Synthesis of a Highly Azide-Reactive and Thermosensitive

May 16, 2017 - O-linked β-N-acetylglucosamine (O-GlcNAc) is a ubiquitous post-translational modification of proteins in eukaryotic cells. Despite the...
1 downloads 13 Views 2MB Size
Article pubs.acs.org/ac

Synthesis of a Highly Azide-Reactive and Thermosensitive Biofunctional Reagent for Efficient Enrichment and Large-Scale Identification of O‑GlcNAc Proteins by Mass Spectrometry Wanjun Zhang,‡,§ Tong Liu,†,§ Hangyan Dong,‡ Haihong Bai,‡ Fang Tian,‡ Zhaomei Shi,‡ Mingli Chen,† Jianhua Wang,*,† Weijie Qin,*,‡ and Xiaohong Qian*,‡ †

Research Center for Analytical Sciences, College of Sciences, Northeastern University, Shenyang 110819, PR China National Center for Protein Sciences Beijing, State Key Laboratory of Proteomics, Beijing Proteome Research Center, Beijing Institute of Radiation Medicine, Beijing 102206, PR China



S Supporting Information *

ABSTRACT: O-linked β-N-acetylglucosamine (O-GlcNAc) is a ubiquitous post-translational modification of proteins in eukaryotic cells. Despite their low abundance, O-GlcNAcmodified proteins play many important roles in regulating gene expression, signal transduction, and cell cycle. Aberrant O-GlcNAc proteins are correlated with many major human diseases, such as Alzheimer’s disease, diabetes, and cancer. Because of the extremely low stoichiometry of O-GlcNAc proteins, enrichment is required before mass spectrometry analysis for large-scale identification and in-depth understanding of their cellular function. In this work, we designed and synthesized a novel thermosensitive immobilized triarylphosphine reagent as a convenient tool for efficient enrichment of azide-labeled O-GlcNAc proteins from complex biological samples. Immobilization of triarylphosphine on highly water-soluble thermosensitive polymer largely increases its solubility and reactivity in aqueous solution. As a result, facilitated coupling is achieved between triarylphosphine and azide-labeled O-GlcNAc proteins via Staudinger ligation, due to the increased triarylphosphine concentration, reduced interfacial mass transfer resistance, and steric hindrance in homogeneous reaction. Furthermore, solubility of the polymer from complete dissolution to full precipitation can be easily controlled by simply adjusting the environmental temperature. Therefore, facile sample recovery can be achieved by increasing the temperature to precipitate the polymer-O-GlcNAc protein conjugates from solution. This novel immobilized triarylphosphine reagent enables efficient enrichment and sensitive detection of more than 1700 potential O-GlcNAc proteins from HeLa cell using mass spectrometry, demonstrating its potential as a general strategy for low-abundance target enrichment.

G

occurrence and development of neurodegenerative diseases, diabetes, and cancer.12,17,18 Abnormal O-GlcNAcylation is increasingly accepted as a general feature of many types of cancer, indicating its potential as a new hallmark of cancer.19−23 Therefore, global mapping and in-depth study of protein OGlcNAcylation is important for understanding its cellular function and may lead to the discovery of potential therapeutic targets for disease treatment. Because of its high throughput, accuracy, and resolution, mass spectrometry (MS) is the method of choice for large-scale identification of protein post-translational modification. However, sensitive identification of O-GlcNAc proteins in complex biological samples by “shotgun” proteomics strategy is a highly challenging task, because O-GlcNAc is often substoichiometri-

lycosylation, one of the most abundant types of protein modification, is found in more than 50% of proteins in cells. 1−3 Among the many different types of protein glycosylation, O-linked β-N-acetylglucosamine (O-GlcNAc) is a unique and essential type of glycosylation that dynamically cycles on and off serine and threonine residues in nucleocytoplasmic proteins, including transcription factors, RNA binding proteins, signaling kinase, and cytoskeletal proteins.4−6 O-GlcNAc modification plays critical roles in numerous cellular functions, such as gene expression, circadian rhythms, signal transduction, cell cycle, and protein localization.7−9 The extent of GlcNAcylation on nucleocytoplasmic proteins is highly sensitive to the concentrations of glucose and other nutrients surrounding cells and therefore it serves as a nutrition sensor in regulating multiple metabolic pathways.9−13 Altered O-GlcNAcylation is correlated with many major chronic diseases and the cross-talk between O-GlcNAcylation and phosphorylation14−16 has been studied to understand the © XXXX American Chemical Society

Received: December 13, 2016 Accepted: May 4, 2017

A

DOI: 10.1021/acs.analchem.6b04960 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

