Analytica Chimica Acta 1066 (2019) 58e68

Contents lists available at ScienceDirect

Analytica Chimica Acta

journal homepage: www.elsevier.com/locate/aca

A sensitive and selective phosphopeptide enrichment strategy by

combining polyoxometalates and cysteamine hydrochloride-modified
chitosan through layer-by-layer assembly
Dandan Jiang a, Zheng Li a, Qiong Jia a, b, *
a
College of Chemistry, Jilin University, Changchun, 130012, China
b
Key Laboratory for Molecular Enzymology and Engineering of Ministry of Education, School of Life Sciences, Jilin University, Changchun, 130012, China

h i g h l i g h t s g r a p h i c a l a b s t r a c t

Polyoxometalates and cysteamine

hydrochloride-modified chitosan
were assembled onto magnetic
nanoparticles via a layer-by-layer
approach.

The prepared material was applied to
enrich phosphopeptides coupled
with MALDI-TOF MS determinations.

The prepared material showed effi-
cient enrichment performance with
high sensitivity, selectivity, and good
reusability for phosphopeptides.

a r t i c l e i n f o a b s t r a c t

Article history: Highly efficient enrichment of phosphopeptides in low abundance prior to mass spectrometry analysis is
Received 19 December 2018 essential in phosphopeptidomics analysis. Herein, an affinity probe possessing both interactions of
Received in revised form polyoxometalates (POMs) and hydrophilic interaction chromatography (HILIC) was facilely prepared.
30 March 2019
POMs and cysteamine hydrochloride-modified chitosan (CYECS) were assembled onto magnetic nano-
Accepted 1 April 2019
particles (MNPs) via a layer-by-layer approach. The developed MNPs-(POM/CYECS) material owned
Available online 3 April 2019
merits of large metal oxide surfaces, positive charge, and hydrophilicity as well as superparamagnetism.
By combining with matrix-assisted laser desorption ionization time-of-flight mass spectrometry
Keywords:
Polyoxometalates
(MALDI-TOF MS), the prepared material was utilized for effective enrichment of phosphopeptides. High
Chitosan selectivity (mass ratio of b-casein and bovine serum albumin digests of 1:5000) and low detection limit
Layer-by-layer assembly (0.02 fmol) toward phosphopeptides were achieved. The recovery of phosphopeptides was measured to
Magnetic be 92.6%. The specific enrichment of phosphopeptides from complex samples (nonfat milk, human saliva,
Enrichment serum, and A549 cell lysate) with MNPs-(POM/CYECS) material further confirmed its bright prospects for
Phosphopeptide isolation of phosphopeptides from biological samples.
© 2019 Elsevier B.V. All rights reserved.

1. Introduction

* Corresponding author. College of Chemistry, Jilin University, Changchun, Protein phosphorylation, an important and universal post-
130012, China; Key Laboratory for Molecular Enzymology and Engineering of
translation modification, plays a significant part in regulating
Ministry of Education, School of Life Sciences, Jilin University, Changchun, 130012,
China. multifarious vital processes involving in protein folding, molecular
E-mail address: [email protected] (Q. Jia). recognition, cell growth, and metabolism [1e3]. To comprehensively

https://doi.org/10.1016/j.aca.2019.04.001
0003-2670/© 2019 Elsevier B.V. All rights reserved.
 D. Jiang et al. / Analytica Chimica Acta 1066 (2019) 58e68 59

understand the phosphorylation pathways in vital processes, the selectivity and sensitivity for the enrichment of phosphopeptides
determination of protein phosphorylation is essential [4,5]. made it a favorable competitor in phosphoproteome.
Recently, mass spectrometry (MS) has been the chief method for the
analysis of protein phosphorylation [6e8]. Unfortunately, the low
content and ionization efficiency of phosphopeptides and serious 2. Experimental section
signal suppression arose from the co-existence of non-
phosphopeptides lead to the direct MS detection still being a diffi- 2.1. Synthesis of Gd-POM
culty [9,10]. Therefore, an effective enrichment step before MS
analysis is required. 1.0375 g Na2WO4$2H2O was dissolved in 20 mL H2O to form
Reported strategies include immunoprecipitation [11], ion ex- homogeneous solution. Then, the mixture was adjusted to pH 7.5
change chromatography [12], hydrophilic interaction chromatog- with HAc (1 mol L1). Subsequently, a solution containing 0.12 g
raphy (HILIC) [13], immobilized metal ion affinity chromatography GdCl3$6H2O was added drop-by-drop to the above solution under
(IMAC) [14e16], and metal oxide affinity chromatography (MOAC) stirring and heated up to 85  C to obtain Gd-POM.
[17,18]. Among the various methods, MOAC is a commonly used
enrichment technology in phosphoproteomics based upon the af-
finity of the phosphate groups to the surface of metal oxides via 2.2. Synthesis of CYECS
bridging bidentate interaction [19,20]. Recently, polyoxometalates
(POMs) with diverse nano-structures have received great attention, The synthesis process of CYECS was listed in Scheme 1. 0.34 g CS
which are a unique kind of metal-oxygen clusters [21e23]. Similar and 0.83 g N-Phth were dissolved in 6 mL DMF and then stirred at
to MOAC, POMs exhibit remarkable features of metal oxide sur- 120  C for 8 h under N2 atmosphere. Afterwards, H2O was added to
faces, which endowed them with great potential in the field of obtain taupe solid of N-Phth modified-CS, N-PhthCS. Then N-
enrichment area [24]. Accompanied with the features of highly PhthCS were washed with methanol and then dried at 60  C under
electronegative properties and abundant chemical combinations, vacuum.
POMs have been applied to the selective adsorption of metal cat- 0.34 g N-PhthCS was dissolved in 10 mL DMF, followed by add-
ions, cationic dyes, and proteins [25e27]. ing 0.52 g EN and stirring at 60  C for 0.5 h. IPA and KOH were
Unfortunately, the electronegative properties of POMs lead to added to the above solution and then stirred at 60  C for 12 h to
their limited enrichment performance for negatively charged obtain EN modified-CS (ENCS). ENCS was washed with methanol
phosphopeptides [28]. Multifunctional affinity probe combining and then dried at 60  C under vacuum.
MOAC with IMAC or HILIC materials is competitive for obtaining 0.3 g ENCS was dissolved in 10 mL DMF. 0.6 g CYE was rapidly
comprehensive information on proteome [29,30]. For instance, an added to the solution under stirring and heated up to 25  C for 1 h.
affinity probe owning both interactions of MOAC and IMAC was 4 mL TEA was added to the above solution and then stirred at 25  C
synthesized by combing zirconium-organic framework with zir- for 8 h. The product was concentrated and washed with methanol
conium (IV) [31]. This material could be regarded as an effective and dispersed in 20 mL N2H4$H2O, followed by stirring at 60  C for
affinity probe for capturing phosphopeptides. A literature survey 6 h to obtain CYE modified-CS (CYECS).
reveals that POMs-based multifunctional affinity materials still
remain undeveloped in phosphoproteomics.
Chitosan (CS), a partially deacetylated form of chitin, is a posi-
2.3. Synthesis of MNPs-NH2
tively charged natural linear polysaccharide [32,33]. CS-based
materials have gained increasing popularity as HILIC adsorbents
MNPs were synthesized through a solvothermal method be-
in separation and enrichment of proteins due to their biological
tween FeCl3$6H2O and ethylene glycol. 1.35 g FeCl3$6H2O and
compatibility, degradability, and chemical activity raised from
3.60 g NaAc were dissolved in 75 mL ethylene glycol. The above
active functional groups such as eOH and eNH2 [34,35]. Based on
mixture was stirred for 1 h to form homogeneous yellow solution,
these considerations, we herein developed a method combining
and then transferred to a Teflon-lined stainless-steel autoclave. The
POMs and HILIC strategies focusing on effective enrichment of
reaction was maintained at 200  C for 16 h. The obtained MNPs
phosphopeptides. CS was modified with cysteamine hydrochloride
were separated with the help of an external magnetic field, washed
to enhance its positive charge and hydrophilicity. Afterwards, Gd-
with ethanol and H2O successively, and then dried at 60  C under
POM and cysteamine hydrochloride-modified chitosan (CYECS)
vacuum.
were layer-by-layer (LbL) assembled onto magnetic nanoparticles
2.0 g MNPs were dispersed into 102.5 mL solution of ethanol/
(MNPs) to form multilayer MNPs-(POM4/CYECS4). The introduction
H2O/NH3$H2O/TEOS (v/v ¼ 80:20:1.5:1). After stirring at 40  C for
of magnetism accelerated the enrichment process with the help of
24 h, the obtained product was separated using a magnet, followed
an external magnetic field. The novel material processed the
by rinsing with ethanol. Then, the product was dispersed in 30 mL
prominent advantages of large metal oxide surfaces, positive
ethanol. After adding 1.3 mL APTES, the mixture was stirred at 25  C
charge, hydrophilicity, and rapid magnetic separation. The high
for 1 h to obtain MNPs-NH2.

