Ti

ZRa

Ab

a

ARRAA

KPDIMFP

1

dWtoAcataps

t

0h

Journal of Chromatography A, 1334 (2014) 55–63

Contents lists available at ScienceDirect

Journal of Chromatography A

j o ur na l ho me page: www.elsev ier .com/ locate /chroma

he on-bead digestion of protein corona on nanoparticles by trypsinmmobilized on the magnetic nanoparticle

hengyan Hua,b, Liang Zhaoa,b, Hongyan Zhanga,b, Yi Zhanga,b,en’an Wua,∗, Hanfa Zoua,∗∗

CAS Key Laboratory of Separation Science for Analytical Chemistry, National Chromatographic R & A Center, Dalian Institute of Chemical Physics, Chinesecademy of Sciences (CAS) , Dalian 116023, ChinaGraduate School of Chinese Academy of Sciences, Beijing 100049, China

r t i c l e i n f o

rticle history:eceived 12 December 2013eceived in revised form 23 January 2014ccepted 29 January 2014vailable online 5 February 2014

eywords:rotein coronaigestion

mmobilized trypsinagnetic nanoparticle

a b s t r a c t

Proteins interacting with nanoparticles would form the protein coronas on the surface of nanoparticles inbiological systems, which would critically impact the biological identities of nanoparticles and/or resultin the physiological and pathological consequences. The enzymatic digestion of protein corona was theprimary step to achieve the identification of protein components of the protein corona for the bottom-up proteomic approaches. In this study, the investigation on the tryptic digestion of protein corona bythe immobilized trypsin on a magnetic nanoparticle was carried out for the first time. As a comparisonwith the usual overnight long-time digestion and the severe self-digestion of free trypsin, the on-beaddigestion of protein corona by the immobilized trypsin could be accomplished within 1 h, along withthe significantly reduced self-digestion of trypsin and the improved reproducibility on the identificationof proteins by the mass spectrometry-based proteomic approach. It showed that the number of iden-

e3O4

lasmatified bovine serum (BS) proteins on the commercial Fe3O4 nanoparticles was increased by 13% for theimmobilized trypsin with 1 h digestion as compared to that of using free trypsin with even overnightdigestion. In addition, the on-bead digestion of using the immobilized trypsin was further applied on theidentification of human plasma protein corona on the commercial Fe3O4 nanoparticles, which leads theefficient digestion of the human plasma proteins and the identification of 149 human plasma proteinscorresponding to putative critical pathways and biological processes.

. Introduction

Engineered nanosized materials have been widely applied inrug/gene delivery, disease diagnosis and biosensing etc. [1–5].hen nanoparticles are introduced into biological systems, pro-

eins would interact with nanoparticles and lead to the formationf so called “protein corona” on the surface of the nanoparticles.s reported previously, the protein corona on nanoparticles wouldritically impact the biological identities of nanoparticles [6–9]nd/or result in the physiological and pathological consequences ofhe macrophage uptake, blood coagulation, complement activationnd cellular toxicity, etc. [10–16]. The elaborate investigation onrotein corona would accordingly make great sense for the exten-

ive application of nanomaterials.

To characterize proteins adsorbed on the surface of nanopar-icles, technologies such as the dynamic light scattering (DLS)

∗ Corresponding author. Tel.: +86 411 84379828; fax: +86 411 84379617.∗∗ Corresponding author. Tel.: +86 411 84379610; Fax: +86 411 84379620.

E-mail addresses: wurenan@dicp.ac.cn (R. Wu), hanfazou@dicp.ac.cn (H. Zou).

021-9673/$ – see front matter © 2014 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.chroma.2014.01.077

© 2014 Elsevier B.V. All rights reserved.

[15,17,18], transmission electron microscopy (TEM) [18], circu-lar dichroism (CD) [18,19], size exclusion chromatography (SEC)[20,21], isothermal titration calorimetry (ITC) [20,22], fluorescencequenching technology [23] and surface plasmon resonance (SPR)[21,22], and mass spectrometry (MS) etc. have been performedto determine the thickness, density, arrangement, conformation,affinity and identification of proteins on nanoparticles [9]. Fromwhich, the identification of proteins involved in a protein corona onthe surface of nanoparticles would be one of the key issues for theunderstanding of biological effects and/or responses of nanopar-ticles with protein corona [11]. Generally, protein corona formedon nanoparticles should be eluted and then enzymatically digestedfor the following analysis of mass spectrometry (MS). Generally,detergent could be applied to elute proteins from nanoparticlesas the denaturation of proteins by SDS [24]. However, SDS deter-gent existed in elution process would interfere the followingenzymatic digestion and identification of protein corona by mass

spectrometry. The isolation of proteins from SDS elution was thusrequired usually via the precipitation of using organic solvent suchas acetone overnight [24] and/or with a further SDS-PAGE sepa-ration [17,25,26], which unfortunately were time-consuming and

5 togr.

ltttotaddect[cmstt

tslttcFcpdpttaaons

2

2

e(a(Af(pfSSbMiiob

2

s

6 Z. Hu et al. / J. Chroma

aborious. Moreover, the sample loss for low abundant proteins,he low digestion efficiency and the cross-contamination werehe nerve-racking problems during protein isolation and diges-ion, which would bring the great adverseness to the determinationf proteins, especially for minute biological samples. Rather thanhe isolation of protein corona from nanoparticles via SDS elutionnd the following precipitation with organic solvents, the directigestion of protein corona on nanoparticles (also called on-beadigestion) might be an efficient approach to avoid the interfer-nce of SDS. With free trypsin, the protein corona on nanoparticlesould be directly digested, which could simplify the procedures ofhe isolation and digestion of protein corona from nanoparticles27]. However, the long time digestion of free trypsin is required toomplete the digestion of proteins from nanoparticles for above-entioned on-bead digestion or SDS elution assay. Moreover, the

elf-digestion of the applied free trypsin in solution would bringhe interference in the identification of protein corona by MS andhe followed protein database searching.

