J O U R N A L O F P R O T E O M I C S 8 4 ( 2 0 1 3 ) 1 7 6 – 1 8 4

Ava i l ab l e on l i ne a t www.sc i enced i r ec t . com

www.e l sev i e r . com/ loca te / j p ro t

Comparative proteomics of Hirschsprung's disease

Yang Fana, Lili Wanga, Yi Zhanga, Tianchu Huanga, Hui Lia, Hui Gua,Weilin Wangb, Zhengwei Yuana,⁎aKey Laboratory of Health Ministry for Congenital Malformation, Shengjing Hospital, China Medical University, Shenyang 110004, PR ChinabDepartment of Pediatric Surgery, Shengjing Hospital, China Medical University, Shenyang 110004, PR China

A R T I C L E I N F O

⁎ Corresponding author at: Key Laboratory ofNo. 36, Sanhao Street, Heping District, Sheny

E-mail address: yuanzw@hotmail.com (Z.

1874-3919/$ – see front matter © 2013 Elseviehttp://dx.doi.org/10.1016/j.jprot.2013.03.024

A B S T R A C T

Article history:Received 19 November 2012Accepted 22 March 2013Available online 3 April 2013

Keywords:2-D electrophoresis

Hirschsprung's disease (HSCR) is a developmental disorder of the enteric nervous systemcharacterized by aganglionosis in distal gut. The estimated population incidence of HSCR is1/5000 live births, but higher in Asian populations. As the disease mainly manifested bowelmotility disturbance, the underlyingmechanism is unknown. Furthermore, in the long term upto 75% of patients showed unsatisfactory postoperative bowel function like incontinence orconstipation, and 10% required a permanent colostomy. Improved therapy of HSCR is required,but the pathophysiological mechanism for postoperative bowel dysfunction is not clear. In thisstudy, we perform a proteomics study in HSCR patients, expecting some findings in proteinalterations to provide more information to reveal the pathophysiological mechanisms ofdisturbed bowel function before and after surgery therapy. As a result, we identified 16 proteinsexpressed differently in aganglionic segment of HSCR patients. These proteins functiondiversely, and included cytoskeleton proteins, regulatory proteins and some enzymes.

Biological significanceIn the present study, we performed a 2-DE based proteomic research on HSCR patients, inorder to find some clue for the pathomechanism of bowel motility of HSCR disease. As acharacter of this study, we also compared the expression of altered proteins in ganglionicsegment of HSCR patients with that in normal children. Our results showed that somealtered proteins found in aganglionic segment had also changed their expression inganglionic segment comparing with normal children. This result suggested that theganglionic segment of HSCR patients was not completely normal, and this is importantbecause it provided more information to understand the pathophysiological mechanismsof bowel dysfunction and will help to the therapy of HSCR disease.

© 2013 Elsevier B.V. All rights reserved.

Hirschsprung's diseaseBowel motility

1. Introduction

Hirschsprung's disease (HSCR) is a developmental disorder ofthe enteric nervous system (ENS) characterized by the lack ofganglionic cells in distal gut (aganglionosis). The estimatedincidence of HSCR is 1/5000 live births, which is higher inAsian populations (2.8/10,000) [1]. Multiple genes such as RET,

Health Ministry for Congeang 110004, PR China. TeYuan).

r B.V. All rights reserved

NRG1, Hedgehog, and BMPs are reported to be involved in theetiology of HSCR [2–6]. However, as a disease characterized bymotility disorder of aganglionic segment of the gut, lessattention was paid to the pathophysiology underlying themotility dysfunction inHSCRwhich is still unclear. Furthermore,in the long term up to 75% of patients showed unsatisfactorypostoperative bowel function like incontinence or constipation,

nital Malformation, Shengjing Hospital, China Medical Universityl./fax: +86 024 23929903.

.

.

177J O U R N A L O F P R O T E O M I C S 8 4 ( 2 0 1 3 ) 1 7 6 – 1 8 4

and 10% required a permanent colostomy [7–9]. But the exactmechanism causing the post-operative motility impairment isalso unknown. Actually, the understanding of the pathophysi-ology for the gut motility dysfunction in children with HSCRmight help formulating novel treatments in the future.

Gut motility is a complex process mediated by interactionbetween intestinal smoothmuscle, interstitial cells of Cajal (ICC),and the ENS. ENS innervates the gastrointestinal tract intrinsi-cally, and its formation requires the co-ordinated migration,proliferation, differentiation and survival of neural crest cells(NCC) within the developing gut. During the establishment ofENS, environmental signals like SOX9, PAX3, neurotrophin-3 andSEMA3A play a critical role [10–12]. Aganglionic segment wasreported to be lack of nitrergic neurons in ENS causing inabilityto relax [13]. InHSCR, thedistal colon is devoidof enteric neuronsand glia which are derived from NCCs. The congenital defect ofENS development in HSCR may cause the fail in coordinationwith the gut smooth muscle, and participate in the pathophys-iology for the gut motility dysfunction.

Except for the congenital defect of ENS development,derangement of ICCs also plays a role in the gut motilitydysfunction of HSCR. There was a report that in the aganglioniccolon, the network of ICCs was disrupted and ICCs weremarkedly reduced [14].

As a powerful tool to investigate the mechanism of acomplex biological process, the proteomics approach can revealthe protein alterations corresponding to certain pathologicalcondition in an integratedway. Considering the highly complexpathophysiology underlying the gut motility dysfunction inHSCR, we performed a proteomics study in HSCR patients,expecting to disclose the protein alterations related to bowelmalfunction of HSCR.

