Design, Synthesis and Preliminary Biological Evaluation of a Biotin-S-S-Phosphine Reagent

Abstract

Biotin-S-S-Phosphine was designed and synthesized as a potential tool for a proteomic study of O-GlcNAcmodified proteins. This reagent features a disulfide linker between a triarylphosphine moiety, which allows selective conjugation to azide-containing proteins, and a biotin moiety that can allow easy isolation through its strong affinity toward avidin-coated solid beads. The disulfide linkage within this reagent can allow the easy release of the bound molecules of interest, which is difficult to achieve when a biotin:avidin pair is used alone, by reducing the disulfide bond of the reagent with DTT. Preliminary in vitro biological assays with azidelabeled and unlabeled cell lysates and a pure protein Nup62 showed that the Biotin-S-S-Phosphine reagent is highly reactive toward the free thiol groups of proteins. When a molecular tool with a disulfide linker is applied to the enrichment of the molecules of interest from other species, it is important to block the free-thiols of the sample using exhaustive alkylation prior to the Staudinger ligation reactions to restore the bioorthogonal nature of this reaction.

Introduction

Protein phosphorylation is the most common form of posttranslational modification and has long been considered a major regulatory mechanism in cellular processes. However, cumulative data has revealed protein O-GlcNAc modification to be another major regulator of cellular signaling. A major impediment to determining the molecular mechanistic roles of O-GlcNAc in the specific signaling cascades has been the lack of sensitive and easy-to-use tools for identifying O-GlcNAcylated proteins and site localization. Mapping the modification site(s), however, is a particularly challenging task due to the low stoichiometry of O-GlcNAc modification at each site on the proteins and the labile property of the glycosidic linkage between the GlcNAc moiety and the Ser/Thr residues of the proteins. Furthermore, in MS, if a small portion of unmodified peptides co-exist, unmodified peptides are preferentially ionized, which suppresses the signal of the glycopeptides. Therefore, it is essential to isolate or enrich O-GlcNAcylated proteins/peptides from other non-glycosylated species. Several enrichment strategies that allow for the efficient isolation of the molecules of interest from other species have been developed. One of the useful strategies for detecting and isolating glycoproteins involves the tagging of target proteins with a bioorthogonal functional group, a “chemical handle”, and conjugation of the chemical handle with an appropriate epitope probe through highly selective chemical reactions (chemoselective reactions). Among the chemical handles available, azide has proven to be valuable as it is not typically found in biological systems and is unreactive with native biological functional groups (i.e., bioorthogonal). The azide-moiety can be displayed on the target proteins using either the cell’s biosynthetic machinery1-4 or the in vitro enzymatic method using a genetically engineered enzyme and an azide-labeled unnatural UDP-sugar analog,5 and it has unique reactivity with phosphine-and alkyne-based probes. The most commonly used reactions involving azide-functionality include Staudinger ligation with a triaryl phosphine probe,1,6,7 a copper(I)-catalyzed cycloaddition with a terminal alkyne probe,8-10 which is called a ‘click’ reaction,11 and the strain-promoted [3+2] azide-alkyne cycloaddition, which is also known as copper-free ‘clikc’ chemistry.12-16 Using these chemoselective reactions, azidelabeled glycoproteins can be ligated covalently to either a visualization tag or an enrichment tag. A phosphine or alkyne reagent with an appropriate epitope tag has been used as an effective tool for a range of biological and physiological studies of post-translationally modified proteins. Biotin has been chosen as the most popular epitope tag for the detection and enrichment of the molecules of interest owing to its very strong affinity for avidin (or streptavidin). The strong interaction between biotin and avidin can tolerate harsh washing conditions, allowing the removal of nonspecific bound species, thereby minimizing the contamination from unmodified peptides and decreasing the ion suppression effect in MS analysis. Their strong binding, however, becomes problematic when the bound biotinylated glycopeptides are eluted from the avidin solid support. This is because the harsh conditions needed to interrupt the strong biotin:avidin interaction to release the bound glycopeptides from the solid support can lead to a decomposition of the glycopeptides, resulting in low product recovery. To tackle this problem, azide-reactive reagents have been engineered further by inserting a readily cleavable linker between the biotin moiety and azide-reactive moiety. A photocleavable biotin reagent, which utilizes the ultraviolet light-catalyzed photochemical reaction for the release of the molecules of interest, has been introduced,17 and Wang et al. exploited this photocleavable linker in the form of the biotin-photocleavable linker-alkyne reagent for a proteomic study of some neuronal proteins modified with O-GlcNAc.18 However, the synthetic challenge and instability of that reagent have limited its widespread utility. Therefore, this study developed a new linker that can be readily synthesized and stable at room temperature. A previous study reported the use of a disulfide bond as a cleavable linker in the selection of their enzymes of interest.19 Figure 1 presents a strategy using a disulfide linker-containing reagent: glycosylated proteins (or peptides) with azide-functionality are labeled with biotin through chemoselective reactions such as Staudinger ligation or Click reaction. The biotin-labeled proteins are then isolated (or enriched) from the unlabeled proteins using the avidin-coated solid beads. After removing all the contaminants from the bound proteins by several washes, a disulfide linker is then cleaved by reducing the disulfide bond with DTT, and the products released are thiol-alkylated, and then subjected to mass analysis for the characterization of their identities and elucidation of the modification site(s).