Scheme 1. Schematic Overview of the Strategy of Liquid Phase Homogenous Enrichment of O-GlcNAc Protein Using Highly Soluble and Thermosensitive PNIPAM-Triarylphosphine

cally modified on low-abundance regulatory proteins. A number of enrichment strategies were developed to facilitate O-GlcNAc protein mapping using mass spectrometry, including lectin (WGA), antibody, chemoenzymatic, and metabolic labeling based methods.24−35 Chemoenzymatic-based labeling, one of most widely used strategies for O-GlcNAc protein enrichment, specifically attaches an azido-derivatized galactose to the Nacetylglucosamine residue of the O-GlcANc protein using the mutated β-1,4-galactosyltransferase (Y298L, GalT).7,25,27,32,36,37 The azide-attached O-GlcANc protein can then react with alkyne resin via copper-catalyzed azide−alkyne cycloaddition (“click chemistry”) for enrichment.38 Though highly specific, azide−alkyne “click chemistry” based O-GlcANc protein enrichment requires a large amount of copper catalyst, which may cause nonspecific background protein adsorption.38 Additionally, the copper catalyst is a contaminant in the subsequent mass spectrometry (MS) analysis. Improved enrichment selectivity and compatibility with MS can be achieved by substituting the alkyne with triarylphosphine, which bioorthogonally reacts with azide via copper-free Staudinger ligation.30,39 However, efficient enrichment and sensitive identification of O-GlcNAc proteins is still difficult. The enrichment efficiency is mainly governed by the nature of the enrichment reaction and the reagent concentrations. Considering the extremely low abundance of O-GlcNAcylation in real biological samples, the concentration of the enrichment reagent is the key parameter. However, the intrinsically poor solubility of triarylphosphine and alkyne reagents in aqueous solution largely limits their reactivity and enrichment yielding.40 Prolonged incubation for 18−24 h is required,41,42 which limits the analysis throughput. Therefore, the low concentration of triarylphosphine or alkyne reagents in aqueous solution is the main constraint of their application to large-scale O-GlcNAc protein enrichment and sensitive mapping.40 In this work, we designed and synthesized a highly watersoluble and reactive triarylphosphine reagent via immobilization on thermosensitive poly(N-isopropylacrylamide) (PNIPAM) copolymer for efficient enrichment of azide-labeled lowabundance O-GlcNAc proteins. Conjugation of triarylphosphine to the highly water-soluble PNIPAM enhances the solubility by up to 90 times. Furthermore, the soluble PNIPAM matrix forms a liquid-phase homogeneous system with the target protein in solution, which is advantageous compared with the commonly adopted solid−liquid two-phase heterogeneous enrichment system using solid/insoluble matrix materials. Reduced interfacial mass transfer resistance and improved accessibility of the immobilized ligands can be expected.43−45

More interestingly, the solubility of NIPAM in water is sensitive to environmental temperature.46,47 Therefore, the O-GlcNAc proteins captured by PNIPAM-triarylphosphine can be easily recovered by heat-induced polymer precipitation by increasing the temperature above its lower critical solubility temperature (LCST) (Scheme 1). Compared with other types of environmental condition-sensitive polymer, temperature sensitive PNIPAM has the advantage of improved application feasibility and robustness. No harsh pH or ion strength condition is required, which is particularly compatible with the chemically unstable O-GlcNAc modification. Combination of the increased concentration of triarylphosphine in solution, facilitated liquid-phase homogeneous enrichment, and facile sample recovery enables successful enrichment and identification of more than 1700 potential O-GlcNAc proteins from HeLa cell using mass spectrometry. Furthermore, considering the broad application of azide labeling in various types of posttranslation modification, in nascent and other low abundant proteins, we expect that this new enrichment material can be applied as a general strategy for sensitive identification.



EXPERIMENTS Synthesis of Thermosensitive Poly(N-isopropylacrylamide-co-methyl Acrylate) and Triarylphosphine Functionalization. Methyl acrylate (0.15 mL), 0.8 g of Nisopropylacrylamide (MA/NIPAM molar ratio of 1:4) and 80 mg of potassium persulfate were dissolved in 50 mL of degassed 50% methanol. The mixture was allowed to react in a nitrogen environment at 60 °C for 8 h with vigorous stirring. The obtained poly(NIPAM-co-MA) polymer was precipitated and recovered by the addition of pure ethanol. Next, the purified poly(NIPAM-co-MA) was redispersed in 50% methanol. The methyl ester groups of the copolymer were converted to hydrazide by drop by drop addition of 1.8 mL of hydrazine monohydrate while cooling in an ice bath. The reaction was allowed to proceed for 5 h at RT under stirring. Excess hydrazine monohydrate was removed by ethanol precipitation and washing of the copolymer. After redissolving 1 g of hydrazide modified PNIPAM in 2 mL of 50% methanol, 0.5 g of 2-(diphenylphosphino) terephthalic acid 1-methyl 4pentafluorophenyl diester dispersed in 500 μL of DMF was added and kept stirred under a nitrogen environment at RT overnight to convert the hydrazide in the copolymer to triarylphosphine. Finally, the obtained triarylphosphine functionalized PNIPAM (PNIPAM-TP) was purified by ethanol precipitation and dialysis (3000 cutoff) to remove the residual B

DOI: 10.1021/acs.analchem.6b04960 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

Figure 1. Synthesis route of the triarylphosphine functionalized PNIPAM (PNIPAM-TP).

