ORIGINAL PAPER

Comparative studies on the proteomic expression patternsin the third- and fifth-stage larvae of Angiostrongylus cantonensis

Kuang-Yao Chen & Chien-Ju Cheng & Chuan-Min Yen &

Petrus Tang & Lian-Chen Wang

Received: 1 May 2014 /Accepted: 4 July 2014 /Published online: 16 July 2014# Springer-Verlag Berlin Heidelberg 2014

Abstract Angiostrongylus cantonensis is an important zoo-notic parasite causing eosinophilic meningitis and eosinophil-ic meningoencephalitis in humans. In this study, the proteinexpression profiles of the infective third- and pathogenic fifth-stage larvae (L3 and L5) of this parasite were compared byproteomic techniques. Isolated protein samples were separatedby two-dimensional gel electrophoresis (2-DE), stained withsilver nitrate, and analyzed by matrix-assisted laserdesorption/ionization-time-of-flight mass spectrometry(MALDI-TOF MS). Proteins from L5 were mainly at pH 5–7 and with molecular weight (MW) 40–100 kDa, whereasthose from L3 were at pH 5–6 and with 5–35 kDa. Of 100protein spots identified, 33 were from L3 whereas 67 from L5and 63 had known identities, whereas 37 were hypotheticalproteins. There were 15 spots of stress proteins, and HSP60was the most frequently found heat stress proteins in L5.Morebinding and protein transport-related proteins were found inL5 including peptidylprolyl isomerase (cyclophilin)-like 2,serum albumin, preproalbumin precursor, and dilute class

unconventional myosin. L3 had a higher expression of cyto-skeleton and membrane proteins than L5. In addition, fourprotein spots were identified in the sera of the rat host byWestern blot analysis. The present proteomic study revealeddifferent protein expression profiles in L3 and L5 ofA. cantonensis. These changes may reflect the developmentof L3 from the poikilothermic snails to L5 in thehomoeothemic rats. This information may be useful for thefinding of stage-specific proteins and biomarker for diagnosisof angiostrongyliasis.

Keywords Angiostrongylus cantonensis . HSP60 .

Two-dimensional gel electrophoresis (2-DE) .MALDI-TOFMS

Introduction

Angiostrongylus cantonensis, the rat lungworm, is an impor-tant zoonotic parasite. Humans acquire the infection by eatingraw or undercooked intermediate hosts (snails or slugs),paratenic host (frogs), or vegetables contaminated with thethird-stage larvae (L3) (Alicata 1965, 1988). After penetratingthe intestinal wall, L3 migrates to the central nervous systemvia the blood circulation. At this site, they develop to the fifth-stage larvae (L5) and cause eosinophilic meningitis and eo-sinophilic meningoencephalitis. The main clinical manifesta-tions include fever, headache, nausea, vomiting, photophobia,neck stiffness, and increased intracranial pressure. This dis-ease may lead to fatal outcomes in severe cases. SoutheastAsia and the Pacific islands have been considered to be themain endemic regions for this infection. However, more than2,800 cases have been recorded in 31 countries worldwide(Wang et al. 2008, 2012).

Proteomic analysis is concerned with global changes inprotein expression. These techniques employ two-

Electronic supplementary material The online version of this article(doi:10.1007/s00436-014-4024-4) contains supplementary material,which is available to authorized users.

K.<Y. Chen : C.<J. Cheng : P. Tang : L.<C. Wang (*)Graduate Institute of Biomedical Sciences, College of Medicine,Chang Gung University, Taoyuan 333, Taiwane-mail: wanglc@mail.cgu.edu.tw

P. Tang : L.<C. WangDepartment of Parasitology, College of Medicine, Chang GungUniversity, Taoyuan 333, Taiwan

C.<M. YenDepartment of Parasitology, Faculty of Medicine, College ofMedicine, Kaohsiung Medical University, Kaohsiung 807, Taiwan

P. TangBioinformatics Center, Chang Gung University, Kwei-Shan,Tao-Yuan 333, Taiwan

Parasitol Res (2014) 113:3591–3600DOI 10.1007/s00436-014-4024-4

dimensional gel electrophoresis (2-DE) combined withmatrix-assisted laser desorption/ionization-time-of-flight(MALDI-TOF) mass spectrometry (MS) and liquidchromatography-MS/MS to elucidate the identifications andexpression patterns of intracellular proteins. Applications ofthe proteomic techniques to identify expression patterns ofproteins in parasites are not uncommon. Secreted proteins ofBrugia malayi in different stages and genders have also beenanalyzed by these techniques to study the parasite-host inter-action (Bennuru et al. 2009). Moreover, Caenorhabditiselegans proteomics is considered to be a versatile new toolbox for the systematic analysis of gene functions throughoutthe entire life cycle of this nematode (Shim and Paik 2010).

