Materials and reagents

Hexadecyltrimethylammonium chloride (≥98%, CTAC), tetraethyl orthosilicate (98%, TEOS), 1,6-diisocyanatohexane (≥99.0%), tridecane (≥99%), ammonium hydroxide solution (28 wt%, NH₃·H₂O), ammonium fluoride (≥98%, NH₄F), 3- aminopropyltriethoxysilane (99%, APS), formaldehyde (37 wt% in water), sodium cyanoborohydride (NaBH₃CN), sodium cyanoborodeuteride (NaBD₃CN), trypsin (from bovine pancreas), protease inhibitor cocktail, formic acid (FA), urea, trifluoroacetic acid (TFA), dithiothreitol (DTT), iodoacetamide (IAA), and ammonium acetate were purchased from Sigma-Aldrich (St. Louis, MO, USA). Phosphatase inhibitor cocktail was ordered from Thermo Fisher Scientific (Rockford, USA). ¹³C-D₂-formaldehyde was purchased from Cambridge Isotope Laboratories, Inc. (Andover, USA). Dry pyridine was provided by Aladdin (99.8%, Shanghai, China). Toluene and chloroform (≥99.0%) were bought from Kaixin (Tianjin, China), and toluene was purified by refluxing (110 °C) over sodium and distilling (136 °C). Acetone (≥99.5%) and ethanol (≥99.8%) were obtained from Kermel (Tianjin, China). Acetonitrile (≥99.9%, ACN) was purchased from Merck (Darmstadt, Germany). Deionized water was used throughout the experiments (Millipore, Milford, MA, USA). Other reagents were all analytical grade. All the standard non-phosphorylated peptides were purchased from ChinaPeptides (Shanghai, China).

Characterizations

Hydrogen NMR (¹H-NMR) was measured by a bruker AVANCE III HD 400 MHz spectrometer at room temperature (Daltonios, Germany). Chemical shifts were recorded in parts per million (p.p.m.) using residual solvent peaks as internal references [CDCl₃ δ: 7.26 (1H)]. Molecular mass was detected by orbitrap LTQ (Thermo, USA). TEM images were collected on a transmission electron microscope operated at 120 kV (JEM-2000EX, JEOL, Tokyo, Japan). SEM images were collected on a Zeiss Merlin FEG-SEM instrument. TGA was performed under an air atmosphere with a heating rate of 10 °C/min using a Netzsch STA449F3 thermogravimetric analyzer (Netzsch, Bavaria, Germany). Zeta potential was measured using the Nano-ZS90 zetasizer (Malvern, UK). X-ray photoelectron spectroscopy (XPS) characterization was carried out on an Thermo ESCALAB250Xi spectrometer with Al Kα radiation as the X-ray source (Thermo, Waltham, USA). The nitrogen adsorption and desorption isotherms were measured by QuadrasorbSI (Quadraorb, Wisconsin, USA). The Brunauer–Emmett–Teller (BET) method was used to calculate the specific surface areas with adsorption data in a relative pressure range from 0.049 to 0.252. Contact angle was measured by DSA100 (Krüss, Hamburg, Germany). MALDI-TOF spectra were recorded on Bruker Ultraflex III MALDI-TOF/TOF MS (Daltonios, Germany) and MALDI-TOF/TOF 5800 System (AB SCIEX, Foster City, CA) under positive reflection mode. ITC experiments were carried out on a MicroCal iTC200 instrument (Malvern,Sweden).

