear to inhibit sister chromatid separation at least transiently (38Gorbsky G.J. Chen R.H. Murray A.W. J. Cell Biol. 1998; 141: 1193-1205Crossref PubMed Scopus (199) Google Scholar, 39Rieder C.L. Cole R.W. J. Cell Biol. 1998; 142: 1013-1022Crossref PubMed Scopus (107) Google Scholar, 40Rieder C.L. Cole R.W. Khodjakov A. Sluder G. J. Cell Biol. 1995; 130: 941-948Crossref PubMed Scopus (581) Google Scholar, 41Rieder C.L. Schultz A. Cole R. Sluder G. J. Cell Biol. 1994; 127: 1301-1310Crossref PubMed Scopus (444) Google Scholar). It was demonstrated that sister chromatid separation is defective when CENP-E function is blocked by microinjection of affinity-purified antibodies to CENP-E (23Lombillo V.A. Nislow C. Yen T.J. Gelfand V.I. McIntosh J.R. J. Cell Biol. 1995; 128: 107-115Crossref PubMed Scopus (196) Google Scholar, 28Wood K.W. Sakowitz R. Goldstein L.S.B. Cleveland D.W. Cell. 1997; 91: 357-366Abstract Full Text Full Text PDF PubMed Scopus (335) Google Scholar, 29Schaar B.T. Chan G.K.T. Maddox P. Salmon E.D. Yen T.J. J. Cell Biol. 1997; 139: 1373-1382Crossref PubMed Scopus (264) Google Scholar). CENP-F/mitosin, which is one of the first proteins that associates with the kinetochores in prophase, is expressed only during mitosis and is rapidly degraded thereafter (34Liao H. Winkfein R.J. Mack G. Rattner J.B. Yen T.J. J. Cell Biol. 1995; 130: 507-518Crossref PubMed Scopus (301) Google Scholar). These findings signify the importance of functional CENP-E and CENP-F/mitosin during mitosis and make them ideal candidate proteins for FTI targets. A detailed study was performed on CENP-E and CENP-F/mitosin to determine whether these proteins were substrates for the enzyme FPT in vitro and in vivo and to determine whether prenylation had a role in the function of the two proteins. The hTCLs, A549, and NCI-H460 (derived from lung carcinomas), MIA Pa Ca-2 (derived from pancreatic carcinoma), and DLD-1 (derived from colon carcinoma) were obtained from American Type Culture Collection (Manassas, VA) and maintained according to their recommendation. The farnesyl transferase inhibitor SCH 66336 was described previously (16Njoroge F.G. Vibulbhan B. Rane D.F. Bishop W.R. Petrin J. Patton R. Bryant M.S. Chen K.J. Nomeir A.A. Lin C.C. Liu M. King I. Chen J. Lee S. Yaremko B. Dell J. Lipari P. Malkowski M. Li Z. Catino J. Doll R.J. Girijavallabhan V. Ganguly A.K. J. Med. Chem. 1997; 40: 4290-4301Crossref PubMed Scopus (54) Google Scholar). To collect total protein, cells grown in appropriate medium were washed with PBS and scraped off of 10-cm dishes. After one more wash with PBS, the cell pellets were resuspended in a 1× protease inhibitor mixture mix (Complete Tablets from Roche Molecular Biochemicals), and 1% Nonidet P-40 was added to the pellets. After a 30-min incubation on ice, the pellets were sonicated, and the insoluble cell debris was removed by centrifugation at 15,000 × g for 5 min. The protein concentration was measured using the BCA protein assay from Pierce. 6 or 8% gels were used to detect CENP-E and CENP-F/mitosin from 60 μg of total protein. Proteins were transferred to polyvinylidene difluoride membranes overnight at 20 V and 4 °C and probed with the specific antibodies. For CENP-F/mitosin, the mitosin antibody from Transduction Laboratories was diluted 1:1000. For CENP-E, a polyclonal antibody was synthesized by immunizing two rabbits (SYNPEP Corp., Dublin, CA). This antibody was diluted 1:1000. For actin, the ascites fluid (Ab-1 antibody; Oncogene Research Products) was diluted 1:500. Horseradish peroxidase-linked secondary antibodies were obtained from New England Biolabs. For FACs analysis 0.5–1 × 106 cells were grown in 10-cm tissue culture plates with or without 1 μm SCH 66336. After 72 h, cells were collected from the culture medium. These were mixed with cells from the tissue culture plate after trypsinization. Cells were then washed with PBS, fixed with 70% methanol for 30 min at 4 °C, and stained directly with propidium iodide solution (0.05 mg/ml propidium iodide, 0.1% Triton X-100, 0.1 mm EDTA, 0.05 mg/ml RNase A). 