reactants and was lyophilized and stored at 4 °C until further use. Azide Labeling of O-GlcNAc Protein, Enrichment by PNIPAM-TP and Trypsin Digestion. One mg of nucleocytoplasmic protein (2 μg/μL) from HeLa cell was treated with 100 Unit PNGase F in 25 mM ammonium bicarbonate at 37 °C overnight for N-glycosylation removal. Next, the protein sample was precipitated using standard chloroform/methanol precipitation and redissolved in the azide labeling buffer containing 0.5 mM UDP-GalNAz, 50 mM NaCl, 20 mM HEPES, 5 mM MnCl2, 2% NP-40, pH 7.9. After adding 75 μL of Gal-T1 (Y289L) enzyme, the solution was thoroughly mixed and incubated at 4 °C overnight with gentle agitation for azide labeling of the O-GlcNAc proteins. The azide labeled OGlcNAc proteins were thoroughly mixed with 1 mg (for mixture of α-Crystallin and BSA) or 10 mg (for nucleocytoplasmic protein from HeLa cell) of PNIPAM-TP in 20 mM HEPES (pH 7.9) and incubated in 25 °C for 1 h while agitating. After conjugating and enriching the azide labeled OGlcNAc proteins to PNIPAM-TP, the solution temperature was raised to above 40 °C to precipitate the PNIPAM-TP. The precipitated PNIPAM-TP and the enriched O-GlcNAc protein was recovered by centrifugation and the supernatant was removed. Next, 3 × 300 μL of 20 mM HEPES (pH 7.9) containing 8 M urea was used to thoroughly wash the sample to remove the nonspecifically adsorbed non-O-GlcNAc proteins. After recovery by the same high-temperature-precipitation and low-temperature-dissolution cycle, the enriched O-GlcNAc proteins were redissolved in 25 mM ammonium bicarbonate (pH 8.0) containing 8 M urea and on polymer digested while still attached on PNIPAM-TP. After protein denaturation by DTT reduction and IAA alkylation, the solution was diluted with 25 mM ammonium bicarbonate to reduce the urea concentration below 1 M. A volume of 10 μL of slurry of immobilized-trypsin prepared using protocols from our previous work48 was added and vortexed at RT for 1 min for trypsin digestion. Next, the PNIPAM-TP was removed by hightemperature-precipitation and centrifugation and the supernatant containing digested peptides from O-GlcNAc proteins was collected and desalted using C18 Zip-Tips (Millipore)

according to the manufacturer’s instructions. Finally, the peptides were resuspended in water containing 0.1% FA and subjected to mass spectrometry analysis.



RESULTS AND DISCUSSION

Synthesis, Functionalization, and Characterization of the Aqueous-Soluble and Thermosensitive PNIPAM-TP Polymer. The synthesis of triarylphosphine-functionalized poly-N-isopropylacrylamide copolymer (PNIPAM-TP) is shown in Figure 1. NIPAM and methyl acrylate was copolymerized in a 4:1 molar ratio via free radical polymerization. A high ratio between NIPAM and methyl acrylate was adopted to ensure high water solubility and sensitive thermoresponse of the copolymer. After converting the methyl ester groups in the copolymer to hydrazide by hydrazine treatment, the copolymer was functionalized with 2-(diphenylphosphino) terephthalic acid 1-methyl 4-pentafluorophenyl diester (DTAMPD) to introduce triarylphosphine groups, via reaction between the hydrazide and the activated pentafluorophenyl ester, which undergoes ester-amine coupling under mild conditions. The obtained PNIPAM-TP was characterized by X-ray photoelectron spectroscopy (XPS) to verify successful attachment of the triarylphosphine group. After conjugation with DTAMPD, a characteristic phosphorus absorption peak at ∼128.5 eV was observed in the XPS spectrum (Figure 2) of the copolymer. In contrast, no corresponding peak was observed for the two other PNIPAM copolymers without DTAMPD modification. The mass content of phosphorus in PNIPAM-TP was further determined using ICP-OES. Impressively, the phosphorus content reached 1.2%, indicating that DTAMPD modification is a highly efficient way to introduce densely packed triarylphosphine groups on the polymer chain. Thermosensitive Behavior Evaluation of PNIPAM-TP. Aqueous solubility, sensitive and reproducible thermo-response is crucial for complete recovery of PNIPAM-TP and efficient O-GlcNAc protein enrichment. NIPAM based thermoresponsive polymer has inverse solubility upon heating. The polymer molecules undergo rapid change from a coil hydrated state to a globule dehydrated state in their microstructure and phase separate/agglomeration upon temperature rising above their C

DOI: 10.1021/acs.analchem.6b04960 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

