Solid-Phase Synthesis of a Radiolabeled ... - ACS Publications

Nov 15, 2000 - Solid-Phase Synthesis of a Radiolabeled, Biotinylated, and Farnesylated Ca1a2X Peptide Substrate for Ras- and a-Mating Factor Convertin...
0 downloads 0 Views 101KB Size
Download PDF
Bioconjugate Chem. 2001, 12, 35−43

35

Solid-Phase Synthesis of a Radiolabeled, Biotinylated, and Farnesylated Ca1a2X Peptide Substrate for Ras- and a-Mating Factor Converting Enzyme E. Kurt Dolence, Julia M. Dolence, and C. Dale Poulter* Department of Chemistry, University of Utah, Salt Lake City, Utah 84112-0850. Received April 13, 2000; Revised Manuscript Received July 11, 2000

Eukaryotic proteins with carboxyl-terminal Ca1a2 motifs undergo three posttranslational processing reactionssprenylation, endoproteolysis, and carboxymethylation. Two genes in yeast encoding Ca1a2X endoproteases, AFC1 and RCE1, have been identified. Rce1p is solely responsible for proteolysis of yeast Ras proteins. When proteolysis is blocked, localization of Ras2p to the outer membrane is impaired. The mislocalization of undermodified Ras in the cell suggests that Rce1p is an attractive target for cancer therapeutics. A biotinylated, farnesylated Ca 1a2X peptide [(1-N-biotinyl(13-N-succinimidyl-(S-(E,E-farnesyl)- L -cysteinyl)- L -valinyl- L -isoleucinyl- L -alanine))-4,7,10-trioxatridecanediamine] 1 containing a poly(ethylene glycol) linker was prepared by solid-phase synthesis for use in an assay for Ca1a2X endoprotease activity that relies on the strong affinity of avidin for biotin. The peptide was radiolabeled in the penultimate step of the synthesis by cleavage of the biotinylated, farnesylated Ca1a2 precursor from Kaiser’s oxime resin with [14C]-L-alanine methyl ester. [14C]1 was a good substrate for yRce1p with KM ) 1.3 ( 0.3 µM. Analysis of the carboxyl terminal products by reverse phase HPLC confirmed that VIA was the only radioactive fragment released upon incubation of [14C]1 with a yeast membrane preparation of recombinant yRce1p. The solid-phase methodology developed using Kaiser’s benzophenone oxime resin to synthesize [14C]1 should be generally applicable for peptides containing sensitive side chains. In addition, introduction of the radiolabeled unit at the end of the synthesis mostly circumvents problems associated with handling radioactive materials.

INTRODUCTION

Some eukaryotic proteins must be modified posttranslationally in order to perform their relevant biological functions (1-3). The modification of Ca1a2X proteins, where “C” is cysteine, “a” is normally a small aliphatic amino acid, and “X” determines whether the protein is modified with a C15 farnesyl or C20 geranylgeranyl group, is a three-step process involving prenylation, endoproteolytic cleavage to remove the “a1a2X” moiety, and carboxymethylation of the new C-terminal prenylcysteine (Scheme 1). Prenylated Ca1a2X proteins include the Ras and Rho subfamilies of the Ras superfamily of small GTPases, fungal pheromones, enzymes, hepatitis delta coat proteins, nuclear lamins, and GTP-binding proteins (1-3). Several forms of cancer, including those of the pancreas, colon, and lung are related to oncogenic forms of Ras. Farnesylation and proteolysis are required for Ras proteins to correctly localize in the cell where they participate in the signal transduction network for division of normal and transformed cells. Recently, two genes encoding proteins with Ca1a2X prenyl protease activity, a-factor converting enzyme (AFC1) and Ras and a-factor converting enzyme (RCE1), were identified in yeast (4). In vivo studies indicated that the gene product of RCE1, Rce1p, is solely responsible for processing of yeast Ras proteins (4). Interestingly, in the same studies null mutations of RCE1 did not seem * To whom correspondence should be addressed. Phone: (801) 581-6685. Fax: (801) 581-4391. E-mail: poulter@ chemistry.utah.edu.

Scheme 1. Endo-Protease Cleavage of Farnesylated Ca1a2X Peptides

to affect cell viability. Subsequent experiments with RCE1-deficient mice indicated that Ras proteins mislocalized in the cell when endoproteolytic processing was blocked (5). These studies suggest that selective inhibitors of Rce1p are attractive targets for the development

10.1021/bc000036g CCC: $20.00 © 2001 American Chemical Society Published on Web 11/15/2000

36 Bioconjugate Chem., Vol. 12, No. 1, 2001

of potential cancer therapeutics. Until recently, most of the efforts to develop drugs to reduce the activity of oncogenic Ras proteins has focused on the inhibition of protein farnesyltransferase (PFTase). However, PFTase inhibitors may have limited utility against many cancers since some Ras proteins, including oncogenic forms of Ki4B-Ras are geranylgeranylated in a compensatory manner when farnesylation is blocked by inhibition of PFTase (6). Inhibition of Ca1a2X prenyl protease will disrupt the processing of both types of prenylated Ras proteins and could provide another approach to reducing the activity of oncogenic Ras. Ca1a2X prenyl protease (7) activity has been studied in yeast (8, 9), rat (10), and bovine (11) membrane preparations. The rat and bovine enzymes were partially purified (11-13). Prenylated tri-C(farnesyl)a1a2 and tetraC(farnesyl)a1a2X peptides are substrates for proteolysis. Recently, the yeast and human genes for Rce1p were expressed in Sf9 insect cells by infection with a recombinant baculovirus (14). The yeast gene was also expressed in a protease-deficient strain of Saccharomyces cerevisiae. Biochemical and site-directed mutagenesis studies suggest that yRce1p is a cysteine protease (15). A variety of assays have been developed for the enzyme. These include HPLC detection of a radioactive cysteine product, a fluorometric HPLC assay (16), and a coupled proteolysis/methylation assay (8, 9, 14). Liu et al. recently reported an assay that relies on the strong affinity of avidin for biotin (17). We were attracted to this assay since it does not require a chromatographic separation or use of a coupling enzyme, is sensitive, and has potential for development as a high throughput system. The peptide synthesized by Liu et al. consisted of a C(Sfarnesyl)-VI-[3H]S tetrapeptide attached to biotin by a hydrophobic spacer. Their synthesis combined a solidphase assembly of a biotinylated CVI tripeptide with solution phase additions of the farnesyl moiety [3H]serine. HPLC purifications were required to purify material following each solution phase step. We encountered considerable difficulties with insoluble intermediates and low yields when attempting to adapt these protocols to our system and decided to develop a solid-phase approach for the entire synthesis on a resin whose chemistry was compatible with acid sensitive side chains. In addition, the hydrophobic linker was replaced by a PEG-linker to improve overall solubility of the biotinylated substrate used in the coupling reaction. We now describe the solidphase synthesis of a radiolabeled, biotinylated, and farnesylated Ca1a2X peptide and its use to assay for Ca1a2X proteolytic activity in membranes enriched with recombinant yRce1p. EXPERIMENTAL PROCEDURES