Trypsin immobilized on stationary phases has demonstratedhe highly efficiency as well as the correspondingly reducedelf-digestion in the digestion of proteins [28–30]. To avoid theimitation of using SDS elution as well as to improve the diges-ion of protein corona from nanoparticles with a high efficiency, inhis work, the on-bead digestion of protein corona on nanoparti-les was carried out by using the trypsin immobilized on magnetice3O4 nanoparticles. It showed that the digestion of proteinorona on the surface of nanoparticles could be efficiently com-leted within 1 h rather than the usual long time (overnight)igestion, with the significantly reduced self-digestion as com-ared to the severe self-digestion of free trypsin. In addition,he reproducibility for the digestion and identification of pro-ein corona on nanoparticles was improved obviously due to thebsence of SDS as well as the reduced self-digestion of trypsinfter the immobilization on nanoparticles. Also, with the helpf the magnet property, the trypsin immobilized on the mag-etic nanoparticles could be simply isolated from the digestionolution.

. Materials and methods

.1. Chemicals and materials

Fe3O4 nanoparticle (∼30 nm), Dithiothreitol (DTT), iodoac-tamide (IAA), TPCK-treated trypsin, bovine serum albuminBSA), formic acid (FA), tetraethyl orthosilicate (TEOS) and (3-minopropyl) triethoxysilane (APTES), sodium cyanoborohydrideNaCNBH3) were obtained from Sigma-Aldrich (St. Louis, MO, USA).cetonitrile (ACN, HPLC grade) and ammonium solution (25%) was

rom Merck (Darmstadt, Germany). Iron (III) chloride hexahydrateFeCl3.6H2O), sodium acetate anhydrous, ethylene diamine, iso-ropanol, ethylene glycol (EG), glutaraldehyde and ethanol wererom Tianjin Kermel plant of chemical reagent (Tianjin, China).DS-PAGE marker and loading buffer were obtained from Thermocientific (San Jose, CA). Water used in experiments was dou-ly distilled and purified by a Mill-Q system (Millipore, Bedford,A, USA). Bovine serum (BS) was bought from Tianhang Biolog-

cal Technology Co. (Zhejiang, China). The human plasma usedn experiments was obtained from the Second Affliated Hospitalf Dalian Medical University (Dalian, China) and stored at −80 ◦Cefore use.

.2. Synthesis of the immobilized trypsin

Fe3O4 nanoparticle with the size of 250 nm was firstly synthe-ized according to a literature procedure [31]. Then the synthesized

A 1334 (2014) 55–63

Fe3O4 nanoparticle was coated by a silica shell via sol-gel process(noted as Fe3O4@SiO2 nanoparticles) and followed by the reac-tion with APTES to yield the amine-functionalized Fe3O4@SiO2nanoparticles (noted as Fe3O4@SiO2@NH2 nanoparticles) [29]. Theamine functionalized magnetic Fe3O4 nanoparticles were thenreacted with glutaraldehyde in PBS solution to introduce the alde-hyde groups on [32]. After that, the Fe3O4 nanoparticles werereacted with TPCK treated trypsin to immobilize the trypsin onFe3O4 nanoparticles, followed with a capping procedure of theremaining aldehyde groups on Fe3O4 nanoparticles by glycine. Theobtained trypsin immobilized Fe3O4 nanoparticles were preservedin PBS buffer containing 0.02% sodium azide for later use to digestprotein corona on nanoparticles (the detailed synthesis procedureswere illustrated in supporting information).

2.3. The digestion of standard BSA on Fe3O4 nanoparticles by theimmobilized trypsin

Standard BSA solution (2 mg/mL) was prepared by dissolving6 mg BSA in 3 mL NH4HCO3 solution (50 mM, pH 8). 100 �L of 30 nmFe3O4 nanoparticles (0.2% w/w in PBS, pH 7.4) suspension weremixed with 100 �L of standard BSA solution, and incubated by shak-ing at 1400 rpm for 1 h at 37 ◦C to prepare the composite of BSAand Fe3O4 nanoparticle. The resulting composite of BSA and Fe3O4nanoparticle was held by a magnet, washed with PBS for at least 3times, resuspended in a solution of 8 M urea and 50 mM NH4HCO3,and then reduced with 10 mM DTT at 37 ◦C for 2 h and followed byalkylation with 20 mM IAA at room temperature with shaking indark for 30 min. Then the solution was diluted to ∼1 M urea and theprepared BSA-Fe3O4 complexes were then digested by the immo-bilized trypsin (10 �g) with digestion time of 15 min, 30 min and1 h, respectively. After that, the immobilized trypsin were retainedby a magnet, and the tryptic digests of BSA from the BSA-Fe3O4complexes were collected, acidified (pH 2∼3), desalted, and finallylyophilized to dryness for LC-MS/MS analysis. For comparison, theBSA-Fe3O4 complexes were also digested by free trypsin (4 �g) for15 min, 30 min, 1 h and overnight, respectively.