Scheme 1. Synthetic scheme of CYECS.

60 D. Jiang et al. / Analytica Chimica Acta 1066 (2019) 58e68

2.4. Synthesis of MNPs-(POM4/CYECS4) and pipetted up and down every 10 min. The resulting cell lysates
were centrifuged and then the supernatant containing soluble
Gd-POM methanol solution (3 mg mL1) and CYECS HAc solu- proteins was separated and used for subsequent experiments.
tion (3 mg mL1) were firstly prepared. Briefly, 90 mg Gd-POM was
dispersed in 30 mL methanol, and the solution was vibrated for 2.7. Selective enrichment of phosphopeptides
0.5 h to obtain Gd-POM methanol solution. Similarly, 90 mg CYECS
was dissolved in 30 mL HAc solution (2%, v/v) with stirring to obtain In a typical experiment, 0.5 mg MNPs-(POM4/CYES4) was added
CYECS HAc solution. into 100 mL sample solution, and the solution was vibrated for 0.5 h.
The fabrication of MNPs-(POM4/CYECS4) was illustrated in Subsequently, MNPs-(POM4/CYES4) were separated by an external
Scheme 2. 30 mg MNPs-NH2 were dispersed in 30 mL Gd-POM magnet. After washing with 100 mL above buffer for 3 times, the
methanol solution (3 mg mL1) and stirred at 50  C for 3 h to trapped phosphopeptides were eluted from MNPs-(POM4/CYECS4)
obtain MNPs-POM1. After that, the prepared MNPs-POM1 was using 20 mL 5% NH3$H2O. Then, the eluent was analyzed by MALDI-
dispersed in 30 mL CYECS HAc solution (3 mg mL1) and stirred at TOF MS.
50  C for 3 h. The obtained product, MNPs-(POM1/CYECS1), was
washed with methanol for 3 times. The layer-by-layer assembly 2.8. Recovery measurement
procedures were repeated to obtain multilayer MNPs-(POM4/
CYECS4). The standard solution of phosphopeptide (LRRApSLGGK) was
split into two equal parts and then labelled with heavy and light
2.5. Adsorption of pNPP with MNPs-(POM/CYES) isotopes, respectively. Then, the heavy labelled phosphopeptides
were captured with MNPs-(POM4/CYECS4) following the above-
2 mg above-prepared material was mixed with 3 mL pNPP so- mentioned enrichment procedure. The acquired eluent was
lution within the concentration range of 0.02e0.15 mg mL1. After mixed by the equal volume of the phosphopeptides with light la-
vibrating for 0.5 h, the supernatant was separated with a magnet, beling for MALDI-TOF-MS detection. To divide the MS peak area
and then subjected to UVeVis analysis. The absorbance at the ratio of phosphopeptides with the heavy labeling to that of the light
wavelength of 285 nm was monitored and selected as the data one could obtain the recovery of LRRApSLGGK.
source for calculating the remained pNPP in the supernatant ac-
cording to the calibration curve. 2.9. MALDI-TOF MS analysis

2.6. Sample preparation Typically, 0.5 mL eluent was mixed with 0.5 mL DHB (25 mg mL1,
70% ACN and 1% H3PO4) on the MALDI plate and dried at 25  C.
1 mg b-casein, ovalbumin or bovine serum albumin (BSA) was MALDI-TOF instrument (Model 5800, Applied Biosystems, Foster
dissolved in 1 mL NH4HCO3 (100 mmol mL1) and digested with City, CA) was performed in the reflection positive-ion mode with
20 mL trypsin (1 mg mL1) at 37  C for 16 h. For the treatment of the m/z scan range of 1000e3500 and a laser pulse rate of 400 Hz.
nonfat milk, 60 mL non-fat milk was diluted by 2 mL above
NH4HCO3 solution, and centrifuged for 5 min. The obtained su- 3. Results and discussion
pernatant was treated at 100  C and digested at 37  C overnight.
50 mL human saliva or serum was diluted with 450 mL NH4HCO3, 3.1. Characterization of MNPs-(POM4/CYECS4)
and was stored at 20  C after centrifuged.
A549 cell lysates were lysed by incubation with 1 mL per FT-IR and XPS determinations were used to obtain structure and
1  107 cells of a lysis buffer containing a cocktail of protease and elemental composition of Gd-POM. In the FT-IR spectrum of Gd-
phosphatase inhibitors for 45 min on ice. The vials were vortexed POM (Fig. S1A), the absorption bands at 755, 850, and 940 cm1

Scheme 2. Synthetic scheme of MNPs-(POM4/CYECS4).