2. Methods

2.1. Patients and specimens

Colon tissueswere obtained from32patients (including 30maleand 2 female patients) diagnosed as HSCR in the Department ofPediatric Surgery, Shengjing Affiliated Hospital of China Medi-cal University. The age of the patients ranged from 1 to96 months old, with an average age of 12.8 months. HSCR wasdiagnosed by a barium enema x-ray and suction rectal biopsy.All HSCR patients showed aganglionosis restricted to thesigmoid colon. The aganglionic and ganglionic segments ofthe colon were immersed in liquid nitrogen immediately aftersurgery removal, and then stored in −80 °C. Colon tissues from8 patientswere used in 2-DE analysis, and those from the left 24patients were used in verification by Western blot. In addition,there were six cases of newborn infants died from non-nervousor digestive system diseases as normal control. The study wasapproved by the local ethical committee and all the subjectsinvolved in the study gave written informed consent.

2.2. Extraction of proteins from colon tissues

For 2-DE analysis, colon tissues were homogenized in freshgrinding solution containing 10% TCA and 2 mM TBP(tributylphosphine) in acetone. Proteins were extracted and

precipitated by TCA/acetone. The homogenate was treatedwith precooled (−20 °C) solution of 10% TCA in acetone with20 mM DTT. Proteins were allowed to precipitate overnight at−20 °C. After centrifugation at 20,000 ×g for 15 min at 4 °C, thepellet was washed with ice-cold acetone containing 20 mMDTT. The supernatant was discarded and the pellet dried in aSPD1010 SpeedVac system (Thermo Savant, Holbrook, NY,USA). The pellet was resolubilized in lysis buffer (7 M urea,2 Mthiourea, 4% [V/W] CHAPS, 2% [V/V] IPG buffer [pH 3–10],40 mM 1,4-dithioerythritol, 1 mM PMSF and 1% [V/V] cocktailprotease inhibitor, Sigma). For immunoblot analysis, differentspecimens were homogenized with 5 volumes of lysis buffercontaining 50 mM Tris (pH 7.4),150 mM NaCl,1% TritonX-100,1% sodium deoxycholate,0.1% SDS and 1% (V/V) cocktailprotease inhibitor (Sigma) and sonicated as above. Then thesolution was centrifuged at 20,000 ×g for 45 min at 4 °C. Thesupernatant was used as protein extract for immunoblotanalysis. The protein concentration was determined by theBradford method, and then stored in aliquots at −80 °C.

2.3. 2-DE

2-DE was performed as our previous report [15]. We loaded900 μg protein extract onto an IPG strip (24 cm, pH 3–10; GEHealthcare, Uppsala, Sweden). For the first-dimension isoelectricfocusing, the IPG strip was rehydrated with 450 μl of solubilizedsample at 30 V for 12 h on an IPGphor (GE Healthcare). IEFfollowed amulti-step protocol: 100 V for 2 h, 300 V for3 h, 600 Vfor 2.5 h,1000 V for 2.5 h, gradient from 1000 V up to 8000 V in2 h and finally 8000 V for held for 66,000 Vh at 20 °C. The IPGstrips were equilibrated in 10 ml equilibration solutions (6 Murea, 30% glycerol, 2% sodium dodecyl sulfate [SDS], 115 mMTris–Cl [pH 8.8], 20 mM dithiothreitol [DTT]) for 15 min, thenequilibrated in the same solution containing 100 mMiodoacetamide instead of DTT. SDS-PAGE involved use of 12.5%polyacrylamide gels in the Ettan DALT twelve system (GEHealthcare). Following SDS-PAGE, gels were stained with mod-ified colloidal Coomassie Brilliant Blue (mcCBB) G-250 asdescribed [16]. For 2-DE, we pooled the protein extracts ofaganglionic and ganglionic colon tissues from 8 HSCR patients(Fig. 2). The experiment was repeated for three times.

2.4. Image acquisition and data analysis

CBB-stained gels were scanned by use of a PowerLook 2100XLimage scanner (Umax, Taiwan). Spot detection, quantification,and matching involved use of 2-D gel analysis software(ImageMaster 2D platinum 6.0, GE Healthcare) with theCBB-stained gels. Percentage of volume of a spot representinga particular protein was determined in comparison with thetotal proteins present in the area of interest (relative volume, %vol). % vol of spots was obtained from 3 parallel experiments.Spots with at least 1.5-fold difference in % vol showingstatistical significance (P < 0.05) were defined as differentiallyexpressed proteins and were excised for further analysis.

2.5. In-gel digestion and MALDI-TOF MS

Selected spots were chosenmanually. CBB-stained spots weredestained in 50% acetonitrile (ACN) in 25 mM ammonium

178 J O U R N A L O F P R O T E O M I C S 8 4 ( 2 0 1 3 ) 1 7 6 – 1 8 4

bicarbonate buffer and dried in the SpeedVac. The dried gelfragments were re-hydrated in trypsin solution (15 μg/ml) for 1 hat 4 °C, followed by the addition of 5 ml 25 mM ammoniumbicarbonate buffer to completely immerse the gel fragments.After incubation for 16 h at 37 °C, the digested peptides wereextracted from the gel fragments with use of 5% trifluoroaceticacid (TFA) and 2.5% TFA/50% ACN at 37 °C for 1 h separately.Tryptic peptideswere finally dissolved inMALDImatrix (5 mg/mlα-cyana-4-hydroxycinnamic acid in 0.1% TFA and 50% ACN),spotted onto 192-well stainless steel MALDI target plates, andanalyzed by use of an ABI 4800 Proteomics Analyzer MALDI-TOF/TOF mass spectrometer (Applied Biosystems, USA). The MS andMS/MS spectra were searched against the International ProteinIndex (IPI) human database, version 3.18, with use of GPSExplorerTM v3.0 and MASCOT database search algorithms(version 2.0) with the search criteria trypsin specificity, cysteinecarbamidomethylation (C) and methionine oxidation (M) asvariable modifications; 1 trypsin miscleavage allowed; 0.2-DaMS tolerance; and 0.3-DaMS/MS tolerance. Protein identificationswere accepted with a Mowse score ≥ 60 and a P value < 0.05.