Figure 1.Illustration of the use of a Biotin-S-S-Phosphine reagent for a proteomic study of post-translationally modified glycoproteins.

Therefore, Biotin-S-S-Phosphine 1 was designed as a potential molecular tool that can be used in the proteomic study of post-translationally modified protein. Reagent 1 contains the biotin and azide-reactive moieties at each end separated by a disulfide linker. This article reports the synthesis and preliminary results of some biological assays of reagent 1.

 

Experimental

Materials. Unless stated otherwide, all chemicals were purchased from Sigma-Aldrich. All solutions were prepared using ultrapure deionized water. HeLa cells were supplied by ATCC (Manassas, VA, USA). DMEM, FBS, PBS, NuPAGE pre-cast gels, and 4 × NuPAGE LDS sample buffer were purchased from Invitrogen (Carlsbad, CA, USA). Complete, mini, EDTA-free protease inhibitor cocktail tablets were acquired from Roche Applied Science (Indianapolis, IN, USA). M-PER lysis buffers and BCA protein assay reagent were obtained from Thermo Scientific Pierce (Rockford, IL, USA). IRDye800CW-Streptavidin was purchased from LI-COR Biosciences (Lincoln, NE, USA). Recombinant human nuclear pore glycoprotein, p62 (Nup62) was supplied by Bioclone (San Diego, CA, USA). Biotin–Alkyne with copper(II)-salt and reducing agent were contained in the Click-iT™ Protein Analysis Detection Kits (Invitrogen). Synthetic intermediates, and target compound were confirmed by high-resolution (HR) ESI MS analysis. High-resolution mass measurements (HRMS) were performed on a Micromass/Waters LCT Premier Electrospray Time of Flight (TOF) mass spectrometer coupled with a Waters HPLC system. The immunoblots were imaged according to the manufacturer’s instructions using the Odyssey Infrared Imaging System (LI-COR Biosciences).

Synthesis of the Biotin-S-S-Phosphine Reagent 1.

Compound 4: HATU (1.59 g, 4.06 mmol), diisopropylethylamine (571 mg, 4.42 mmol), and compound 3 (1.01 g, 4.07 mmol) were adeed to a solution of biotin 2 (980 mg, 4.01 mmol) in DMF (9 mL). The reaction mixture was stirred for 20 h at room temperature. Subsequently, DMF was removed under low pressure. The oily residue was purified by flash column chromatography eluting with methylene chloride/methanol (20:1 to 4:1, v/v) to afford compound 4 (55%). HRMS (ESI): [M+H]+ calculated for C21H39N4O6S: m/z = 475.2590; found: m/z = 475.2598.

Compound 7: Sodium bicarbonate (1.04 g, 12.4 mmol), and Fmoc-Cl solution [1.4 g, 5.41 mmol in 1,4-dioxane (25 mL)] were added to a solution of 3-(2-aminoethyldisulfanyl) propanoic acid (805 mg, 4.44 mmol) in water (50 mL). The reaction mixture was stirred for 16 h at room temperature and then washed with ethyl ether (20 mL). The aqueous layer was acidified with 2 M HCl to pH 1-2 and extracted with ethyl acetate. The organic layer was washed with water, dried with magnesium sulfate, filtered, and concentrated under vacuum. The residue was purified by flash column chromatography eluting with n-hexane/ethyl acetate/acetic acid (40:10:1 to 10:10:1, v/v/v) to yield compound 6 (63%). HRMS (ESI): [M+H]+ calculated for C20H22NO4S2: m/z = 404.0990; found: m/z = 404.1005.