Proteins expressed in adult worms of A. cantonensis havebeen identified by differentially proteomics analysis (Songet al. 2012). This study provides differences in protein expres-sion profiles and the ability to survive in the final host betweenfemale and male adults of A. cantonensis. Although heavyinfections by adult wormsmay cause severe obstruction of thepulmonary arteries and respiratory failure in rats, this sugges-tion requires further histopathological studies (Wang et al.2012). Recently, development-specific differences in the pro-teomics of this parasite have been reported (Huang et al.2013). However, this study provided only limited informationon the protein expressed in the developmental stages. Com-parison of the protein expression profiles of L3 and L5 mayrespectively provide information on the host invasion andpathogenesis of human eosinophilic meningitis and eosino-philic meningoencephalitis. In this study, we used proteomictechniques to analyze protein expression profiles in L3 and L5of A. cantonensis. The results indicate that metabolism-relatedproteins are more abundant in L3 and stress-related proteins inL5.

Materials and methods

Parasite

A strain of A. cantonensis has been maintained throughBiomphalaria glabrata snails and Sprague-Dawley (SD) ratsin our laboratory since 1980 (Wang et al. 1989). From onesnail, more than 10,000 L3 were isolated. The L3 were inoc-ulated into SD rats (8 weeks old), and the first-stage larvae(L1) were collected from rat feces 50 days later. B. glabratawas infected with L1. After 20 days, L3 was isolated withorganization homogenizer (Cole-Parmer Instrument Co.,USA) and artificial gastric juice (0.6 % (w/v) pepsin, pH 2-3) (Wang et al. 1991). SD rats were purchased from theNational Laboratory Animal Center, Taipei. After infectingwith 100 L3 to each rat, they were then housed in plastic cagesand provided with food and water ad libitum. L5 were recov-ered from the brains of infected rats 3 weeks postinfection.

This study was reviewed and approved by the Chang GungUniversity Institutional Animal Care and Use Committee.

Protein extraction and purification

After washing with saline, phosphate-buffered saline, anddistilled water for three times, the worms were disrupted bysonication in a lysis buffer (8M urea, 4 %CHAPS) containingprotease inhibitors (Roche Diagnostics, Basel, Switzerland)on ice for 30 times. The cell lysate was centrifuged at 10,000×g and 4 °C for 15 min, and impurities from the total cell lysatewere removed using the 2-D Cleanup Kit (GE Healthcare Bio-Science Corp., Piscataway, USA). The protein concentrationin the preparations was determined with the Bio-Rad ProteinAssay Kit (Bio-Rad, Hercules, CA, USA) according to theinstructions of the manufacturer.

2-D gel electrophoresis

For first dimensional electrophoresis, 250 μg of the proteinpreparation was diluted to a final volume of 250 μl in arehydration buffer containing a trace amount of bromophenolblue (8 M urea, 2 % (w/v) CHAPS) and then applied to a 13-cm immobilized pH gradient (IPG) gel strip (GE Healthcare)with a linear separation range of pH 3–10 and pH 4–7.Rehydration and isoelectric focusing were carried out in theEttan IPGphor II (GE Healthcare) using the following set-tings: 30 V for 12 h, 50 V for 0.5 h, 100 V for 0.5 h, 250 V for0.5 h, 500 V for 0.5 h, 1,000 V for 0.5 h, 4,000 V for 0.5 h, andgradient to 8,000 V for 45,000 V/h. The IPG strip was incu-bated in an equilibration buffer (50 mM Tris-HCl (pH 8.8),6 M urea, 30 % glycerol, 2 % sodium dodecyl sulfate (SDS),and a trace amount of bromophenol blue) containing 1 % (w/v) dithiothreitol for 15 min and then in an equilibration buffercontaining 2.5 % (w/v) iodoacetamide for 15 min.