2.6. Immunoblot analysis

In total, 50 μg protein extract was separated by 12.5% SDS-PAGEand then transferred with Tris–HCl methanol (20 mM Tris,150 mM glycine, 20% methanol) onto polyvinylidene difluoridemembranes (Millipore, USA) in a trans-blot electrophoresistransfer cell (Bio-Rad). Blotting was probed with antibodiesagainst ATP synthase subunit beta (ATP5B, abcam, ab85068),calponin-1 (CNN1, abcam, ab78491), Heat shock protein beta-1(HSPB1, Enzo, ADI-SPA-803), Annexin A2 (ANXA2, proteintech,11256-1-AP) or GAPDH (Kangcheng, E021010-02) as described[17]. All immunoblots were run at least in triplicate. 24 patientsdifferent from those in 2-DEanalysiswereused. Visualization ofthe antigen–antibody complexes involved use of enhancedchemiluminescence reagents (GE healthcare). Detected bandswere quantified by Gel-pro 4.0 software (Media Cybernetics, LP).The relative density of each protein was calculated by dividingthe optical density value of each protein by that of loadingcontrol (GAPDH).

2.7. Statistical analysis

Data are expressed as means ± SD. Statistical significance wasdetermined by using Student's t test, and statistical significancewas considered at the P < 0.05.

3. Results

3.1. Protein profiles of aganglionic and ganglionic segmentsin Hirschsprung's disease

In order to validate the reproducibility, 2-DE was performed intriplicate. The proteome of aganglionic and ganglionic seg-ments of affected colon in patients with HSCR contained745–930 detectable proteins on a single mcCBB-stained 2-DEgel, with an average matching rate of 80.1% on total gels(Fig. 1). 2-DE images with high resolution and reproducibilitywere obtained, and representative images are shown in Fig. 1.

We found 29 protein spots with at least 1.5-fold difference in% vol (P < 0.05) between aganglionic and ganglionic segmentsand dissected them for further analysis.

Of 29 dissected spots, 8 spots were rejected because of lowMowse scores, and the left 21 spots were identified by MS asdifferentially expressed. Among the 21 spots, six spots wereidentified to be the same protein although their expressionswere different (Tables 1, 2, see Supplemental Table for detailedinformation). Totally, 16 proteins were identified as differentlyexpressed.

3.2. Immunoblot analysis of selected proteins in HSCRpatients

We selected 4 proteins for immunoblot analysis (Fig. 2). Inaganglionic segment, HSPB1 expression increased by 1.7 fold(P < 0.05) and CNN1 was up-regulated by 4.4-fold (P < 0.01,Table 3, Fig. 2B,C). ANXA2 and ATP5B were down-regulated by45% and 68% respectively (P < 0.05, Table 3, Fig. 2B,C), whichwere consistent with 2-DE analysis (Fig. 2A and Table 1). In 24patients used for verification, majority of patients showedconsistent results with 2-DE. In detail, 71% of patients (17 in 24patients) showed up-regulated expression in CNN1 and HSPB1,while 75% and 79% of patients (18 in 24 patients and 19 in 24patients) showed down-regulated expression in ANXA2 andATP5B respectively.

3.3. Immunoblot analysis of selected proteins in newborninfants

We also analyzed the expression for ATP5B, HSPB1, CNN1 andANAX2 in the colon of newborn infants without any disease ofgut and nervous system. In all of our aganglionic colonsamples that are restricted to the sigmoid colon, we chosethe transversal and sigmoid colons of newborn infants forcomparing study (Fig. 3). There were no differences for theexpression of the four proteins between the two segments ofnormal colon (Table 3, Fig. 3B). But we found an increasedATP5B and decreased ANXA2 expression in ganglionic seg-ment of HSCR patient comparing that in the transversal colonof newborn infants. ATP5B expression increased by 2.6 folds,while ANXA2 decreased by 88% in ganglionic segment of HSCRpatients. There were no differences showed for the left twoprotein expression between the two groups (Table 3, Fig. 3C).

4. Discussion

HSCR is a multigene disease caused by congenital absence ofganglion cells in the myenteric and submucosal plexuses ofthe gastrointestinal tract (aganglionosis). By far, many genesare reported to be involved in the etiology of HSCR. Themechanism of motility dysfunction in HSCR is still unclearalthough colonic motility dysfunction is a main manifesta-tion. Furthermore, for HSCR patients, persistent postoperativedisturbances in bowel motility even after the operation areone of the major problems, which the underlying patho-mechanism remains unclear, too. In this study, we used 2-DEbased proteomics study to investigate the differential proteinexpression between the aganglionic and ganglionic segments

Fig. 1 – 2-DE of protein profile for aganglionic and ganglionic segments of colon tissue from HSCR patients. RepresentativeCoomassie-stained 2-DE gels of expressionmaps of proteins in aganglionic segment (A) and ganglionic segment of colon tissuefrom HSCR patients (B). Numbers indicate the differently expressed protein spots listed in Table 1. Accession numbers areindicated. Star mark indicated protein spots rejected by low Mowse scores.

179J O U R N A L O F P R O T E O M I C S 8 4 ( 2 0 1 3 ) 1 7 6 – 1 8 4

of colon tissues from HSCR patients in order to get moreinformation about bowel motility disturbance. We identified 16different expression proteins, which included cytoskeletonproteins and some regulate the smooth muscle movement(Tables 1, 2). We confirmed four identified proteins by

immunoblot analysis. Furthermore, we also compared theexpression alterations of the four proteins between the normaltransversal colon and ganglionic segment of affected colonfrom the patients with HSCR. We demonstrated two proteinsaltered in ganglionic segment comparing with normal control.