Synthesis of Dpa (1)

According to the literature⁵⁵, 2,2′-dipicolylamine (3.37 g, 16.9 mmol) and paraformaldehyde (0.813 g, 27.1 mmol) solution (water/i-PrOH = 5:3 (v/v), 48 mL) was adjusted to pH 8.0 by 1 N HCl aq. After stirring at 80 °C for 30 min, Boc-L-tyrosine-OMe (2.0 g, 6.77 mmol) was added, and the mixture was refluxed at 110 °C for 13 h. Then, the mixture was cooled to room temperature, and i-PrOH was removed by evaporation. After cooling on an ice-bath, the solution was removed by decantation, and the precipitated viscous oil was dissolved in 50 mL of AcOEt. The solution was washed with saturated NaHCO₃ and brine followed by drying over Na₂SO₄. After removal of the solvent in vacuo, the residue was purified by column chromatography (SiO₂, CH₂Cl₂/MeOH/TEAB aq = 30/1/0.1 (v/v/v)) to give 1 (3.30 g, 68%) as a yellow oil. ¹H-NMR (400 MHz, CDCl₃): δ 8.53 (4H, d, J = 6 Hz), 7.62 (4H, td, J = 7.5 Hz), 7.47 (4H, d, J = 9 Hz), 7.13 (4H, td, J = 6 Hz), 6.99 (2H, s), 5.23 (1H, d, J = 6 Hz), 4.49–4.51 (1H, m), 3.85 (8H, s), 3.76 (4H, s), 3.59–3.61(3H, s), 2.99 (2H, s), 1.35 (9H, s). ESI-MS m/z for C₄₁H₄₇N₇O₅ [M+H]⁺ 718.35 and [M+Na]⁺ 740.20 (Supplementary Figs. 15 and  16).

Synthesis of amino terminal Dpa (2)

In a typical reaction, TFA (20 mL) was added dropwise to a stirring solution of 1 (3.30 g, 4.60 mmol) in anhydrous CH₂Cl₂ (20 mL) on ice-bath, and the solution was stirred at room temperature for 2 h. After concentration in vacuo, the residue was dissolved in water and alkalized with ammonium hydroxide solution on ice-bath. The resulting mixture was extracted with CH₂Cl₂. The combined organic layers were washed with brine and dried over Na₂SO₄. After removal of the solvent in vacuo, 2 (2.86 g, 97%) was obtained as a pale viscous oil. ¹H-NMR (400 MHz, CDCl₃): δ 8.53 (4H, d, J = 6 Hz), 7.61 (4H, td, J = 6 Hz), 7.46 (4H, d, J = 9 Hz), 7.13 (4H, td, J = 6 Hz), 7.03 (2H, s), 3.89 (8H, s), 3.81 (4H, s), 3.75–3.70 (4H, m), 3.68 (3H, S), 3.01–2.96 (1H, m), 2.80–2.75 (1H, m). ESI-MS m/z for C₄₁H₄₅N₇O₆ [M+H]⁺ 618.48, [M+Na]⁺ 640.34 (Supplementary Figs. 15 and  17).

Synthesis of carboxyl terminal Dpa (3)

In a typtical reaction, glutaric anhydride (0.635 g, 5.57 mmol) was added to a solution of 2 (2.86 g, 4.64 mmol) in anhydrous CH₂Cl₂ (110 mL). The mixture was stirred and refluxed at 50 °C overnight. After removing the solvent by evaporation, 3 was obtained as a light yellow viscous oil. ¹H-NMR (400 MHz, CDCl₃): δ 8.53 (4H, d, J = 6 Hz), 7.61 (4H, td, J = 7.5 Hz), 7.46 (4H, d, J = 9 Hz), 7.13 (4H, td, J = 6 Hz), 7.03 (2H, s), 3.86 (8H, s), 3.79 (4H, s), 3.70–3.65 (4H, m), 3.01–2.96 (1H, m), 2.80–2.75 (1H, m). ESI-MS m/z for C₄₁H₄₅N₇O₆ [M+H]⁺ 732.53, [M+Na]⁺ 754.41 (Supplementary Figs. 15 and  18).