2.4. The digestion of protein corona of BS on Fe3O4 nanoparticlesby immobilized trypsin

To evaluate the digestion efficiency of immobilized trypsin, theBS protein corona was prepared on the 30 nm Fe3O4 nanoparti-cles. Briefly, 200 �L of BS was incubated with 200 �L of Fe3O4nanoparticles suspension (0.2% w/w in PBS) with shaking at1400 rpm on a multi-thermo-shaker (Hangzhou Allsheng Instru-ments Co., Ltd.) at 37 ◦C for 1 h to yield the composite of BS proteincorona@Fe3O4 nanoparticles. The as-formed composite of BS pro-tein corona@Fe3O4 nanoparticles was washed with PBS for 3 timeswith a help of a magnet, resuspended in a 200 �L of 8 M urea,50 mM NH4HCO3 solution, and followed with the reduction by10 mM DTT at 37 ◦C for 2 h and the alkylation by 20 mM IAA atroom temperature with shaking in dark for 30 min subsequently.Then the solution was diluted to ∼1 M urea and the digestion ofthe BS protein corona bound on Fe3O4 nanoparticles was carriedout by incubating the composite with immobilized trypsin (20 �g)at 37 ◦C for 1 h. For comparison, the same amount of BS protein-Fe3O4 complexes were also digested by free trypsin (8 �g) for 1 hand overnight, respectively. The digested peptides of the BS protein

corona from the composite of BS protein corona@Fe3O4 nanoparti-cles were collected by removing the residual Fe3O4 nanoparticles aswell as the magnetic immobilized trypsin by a magnet, which wereacidified to pH 2–3, desalted, lyophilized and stored at −30 ◦C.

togr.

2n

Pwovpttn5ailtdFtwM

2

0MwiBFtlrTeeoo

wsoA0tosemaf1

2

sTdITsar

Z. Hu et al. / J. Chroma

.5. The analysis of protein corona of human plasma on Fe3O4anoparticles by the immobilized trypsin

200 �L of 30 nm Fe3O4 nanoparticles suspensions (0.2% w/w inBS) were incubated with 200 �L of human plasma for 1 h at 37 ◦Cith shaking at 1400 rpm, respectively. The as-formed composites

f human plasma protein corona@Fe3O4 nanoparticles were har-ested by a magnet, and washed by PBS for 3 times to get the hardlasma corona on the Fe3O4 nanoparticles. To identify proteins ofhe human plasma protein corona formed on the Fe3O4 nanopar-icles, the composites of human plasma protein corona@Fe3O4anoparticles were resuspended in a solution of 8 M urea and0 mM NH4HCO3. After reduced by 10 mM DTT at 37 ◦C for 2 hnd alkylated by 20 mM IAA at room temperature with shakingn dark for 30 min, the solution was diluted to ∼1 M urea and fol-owed with the digestion of the human plasma protein corona byhe immobilized trypsin (20 �g) at 37 ◦Cfor 1 h with shaking. Theigested peptides of the human plasma protein corona from thee3O4 nanoparticles were collected by isolating the Fe3O4 nanopar-icles from the digestion solution with the help of a magnet, whichere further acidified, lyophilized and stored at −30 ◦C for RPLC-S/MS analysis.

.6. RPLC-MS/MS analysis

The tryptic digested peptides of BSA were dissolved in 100 �L.1% FA solution, and 15 �L of the peptides was analyzed by 1-D LC-S/MS system each time. The C18 trap column (7 cm × 200 �m i.d.)as connected to a reverse phase (RP) C18 column (12 cm × 75 �m

.d.) packed with C18 AQ beads (5 �m, 120 A pore size) from MicromioResources in tandem by a union. For the RPLC separation, 0.1%A in water and ACN were used as mobile phases of A and B, respec-ively, with the total flow rate of 200 nL/min. After the sampleoaded onto the C18 column automatically, a binary gradient sepa-ation with mobile phase B from 5% to 35% was performed in 30 min.andem mass spectrometry was performed on a Finnigan LTQ lin-ar ion trap mass spectrometer (Thermo, Finnigan, San Jose, CA)quipped with an ESI nanospray source. The LTQ instrument wasperated in positive ion mode with a spray voltage of 1.8 kV appliedn the cross union.

The digested peptide of BS and human plasma protein coronaere resuspended in 100 �L 0.1% FA solution and then the analy-

is was carried out by the 1-D nano LC-MS/MS system with 5 �Lf the resuspended peptides injected each time. A Waters Nano-CQUITY UPLC system was used to deliver the mobile phases of.1% FA in water (mobile phase A) and ACN (mobile phase B) withhe total flow rate of 350 nL/min and a binary separation gradientf mobile phase B changed from 2% to 30% in 60 min. Tandem masspectrometry was performed on a triple TOF 5600 mass spectrom-try (AB SCIEX, USA), operated in positive ion data dependent (IDA)ode with one MS survey scan followed by 30 MS/MS scans using

120 s exclusion window. The scan range of full MS was set for m/zrom 350 to 1250, while the MS/MS was set for m/z from 100 to500.

