Highly Selective Capture of Monophosphopeptides by Two

Jun 16, 2019 - Separation of monophosphopeptides from multi-phosphopeptides in complex biological samples is significant in the study of protein kinas...
0 downloads 0 Views 6MB Size
Download PDF
Article pubs.acs.org/ac

Cite This: Anal. Chem. 2019, 91, 9093−9101

Highly Selective Capture of Monophosphopeptides by TwoDimensional Metal−Organic Framework Nanosheets Jing Xiao,† Shi-Shu Yang,† Jian-Xiang Wu,† He Wang,§ Xizhong Yu,‡ Wenbin Shang,‡ Gui-Quan Chen,§ and Zhi-Yuan Gu*,†

Downloaded via UNIV OF SOUTHERN INDIANA on July 19, 2019 at 03:35:49 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

†

Jiangsu Key Laboratory of Biofunctional Materials, Jiangsu Collaborative Innovation Center of Biomedical Functional Materials, College of Chemistry and Materials Science, Nanjing Normal University, Nanjing 210023, China ‡ Key Laboratory for Metabolic Diseases in Chinese Medicine, First College of Clinical Medicine, Nanjing University of Chinese Medicine, Nanjing 210023, China § State Key Laboratory of Pharmaceutical Biotechnology, MOE Key Laboratory of Model Animal for Disease Study, Model Animal Research Center, Nanjing University, 12 Xuefu Avenue, Nanjing, 210061, China S Supporting Information *

ABSTRACT: Separation of monophosphopeptides from multiphosphopeptides in complex biological samples is significant in the study of protein kinase signal transduction pathways. To the best of our knowledge, very few materials have been reported that could selectively enrich monophosphopeptides because of the chemical difficulty in retaining the intermediate monophosphopeptides and excluding both non-phosphopeptides and multiphosphopeptides in acidic conditions, which requires unique interactions to balance the metallic affinity and the hydrophobicity. With the large surface area, abundant accessible active sites, and ultrathin structures, two-dimensional (2-D) metal− organic framework (MOF) Hf-1,3,5-tris(4-carboxyphenyl)benzene (BTB) nanosheets were rationally selected. Due to the elongated organic ligands and the balance between metallic affinity of clusters and hydrophobicity from ligands, the 2-D Hf-BTB nanosheets exhibited unique enrichment selectivity toward monophosphopeptides. The 2-D MOF nanosheets demonstrated excellent sensitivity (detection limit of 0.4 fmol μL−1) and selectivity [1:1000 molar ratios of β-casein/BSA (bovine serum albumin)] in model phosphopeptides enrichment. The nanosheets were implemented for the analysis of nonfat milk and human saliva samples as well as in situ isotope labeling for dysregulated phosphopeptides from patients’ serum with anal canal inflammation, exhibiting 6.6-fold upregulation of serum phosphopeptide HS4 (ADpSGEGDFLAEGGGVR) compared to the control healthy serum. The proteomics analysis of mouse brain cortical samples associated with Alzheimer’s disease, which were from Akt (protein kinase B) conditional knockout mouse and littermate control mouse, was further established with 2-D Hf-BTB nanosheets. With high capture efficiency for monophosphopeptides, this method was capable of distinguishing the difference of monophosphopeptides from microtubuleassociated protein τ (MAPT/τ) between the Akt knockout sample and control sample.

P

ylation of myosin light chain differentially regulated adhesion and polarity in the migrating cells.6 Therefore, it is imperative to develop new methods that could isolate monophosphopeptides from both non-phosphopeptides and multi-phosphopeptides. The conventional analysis of phosphopeptides requires enrichment due to the low concentration and the suppression from abundant non-phosphopeptides. Diverse strategies have been reported that can effectively enrich both mono- and multi-phosphopeptides, such as immobilized metal affinity

rotein phosphorylation is a significant cellular regulatory mechanism since many receptors and enzymes are activated or deactivated through the phosphorylation and dephosphorylation with the assistance of kinases and phosphatases, respectively.1 Monophosphorylation is usually neglected in the processive multi-phosphorylation cascades, although recent discoveries reveal that monophosphorylation instead of multi-phosphorylation has regulated many processes, such as the activation of retinoblastoma tumor suppressor protein,2 phosphorylation of cardiac troponin-I in the myocardial relaxation,3 the endothelial cell retraction,4 and the chronic inflammatory environment process in Kaposi’s sarcoma lesion.5 On the other hand, the functions of monophosphopeptides could be different from those of the multi-phosphopeptides. For example, mono- and diphosphor© 2019 American Chemical Society