Preparation of non-porous silica

Non-porous microspheres were prepared by a seed-growth approach⁵⁶. It contains three steps. Firstly, hydrolysis solution, containing 6.7 mL of NH₃·H₂O, 5.1 mL of H₂O, and 70 mL of CH₃CH₂OH, was placed in water bath at 22 °C, and 4.0 mL of TEOS was added. The mixtures were reacted for 40 min under constant mechanical stirring. After the temperature was increased to 55 °C, 0.64 mL of H₂O and 4.0 mL of pre-heated TEOS were added to the mixtures and reacted for 40 min (denoted as growth process), and the growth process was repeated another three times. The obtained suspension was denoted as seeds and divided into four pieces. Secondly, we used 1 piece seed replaced equal volume hydrolysis solution, and hydrolysis solution containing seed was heated to 55 °C, and the growth process was repeated three times. Afterward, the hydrolysis solution was added to the mixtures, and growth process was repeated four times. The obtained particles were centrifuged at 4000 × g for 5 min, with 95% (v/v%) CH₃CH₂OH and H₂O. The obtained particles were divided to four pieces which were denoted as large particles. Finally, one piece of large particle was re-dispersed in the hydrolysis solution at 55 °C, and the growth process was repeated three times. Non-porous silica was obtained by centrifugation at 3300 × g for 3 min using 95% (v/v%) CH₃CH₂OH and H₂O.

Preparation of sub-2-μm core–shell silica

Firstly, 1.0 g of above non-porous silica with were added to the growth system which included 100 mL of H₂O, 5.8 mL of tridecane, and 1.0 g of CTAC. Then, 26 mg of NH₄F and 6.0 mL of NH₃·H₂O were added and the mixtures were carried out for 24 h at 90 °C under stirring. The product was obtained by centrifugation at 3300 × g for 3 min. Finally, the produce was dried at 65 °C for 2 h and calcined with a Ceramic Fibre Muffle Furnace (Michem, Beijing, China) at 550 °C for 6 h in air. The heating procedure was from 25 to 550 °C at a ramp rate of 1 °C/min. The pore size was enlarged by etching in 5 M HCl at 120 °C for 12 h.

Amino-functionalized sub-2-μm superficially porous silica

In a typical reaction, 0.6 g of sub-2-μm superficially porous silica particles and 2.21 mL of APTES were added to 30 mL of anhydrous toluene, and the mixtures were stirred and refluxed at 110 °C for 24 h. The obtained particles were centrifuged at 4000 × g for 5 min and sequentially washed with toluene, CH₃OH, and CH₃COCH₃.

Preparation of SiO₂@DpaZn

In a typical reaction, a solution of carboxyl terminal Dpa 3 (750 mg, 1.1 mmol), amino-functionalized sub-2-μm superficially porous silica particles (600 mg), EDC (251 mg, 1.3 mmol), HOBt·H₂O (198 mg, 1.3 mmol), and DIEA (464 mg, 3.57 mmol) in dry DMF (21 mL) was stirred at room temperature for 3 days. After filtration, the particles were subjected to sequential washing with toluene, methanol, and acetone. Then, bis-Zn (GA-DpaTyr-OMe)-sub-2-μm superficially porous silica particles (570 mg) were dissolved in water/MeOH (1:1, v/v, 10 mL), and an water/MeOH solution (1:1, v/v, 20 mL) of Zn(NO₃)₂·6H₂O (2.97 g, 10 mmol) was added drop-wise to the silica particles suspension. After stirring at room temperature for 72 h, the obtained particles were centrifuged at 4000 × g for 5 min, and sequentially washed with H₂O, CH₃CH₂OH, and acetone.

NMR study

¹H-NMR and ³¹P-NMR studies of DpaZn with pho-His were carried out on a Brucker ADVANCE III (500 MHz) at 25 °C. Samples were prepared using pho-His in D₂O in the absence or presence of DpaZn. TSP was used as an internal standard, and the observed peaks were individually assigned by the COSY experiments. In ³¹P-NMR experiment, 85% H₃PO₄ in H₂O was used as an external standard.