.7. Database searching

The MS/MS spectra acquired by the LTQ linear ion trap masspectrometer were submitted to SEQUEST for database searching.he MS/MS spectra acquired from BSA were searched against aatabase containing BSA and trypsin from International Protein

ndex. The database searching parameters were set as follows.

rypsin was selected as the enzyme, with KR/P as the cleavageite. Enzyme limits were set to the fully enzymatic and cleavingt both ends. The tolerance of missed cleavage was 2. Cysteineesidues were searched as the fixed modification of 57.0215 Da,

A 1334 (2014) 55–63 57

while methionine residues were as the variable modification of15.9949 Da. The peptides from BSA were considered positivelyidentified, if �Cn were higher than 0.1, and the Xcorr were higherthan 1.9, 2.2 and 3.75 for singly, doubly and triply charged peptides,respectively. The spectral counts for BSA and self-digested trypsinwould be obtained from the positively identified peptides.

The MS/MS spectral acquired by the triple TOF 5600 mass spec-trometry were searched by the “Paragon Algorithm” [33] of theProtein Pilot software (AB SCIEX, USA) with trypsin specificity in“Thorough ID” mode. The peptide false discovery rate (FDR) wascontrolled less than 1% by setting the criterion of correspond-ing confidence of global FDR of hit. The proteins were identifiedby the obtained peptides (FDR < 1%) and then the spectral countsfor proteins identified could be obtained. The results of BS pro-tein corona were searched against International Protein Index(ipi.bovine.3.54.fasta), while the results of human plasma pro-tein corona were searched against the International Protein Index(ipi.human.3.80.fasta). Proteins listed in the study were identifiedwith at least 2 unique peptides.

3. Results and discussion

3.1. Characterization of the magnetic Fe3O4 nanoparticles fortrypsin immobilization

The synthesized raw Fe3O4 nanoparticles, the SiO2 coatedFe3O4 nanoparticles for the preparation of immobilized trypsinwere characterized by the transmission electron microscopy (TEM,Fig. 1). The TEM image in Fig. 1A displays the uniform mag-netic Fe3O4 microspheres synthesized through the solvothermalmethod. Fig. 1B presents the TEM images of the synthesized 250 nmFe3O4 nanoparticles after coating of silica layer, the zeta potential ofwhich was ca. −19.5 mV. After the modification of the SiO2 coatedFe3O4 nanoparticles by amine groups [29], the zeta potential of theamine functionalized Fe3O4@SiO2 nanoparticles was ca. 34.4 mV,indicating the turnover of the exposed negatively charged silanolgroups to the positively charged amine groups. After that, the aminemodified Fe3O4@SiO2 nanoparticles were further applied to immo-bilize trypsin with the calculated amount of immobilized trypsin ofca. 65 �g/mg of the magnetic nanoparticle by monitoring the con-centrations of trypsin solutions before and after the immobilizationreaction with the magnetic nanoparticles.

3.2. The on-bead tryptic digestion of BSA adsorbed on commercialFe3O4 nanoparticles

To evaluate the on-bead tryptic digestion performance of thetrypsin immobilized on the synthesized magnetic Fe3O4 nanopar-ticle, the standard BSA adsorbed on the surface of the commercialFe3O4 nanoparticles was used to mimic the protein corona.After the incubation of the commercial Fe3O4 nanoparticles withstandard BSA solution, the resulting BSA-nanoparticle complex wascollected and washed with PBS buffer for at least 3 times. Then, theprepared BSA-nanoparticle complex was directly digested by theimmobilized trypsin with different time periods. As a comparison,the free trypsin was also applied to digest the BSA-nanoparticlecomplex. With the immobilized trypsin, it was observed that thecalculated sequence coverage for the standard BSA were ca. 46%,58% and 63% for incubation time of 15 min, 30 min and 1 h, respec-tively, which were significantly higher than that (15%, 27% and32%) of using free trypsin (Fig. 2A). For free trypsin, even taking

a long time of digestion period, the sequence coverage for BSA wasca. 52%, which was yet slightly lower than those of using thetrypsin immobilized on magnetic nanoparticles for only 30 minand 1 h, respectively, indicating an improved tryptic digestion

58 Z. Hu et al. / J. Chromatogr. A 1334 (2014) 55–63

gnetic

eo3

tilaftstifos

Faa

Fig. 1. TEM images of the synthesized 250 nm Fe3O4 ma

fficiency for immobilized trypsin with ca. 12–24 folds shortingf the digestion time from overnight (12–24 h) to 1 h or even0 min.

On the other hand, the self-digestion would occur for freerypsin in the chronic digestion process, which may not onlympact the efficiency of digestion but also the identification ofow abundant proteins by MS. The monitored sequence coveragettributed to the self-digestion of trypsin was ca. 70% and 32%or free trypsin digestion overnight and immobilize trypsin diges-ion for 1 h, respectively (Fig. 2B). Correspondingly, the obtainedequence coverage of BSA was ca. 52% and 63%, respectively, forhe free trypsin and the immobilized trypsin as well. This result

ndicates the less self-digestion and the higher identification per-ormance of mass spectrometry of using the trypsin immobilizedn the magnetic nanoparticle as compared with the free trypsin inolution.

ig. 2. The sequence coverage of standard BSA on the commercial Fe3O4 nanoparticles (3nd the sequence coverage of self-digested trypsin during the digestion of standard BSA and the immolized trypsin for 1 h (B).

nanoparticles (A), and coated with the shell of SiO2 (B).