Received: March 29, 2019 Accepted: June 16, 2019 Published: June 16, 2019 9093

DOI: 10.1021/acs.analchem.9b01581 Anal. Chem. 2019, 91, 9093−9101

Article

Analytical Chemistry chromatography (IMAC)7−10 and metal oxide affinity chromatography (MOAC).11−14 In the above strategies, the acidic conditions are usually chosen to prevent nonphosphopeptides from binding to the immobilized metal, while acetonitrile is used to break up interactions between hydrophobic peptides and the hydrophobic parts of adsorbents.15,16 They are beneficial to the separation of phosphopeptides (both mono and multi) from non-phosphopeptides. Compared with monophosphopeptides and nonphosphopeptides, the multi-phosphopeptides with more phosphate groups have been separated by either the strong bonding to the specific enrichment materials, including the NiZnFe2O4 nanoparticles,17 PNI-co-ATBA0.2@SiO2 polymers,18 and two-dimensional (2-D) Egyptian Blue nanosheets19 or the additional modulator, such as β-glycerophosphate disodium20 and fructose-1,6-diphosphate.21 However, it is much more difficult to retain the intermediate monophosphopeptides and exclude both non-phosphopeptides and multi-phosphopeptides in the acidic conditions, which requires the unique interactions to balance the metallic affinity and the hydrophobicity. To the best of our knowledge, almost no strategy was reported that could selectively enrich the monophosphopeptides, so finding an effective enrichment strategy for monophosphopeptides is extremely urgent. Metal−organic frameworks (MOFs) have a wide range of applications in gas adsorption and separation, catalysts, and energy storage due to the unique properties of high surface areas, tunable pores sizes, and diverse functionalities.22−28 Recently, the ultrathin 2-D MOF nanosheets as an emerging type of 2-D nanomaterials have demonstrated more abundant accessible active sites than MOFs, especially in the solution phase applications, such as sample separation,29,30 effective inhibition of enzyme,31 and fluorescence sensing.32,33 The hydrophilic properties and abundant metal clusters of 2-D MOF nanosheets, Ti2(HDOBDC)2(H2DOBDC) (H4DOBDC is 2,5-dihydroxyterephtalic acid), are beneficial to the enrichment of phosphopeptides. Our group has reported that Tibased 2-D MOF nanosheets could efficiently enrich both mono- and multi-phosphopeptides.34 However, it is still a big challenge to enrich the monophosphopeptides with 2-D MOF nanosheets as well as other materials. Compared to other enrichment materials, the advantages of 2-D MOF nanosheets are the precise regulation at the angstrom (Å) level of the distance between the adjacent metal clusters. This precise regulation aims to balance the metallic affinity of the clusters and hydrophobicity from the ligands, which will probably generate suitable materials with a proper surface that could retain only monophosphopeptides and exclude both non-phosphopeptides and multi-phosphopeptides in the acidic conditions. This regulation considers three factors, the hydrophobicity, the metallic affinity and the molecular arrangement between them, which is very different from the traditional enrichment methods, such as IMAC and MOAC. Herein, the rational selection from a series of MOFs and MOF nanosheets has been shown. The 2-D Hf-BTB nanosheets with elongated organic ligands of H3BTB [BTB is 1,3,5-tris(4-carboxyphenyl)benzene] and high-valence Hf(IV) metal clusters together with other two MOF nanosheets [Hf-TATB (TATB is 2,4,6-tris(4-carboxyphenyl)-1,3,5-triazine) and Zr-BTB] are obtained for the successful enrichment of monophosphopeptides. The 2-D MOF Hf-BTB nanosheets demonstrated the excellent sensitivity and selectivity in

phosphopeptides enrichment. Furthermore, the 2-D Hf-BTB nanosheets were implemented for in situ isotope labeling for dysregulated phosphopeptides analysis from patients’ serum with anal canal inflammation and for the enrichment of phosphopeptides from the tryptic digests of brain cortical samples of Akt (protein kinase B) conditional knockout mouse and littermate control mouse associated with Alzheimer’s disease.

■

EXPERIMENTAL SECTION Synthesis of 2-D Hf-BTB Nanosheets. The 2-D Hf-BTB nanosheets were synthesized via a solvothermal method.35 Briefly, HfCl4 (14 mg) and H3BTB (12.5 mg) were dissolved in a mixture of dimethylformamide (DMF) (5 mL), HCOOH (1.113 g), and water (120 μL) in a Pyrex vial. Then, the vial was heated at 120 °C for 48 h in the oven. The 2-D Hf-BTB nanosheets were collected by centrifugation and washed three times with DMF and ethanol, respectively. After that, the 2-D Hf-BTB nanosheets were dried at 80 °C for 8 h in a vacuum oven. Enrichment of Phosphopeptides for MALDI-TOF MS Analysis. The 2-D Hf-BTB nanosheets (1 mg) were dispersed in 100 μL of loading buffer [50% acetonitrile (ACN) (v/v) and 10% trifluoroacetic acid (TFA) (v/v)] containing the digest of model protein or complex samples, and then vortexed for 30 min. Afterward, the nanosheets were washed with washing buffer [50% ACN (v/v) and 0.1% TFA (v/v)] for three times in order to remove unadsorbed peptides. Finally, the enriched phosphopeptides were eluted with 20 μL of 30% NH3·H2O before matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS) analysis. In Situ Isotope Dimethyl Labeling and Relative Quantification of Human Serum Phosphopeptides. Human serum samples were obtained from eight healthy adults in Jiangsu Province Hospital of TCM (Affiliated Hospital of Nanjing University of Chinese Medicine) and three patients with anal canal inflammation in Nanjing Chinese Medicine Hospital according to their standard clinical procedures. The mixed sera from eight healthy adults was used as control sample. All the human serum samples were obtained with consent of both healthy adults, and patients and were dealt in conformity with the relevant laws and institutional guidelines. In situ isotope labeling was achieved after phosphopeptides loading and non-phosphopeptides washing steps and before the absorbed phosphopeptides eluting step (see the Supporting Information). For relative quantification, 0.5 μL of eluent from the healthy adults or patients (light-labeled) was dropped upon the MALDI plate and the equal eluent from the control sample (heavy-labeled) was deposited in the same position. Enrichment of Phosphopeptides from Mouse Brain Tissues for LC/MS/MS Analysis. The digest of mouse brain cortical sample lysate (1 mg) was enriched with 2-D Hf-BTB nanosheets (1 mg) in 100 μL of optimal loading buffer [50% ACN (v/v) and 40% TFA (v/v)], and then vortexed for 30 min. Afterward, the nanosheets were washed with the same washing buffer in order to remove unadsorbed peptides. Finally, the enriched phosphopeptides were eluted with 20 μL of 30% NH3·H2O before MS analysis. Enriched phosphopeptides were desalted through the StageTip C18 column. Thereafter, the samples were detected by a hybrid quadrupole−TOF LC/MS/MS mass spectrometer (TripleTOF 5600+, AB Sciex) with a nanospray source. The detailed 9094

DOI: 10.1021/acs.analchem.9b01581 Anal. Chem. 2019, 91, 9093−9101

Article

Analytical Chemistry

Figure 1. (a) Powder XRD patterns for simulated and as-synthesized 2-D Hf-BTB nanosheets. (b) SEM and (c) TEM images of the 2-D Hf-BTB nanosheets. (d) AFM image of the 2-D Hf-BTB nanosheets. The inset shows the height of 10 nm outlined along the green line.