TFA (3 mL) was added to a solution of compound 4 (204 mg, 0.430 mmol) in methylene chloride (3 mL). The reaction mixture was stirred for 1.5 h at room temperature. Methylene chloride and TFA were removed and the resulting residue was kept under vacuum overnight. The residue (5) was dissolved in 1,4-dioxane (6 mL) and compound 6 (200 mg, 0.496 mmol), HATU (205 mg, 0.522 mmol), and diisopropylethylamine (312 mg, 2.41 mmol) were added. The reaction mixture was stirred for 24 h at room temperature and 1,4-dioxane was then removed under low pressure. The residue was purified by flash column chromatography eluting with methylene chloride/methanol (30:1 to 7:1, v/v) to afford the compound 7 (54%, two steps overall). HRMS (ESI): [M+H]+ calculated for C20H22NO4S2: m/z = 404.0990; found: m/z = 404.1005.

Compound 1 from Compound 7 (Scheme 1): To a solution of compound 7 (28.3 mg, 0.0372 mmol) in DMF (0.8 mL) was added piperidine (0.2 mL). The reaction mixture was stirred for 30 minutes at room temperature, and the reaction was quenched with water (3 mL) and subjected to lyophilization. The resulting solid (8) was dissolved in DMF (1.5 mL). Subsequently, compound 93 (11.2 mg, 0.0307 mmol), HATU (22 mg, 0.0561mmol), and diisopropylethylamine (15 mg, 0.116 mmol) were added at room temperature. The reaction mixture was stirred for 16 h at room temperature. After removing the DMF by vacuum, the resulting residue was purified by flash column chromatography eluting with CH2Cl2/CH3OH/CH3COOH (200:10:1 to 40:10:1, v/v/v) to yield compound 1 (4％). HRMS (ESI): [M+H]+ calculated for C42H55N5O8PS3: m/z = 884.2950; found: m/z = 884.2957.

Compound 12: Pentafluorophenol (120 mg, 0.652 mmol), EDC (102 mg, 0.532 mmol) and DMF (0.05 mL) were added to a solution of 2-iodoterephthalic acid-1-methyl ester 10 (110 mg, 0.359 mmol) in 1,4-dioxane (1.5 mL), and the reaction mixture was stirred for 16 h at room temperature. 1,4-Dioxane was removed and the residing residue was separated by flash column chromatography eluting with n-hexane/ ethyl acetate (20:1 to 2:1, v/v) to afford compound 11 (98％).

To a solution of compound 11 (164 mg, 0.309 mmol) in 1,4-dioxane (3 mL) were added 3-(2-aminoethyldisulfanyl)-propanoic acid (134 mg, 0.739 mmol), and diisopropylethylamine (148 mg, 1.15 mmol). The reaction mixture was stirred for 16 h at room temperature, and concentrated by vacuum. The resulting residue was dissolved in 1,4-dioxane (4 mL) and pentafluorophenol (195 mg, 1.06 mmol), EDC (185 mg, 0.965 mmol) and DMF (0.05 mL) were added. The reaction mixture was stirred for 20 h at room temperature. After removing the 1,4-dioxane, the resulting residue was separated by flash column chromatography eluting with n-hexane/ ethyl acetate (10:1 to 1:1, v/v) to give compound 12 (38％, two steps overall). HRMS (ESI): [M+H]+ calculated for C20H16F5INO5S2: m/z = 635.9435; found: m/z = 635.9461.

Compound 13: Compound 5 (107 mg, 0.219 mmol) and diisopropylethylamine (223 mg, 1.72 mmol) were added to a solution of compound 12 (72.8 mg, 0.115 mmol) in 1,4-dioxane (4 mL) and DMF (0.5 mL). The reaction mixture was stirred for 16 h at room temperature and the solvents were removed. The resulting residue was separated by flash column chromatography eluting with methylene chloride/ methanol (20:1 to 4:1, v/v) to afford compound 13 (39%). HRMS (ESI): [M+H]+ calculated for C30H45IN5O8S3: m/z = 826.1475; found: m/z = 826.1468.

Compound 14: Diphenylphosphine (18 mg, 0.0967 mmol), palladium(II) acetate (0.1 mg) and triethylamine (20.5 mg, 0.203 mmol) were added to a solution of compound 13 (30.0 mg, 0.0363 mmol) in acetonitrile (3 mL) at room temperature. The reaction mixture was stirred for 16 h at 65°C, cooled to room temperature, and then subjected to flash column chromatography eluting with methylene chloride/methanol (40:1 to 4:1, v/v) to yield compound 14 (5％). HRMS (ESI): [M+H]+ calculated for C42H55N5O9PS3: m/z = 900.2900; found: m/z = 900.2910.