3.3. The identification of BS protein corona on the Fe3O4nanoparticle

To further evaluate the on-bead digestion performance of thetrypsin immobilized on the magnetic nanoparticle, the BS pro-tein corona was applied as the model protein corona consistingof complex proteins from BS bound on the commercial Fe3O4nanoparticles with the size of 30 nm. Briefly, the BS protein coronawas prepared by incubating the 30 nm Fe3O4 nanoparticles with BSunder a shaking of 1400 rpm at 37 ◦C for 1 h, followed by washingthe nanoparticles with PBS buffer for at least 3 times. To identifythe proteins of the BS protein corona on the Fe3O4 nanoparticles,

the yielded composite of BS protein corona@Fe3O4 nanoparticleswas on-bead digested by the immobilized trypsin on the magneticnanoparticle for 1 h at 37 ◦C. As a comparison, the same amountsof the BS protein corona were also digested by free trypsin for 1 h

0 nm) digested by free trypsin and immobilized trypsin for different time points (A)dsorbed on the commercial Fe3O4 nanoparticles (30 nm) by free trypsin overnight

Z. Hu et al. / J. Chromatogr. A 1334 (2014) 55–63 59

Table 1The identification of BS proteins bound on the commercial Fe3O4 nanoparticles (30 nm) via three different protocols (*indicates 13% increase in the identified proteinscompared with free trypsin digestion overnight).

No. of identifiedbovine serumproteins

Unique peptidesfrom trypsinself-digestion

Unique peptidesexcept that fromtrypsin self-digestion

Spectral countsfrom trypsinself-digestion

Total spectralcounts excepttrypsin

Free trypsin @ overnight 153 574 3868 1297 6417Free trypsin @ 1 h 120 301 2130 937 3355Immobilized trypsin @ 1 h 173* 155 4250 308 7322

Fig. 3. (A) BS protein corona on the commercial 30 nm Fe3O4 nanoparticles digested by free trypsin for 1 h (Green Circle) and overnight (Blue Circle), and the immobilizedt by imf identir

ait1dtpd

Fp

rypsin for 1 h (Red Circle); (B) Spectral counts of BS proteins respectively digestedor the 30 nm Fe3O4 nanoparticles. Each point represents the individual BS protein

eferred to the web version of the article.)

nd overnight, respectively. As a result, there were 173 BS proteinsdentified from the protein corona on the 30 nm Fe3O4 nanopar-icles with the digestion of the immobilized trypsin within only

h. While, there were only 120 BS proteins identified with theigestion of using free trypsin within the 1 h. Even the digestion

ime prolonged to the usual 24 h for free trypsin, the number ofroteins (153 BS proteins) identified by free trypsin digestion yetid not reach that of using immobilized trypsin (Table 1). As shown

ig. 4. MW distributions of the total peptides (A) and unique peptides (B) identified for BSrotocols.

mobilized trypsin and free trypsin for 1 h versus that by the free trypsin overnightfied. (For interpretation of the color information in this figure legend, the reader is

in Fig. 3A, the overlap of the identified BS proteins from the com-mercial Fe3O4 nanoparticles was nearly up to 80% (128 proteins)for the on-bead digestion of using immobilized trypsin for 1 h andfree trypsin overnight. On the other hand, as shown in Fig. 3B, thespectral counts of proteins except trypsin identified by the diges-

tion of immobilized trypsin for 1 h were nearly the same as thatidentified by the digestion of free trypsin overnight, while whichidentified by the digestion of free trypsin for 1 h were much lower.

proteins on the commercial Fe3O4 nanoparticles obtained from different digestion

60 Z. Hu et al. / J. Chromatogr. A 1334 (2014) 55–63

Fig. 5. (A) BS protein corona on the 250 nm Fe3O4 nanoparticles digested by free trypsin for 1 h (Green Circle) and overnight (Blue Circle), and the immobilized trypsin for 1 h(Red Circle); (B) Spectral counts of BS proteins respectively digested by immobilized trypsin and free trypsin for 1 h versus that by the free trypsin overnight for the 250 nmF interw

TtcowtofbbBc111fidaFrp[BwnfioiiftaFfc(ctp2

nanoparticles but also impact the identification of protein coronaby mass spectrometry. As a result, the reduced self-digestion oftrypsin immobilized on the magnetic Fe3O4 nanoparticles may be

e3O4 nanoparticles. Each point represents the individual BS protein identified. (Foreb version of the article.)

his result indicates the high efficiency of immobilized trypsin onhe magnetic nanoparticles for the on-bead digestion of proteinorona from nanoparticles. Also, as shown in Fig. 4, the numbersf identified peptides digested by immobilized trypsin for only 1 here similarly equal to that by free trypsin digestion overnight in

he ranges of MW distributions. It can be seen that the numbersf peptides (Fig. 4A) and unique peptides (Fig. 4B) obtained fromree trypsin digestion for 1 h were significantly much lower thanoth of overnight free trypsin digestion overnight and 1 h immo-ilized trypsin digestion. On the other hand, the digestion of theS protein corona bound on 250 nm Fe3O4 nanoparticles was alsoarried out with the immobilized trypsin for 1 h, free trypsin for

h and overnight, respectively (Fig. 5A). We found that 96 and27 proteins were identified with the digestion of free trypsin for