Amino acids, commercially available N-BOC protected amino acids, farnesyl bromide, di-tert-butyl dicarbonate, N,N′-dicyclohexylcarbodiimide, and N-hydroxybenzotriazole, and other chemicals were purchased from Aldrich Chemical Co. Polystyrene resin was purchased from BioRad. [14C]-L-Alanine was obtained from NEN DuPont Corp. Biochemical reagents were purchased from Gibco or Sigma. Kaiser’s benzophenone oxime resin was synthesized by the method of DeGrado and Kaiser (18). Avidin resin was purchased from Pierce. THF was dried and distilled from sodium benzophenone. CH2Cl2 was distilled from CaH2. DMF was vacuum distilled from CaH2. Analytical thin-layer chromatography was carried out on either E. Merck or Whatman precoated silica gel

Dolence et al.

60 (0.2 mm, aluminum support) TLC plates. Preparative TLC was carried out using E. Merck silica gel 60 glass backed TLC plates. C18 reverse phase HPLC was carried out using a Vydac Protein and Peptide column with monitoring at 214 nm. Solvents for C18 HPLC were a gradient of acetonitrile and water containing 0.1% v/v TFA. 1H, 13C, and DEPT NMR spectra were obtained using a UNITY Varian 300 MHz NMR and were referenced to either TMS or the residual NMR solvent signal for d6DMSO at 2.50 ppm for 1H or 39.5 ppm for 13C. Optical rotations were obtained in spectral grade solvents. Infrared spectra were obtained as thin films on NaCl plates. FABMS (positive or negative ion) were obtained from the Mass Spectrometry Service Center in the Chemistry Department. MALDI spectra were provided by the Mass Spectral facility in the Department of Medicinal Chemistry. Syntheses. N-tert-Butoxycarbonyl-L-Valinyl-L-Isoleucinyl Kaiser’s Oxime Ester Resin (2). To a solution of N-BOC-L-isoleucine (1.64 g, 7.11 mmol) and HOBt (0.96 g, 7.11 mmol) in 20 mL of anhydrous THF at 0 °C under nitrogen was added dropwise DCC (1.47 g, 7.11 mmol) in 10 mL of anhydrous THF. The mixture was stirred for 3 h while it was allowed to warm to room temperature. Dicyclohexylurea was removed by filtration through a plug of Celite, and the plug was washed with ice-cold THF. Solvent was removed at reduced pressure to give L-isoleucine active ester. The isoleucine active ester was dissolved in 100 mL of anhydrous CH2Cl2 and added to 10.0 g of Kaiser’s resin. Anhydrous triethylamine (0.91 mL, 6.52 mmol) was added, and the mixture rotated under N2 overnight. The resin was collected by suction filtration on a medium porosity sintered glass funnel and washed with the following standard protocol: (1) four 100 mL portions of CH2Cl2; (2) four 100 mL portions of DMF; (3) four 100 mL portions of 2-propanol; and (4) four 100 mL portions of CH2Cl2. The resin was dried overnight under high vacuum. N-BOC-L-valine (1.54 g, 7.11 mmol) was converted to the HOBt active ester as described previously for N-BOCL-isoleucine. The active ester was isolated as described above. N-BOC-L-isoleucinyl Kaiser’s oxime resin was added to 160 mL of anhydrous CH2Cl2. TFA (40 mL) was added and the suspension rotated for 1 h under nitrogen. The resin was collection by suction on a medium porosity sintered glass funnel, washed with four 100 mL portions of CH2Cl2 and added to a solution of L-valine active ester in 100 mL of anhydrous CH2Cl2. Anhydrous triethylamine (2.0 mL, 14.2 mmol) was added, and the suspension was rotated overnight under N2. The resin was collected by suction on a sintered glass funnel and washed with the four solvent protocol as described above. The resin was dried under high vacuum overnight to afford 11.32 g of a tan solid. (S-(E,E-Farnesyl)-L-cysteinyl)-L-valinyl-L-isoleucinyl Kaiser’s Oxime Ester Resin Trifluoroacetate Salt (3). A suspension of N-(triphenylmethyl)-(S-(E,E-farnesyl)-Lcysteine) (4.48 g, 7.89 mmol) (19), HOBt (1.07 g, 7.89 mmol), and 30 mL of anhydrous THF was cooled to 0 °C under N2. DCC (1.63 g, 7.89 mmol) in 30 mL of anhydrous THF was slowly added dropwise. The reaction mixture was stirred for 24 h while slowly warming to room temperature. Dicyclohexlyurea was removed by filtration through a plug of Celite, and the Celite was washed with 25 mL of ice-cold THF. Solvent was removed in vacuo to afford the N-trityl active ester as a thick oil.