Compound 16: Sodium bicarbonate (233 mg, 2.77 mmol), and di-tert-butyl dicarbonate (590 mg, 2.70 mmol) were added to a solution of 3-(2-aminoethyldisulfanyl)propanoic acid (220 mg, 1.21mmol) in water (2 mL) and THF (2 mL). The reaction mixture was stirred for 24 h at room temperature and then lyophilized to remove water. The residual residue was purified by flash column chromatography eluting with n-hexane/ethyl acetate/acetic acid (40:10:1 to 10:10:1, v/v/v) to produce compound 15 (40％). HRMS (ESI): [M+H]+ calculated for C10H19NO4S2: m/z = 281.0755; found: m/z = 281.0763.

To a solution of compound 5 (25.5 mg, 0.0522 mmol) in 1,4-dioxane (2 mL) were added compound 15 (14.7 mg, 0.0522 mmol), HATU (60 mg, 0.153 mmol) and diisopropylethylamine (26 mg, 0.201mmol). The reaction mixture was stirred for 24 h at room temperature, and 1,4-dioxane was then removed. The resulting residue was purified by flash column chromatography eluting with methylene chloride/methanol (20:1 to 4:1, v/v) to afford compound 16 (45％). HRMS (ESI): [M+H]+ calculated for C26H47N5O7S3: m/z = 638.2716; found: m/z = 638.2745.

Compound 1 from Compound 16 (Scheme 3): To a solution of compound 16 (7.0 mg, 0.110 mmol) in methylene chloride (0.5 mL) was added TFA (0.5 mL). The mixture was stirred for 1 h at room temperature. Methylene chloride and TFA were removed and kept under vacuum overnight. The resulting residue (8) was dissolved in methylene chloride (1.5 mL) and commercially available compound 17 (6.8 mg, 0.128 mmol) and diisopropylethylamine (74 mg, 0.573 mmol) were added at room temperature. The reaction mixture was stirred for 24 h at room temperature and methylene chloride was then removed. The residue was purified by flash column chromatography eluting with methylene chloride/methanol (40:1 to 4:1, v/v) to yield compound 1 (33％, two steps overall). HRMS (ESI): [M+H]+ calculated for C42H55N5O8PS3: m/z = 884.2950; found: m/z = 884.2954.

Biological Assays.

Preparation of Azide-labeled and Unlabeled HeLa Cell Lysates: Azide-labeling of the cellular proteins was achieved using a metabolic engineering method with an unnatural sugar, peracetylated N-azidoacetylglucosamine (Ac4GlcNAz, 18, Fig. 2). The cells were cultured in a 10 mL DMEM supplemented with 10％ FBS in the presence of 100 μM of peracetylated N-azidoacetyl-glucosamine (Ac4GlcNAz) in a humidified 5％ CO2 incubator for 2 days. The cultured cells were washed with PBS at pH 7.2, transferred to a 15 mL-felcon tube and centrifuged at 2500 × g for 10 min at 4 °C. The pellets were lysed in 0.6 mL of a M-PER protein extraction buffer containing an EDTA-free protease inhibitor cocktail, according to the manufacturer’s protocol. The supernatant was obtained after centrifugation at 20,000 × g for 15 min at 4 °C and stored in aliquots at −80 °C until needed.

Reactions of Azide-labeled and Unlabeled Cell Lysates with Various Azide-specific Reagents: Before carrying out the reaction of the cell lysates with various azide-specific reagents, buffer-exchange into an appropriate buffer was performed. For the copper-catalyzed ‘click’ reactions, 50 mM Tris–HCl, pH 7.4 buffer containing 0.1％ SDS was used as the reaction buffer. PBS was used for the Staudinger ligation reaction.