Table 1 – Differentially expressed proteins identified by MOLDI-TOF MS.

Spot no. Protein name Accession no. Mr(kDa)/pI

Mascotscore

Matchedpeptides

Cover% Expressionalfold change

Pvalue

1 Keratin, type II cytoskeletal 1 (KRT1) IPI00220327 66/8.2 99 19 27 3.1↑ 0.0182 Calponin-1 (CNN1) IPI00021264 33/9.1 121 11 54 3.4↑ 0.0213 Heat shock protein beta-1 (HSPB1) IPI00025512 22/5.9 299 10 76 1.7↑ 0.00134 Transgelin (TAGLN) IPI00216138 22.5/8.9 191 14 40 2.2↑ 0.0195 Transgelin (TAGLN) IPI00216138 22.5/8.9 184 15 67 3.0↑ 0.0086 Transgelin (TAGLN) IPI00216138 22.5/8.9 331 16 32 2.0↑ 0.0227 Transgelin (TAGLN) IPI00216138 22.5/8.9 176 14 48 4.8↑ 0.0028 Probable arginyl-tRNA synthetase (RARS2) IPI00549566 65/8.4 61 17 11 3.8↑ 0.0199 SH3 domain-binding glutamic acid-rich-like

protein (SH3BGRL)IPI00025318 12.7/5.2 135 6 72 1.3↓ 0.009

10 ALB 23 kDa protein IPI00878282 23/5.9 108 9 31 2.2↓ 0.02411 Tropomyosin isoform 1/alpha chain isoform 3 IPI00018853/

IPI0060453728/4.932/4.7

388/514 22/27 67/77 1.4↓ 0.019

12 Carbonic anhydrase 1 (CA1) IPI00215983 29/6.6 478 18 72 1.4↓ 0.02213 Sialic acid synthase (NANS) IPI00147874 40/6.3 113 15 30 1.9↓ 0.02414 Annexin A2 (ANXA2) IPI00455315 39/7.57 282 20 49.8 2.3↓ 0.01615 Fibrinogen beta chain (FGB) IPI00298497 56/8.5 90 13 44 1.4↓ 0.02516 Transgelin (TAGLN) IPI00216138 22.5/8.9 76 11 23 3.2↓ 0.02217 Transgelin (TAGLN) IPI00216138 22.5/8.9 160 13 38 × 0.001718 Lamin A/C (LMNA/C) IPI00514204 53/6.1 67 15 35 × 0.00419 IGHG1 protein (IGHG1) IPI00423463 52/8.6 80 11 15 × 0.01920 ATP synthase subunit beta (ATP5B) IPI00303476 57/5.3 235 11 28 × 0.01721 Desmin (DES) IPI00465084 54/5.2 242 22 41 × 0.020

Spot no. was defined according to spot positions in the 2-DE gel as indicated in Fig. 1. Expressional fold change: the relatively quantitativealterations of proteins were determined based on the spot relative volume (% vol) of protein. ↑ means that the protein level in aganglionicsegment increased compared with ganglionic segment; ↓ means that the protein level in aganglionic segment decreased compared withganglionic segment; × means that the protein did not express in ganglionic segment.

180 J O U R N A L O F P R O T E O M I C S 8 4 ( 2 0 1 3 ) 1 7 6 – 1 8 4

Our result might shed light on understanding the disease aswell as the pathomechanism of gut motility disturbance.

Colonic motility is a complex biological behavior requiringdelicate coordination and regulation. The movements ofbowel are due to coordinated contractions and relaxations ofthe internal circular smooth muscle layer and the externallongitudinal smooth muscle, which are controlled by myo-genic, neural (via ENS), and chemical control mechanisms[18,19]. In the differently expressed proteins, we found thatmany belonged to intermediate filament family (Tables 1, 2),such as keratin, type II cytoskeletal 1, lamin A/C, and desmin.Intermediate filament is an important component of cyto-skeletal system in cell, which serves as a structural frame-work [20] and plays an active role in force transmission [21].HSCR is a disease characterized by colonic motility disorder.Nemeth et al. reported the altered cytoskeleton in smoothmuscle of aganglionic bowel, in which lacking of desmin wasshowed [22]. Although we did not get the other two cytoskel-eton protein alterations in aganglionic colon that are reportedby Nemeth et al. for the limitation of 2-DE technique itself, wehad found several other altered cytoskeleton proteins. Thesealtered cytoskeletons in aganglionic tissue reflected the disor-der of mechanochemical signal in the affected bowel, and thatmay partially explain the tonic contraction of affected colon.Besides, smooth muscle cells were demonstrated to influenceneuronal development by co-culturingneuronswith aganglionicsmoothmuscle cells. Langer et al. found that the smoothmuscleof the aganglionic colon was less favorable for neuronaldevelopment than that of a normal colon [23]. The nerve plexusof myenteric and submucosal and enteric smooth muscle cells

may be involved in the pathomechanism for postoperative gutmotility disorder. Therefore, in our study the altered expressingproteins in intermediate filament family might attribute to theimpaired development of enteric neurons and glia, theninvolved in the disturbance of gut motility in HSCR patients.