Substrate for Ras- and a-Factor Converting Enzymes

A slurry of 10.10 g of N-BOC-L-valinyl-L-isoleucinyl Kaiser’s oxime ester resin, 100 mL of anhydrous CH2Cl2 and 25 mL of TFA were rotated under N2 for 1.5 h. The resin was collected by suction filtration and washed with four 100 mL portions of CH2Cl2. The deprotected resin was added to a solution of the N-trityl active ester in 100 mL of anhydrous CH2Cl2 and anhydrous triethylamine (2.2 mL, 15.78 mmol). The suspension was rotated under nitrogen for 24 h. The resin was collected on a sintered glass funnel, washed with the four solvent wash protocol, and dried overnight under high vacuum to afford 10.24 g of a tan solid. The N-trityl Kaiser’s oxime ester resin (9.45 g) was suspended in 100 mL of anhydrous CH2Cl2. Ethylmethyl sulfide (5.52 mL, 61 mmol) was added by syringe to scavenge advantageous cations, followed by 1.18 mL (15.3 mmol) of TFA. The resin was rotated for 30 min, collected by suction, washed by mixing with four 125 mL portions of CH2Cl2, and dried in vacuo for 15 min to give the TFA salt of Kaiser’s resin 3. N-(13-Amino-4,7,10-trioxatridecanyl)biotinamide (4) was synthesized by the method of Wilbur et al. (20). (N-(13-N-Succinimidyl-4,7,10-trioxatridecanyl)biotinamide-(2,3,5,6-tetrafluorophenyl Active Ester) (5). To a solution of 1.0 g (2.24 mmol) of 4 and 0.25 g (2.46 mmol) of succinic anhydride in 20 mL of anhydrous DMF under N2, 0.35 mL (2.46 mmol) of anhydrous triethylamine was added by syringe. The reaction was stirred for 16 h before the DMF was removed in vacuo at 40 °C. The residue was dissolved with vigorous stirring in 25 mL of 1 M acetic acid and 25 mL of n-butanol. The mixture was extracted with two 25 mL portions of n-butanol, and the pooled organic phases were evaporated in vacuo. The residue was repeatedly evaporated from 25 mL portions of CHCl3 to afford a white solid. The solid was purified by silica column chromatography upon elution with 1:5 MeOH:CHCl3, followed by 1:5 MeOH:CHCl3 containing 5% v/v acetic acid. Fractions that were homogeneous by silica TLC (1:5 MeOH:CHCl3 containing 5% v/v acetic acid) were pooled and concentrated in vacuo to afford 1.01 g (82%) of a white solid; 1H NMR (300 MHz, DMSO) δ 1.20-1.34 (m, 2 H), 1.40-1.68 (m, 6 H), 2.04 (t, 2 H, biotin CH2CO), 2.28 (t, 2 H, succinimidyl CH2), 2.40 (t, 2 H, succinimidyl CH2), 2.55 (d, 1 H), 2.73-2.89 (m, 1 H), 3.03-3.12 (m, 5 H), 3.38 (t, 4 H), 3.42-3.52 (m, 8 H), 4.09-4.15 (m, 1 H), 4.28-4.32 (m, 1 H), 6.38 (s, 1 H, biotin NH), 6.46 (s, 1 H, biotin NH), 7.76 (t, 1 H, NH), 7.82 (t, 1 H, NH); 13C NMR (75 MHz, DMSO) δ 173.96, 171.94, 170.87, 162.78, 69.79, 69.56, 68.12, 68.06, 61.07, 59.08, 59.22, 55.47, 39.89, 35.82, 35.74, 35.23, 30.04, 29.43, 29.37, 29.24, 28.26, 28.07, 25.34 ppm. Positive ion FABMS: calcd for C24H43N4O8S (M + H) 547; found (M + H) m/z 547. To a solution of 1.67 g (3.05 mmol) of N-(13-Nsuccinimidyl-4,7,10-trioxatridecanyl)biotinamide and 0.86 mL (6.11 mmol) of anhydrous triethylamine in 32 mL of anhydrous DMF was added dropwise a solution of 1.20 g (4.58 mmol) of 2,3,5,6-tetrafluorophenyltrifluoroacetate (21) in 5 mL of anhydrous DMF. The reaction was stirred for 1 h under N2. DMF was removed in vacuo with a water bath at e50 °C to afford a pale yellow oil. Diethyl ether (200 mL) was added with stirring under N2. Slowly, over 30 min a white solid formed on the walls of the flask. The ether was removed by suction using a pipet, and a second 200 mL portion of diethyl ether was added. The suspension was stirred for 30 min, and the ether was removed as before. The white solid was dried under high vacuum for 30 min and used immediately.

Bioconjugate Chem., Vol. 12, No. 1, 2001 37

(1-N-Biotinyl-(13-N-(succinimidyl-(S-(E,E-farnesyl)-Lcysteinyl)-L-valinyl-L-isoleucinyl)-4,7,10-trioxatridecanediamine) Kaiser’s Resin Oxime Ester (6). To resin 3 (9.45 g) and newly synthesized tetrafluorophenyl active ester of N-(13-N-succinimidyl-4,7,10-trioxatridecanyl)biotinamide 5 was added 60 mL of DMF and 60 mL of CH2Cl2, followed by dropwise addition of 1.75 mL (12.51 mmol) of anhydrous triethylamine. The suspension was rotated under N2 for 48 h. The slurry was collected on a sintered glass funnel and washed with the four solvent protocol described previously. The resin was dried under high vacuum for 18 h to afford 9.79 g of a tan solid. (1-N-Biotinyl-(13-N-succinimidyl-(S-E,E-farnesyl)- Lcysteinyl)-L-valinyl-L-isoleucinyl-L-alaninyl))-4,7,10-trioxatridecanediamine) Benzyl Ester (8). To a suspension of 0.50 g of (1-N-biotinyl-(13-N-(succinimidyl-(S(E,E-farnesyl)-L-cysteinyl)-L-valinyl-L-isoleucinyl)-4,7,10trioxatridecanediamine), Kaiser’s resin 6, were added 0.33 g (0.97 mmol) of L-alanine benzyl ester tosylate and 5 mL of anhydrous CH2Cl2. Glacial acetic acid (56 µL, 0.97 mmol) and anhydrous triethylamine (136 µL, 0.97 mmol) were added. The flask was rotated under N2 for 24 h. The resin slurry was poured into a sintered glass funnel that contained 20 mL of 2-propanol and was washed with five 40 mL portions of CHCl3 and 2-propanol. The CHCl3/2-propanol filtrate was concentrated in vacuo and redissolved in 45 mL of CHCl3. The solution was washed with 15 mL of saturated NaHCO3 solution and two 15 mL portions of 1 M acetic acid. The organic layer was dried over anhydrous MgSO4, and the solvent was removed at reduced pressure. The last traces of acetic acid were removed by high vacuum overnight. The residue (0.14 g) was dissolved in CHCl3 and applied to two preparative silica gel thin-layer chromatography plates. The plates were eluted with 1:10 MeOH: CHCl3. The UV active band at Rf ) 0.24 was removed by scraping and eluted from the silica with a total of 200 mL of 1:5 MeOH:CHCl3. Solvent was removed in vacuo to afford 8 (0.11 g, 26% overall) as a white solid; HPLC analysis was performed on a Vydac C18 reverse phase column with detection at 214 nm and elution with acetonitrile:water containing 0.1% TFA. The gradient profile was 45% acetonitrile:55% water to 80% acetonitrile:20% water over 20 min. The retention time of 8 under these conditions was 14.82 min. 1H NMR (300 MHz, DMSO) δ 0.70-0.82 (m, 12 H, valine and isoleucine CH3), 0.97-1.09 (m, 2 H), 1.27 (d, 3 H, J ) 7.2 Hz, alanine CH3), 1.30-1.70 (m, 10 H), 1.54 (s, 6 H, farnesyl CH3), 1.60 (s, 3 H, farnesyl CH3), 1.62 (s, 3 H, farnesyl CH3), 1.86-2.10 (m, 10 H, biotin CH2CO and farnesyl allylic CH2), 2.24-2.40 (m, 4 H, succinimidyl CH2), 2.56 (d, 2 H, J ) 12.3 Hz, CH2S), 2.70-2.85 (m, 2 H, SCH2), 3.00-3.18 (m, 7 H), 3.37 (t, 4 H, J ) 6.3 Hz, PEG CH2), 3.42-3.52 (m, 8 H, PEG CH2), 4.09-4.22 (m, 3 H, cysteine, valine, and isoleucine CH), 4.25-4.34 (m, 2 H, biotin CH), 4.45 (q, 1 H, J ) 7.8 and 8.1 Hz, alanine CH), 5.00-5.10 (m, 4 H, farnesyl olefinic and benzylic CH2), 5.15 (t, 1 H, J ) 7.8 and 8.1 Hz, farnesyl olefinic), 6.47 (br s, 2 H, biotin NH), 7.33 (s, 5 H, aromatic H), 7.71 (d, 1 H, J ) 8.7 Hz, NH), 7.77 (t, 1 H, J ) 5.4 Hz, PEG NH), 7.86 (t, 1 H, J ) 5.4 Hz, PEG NH), 7.89 (d, 1 H, J ) 9.0 Hz, NH), 8.21 (d, 1 H, J ) 8.1 Hz, NH), 8.38 (d, 1 H, J ) 6.6 Hz, NH); 13C NMR (75 MHz, DMSO) δ 172.20, 171.92, 171.64, 171.20, 170.65, 170.42, 170.38, 162.76, 138.30, 135.95, 134.55, 130.65, 128.38, 127.99, 127.75, 124.13, 123.68, 120.27, 69.79, 69.57, 68.12, 65.85, 61.07, 59.23, 57.83, 56.36, 55.47, 52.55, 47.66, 39.88,