h and overnight, respectively, while 140 proteins were identi-ed by the digestion of immobilized trypsin for 1 h (Fig. 5A, theetailed procedures were shown in Supporting Information). Therere 153 and 127 BS proteins identified on the 30 nm and 250 nme3O4 nanoparticles by the digestion of free trypsin overnight,espectively. It was possibly attributed to the different bindingatterns of proteins on nanoparticles with different particle size24]. This phenomenon was also observed in the identification ofS protein corona by the digestion of immobilized trypsin for 1 h,ith 173 and 140 BS proteins on the 30 nm and 250 nm Fe3O4anoparticles, respectively. The increased number of the identi-ed BS proteins from 153 to 173 on 30 nm and that 127 to 140n 250 nm Fe3O4 nanoparticles confirm the higher efficiency ofmmobilized trypsin. The examined decline tendency of BS bind-ng on the 30 nm and 250 nm Fe3O4 nanoparticles either by theree or the immobilized trypsin implies the unbiased digestion ofhe immobilized trypsin for proteins on different size nanoparticless compared with the standard overnight free trypsin digestion.urthermore, the spectral counts for the BS proteins identifiedrom the digestion of immobilized trypsin for 1 h were linearlyorrelated with that from the digestion of free trypsin overnightthe diagonal line for the red color points in Fig. 5B), which indi-

ated the highly coincident digestion efficiency of the immobilizedrypsin as compared to the free trypsin digestion overnight forroteins on nanoparticles with the size changed from 30 nm to50 nm.

pretation of the color information in this figure legend, the reader is referred to the

Moreover, the self-digestion of trypsin during the on-beaddigestion of protein corona by immobilized trypsin and free trypsinwas further investigated. As shown in Fig. 6, the numbers of theidentified unique peptides of trypsin from immobilized trypsinand free trypsin were 155, 301 and 574, respectively for diges-tion periods of 1 h, 1 h and overnight. Correspondingly, the detectedspectral counts of trypsin were 308, 937 and 1297 during the on-bead digestion with immobilized trypsin for 1 h, free trypsin for 1 hand overnight, respectively. This result clearly indicates that theself-digestion of trypsin has been significantly reduced after theimmobilization of trypsin on the Fe3O4 nanoparticle as compar-ing with the free trypsin during the on-bead digestion of proteincorona. It has to be noted that the self-digestion of trypsin wouldnot only influence the digestion efficiency of protein corona from

Fig. 6. The identified unique peptides and spectral counts for trypsin arising fromself-digestion of immobilized trypsin for 1 h, free trypsin for 1 h and overnight duringthe on-bead digestion of BS proteins on the 30 nm Fe3O4 nanoparticles.

Z. Hu et al. / J. Chromatogr. A 1334 (2014) 55–63 61

Fig. 7. The reproducibility for the on-bead tryptic digestion by immobilized trypsin compared with that by free trypsin (NO. of proteins shown here were identified by atl on thw

aBadwTobew

interfacial physicochemical properties of the nanoparticles inpractical biological systems, which may play essential roles inthe biological effects such as the adsorption, blood circulation,

F(

east two unique peptides). Three process replicates of identified BS proteins boundith free trypsin (A) and immobilized trypsin (B).

ccounted for the 13% increase of BS proteins identified from theS protein corona on the commercial 30 nm Fe3O4 nanoparticless compared with free trypsin digestion. Additionally, the repro-ucibility of the identification of protein corona on nanoparticlesas evaluated by using free trypsin and immobilized trypsin for 1 h.

he overlap of the triplicating identification of BS protein coronan the commercial Fe3O4 nanoparticles was ca. 73% for the immo-ilized trypsin, which was ca. 62% for free trypsin, indicating anlevated reproducibility of the on-bead digestion of protein corona

hen using immobilized trypsin (Fig. 7).

ig. 8. Classification of the human plasma proteins bound on the commercial Fe3O4 nanoC) by GoMiner and Panther database.

e commercial Fe3O4 nanoparticles (30 nm) through the on bead digestion protocol

3.4. The analysis of human plasma protein corona by theimmobilized trypsin

Nanoparticles with sizes smaller than 100 nm were extensivelyapplied in drug delivery and cell tracking etc. The protein coronaformed on the surface of nanoparticles would represent the first

bio-distribution, elimination as well as the cellular uptake and

particles (30 nm) according to protein class (A), biological process (B) and pathway

6 togr.

cobodicct

wFocsta

N

ctftptSaZga3ab(hinsiFaot

tapttrtcciiratFwap

[

[[

[

[

[

[[

[[

[

[

2 Z. Hu et al. / J. Chroma

ytotoxicity etc. Herein, the human plasma protein corona formedn the commercial Fe3O4 nanoparticles (∼30 nm) were analyzedy utilizing the on-bead digestion of using the immobilized trypsinn magnetic nanoparticles and the following mass spectrometryetection. Briefly, the commercial Fe3O4 nanoparticles were firstly

ncubated with human plasma to obtain the human plasma proteinorona. The resulting human plasma protein-Fe3O4 nanoparticleomplexes were then digested by the immobilized trypsin andhen analyzed by LC-MS/MS.