Enrichment of N-phosphopeptide from BSA digests

Five micrograms of standard N-phosphopeptide (TS_pHYSIMAR) and 500 μg of BSA digests were mixed and dissolved in 200 μL of 80% ACN (20 mM HEPES, pH 7.7). After incubation with 200 μg of SiO₂@DpaZn at room temperature for 20 min, total volume of 400 μL 80% ACN, 400 μL 65% ACN, and 400 μL 0.001% NH₃·H₂O comprised washing steps, respectively. Phosphopeptides were eluted with 100 μL of 0.1% NH₃·H₂O and analyzed by MALDI-TOF MS.

Adsorption kinetics of SiO₂@DpaZn

A total of 0.4 mg of SiO₂@DpaZn was incubated with 1.6 mL of 80% ACN (20 mM HEPES, pH 7.7) containing 10 μg of O-phosphopeptide GK8 at room temperature. After incubation for 2, 5, 10, 15, 20, and 30 min, respectively, 40 μL of the mixture was centrifuged, and the concentration of GK8 was measured by UV analysis. The adsorption capacity (Q) was calculated as follows: Q = (C₀−C) V/m, where m is the mass (mg) of SiO₂@DpaZn, V (mL) the volume of incubation, C₀ (mg/mL) and C (mg/mL) the concentrations of GK8 in the initial solution and the supernatant after the adsorption, respectively. Three sets of parallel experiments were conducted simultaneously.

On-tip enrichment of phosphopeptides by SiO₂@DpaZn tips

A total of 1 mg of SiO₂@DpaZn in 80% ACN (20 mM HEPES, pH 7.7) was packed into a 20-μL micropipette tip by centrifugation at 5400 × g for 10 min, and then equilibrated with 10 μL of 80% ACN (20 mM HEPES, pH 7.7). Sample loading, washing, and elution were realized by centrifugation at 5400 × g for 5, 12, and 5 min, respectively. Dissolved in 20 μL of 80% ACN, peptides were load onto SiO₂@DpaZn tips. After washing three times with 80 μL of 50% ACN (20 mM HEPES, pH 7.7), 80 μL of 0.001% NH₃·H₂O, and 80 μL of 0.005% NH₃·H₂O orderly, samples were eluted with 10 μL of 1% NH₃·H₂O. Eluates were analyzed by MALDI-TOF MS.

Preparation of protein extracts from E. coli

E. coli (strain K12) was cultured in Luria–Bertani medium (LB, 5 g/L yeast extracts, 10 g/L NaCl, and 10 g/L tryptone) at 37 °C overnight or in M9 minimal medium (consisting of 6 g/L Na₂HPO₄, 3 g/L KH₂PO₄, 0.5 g/L NaCl, 1 g/L NH₄Cl, supplemented with either additional 0.5% (w/v) glucose or 0.5% glycerol) at 37 °C to exponential phase and stationary phase, and were harvested by centrifugation at 4000 × g for 2 min at 4 °C. Cells were washed three times with cold PBS and resuspended in lysis buffer (8 M urea, 10 mM PBS, 1% (v/v) protease inhibitor cocktail, 10 mM EDTA, and 1% (v/v) phosphatase inhibitor cocktail). After ultrasonication for 120 s (20 s interval every 10 s) in an ice bath, insoluble portions were separated from the soluble ones by centrifugation at 16,000 × g for 30 min at 4 °C, and protein concentrations were determined using BCA assay.

Tryptic digestion of E. coli

E. coli protein extracts were denatured at 95 °C for 1 min. Then, the disulfide bond was reduced with 25 mM DTT at 37 °C for 75 min and the resulting cysteine residues were alkylated with 50 mM IAA at room temperature in the dark for 30 min. The excess IAA was quenched with 50 mM DTT. The resulting mixture was diluted with 20 mM ammonium bicarbonate to the final concentration of urea <1 M. Afterward, trypsin (1:3) was added and incubated at 37 °C for 1.5 h. The resultant peptides were desalted by RP-LC.