homogeneous solution, providing a soluble matrix for homogeneous reaction-based enrichment. Heating the solution in a 40 °C water bath leads to immediate aggregation of PNIPAM-TP and transforms the solution into a milky white heterogeneous suspension. The PNIPAM-TP polymer precipitates from the solution and can be easily collected by gentle centrifugation. The low-temperature dissolution and hightemperature precipitation feature makes PNIPAM-TP particularly advantageous for homogeneous reaction-based enrichment. To further optimize the thermosensitive behavior of the soluble polymer, PNIPAM-TP of different molecular weights (Mn) was prepared by varying the concentration of the polymerization initiator (potassium persulfate). The molecular weight measured by GPC is listed in Table S1 (Supporting Information), and the thermosensitive behavior of the corresponding PNIPAM-TP is shown in Figure S2 (Supporting Information). The molecular weight of PNIPAM-TP decreases as the concentration of potassium persulfate increases, which is expected, since a greater amount of polymerization initiator generates more “seed” polymer chains and therefore fewer monomers are left for each polymer chain to grow. As shown in Figure S2, although all the solution becomes cloudy when the temperature changes from 28 to 40 °C, higher molecular weight PNIPAM-TP is more sensitive to temperature changes and is more easily precipitated from the solution. The 122k Mn PNIPAM-TP has the strongest thermo-response, but its solubility is the lowest. Therefore, 1 mM potassium persulfate was selected as the optimal synthesis condition, and 65k Mn PNIPAM-TP with up to 125 mg/mL solubility in water was used in the subsequent enrichment experiments. Because of the high aqueous solubility and the heavy functionalization of the triarylphosphine group of PNIPAM-TP (1.2% mass content of phosphorus), the molar concentration of the PNIPAM-TP immobilized triarylphosphine group in water reaches as high as 46.9 mM, which is 90 times higher than that of typical commercial reagents (0.5 mM maximum solubility for EZ-Link Phosphine-PEG3-Biotin). The largely increased concentration of the triarylphosphine group effectively enhances the reaction rate according to the rate law of chemical reactions and is particularly beneficial for enrichment of the extremely low abundance azide-labeled O-GlcNAc proteins from a complex mixture via Staudinger ligation. Highly reproducible and complete transformation between dissolution and precipitation of the soluble polymer matrix is a prerequisite for efficient enrichment, since repeated washing steps are necessary to remove nonspecifically adsorbed protein. The temperature-induced transmittance variation of the PNIPAM-TP solution was determined by UV adsorption at 280 nm. The UV adsorption of PNIPAM-TP at complete dissolution (28 °C) and full precipitation (40 °C) in the first temperature cycle was set as 100% and 0% transmittance. As shown in Figure S3, the transmittance of the PNIPAM-TP solution still reaches >95% and 0% at 28 and 40 °C after six temperature cycles, indicating that repeated dissolution and precipitation cycles do not impair the solubility and thermoresponse of PNIPAM-TP. The high reproducibility of the thermo-sensitive behavior of PNIPAM-TP makes it particularly suitable for protein enrichment, and repeated washing only results in negligible sample loss. Low abundance O-GlcNAc proteins need to be labeled with azide-modified galactose (UDP-GalNAz) using Gal-T1 (Y289L) enzyme, which conjugates UDP-GalNAz to the O-

Figure 2. XPS spectra of phosphorus of poly(N-isopropylacrylamideco-methyl acrylate) (green), after hydrazine treatment (blue) and after triarylphosphine functionalization (red).

lower critical solution temperature (LCST). This conformation change is reversible and the polymer is capable of returning to its initial state when the temperature drops below its LCST. The thermosensitive behavior of PNIPAM-TP was evaluated using dynamic light scattering (DLS), gel permeation chromatography (GPC), and UV adsorption. In Figure 3, the

Figure 3. DLS analysis of the hydrodynamic diameters of PNIPAMTP under different temperature. Inserted picture: TEM image of PNIPAM-TP collected at 40 °C.

hydrodynamic diameter of PNIPAM-TP measured by DLS shows a sharp increase from less than 5 nm to more than 1500 nm within a 12 °C range, indicating a dramatic change in the microscale morphology of PNIPAM-TP from dissolved linear polymer chains to aggregated self-assembled micrometer-sized particles. The aggregated PNIPAM-TP collected at 40 °C was characterized by TEM and is shown in the inset in Figure 3. We can see linear polymer chains of PNIPAM-TP agglomerate into round particles at temperature above its LCST. The obviously increased hydrodynamic diameter of PNIPAM-TP remarkably alters its dissolution in aqueous solution, as shown in Figure S1 (Supporting Information). Below 28 °C, PNIPAM-TP completely dissolves in water to form a transparent and D

DOI: 10.1021/acs.analchem.6b04960 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

Figure 4. Gel electrophoresis pictures of the supernatant of PNIPAM-TP enrichment of (azide)-α-Crystallin under different conditions. (a) Different amounts of PNIPAM-TP were incubated with (azide)-α-Crystallin. (b) PNIPAM-TP was incubated with azide-α-Crystallin for different amounts of time. (c) Selectivity evaluation of PNIPAM-TP enrichment using a mixture of (azide)-α-Crystallin and BSA.

GlcNAc residues on the target proteins,26 before they can react with the triarylphosphine groups on PNIPAM-TP via Staudinger ligation. Azide-labeled α-Crystallin was used as a standard O-GlcNAc protein to evaluate the feasibility of soluble polymer-based homogeneous enrichment using PNIPAM-TP. Figure 4a shows the gel electrophoresis image of the supernatant of (azide)-α-Crystallin after enrichment by different amounts of PNIPAM-TP. The amount of the two subunits of azide-α-Crystallin (A and B chains) remaining in the supernatant after enrichment decreases as the amount of PNIPAM-TP increases. When 1 mg PNIPAM-TP was used, no residual azide-α-Crystallin remained in the supernatant, indicating complete enrichment. In contrast, when 1 mg PNIPAM-TP was incubated with α-Crystallin without azide labeling (lane C), a great amount of protein remained in the supernatant. This result demonstrates that specific Staudinger ligation, rather than nonspecific adsorption is responsible for the conjugation of azide-α-Crystallin on PNIPAM-TP and removal from the supernatant. In Figure 4b, we analyzed the azide-αCrystallin remaining in the supernatant after incubation with PNIPAM-TP for different amounts of time to investigate the reaction rate of the soluble polymer with the target protein. The azide-α-Crystallin remaining in the supernatant sharply decreases after only 10 min incubation, and the enrichment is nearly completed within 60 min, which is approximately 10 times faster than that using micrometer-sized agarose beads (overnight incubation is necessary for complete enrichment). The improved enrichment reaction rate is presumably attributed to the heavily functionalized and highly soluble PNIPAM-TP, which leads to largely increased concentration of triarylphosphine groups. The enrichment is further improved by the soluble polymer-based homogeneous enrichment. There is no mass transfer resistance and low steric hindrance between the completely dissolved PNIPAM-TP and the target azide-αCrystallin, which allows them to more readily react with other. Next, we evaluated the enrichment selectivity of PNIPAM-TP using a mixture of azide-α-Crystallin and BSA in a 1:10 mass ratio. Lanes 1−3 in Figure 4c are mixtures of azide-α-Crystallin and BSA, the supernatant of α-Crystallin without azide labeling and BSA after incubation with PNIPAM-TP, and the supernatant of azide-α-Crystallin and BSA after incubation with PNIPAM-TP. No obvious difference was observed in the gray scale between the gel bands of (azide) α-Crystallin and BSA in lanes 1 and 2. In contrast, there is no azide-α-Crystallin left with intact amount of BSA in lane 3. These results demonstrate that PNIPAM-TP is resistant to nonspecific proteins adsorption, which is a key feature to achieve high