Reactions of Azide-labeled and Unlabeled HeLa Cell Lysates with Terminal Alkyne and Copper Catalyst (copper-catalyzed ‘click’ reaction): Reactions of azide-labeled and unlabeled HeLa cell lysates with Biotin–Alkyne (19, Fig. 2) were carried out using a Click-iTTM Protein Analysis Detection Kit, according to the manufacturer’s protocol. Briefly, 150 μg of GlcNAz-labeled and unlabeled HeLa cell lysates in 30 μL of 50 mM Tris–HCl buffer containing 0.1％ SDS were added to 50 μL of a 2×Click-iTTM reaction buffer containing Biotin–Alkyne (19). Subsequently, 5 μL of 40 mM CuSO4 solution, and 5 μL of Click-iTTM reaction buffer additive 1 were added. After 2–3 min, 10 μL of a Click-iTTM reaction buffer additive 2 was added. The reaction tube was agitated at room temperature for 20 min. The reaction was quenched by removing the excess reagent by performing buffer-exchange into a total 80 μL volume of 50 mM Tris–HCl buffer containing 0.1％ SDS, using a 10 K-cutoff Amicon Ultra-0.5 centrifugal filter device. The reaction results were determined by (non-reducing) gel electrophoresis followed by Western blot analysis. The samples were prepared for gel electrophoresis by adding 10 μL of distilled water and 30 μL of a 4× NuPAGE® LDS sample buffer to give a total volume of 120 μL. After incubation at 80 °C for 5 min, 20 μL of each reaction sample was loaded onto a SDS–PAGE gel (10％ or 4–12％ NuPAGE® Bis–Tris gel) and run with 1× NuPAGE® MOPS SDS Running Buffer for 50 min at 200 V. The proteins were transferred electrophoretically onto a nitrocellulose membrane for Western blot analysis. The membrane was probed with IRDye® 800CW-Streptavidin at 1:5000 dilution for an evaluation of the reactions with Biotin-tagged reagent. The blot was imaged according to the manufacturer’s instructions using the Odyssey Infrared Imaging System.

Reactions of Azide-labeled and Unlabeled Cell Lysates Either with Biotin–Phosphine (20) or with Biotin-S-SPhosphine (1) (Staudinger ligation): Staudinger ligation reactions of the azide-labeled and unlabeled HeLa cell lysates were performed using either Biotin–Phosphine (20, Fig. 2) or Biotin-S-S-Phosphine (1). PBS (52 μL) and Biotin–Phospine (8 μL, 10 mM) (20) (or Biotin-S-S-Phosphine 1 (8 μL, 10 mM)) were added to 20 μL of 0.1％ SDS, 50 mM Tris –HCl buffer containing either 100 μg of GlcNAz-labeled or unlabeled HeLa cell lysate. The reaction mixture was agitated at 37 °C for 90 min. Subsequently, buffer-exchange into a total 60 μL volume of PBS containing 0.1％ SDS was performed to remove the excess, unreacted reagent using a 10 K-cutoff Amicon Ultra-0.5 centrifugal filter device. The reaction results were determined by (non-reducing) gel electrophoresis followed by Western blot analysis.

Reactions of Unlabeled and Azide-labeled Nup62 Either with Biotin-Phosphine (20) or with Biotin-S-S-Phosphine (1) (Staudinger ligation): Staudinger ligation reactions of unlabeled Nup62 were performed using either Biotin–Phosphine (20) or Biotin-S-S-Phosphine (1). To 20 μL of PBS containing 1 μg of unlabeled Nup62 were added 56 μL of PBS and 4 μL of 10 mM Biotin–Phospine (20) (or Biotin-SS- Phosphine, 1). The reaction mixture was agitated at 37 oC for 90 min. Subsequently, buffer-exchange into a total 60 μL volume of PBS containing 0.1％ SDS was performed to remove the excess, unreacted reagent using a 10 K-cutoff Amicon Ultra-0.5 centrifugal filter device. The reaction results were determined by (non-reducing) gel electrophoresis followed by Western blot analysis. The protocols for gel electrophoresis and Western blot analysis were identical as described above.

Treatment of Unlabeled Nup62 with TAMRA–Mal and Reaction of Prior TAMRA–Mal-treated Nup62: 1 μg of Nup62 in 49 μL of PBS was treated with and without 1 μL of a 10 mM tetramethylrhodamine–5-maleimide (TAMRA–Mal, 21, Fig. 2) solution at room temperature in the dark for 1 h. The reaction was quenched by removing the excess, unreacted TAMRA–Mal (21) reagent by performing buffer-exchange into a total volume of 76 μL of PBS containing 0.1％ SDS, using a 10 K-cutoff Amicon Ultra-0.5 centrifugal filter device. Subsequently, the resulting Nup62 was reacted with 4 μL of 10 mM Biotin-S-S-Phosphine 1 at 37 °C for 90 min. Subsequently, buffer-exchange into a total 60 μL volume of PBS containing 0.1％ SDS was performed to remove the excess, unreacted reagent using a 10 K-cutoff Amicon Ultra-0.5 centrifugal filter device. The reaction results were determined by (non-reducing) gel electrophoresis followed by Western blot analysis.