38 Bioconjugate Chem., Vol. 12, No. 1, 2001

39.12, 36.69, 35.85, 35.72, 35.22, 32.69, 30.82, 30.40, 29.43, 29.36, 28.96, 28.25, 28.06, 26.20, 25.93, 25.51, 25.34, 24.18, 19.18, 18.09, 17.56, 16.81, 15.80, 15.07, 10.94 ppm. Positive ion FABMS: calcd for C65H103N8O12S2 (M + H) 1228; found (M + H) m/z 1228. 1-N-Biotinyl-(13-N-succinimidyl-(S-(E,E-farnesyl)- Lcysteinyl)-L-valinyl-L-isoleucinyl-L-alanine))-4,7,10-trioxatridecanediamine (1). To 0.11 g (0.088 mmol) of benzyl ester 8 in 2.2 mL of dry THF was added 0.45 mL (0.110 mmol) of 0.25 M KOH. The mixture was stirred under N2 for 24 h. TLC analysis on silica gel (1:10 MeOH: CHCl3) showed the presence of a small amount of starting benzyl ester. A 0.17 mL (0.042 mmol) portion of 0.25 M KOH was added, and the mixture stirred for 4 h. TLC analysis showed no benzyl ester. Solvent was removed at reduced pressure and distilled deionized water (10 mL) was added. The solution was extracted with 40 mL of diethyl ether; the aqueous layer was acidified with 20 mL of 1 M acetic acid, and the resulting milky suspension was extracted with three 20 mL portions of water saturated n-butanol. The extracts were combined and the n-butanol was removed in vacuo with a water bath at e40 °C. The residue was dissolved in 1:5 MeOH:CHCl3 and transferred to a tared flask. Solvent was removed in vacuo to give 0.092 g of a white solid. HPLC analysis as described above gave a retention time of 10.20 min for 1. The material was purified by silica column chromatography upon elution with 1:5 MeOH:CHCl3 containing 5% v/v acetic acid. Homogeneous fractions were pooled, and the solvent removed in vacuo to give 0.072 g (72%) of 1 as a white solid, [R]D20 ) -3.80 (1:5 MeOH: CHCl3 containing 5.0% v/v glacial acetic acid, c ) 5.0); IR TF (NaCl) 3353, 3211, 3082, 2961, 2930, 2869, 1707, 1633, 1549, 1454, 1385, 1222, 1115, 755, 709 cm-1; TLC, Rf ) 0.57 using 1:5 MeOH:CHCl3 with 5.0% v/v glacial acetic acid on silica; 1H NMR (300 MHz, DMSO) δ 0.700.84 (m, 12 H, valine and isoleucine CH3), 0.94-1.12 (m, 2 H), 1.16-1.30 (m, 3 H, alanine CH3), 1.32-1.74 (m, 10 H), 1.51 (s, 6 H, farnesyl CH3), 1.57 (s, 3 H, farnesyl CH3), 1.59 (s, 3 H, farnesyl CH3), 1.84-2.06 (m, 10 H, biotin CH2CO and farnesyl allylic CH2), 2.20-2.40 (m, 4 H, succinimidyl CH2), 2.53 (d, 2 H, J ) 12.3 Hz, CH2S), 2.66-2.84 (m, 2 H, S CH2), 2.94-3.16 (m, 7 H), 3.34 (t, 4 H, J ) 6.3 Hz, PEG CH2), 3.38-3.52 (m, 8 H, PEG CH2), 4.00-4.20 (m, 4 H, cysteine, valine, isoleucine CH, and biotin CH), 4.25-4.34 (m, 1 H, biotin CH), 4.41 (q, 1 H, J ) 7.5 and 8.1 Hz, alanine CH), 4.96-5.08 (m, 2 H, farnesyl olefinic), 5.12 (t, 1 H, J ) 7.5 Hz, farnesyl olefinic), 6.35 (s, 1 H, biotin NH), 6.43 (s, 1 H, biotin NH), 7.60-7.96 (m, 4 H, NH), 8.00-8.10 (m, 1 H, NH), 8.17 (d, 1 H, J ) 7.5 Hz, NH); 13C NMR (75 MHz, DMSO) δ 173.92 (carbonyl), 171.90 (carbonyl), 171.60 (carbonyl), 171.18 (carbonyl), 170.40 (carbonyl), 162.76 (urea carbonyl), 138.30 (quaternary), 134.55 (quaternary), 130.65 (quaternary), 124.13 (CH), 123.68 (CH), 120.26 (CH), 69.78 (CH2), 69.56 (CH2), 68.10 (CH2), 61.06 (CH), 59.21 (CH), 57.81 (CH), 56.48 (CH), 55.46 (CH), 52.53 (CH), 47.60 (CH), 39.88 (CH2), 39.22 (CH2), 39.12 (CH2), 36.67 (CH), 35.84 (CH2), 35.72 (CH2), 35.22 (CH2), 32.70 (CH2), 30.81 (CH2), 30.41 (CH), 29.42 (CH2), 29.35 (CH2), 28.94 (CH2), 28.24 (CH2), 28.05 (CH2), 26.19 (CH2), 25.93 (CH2), 25.52 (CH3), 25.33 (CH2), 24.16 (CH2), 19.20 (CH3), 18.08 (CH3), 17.57 (CH3), 17.18 (CH3), 15.80 (CH3), 15.21 (CH3), 11.01 (CH3) ppm; negative FABMS (glycerol/25 mM NH4HCO3): calcd m/z 1135.3 (M - H); found m/z 1135.4 (M - H). The derived isotopic pattern for the M-H signal was found to be identical to the theoretical spectrum: 1135 (100.00%), 1136 (66.87%), 1137 (37.69%), 1138 (16.71%), 1139 (7.87%); MALDI mass spectroscopy: calcd