Scheme 1. Phase Diagram for the Enrichment of Phosphopeptides by MOFs with Different Hydrophobicities and Distances between Adjacent Metal Clusters (Left) and Variation of Organic Ligands and Metal Clusters for the Different MOFs and MOF Nanosheets Studied (Right)a

a

The 2-D Hf-TATB and Zr-BTB nanosheets are omitted since they are similar to 2-D Hf-BTB nanosheets.

(Figure 1a). Due to the ultrathin thickness, the 2-D Hf-BTB nanosheets had curled up in the scanning electron microscopy (SEM) image (Figure 1b). Nevertheless, the uniform sheet structure was observed in transmission electron microscopy (TEM) (Figure 1c) and atomic force microscopy (AFM) images with a thickness of 10 nm in accordance to the eightlayer thickness with a monolayer of 1.2 nm (Figure 1d). Moreover, the N2 adsorption−desorption isotherm of 2-D HfBTB nanosheets was measured to show the porosity of the material, giving the Brunner−Emmet−Teller (BET) surface

procedures and parameters of protein extraction, protein digestion, liquid chromatography/tandem mass spectrometry (LC/MS/MS), and database searching are shown in the Supporting Information.

■

RESULTS AND DISCUSSION Characterization of 2-D Hf-BTB Nanosheets. The 2-D Hf-BTB nanosheets were synthesized by a solvothermal method. The X-ray diffraction (XRD) pattern of the 2-D HfBTB nanosheets was consistent with the simulated pattern 9095

DOI: 10.1021/acs.analchem.9b01581 Anal. Chem. 2019, 91, 9093−9101

Article

Analytical Chemistry area of 378 m2 g−1 (Figure S1). These results indicated that we successfully synthesized 2-D Hf-BTB nanosheets. The characterization of other MOFs and MOF nanosheets is shown in the Supporting Information (Figures S2−S6). Rational Selection of MOFs Enrichment Materials for Monophosphopepetides. Compared with the traditional enrichment materials, MOFs could be precisely regulated at the molecular level through the rational selection of metal clusters and organic ligands. It provides the practical approach to screen possible enrichment materials with the hydrophobicity, the metallic affinity, and the molecular arrangement within them, which are three possible key factors to obtain the enrichment materials toward monophosphopeptides. Therefore, six high-valence Hf(IV) or Zr(IV) MOFs with different topologies (3-D or 2-D) and different organic ligands [terephthalic acid (BDC), biphenyl-4,4′-dicarboxylic acid (BPDC), BTB, or TATB] were selected, including 3-D HfUiO-66, 3-D Hf-UiO-67, 2-D Hf-UiO-67 nanosheets, 2-D HfBTB nanosheets, 2-D Hf-TATB nanosheets, and 2-D Zr-BTB nanosheets, as enrichment materials to evaluate the selectivity toward monophosphopeptides (Scheme 1). The hydrophobicity of the enrichment materials was calculated according to the surface area of the ligands (Supporting Information). The metallic affinity was from Zr(IV) and Hf(IV), which showed not much difference in this work. The distance between two metal clusters was selected as the geometric indicator, which was measured as the average distance of the shortest and longest distance between adjacent metallic clusters, while the pore size (8.5, 11.5, 11.5, 8.8, 8.8, and 8.8 Å for 3-D Hf-UiO-66, 3-D Hf-UiO-67, 2-D Hf-UiO-67 nanosheets, 2-D Hf-BTB nanosheets, 2-D Hf-TATB nanosheets, and 2-D Zr-BTB nanosheets, respectively) was not selected because it was not a direct indicator and showed inconsistency with the phosphopeptides selectivity. We rationalized the six MOFs with the hydrophobicity and adjacent distance of metal clusters as two orthogonal factors (Scheme 1). With the smallest hydrophobicity and shortest distance, 3-D Hf-UiO-66 could enrich not only the monophosphopeptides but also the multi-phosphopeptides with their dephosphopeptides, respectively (Figure 2a and Scheme 1). With increased hydrophobicity and distance, 3-D Hf-UiO67 still had both mono- and multi-phosphopeptides in the enrichment experiment (Figure 2b and Scheme 1). It was worth noting that the peak intensity of multi-phosphopeptide β4m in 3-D Hf-UiO-67 was lower than that of 3-D Hf-UiO-66, indicating that the increased hydrophobicity and cluster distance benefited the selectivity toward monophosphopeptides. The further increase of hydrophobicity from 3-D HfUiO-67 to 2-D Hf-UiO-67 nanosheets was established to examine the selectivity toward phosphopeptides. The enrichment result showed that both mono- and multi-phosphopeptides were detected by 2-D Hf-UiO-67 nanosheets with lower peak intensity of β4m (Figure 2c and Scheme 1), indicating that the better selectivity toward monophosphopeptides required higher hydrophobicity. On the basis of the above experiments, three of the 2-D MOF nanosheets, the 2-D Hf-BTB nanosheets, 2-D Hf-TATB nanosheets, and 2-D Zr-BTB nanosheets, which all have larger organic linkers and higher hydrophobicity than the previous Hf-UiO-66 and Hf-UiO-67, were synthesized to enrich phosphopeptides. Among them, the 2-D Hf-TATB nanosheets were synthesized for the first time. After the enrichment with these three materials, the same results were observed that only

Figure 2. MALDI-TOF mass spectra of phosphopeptides enriched from the tryptic digest of β-casein (4 × 10−6 M) with the different materials: (a) 3-D Hf-UiO-66, (b) 3-D Hf-UiO-67, (c) 2-D Hf-UiO67 nanosheets, (d) 2-D Hf-BTB nanosheets, (e) 2-D Hf-TATB nanosheets, and (f) 2-D Zr-BTB nanosheets [#, dephosphopeptides (m/z [M + H − HPO3]+); all unmarked peaks are from nonphosphopeptides; β1s, the first monophosphopeptide from β-casein; β4m, the fourth multi-phosphopeptide from β-casein; NL, normalized level].