 

Results and Discussion

Synthesis of the Target Compound 1. Reagent 1 contains a triarylphosphine moiety for covalent ligation with an azide group-containing molecule and a biotin tag for efficient separation and enrichment from untagged other species. The two entities were connected through a disulfide linker. A series of reactions described in Scheme 1 was performed to synthesize this reagent. First, biotin 1 was reacted with Bocprotected diamine 3 and HATU to obtain amide 4. The compound was then treated with TFA/CH2Cl2 to remove the protecting Boc group followed by a reaction with Fmocprotected 3-(2-aminoethyldisulfanyl)propanoic acid 6 in the presence of HATU to afford a condensed amide compound 7. After removing the Fmoc group with piperidine, the resulting compound (8) was treated with 2-(diphenylphosphino) terephthalic acid 1-methyl ester 9 to prepare the target compound 1. The yield for this final amide bond formation was only 4％ (Scheme 1). This very low yield for the last coupling reaction was attributed to the low reactivity of the carboxylic group of compound 9. Therefore, an amide bond formation reaction was performed using a carboxylic group activated intermediate, particularly active esters. The pentafluorophenyl ester (PFPE) is a commonly used active ester,20-22 that is normally prepared from a reaction of a carboxylic acid with pentafluorophenol using a diimide as a coupling agent. The active PFPE is relatively non-polar, stable to chromatographic purification and extended storage.21,23 The reactions outlined in Scheme 2 were preformed by exploiting a reactive PFPE ester for the synthesis of amide. 2-Iodoterephthalic acid 1-methyl ester 10 was treated with pentafluorophenol and EDC to afford pentafluorophenyl ester 11, which was then reacted with 3-(2-aminoethyldisulfanyl)-propanoic acid to form the respective amide adduct. The carboxylic functionality of this disulfanyl propanoic acid intermediate was transformed to PFPE by a treatment with pentafluorophenol and EDC. Second amide formation was achieved using a newly formed PFPE 12 and a biotin derivative 4 to produce a biotin-conjugated disulfanyl 2-iodoterephthalic methyl ester 13. A reaction of compound 13 with diphenylphosphine, and palladium (II) acetate, however, did not give the target compound 1, but instead produced its oxidized form, a diphenylphosphine oxide 14.

Scheme 1.Reagents and conditions: i) HATU, diisoproylethylamine, DMF, 55%; ii) TFA/CH2Cl2; iii) HATU, diisopropyl-ethylamine, 1,4-dioxane, 54% two steps; iv) piperidine, DMF; v) HATU, diisopropylethylamine, DMF, 4%.

Scheme 2.Reagents and conditions: i) pentafluorophenol, EDC, DMF, 1,4-dioxane, 98%; ii) 3-(2-aminoethyldisulfanyl)propionic acid, diisopropylethylamine, 1,4-dioxane; iii) pentafluorophenol, EDC, DMF, 1,4-dioxane, 38% two steps; iv) 4, diisopropylethyl-amine, 1,4-dioxane, 39%; v) diphenylphosphine, palladium(II) acetate, triethylamine, acetonitrile, 5%.

Scheme 3.Reagents and conditions: i) HATU, diisopropylethylamine, 1,4-dioxane, 45%; ii) TFA/CH2Cl2; iii) 17, diisopropylethyl-amine, dichloromethane, 33% two steps.

To avoid the oxidation problem of triarylphosphine under the reaction conditions, an alternative method using a reactive pentafluorophenyl ester of triarylphosphine, which is commercially available, was used as described in Scheme 3. In this method, the amino functionality of 3-(2-aminoethyl-disulfanyl) propanoic acid was protected with a Boc group instead of a Fmoc group, and the resulting NHBoc-protected compound 15 was reacted with biotin-linked ammonium trifluoroacetate 5 to afford compound 16. This compound was treated with TFA/CH2Cl2 for Boc deprotection and then reacted with 2-(diphenylphosphino)terephthalic acid 1-methyl 4-pentafluorophenyl diester 17 to produce the target compound 1 in moderate yield (33% for the last two steps). The synthetic route shown in Scheme 3 prevented the formation of the by-product, the triarylphosphine oxide 14, successfully producing the target compound 1 in reasonable yield.

Biological Evaluation of the Biotin-S-S-Phosphine (1). To examine the biological reactivity of reagent 1, which includes three distinctive regions, a biotin tag for efficient isolation from other species, a disulfide linker for easy cleavage, and a triarylphosphine moiety for its conjugation to azide-labeled proteins, azide-labeled and unlabeled HeLa cell lysates were prepared. The preparation of the azidelabeled and unlabeled cell lysates is described in detail in the experimental section. Briefly, azide labeling of glycoproteins was achieved metabolically by incubating the cells in a growth medium containing peracetylated N-azidoacetyl-D-glucosamine ((Ac4GlcNAz, 18, Fig. 2) at 37 °C in an incubator. Like a natural sugar, the unnatural azide-labeled sugar enters the cells and is converted to UDP-GlcNAz (22, Fig. 2) by hijacking the hexosamine biosynthetic pathway and GlcNAc salvage pathway.3 O-GlcNAc transferase can catalyze the attachment of this unnatural sugar, GlcNAz from UDP-GlcNAz (22) to the serine or threonine residues of nucleocytoplasmic proteins.3