Silver staining of 2-D gels

Gels were fixed in 50 % (v/v) methanol and 25 % (v/v) aceticacid for 2 h, 30 % (v/v) methanol for 15 min, followed bysensitizing for 2 min in 0.8 mM sodium thiosulfate, andrinsing with Milli-Q water twice. Silver nitrate (0.2 % (v/v))was then added and incubated for 30 min. The gels wererinsed quickly with Milli-Q water twice and developed infresh 0.28 M sodium carbonate and 0.016 M Na2S2O3-2H2O with 0.085 % (v/v) formaldehyde. The developmentwas stopped by adding 0.042 M EDTA for 10 min.

Protein visualization and image analysis

The equilibrated IPG strip was separated across 15 % SDS-polyacrylamide gel electrophoresis with a solution of 0.5 %(w/v) agarose containing a trace of bromophenol blue. The gel

3592 Parasitol Res (2014) 113:3591–3600

was run at 35 mA/gel and 4 °C until the tracking dye migratedto the bottom and then stained with silver. The silver-stained2-DE gel of isoelectric point (pI) 3–10, pI 4–7, and pI 6–11was scanned, and images were analyzed by the Phoretix™ 2-D analysis software. The relative abundance or difference ofeach protein spot was determined as a percentage of the totalspot intensity score (0 and −100,000). Experimental molecularmass (Mr) and pI of each spot identified by 2-DE gels wereobtained by comparing with a set standardMr and pImarkers.

In-gel trypsin digestion for MALDI-TOF MS

Protein spots of interest were excised and transferred to0.5-ml tube for in-gel trypsin digestion. The gel pieceswere destained in a solution containing 30 mM potassiumferricyanide and 100 mM sodium thiosulfate. After wash-ing and shrinking with 50 mM ammonium bicarbonateand acetonitrile (ACN), the gel pieces were dried undervacuum centrifuge to remove the excess ACN andrehydrated in threefold volume of trypsin (Promega) so-lution (20 μg/ml in 25 mM mmonium bicarbonate). Afteradding 2 μl of extraction buffer (100 % ACN with 1 %trifluoroacetic acid), peptides were extracted by sonicationfor 30 min. These extracts were eluted and thencocrystallized with a saturated solution on a MALDI-TOF sample plate using 0.5 μl of sample and 0.5 μlmatrix. The peptides were analyzed using the UltraflexMALDI-TOF MS (Bruker Daltonic).

MALDI-TOF MS analysis

The NCBI database was explored by the MASCOT databasesearch engine (Matrix Science; http://www.matrixscience.com/serach_form_select.html). The following searchparameters were used: Trypsin was used as an enzyme, thepeptide tolerance window was set to 100 ppm, one missedcleavage was allowed, and carbamidomethyl and oxidizedmethionine were respectively set as fixed and variablemodifications. Protein identifications were consideredaccurate when the overall MASCOT score was greater than50.

MALDI-TOF/TOF MS analysis

Abundance of each protein spot was identified by MALDI-TOF. Significant differences in sequences were determinedand reconfirmed by MALDI-TOF/TOF.

Western blotting

Protein spots were also analyzed by 12.5 % SDS-PAGE.The spots were transferred from gels to nitrocellulosemembranes using a semidry transfer unit under 0.04 mA

for 50 min. After washing three times with Tris-bufferedsaline with Tween 20 (TBS/T) and then with a blockingbuffer, the membranes were incubated with primary anti-bodies (serum from rats after infecting A. cantonensis21 days, 1:1,000) at 4 °C overnight and then washed threetimes with TBS/T. The membranes were then incubatedwith secondary antibodies (anti-rat IgG in rabbit, Sigma-Aldrich, 1:100,000) for 450 min and washed three timeswith TBS/T. Further incubation of the membranes with amixture of 500 μl stable peroxide solution and 500 μlenhanced solution was performed for 5 min in the dark.

Results

2-DE protein maps and image analysis

Proteins extracted from L3 and L5 were separated by 2-DE using a wide range IPG strip of pH 3–10 and a narrowrange IPG strip of pH 4–7. Analysis of the protein prep-arations from L5 revealed differences in three main re-gions: molecular weight (MW) of 15–25 kDa, pI 3–7;MW around 20 kDa, pI 8–10; and MW around 10 kDa,pI 8–10. Since proteins from L5 were found to be tooconcentrated in the gels, a narrow range pH (pH 4–7) ofthe IPG strip was employed for further proteomicanalysis.