Preparation and digestion of protein extracts from mammalian cells

The HeLa and HEPG2 cells were cultured in DMEM medium containing 10% FBS and 1% PBS in a humidified incubator of 5% CO₂ and 95% air at 37 °C. All cells reaching 80% confluence were detached with 0.25% trypsin/EDTA and centrifugated (540 × g, 5 min) to collect the cells. Cells were washed three times with cold PBS and resuspended in lysis buffer (8 M urea, 10 mM PBS, 1% (v/v) protease inhibitor cocktail, 10 mM EDTA, and 1% (v/v) phosphatase inhibitor cocktail). After ultrasonication for 120 s (20 s interval every 10 s) in an ice bath, insoluble portions were separated from the soluble ones by centrifugation at 16,000 × g for 30 min at 4 °C, and protein concentrations were determined using BCA assay. After treating with DTT and IAA, the proteins were precipitated with CHCl₃/CH₃OH. Afterward, trypsin (1:3) was added and incubated at 37 °C for 1.5 h.

Enrichment of E. coli and HeLa cells phosphopeptides by SiO₂@DpaZn tips

A total of 36 mg of SiO₂@DpaZn in 80% ACN was packed into 200-μL micropipette tips equally by centrifugation at 5400 × g for 10 min. Disolved in 200 μL of 80% ACN, 12 mg of E. coli digests were loaded onto SiO₂@DpaZn tips by centrifugation at 5400 × g for 10 min. Washing steps were carried out with 100 μL of 50% ACN, 100 μL of 30% ACN, 100 μL of 0.001% NH₃·H₂O, and 80 μL of 0.005% NH₃·H₂O, respectively, and phosphopeptides were eluted with 20 μL of 1% NH₃·H₂O for two times. Elutes was mixed up for the following high pH-RP fractionation. For mammalian cell phosphopeptides, 30 mg of SiO₂@DpaZn was used for 6 mg of cell lysate digests enrichment.

MALDI-TOF-MS

One microliter of the sample was deposited on the sample plate and dried, then 1 µL of 20 mg/mL DHB substrate (in 60% (v/v) acetonitrile and 0.1% (v/v) TFA aqueous solution) was introduced. MS spectra were acquired on an Ultraflex III TOF/TOF mass spectrometer (Bruker Daltonik GmbH, Germany) equipped with a 200-Hz smart-beam 1 laser at 355 nm or a MALDI-TOF/TOF 5800 System (AB SCIEX, Foster City, CA) quipped with a 1-kHz optibeam™ on-axis laser in the positive reflection mode, and controlled by the Flex Control 2.4 software and Data Explorer software 4.11. Flex Analysis software 3.3 and TOF/TOF Series Explorer software 4.1 were used to analyze MALDI-TOF MS data. The threshold for peak acceptance was a signal-to-noise ratio of 6.