Via the detection of LC-MS/MS, 149 human plasma proteinsere identified in human plasma protein corona on the 30 nm

e3O4 nanoparticles (detailed proteins identified see Table S1). Inrder to take into account of the protein size and evaluate the realontribution of each protein to the hard corona composition, thepectral count (SpC) of each protein identity was normalized tohe protein mass and described as the relative protein quantity bypplying the following equation [24,34]:

ormalized SpCk =(

(SpC/Mw)k∑ni=1(SpC/Mw)i

)× 100

Where, normalized SpCk is the percentage normalized spectralount for protein k, SpC is the spectral count identified, and Mw ishe molecular weight in kDa for the protein k (the normalizied SpCor human plasma proteins identified see Table S1). It was foundhat the abundance of albumin were the 32th of the most abundantroteins in protein corona on the commercial Fe3O4 nanoparticleshough the albumin is the most abundant protein in plasma (Table1). Although proteins such as apolipoprotein B-100/E, fibrinogenlpha/beta chain etc. have been detected in protein coronas on SiO2,nO, TiO2, etc. [24,35], proteins including prenylcysteine oxidase 1,lutathione peroxidase 3, multimerin-1, monocyte differentiationntigen CD14, etc. were found only in the protein corona on the0 nm Fe3O4 nanoparticles. Moreover, we noticed that the neg-tively charged proteins (pI < 7) in plasma seemed to prefer theinding on the negatively charged Fe3O4 nanoparticles (pH 7.4)Fig. S1). This observation is similar to the report of the binding ofuman plasma protein on negatively charged nanoparticles of sil-

ca [24]. The phenomenon of the negatively charged proteins on theegatively charged Fe3O4 nanoparticles might be explained as theequential adsorption of proteins on nanoparticles that the pos-tively charged proteins would adsorb on the negatively chargede3O4 nanoparticles at first and then followed with the subsequentdsorption of negatively charged proteins on the positively chargeduter layer proteins. Indeed, the exact explanation of this adsorp-ion phenomenon needs to be further investigated.

Furthermore, the GoMiner and Panther database were appliedo analyze the plasma proteins according to protein class, pathwaynd biological processes. As demonstrated in Fig. 8A, the adsorbedroteins on the 30 nm Fe3O4 nanoparticles were mainly classifiedo enzyme modulator, hydrolase and the defense/immunity pro-eins, and the proteins of transfer/carrier protein and transporter,espectively. Although the adsorption of defense/immunity pro-eins which consists of complement factor H, apolipoprotein C-I,oagulation factor IX and serum amyloid P-component, etc., wereonsidered to facilitate the recognition by macrophages and elim-nation from human organism [36,37], proteins including vinculin,nsulin-like growth factor-binding protein 5 and periostin, etc., areelated in mediating cell adhesion and cell-matrix interaction, etc.nd may be beneficial for the internalization of nanoparticles intoarget cells. The classification of human plasma proteins on the

e3O4 nanoparticles according to pathways and biological processere also illustrated in Fig. 8B and C. It was observed that the

dsorbed proteins on the 30 nm Fe3O4 nanoparticles were mainlyarticipated in critical pathways and biological processes, which

[

[

A 1334 (2014) 55–63

could impact the following bio-distributions and cellular responsesof nanoparticles.

4. Conclusions

The enzymatic digestion of protein corona on nanoparticles wasinvestigated by the immobilized trypsin on magnetic nanoparti-cles as well as the free trypsin. It was observed that the digestionof protein corona on the surface of nanoparticles could be effi-ciently accomplished within 1 h rather than the usual overnightdigestion by free trypsin. The self-digestion of trypsin was reducedsignificantly due to the immobilization of trypsin on the magneticnanoparticle, which guaranteed the high digestion efficiency and ahigh reproducibility for the identification of proteins for the com-position illustration of protein corona on nanoparticles. The humanplasma proteins on the commercial 30 nm Fe3O4 nanoparticleswere analyzed by LC-MS/MS with the on-bead digestion approachof using the immobilized trypsin. It was observed that the plasmaprotein corona bound on the Fe3O4 nanoparticles seemed relatedto some critical pathways and biological processes which may posethe potential impacts on the bio-distribution and fates of Fe3O4nanoparticles when they enter into biological system.

Acknowledgements

We thank the financial supports of the National Natural ScienceFoundation of China (Nos. 21175134, 21375125) and the CreativeResearch Group Project of National Natural Science Foundation ofChina (21321064).

Appendix A. Supplementary data

Supplementary material related to this article can be found,in the online version, at http://dx.doi.org/10.1016/j.chroma.2014.01.077.

References

[1] S. Laurent, D. Forge, M. Port, A. Roch, C. Robic, L.V. Elst, R.N. Muller, Chem. Rev.108 (2008) 2064.

[2] M.G. Lines, J. Alloy Compd. 449 (2008) 242.[3] J.A. Barreto, W. O’Malley, M. Kubeil, B. Graham, H. Stephan, L. Spiccia, Adv.