 D. Jiang et al. / Analytica Chimica Acta 1066 (2019) 58e68 61

existed, which were in accordance with previous reports [36,37]. In addition, we determined contact angles of CS and CYECS. The
XPS results were demonstrated in Fig. S1B, in which the peaks at analysis results were presented in Fig. S4. The contact angle values
35.4, 141.6, 418.4, 530.6, and 1072.2 eV corresponded to W4f, Gd4d, of CS and CYECS were 76.2 and 53.4 , respectively. Such results
Gd3d, O1s, and Na1s, respectively. Such results indicated the suc- implied that the positive charge and hydrophilicity substantially
cessful synthesis of Gd-POM. improved after CS was modified with cysteamine hydrochloride.
Afterwards, zeta potential values of Gd-POM at different HAc The effect of acidity on the assembly procedure was verified by
concentrations (1%e4%) were measured (Fig. S2). The negative zeta assembling POM/CYECS on MNPs at different HAc concentrations
potential values between 18 and 45 mV could be observed in the (1%e4%) [42]. The average diameters and polydispersity index (PDI)
whole HAc concentration range. Such a result was because of the values of MNPs-(POM4/CYECS4) were presented in Table S2. It could
existent of counterion-mediated attraction between polyanions be observed that the size distribution range was 185e275 nm. This
that could cause a partial charge shielding [38]. It could also be seen phenomenon was because that the positive charge of CYECS
from Fig. S2 that the negative charge of Gd-POM was lowest when increased with the increase of acidity (Fig. S2), making the mole-
HAc concentration was 2%, which was beneficial to combine with cule more rigid and the hydrodynamic radius bigger due to intra-
CYECS to prepare the suitable material. molecular repulsions [43]. The average particle size was 275 nm at
CYECS was analyzed by FT-IR spectra together with CS, N- high acidity (4%), at which the measured PDI value was 0.98.
PhthCS, and ENCS. As shown in Fig. S3, N-PhthCS and ENCS Whereas, the PDI value was 0.37 at low acidity (1%), indicating that
possessed all characteristic peaks of CS [39]. For N-PhthCS (curve the synthesized MNPs-(POM4/CYECS4) were relatively uniform
b), the peaks at 1715 and 1770 cm1 belonged to C]O vibrations, (size of 185 nm). However, the amount of introduced POM/CYECS
confirming the successful introduction of phthalic anhydride [40]. was small, which was not conducive to obtain high adsorption ef-
For ENCS (curve c), the band at 1275 cm1 corresponded to CeOeC ficiency. Therefore, 2% HAc was finally selected as the optimum
vibrations. Moreover, for CYECS (curve d), the intensity of C]O acidity to prepare POM/CYECS assemblies on MNPs. It was inter-
peak reduced compared with that of ENCS. Such results were in esting to note that this condition was coincided with the zeta po-
accordance with those reported in previous literature, indicating tential value of Gd-POM, i.e., the negative charge was lowest at 2%
that CYECS was prepared with success [41]. The result of elemental HAc solution, benefiting to combine with CYECS to prepare the
analysis was listed in Table S1. The contents of C, H, N, and S in suitable material.
CYECS were 47.48%, 6.72%, 5.95%, and 3.89%, which were in accor- The morphologies and dimensions of MNPs and MNPs-(POM4/
dance with the structure of product [39]. CYECS4) were investigated by SEM and TEM. SEM analyses showed
The zeta potential values of CS and CYECS were measured to that MNPs exhibited a spherical shape with agglomeration (Fig. 1A).
explore the charge variations (Fig. S2). Zeta potential values of CS After the LbL process, MNPs-(POM4/CYECS4) were uniformly
were determined to be 40.1e47.9 mV at different HAc concentra- dispersed and the surface became rougher (Fig. 1B). Fig. 1D illus-
tions (1%e4%). After eNH2 groups were introduced onto CS, zeta trated TEM image of MNPs-(POM4/CYECS4), exhibiting a typical
potential values shifted to 53.2e59.7 mV. It could be concluded that core-shell structure. Furthermore, TEM images of MNPs-(POM1/
the zeta potential values of CYECS changed little at different HAc CYECS1), MNPs-(POM2/CYECS2), and MNPs-(POM3/CYECS3) were
concentrations. shown in Fig. S5. It could be concluded that the thickness of the

Fig. 1. SEM images, TEM images, and particle size distributions of (A, C, E) MNPs and (B, D, F) MNPs-(POM4/CYECS4).
62 D. Jiang et al. / Analytica Chimica Acta 1066 (2019) 58e68

Fig. 2. (A) Zeta potential values and (B) FT-IR spectra of (a) MNPs, (b) MNPs-NH2, (c) MNPs-(POM1), (d) MNPs-(POM1/CYECS1), (e) MNPs-(POM2/CYECS1), (f) MNPs-(POM2/CYECS2),
(g) MNPs-(POM3/CYECS2), (h) MNPs-(POM3/CYECS3), (i) MNPs-(POM4/CYECS3), and (j) MNPs-(POM4/CYECS4).