Except for the altered proteins of cytoskeleton family, wealso found some differently expressed proteins that regulatesmooth muscle movement in function (Tables 1, 2), such asCNN1, HSPB1, tropomyosin and ANXA2. CNN1, and HSPB1were reported to be participated in smooth muscle contrac-tion [24,25]. Tropomyosin and ANXA2 were reported to playimportant roles in the development of ENS [26,27]. For thecoordinated regulation of smooth muscle contraction andrelaxation that is required for colonic motility, the motionanomaly of bowel in HSCR may be the result of the alteredproteins that regulated the smooth muscle movement.Therefore, we verified CNN1, HSPB1 and ANXA2 by Westernblot analysis. As shown in Fig. 2, majority of patients showedup-regulated CNN1 (17 in 24 patients) and HSPB1 (17 in 24patients) expression and down-regulated ANXA2 (18 in24 patients) expression, which were consistent with 2-DEanalysis. To investigate whether these proteins had alreadyaltered in ganglionic segment of HSCR patients as well as thealterations in aganglionic segments, we also compared thisprotein expression with normal controls coming from sixnewborn infants who died from non-nervous or digestivesystem diseases. In all of our samples that are restricted insigmoid colon, we chose sigmoid colon and transversal colonas control for aganglionic and ganglionic segments of HSCRpatients.

Table 2 – The function of differentially expressed proteins identified.

Spot no. Protein name Location a Function b

1 Keratin, type II cytoskeletal 1(KRT1)

Cell membrane May regulate the activity of kinases such as PKC and SRC via bindingto integrin beta-1 (ITB1) and the receptor of activated protein kinase C.

2 Calponin-1 (CNN1) Cytoplasm Thin filament-associated protein that is implicated in theregulation and modulation of smooth muscle contraction. It iscapable of binding to actin, calmodulin, troponin C and tropo-myosin. The interaction of calponin with actin inhibits theactomyosin Mg-ATPase activity.

3 Heat shock protein beta-1 (HSPB1) Cytoplasm Involved in stress resistance and actin organization.4–7, 16, 17 Transgelin (TAGLN) Cytoplasm A transformation and shape-change actin cross-linking protein

found in smooth muscle.8 Probable arginyl-tRNA synthetase

(RARS2)Mitochondrion Protein biosynthesis, aminoacyl-tRNA synthetase.

9 SH3 domain-binding glutamicacid-rich-like protein (SH3BGRL)

Cytoplasm/nucleus Act as a modulator of glutaredoxin biological activity.

10 ALB 23 kDa protein Extracellular space Albumin functions primarily as a carrier protein for steroids, fattyacids, and thyroid hormones and plays a role in stabilizingextracellular fluid volume.

11 Tropomyosin isoform 1/alphachain isoform 3

Cytoplasm Affect the stability of actin filaments and have been shown to beimplicated in various cellular functions including the regulationof cell transformation, cytokinesis, motility and morphogenesis.

12 Carbonic anhydrase 1 (CA1) Cytoplasm Reversible hydration of carbon dioxide. Can hydrate cyanamide tourea.

13 Sialic acid synthase (NANS) Cytoplasm Produces N-acetylneuraminic acid (Neu5Ac) and2-keto-3-deoxy-D-glycero-D-galacto-nononic acid (KDN). Can alsouseN-acetylmannosamine 6-phosphate andmannose 6-phosphateas substrates to generate phosphorylated forms of Neu5Ac andKDN, respectively. Lipopolysaccharide biosynthetic process.

14 Annexin A2 (ANXA2) Cytoplasm/secreted Calcium-regulated membrane-binding protein whose affinity forcalcium is greatly enhanced by anionic phospholipids. It binds twocalcium ions with high affinity. May be involved in heat-stressresponse. It may cross-link plasma membrane phospholipids withactin and the cytoskeleton and be involved with exocytosis.

15 Fibrinogen beta chain (FGB) Secreted Fibrinogen has a double function: yielding monomers thatpolymerize into fibrin and acting as a cofactor in plateletaggregation. Blood coagulation hemostasis.

18 Lamin A/C (LMNA/C) Cytoplasm/nucleus Structural molecule activity.19 IGHG1 protein (IGHG1) Secreted Complement activation, classical pathway, innate immune response.20 ATP synthase subunit beta

(ATP5B)Cytoplasm Produces ATP from ADP in the presence of a proton gradient

across the membrane ATP synthesis, hydrogen ion transport, andion transport.

21 Desmin (DES) Cytoplasm Desmin is a structural constituent of cytoskeleton. Functioning toform a fibrous network connecting myofibrils to each other andmuscle filament sliding.

a Location is based on information from http://www.uniprot.org/.b All the information about function comes from the SWISS-PROT, NCBI database.

181J O U R N A L O F P R O T E O M I C S 8 4 ( 2 0 1 3 ) 1 7 6 – 1 8 4

Calponin inhibits actin-activated myosin ATPase in smoothmuscle. Calponin-1 (CNN1) is a basic isoform of calponin whichis a calcium binding protein. Calcium plays a fundamental rolein the contraction of muscle and neuronal excitation of cells.CNN1 plays a role in contractile function, which is specific todifferentiated smooth muscle cells and up-regulated duringpost-natal development [28]. CNN1 is considered as a mechan-ical strain-induced contractile marker [29]. We found in ourpresent result that expression of CNN1 increased in aganglionicsegment (up-regulated by 4.4-fold), this might be an explana-tion for the motive dysfunction of affected bowel.

Heat shock protein beta-1 (HSPB1) is a member of themammalian small heat-shock protein family and plays a rolein protecting cells from environmental stresses by regulatingapoptosis and protein folding. HSPB1 interacts with F-actin

[30], and has also been found to regulate the actin-cytoskeletaldynamics [31,32]. Ibitayo et al. reported HSPB1 associationwith contractile proteins resulting in sustained smoothmuscle contraction [33]. Elevated HSPB1 was also reported toplay a role in protecting or repairing microfilaments in agedsmooth muscle cells [30]. We found an increased HSPB1expression in the aganglionic segment (up-regulated by1.7-fold), which is consistent with Gao's report [34]. The highlevel of HSPB1 expression in the aganglionic segment mightbe considered as a repairing respond to the smooth musclelacking of glia, and the altered expression may change thecontractile ability of smooth muscle cells, hence disturbcolonic motor activity. It is worth to be mentioned that Gaoet al. also performed a comparative proteomics research onHSCR disease. In our study, we improved Gao's method,

Table 3 – Expression of ATP5B, HSPB1,CNN1 and ANX2 incolon of HSCR patients and normal newborn infants.