After ImageMaster software analysis, the gels were foundto have 552 protein spots in L3 and 542 spots in L5. Proteinsfrom L5 were found to have pH between 5 and 7 and MWbetween 40 and 100 kDa. The corresponding values for L3were 5 to 6 in pH and 5 to 35 kDa in MW.

Mass spectrometric analysis

From L3 and L5, 119 protein spots were explored with theNCBI database. Of 100 proteins identified (19 proteins un-identified), 33 proteins were from L3 and 67 from L5 (Fig. 1).Moreover, 63 had known identities and 37 were hypotheticalproteins (Tables S1 and S2).

Functional analysis to different proteins

The functional annotations of these protein spots werefurther explored with the Gene Ontology (GO) database(29 functions in 33 L3 spots and 98 functions in 67 L5spots), including biological process (7 functions in L3 and54 functions in L5), cellular component (7 functions in L3and 21 functions in L5), and molecular function (15functions in L3 and 23 functions in L5). In biologicalprocess, L3 spot functions only found in metabolic pro-cess (n=3) and cellular process (n=4), whereas L5 spot

Parasitol Res (2014) 113:3591–3600 3593

functions in response to stimulus (n=6), cellular process(n=8), regulation of biological process (n=6), and biolog-ical regulation (n=6) (Fig. 2). Moreover, most of theproteins identified were stress proteins, nucleotide-binding proteins, and protein-binding proteins (Fig. 3).There were 15 spots of stress proteins (Fig. 4) (Table 1).These proteins were divided into three categories by func-tion: three in nutritional deficiencies, four in oxidative

stress, and eight in heat stress. HSP60 was the mostfrequently found (n=7) among the heat stress proteins.

Stress responses

There were 15 stress-related proteins in L5, and no one wasfound in L3. The stress-related proteins in L5 included heat-shock protein 68, putative protein disulfide isomerase 1,

Fig. 1 Two dimensional gelelectrophoresis map of solubleproteins from the third-stagelarvae (a) and fifth-stage larvae(b) of Angiostrongyluscantonensis separated at pH range4–7: 67 of 71 protein spots wereidentified in the female fifth-stagelarvae and 33 of 48 in the third-stage L3

3594 Parasitol Res (2014) 113:3591–3600

Fig. 2 Functional annotations of100 protein spots from the third-stage larvae and fifth-stage larvaeof Angiostrongylus cantonensisidentified by MALDI-TOF PMFclassified through UniProt (http://www.uniprot.org/). a Biologicalprocess. b Cellular component. cMolecular function

Parasitol Res (2014) 113:3591–3600 3595

putative heat-shock protein, serum albumin, preproalbuminprecursor, and oxidoreductase (short chain dehydrogenase/reductase family protein). Using MALDI-TOF/TOF MS, pu-tative heat-shock protein (no. 24–30) was determined to be

most abundant protein in A. cantonensis. These proteins wereidentified to be heat-shock protein 60.

Protein transport, protein binding, and nucleotide binding

More nucleotide-binding proteins, protein-binding proteins,and protein transport-related proteins were found in L5 thanL3 in the two-dimensional protein map. These proteins in-cluded peptidylprolyl isomerase (cyclophilin)-like 2, serumalbumin, preproalbumin precursor, and dilute class unconven-tional myosin.

Metabolism and ATP-related proteins

L3 and L5 had the same number of metabolism andenergy-related proteins. Among the seven protein func-tions in L5, four were metabolism-related and threeenergy-related proteins. Of seven protein functions inL3, four were metabolism-related and three wereenergy-related proteins (Fig. 3). However, L3 had ahigher the proportion of known metabolism andenergy-related proteins than L5.

Cell skeletal and membrane

L3 had a higher cytoskeleton and membrane proteinexpression than L5. These proteins included connexin40, prohibitin in L5, and actin in the L3.