(POM/CYECS) shell increased almost with the number of layers absorption bands at 755, 850, and 940 cm1 appeared due to the
[44]. introduction of Gd-POM [36]. FT-IR spectrum of MNPs-(POM1/
The image analysis program (Digital Micrograph) was employed CYECS1) illustrated characteristic peaks derived from CYECS (curve
to measure the sizes of the prepared materials. The obtained results d), which clearly indicated that CYECS was conjugated onto MNPs-
were shown in Fig. 1E and F. The number-average diameter (Dn) was POM1 [41]. In curves e-j, the absorption peaks were similar to those
calculated by Equation (1) [45]: of MNPs-(POM1/CYECS1), implying the successful LbL assembly
process.
P
niDi The magnetization saturation values of MNPs, MNPs-(POM1/
Dn ¼ P (1) CYECS1), MNPs-(POM2/CYECS2), MNPs-(POM3/CYECS3), and MNPs-
ni
(POM4/CYECS4) were determined as 67.9, 56.5, 45.9, 38.5, and 28.6
in which ni and Di represent the number and diameter of particles, emu g1, respectively (Fig. 3A). The reversible curve indicated that
respectively. Di values of MNPs and MNPs-(POM4/CYECS4) were MNPs-(POM4/CYECS4) exhibited characteristic superparamagnetic
174.8 and 200.1 nm. behaviors. The superparamagnetism could guarantee its rapid
The LbL process was characterized by zeta potential experi- separation from solution by a magnet (Fig. 3A inset). The results
ments. We firstly detected the zeta potential values of MNPs and also illustrated that the saturation magnetization values decreased
MNPs@NH2, which all displayed positive zeta potential. Therefore, as layer numbers increased. This phenomenon was due to the
after further introduction of (POM/CYECS) layer by layer, charge introduction of non-magnetic POM/CYECS shell [46,47]. Too low
inversions appeared about the zeta potential values (Fig. 2A). These magnetization saturation value did not benefit magnetic separa-
results suggested that the LbL process of (POM/CYECS) onto MNPs tions, therefore, more (POM/CYECS) layers were not considered.
was successful [28,44]. The surface area and porous structure of MNPs, MNPs-(POM1/
FT-IR results of MNPs, MNPs-NH2, and MNPs-(POM/CYECS) se- CYECS1), MNPs-(POM2/CYECS2), MNPs-(POM3/CYECS3), and MNPs-
ries were given in Fig. 2B. For all curves, the peak at 595 cm1 (POM4/CYECS4) were analyzed by N2 adsorption-desorption ex-
corresponded to FeeOeFe vibrations while that at 3470 cm1 was periments. As depicted in Fig. 3B, the Brunauer-Emmett-Teller
due to OeH vibrations, suggesting the existence of Fe3O4 in all (BET) surface areas of MNPs and MNPs-(POM4/CYECS4) were
products [14]. As for MNPs-NH2 (curve b), the peaks at 1080 and about 51 and 149 m2 g1, respectively. The curve of MNPs-(POM4/
1630 cm1 could be ascribed to SieOeSi and NeH vibrations, CYECS4) displayed a combination of Type I and IV shapes. The
respectively, which were direct proofs for the existence of eNH2 typical hysteresis loop (0.8 < P/P0 < 1) illustrated the existence of
groups [44]. In the FT-IR spectrum of MNPs-POM1 (curve c), the mesopores in MNPs-(POM4/CYECS4). The distribution of the

Fig. 3. (A) Magnetization curves and (B) N2 adsorption-desorption isotherms of (a) MNPs, (b) MNPs-(POM1/CYECS1), (c) MNPs-(POM2/CYECS2), (d) MNPs-(POM3/CYECS3), and (e)
MNPs-(POM4/CYECS4).
 D. Jiang et al. / Analytica Chimica Acta 1066 (2019) 58e68 63

mesopores was 10e50 nm by Barrett-Joyner Halenda (BJH) method phosphopeptides, different concentrations of loading buffers and
[48]. eluents were used for the enrichment of phosphopeptides from b-
The enrichment ability of MNPs-(POM/CYECS) composites was casein digests with MNPs-(POM4/CYECS4). The effects of TFA con-
investigated. P-nitrophenylphosphate (pNPP), as a phosphate tents in the loading buffer on the MS intensity were displayed in
molecule, was employed because it contained a phosphate group Fig. S7. When 50% ACNþ3% TFA (v/v) was used as the loading buffer,
that could be captured by affinity adsorbent [49]. The saturated the highest capture efficiency could be achieved. Fig. S8 revealed
adsorption capacity of pNNP with MNPs-(POM4/CYECS4) was that the signal intensity was highest in the mass spectrum when
73.6 mmol g1 and those with MNPs, MNPs-(POM1/CYECS1), MNPs- using 5% NH3$H2O as the eluent. Hence, 50% ACNþ3% TFA and 5%
(POM2/CYECS2), and MNPs-(POM3/CYECS3) were 21.4, 46.5, 61.8, NH3$H2O were employed as the loading buffer and eluent for
and 67.1 mmol g1 (Fig. 4). Compromising the magnetic property, subsequent experiments, respectively.
MNPs-(POM4/CYECS4) were considered in the following studies. Under the above experimental conditions, the mass spectra
The elemental composition of MNPs and MNPs-(POM4/CYECS4) obtained through direct detection and after treatment with MNPs-
was explored by XPS analysis. As depicted in Fig. S6A, O1s (POM4/CYECS4) were compared. As demonstrated in Fig. S9A, many
(532.89 eV) and Fe2p (711.04 eV) peaks appeared in the spectrum of non-phosphopeptides peaks existed through direct analysis. It
MNPs. However, W4f, Si2p, Gd4d, C1s, N1s, Gd3d, and Na1s peaks could be clearly observed from Fig. S9B that the signals of phos-
corresponding to 35.44, 102.81, 153.59, 284.67, 399.72, 418.1, and phopeptides appeared after enrichment [51]. Detailed sequence of
1068.18 eV could be observed except those of O1s and Fe2p in the the identified phosphopeptides was displayed in Table S3. Above
spectrum of MNPs-(POM4/CYECS4). Results illustrated the suc- results demonstrated that the present material could efficiently
cessful modification of the POM/CYECS shell onto MNPs. capture phosphopeptides. The extraction mechanism might be as
The crystal structures of MNPs and MNPs-(POM4/CYECS4) were the followings. Firstly, the abundant metal oxide surfaces of POMs
validated by XRD determinations. It could be seen from Fig. S6B in MNPs-(POM4/CYECS4) provided Lewis acid-base interactions
that 2q diffraction peaks appeared at 30.27, 35.66 , 43.30 , with phosphate groups. Secondly, hydrogen bondings and hydro-
53.58 , 57.21, and 62.78 in the pattern of MNPs-(POM4/CYECS4). philic interactions between phosphopeptides and CYECS in MNPs-
The results was in accord with those of standard Fe3O4 (JCPDS Card (POM4/CYECS4) might contribute to the enrichment efficiency [52].
No. 19e629), indicating that the crystalline structure of MNPs was In addition, MNPs-(POM4/CYECS4) displayed positive charge
not changed in the LbL assembly process [50]. properties, leading to the strong affinity toward negatively charged
Fig. S6C revealed TGA-DTG results of MNPs and MNPs-(POM4/ phosphopeptides via electrostatic interactions.
CYECS4) when the experiments were performed from 20 to 800  C. The sensitivity is deemed to be vital because of the low content
It could be seen from Fig. S6C (inset) that MNPs existed a slight of phosphopeptides in the real samples. b-casein tryptic digests
weight loss in the whole heating process. For MNPs-(POM4/ with lower concentrations (20, 2, 0.2, and 0.02 fmol) were
CYECS4), the mass loss from 20 to 200  C was attributed to the employed as test samples to assess the sensitivity for phospho-
volatilization of adsorbed water. In the range of 200e500  C, the peptides with MNPs-(POM4/CYECS4). As shown in Fig. S10, phos-
mass loss was caused by the decomposition of POM/CYECS shell on phopeptides peaks could be distinctly seen with clean background
MNPs. A slight weight loss appeared in the curve after 500  C, when the b-casein tryptic digests concentration was as low as
which was due to the lattice decomposition of MNPs. The tiptop 0.02 fmol. The results showed that this method had high detection
temperatures of pyrolysis were determined to be 295 and 356  C, sensitivity.
respectively. In order to examine the selectivity of MNPs-(POM4/CYECS4), the
mixtures of b-casein and BSA digests were utilized as test samples.
3.2. Application of MNPs-(POM4/CYECS4) to phosphopeptides It could be seen from Fig. 5 that nearly no phosphopeptide exited
enrichment from standard protein through direct analysis. After treatment with MNPs-(POM4/
CYECS4), 9 phosphopeptides peaks dominated mass spectrum
To investigate the optimal experimental conditions to enrich when the mass ratio of b-casein to BSA was 1:1000 (Fig. 5B). Even
when the mass of BSA was 2000 and 5000-fold than b-casein
(Fig. 5D and F), the phosphopeptides peaks could be evidently
identified with high signal intensity [53]. These results indicated
that MNPs-(POM4/CYECS4) possessed of satisfactory selectivity for
phosphopeptides enrichment.
A mixture of b-casein, BSA, and ovalbumin digests with a mass
ratio (1:2000:2000) was chosen to further evaluate the perfor-
mance of MNPs-(POM4/CYECS4) for phosphopeptides enrichment.
As demonstrated in Fig. S11A, no phosphopeptides could be
observed through direct analysis. However, the signals of phos-
phopeptides appeared after enrichment with MNPs-(POM4/
CYECS4) (Fig. S11B) [54], which further hinted the high selectivity of
the material for phosphopeptides.
The reusability of MNPs-(POM4/CYECS4) was investigated with
b-casein tryptic digests. The obtained phosphopeptides peaks after
capturing 20 times were similar as those after 1 time (Fig. S12). The
results illustrated that MNPs-(POM4/CYECS4) had good reusability.
As displayed in Fig. S12D, it had no evident changes in the mass
spectrum after storage for 3 months. The results indicated that the
prepared MNPs-(POM4/CYECS4) had adequate stability and effi-
Fig. 4. Saturated adsorption isotherms for pNPP adsorbed with (a) MNPs, (b) MNPs-
cient trapping capability. The respective m/z values of the captured
(POM1/CYECS1), (c) MNPs-(POM2/CYECS2), (d) MNPs-(POM3/CYECS3), and (e) MNPs- phosphopeptides were used to calculate the standard deviations
(POM4/CYECS4). (Table S4). The highest value was 0.569 at m/z of 1943, while the
64 D. Jiang et al. / Analytica Chimica Acta 1066 (2019) 58e68