two monophosphopeptides and their respective dephosphopeptides were detected, suggesting that they had the same selectivity toward monophosphopeptides (Figure 2d−f and Scheme 1). The high selectivity toward monophosphopeptides 9096

DOI: 10.1021/acs.analchem.9b01581 Anal. Chem. 2019, 91, 9093−9101

Article

Analytical Chemistry was successfully achieved by the rational selection of high hydrophobicity and long distance between metal clusters in MOFs. To confirm the hydrophobicity trend of MOFs and MOF nanosheets, the contact angles were measured (Table S1). The contact angles of 3-D Hf-UiO-66, 3-D Hf-UiO-67, 2D Hf-UiO-67 nanosheets, and 2-D Hf-BTB nanosheets are 33.9°, 39.1°, 50.5°, and 67.0°, respectively, confirming that the hydrophobicity is one of the key features of 2-D Hf-BTB nanosheets (Table S1). Moreover, for Zr(IV) and Hf(IV) as well as BTB and TATB, the universal selectivity is independent of the types of metal clusters or the organic ligands. The 2-D Hf-BTB/Hf-TATB/Zr-BTB nanosheets have the longer distance of adjacent metal clusters (20.1 Å) than HfUiO-66 (14.9 Å) and Hf-UiO-67 (19.0 Å) as well as the elongated organic ligands with higher hydrophobicity. Considering that the open metal sites of metallic clusters in Hf-BTB/Hf-TATB/Zr-BTB were sterically separated on each side of the nanosheet, each metallic cluster could only achieve the enrichment of monophosphopeptides. When the distance between two metallic clusters was short enough, the adjacent metallic clusters could enrich multi-phosphopeptides synergetically. However, when the longer distance between two metallic clusters was obtained, the adjacent metallic clusters could not enrich multi-phosphopeptides. At the same time, because there were more hydrophilic sites in the multi-phosphopeptides than the monophosphopeptides, the hydrophobicity of ligands was beneficial to retain the monophosphopeptides and also to exclude the multi-phosphopeptides. Therefore, we could distinguish the monophosphopeptides from the multiphosphopeptides through the hydrophobicity of organic ligands and the distance between adjacent metallic clusters. However, because of the hydrophobicity of the enrichment material, the non-phosphopeptides were detected easily in the enrichment process. Therefore, the acid enrichment condition was applied to eliminate the interference from nonphosphopeptide, which can distinguish the monophosphopeptides from non-phosphopeptides. The unique balance between metallic affinity of clusters and hydrophobicity from ligands can retain the intermediate monophosphopeptides and exclude both non-phosphopeptides and multi-phosphopeptides in the acidic conditions. Furthermore, considering that the synthesis of TATB ligand is time-consuming, the 2-D Hf-BTB nanosheets were further explored as the enrichment material for the convenience of synthesis in the following experiments. Selective Enrichment of Monophosphopeptides by 2-D Hf-BTB Nanosheets. To evaluate the performance of the 2-D Hf-BTB nanosheets, model bovine β-casein digest (4 × 10−6 M) was pretreated with the materials for loading, washing, and eluting sequentially; then, the supernatant was analyzed by MALDI-TOF MS. Moreover, the direct analysis and commercial HfO2 nanoparticles were also studied for the comparison. Without enrichment, three phosphopeptide peaks with the low signal intensities were observed accompanied with abundant non-phosphopeptide peaks (Figure 3c). Although the signal-to-noise (S/N) ratio of phosphopeptide peaks was enhanced after the enrichment by HfO2 nanoparticles, a number of non-phosphopeptide peaks could still be observed, especially in the range of m/z = 2750−3000 (Figure 3b). In contrast, after treatment with 2-D Hf-BTB nanosheets, very few non-phosphopeptide peaks could be obtained, giving a clean MS background (Figure 3a). In addition, the peaks of multi-phosphopeptide β4m (m/z = 3122) and its dephosphopeptide (m/z = 3042) could not be observed with 2-D Hf-BTB

Figure 3. MALDI-TOF mass spectra of phosphopeptides enriched from the tryptic digest of β-casein (4 × 10−6 M) with different conditions: (a) 2-D Hf-BTB nanosheets, (b) HfO2 nanoparticles, and (c) direct analysis [#, dephosphopeptides (m/z [M + H − HPO3]+); all unmarked peaks are from non-phosphopeptides; β1s, the first monophosphopeptide from β-casein; β4m, the fourth multiphosphopeptide from β-casein; NL, normalized level]. The details of the phosphopeptides are listed in Table S2.

nanosheets compared with the direct analysis and the HfO2 nanoparticles. It was worth noting that only monophosphopeptides were successfully enriched, while the peaks of β1s (m/ z = 2061), β2s (m/z = 2556) and their dephosphopeptides were identified in the MS spectrum with significant intensities. To the best of our knowledge, almost no materials were reported that could selectively enrich the monophosphopeptides. At the same time, the enrichment performance of 2-D Hf-BTB nanosheets was highly reproducible (Figure S7). Therefore, the 2-D Hf-BTB nanosheets exhibited unique selectivity toward monophosphopeptides. The details of the phosphopeptides are listed in Table S2. Meanwhile, enrichment of bovine α-casein digest (4 × 10−6 M) also confirmed that 2-D Hf-BTB nanosheets had excellent selectivity for monophosphopeptides compared with the direct analysis and the HfO2 nanoparticles (Figure S8). The selectivity of 2-D Hf-BTB nanosheets toward monophosphopeptides was quite stable and had not been significantly affected by the pH of the loading buffer, the polarity of loading buffer, and the pH of the elution solution (Figures S9−S11). To facilitate the analysis of complex samples, a mixture of 50% ACN (v/v) and 10% TFA (v/v) was chosen as the optimal loading buffer and 30% NH3·H2O was adopted as the optimal elution buffer in the following experiment. Analytical Performance of Monophosphopeptides Enrichment by 2-D Hf-BTB Nanosheets. For further confirming the enrichment performance of 2-D Hf-BTB nanosheets, the sensitivity and selectivity of 2-D Hf-BTB nanosheets toward monophosphopeptides were also inves9097