NanoRPLC-ESI-MS/MS methods

For E. coli lysate digests analysis, each fraction of the enriched phosphopeptides was automatically loaded onto a RP trap column (150 μm i.d. × 5 cm) and separated by a C18 capillary column (150 μm i.d. × 15 cm). The trap column and the analytical column were both packed in-house with 5 μm, 100 Å Venusil XBP C18 silica particles. In order to study the phosphorylated peptides in E. coli cultured in LB medium, two mobile phases (A: 2% (v/v) ACN with 0.1% (v/v) FA and B: 98% (v/v) ACN with 0.1% (v/v) FA) were used to generate a 75-min gradient with the flow rate of 500 nL/min (45 min from 7 to 25% B, 20 min from 25 to 40% B, 5 min from 40 to 80% B, and 5 min to kept at 80% B). A Q-Exactive mass spectrometer (Thermo-Fisher, San Jose, CA, USA) was operated in full scan (70,000 FWHM, 350–1800 m/z) and product scan (17,500 FWHM, 100–1000 m/z) modes at positive ion mode. The electro-spray voltage was 2.3 kV, and the heated capillary temperature was 270 °C. All the mass spectra were recorded with Xcalibur software (version 3.1, Thermo Fisher Scientific, USA). MS/MS spectra were acquired by data-dependent acquisition mode, and the 20 most intense peaks with charge state ≥2 were selected for sequencing in the HCD collision cell with stepped collision energy of 28%, 30%, and 32%. Each fraction was analyzed in duplicate. Furthermore, in order to study the phosphorylated peptides in E. coli cultured in M9 medium, two mobile phases (A: 0.1% (v/v) FA in water and B: 98% (v/v) ACN with 0.1% (v/v) FA) were used to generate a 121-min gradient with the flow rate of 600 nL/min (90 min from 7 to 25% B, 20 min from 25 to 40% B, 1 min from 40 to 80% B, and 10 min to kept at 80% B). A Orbitrap Fusion Lumos mass spectrometer (Thermo-Fisher, San Jose, CA, USA) was operated in full scan (60,000 FWHM, 350–1500 m/z) and product scan (15,000 FWHM, 100–1000 m/z) modes at positive ion mode with orbitrap detector. The electro-spray voltage was 2.3 kV, and the heated capillary temperature was 320 °C. All the mass spectra were recorded with Xcalibur software (version 3.1, Thermo Fisher Scientific, USA). MS/MS spectra were acquired by data-dependent acquisition mode, and total cycle time was set to 3 s. Peptides with charge state ≥2 were selected for sequencing in the HCD collision cell with collision energy of 30%. HPLC separation for Hela lysate is as same as that for E. coli cultured in M9 medium, and an Orbitrap Fusion Lumos mass spectrometer (Thermo Fisher, San Jose, CA, USA) was operated in full scan (120,000 FWHM, 350–1500 m/z) mode at positive ion mode with orbitrap detector. The electro-spray voltage was 2.5 kV, and the heated capillary temperature was 320 °C. All the mass spectra were recorded with Xcalibur software (version 3.1, Thermo Fisher Scientific, USA). MS/MS spectra were acquired by data-dependent acquisition mode, and total cycle time was set to 3 s. Peptides with charge state ≥2 were selected for sequencing with HCD (collision energy 32%, max injection time 35 ms) and neutral-loss-triggered (Δ97.98) EThcD (ETD reaction time 50 ms, max ETD reagent injection time 200 ms, supplemental activation energy 25%, max injection time 35 ms) for fragmentation. All product ions were detected in the ion trap (rapid mode).

Database search

Proteome Discoverer 2.1 (for analysis of the phosphorylated peptides in E. coli cultured in LB medium) with MASCOT 2.4 and Maxquant 1.6.0 (for the phosphorylated peptides in E. coli cultured in M9 medium and HeLa) were applied for database searching against the NCBI_E.coli_k12 database (updated on 01/18/2016, 4127 proteins) or Uniprot_Homo Sapiens database (updated on 04/04/2019, 42432 proteins). The corresponding reversed database was also performed to evaluate the false discovery rate (FDR) of peptide identification in the database searching process. The parameters of database searching included: up to three missed cleavages allowed for full tryptic digestion, precursor ion mass tolerance 10 p.p.m., product ion mass tolerance 20 m.m.u., carbamidomethylation (C) as a fixed modification, and oxidation (M), phosphorylation (STYHRK), acetyl (protein N-term), deamidated (NQ) as variable modifications. PSMs were validated using perculator based on q-value at a 1% FDR. GO analysis was carried out on http://pantherdb.org/ (Panther 15.0) and http://revigo.irb.hr/. Sequence motif analysis was carried out on Weblogo 2.8.2.

ITC experiment

ITC titration was performed on an Isothermal Titration Calorimeter from MicroCal Inc. All measurements were conducted at 298 K. In general, a solution of the peptide (1–2 mM) in 50 mM HEPES buffer (pH 7.2) was injected stepwise (10 μL × 24 times) into a solution of DpaZn (25–100 μM) dissolved in the same solvent system. The measured heat flow was recorded as function of time and converted into enthalpies (ΔH) by integration of the appropriate reaction peaks. The binding parameters (K_D, ΔH, ΔS, n) were evaluated by applying one site model using the software Origin (MicroCal Inc.).

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.