Fig. 5. MALDI-TOF mass spectra of tryptic digests of b-casein and BSA mixtures. (A, C, and E) Direct analysis and (B, D, and F) analysis after enrichment with MNPs-(POM4/CYECS4).
Mass ratios of b-casein to BSA were 1:1000 (A and B), 1:2000 (C and D), and 1:5000 (E and F). “✦” and “#” indicated phosphopeptides and dephosphorylated fragments, respectively.

lowest value was 0.145 at m/z of 2556, implying that there was a
little difference between theoretical and experimental masses [6].
The enrichment recovery of phosphopeptides with MNPs-
(POM4/CYECS4) was evaluated by a stable isotope labeling method
[17,55,56]. Two samples containing same amount of phosphopep-
tide (LRRApSLGGK) were treated with CH2O or CD2O to cause a
mass shift (28 or 32 Da). Then, the heavy labelled one was treated
with MNPs-(POM4/CYECS4) and the eluent was blended with the
light one for detection. As depicted in Fig. 6, the recovery of
phosphopeptide was determined to be 92.6%. The enrichment
performance of MNPs-(POM4/CYECS4) was compared with previ-
ous materials (Table 1), illustrating its high selectivity, low detec-
tion limit, and high recovery for capturing phosphopeptides
[14,17,19,31,51,53,56].

3.3. Biological properties assessment

Fig. 6. MALDI-TOF mass spectra of standard phosphopeptide LRRApSLGGK (a phos-

Human serum and saliva consist of various endogenous pep-
phopeptide mixture including unenriched fraction labelled by CH2O and an equal
tides including phosphopeptides, and these peptides have obtained number of enriched fraction labelled by CD2O).
interests for the disease biomarker discovery. Matrix effect (ME)
was evaluated because it may have an effect on the result accuracy.
ME during validation of analytical method in biological samples where AM is the MS peak area of blank human serum; AS and AMS
was calculated by Equation (2) [57]: are the MS peak area of the standard solution in loading buffer and
blank sample obtained after enrichment. In the work, ME values
AMS  AM were 1.15e1.37 and 0.97e1.21 for human serum and saliva (in
ME ¼ (2)
AS triplicate), illustrating that the ME of this method could be ignored.
Hemolysis ratio (HR) was measured to investigate the blood
 D. Jiang et al. / Analytica Chimica Acta 1066 (2019) 58e68 65

Table 1
Comparison of the present probe with reported materials.

Adsorbent material Sensitivity Selectivity Recovery Ref.

Fe3O4@SiO2@(HA/CS)10eTi4þ 0.5 fmol b-casein:BSA ¼ 1:2000 85.45% [14]

PAA-Ti/TiO2 2 fmol b-casein:BSA ¼ 1:1000 78% [17]
MoO3/GO 1 fmol b-casein:BSA ¼ 1:1000 91.13% [19]
DZMOF 0.07 fmol b-casein:BSA ¼ 1:5000 91.4% [31]
mag GOeCSeTi4þ 0.5 fmol b-casein:BSA ¼ 1:400 93.11% [51]
Fe3O4@MIL-101(Fe) 0.5 fmol b-casein:BSA ¼ 1:500 84.47% [53]
DFMMOF 5 fmol b-casein:BSA ¼ 1:1000 84.8% [56]
MNPs-(POM4/CYECS4) 0.02 fmol b-casein:BSA ¼ 1:5000 92.6% This work

compatibility of MNPs-(POM4/CYECS4). HR was calculated accord- accorded with the requirement of biomaterials.
ing to Equation (3) [54]:

3.4. Application of MNPs-(POM4/CYECS4) to phosphopeptides

Dt  Dnc enrichment from real biosamples
HR ¼ (3)
Dpc  Dnc
To further confirm the practicability of MNPs-(POM4/CYECS4),
where Dt is the absorbance of the sample, Dnc and Dpc are absor- nonfat milk, human saliva, serum, and A549 cell lysate were
bances of the negative control (normal saline) and positive control employed as real samples for the capture of phosphopeptides. For
(water). The HR of MNPs-(POM4/CYECS4) (2.39%) was low than that the non-fat milk tryptic digests, no phosphopeptides were
of clinic limit (5%). The results indicated that MNPs-(POM4/CYECS4) observed through direct analysis owing to the severe interference