Dolence et al.

for C56H97N8O12S2 (M + H) 1137.6667, found 1137.6810 (4%); calcd for (M + Na) 1159.6487, found 1159.6715 (59%); calcd for (M + K) 1175.6226, found 1175.6484 (37%). Synthesis of [14C]1. A solution of 116 µL (3 µmol, 5 mg/ mL) p-toluenesulfonic acid in anhydrous THF was evaporated to dryness before 5.0 mL of [14C]-L-alanine (2.9 µmol, 0.5 mCi, 174.6 mCi/mmol) in 2:98 ethanol:water was added by syringe. Two 1.0 mL portions of 2:98 ethanol:water were used to rinse the syringe. The solution was flash frozen in a dry-ice/acetone bath and lyophilized to give [14C]-L-alanine tosylate. Spectral grade methanol (250 µL) was added to the residue followed by 30 µL of 2 M thionyl chloride in anhydrous CH2Cl2. The mixture was stirred 4 h at which time 250 µL of spectral grade methanol and 30 µL of thionyl chloride/CH2Cl2 solution were added. The mixture was stirred for 18 h before addition of third portion of 30 µL of thionyl chloride/CH2Cl2 solution. The reaction was stirred for 2 h before the solvent was removed with a stream of argon. The flask was placed under high vacuum for 15 min. To [14C]-L-alanine methyl ester tosylate and 0.020 g of (1-N-biotinyl-(13-N-succinimidyl-(S-(E,E-farnesyl)-L-cysteinyl)-L-valinyl-L-isoleucinyl)-4,7,10-trioxatridecanediamine) Kaiser’s resin 6 were added a stock solution of glacial acetic acid (5.8 µmol) in CH2Cl2 (200 µL) and triethylamine (5.8 µmol) in CH2Cl2 (200 µL). The slurry was vigorously stirred for 96 h. The suspension was filtered through a Pasteur pipet/cotton plug. The resin was washed with four 1.0 mL portions of CHCl3, three 1.0 mL portions of 2-propanol, and five 1.0 mL portions of 1:5 MeOH:CHCl3. Solvent was removed from the filtrate with a gentle stream of argon. The residue was dissolved in a minimal amount of 1:5 MeOH:CHCl3 and applied to an analytical Merck glassbacked silica gel 60 thin-layer chromatography plate. The plate was eluted with 1:5 MeOH:CHCl3 containing 5% v/v acetic acid. A faint UV-positive band of Rf value 0.680.80 was carefully removed by scraping, and the silica was placed in a Pasteur pipet plugged with cotton. The silica was eluted with twelve 1.0 mL portions of 1:5 MeOH:CHCl3 containing 0.5% v/v acetic acid. Solvent was removed with a gentle stream of argon followed by high vacuum to remove final traces of acetic acid. To the residue were added 200 µL of THF and 200 µL of 50 mM aqueous KOH. The progress of the reaction was monitored by silica TLC (1:5 MeOH:CHCl3 containing 5% v/v acetic acid). Two hundred microliter portions of 50 mM aqueous KOH were added every 15 min until TLC analysis indicated the starting methyl ester had been saponified. Acetic acid (50 µL, 1 M) was added. Solvent was removed with a gentle stream of argon, followed by high vacuum to removed traces of acetic acid and water. DMF (500 µL) was added, and the solution was divided into portions, which were stored at -80 °C. The overall radiochemical yield was 45% from [14C]-Lalanine. Growth of JDY101/pJel-RyRcemyc (15). Following the procedure of Worland and Wang (22), single colonies from freshly streaked plates were used to inoculate 10 mL cultures of synthetic minimal media lacking uracil and supplemented with 2% (w/v) glucose to repress PGAL1. Cultures were grown to late log phase, and a 3 mL portion was used to inoculate 300 mL of synthetic minimal media lacking uracil and supplemented with 3% (w/v) glycerol and 2% (w/v) lactic acid. The cultures were grown at 30 °C for 20-24 h (OD600 ) 0.5-0.7). Galactose was added to a final concentration of 2%, and the cells

Substrate for Ras- and a-Factor Converting Enzymes

were grown for an additional 6 h. Cells were pelleted by centrifugation at 5000g for 15 min, washed with cold sterile water, and stored at -80 °C. Preparation of Membrane Fractions. Alkaline carbonate-leached membranes (AC-P139) were prepared according to the procedure of Ashby and Rine (9), with slight modifications. Briefly, 4 mL of frozen cell paste was thawed on ice and resuspended in 8 mL of SST buffer (0.3 M sorbitol, 0.1 M NaCl, 10 mM Tris-HCl, pH 7.4, 5 mM EDTA, pH 8, 1 mM PMSF, 0.5 mM 1,10-phenanthroline). Acid washed glass beads (212-300 µm) were added (8 mL), and the suspension chilled on ice for 5 min prior to vortexing at maximum setting for 1 min and then chilled on ice for 1 min. The vortex-chilling cycle was repeated 10 times. The cellular lysate was clarified by centrifugation at 3500g for 15 min at 4 °C. The centrifugation was repeated for the supernatant. The final 3500g supernatant was transferred to ultracentrifuge tubes, mixed with 1/10 volume of 2 M Na2CO3, pH 11.5, and spun at 139000g (40 000 rpm, TY65 rotor) for 1 h at 2 °C. The P139 pellet was gently washed twice with cold SST buffer (8 mL), resuspended in 8 mL of cold SST buffer with a Dounce homogenizer. The sample was spun again at 139000g for 1 h at 2 °C. The pellet was washed as before and was resuspended in 2 mL of cold SST buffer. The AC-P139 membrane preparation was divided into portions, quick frozen in dry-ice/acetone, and stored at -80 °C. Protease Assays. Biotin/Avidin-Binding Assay. Assays (50 µL) contained AC-P139 membranes (1-5 µg) and 1-N-biotinyl-(13-N-succinimidyl-(S-(E,E-farnesyl)L-cysteinyl)-L-valinyl-L-isoleucinyl-L-alanine)-4,7,10-trioxatridecanediamine ([14C]1) in 100 mM Tris, pH 7.4, 0.5 mM 1,10-phenanthroline, and 1 mM PMSF. Reactions were initiated by the addition of [14C]1 in 2.5 µL of DMF. After incubation for 15-30 min at 37 °C, the enzyme was inactivated by heating at 80 °C for 5 min. The incubation mixtures were briefly cooled on ice, avidin resin (150 µL, Pierce) was added, and the material was allowed to stand at room temperature for 15 min with frequent mixing. Assay buffer (150 µL) was added, the tubes were centrifuged for 1 min, and radioactivity in 200 µL of supernatant was measured. Radioactive HPLC Assay to Confirm Endo-Protease Activity. Assays (50 µL) contained AC-P139 membranes (1-100 µg) and [14C]1 in 100 mM Tris, pH 7.4, 0.5 mM 1,10-phenanthroline, and 1 mM PMSF. Reactions were initiated by the addition of [14C]1. Incubation proceeded for 15-30 min at 37 °C followed by termination in boiling water for 5 min. Assay tubes were briefly centrifuged, and the supernatants were transferred to fresh tubes. Avidin resin (150 µL) was added and the tubes were allowed to stand at room temperature for 15 min with frequent mixing. Assay buffer (150 µL) containing a mixture of cold A, I-A, and V-I-A was added, and the tubes were centrifuged for 1 min. A 200 µL portion of the peptide/reaction mixture was transferred to fresh tubes and an additional 150 µL of the mixture containing cold peptide was added. These samples were frozen at -80 °C until HPLC analysis. Samples were analyzed on a 4.6 × 250 mm Vydac reversed phase C18 column with isocratic elution using 3.5% acetonitrile/0.06% TFA. After the first min, 1 min fractions were collected, added to 10 mL of scintillation cocktail, and counted. RESULTS AND DISCUSSION