10,000 stained cells were analyzed by flow cytometry in a Becton Dickinson flow cytometer and analyzed using MODFIT LT (Verity Software Inc.) Biotinylated peptides derived from the carboxyl terminus of Ha-Ras, a mutant form of lamin B (CAIM → CAIL), CENP-E and CENP-F/mitosin were synthesized by SYNPEP. The sequences for the peptides are as follows: Ha-Ras, biotin-DESGPGCMSCKCVLS; mutant lamin B, biotin-SGSGYRASNRSCAIL; CENP-E, biotin-GKDVPECKTQ; CENP-F, biotin-SKGSENCKVQ. Biotinylated peptides were used as substrates to detect the transfer of the [3H]isoprenyl group from the farnesyl pyrophosphate or the geranylgeranyl pyrophosphate in the presence of purified recombinant human FPT or GGPT-I in vitro as described previously (18Zhang F.L. Kirschmeier P. Carr D. James L. Bond R.W. Wang L. Patton R. Windsor W.T. Syto R. Zhang R. Bishop W.R. J. Biol. Chem. 1997; 272: 10232-10239Abstract Full Text Full Text PDF PubMed Scopus (201) Google Scholar, 43Bishop W.R. Bond R. Petrin J. Wang L. Patton R. Doll R. Njoroge G. Catino J. Schwartz J. Windsor W. Syto R. Schwartz J. Carr D. James L. Kirschmeier P. J. Biol. Chem. 1995; 270: 30611-30618Abstract Full Text Full Text PDF PubMed Scopus (165) Google Scholar, 44Moomaw J.F. Casey P.J. J. Biol. Chem. 1992; 267: 17438-17443Abstract Full Text PDF PubMed Google Scholar). After a 35-min incubation, peptides were collected with streptavidin-coated beads and the incorporation of [3H]isoprene was detected using the scintillation proximity assay. Ha-Ras, which is a substrate for FPT but not GGPT-I, and a mutant lamin B peptide, which is a substrate for GGPT-I but not FPT, were used as controls. A549 cells (control (untreated), and cells treated with 1 μm 66336 for 4 days) were fixed in 3.7% formaldehyde, washed three times with PBS, and treated with 0.1% Triton X-100 to permeablize cells. After three washes with PBS, they were incubated with the primary antibody for 1 h, washed three times with PBS, and subsequently incubated with a fluorescein isothiocyanate-conjugated secondary antibody. They were also stained with propidium iodide solution (0.05 mg/ml propidium iodide, 0.1% Triton X-100, 0.1 mm EDTA, 0.05 mg/ml RNase A). After five washes in PBS and one wash in water, cells were immobilized in elvanol and visualized by fluorescence microscopy (× 1000). For CENP-F/mitosin, a monoclonal antibody was obtained from Transduction Laboratories. For CENP-E, a polyclonal antibody to the peptide LRRSQQAQDTSVISEHTDPQPSNKPL, near the COOH terminus of CENP-E, was synthesized by immunizing two rabbits (SYNPEP Corp., Dublin, CA). The antibodies from the two rabbits (enzyme-linked immunosorbent assay titers 12,500 and 10,500, while titer from preimmune serum was <50) were pooled together and affinity-purified by the manufacturer. For immunohistochemistry, the affinity-purified CENP-E antibody was diluted 1:300, and the mitosin antibody was diluted 1:100. Fluorescein isothiocyanate-conjugated secondary antibodies were obtained from ICN chemicals and used according to the manufacturer's specifications. pMev-DLD-1, DLD-1 cells expressing a mevalonate transporter gene (15Whyte D.B. Kirschmeier P. Hockenberry T.N. Nunez-Oliva I. James L. Catino J.J. Bishop W.R. Pai J. J. Biol. Chem. 1997; 272: 14459-14464Abstract Full Text Full Text PDF PubMed Scopus (728) Google Scholar), were grown in medium containing 20 μm mevastatin and 10% dialyzed FBS for 2 h. The culture medium was then replaced with culture medium with 10% dialyzed FBS containing 1 μm SCH 66336 (or Me2SO for control cells), 150 μCi/ml [3H]mevalonolactone, and 20 μm mevastatin to metabolically label prenylated proteins overnight. After washing twice with cold PBS, the cells were harvested in radioimmune precipitation buffer containing protease inhibitors. The primary CENP-F/mitosin antibody was added to the total protein lysate along with protein G/protein A-Sepharose beads, and the tubes with the lysate were rotated for 2 h. The beads were collected by centrifugation and washed twice before sample buffer was added. The samples were boiled in water bath and loaded on 8% SDS-polyacrylamide gel electrophoresis gels. The gels were dried on a Whatman paper after electrophoresis and viewed by autoradiography. The lysate depleted of CENP-F/mitosin was used for detection of labeled CENP-E and subsequently for K-Ras using the antibody K-Ras (F-234) from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). A549, NCI-H460, and MIA Pa Ca-2 cells were grown for 3 days without and with 1 μm SCH 66336. Microtubule-associated proteins were isolated from these human tumor cell lines using the previously described method for high microtubule-associated protein (MAP) content (45Keates R.A.B. Can. J. Biochem. Cell Biol. 1984; 62: 803-813Crossref PubMed Scopus (12) Google Scholar) after two cycles of depolymerization-polymerization of the microtubules. The depolymerization of the microtubules was achieved by incubation on ice. This was followed by polymerization of the microtubules at 32 °C in the presence of GTP, PEP, and pyruvate kinase as described previously (45Keates R.A.B. Can. J. Biochem. Cell Biol. 1984; 62: 803-813Crossref PubMed Scopus (12) Google Scholar). The microtubule-associated proteins were used for Western blot analysis of CENP-E and for silver staining using the Silver XPress kit from NOVEX. To identify farnesylated proteins that could contribute to a G2 → M pause, the FINDPATTERNS algorithm was used to obtain proteins with CAAX boxes at their COOH-terminal ends from the Swiss-Prot data base. This search revealed more than 300 CAAX box peptide sequences that could potentially be prenylated. Three of these proteins, CENP-E, CENP-F, and mitosin, with CAAX boxes CKTQ, CKVQ, and CKVQ, are known to be kinetochore-binding proteins that are expressed during mitosis (28Wood K.W. Sakowitz R. Goldstein L.S.B. Cleveland D.W. Cell. 1997; 91: 357-366Abstract Full Text Full Text PDF PubMed Scopus (335) Google Scholar,33Rattner J.B. Rao A. Fritzler M.J. Valencia D.W. Yen T.J Cell Motil. Cytoskeleton. 1993; 26: 216-226Crossref Scopus (182) Google Scholar, 34Liao H. Winkfein R.J. Mack G. Rattner J.B. Yen T.J. J. Cell Biol. 1995; 130: 507-518Crossref PubMed Scopus (301) Google Scholar, 35Zhu X. Mancini M. Chang K. Liu C. Chen C. Shan B. Jones D. Yang-Feng T.L. Lee W. Mol. Cell. Biol. 1995; 15: 5017-5029Crossref PubMed Scopus (124) Google Scholar). All three are expressed in a cell cycle dependent fashion, are transiently associated with kinetochores/centromeres during mitosis, and regulate progression through the G2 → M checkpoint (19Pennisi E. Science. 