Proteins Mean value of relative density a(mean ± SD)

Sample from patients(24) b

Sample from normalcontrol (6) c

Ganglionicsegment

Aganglionicsegment

Transversalcolon

Sigmoidcolon

ATP5B 2.41 ± 1.19 0.77 ± 0.15 ⁎ 0.67 ± 0.41△ 0.69 ± 0.19HSPB1 2.96 ± 1.19 5.03 ± 1.58 ⁎ 2.03 ± 0.71 2.27 ± 0.10CNN1 1.28 ± 0.22 6.92 ± 1.89 ⁎⁎ 2.33 ± 1.06 2.21 ± 0.13ANXA2 0.17 ± 0.07 0.09 ± 0.02 ⁎ 1.34 ± 0.65△ 1.19 ± 0.07

a The mean value of relative density is calculated by dividing theoptical density value of eachprotein by that of loading control (GAPDH).b Samples come from ganglionic and aganglionic segments ofHSCR patients. The number in the parentheses indicated thenumber of patients used in the study.c Samples come fromnormal newborn infants died fromnon-nervousor digestive systemdiseases. The number in the parentheses indicatedthe number of normal infants used in the study.⁎ Indicate P value < 0.05 when comparing ganglionic segment with

aganglionic segment.⁎⁎ Indicate P value < 0.01 when comparing ganglionic segment withaganglionic segment.△ Indicate P value < 0.05 when comparing ganglionic segment ofHSCR with the transversal colon of normal infants.

182 J O U R N A L O F P R O T E O M I C S 8 4 ( 2 0 1 3 ) 1 7 6 – 1 8 4

especially the sample preparation [35,36] for which is themost crucial procedure for 2DE-analysis. By our modifiedmethod, we got several new differently expressed proteins,except for some common ones including HSPB1 that arediscussed above. Meanwhile, there were also proteins reportedby Gao's study that are missed in ours. We inferred that thismight be due to the lower protein recovery by our preparationmethod, although it helped us get more clear reproducibleelectrophoresis result.

Annexin A2 (ANXA2) is a member of annexin which is afamily of calcium-dependent phospholipid binding proteins.ANXA2 is most correlated with carcinogenesis and theprogression of invasive cancer [37,38]. This is the first timethat ANXA2 is being reported to be involved in HSCR. ANXA2could play the role of a scaffold protein linking plasmamembrane domain with the actin polymerization machinery[39], interfering with ANXA2 function can lead to failure ofactin polymerization [40]. ANXA2 has been implicated inmembrane–cytoskeleton interactions and in regulations ofion currents and substances across the membrane [41]. Thedecreased expression of ANXA2 in the aganglionic segment(decreased by 45%) might attribute to the bowel motilitydisorder. More important, we found that ANXA2 expressionwas altered not only in aganglionic segment of HSCR patient,but also in ganglionic segment (Table 3, Fig. 3C). Our resultsuggested that the ganglionic colon of HSCR patients is notcompletely the same with normal colon, several proteins mighthave changed in their expressions. In our study, down-regulatedANXA2 expression (by 88% comparing to normal control)appeared in ganglionic segment. This result might elucidatethat the bowel motility disorder persistently existed evenafter the affected colon was removed by surgery. Our resultsimplicated the important role ANXA2 played in bowel motilitydysfunction of HSCR including the persistent symptom aftersurgery.

In the present study, in the aganglionic segment there is analtered expression in protein whose function is related withenergy metabolism, like ATP synthase subunit beta (ATP5B).ATP5B is the major catalytic subunit of ATP synthase complex,which produces ATP from ADP in the presence of a proton

Fig. 2 – Confirmation of 2-DE results by immunoblot analysis of ATPATP5B, HSPB1, CNN1 and ANXA2. B. Immunoblot analysis for ATPHSCR was shown. GAPDH is used as the internal control. C. Quantexpression ATP5B, HSPB1, CNN1 and ANXA2 was showed in histowith HSCR was indicated by N, aganglionic segment of affected co

gradient across the mitochondrial membrane [42]. We verifiedthe expression of ATP5B decreased by 68% in the aganglionicsegment compared with the ganglionic ones (Table 3, Fig. 2B),which was in agreement with the 2-DE results (Fig. 2A).Interestingly, we found that ATP5B expression in ganglionicsegment of HSCR patients increased comparing with thetransversal colon of normal control (by 2.6 folds, see Table 3,Fig. 3C). This result implied that the energy metabolism wasabnormal in ganglionic colon of HSCR patient. There are noreports about the role ATP5B plays in HSCR yet, especially thealteration in ganglionic segment of HSCR patients. Consideringthe important role ATP5B has in energy machinery, wespeculated that its abnormal expression in both aganglionicand ganglionic segments of HSCR patients must affect the

5B, HSPB1, CNN1 andANXA2. A. 2-DE gel of spots identified as5B, HSPB1, CNN1 and ANXA2 expression from 4 patients withification of immunoblot result. Relative density of proteingram. Ganglionic segment of affected colon from the patientslon tissue is indicated by A. * P < 0.05, ** P < 0.01.