Synthesis of Nonradioactive and Radiolabeled Biotinylated Farnesylated Tetrapeptide 1. Biotin-

Bioconjugate Chem., Vol. 12, No. 1, 2001 39 Scheme 2. Resina

Synthesis of S-Farnesylated Kaiser’s

a

Reagents: (a) TFA in CH2Cl2, 1.5 h; (b) N-trityl-(S-farnesyl)THF, DCC, HOBt, 0 °C; (c) Et3N, CH2Cl2; (d) wash protocol, dried in vacuo; (e) 2.5 equiv of TFA, 10 equiv of EtSMe, CH2Cl2. L-cysteine,

ylated 1 was synthesized using solid-phase methods based on Kaiser’s oxime resin (18) as outlined in Schemes 2, 3, and 4. This approach allowed us to take advantage of solid-phase chemistry to assemble the entire biotinylated, farnesylated, radiolabeled tetrapeptide and to introduce the radioactive unit near the end of the synthetic sequence. The reactions needed to assemble and cleave the peptide from the resin are compatible with the sensitive isoprenoid group attached to the cysteine residue. Kaiser’s benzophenone oxime resin was synthesized on a 100 g scale using the method of DeGrado and Kaiser (18). N-BOC-L-valinyl-L-isoleucinyl resin 2 was produced by a standard sequence using HOBt/DCC activation and TFA mediated N-BOC deprotection steps. As outlined in Scheme 2, N-BOC-L-valinyl-L-isoleucinyl resin 2 was deprotected using TFA and then allowed to react with the HOBt-derived active ester of N-triphenylmethyl-S(E,E-farnesyl)-L-cysteine. A standard wash protocol of four 100 mL portions of CH2Cl2, four 100 mL portions of DMF, four 100 mL portions of 2-propanol, and finally four 100 mL portions of CH2Cl2 was used after all peptide or linker coupling steps. After drying in vacuo, the N-trityl resin was suspended in CH2Cl2 containing ethylmethyl sulfide as a cation scavenger and deprotected with TFA. The TFA salt of the resin 3 was used immediately in the next step. Attempts using solution phase chemistry to attach biotin to farnesylated cysteine by an L-alanine linker were unsatisfactory. Although coupling between biotinL-alanine and S-(E,E-farnesyl)-L-cysteine esters appeared to proceed as judged by TLC, the product formed an intractable gel in all solvents we studied, including DMSO-d6, typically used for NMR spectra. Lui et al. (17) had reported a similar synthesis; however, in their case the cysteine was not farnesylated, and the isoprenoid moiety was added in an additional step near the end of the synthesis. To avoid problems with solubility, a PEG linker (20) was used to tether biotin to the tetrapeptide (Scheme 3). The biotinylated-PEG linker (4) was synthesized as described by Wilbur et al. (20). The free amine of 4 was converted to the succinate half-amide, and the free carboxylate group was activated using 2,3,5,6tetrafluorophenyltrifluoroacetate (21). This activating agent was particularly useful, since the biotinylated

40 Bioconjugate Chem., Vol. 12, No. 1, 2001 Scheme 3. Synthesis of Biotinylated Linkera

a Reagents: (a) succinic anhydride, DMF, Et N; (b) 2,3,5,63 tetrafluorophenyltrifluoroacetate, DMF, Et3N.

active ester could easily be purified by precipitation using diethyl ether to remove excess activating agent and triethylamine salts. Active ester 5 was isolated by ether precipitation, dried in vacuo, and used immediately in the next step. The final reactions for syntheses of both the nonradioactive and radiolabeled forms of the biotinylated sub-

Dolence et al.

strate 1 are shown in Scheme 4. The TFA salt of the farnesylated tripeptide resin 3 and the biotinylated tetrafluorophenyl active ester 5 were successfully coupled to give the biotinylated farnesylated Kaiser’s oxime ester resin 6. Resin 6 was used to synthesize both the nonradioactive substrate 1 and radioactive [14C]1. The final peptides were cleaved from the resin as also outlined in Scheme 4. The coupling procedures for radioactive and nonradioactive peptides differed slightly with respect to the type of alanine ester used. Initially, L-alanine benzyl ester was used for the nonradioactive synthesis, to take advantage of the UV-active benzyl group during purification. Accordingly, biotinylated farnesylated Kaiser’s oxime ester resin 6 was treated with L-alanine benzyl ester tosylate 7 (R ) Bz) to give 8 (R ) Bz) in 26% yield overall from Kaiser’s oxime resin. Saponification of 8 (R ) Bz) with aqueous KOH in THF gave nonradioactive 1 in 72% yield. Methyl ester [14C]7 (R ) CH3) was used to synthesize [14C]1. Initial attempts to esterify L-alanine as a benzyl ester on a scale compatible with the synthesis of radioisotopes gave unsatisfactory yields. Accordingly, [14C]-Lalanine was esterified using p-toluenesulfonic acid and thionyl chloride in methanol to afford the tosylate salt of [14C]7 (R ) CH3). [14C]7 (R ) CH3) was coupled to

Scheme 4. Synthesis, Radiosubstrate Cleavage, and Saponificationa

a Reagents: (a) 1:1 DMF:CH Cl , Et N; (b) 1.05 equiv of p-TsOH, (c) (1) BzOH, PhH, reflux; (2) MeOH, SOCl ; (d) HOAc, Et N, 2 2 3 2 3 CH2Cl2; (e) KOH, THF.