enrichment specificity due to the particularly low abundance of O-GlcNAc proteins in real biological samples. To further challenge the enrichment selectivity of PNIPAM-TP, a highly diluted O-GlcNAc protein sample containing azide-α-Crystallin and BSA in a 1:100 mass ratio was prepared to mimic a real complex sample. Because of the limitation of gel electrophoresis in analyzing protein mixtures with high BSA to azide-αCrystallin ratio, we used matrix-assisted laser desorption ionization-time-of-flight-mass spectrometry (MALDI-TOFMS) instead to characterize the tryptic digests of the proteinPNIPAM-TP conjugates after enrichment. Using this method, we can actually examine the protein enriched by PNIPAM-TP, which provides direct information about the enrichment efficiency and selectivity. The PNIPAM-TP enriched proteins were on polymer digested at room temperature (RT) using the immobilized trypsin, which exhibits excellent digestion efficiency at RT as demonstrated in our previous work,48 to prevent polymer aggregation at 37 °C. Without enrichment, the spectrum is saturated with the signals of the abundant BSA peptides and no α-Crystallin peptide can be identified due to the strong signal suppression by BSA peptides (Figure S4a). In contrast, most of the BSA peptides are removed (BSA sequence coverage drops from 90% to 22%) and the sequence coverage of the A and B chains of α-Crystallin is increased to 75% and 91% after PNIPAM-TP enrichment (Figure S4b). The detected amount of the α/β chain of α-Crystallin is 13.5 and 24.6 times of that of the residue BSA using the top three protein quantification workflows.49 These results confirm the excellent enrichment selectivity of this method, because only a negligible amount of interfering peptide remains after enrichment, even though its concentration was 100 times higher than that of αCrystallin. Finally, we applied PNIPAM-TP for large-scale specific enrichment of O-GlcNAc proteins from HeLa cells. Nucleocytoplasmic proteins were extracted using commercial cell fractionation kits to prevent interference from oligosaccharidemodified N/O-lined glycoproteins. The extracted nucleocytoplasmic proteins were first treated with PNGase F to remove possible residual N-glycans before labeling with UDP-GalNAz by the Gal-T1 (Y289L) enzyme and enriching with PNIPAMTP via Staudinger ligation. After repeated washing to remove nonspecifically adsorbed proteins, PNIPAM-TP and the enriched O-GlcNAc proteins were on polymer digested by the immobilized trypsin to enable mass spectrometry analysis of the peptides that were not covalently bound to PNIPAM-TP. Although chemoenzymatic labeling of O-GlcNAc protein and azide-phosphine Staudinger ligation is specific and bioorthogonal, possible false azide labeling of glycans with GlcNAc E

DOI: 10.1021/acs.analchem.6b04960 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

first immunoprecipitated from nucleocytoplasmic protein extract of HeLa cell using corresponding protein-specific antibody (labeled as Sample) or IgG (negative control) and next subjected to western-blotting using anti-O-GlcNAc antibody (RL2). Western-blotting was also conducted using corresponding protein-specific antibody as the positive control (labeled as Input). All of the IP products obtained by the corresponding antibodies of the six tested proteins gave positive signal upon RL2 (Sample) or protein-specific antibody blotting (Input), while no band was observed in the negative control (IgG). To gain a deeper understanding of the functional role of the potential O-GlcNAc proteins identified using PNIPAM-TP enrichment, Gene Ontology (GO) annotation analysis was conducted using the bioinformatics tool FunRich52 and Uniport. The identified potential O-GlcNAc proteins are mainly intracellular proteins (91.3%) located in nucleus, cytoplasm, mitochondrion and cytoskeleton, which is consistent with the literature. In terms of molecular function, RNA binding, transcription regulator activity, structural constituent of ribosome, ubiquitin-specific protease activity, and DNA binding are among the top five most enriched ones. In the category of biological processes, regulation of nucleobase, nucleoside, nucleotide, and nucleic acid metabolism, protein metabolism, metabolism, and energy pathways constitute a large proportion of the identified potential O-GlcNAc proteins. Considering the potential therapeutic role of O-GlcNAc transferase (OGT) inhibitor in cancer treatment,53 we further investigate the O-GlcNAc proteins that respond to the activity regulation of OGT in OGT inhibitor (OSMI-1) treated HeLa cell. Among the 1707 potential O-GlcNAc proteins, 621 proteins showed quantitative decrease after OGT inhibitor treatment compared with that of the nontreated cells (Table S3), which may be considered as the highly confident OGlcNAc proteins and potential drug targets.