To ensure the GlcNAz-incorporation of the cellular proteins, the azide-labeled and unlabeled cell lysates were reacted with or without the biotin-conjugated terminal alkyne (19) in combination with copper (I) at room temperature for 20 min (copper-catalyzed ‘click’ reaction) or Biotin-Phosphine (20) at 37 °C for 1.5 h (Staudinger ligation). Figure 3 shows the reaction results determined by gel electrophoresis follow- ed by Western blot analysis. As expected, both copper-catalyzed ‘click ’ reaction and Staudinger ligation exhibited high specificity for azide-functionality. Immuno-reactive bands were detected only in the azide-labeled lysates (Fig. 3, lanes 4 and 6, respectively) and not in the lanes containing unlabeled lysates (lanes 3 and 5). The one band observed in the negative control lanes, 2 and 7, of Figure 3 represents most likely a non-covalent interaction of the IR-dye conjugated streptavidin with the cell lysates.

Figure 2.Chemical structures of the compounds described in the biological reactivity assays.

Figure 3.Ponceu S-stained nitrocellulose membrane (left) and immunoblot (right) of HeLa cell extracts treated with and without 19/Cu(I) or 20. The immunoblot was probed with IRDye800CW-Streptavidin. The protein marker is shown in lanes 1 and 8.

Figure 4.Ponceu S-stained nitrocellulose membrane (left) and immunoblot (right) of HeLa cell extracts treated with and without 1. Immunoblot was probed with IRDye800CW-Streptavidin.

In contrast, Biotin-S-S-Phosphine (1) produced numerous strong signals in both the absence and presence of azide-labeled cell lysates (Fig. 4, lanes 2 and 4, respectively). Because the reactions of unlabeled proteins with other Biotin-containing reagents (19 and 20) imparted little to no reactivity (Fig. 3), these signals did not arise from the nonspecific interactions of the IR-dye conjugated streptavidin with cell lysates.

The strong signals shown in the reaction of the proteins lacking azide-moieties and reagent (1) (Fig. 4, lane 2) are believed to arise from the nonspecific covalent adduct formation between them. A nucleophile, most likely the sulfhydryl group in the free cysteine residue of the protein is believed to attack a sulfur atom in the disulfide bridge of the reagent 1 to form a new disulfide bond. As shown in lanes 2 and 4 of Figure 4, treatment of the azide-labeled HeLa cell extracts with Biotin-S-S-Phosphine resulted in similar signal intensity in the immunoblot to the signal intensity arising from the reaction with the unlabeled cell extracts. The observation of no reinforcement resulting from the conjugation reaction between triarylphosphine and azide-group, suggests that the reactivity of this reagent toward the free cysteine residue is higher (or faster) than that of the azide moiety.

Purified recombinant human nuclear pore protein, Nup62, was used to further test this speculation. Nup62 is heavily modified with O-GlcNAc and contains three cysteine residues with at least one cysteine, which is not part of a disulfide-bridge. The presence of a free thiol moiety in Nup62 was confirmed using the fluorophore-labeled sulfhydryl alkylating tetramethylrhodamine-5-maleimide (TAMRA-Mal, 21). A reaction of Nup62 and TAMRA-Mal (21) at room temperature in the dark for 1 h generated a pink-color band corresponding to TAMRA-conjugated Nup62 on the nitrocellulose membrane even before staining with the Ponceu S staining solution (Fig. 5).

Figure 5.Nitrocellulose membrane of Nup62 reacting with and without TAMRA-Mal (21) before (left) and after (right) Ponceu S staining.

Figure 6.Ponceau S-stained (left) and corresponding immunoblot (right) of Nup62: lane 1 is a protein marker, lane 2 is Nup62 treated with only Biotin-S-S-Phosphine (1), and lane 3 is Nup62 subsequently treated with TAMRA-MAL (21) and then with Biotin- S-S-Phosphine. The immunoblot was analyzed using IRDye800 CW-Streptavidin.

Exposing the unlabeled Nup62 to Biotin-S-S-Phosphine (1) produced a distinct signal (Fig. 6, lane 2). On the other hand, a prior treatment of Nup62 with TAMRA-Mal (21) to block the free cysteine’s thiol resulted in the almost com-plete blocking of adduct formation between the Biotin-S-S-Phosphine reagent and Nup62 (Fig. 6, lane 3).