DOI: 10.1021/acs.analchem.9b01581 Anal. Chem. 2019, 91, 9093−9101

Article

Analytical Chemistry

Figure 4. Normalized MALDI-TOF mass spectra of HS4 phosphopeptides from the serum of (a) three healthy adults (light-labeled) and (b) three patients with inflammation (light-labeled) vs the control sera sample (heavy-labeled), respectively. Logarithms of the average isotope cluster peak area ratios for each phosphopeptide (HS1, HS2, HS3, and HS4) from (c) healthy serum (light-labeled) to that from the control sera (heavylabeled) and (d) patients serum (light-labeled) to the control sera (heavy-labeled).

tigated. The different concentrations of bovine β-casein digests were used to evaluate the detection limit. When the concentration of bovine β-casein digest was decreased from 4 × 10−7 to 4 × 10−10 M, one of monophosphopeptides and its dephosphopeptide could still be detected after the enrichment by 2-D Hf-BTB nanosheets (Figure S12). Undoubtedly, for monophosphopeptides enrichment, 2-D Hf-BTB nanosheets possessed excellent sensitivity, suggesting that the nanosheets could be used to capture monophosphopeptides even if the phosphopeptide concentration was extremely low. To evaluate the selectivity of 2-D Hf-BTB nanosheets toward monophosphopeptides, the tryptic digest mixture of bovine serum albumin (BSA) and bovine β-casein at different molar ratios of 10:1, 100:1, and 1000:1 were used to simulate the complex biological samples, respectively. The essential point here is to distinguish intermediate monophosphopeptides from not only multi-phosphopeptides but also abundant non-phosphopeptides, which is a strong indicator to show the feasibility in real sample analysis. When the ratio of BSA/βcasein = 10:1 or 100:1, two monophosphopeptides and their respective dephosphopeptides could be detected after the enrichment by 2-D Hf-BTB nanosheets, though there were a few non-phosphopeptide peaks in the spectrum (Figure S13, parts a and b). Moreover, one of the monophosphopeptides and its dephosphopeptide peaks could still be observed even with BSA/β-casein = 1000:1 (Figure S13c), which indicated

that the 2-D Hf-BTB nanosheets could demonstrate better selectivity for monophosphopeptides in complex biological samples. In order to confirm the feasibility of 2-D Hf-BTB nanosheets for enrichment of monophosphopeptides in complex biological samples, the nonfat milk digest and endogenous phosphopeptides from human saliva were analyzed, respectively. For nonfat milk, before enrichment, a few non-phosphopeptides with high intensities were detected, though one phosphopeptide peak and its dephosphopeptide peak with low S/N ratio could be observed (Figure S14a). After enrichment with 2-D Hf-BTB nanosheets, the signal intensities of phosphopeptides were significantly increased and four monophosphopeptides and their respective dephosphopeptides were identified (Figure S14b and Table S2). In addition, human saliva was also used as a sample to further verify the enrichment efficiency of 2-D Hf-BTB nanosheets (Figure S15). Seventeen phosphopeptides were enriched by 2-D Hf-BTB nanosheets and identified according to the published results,36−42 while two more peptides were also detected and had not been reported previously (Table S3). We speculated that these peaks could possibly be phosphopeptides due to the difference of enrichment materials between our method and references. These results exhibited that the phosphopeptides enrichment of 2-D Hf-BTB nanosheets was feasible in complex biological samples, indicating 9098

DOI: 10.1021/acs.analchem.9b01581 Anal. Chem. 2019, 91, 9093−9101

Article

Analytical Chemistry

Figure 5. WebLogos of phosphorylated serine sites from the tryptic digests of mouse brain cortical samples lysate: (a) the Akt knockout mouse sample and (b) the control sample. (c) The schematic structure of 2-D Hf-BTB nanosheets with metallic clusters (orange) and hydrophobic ligands (gray).

with the control samples (CD2O, heavy-labeled) are shown in Figure 4, parts a and b, respectively, by setting the m/z = 1647.4 peak intensity of control samples as 100%. The average values for healthy adults and inflammation patients of the isotope cluster peak area of human serum phosphopeptides HS1, HS2, HS3, and HS4 were calculated and compared (Figure 4, parts c, and d, and Table S4). There was not a significant difference for the four phosphopeptides between the healthy serum and the control sera, while the significant difference for the four phosphopeptides was observed in patient serum compared to the control sera. All of the four phosphopeptides in the serum were observed to be upregulated from patients with inflammation, especially the phosphopeptide HS4. As shown in the logarithmic ratios of the isotope clusters areas, the phosphopeptide HS4 underwent a 6.6-fold upregulation (Figure 4d). The above results could possibly show that the intensity of endogenous phosphopeptides from the healthy serum and the patients’ serum with inflammation were different. A larger number of samples are needed in further study. Application of 2-D Hf-BTB Nanosheets in the Alzheimer’s Disease-Related Mouse Brain Samples. Inspired by the enrichment capability toward monophosphopeptides of 2-D Hf-BTB nanosheets, we further evaluated its performance in the Alzheimer’s disease-related mouse brain samples. The target of Akt is glycogen synthase kinase-3β (GSK-3β), which is inhibited by phosphorylation of its Ser9 residue. The knockout of Akt will activate GSK-3β and cause specific hyperphosphorylation of τ that forms neurofibrillary tangles as well as production and deposition of β-amyloid (Aβ).49 The cortical samples of a specific conditional knockout (Akt) mouse and an age-matched littermate control mouse were analyzed. A total of the sample of 1 mg was enriched by 2-D Hf-BTB nanosheets in the optimized enrichment condition, respectively, and the eluent was analyzed with reversed-phase liquid chromatography−electrospray ionization tandem mass spectrometry (RPLC−ESI-MS/MS). After comparing the MS data with the UniProt database, we identified 78 and 153 phosphopeptides as well as a total of 54 and 104 phosphorylated proteins from the control cortical sample and Akt knockout cortical sample, respectively. It was worth noting that all phosphopeptides from the control mouse