Substrate for Ras- and a-Factor Converting Enzymes

Figure 1. Ca1a2X endoprotease activity increases with increasing amounts of alkaline carbonate leached membranes. [14C]1 (25 µM) was incubated with yRce1p membranes (0.1-1.0 µg of protein) at 37 °C for 15 min. The reaction was stopped and the mixture was treated with avidin resin as described in the Experimental Section.

biotinylated farnesylated 6 using conditions similar to those for unlabeled material, followed by aqueous workup, to afford [14C]8 (R ) CH3). The tetrapeptide was purified easily by analytical TLC to remove any unreacted radioactive methyl ester in order to minimize any radioactive contaminates that would contribute to a higher background in the enzymatic assay. The methyl ester was saponified to afford [14C]1 in an overall radiochemical yield of 45% for the coupling and hydrolysis steps. This material comigrated with unlabeled 1 on reverse-phase HPLC. Endoprotease Assays. yRce1 activity was measured by release of the radioactive C-terminal tripeptide from [14C]1 using biotin/avidin binding to separate the radioactive product and substrate. The assay involved incubation of [14C]1 with yRce1p membranes (15) followed by heat inactivation. Unreacted substrate was separated from the VIA fragment by incubation of the reaction mixture with avidin resin at room temperature. The resin was pelleted by centrifugation, and the radioactivity of VI[14C]A in the supernatant was measured by liquid scintillation counting. A proportional increase in protease activity was observed upon incubation of a saturating concentration of [14C]1 with increasing amounts of membranes (Figure 1). At saturating substrate concentrations (>25 µM), it was necessary to use 150 µL of avidin resin in order to achieve 10:1 signal-to-noise. Incubation of the assay mixture with avidin resin for 15 min with frequent mixing was sufficient to remove radioactive substrate. Steady-state kinetic constants were determined for membrane-bound yRce1p. A plot of initial velocity versus [14C]1 is shown in Figure 2. The experimental data were fit to the Michaelis-Menten equation to give KM ) 1.3 ( 0.3 µM and VMax ) 46.1 ( 2.7 nmol min-1 mg-1. Addition of biotin and the PEG linker to the farnesylated tetrapeptide apparently had a minimal impact on the ability of yRce1p to hydrolyze the substrate. A KM of 2.8 µM was obtained with membranes from native yeast using the peptide substrate KWDPA(farnesyl)CV[4,5-3H]IA (9). In addition, a non-hydrolyzable inhibitor of mammalian Ca1a2X endoproteases reported by Ma et al. (23), inhibited membrane-bound recombinant yRce1p with an IC50 ) 103 ( 9 nM (data not shown). Although alanine and the IA dipeptide are potential radioactive products from [14C]1 by nonspecific proteolysis by Rce1p or contaminating proteases in the crude membrane preparation, analysis of the reaction mixture by

Bioconjugate Chem., Vol. 12, No. 1, 2001 41

Figure 2. A plot of the initial velocity for proteolysis versus [14C]1 concentration. yRce1p membranes (0.29 µg) were incubated with [14C]1 for 15-30 min at 37 °C. VI[14C]A was separated from unreacted substrate with avidin resin and counted as described in the Experimental Section.

Figure 3. Reversed-phase HPLC analysis of the products from farnesylated peptide, [14C]1 upon incubation with yRce1p. Yeast membranes (5-20 µg of protein) were incubated with 50 µM [14C]1 for 45 min at 37 °C. The reactions were stopped by boiling for 5 min and were treated with avidin resin as described in the Experimental Section. Cold A, IA, and VIA were added to the nonbiotinylated products. The mixture was analyzed by HPLC on a C18 reversed-phase column as described in the Experimental section. Fractions were collected in 1 min intervals. Detection of the cold standards was monitored at 214 nm.

reversed phase HPLC confirmed that VI[14C]A was the only radioactive product. After incubating [14C]1 with Rce1p under standard assay conditions, the avidintreated supernatant was mixed with cold A, IA, and VIA, and the mixture was analyzed by HPLC to give the results shown in Figure 3. The large peak of radioactivity coeluted with VIA, and no significant radioactivity above background was detected with the fractions correspond-

42 Bioconjugate Chem., Vol. 12, No. 1, 2001

ing to alanine (Figure 3A). In contrast, prolonged incubation with membranes prepared from a yeast strain deficient in Rce1p activity (15) showed no radioactivity coeluting with VIA and only a small amount of radioactivity coeluting with alanine. A similar profile was seen when no membranes were added. The peak coeluting with VIA decreased when the RPI inhibitor was added to the assay (Figure 3B). In contrast, membranes isolated from rat liver produced tripeptide VI-[3H]S, dipeptide I-[3H]S, and, in some cases, [3H]S (17). The smaller radioactive fragments, released from substrate and the tripeptide product by contaminating protease activities, limited the utility of the assay to precisely measure the activity of the Ras and a-mating factor proteases. Our recombinant membrane preparations were essentially free of activities that compete with yRce1p for the substrate. [14C]1 and recombinant yRce1p were recently used to assess the sensitivity of the enzyme to metal ion chelators, heavy metals, and a variety of protease inhibitors (15) and to screen a library of N-acetyl-S-(E,Efarnesyl) Ca1a2X tetrapeptides (24). Spacers separate biotin from bulky groups elsewhere in the molecule that can interfere with its binding avidin and streptavidin (25) and help maintain strong biotin/ avidin binding (20). We used a commercially available water soluble spacer, 4,7,10-trioxa-1,13-tridecanediamine and a succinic diamide unit to link biotin to the farnesylated Ca1a2X peptide. The PEG spacer improved the solubility of the biotinylated prenyl Ca1a2X peptide in common organic solvents and greatly facilitated its synthesis and ease of handling. The tetrafluorophenyl active ester used to couple the biotinylated-PEG linker to the resin bound tripeptide proved to be ideal. The ester was easy to prepare, easy to purify by ether precipitation, and coupled smoothly to resin-bound peptide in good yield. A completely solid-phase method based on Kaiser’s resin provides several advantages over previously reported syntheses of compounds of this kind. Acid-sensitive groups such as the farnesyl and geranylgeranyl moieties found in prenylated proteins can be introduced regiospecifically as modified amino acids (i.e., S-farnesyl cysteine) during the peptide elongation process. The mild conditions used to cleave the peptide from the resin are ideal for preserving acid-sensitive side chains. Since the nucleophile employed to cleave the peptide becomes the N-terminal amino acid in the released peptide, radiolabel can be introduced into the final product at a late stage in the synthesis in a manner that minimizes the handling of radioactive materials. The high specific activity of [14C]1, and corresponding high sensitivity of the assay, has proved to be useful for screening and characterization of inhibitors for Ca1a2X proteases. ACKNOWLEDGMENT

We would like to thank Dr. Lance Steward for helpful discussions and Dr. Elliott Rachlin of the University of Utah, Department of Chemistry, Mass Spectral Facility, for mass spectral determinations. This work was supported by Acacia Biosciences and a University of Utah SEED Grant. LITERATURE CITED (1) Maltese, W. A. (1990) Posttranslational modification of proteins by isoprenoids in mammalian cells. FASEB J. 4, 3319-3328.