1998; 279: 477-478Crossref PubMed Scopus (33) Google Scholar). A protein pileup for CENP-F and mitosin revealed that they have identical amino acid sequences, except for a stretch of 96 amino acids that is missing in mitosin. Thus, the two proteins may be products of alternatively spliced transcripts from the same gene, and since we do not distinguish between the two forms, we refer to them as CENP-F/mitosin. The fact that CENP-E and CENP-F/mitosin have CAAX boxes and the fact that they play a role in G2 → M progression suggest that these proteins may be important in mediating some of the cell cycle changes induced by FTIs. To determine whether CENP-E and CENP-F/mitosin were substrates for FPT and GGPT-I, biotinylated peptides from the COOH-terminal ends of CENP-E and CENP-F/mitosin were synthesized. The peptides were tested as substrates for FPT and GGPT-I using 3H-labeled farnesyl pyrophospate and3H-labeled geranylgeranyl pyrophospate, respectively, as isoprene donors. A COOH-terminal Ha-Ras peptide and a mutated lamin B peptide (in which the CAAX box CAIM was changed to CAIL) were used as controls for farnesylation and geranylgeranylation, respectively. Both peptides corresponding to the COOH-terminal CAAX sequences of CENP-E and CENP-F/mitosin were substrates for FPT (Fig.1 A). The apparent Km of CENP-E and CENP-F/mitosin for FPT was approximately 200 nm as compared with the Ha-Ras peptide, which has a Km of approximately 550 nm. This indicates that in vitro, the COOH-terminal peptides of both CENP-E and CENP-F/mitosin had higher affinity for FPT than the synthetic Ha-Ras peptide. Significantly, the CENP-E and CENP-F/mitosin peptides were not substrates for GGPT-I in vitro. As expected, the mutated lamin B peptide was a good substrate for GGPT-I under the same conditions, indicating that the enzyme was functional in the experiments (Fig. 1 B). These results suggested that CENP-E and CENP-F/mitosin may not be in vivo substrates for GGPT-I. This is in contrast to K-Ras and N-Ras, which were shown to be substrates for both GGPT and FPT in vitro (18Zhang F.L. Kirschmeier P. Carr D. James L. Bond R.W. Wang L. Patton R. Windsor W.T. Syto R. Zhang R. Bishop W.R. J. Biol. Chem. 1997; 272: 10232-10239Abstract Full Text Full Text PDF PubMed Scopus (201) Google Scholar). In vivo, K-Ras and N-Ras are farnesylated proteins, but in FTI-treated cells these proteins are alternatively prenylated by GGPT-I (14James G.L. Goldstein J.L. Brown M.S. J. Biol. Chem. 1995; 270: 6221-6226Abstract Full Text Full Text PDF PubMed Scopus (295) Google Scholar, 15Whyte D.B. Kirschmeier P. Hockenberry T.N. Nunez-Oliva I. James L. Catino J.J. Bishop W.R. Pai J. J. Biol. Chem. 1997; 272: 14459-14464Abstract Full Text Full Text PDF PubMed Scopus (728) Google Scholar). The prenylation of CENP-E and CENP-F/mitosin was examined in pMev-DLD-1 cells. Studying the prenylation pattern of endogenous proteins is difficult in human tumor cell lines, where incorporation of mevalonate is limited. pMev-DLD-1 is a cell line developed to increase the incorporation of mevalonate (15Whyte D.B. Kirschmeier P. Hockenberry T.N. Nunez-Oliva I. James L. Catino J.J. Bishop W.R. Pai J. J. Biol. Chem. 1997; 272: 14459-14464Abstract Full Text Full Text PDF PubMed Scopus (728) Google Scholar). In addition, for this experiment, antibodies were required to immunoprecipitate endogenous CENP-E and CENP-F/mitosin from the pMev-DLD-1 cells. The CENP-F/mitosin antibody was commercially available. The CENP-E antibody was generated against a peptide from the COOH-terminal region of CENP-E as described under "Experimental Procedures." Western blot analysis performed with the affinity-purified CENP-E antibody revealed a band migrating at ∼250 kDa in HCT 116 (Fig. 2 A) and all other human tumor cell lines that were tested (data not shown) as reported earlier (23Lombillo V.A. Nislow C. Yen T.J. Gelfand V.I. McIntosh J.R. J. Cell Biol. 1995; 128: 107-115Crossref PubMed Scopus (196) Google Scholar). The antibody was shown to be specific for CENP-E because incubation with the peptide blocked immunoreactivity (Fig.2 