Fig. 3 – Immunoblot analysis of ATP5B, HSPB1, CNN1 and ANXA2 in the transversal and sigmoid colons of newborn infants.A. Expression for ATP5B, HSPB1, CNN1 and ANXA2 in the transversal and sigmoid colons of newborn infants. GAPDH is used asthe internal control. B. Quantification of immunoblot result of ATP5B, HSPB1, CNN1 and ANXA2 expression in the transversaland sigmoid colons of newborn infants. Relative density of proteins expressionwas showed in histogram. Transversal colon isindicated by T and sigmoid colon is indicated by S. C. Comparison of ATP5B, HSPB1, CNN1 and ANXA2 expression between thetransversal colon of newborn infant (T) and ganglionic segment of affected colon from the patients with HSCR (N).Quantification of immunoblot result showed in histogram. ATP5B expression increased while ANXA2 expression decreased inganglionic segment of HSCR patient comparing with transversal colon of newborn infant, but there were no differencesshowed for the left two protein expression.* P < 0.05.

183J O U R N A L O F P R O T E O M I C S 8 4 ( 2 0 1 3 ) 1 7 6 – 1 8 4

movement of bowel inHSCR, in some extent, andmight involvein the post-operative gut motility dysfunction.

5. Conclusions

We provided an overview of proteomic changes in theaganglionic and ganglionic segments of colon tissues fromHSCR patients. We found 16 differentially expressed proteinsin the aganglionic segment primary or secondary to theaganglionosis. These proteins included intermediate fila-ment forming cytoskeleton, ones that regulate the smoothmuscle movement, and some enzymes. However, the preciserole that they play in HSCR needs further investigation. Wealso found that ATP5B and ANXA2 expression was altered inganglionic segment of HSCR patient, except for an alterationin aganglionic segment. Our study may provide more insightsinto the motility disturbance of bowel in HSCR. Manipulationsfocusing on the altered proteins that we found might providehelp to the therapy of HSCR.

Acknowledgments

This study was supported by the National Natural Foundationof China (Grant numbers: 81100434), the Research Fund for theDoctoral Program of Higher Education of China (Grant number:20092104120010) and the Educational Commission of LiaoningProvince, China (Grant number: L2010637). We thankDr. ShawnChen for the help with the MS analysis.

Appendix A. Supplementary data

Supplementary data to this article can be found online athttp://dx.doi.org/10.1016/j.jprot.2013.03.024.

R E F E R E N C E S

[1] Chakravarti A, Lyonnet S. In: Scriver CR, SlyWS, Valle D, BeaudetAL, editors. The metabolic and molecular basis of inheriteddiseases. 8th ed. New York: McGraw-Hill; 2001 [chapter 251].

[2] Gershon MD. Genes, lineages, and tissue interactions in thedevelopment of the enteric nervous system. Am J Physiol1998;275:G869–73.

[3] Ramalho-Santos M, Melton DA, McMahon AP. Hedgehogsignals regulate multiple aspects of gastrointestinalevelopment. Development 2000;127:2763–72.

[4] Garcia-Barcelo MM, Tang CS, Ngan ES, Lui VC, Chen Y, So MT,et al. Genome-wide association study identifies NRG1 as asusceptibility locus for Hirschsprung's disease. Proc NatlAcad Sci U S A 2009;106:2694–9.

[5] Tam PK, Garcia_Barcelo' M. Genetic basis of Hirschsprung'sdisease. Pediatr Surg Int 2009;25:543–58.

[6] RomeoG, Ronchetto P, LuoY, BaroneV, SeriM, Ceccherini I, et al.Point mutations affecting the tyrosine kinase domain of the RETproto-oncogene in Hirschsprung's disease. Nature 1994;367:377–8.

[7] Kim HY, Kim JH, Jung SE, Lee SC, Park KW, Kim WK. Surgicaltreatment and prognosis of chronic intestinalpseudo-obstruction in children. J Pediatr Surg 2005;40:1753–9.

[8] Kenny SE, Tam PK, Garcia-Barcelo M. Hirschsprung's disease.Semin Pediatr Surg 2010;19(3):194–200.

[9] Ieiri S, Nakatsuji T, Akiyoshi J, Higashi M, Hashizume M, SuitaS, et al. Long-term outcomes and the quality of life ofHirschsprung disease in adolescents who have reached18 years or older—a 47-year single-institute experience.J Pediatr Surg 2010;45:2398–402.

[10] Harris ML, Erickson CA. Lineage specification in neural crestcell pathfinding. Dev Dyn 2007;236:1–19.

[11] Anderson RB, Bergner AJ, Taniguchi M, Fujisawa H, Forrai A,Robb L, et al. Effects of different regions of the developing guton the migration of enteric neural crest-derived cells: a rolefor Sema3A, but not Sema3F. Dev Biol 2007;305:287–99.

[12] Wang LL, Fan Y, Zhou FH, Li H, Zhang Y, Miao JN, et al.Semaphorin 3A expression in the colon of Hirschsprungdisease. Birth Defects Res A Clin Mol Teratol 2011;91(9):842–7.

184 J O U R N A L O F P R O T E O M I C S 8 4 ( 2 0 1 3 ) 1 7 6 – 1 8 4

[13] Ciftci AO, Sara Y, Tanyel FC, Bozdag O, Orer HS, Onur R. Therole of nitrergic system on the contractility of colonic circularsmooth muscle in Hirschsprung's disease. J Pediatr Surg1999;34:1477–81.

[14] Vanderwinden JM, Rumessen JJ, Liu H, Descamps D, De LaetMH, Vanderhaeghen JJ. Interstitial cells of Cajal in humancolon and in Hirschsprung's disease. Gastroenterology1996;111:901–10.

[15] Fan Y, Wang L, Zhou F, Zhang Y, Li H, Shan L, et al.Comparative proteomics of spinal cords of rat fetuses withspina bifida aperta. J Proteomics 2011;75:668–76.