Dolence et al. (2) Clarke, S. (1992) Protein Isoprenylation and Methylation at Carboxyl-Terminal Cysteine Residues. Annu. Rev. Biochem. 61, 355-386. (3) Omer, C. A., and Gibbs, J. B. (1994) Protein prenylation in eukaryotic microorganisms: genetics, biology and biochemistry. Mol. Microbiol. 11, 219-225. (4) Boyartchuk, V. L., Ashby, M. N., and Rine, J. (1997) Modulation of Ras and a-Factor Function by CarboxylTerminal Proteolysis. Science 275, 1796-1800. (5) Kim, E., Ambroziak, P., Otto, J. C., Taylor, B., Ashby, M. N., Shannon, K., Casey, P. J., and Young, S. G. (1999) Disruption of the Mouse Rce1 Gene Results in Defective Ras Processing and Mislocalization of Ras within Cells. J. Biol. Chem. 274, 8383-8390. (6) Gibbs, J. B., Graham, S. L., Hartman, G. D., Koblan, K. S., Kohl, N. E., and Omer, C. A. (1997) Farnesyltransferase inhibitors versus Ras inhibitors. Curr. Opin. Chem. Biol. 1, 197-203. (7) Hancock, J. F., Cadwallader, K., and Marshall, C. J. (1991) Methylation and proteolysis are essential for efficient membrane binding of prenylated p21K-ras(B). EMBO J. 10, 641646. (8) Ashby, M. N., King, D. S., and Rine, J. (1992) Endoproteolytic processing of a farnesylated peptide in vivo. Proc. Natl. Acad. Sci. U.S.A. 89, 4613-4617. (9) Ashby, M. N., and Rine, J. (1995) Ras and a-Factor Converting Enzyme. Methods Enzymol. 250, 235-251. (10) Jang, G.-F., Yokoyama, K., and Gelb, M. H. (1993) A prenylated protein-specific endoprotease in rat liver microsomes that produces a carboxyl-terminal tripeptide. Biochemistry 32, 9500-9507. (11) Ma, Y.-T., and Rando, R. R. (1992) A microsomal endoprotease that specifically cleaves isoprenylated peptides. Proc. Natl. Acad. Sci. U.S.A. 89, 6275-6279. (12) Chen, Y., Ma, Y.-T., and Rando, R. R. (1996) Solubilization, Partial Purification, and Affinity Labeling of the MembraneBound Isoprenylated Protein Endoprotease. Biochemistry 35, 3227-3237. (13) Jang, G.-F., and Gelb, M. H. (1998) Substrate Specificity of Mammalian Prenyl Protein-Specific Endoprotease Activity. Biochemistry 37, 4473-4481. (14) Otto, J. C., Kim, E., Young, S. G., and Casey, P. J. (1999) Cloning and characterization of a mammalian prenyl proteinspecific protease. J. Biol. Chem. 274, 8379-8382. (15) Dolence, J. M., Steward, L. E., Dolence, E. K., Wong, D. H., and Poulter, C. D. (2000) Studies with Recombinant Saccharomyces cerevisiae CaaX Prenyl Protease Rce1p. Biochemistry 39, 4096-4104. (16) Nishii, W., Muramatsu, T., Kuchino, Y., Yokoyama, S., and Takahashi, K. (1997) Partial Purification and Characterization of a CAAX-Motif-Specific Protease from Bovine Brain Using a Novel Fluorometric Assay. J. Biochem. 122, 402408. (17) Lui, L., Jang, G.-F., Farnsworth, C., Yokoyama, K., Glomset, J. A., and Gelb, M. H. (1995) Synthetic Prenylated Peptides: Studying Prenyl Protein-Specific Endoprotease and Other Aspects of Protein Prenylation. Methods Enzymol. 250, 189-206. (18) DeGrado, W. F., and Kaiser, E. T. (1980) Polymer-Bound Oxime Esters as Supports for Solid-Phase Peptide Synthesis. Preparation of Protected Peptide Fragments. J. Org. Chem. 45, 1295-1300. (19) Sim, T. B., and Rapoport, H. (1999) N-Trityl- and NPhenylfluorenyl-N-carboxyanhydrides and Their Use in Dipeptide Synthesis. J. Org. Chem. 64, 2532-2536. (20) Wilbur, S., Hamlin, D. K., Vessella, R. L., Stray, J. E., Buhler, K. R., Stayton, P. S., Klumb, L. A., Pathare, P. M., and Weerawarna, S. A. (1996) Antibody Fragments in Tumor Pretargeting. Evaluation of Biotinylated Fab’ Colocalization with Recombinant Streptavidin and Avidin. Bioconjugate Chem. 7, 689-702.

Substrate for Ras- and a-Factor Converting Enzymes (21) Gamper, H. B., Reed, M. W., Cox, T., Virosco, J. S., Adams, A. D., Gall, A. A., Scholler, J. K., and Meyer, R. B. (1993) Facile preparation of nuclease resistant 3′ modified oligodeoxynucleotides. Nucleic Acids Res. 21, 145-150. (22) Worland, S. T., and Wang, J. C. (1989) Inducible overexpression, purification, and active site mapping of DNA topoisomerase II from the yeast Saccharomyces cerevisiae. J. Biol. Chem. 264, 4412-4416. (23) Ma, Y.-T., Gilbert, B. A., and Rando, R. R. (1993) Inhibitors of the Isoprenylated Protein Endoprotease. Biochemistry 32, 2386-2393.

Bioconjugate Chem., Vol. 12, No. 1, 2001 43 (24) Dolence, E. K., Dolence, J. M., and Poulter, C. D. (2000) Solid-Phase Synthesis of a Farnesylated CaaX Peptide Library: Inhibitors of Recombinate Yeast CaaX Prenyl Protease. J. Comb. Chem. 2, 522-536. (25) Green, N. M. (1999) Avidin. Adv. Protein Chem. 29, 85-133.

BC000036G