Figure 7.Unstained (left), Ponceau S-stained nitrocellulose membrane (middle), and immunoblot (right) of unlabeled HeLa cell extracts sequentially treated with or without TAMRA-Mal (21), and Biotin-S-S-Phosphine (1). Immunoblot was analyzed using IRDye800CW-Streptavidin.

Similarly, unlabeled HeLa cell extracts formerly reacted with TAMRA-MAL (21) before exposure to Biotin-S-S-Phosphine (1) exhibited almost no signals in the Western blot, suggesting that a sulfhydryl was indeed the only group that underwent addition with the disulfide functionality in the reagent 1 (Fig. 7). Similarly, blocking the free thiols of the unlabeled cell lysates using another sulfhydryl-alkylating reagent, 2-iodoacetamide followed by exposure to the Biotin-S-S-Phosphine (1) probe almost completely abolished the adduct formation (data not shown).

These in vitro preliminary experimental results clearly show that the Biotin-S-S-Phosphine (1) reagent has highly reactive disulfide linker. Substitution reaction of disulfide bond of the reagent 1 by a free cysteine residue of proteins is likely to occur more rapidly than the Staudinger ligation reaction between the triarylphosphine moiety of the reagent and azide-functionality of the proteins. In contrast to this observation of a highly reactive disulfide linker property in reagent 1, a previous study,19 reported the utility of their reagent compound, which has a disulfide linker between the two functional entities in isolation and selection of their enzymes of interest.

Currently, there are several questions remaining to be answered: can all cysteine residues in an individual protein be equally sensitive to the Biotin-S-S-Phosphine reagent? If so, why this reagent is so reactive and how can this reagent be modulated to improve its selectivity? If not, which cysteines are more prone to react with this reagent? The cysteine reactivity is associated with its ionization constants (pKa) and depends on the local protein microenvironment. Low pKa protein thiols, particularly those ionized at physiological pH, are often referred to as “reactive cysteines”,24 because thiolates are much stronger nucleophiles than thiol groups. Thiol ionization in the context of proteins can be facilitated if those groups are located near the positively charged amino acid,25 or hydrogen bonding,26 or at the N-terminal end of an α-helix.27,28 A further systematic investigation will be required to determine the important factors that can affect and modulate the reactivity of the disulfide bond of the Biotin-S-S-Phosphine reagent. Considering the factors, such as the pH and temperature, more study will be needed on the relative nucleophilicity of the various free cysteine residues, such as low-molecular thiols, simple peptidyl thiols, and proteins’ cysteine residues under different conditions. These further studies will help to determine if this reagent can be used as a useful tool for the proteomic study of O-GlcNAc, as envisioned.

 

Conclusion

Protein post-translational modifications play important biological roles. In concert with protein phosphorylation, which has been known as the most common form of the protein post-translational modification, and as the major regulatory mechanism in a variety of cellular functions, O-GlcNAc modification also plays important roles in cellular signaling. To understand the molecular basis of this dynamic modification, it is important to identify the proteins it modifies and determine the modification sites.

Biotin-S-S-Phosphine (1) was designed, synthesized and partially evaluated as a potential small molecular tool available for the proteomic study of post-translationally modified proteins, such as an O-GlcNAc proteomic study. The features of this reagent include a biotin moiety, a disulfide linker, and a triarylphospphine moiety. The disulfide linkage within the reagent 1 can allow the efficient release of the molecules of interest by reducing the disulfide bond of the reagent with DTT after enrichment of the target molecules.

Biotin-S-S-Phosphine (1) was synthesized successfully by a reaction of an active carboxylic acid in the form of pentafluorophenyl ester (PFPE) (17) and the NHBoc-protected reactant (16) as shown in Scheme 3 in moderate yield.

Based on the biological assays with the unlabeled cell lysates and pure protein Nup62, this reagent was found to be highly reactive toward sulfhydryls of free cysteine residues of proteins. Although more evaluation of this reagent is needed, these preliminary empirical results show that exhaustive thiol-blocking step prior to the use of this reagent in the isolation and enrichment of azide-labeled biomolecules in biological contexts is essential to restore the bioorthogonal nature of the reactions between the azide-functionality and triarylphosphine moiety. Obviously, a continuous investigation of the reactivity of the disulfide bond of this reagent on a molecular basis will assist in optimizing this reagent as a useful tool for proteomic studies of O-GlcNAc modified glycoproteins.