that the 2-D Hf-BTB nanosheets had great potential in real applications, such as the quantification of biomarkers and phosphoproteomics. Relative Quantification of Human Serum Phosphopeptides. Human serum is one of the most important and accessible body fluids in clinical assays, and it contains a large number of informative biomolecules, such as proteins and peptides. The concentrations of endogenous phosphopeptides in human serum are potentially related to the occurrence of various diseases.43−45 The expression of phosphopeptides is modulated by a series of endogenous proteases which perform highly distinct activities between healthy and diseased states. For relative quantification of serum phosphopeptides from healthy and patient adults, an in situ isotope dimethyl labeling procedure was performed after pretreatment with 2-D Hf-BTB nanosheets. Stable isotope dimethyl labeling is a reliable method for quantitative proteomics, which generates the mass difference of 28 and 32 Da through the reductive amination with CH2 O and CD 2O, respectively (Table S4). To demonstrate the reliability of in situ isotope dimethyl labeling for relative quantification of phosphopeptides, the ratios of normal and deuterated isotope peak areas of β1 from bovine βcasein digest were first used to plot a calibration curve versus the ratios of sample volume. The volume ratios varied from 0.125:1 to 8:1, and the linear fitting resulted an equation with the slope of 1.06 (R2 = 0.994), confirming the solidity of relative quantification in this range (Figure S16). The dysregulation of phosphopeptide expression in serum has been reported in leukemia and type 2 diabetes.34,46,47 However, to the best of our knowledge, the serum phosphopeptides in patients with inflammation were not investigated as potential biomarkers in the practical medical evaluation although the control of inflammation was recently revealed via the phosphorylation and dephosphorylation at specific serine residue.48 Herein, the sera of three patients with anal canal inflammation and three healthy adults were first collected. Then, the inflammation and healthy samples were enriched by 2-D Hf-BTB nanosheets and labeled in situ with CH2O, while mixed sera from eight volunteers was labeled with CD2O as a control. The normalized MS spectra with serum phosphopeptide HS4 from the serum of healthy adults and inflammation patients (CH2O, light-labeled) associated 9099

DOI: 10.1021/acs.analchem.9b01581 Anal. Chem. 2019, 91, 9093−9101

Article

Analytical Chemistry sample and 99.3% of phosphopeptides from the Akt knockout mouse sample were monophosphopeptides. The WebLogos of phosphorylated serine sites were first plotted to reflect the local domain information, which indicated the recognition feature of 2-D Hf-BTB nanosheets. The overall probability for hydrophobic peptides, such as proline (P), leucine (L), and alanine (A), around the serine (S) sites was greatly increased with 2-D Hf-BTB nanosheets compared to previous less-hydrophobic enrichment materials (Figure 5, parts a and b).34 Especially, more than half of the adjacent sites to serine on both locations of 1 and −1 were occupied by the hydrophobic peptides. This was probably due to the strong hydrophobicity around the metallic clusters and significant distance between the adjacent metallic clusters in 2-D Hf-BTB nanosheets (Figure 5c). This hypothesis is further supported by the increased probability of hydrophilic peptides, such as serine and glutamic acid (E) at locations of 2 and −2, which probably were bonded to the metal clusters at the outer circle (Figure 5c). This domain analysis revealed the strong correlation between the features of 2-D Hf-BTB nanosheets and their excellent selectivity toward monophosphopeptides. The control of hydrophilic/hydrophobic local microenvironments was also vital in other MOF applications, in biomimetic catalysis,50 bioalcohol adsorption,51 and oil capture.52 The enrichment feature of 2-D Hf-BTB nanosheets was also very practical for the analysis of mouse brain cortical samples. The 2-D Hf-BTB nanosheets identified much higher percentage of monophosphopeptides than other materials, such as Ti-based MOF nanosheets, Fe3O4@hYPO4 microspheres, and rGTZ nanosheets,34,53,54 in the mouse brain samples (Table S5). Moreover, one phosphorylated protein lunapark (endoplasmic reticulum junction formation protein), which was involved in the development of limbs and central nervous system in the mouse,55 was detected with the 2-D HfBTB nanosheets enrichment, while this protein could not be detected with other materials. The number of detected phosphorylated proteins in the knockout mouse sample was nearly twice as many as the control sample; this is due to the disorder of phosphorylation in mouse brain neocortex after Akt knockout, indicating that the protein phosphorylation was closely related to the function of the brain.56 At the same time, after Akt knockout, the hyperphosphorylation of τ protein such as microtubuleassociated protein τ (MAPT/τ) was observed and found to be strongly related to Alzheimer’s disease.57 In our Akt knockout mouse and control mouse, after enrichment by 2-D Hf-BTB nanosheets, there was one phosphorylated protein and one monophosphopeptide in the Akt knockout sample, but there were no relevant phosphorylated proteins and phosphopeptides in the control sample. This data was consistent with previous study.57 The detailed information on the identified phosphopeptides and phosphorylated proteins from the control cortical sample (Tables S6 and S7) and the Akt knockout cortical sample (Tables S8 and S9) is listed in the Supporting Information. This result revealed that the 2-D Hf-BTB nanosheets exhibited excellent selectivity toward monophosphopeptides from real biological samples, suggesting it was promising for the proteomics analysis of monophosphopeptides.

excellent sensitivity and selectivity. Relative quantification for human serum and proteomics analysis for Akt knockout cortical samples were analyzed with 2-D Hf-BTB nanosheets, indicating the strong potential in clinical analysis and phosphoproteome research.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.9b01581.