Fig. 7. MALDI-TOF mass spectra of (A and B) nonfat milk, (C and D) human saliva, and (E and F) serum. (A, C, and E) Direct analysis and (B, D, and F) analysis after enrichment with
MNPs-(POM4/CYECS4). “✦” indicated phosphopeptides.
66 D. Jiang et al. / Analytica Chimica Acta 1066 (2019) 58e68

Fig. 8. MALDI-TOF mass spectra of A549 cell lysates: (A) Direct analysis and (B) analysis after enrichment with MNPs-(POM4/CYECS4). (C, D, and E) The enlarged images of MALDI
mass spectra of phosphopeptides in detail. “✦” indicated phosphopeptides.

by a large amount of non-phosphopeptides (Fig. 7A). As depicted in practical applications, MNPs-(POM4/CYECS4) were employed to
Figs. 7B, 15 phosphopeptides peaks were confidently detected with enrich low abundance phosphopeptides from nonfat milk, human
a clear background after enrichment with MNPs-(POM4/CYECS4) saliva, serum, and A549 cell lysate. This study provided a feasible
[53]. The detailed sequences of captured phosphopeptides were method for the discrimination of phosphopeptides and opened a
listed in Table S5. new avenue for the development of POM-based materials for spe-
We studied the practical application of MNPs-(POM4/CYECS4) by cific biological applications.
using human saliva and serum samples. As demonstrated in Fig. 7C
and E, most non-phosphopeptides could be identified through Declaration of interests
direct analysis. After treatment with MNPs-(POM4/CYECS4),
endogenous phosphopeptides peaks were easily detected in the The authors declare that they have no known competing
spectrum (Fig. 7D and F) [31]. Detailed sequence information of financial interests or personal relationships that could have
phosphopeptides was also listed in Table S5. appeared to influence the work reported in this paper.
To further investigate the feasibility of this method in practical
analysis, A549 cell lysate were selected as a real biological sample Acknowledgments
[58]. Direct analysis results without the enrichment step were
presented in Fig. 8A, indicating that no phosphopeptides signals This work was financially supported by State Key Laboratory of
appeared in the MALDI mass spectra. Nevertheless, after enrich- Inorganic Synthesis and Preparative Chemistry, College of Chem-
ment with MNPs-(POM4/CYECS4), phosphopeptides peaks istry, Jilin University, China (2019-4).
appeared in the mass spectrum (Fig. 8B). The detailed MALDI mass
spectra of phosphopeptides containing I (950e1050 Da), II Appendix A. Supplementary data
(1250e1350 Da), and III (1550e1700 Da) were displayed in Fig. 8C,
D, and E, respectively. The identified sequences of phosphopeptides Supplementary data to this article can be found online at
were listed in Table S5. All results suggested that MNPs-(POM4/ https://doi.org/10.1016/j.aca.2019.04.001.
CYECS4) had high selectivity for phosphopeptides from compli-
cated biological samples. References

[1] T. Pawson, Specificity in signal transduction: review from phosphotyrosine-

4. Conclusions SH2 domain interactions to complex cellular systems, Cell 116 (2004)
191e203.
[2] T. Hunter, Signaling-2000 and beyond, Cell 100 (2000) 113e127.
In the study, a novel magnetic adsorbent combining POMs and
[3] R.L. Wiseman, J.W. Kelly, Phosphatase inhibition delays translational recovery,
HILIC strategies with large metal oxide surfaces, positive charge, Science 332 (2011) 44e45.
hydrophilicity, and superparamagnetism was successfully prepared [4] G.Y. Qing, Q. Lu, X.L. Li, J. Liu, M.L. Ye, X.M. Liang, T.L. Sun, Hydrogen bond
for phosphopeptides isolation. The prepared MNPs-(POM4/CYECS4) based smart polymer for highly selective and tunable capture of multiply
phosphorylated peptides, Nat. Commun. 8 (2017) 461.
material showed efficient enrichment performance with high [5] J.M. Irish, R. Hovland, P.O. Krutzik, O.D. Perez, O. Bruserud, B.T. Gjertsen,
sensitivity, selectivity, and good reusability for phosphopeptides. In G.P. Nolan, Single cell profiling of potentiated phospho-protein networks in
 D. Jiang et al. / Analytica Chimica Acta 1066 (2019) 58e68 67