[16] Candiano G, Bruschi M, Musante L, Santucci L, Ghiggeri GM,Carnemolla B, et al. Blue silver: a very sensitive colloidalCoomassie G-250 staining for proteome analysis.Electrophoresis 2004;9:1327–33.

[17] Towbin H, Staehelin T, Gordon J. Electrophoretic transfer ofproteins from polyacrylamide gels to nitrocellulose sheets:procedure and some applications. Proc Natl Acad Sci U S A1979;76:4350–4.

[18] Hennig GW, Gregory S, Brookes SJ, Costa M. Non-peristalticpatterns of motor activity in the guinea-pig proximal colon.Neurogastroenterol Motil 2010;22:e207–17.

[19] Rae MG, Fleming N, McGregor DB, Sanders KM, Keef KD.Control of motility patterns in the human colonic circularmuscle layer by pacemaker activity. J Physiol (Lond) 1998;495510:309–20.

[20] Han Yoon Kyeong, Yoon M, Moir RD, Khuon S, Flitney FW,Goldmana RD. Insights into the dynamic properties of keratinintermediate filaments in living epithelial cells. J Cell Biol2001;153:502–16.

[21] Small JV, Gimona M. The cytoskeleton of the vertebratesmooth muscle cell. Acta Physiol Scand 1998;164:341–8.

[22] Nemeth L, Rolle U, Puri P. Altered cytoskeleton in smoothmuscle of aganglionic bowel. Arch Pathol Lab Med 2002;126:692–6.

[23] Langer JC, Betti PA, Blennerhassett MG. Smooth muscle fromaganglionic bowel in Hirschsprung's disease impairsneuronal development in vitro. Cell Tissue Res 1994;276:181–6.

[24] Winder SJ, Walsh MP, Vasulka C, Johnson JD.Calponin–calmodulin interaction: properties and effects onsmooth and skeletal muscle actin binding and actomyosinATPases. Biochemistry 1993;32:13327–33.

[25] Somara S, Gilmont R, Bitar KN. Role of thin-filamentregulatory proteins in relaxation of colonic smooth musclecontraction. Am J Physiol Gastrointest Liver Physiol 2009;297:G958–66.

[26] Dufour C, Weinberger RP, Gunning P. Tropomyosin isoformdiversity and neuronal morphogenesis. Immunol Cell Biol1998;76:424–9.

[27] Won KJ, Lee P, Jung SH, Jiang X, Lee CK, Lin HY, et al.3-morpholinosydnonimine participates in the attenuation ofneointima formation via inhibition of annexin A2-mediated

vascular smooth muscle cell migration. Proteomics 2011;11:193–201.

[28] Spitz F, Demignon J, Kahn A, Daegelen D, Maire P.Developmental regulation of the aldolase A muscle-specificpromoter during in vivo muscle maturation is controlled by anuclear receptor binding element. J Mol Biol 1999;289:893–903.

[29] Li X, Chu JS, Yang L, Li S. Anisotropic effects of mechanicalstrain on neural crest stem cells. Ann Biomed Eng 2012;40:598–605.

[30] Doshi BM, Hightower LE, Lee J. HSPB1, actin filamentdynamics, and aging cells. Ann N Y Acad Sci 2010;1197:76–84.

[31] Landry J, Huot J. Modulation of actin dynamics during stressand physiological stimulation by a signaling pathwayinvolving p38 MAP kinase and heat-shock protein 27.Biochem Cell Biol 1995;73:703–7.

[32] Somara S, Bitar KN. Tropomyosin interacts withphosphorylated HSP27 in agonist-induced contraction ofsmooth muscle. Am J Physiol Cell Physiol 2004;286:C1290–301.

[33] Ibitayo AI, Sladick J, Tuteja S, Louis-Jacques O, Yamada H,Groblewski G, et al. HSP27 in signal transduction andassociation with contractile proteins in smooth muscle cells.Am J Physiol 1999;277:G445–54.

[34] Gao H, He X, Wu M, Zhang Z, Wang D, Lv L, et al. Proteomicanalysis of differentially expressed proteins between stenoticand normal colon segment tissues derived from patients withHirschsprung's disease. Protein J 2011;30(2):138–42.

[35] Jiang L, He L, Fountoulakis M. Comparison of proteinprecipitation methods for sample preparation prior toproteomic analysis. J Chromatogr A 2004;1023:317–20.

[36] Fic E, Kedracka-Krok S, Jankowska U, Pirog A,Dziedzicka-Wasylewska M. Comparison of proteinprecipitation methods for various rat brain structures prior toproteomic analysis. Electrophoresis 2010;31:3573–9.

[37] Mussunoor S, Murray GI. The role of annexins in tumourdevelopment and progression. J Pathol 2008;216:131–40.

[38] Hayes MJ, Moss SE. Annexins and disease. Biochem BiophysRes Commun 2004;322:1166–70.

[39] Benaud C, Gentil BJ, Assard N, Court M, Garin J, Delphin C,et al. AHNAK interaction with the annexin 2/S100A10complex regulates cell membrane cytoarchitecture. J Cell Biol2004;164:133–44.

[40] Hayes MJ, Rescher U, Gerke V, Moss SE. Annexin–actininteractions. Traffic 2004;5:571–6.

[41] Réty S, Sopkova J, Renouard M, Osterloh D, Gerke V, TabariesS, et al. The crystal structure of a complex of p11 with theannexin II N-terminal peptide. Nat Struct Biol 1999;6(1):89–95.

[42] Kucharczyk R, Zick M, Bietenhader M, Rak M, Couplan E,Blondel M, et al. Mitochondrial ATP synthase disorders:molecular mechanisms and the quest for curativetherapeutic approaches. Biochim Biophys Acta 2009;1793:186–99.