ending and nonspecific protein adsorption on PNIPAM-TP is inevitable. To differentiate actual O-GlcNAc proteins from false positives, two control experiments were conducted. Control 1 is prepared using nucleocytoplasmic proteins without azide labeling. Control 2 is prepared using proteins extracted from cell membrane and golgi apparatus considering the cellular location of the glycan linked glycoproteins with GlcNAc ending and reacted with UDP-GalNAz and Gal-T1 enzyme. Control 1 and control 2 are enriched using PNIPAM-TP and analyzed by LC−MS. Three replicates were conducted for each enrichment and control experiment. Proteins identified in the enrichment experiment and the two control experiments were label-free, quantitatively compared using the data processing methods described in the Supporting Information. The summed MS peak area of all the proteins identified in the enrichment experiment is more than 90 and 200 times higher than that of the two control experiments (Figure S5), indicating that Staudinger ligation is the driving force for protein enrichment and nonspecific/false protein binding is limited in the control experiment. This result is not unexpected, since the nonfouling nature and excellent resistance to nonspecific protein adsorption of PNIPAM has been demonstrated in multiple studies.50 As shown in Figure 5, based on the two strict



CONCLUSION In this work, a triarylphosphine-functionalized thermosensitive polymer is reported for facile enrichment of O-GlcNAc proteins from complex biological samples. The combined application of selective chemoenzymatic labeling of O-GlcNAc proteins, bioorthogonal Staudinger ligation, and highly soluble thermosensitive polymer-based homogeneous enrichment enables specific and efficient enrichment and identification of more than 1700 potential O-GlcNAc proteins from the HeLa cell. Although we did not locate O-GlcNAcylation sites in this study, this new enrichment strategy achieves obviously improved throughput and efficiency for O-GlcNAc protein identification than traditional western-blot and microarray base methods. Furthermore, considering a number of azide-based protein labeling methods are currently available, we expect that this new enrichment material can be applied as a general approach to promote sensitive identification of low abundance proteins.

Figure 5. Scatterplot of label-free quantitative comparison of proteins identified in the enrichment experiment (E) and the two control experiments (C1 and C2). Red spots represent the potential OGlcNAc proteins that meet the two screening criteria and black spots are non-O-GlcNAc proteins.

screening criteria: fold-difference (≥4) between the quantity of the corresponding proteins identified in the enrichment and control experiments and t test (p value ≤ 0.01),41,51 A total of 1707 potential O-GlcNAc proteins were obtained (Table S2, Supporting Information). We attribute the successful large-scale mapping of O-GlcNAc proteins to two possible reasons. First, the new PNIPAM-TP reagent greatly increases the concentration of the azide-reactive triarylphosphine groups and the liquid-phase homogeneous enrichment with improved mass transfer and accessibility of the reactant. Second, the biocompatible and mass spectrometry friendly Staudinger ligation is free from copper-induced protein background and does not require a tedious desalting process. Therefore, sample processing is simplified and sample loss is reduced. We randomly chose six proposed potential O-GlcNAc proteins for “protein IP and O-GlcNAc blotting” validation (Figure S6). ALDOA, NAP1L4, C1orf31, ADAR1, MCM2, and KAP1 were



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.6b04960. Additional experimental details, data, and data analysis (PDF) F

DOI: 10.1021/acs.analchem.6b04960 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry



(21) Ortiz-Meoz, R. F.; Merbl, Y.; Kirschner, M. W.; Walker, S. J. Am. Chem. Soc. 2014, 136, 4845−4848. (22) Rao, X. J.; Duan, X. T.; Mao, W. M.; Li, X. X.; Li, Z. H.; Li, Q.; Zheng, Z. G.; Xu, H. M.; Chen, M.; Wang, P. G.; Wang, Y. J.; Shen, B. H.; Yi, W. Nat. Commun. 2015, 6, 8468. (23) Zeng, Q. H.; Zhao, R. X.; Chen, J. F.; Li, Y. N.; Li, X. D.; Liu, X. L.; Zhang, W. M.; Quan, C. S.; Wang, Y. S.; Zhai, Y. X.; Wang, J. W.; Youssef, M.; Cui, R. T.; Liang, J. Y.; Genovese, N.; Chow, L. T.; Li, Y. L.; Xu, Z. X. Proc. Natl. Acad. Sci. U. S. A. 2016, 113, 9333−9338. (24) Vocadlo, D. J.; Hang, H. C.; Kim, E. J.; Hanover, J. A.; Bertozzi, C. R. Proc. Natl. Acad. Sci. U. S. A. 2003, 100, 9116−9121. (25) Khidekel, N.; Arndt, S.; Lamarre-Vincent, N.; Lippert, A.; Poulin-Kerstien, K. G.; Ramakrishnan, B.; Qasba, P. K.; Hsieh-Wilson, L. C. J. Am. Chem. Soc. 2003, 125, 16162−16163. (26) Clark, P. M.; Dweck, J. F.; Mason, D. E.; Hart, C. R.; Buck, S. B.; Peters, E. C.; Agnew, B. J.; Hsieh-Wilson, L. C. J. Am. Chem. Soc. 2008, 130, 11576−11577. (27) Sakabe, K.; Wang, Z. H.; Hart, G. W. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 19915−19920. (28) Wang, Z. H.; Udeshi, N. D.; O’Malley, M.; Shabanowitz, J.; Hunt, D. F.; Hart, G. W. Mol. Cell. Proteomics 2010, 9, 153−160. (29) Teo, C. F.; Ingale, S.; Wolfert, M. A.; Elsayed, G. A.; Not, L. G.; Chatham, J. C.; Wells, L.; Boons, G. J. Nat. Chem. Biol. 2010, 6, 338− 43. (30) Boyce, M.; Carrico, I. S.; Ganguli, A. S.; Yu, S. H.; Hangauer, M. J.; Hubbard, S. C.; Kohler, J. J.; Bertozzi, C. R. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 3141−3146. (31) Yu, S. H.; Boyce, M.; Wands, A. M.; Bond, M. R.; Bertozzi, C. R.; Kohler, J. J. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 4834−9. (32) Alfaro, J. F.; Gong, C. X.; Monroe, M. E.; Aldrich, J. T.; Clauss, T. R.; Purvine, S. O.; Wang, Z.; Camp, D. G., 2nd; Shabanowitz, J.; Stanley, P.; Hart, G. W.; Hunt, D. F.; Yang, F.; Smith, R. D. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 7280−5. (33) Cao, Z.; Partyka, K.; McDonald, M.; Brouhard, E.; Hincapie, M.; Brand, R. E.; Hancock, W. S.; Haab, B. B. Anal. Chem. 2013, 85, 1689−1698. (34) Zhu, Y. P.; Liu, T. W.; Cecioni, S.; Eskandari, R.; Zandberg, W. F.; Vocadlo, D. J. Nat. Chem. Biol. 2015, 11, 319−U105. (35) Robinson, P. V.; Tsai, C. T.; de Groot, A. E.; McKechnie, J. L.; Bertozzi, C. R. J. Am. Chem. Soc. 2016, 138, 10722−10725. (36) Khidekel, N.; Ficarro, S. B.; Clark, P. M.; Bryan, M. C.; Swaney, D. L.; Rexach, J. E.; Sun, Y. E.; Coon, J. J.; Peters, E. C.; Hsieh-Wilson, L. C. Nat. Chem. Biol. 2007, 3, 339−348. (37) Wang, Z.; Gucek, M.; Hart, G. W. Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 13793−13798. (38) Hahne, H.; Sobotzki, N.; Nyberg, T.; Helm, D.; Borodkin, V. S.; van Aalten, D. M. F.; Agnew, B.; Kuster, B. J. Proteome Res. 2013, 12, 927−936. (39) Sletten, E. M.; Bertozzi, C. R. Acc. Chem. Res. 2011, 44, 666− 676. (40) Agard, N. J.; Baskin, J. M.; Prescher, J. A.; Lo, A.; Bertozzi, C. R. ACS Chem. Biol. 2006, 1, 644−648. (41) Eichelbaum, K.; Winter, M.; Diaz, M. B.; Herzig, S.; Krijgsveld, J. Nat. Biotechnol. 2012, 30, 984−990. (42) Champasa, K.; Longwell, S. A.; Eldridge, A. M.; Stemmler, E. A.; Dube, D. H. Mol. Cell. Proteomics 2013, 12, 2568−2586. (43) Bergbreiter, D. E.; Tian, J. H.; Hongfa, C. Chem. Rev. 2009, 109, 530−582. (44) Yang, Y. F.; Priyadarshani, N.; Khamatnurova, T.; Suriboot, J.; Bergbreiter, D. E. J. Am. Chem. Soc. 2012, 134, 14714−14717. (45) Xu, J. Q.; Huang, S. Y.; Wei, S. B.; Yang, M. Z.; Cao, C. Y.; Jiang, R. F.; Zhu, F.; Ouyang, G. F. Anal. Chem. 2016, 88, 8921−8925. (46) Liu, J.; Bai, S. Y.; Jin, Q. R.; Li, C.; Yang, Q. H. Chemical Science 2012, 3, 3398−3402. (47) Kelley, E. G.; Albert, J. N. L.; Sullivan, M. O.; Epps, T. H. Chem. Soc. Rev. 2013, 42, 7057−7071. (48) Fan, C.; Shi, Z. M.; Pan, Y. T.; Song, Z. F.; Zhang, W. J.; Zhao, X. Y.; Tian, F.; Peng, B.; Qin, W. J.; Cai, Y.; Qian, X. H. Anal. Chem. 2014, 86, 1452−1458.

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. *E-mail: [email protected]. ORCID

Jianhua Wang: 0000-0003-2175-3610 Weijie Qin: 0000-0002-7633-9786 Author Contributions §

W.Z. and T.L. contributed equally to this work.

Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This study was supported by National Key Program for Basic Research of China (No. 2013CB911204, 2016YFA0501403, 2014CBA02001), Innovation Project (16CXZ027), National Natural Science Foundation of China (No. 21235001, 21405175 and 21675172) and Beijing Science and Technology Plan Project (No. Z161100002616036).