■

Additional experimental details, characterization, mass spectra, contact angles, and detailed information of phosphopeptides and phosphorylated proteins (PDF)

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone/Fax: +86-2585891952. ORCID

Zhi-Yuan Gu: 0000-0002-6245-4759 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work is supported the National Natural Science Foundation of China (NSFC Grant No. 21505076), the Young Elite Scholar Support (YESS) Program from CAST, the Program of Jiangsu Specially Appointed Professor, the NSF of Jiangsu Province of China, the Innovation Team Program of Jiangsu Province of China, and the Priority Academic Program Development of Jiangsu Higher Education Institutions. W.S. acknowledges financial support from the National Natural Science Foundation of China (81473391 and 81303277). Thanks are given for the MS technical support from Dr. X. Chen of the Wuhan Institute of Biotechnology.

■

REFERENCES

(1) Ardito, F.; Giuliani, M.; Perrone, D.; Troiano, G.; Lo Muzio, L. Int. J. Mol. Med. 2017, 40, 271−280. (2) Narasimha, A. M.; Kaulich, M.; Shapiro, G. S.; Choi, Y. J.; Sicinski, P.; Dowdy, S. F. eLife 2014, 3, e02872. (3) Martin-Garrido, A.; Biesiadecki, B. J.; Salhi, H. E.; Shaifta, Y.; dos Remedios, C. G.; Ayaz-Guner, S.; Cai, W.; Ge, Y.; Avkiran, M.; Kentish, J. C. J. Biol. Chem. 2018, 293, 8588−8599. (4) Zeng, Q.; Lagunoff, D.; Masaracchia, R.; Goeckeler, Z.; Cote, G.; Wysolmerski, R. J. Cell Sci. 2000, 113, 471−482. (5) King, C. A. J. Virol. 2013, 87, 8779−8791. (6) Vicente-Manzanares, M.; Horwitz, A. R. Biochem. Biophys. Res. Commun. 2010, 402, 537−542. (7) Ficarro, S. B.; Adelmant, G.; Tomar, M. N.; Zhang, Y.; Cheng, V. J.; Marto, J. A. Anal. Chem. 2009, 81, 4566−4575. (8) He, X.-M.; Chen, X.; Zhu, G.-T.; Wang, Q.; Yuan, B.-F.; Feng, Y.-Q. ACS Appl. Mater. Interfaces 2015, 7, 17356−17362. (9) Tape, C. J.; Worboys, J. D.; Sinclair, J.; Gourlay, R.; Vogt, J.; McMahon, K. M.; Trost, M.; Lauffenburger, D. A.; Lamont, D. J.; Jørgensen, C. Anal. Chem. 2014, 86, 10296−10302. (10) Zhou, H.; Ye, M.; Dong, J.; Corradini, E.; Cristobal, A.; Heck, A. J. R.; Zou, H.; Mohammed, S. Nat. Protoc. 2013, 8, 461−480. (11) Leitner, A.; Sturm, M.; Hudecz, O.; Mazanek, M.; Smått, J.-H.; Lindén, M.; Lindner, W.; Mechtler, K. Anal. Chem. 2010, 82, 2726− 2733.

■

CONCLUSIONS In summary, 2-D Hf-BTB nanosheets exhibited the unique enrichment selectivity toward monophosphopeptides with the 9100

DOI: 10.1021/acs.analchem.9b01581 Anal. Chem. 2019, 91, 9093−9101

Article

Analytical Chemistry

(44) Min, Q.; Chen, X.; Zhang, X.; Zhu, J.-J. Lab Chip 2013, 13, 3853−3861. (45) Zhai, G.; Wu, X.; Luo, Q.; Wu, K.; Zhao, Y.; Liu, J.; Xiong, S.; Feng, Y.-Q.; Yang, L.; Wang, F. Talanta 2014, 125, 411−417. (46) Chen, X.; Li, S.; Zhang, X.; Min, Q.; Zhu, J.-J. Nanoscale 2015, 7, 5815−5825. (47) Jiang, D.; Li, X.; Jia, Q. ACS Sustainable Chem. Eng. 2019, 7, 421−429. (48) Clark, A. R.; Dean, J. L. E. Biochem. Soc. Trans. 2016, 44, 1321− 1337. (49) Mudher, A.; Lovestone, S. Trends Neurosci. 2002, 25, 22−26. (50) Shi, W.; Cao, L.; Zhang, H.; Zhou, X.; An, B.; Lin, Z.; Dai, R.; Li, J.; Wang, C.; Lin, W. Angew. Chem., Int. Ed. 2017, 56, 9704−9709. (51) Liu, C.; Liu, Q.; Huang, A. Chem. Commun. 2016, 52, 3400− 3402. (52) Kim, D.; Kim, D. W.; Buyukcakir, O.; Kim, M.-K.; Polychronopoulou, K.; Coskun, A. Adv. Funct. Mater. 2017, 27, 1700706. (53) Cheng, G.; Liu, Y.-L.; Wang, Z.-G.; Li, S.-M.; Zhang, J.-L.; Ni, J.-Z. J. Mater. Chem. B 2013, 1, 3661−3669. (54) Huang, X.; Wang, J.; Liu, C.; Guo, T.; Wang, S. J. Mater. Chem. B 2015, 3, 2505−2515. (55) Spitz, F.; Gonzalez, F.; Duboule, D. Cell 2003, 113, 405−417. (56) Rickle, A.; Bogdanovic, N.; Volkman, I.; Winblad, B.; Ravid, R.; Cowburn, R. F. NeuroReport 2004, 15, 955−959. (57) Gong, C. X.; Iqbal, K. Curr. Med. Chem. 2008, 15, 2321−2328.