cancer cells, Cell 118 (2004) 217e228. [32] M. Rinaudo, Chitin and chitosan: properties and applications, Prog. Polym. Sci.
[6] F. Jabeen, M. Najam-ul-Haq, M. Rainer, C.W. Huck, G.K. Bonn, In-tip lanthanum 31 (2006) 603e632.
oxide monolith for the enrichment of phosphorylated biomolecules, Anal. [33] W. Jiang, W. Wang, B. Pan, Q. Zhang, W. Zhang, L. Lv, Facile fabrication of
Chem. 89 (2017) 10232e10238. magnetic chitosan beads of fast kinetics and high capacity for copper removal,
[7] B. Domon, R. Aebersold, Mass spectrometry and protein analysis, Science 312 ACS Appl. Mater. Interfaces 6 (2014) 3421e3426.
(2006) 212e217. [34] Z.G. Wang, J.L. Zhang, D.H. Sun, J.Z. Ni, Novel Ti4þ-chelated magnetic nano-
[8] E. Lopez, I. Lopez, A. Ferreira, J. Sequi, Retracted: clinical and technical phos- structured affinity microspheres containing N-methylene phosphonic chito-
phoproteomic research, Proteome Sci. 9 (2011) 27. san for highly selective enrichment and rapid separation of phosphopeptides,
[9] X.M. He, X. Chen, B.F. Yuan, Y.Q. Feng, Graft modification of cotton with J. Mater. Chem. B 2 (2014) 6886e6892.
phosphate group and its application to the enrichment of phosphopeptides, [35] L.G. Qin, H.H. Dong, Z.Q. Mu, Y.L. Zhang, G.N. Dong, Preparation and bioactive
J. Chromatogr. A 1484 (2017) 49e57. properties of chitosan and casein phosphopeptides composite coatings for
[10] L. Yi, T.J. Shi, M.A. Gritsenko, C.Y. X’avia Chan, T.L. Fillmore, B.M. Hess, orthopedic implants, Carbohydr. Polym. 133 (2015) 236e244.
A.C. Swensen, T. Liu, R.D. Smith, H. Steven Wiley, W.J. Qian, Targeted quan- [36] Y. Yong, L.J. Zhou, S.S. Zhang, L. Yan, Z.J. Gu, G.J. Zhang, Y.L. Zhao, Gadolinium
tification of phosphorylation dynamics in the context of EGFR-MAPK polytungstate nanoclusters: a new theranostic with ultrasmall size and ver-
pathway, Anal. Chem. 90 (2018) 5256e5263. satile properties for dual-modal MR/CT imaging and photothermal therapy/
[11] H. Zheng, P. Hu, D.F. Quinn, Y.K. Wang, Phosphotyrosine proteomic study of radiotherapy of cancer, NPG Asia Mater. 8 (2016) 273.
interferon alpha signaling pathway using a combination of immunoprecipi- [37] M.A. AlDamen, S. Cardona-Serra, J.M. Clemente-Juan, E. Coronado, A. Gaita-
tation and immobilized metal affinity chromatography, Mol. Cell. Proteomics Arino, C. Marti-Gastaldo, F. Luis, O. Montero, Mononuclear lanthanide single
4 (2005) 721e730. molecule magnets based on the polyoxometalates [Ln(W5O18)2]9- and [Ln(2-
[12] M.L. Hennrich, V. Groenewold, G.J. Kops, A.J. Heck, S. Mohammed, Improving SiW11O39)2]13-(LnIII) Tb, Dy, Ho, Er, Tm, and Yb, Inorg. Chem. 48 (2009)
depth in phosphoproteomics by using a strong cation exchange-weak anion 3467e3479.
exchange-reversed phase multidimensional separation approach, Anal. Chem. [38] G. Liu, T.B. Liu, S.S. Mal, U. Kortz, Wheel-shaped polyoxotungstate
83 (2011) 7137e7143. [Cu20Cl(OH)24(H2O)12(P8W48O184)]25- macroanions form supramolecular
[13] D.E. McNulty, R.S. Annan, Hydrophilic interaction chromatography reduces "blackberry" structure in aqueous solution, J. Am. Chem. Soc. 128 (2006)
the complexity of the phosphoproteome and improves global phosphopep- 10103e10110.
tide isolation and detection, Mol. Cell. Proteomics 7 (2008) 971e980. [39] M.L. Li, J. Xu, R.H. Li, D.G. Wang, T.B. Li, M.S. Yuan, J.Y. Wang, Simple prepa-
[14] Z.C. Xiong, L.Y. Zhang, C.L. Fang, Q.Q. Zhang, Y.S. Ji, Z. Zhang, W.B. Zhang, ration of aminothiourea-modified chitosan as corrosion inhibitor and heavy
H.F. Zou, Ti4þ-immobilized multilayer polysaccharide coated magnetic metal ion adsorbent, J. Colloid Interface Sci. 417 (2014) 131e136.
nanoparticles for highly selective enrichment of phosphopeptides, J. Mater. [40] K. Kurita, H. Ikeda, Y. Yoshida, M. Shimojoh, M. Harata, Chemoselective pro-
Chem. B 2 (2014) 4473e4480. tection of the amino groups of chitosan by controlled phthaloylation: facile
[15] Q. Wang, X.M. He, X. Chen, G.T. Zhu, R.Q. Wang, Y.Q. Feng, Pyridoxal 50 - preparation of a precursor useful for chemical modifications, Bio-
phosphate mediated preparation of immobilized metal affinity material for macromolecules 3 (2002) 1e4.
highly selective and sensitive enrichment of phosphopeptides, J. Chromatogr. [41] J. Fangkangwanwong, R. Yoksan, S. Chirachanchai, Chitosan gel formation via
A 1499 (2017) 30e37. the chitosan-epichlorohydrin adduct and its subsequent mineralization with
[16] X.M. He, X. Chen, G.T. Zhu, Q. Wang, B.F. Yuan, Y.Q. Feng, Hydrophilic carboxyl hydroxyapatite, Polymer 47 (2006) 6438e6445.
cotton chelator for titanium(IV) immobilization and its application as novel [42] G. Fiorani, O. Saoncella, P. Kaner, S.A. Altinkaya, A. Figoli, M. Bonchio,
fibrous sorbent for rapid enrichment of phosphopeptides, ACS Appl. Mater. M. Carraro, Chitosan-polyoxometalate nanocomposites: synthesis, character-
Interfaces 7 (2015) 17356e17362. ization and application as antimicrobial agents, J. Clust. Sci. 25 (2014)
[17] H.F. Sun, Q.Q. Zhang, L. Zhang, W.B. Zhang, L.Y. Zhang, Facile preparation of 839e854.
molybdenum (VI) oxide-modified graphene oxide nanocomposite for specific [43] J.H. Pa, T.L. Yu, Light scattering study of chitosan in acetic acid aqueous so-
enrichment of phosphopeptides, J. Chromatogr. A 1521 (2017) 36e43. lutions, Macromol. Chem. Phys. 202 (2001) 985e991.
[18] G.T. Zhu, X.M. He, S. He, X. Chen, S.K. Zhu, Y.Q. Feng, Synthesis of poly- [44] Z.C. Xiong, H.Q. Qin, H. Wan, G. Huang, Z. Zhang, J. Dong, L.Y. Zhang,
ethylenimine functionalized mesoporous silica for in-pipet-tip phosphopep- W.B. Zhang, H.F. Zou, Layer-by-layer assembly of multilayer polysaccharide
tide enrichment, ACS Appl. Mater. Interfaces 8 (2016) 32182e32188. coated magnetic nanoparticles for the selective enrichment of glycopeptides,
[19] X.N. Wei, H.L. Wang, Facile fabrication of hydrophilic PAA-Ti/TiO2 nano- Chem. Commun. 49 (2013) 9284e9286.
composite for selective enrichment and detection of phosphopeptides from [45] D.D. Jiang, X.Q. Li, Q. Jia, Multilayer cucurbit[6]uril-based magnetic nano-