Protein extracts of worms against host sera

Western blot analysis was then performed for eight proteinspots from L5 in serum samples from rats (Fig. 5). Four spots

Fig. 3 Functional annotations ofprotein spots from the third-stagelarvae (L3) and fifth-stage larvae(L5) of Angiostrongyluscantonensis identified byMALDI-TOF PMF classified byGene Ontology (http://www.geneontology.org/)

Fig. 4 Two-dimensional gel electrophoresis map of stress proteins (n=15) from the third-stage larvae and fifth-stage larvae of Angiostrongyluscantonensis

3596 Parasitol Res (2014) 113:3591–3600

were identified. They were no. 20 (rCG22908), no. 31(synaptotagmin-5), no. 32 (serum albumin), and no. 34(heat-shock protein 8).

Discussion

Although the genome of A. cantonensis has not been se-quenced, information at the transcriptomic level is available.The sequences from L5 and some additional ones were ana-lyzed and found to encode proteins participating in metabo-lism, cellular development, immune evasion, and host-parasite interactions (Xu et al. 2009) and grouped into 13categories (Fang et al. 2010). In ESTs of L3, the most abun-dantly expressed transcripts were found to be cathepsin B-likecysteine protease 1 and 2, metalloprotease I, metalloprotease 1precursor, and extracellular superoxide dismutase. These find-ings suggest the importance of proteases in host invasion andnutrition of A. cantonensis (Chang et al. 2011). Moreover, L3have a gene expression profile toward metabolism and L5toward growth and development (Chang et al. 2014). Investi-gations on the proteomics may confirm the differential ex-pressions of genes in the developmental stages of this parasite.

Proteomics is not only important for the investigations ofproteins in life science research but also became the mostimportant trends in biotechnology. Recently, 2-DE andMALDI-TOF MS techniques have been widely used in thestudy of parasitology. Two new cysteine proteinases associat-ed with host invasion in Trichomonas vaginalis has beenfound by 2-DE and MALDI-TOF MS (Leon-Felix et al.2004). Among the secreted proteins of Toxoplasma gondii,

some sequences similar to malaria proteins were revealed tohave association with host invasion (Zhou et al. 2005). Dif-ferences in surface antigen protein expressions among fourPlasmodium falciparum laboratory culture differences havebeen determined to lead to variations in parasite adhere to redblood cells and also affect the parasite pathogenicity (Wu andCraig 2006).

Associations between protein expression and the develop-ment of parasites have also been determined by the proteomictechniques. Proteins regulating development have been iden-tified in different stages of Leishmania panamensis (Walkeret al. 2006). After analyzing 3,260 proteins in different stageand gender Schistosoma japonicum, these proteins were di-vided functionally into egg development, adult worm growth,and antigens (Liu et al. 2006). Moreover, these techniqueshave also been applied to study on parasite invasion and hostimmune responses modulation. Mice infected with Trichinellaspiralis may induce lipid metabolism, transport, and innateimmune response-related cell cytokine expression (Pembertonet al. 2004). Host cell may induce leukocyte differentiationand infection-resistant-related proteins in Theileria annulata(Oura et al. 2006). Although the proteomic techniques havebeen widely applied in the studies of parasites, information onthe proteome of A. cantonensis is very limited (Song et al.2012; Huang et al. 2013).

Using proteomic techniques, protein expressions in maleand female adult worms of A. cantonensis were compared. Atotal of 28 protein spots were found to be differentiallyexpressed. Of seven protein spots that were identified byMALDI-TOF MS, five were upregulated and two proteinsdownregulated in male worms compared with female ones.Three of the five upregulated proteins had functional

Table 1 Identification of stress proteins from the fifth-stage larvae of Angiostrongylus cantonensis by MALDI-TOF mass spectrometry

Spot Accession number E value Annotation Function

19 gi|126723507 0.035 Preproalbumin precursor [Equus caballus] Starvation

32 gi|76363596 0.05 RecName: Full=Serum albumin; Flags: Precursor Starvation

44 gi|126723507 0.061 Preproalbumin precursor [Equus caballus] Starvation

16 gi|113895895 2.5e-006 Putative protein disulfide isomerase 1 [Dictyocaulus viviparus] Oxidation

57 gi|170591170 0.076 Oxidoreductase [Brugia malayi] Oxidation

58 gi|195121732 0.056 GI19114 [Drosophila mojavensis] Oxidation

69 gi|170591170 0.058 Oxidoreductase [Brugia malayi] Oxidation

3 gi|34398667 0.01 Heat-shock protein 68 [Drosophila simulans] Heat

24 gi|256251570 0.0073 Putative Heat-Shock Protein [Angiostrongylus cantonensis] Heat