(12) Tang, L. A. L.; Wang, J.; Lim, T. K.; Bi, X.; Lee, W. C.; Lin, Q.; Chang, Y.-T.; Lim, C. T.; Loh, K. P. Anal. Chem. 2012, 84, 6693− 6700. (13) Wolschin, F.; Wienkoop, S.; Weckwerth, W. Proteomics 2005, 5, 4389−4397. (14) Yan, J.; Li, X.; Cheng, S.; Ke, Y.; Liang, X. Chem. Commun. 2009, 2929−2931. (15) Posewitz, M. C.; Tempst, P. Anal. Chem. 1999, 71, 2883−2892. (16) Thingholm, T. E.; Jørgensen, T. J. D.; Jensen, O. N.; Larsen, M. R. Nat. Protoc. 2006, 1, 1929−1935. (17) Zhong, H.; Xiao, X.; Zheng, S.; Zhang, W.; Ding, M.; Jiang, H.; Huang, L.; Kang, J. Nat. Commun. 2013, 4, 1656. (18) Qing, G.; Lu, Q.; Li, X.; Liu, J.; Ye, M.; Liang, X.; Sun, T. Nat. Commun. 2017, 8, 461. (19) Yang, S.-S.; Yu, H.-X.; Wang, Z.-Z.; Liu, H.-L.; Zhang, H.; Yu, X.; Shang, W.; Chen, G.-Q.; Gu, Z.-Y. Chem. - Eur. J. 2018, 24, 2109− 2116. (20) Liu, Z.; Wang, F.; Chen, J.; Zhou, Y.; Zou, H. J. Chromatogr. A 2016, 1461, 35−41. (21) Peng, J.; Niu, H.; Zhang, H.; Yao, Y.; Zhao, X.; Zhou, X.; Wan, L.; Kang, X.; Wu, R. A. ACS Appl. Mater. Interfaces 2018, 10, 32613− 32621. (22) Gu, Z.-Y.; Wang, G.; Yan, X.-P. Anal. Chem. 2010, 82, 1365− 1370. (23) Gu, Z.-Y.; Yan, X.-P. Angew. Chem., Int. Ed. 2010, 49, 1477− 1480. (24) James, S. L. Chem. Soc. Rev. 2003, 32, 276−288. (25) Murray, L. J.; Dincǎ, M.; Long, J. R. Chem. Soc. Rev. 2009, 38, 1294−1314. (26) Pagis, C.; Ferbinteanu, M.; Rothenberg, G.; Tanase, S. ACS Catal. 2016, 6, 6063−6072. (27) Rodenas, T.; Luz, I.; Prieto, G.; Seoane, B.; Miro, H.; Corma, A.; Kapteijn, F.; Llabrés i Xamena, F. X.; Gascon, J. Nat. Mater. 2015, 14, 48−55. (28) Zhou, H.-C.; Long, J. R.; Yaghi, O. M. Chem. Rev. 2012, 112, 673−674. (29) Liu, H.-L.; Chang, Y.-J.; Fan, T.; Gu, Z.-Y. Chem. Commun. 2016, 52, 12984−12987. (30) Rodenas, T.; Luz, I.; Prieto, G.; Seoane, B.; Miro, H.; Corma, A.; Kapteijn, F.; Llabrés i Xamena, F. X.; Gascon, J. Nat. Mater. 2015, 14, 48−55. (31) Xu, M.; Yuan, S.; Chen, X.-Y.; Chang, Y.-J.; Day, G.; Gu, Z.-Y.; Zhou, H.-C. J. Am. Chem. Soc. 2017, 139, 8312−8319. (32) Cao, L.; Lin, Z.; Shi, W.; Wang, Z.; Zhang, C.; Hu, X.; Wang, C.; Lin, W. J. Am. Chem. Soc. 2017, 139, 7020−7029. (33) Dong, J.; Zhang, K.; Li, X.; Qian, Y.; Zhu, H.; Yuan, D.; Xu, Q.H.; Jiang, J.; Zhao, D. Nat. Commun. 2017, 8, 1142. (34) Yang, S.-S.; Chang, Y.-J.; Zhang, H.; Yu, X.; Shang, W.; Chen, G.-Q.; Chen, D. D. Y.; Gu, Z.-Y. Anal. Chem. 2018, 90, 13796−13805. (35) Cao, L.; Lin, Z.; Peng, F.; Wang, W.; Huang, R.; Wang, C.; Yan, J.; Liang, J.; Zhang, Z.; Zhang, T.; Long, L.; Sun, J.; Lin, W. Angew. Chem., Int. Ed. 2016, 55, 4962−4966. (36) Lin, H.; Chen, H.; Shao, X.; Deng, C. Microchim. Acta 2018, 185, 562. (37) Lin, H.; Deng, C. Proteomics 2016, 16, 2733−2741. (38) Long, X.-Y.; Li, J.-Y.; Sheng, D.; Lian, H.-Z. RSC Adv. 2016, 6, 96210−96222. (39) Peng, J.; Zhang, H.; Li, X.; Liu, S.; Zhao, X.; Wu, J.; Kang, X.; Qin, H.; Pan, Z.; Wu, R. a. ACS Appl. Mater. Interfaces 2016, 8, 35012−35020. (40) Su, J.; He, X.; Chen, L.; Zhang, Y. ACS Sustainable Chem. Eng. 2018, 6, 2188−2196. (41) Sun, N.; Wang, J.; Yao, J.; Chen, H.; Deng, C. Microchim. Acta 2019, 186, 159. (42) Yao, J.; Sun, N.; Wang, J.; Xie, Y.; Deng, C.; Zhang, X. Proteomics 2017, 17, 1600320. (43) Hu, L.; Zhou, H.; Li, Y.; Sun, S.; Guo, L.; Ye, M.; Tian, X.; Gu, J.; Yang, S.; Zou, H. Anal. Chem. 2009, 81, 94−104. 9101

DOI: 10.1021/acs.analchem.9b01581 Anal. Chem. 2019, 91, 9093−9101