complex biological samples, Anal. Chim. Acta 949 (2017) 67e75. particles prepared by host-guest interaction: remarkable adsorbent for low
[20] X.S. Li, B.F. Yuan, Y.Q. Feng, Recent advances in phosphopeptide enrichment: density lipoprotein removal from plasma, Chem. Eur J. 24 (2018) 2242e2248.
strategies and techniques, Trends Anal. Chem. 78 (2016) 70e83. [46] H. Wang, B.B. Chen, S.Q. Zhu, X.X. Yu, M. He, B. Hu, Chip-based magnetic solid-
[21] R. John Errington, S.S. Petkar, B.R. Horrocks, A. Houlton, L.H. Lie, S.N. Patole, phase microextraction online coupled with microHPLC-ICPMS for the deter-
Covalent immobilization of a TiW5 polyoxometalate on derivatized silicon mination of mercury species in cells, Anal. Chem. 88 (2016) 796e802.
surfaces, Angew. Chem. Int. Ed. 44 (2005) 1254e1257. [47] J. Tan, S. Namuangruk, W.F. Kong, N. Kungwan, J. Guo, C.C. Wang, Manipula-
[22] X. Lopez, J.M. Maestre, C. Bo, J.M. Poblet, Electronic properties of poly- tion of amorphous-to-crystalline transformation: towards the construction of
oxometalates: a DFT study of a/b-[XM12O40]ne relative stability (M¼W, Mo covalent organic framework hybrid microspheres with NIR photothermal
and X a main group element), J. Am. Chem. Soc. 123 (2001) 9571e9576. conversion ability, Angew. Chem. Int. Ed. 55 (2016) 13979e13984.
[23] M.C. Leon, E. Coronado, P. Delhaes, C.J. Gomez-Garcia, C. Mingotaud, Hybrid [48] D.D. Jiang, T.T. Hu, H.J. Zheng, G.X. Xu, Q. Jia, Aptamer-functionalized magnetic
Langmuir-Blodgett films formed by alternating layers of magnetic poly- conjugated organic framework for selective extraction of traces of hydroxyl-
oxometalate clusters and organic donor molecules-towards the preparation of ated polychlorinated biphenyls in human serum, Chem. Eur J. 24 (2018)
multifunctional molecular materials, Adv. Mater. 13 (2001) 574e577. 10390e10396.
[24] A. Nisar, Y. Lu, J. Zhuang, X. Wang, Polyoxometalate nanocone nanoreactors: [49] X.Q. Chen, S.Y. Li, X.X. Zhang, Q.H. Min, J.J. Zhu, Weaving a two-dimensional
magnetic manipulation and enhanced catalytic performance, Angew. Chem. fishing net from titanoniobate nanosheets embedded with Fe3O4 nano-
Int. Ed. 50 (2011) 3187e3192. crystals for highly efficient capture and isotope labeling of phosphopeptides,
[25] A.G. Moaser, R. Khoshnavazi, Facile synthesis and characterization of Fe3O4@ Nanoscale 7 (2015) 5815e5825.
MgAl-LDH@STPOM nanocomposites for highly enhanced and selective [50] J. Liu, Z.K. Sun, Y.H. Deng, Y. Zou, C.Y. Li, X.H. Guo, L.Q. Xiong, Y. Gao, F.Y. Li,
degradation of methylene blue, New J. Chem. 41 (2017) 9472e9481. D.Y. Zhao, Highly water-dispersible biocompatible magnetite particles with
[26] Q. Chen, D.D. Zhang, M.M. Wang, X.W. Chen, J.H. Wang, A novel organic- low cytotoxicity stabilized by citrate groups, Angew. Chem. Int. Ed. 48 (2009)
inorganic hybrid polyoxometalate for the selective adsorption/isolation of b- 5875e5879.
lactoglobulin, J. Mater. Chem. B 3 (2015) 6964e6970. [51] Y.M. Zhao, L.Y. Zhang, Z.Y. Chu, Z.C. Xiong, W.B. Zhang, Ti4þ-immobilized
[27] Q. Chen, X. Hu, D.D. Zhang, X.W. Chen, J.H. Wang, Selective isolation of myosin chitosan-coated magnetic graphene oxide for the highly selective enrichment
subfragment-1 with a DNA-polyoxovanadate bioconjugate, Bioconjug. Chem. of phosphopeptides, Anal. Methods 9 (2017) 443e449.
28 (2017) 2976e2984. [52] Y. Zhang, M.M. Wang, J.X. Hao, X.W. Chen, J.H. Wang, Discrimination and
[28] S.M. Zhang, Y.M. Zheng, D.Y. Fu, W. Li, Y.Q. Wu, B. Li, L.X. Wu, Biocompatible highly selective adsorption of phosphoproteins and glycoproteins with
supramolecular dendrimers bearing a gadolinium-substituted polyanionic arginine-functionalized polyhedral oligomeric silsesquioxane frameworks,
core for MRI contrast agents, J. Mater. Chem. B 5 (2017) 4035e4043. J. Mater. Chem. B 6 (2018) 4116e4123.
[29] M. Wakabayashi, Y. Kyono, N. Sugiyama, Y. Ishihama, Extended coverage of [53] Y.J. Chen, Z.C. Xiong, L. Peng, Y.Y. Gan, Y.M. Zhao, J. Shen, J.H. Qian, L.Y. Zhang,
singly and multiply phosphorylated peptides from a single titanium dioxide W.B. Zhang, Facile preparation of core-shell magnetic metal-organic frame-
microcolumn, Anal. Chem. 87 (2015) 10213e10221. work nanoparticles for the selective capture of phosphopeptides, ACS Appl.
[30] X.J. Zou, J.Z. Jie, B. Yang, Single-step enrichment of N-glycopeptides and Mater. Interfaces 7 (2015) 16338e16347.
phosphopeptides with novel multifunctional Ti4þ-immobilized dendritic [54] D.D. Jiang, X.Q. Li, X.J. Lv, Q. Jia, A magnetic hydrazine-functionalized den-
polyglycerol coated chitosan nanomaterials, Anal. Chem. 89 (2017) drimer embedded with TiO2 as a novel affinity probe for the selective
7520e7526. enrichment of low-abundance phosphopeptides from biological samples,
[31] J.X. Peng, H.Y. Zhang, X. Li, S.J. Liu, X.Y. Zhao, J. Wu, X.H. Kang, H.Q. Qin, Talanta 185 (2018) 461e468.
Z.F. Pan, R.A. Wu, Dual-metal centered zirconium-organic framework: a metal [55] P.J. Boersema, R. Raijmakers, S. Lemeer, S. Mohammed, A.J.R. Heck, Multiplex
affinity probe for highly specific interaction with phosphopeptides, ACS Appl. peptide stable isotope dimethyl labeling for quantitative proteomics, Nat.
Mater. Interfaces 8 (2016) 35012e35020. Protoc. 4 (2009) 484e494.
68 D. Jiang et al. / Analytica Chimica Acta 1066 (2019) 58e68

[56] J.Q. Zhou, Y.L. Liang, X.W. He, L.X. Chen, Y.K. Zhang, Dual-functionalized HPLC-MS/MS, Anal. Chem. 75 (2003) 3019e3030.
magnetic metal-organic framework for highly specific enrichment of phos- [58] X.Y. Long, Z.J. Zhang, J.Y. Li, D. Sheng, H.Z. Lian, Controllable preparation of
phopeptides, ACS Sustain. Chem. Eng. 5 (2017) 11413e11421. CuFeMnO4 nanospheres as a novel multifunctional affinity probe for efficient
[57] B.K. Matuszewski, M.L. Constanzer, C.M. Chavez-Eng, Strategies for the adsorption and selective enrichment of low-abundance peptides and phos-
assessment of matrix effect in quantitative bioanalytical methods based on phopeptides, Anal. Chem. 89 (2017) 10446e10453.