REFERENCES

(1) Zaia, J. Nat. Methods 2011, 8, 55−57. (2) Moremen, K. W.; Tiemeyer, M.; Nairn, A. V. Nat. Rev. Mol. Cell Biol. 2012, 13, 448−462. (3) Cruz, I. N.; Barry, C. S.; Kramer, H. B.; Chuang, C. C.; Lloyd, S.; van der Spoel, A. C.; Platt, F. M.; Yang, M.; Davis, B. G. Chemical Science 2013, 4, 3442−3446. (4) Hanover, J. A.; Krause, M. W.; Love, D. C. Nat. Rev. Mol. Cell Biol. 2012, 13, 312−21. (5) Lazarus, M. B.; Jiang, J.; Kapuria, V.; Bhuiyan, T.; Janetzko, J.; Zandberg, W. F.; Vocadlo, D. J.; Herr, W.; Walker, S. Science 2013, 342, 1235−9. (6) Ruan, H. B.; Dietrich, M. O.; Liu, Z. W.; Zimmer, M. R.; Li, M. D.; Singh, J. P.; Zhang, K.; Yin, R.; Wu, J.; Horvath, T. L.; Yang, X. Cell 2014, 159, 306−17. (7) Rexach, J. E.; Clark, P. M.; Mason, D. E.; Neve, R. L.; Peters, E. C.; Hsieh-Wilson, L. C. Nat. Chem. Biol. 2012, 8, 253−61. (8) Li, M. D.; Ruan, H. B.; Hughes, M. E.; Lee, J. S.; Singh, J. P.; Jones, S. P.; Nitabach, M. N.; Yang, X. Cell Metab. 2013, 17, 303−10. (9) Hardiville, S.; Hart, G. W. Cell Metab. 2014, 20, 208−13. (10) Gloster, T. M.; Zandberg, W. F.; Heinonen, J. E.; Shen, D. L.; Deng, L. H.; Vocadlo, D. J. Nat. Chem. Biol. 2011, 7, 174−181. (11) Wellen, K. E.; Thompson, C. B. Nat. Rev. Mol. Cell Biol. 2012, 13, 270−U1. (12) Erickson, J. R.; Pereira, L.; Wang, L. G.; Han, G. H.; Ferguson, A.; Dao, K.; Copeland, R. J.; Despa, F.; Hart, G. W.; Ripplinger, C. M.; Bers, D. M. Nature 2013, 502, 372−376. (13) Lagerlof, O.; Slocomb, J. E.; Hong, I.; Aponte, Y.; Blackshaw, S.; Hart, G. W.; Huganir, R. L. Science 2016, 351, 1293−1296. (14) Sprung, R.; Nandi, A.; Chen, Y.; Kim, S. C.; Barma, D.; Falck, J. R.; Zhao, Y. M. J. Proteome Res. 2005, 4, 950−957. (15) Tarrant, M. K.; Rho, H. S.; Xie, Z.; Jiang, Y. L.; Gross, C.; Culhane, J. C.; Yan, G.; Qian, J.; Ichikawa, Y.; Matsuoka, T.; Zachara, N.; Etzkorn, F. A.; Hart, G. W.; Jeong, J. S.; Blackshaw, S.; Zhu, H.; Cole, P. A. Nat. Chem. Biol. 2012, 8, 262−9. (16) Wu, J. L.; Wu, H. Y.; Tsai, D. Y.; Chiang, M. F.; Chen, Y. J.; Gao, S. J.; Lin, C. C.; Lin, C. H.; Khoo, K. H.; Lin, K. I. Nat. Commun. 2016, 7, 12526. (17) Ma, J. F.; Hart, G. W. Expert Rev. Proteomics 2013, 10, 365−380. (18) Marotta, N. P.; Lin, Y. H.; Lewis, Y. E.; Ambroso, M. R.; Zaro, B. W.; Roth, M. T.; Arnold, D. B.; Langen, R.; Pratt, M. R. Nat. Chem. 2015, 7, 913−920. (19) Slawson, C.; Hart, G. W. Nat. Rev. Cancer 2011, 11, 678−684. (20) Ma, Z.; Vosseller, K. Amino Acids 2013, 45, 719−33. G

DOI: 10.1021/acs.analchem.6b04960 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry (49) Grossmann, J.; Roschitzki, B.; Panse, C.; Fortes, C.; BarkowOesterreicher, S.; Rutishauser, D.; Schlapbach, R. J. Proteomics 2010, 73, 1740−1746. (50) Biggs, C. I.; Walker, M.; Gibson, M. I. Biomacromolecules 2016, 17, 2626−2633. (51) Eberl, H. C.; Spruijt, C. G.; Kelstrup, C. D.; Vermeulen, M.; Mann, M. Mol. Cell 2013, 49, 368−378. (52) Pathan, M.; Keerthikumar, S.; Ang, C. S.; Gangoda, L.; Quek, C. Y. J.; Williamson, N. A.; Mouradov, D.; Sieber, O. M.; Simpson, R. J.; Salim, A.; Bacic, A.; Hill, A. F.; Stroud, D. A.; Ryan, M. T.; Agbinya, J. I.; Mariadason, J. M.; Burgess, A. W.; Mathivanan, S. Proteomics 2015, 15, 2597−2601. (53) Ortiz-Meoz, R. F.; Jiang, J. Y.; Lazarus, M. B.; Orman, M.; Janetzko, J.; Fan, C. G.; Duveau, D. Y.; Tan, Z. W.; Thomas, C. J.; Walker, S. ACS Chem. Biol. 2015, 10, 1392−1397.

H

DOI: 10.1021/acs.analchem.6b04960 Anal. Chem. XXXX, XXX, XXX−XXX