A). pMev-DLD-1 cells were grown in the presence of 3H-labeled mevalonolactone (precursor for both the farnesyl and the geranylgeranyl pyrophosphate) and mevastatin (which prevents the synthesis of cellular mevalonate) in the presence and in the absence of the FTI. Protein lysates collected from the cells were immunoprecipitated with mitosin antibody and resolved by SDS-PAGE as shown in Fig.2 B. 3H-Labeled mevalonate was incorporated into mitosin, indicating that it is a prenylated protein. However, the incorporation of the label from 3H mevalonolactone was completely blocked in the presence of a farnesyl transferase inhibitor, SCH 66336 (Fig. 2 B), indicating that CENP-F/mitosin is a farnesylated protein and not subject to alternative prenylation. Western analysis confirmed that CENP-F/mitosin was not repressed in FTI-treated cells (Fig. 2 C), suggesting that the loss of signal was due to inhibition of prenylation and not due to a reduction in protein expression. CENP-E was immunoprecipitated using the anti-CENP-E antibody from the protein lysate that had been depleted of mitosin and resolved by SDS-polyacrylamide gel electrophoresis. Like CENP-F, CENP-E was not repressed by FTI treatment (Fig. 2 C), and CENP-E also incorporated 3H-labeled mevalonate only in the absence of FTI (Fig. 2 B). To ensure that 3H-labeled mevalonolate was incorporated in FTI-treated p-Mev-DLD-1, K-Ras, which is alternatively prenylated in the presence of FTIs, was immunoprecipitated from the same lysate. As expected, K-Ras was labeled in both untreated and FTI-treated cells (Fig. 2 B) (15Whyte D.B. Kirschmeier P. Hockenberry T.N. Nunez-Oliva I. James L. Catino J.J. Bishop W.R. Pai J. J. Biol. Chem. 1997; 272: 14459-14464Abstract Full Text Full Text PDF PubMed Scopus (728) Google Scholar). These results confirm that in whole cells, both CENP-E and CENP-F/mitosin are farnesylated proteins, and it is important to note that the presence of the FTI does not lead to the alternative prenylation of either CENP-E or CENP-F/mitosin in cells. The lung carcinoma cells, A549, which harbor a K-ras mutation, were shown to be sensitive to the effects of FTIs (2Nagasu T. Yoshimatsu K. Rowell C. Lewis M.D. Garcia A.M. Cancer Res. 1995; 55: 5310-5314PubMed Google Scholar, 3Sepp-Lorenzino L. Ma Z. Rands E. Kohl N.E. Gibbs J.B. Oliff A. Rosen N. Cancer Res. 1995; 55: 5302-5309PubMed Google Scholar, 46Miquel K. Pradines A. Sun J. Qian Y. Hamilton A.D. Sebti S.M. Favre G. Cancer Res. 1997; 57: 1846-1850PubMed Google Scholar). To evaluate the ability of SCH 66336 to alter the cell cycle distribution, flow cytometry was performed on A549 cells in the presence of SCH 66336. As observed earlier with human tumor cells lacking Ha-ras mutations, A549 cells treated with 1 μmSCH 66336 pause in the G2 → M phase (Fig.3 and Ref. 1Moasser M.M. Sepp-Lorenzino L. Kohl N.E. Oliff A. Balog A. Su D.S. Danishefsky S.J. Rosen N. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 1369-1374Crossref PubMed Scopus (200) Google Scholar). Sebti et al.(46Miquel K. Pradines A. Sun J. Qian Y. Hamilton A.D. Sebti S.M. Favre G. Cancer Res. 1997; 57: 1846-1850PubMed Google Scholar) have also reported a G2 → M pause in the same cells with a structurally distinct FTI. After 60 h of treatment with 1 μm SCH 66336, 25% cells are in G2 → M as compared with 9% in untreated control cells, indicating that in the presence of FTIs the cells are unable to progress through mitosis normally. Immunohistochemistry was performed to determine the precise stage of the mitotic pause. Cells were stained with propidium iodide to identify mitotic cells. Although initially, 10,000 cells were grown on each coverslip, the doubling time of the cells wa