Mater. 23 (2011) H18.[4] N.A. Saunders, Nanomedicine 6 (2011) 271.[5] L. Zhang, T.J. Webster, Nano Today 4 (2009) 66.[6] M.P. Monopoli, F.B. Bombelli, K.A. Dawson, Nat. Nanotechnol. 6 (2011) 11.[7] I. Lynch, K.A. Dawson, Nano Today 3 (2008) 40.[8] I. Lynch, A. Salvati, K.A. Dawson, Nat. Nanotechnol. 4 (2009) 546.[9] C.D. Walkey, W.C.W. Chan, Chem. Soc. Rev. 41 (2012) 2780.10] M.S. Ehrenberg, A.E. Friedman, J.N. Finkelstein, G. Oberdorster, J.L. McGrath,

Biomater. 30 (2009) 603.11] Z.J. Deng, Nat. Nanotechnol. 6 (2011) 39.12] W. Hu, C. Peng, M. Lv, X. Li, Y. Zhang, N. Chen, C. Fan, Q. Huang, Acs Nano 5

(2011) 3693.13] D. Dutta, S.K. Sundaram, J.G. Teeguarden, B.J. Riley, L.S. Fifield, J.M. Jacobs, S.R.

Addleman, G.A. Kaysen, B.M. Moudgil, T.J. Weber, Toxicol. Sci. 100 (2007) 303.14] M.A. Dobrovolskaia, P. Aggarwal, J.B. Hall, S.E. McNeil, Mol. Pharmaceut. 5

(2008) 487.15] A. Lesniak, F. Fenaroli, M.R. Monopoli, C. Aberg, K.A. Dawson, A. Salvati, Acs

Nano 6 (2012) 5845.16] C. Vauthier, B. Persson, P. Lindner, B. Cabane, Biomater. 32 (2011) 1646.17] M. Lundqvist, J. Stigler, G. Elia, I. Lynch, T. Cedervall, K.A. Dawson, Proc. Natl.

Acad. Sci. 105 (2008) 14265.18] W. Shang, J.H. Nuffer, J.S. Dordick, R.W. Siegel, Nano Lett. 7 (2007) 1991.19] S.H.D. Lacerda, J.J. Park, C. Meuse, D. Pristinski, M.L. Becker, A. Karim, J.F. Doug-

las, Acs Nano 4 (2010) 365.20] T. Cedervall, I. Lynch, M. Foy, T. Berggård, S.C. Donnelly, G. Cagney, S. Linse, K.A.

Dawson, Angew. Chem. Int. Ed. 46 (2007) 5754.21] T. Cedervall, I. Lynch, S. Lindman, T. Berggard, E. Thulin, H. Nilsson, K.A. Dawson,

S. Linse, Proc. Natl. Acad. Sci. 104 (2007) 2050.22] S. Lindman, I. Lynch, E. Thulin, H. Nilsson, K.A. Dawson, S. Linse, Nano Lett. 7

(2007) 914.23] C. Rocker, M. Potzl, F. Zhang, W.J. Parak, G.U. Nienhaus, Nat. Nanotechnol. 4

(2009) 577.

togr.

[

[

[

[

[[

[[

[[

[

Z. Hu et al. / J. Chroma

24] S. Tenzer, D. Docter, S. Rosfa, A. Wlodarski, J. Kuharev, A. Rekik, S.K. Knauer, C.Bantz, T. Nawroth, C. Bier, J. Sirirattanapan, W. Mann, L. Treuel, R. Zellner, M.Maskos, H. Schild, R.H. Stauber, Acs Nano 5 (2011) 7155.

25] J. Sund, H. Alenius, M. Vippola, K. Savolainen, A. Puustinen, Acs Nano 5 (2011)4300.

26] C.D. Walkey, J.B. Olsen, H. Guo, A. Emili, W.C.W. Chan, J. Am. Chem. Soc. 134(2012) 2139.

27] H. Zhang, K.E. Burnum, M.L. Luna, B.O. Petritis, J.-S. Kim, W.-J. Qian, R.J. Moore, A.Heredia-Langner, B.-J.M. Webb-Robertson, B.D. Thrall, D.G.I.I. Camp, R.D. Smith,J.G. Pounds, T. Liu, Proteomics 11 (2011) 4569.

28] S. Lin, G. Yao, D. Qi, Y. Li, C. Deng, P. Yang, X. Zhang, Anal. Chem. 80 (2008) 3655.29] L. Zhao, H. Qin, Z. Hu, Y. Zhang, R.a. Wu, H. Zou, Chem. Sci 3 (2012) 2828.

[

[[

A 1334 (2014) 55–63 63

30] L. Sun, Y. Li, P. Yang, G. Zhu, N.J. Dovichi, J. Chromatogr. A 1220 (2012) 68.31] H. Deng, X. Li, Q. Peng, X. Wang, J. Chen, Y. Li, Angew. Chem. Int. Ed. 44 (2005)

2782.32] Y. Li, B. Yan, C. Deng, J. Tang, J. Liu, X. Zhang, Proteomics 7 (2007) 3661.33] I.V. Shilov, S.L. Seymour, A.A. Patel, A. Loboda, W.H. Tang, S.P. Keating, C.L.

Hunter, L.M. Nuwaysir, D.A. Schaeffer, Mol. Cell. Proteomics 6 (2007) 1638.34] M.P. Monopoli, D. Walczyk, A. Campbell, G. Elia, I. Lynch, F.B. Bombelli, K.A.

Dawson, J. Am. Chem. Soc. 133 (2011) 2525.35] Z.J. Deng, G. Mortimer, T. Schiller, A. Musumeci, D. Martin, R.F. Minchin, Nano-

technology 20 (2009).36] P.P. Karmali, D. Simberg, Expert Opin. Drug Del. 8 (2011) 343.37] M.A. Dobrovolskaia, D.R. Germolec, J.L. Weaver, Nat. Nanotechnol. 4 (2009) 411.