25 gi|256251570 0.0073 Putative Heat-Shock Protein [Angiostrongylus cantonensis] Heat

26 gi|256251570 0.0073 Putative Heat-Shock Protein [Angiostrongylus cantonensis] Heat

27 gi|256251570 0.0073 Putative Heat-Shock Protein [Angiostrongylus cantonensis] Heat

28 gi|256251570 0.0073 Putative Heat-Shock Protein [Angiostrongylus cantonensis] Heat

29 gi|256251570 0.0073 Putative Heat-Shock Protein [Angiostrongylus cantonensis] Heat

30 gi|256251570 0.0073 Putative Heat-Shock Protein [Angiostrongylus cantonensis] Heat

Parasitol Res (2014) 113:3591–3600 3597

annotations of galectin-1, proteasome alpha subunit, andperoxiredoxin. The two downregulated proteins were identi-fied as indoleamine dioxygenase-like myoglobin and galectin(Song et al. 2012). In the recent proteomic analysis in differentdevelopmental stages of A. cantonensis, only 37 protein spotswere found to have high confidence scores. Among thesespots, 29 were cytoskeleton-associated proteins and functionalproteins and 8 unnamed proteins (Huang et al. 2013).

This study focuses on L3 and L5 of A. cantonensis. Amongthe 1,000 protein sports detected, 119 were found to be dif-ferentially regulated. Of 100 spots identified by MALDI-TOFMS analysis and then explored with the NCBI database, 33spots were isolated from L3 and 67 spots from L5. Functionalannotations of these proteins were determined with the GOdatabase. These spots were classified into 12 functional cate-gories. In L5, pressure-related proteins, nucleotide-binding-related proteins, protein-binding-related proteins, proteintransport-related proteins, and metabolism-related proteinsare the common ones. In L3, metabolism-related proteins,protein transport-related proteins, and ATP-related proteinsare frequently found. In both developmental stages,pressure-related proteins are the most common category,followed by nucleotide-binding-related proteins, protein-binding-related proteins, and protein transport-relatedproteins.

It is well-known that parasite cells may generate protein toresist environment stress when they suffer from heat stress,oxidative stress, osmotic pressure, and charge gradient. After

infecting final host (rat) by L3 inB. glabrata, these larvae molttwice and develop into L5 in the rat brain in 14 days. At thissite, L5 release different types of stress-related proteins. Heat-shock proteins are important for the survival of parasites.These proteins may help parasite to resist the pressure of theenvironment. HSP90 and HSP70 have been identified in thedevelopment of Leishmania donovani promastigote in mam-mal (homoiothermic) from amastigote in (poikilothermic).After environmental changes, the amastigote released thesestress-related proteins to resist the changes (Bente et al. 2003).In this study, L3 and L5 have stress-related protein spots.These proteins can be classified into three categories: starva-tion stress, oxidation stress, and heat stress. Heat stress is themost abundant category, and HSP60 is the most abundantheat-shock protein. HSP60 has been documented as an im-portant antigenic protein in T. gondii (Ma et al. 2009). Thisprotein has also been identified as an important secretedprotein for the resistance of environmental stress inStrongyloides ratti (Tazir et al. 2009). It also has the functionof resisting stress in intermediate host for Echinostomacaproni (Sotillo et al. 2010). Parasites suffer from oxidationstress and starvation stress in environment changes. In thepresent study, we identified preproalbumin precursor andalbumin in the protein spots. Since albumin has been reportedto regulate P. falciparum growth (Asahi et al. 2011), theputative protein disulfide isomerase 1 and oxidoreductaseidentified in the present study may be associated with oxida-tion stress.

Fig. 5 Western blot analysis wasperformed for eight protein spotsfrom fifth-stage larvae ofAngiostrongylus cantonensisagainst serum samples from rats.Four spots were identified. Theywere no. 20 (rCG22908), no. 31(synaptotagmin-5), no. 32 (serumalbumin), and no. 34 (heat-shockprotein 8)

3598 Parasitol Res (2014) 113:3591–3600

Expressions of cytoskeleton-related proteins have beenrevealed in the growth stages and environment changes inT. vaginalis, L. donovani, and Eimeria tenella (Bente et al.2003; Huang et al. 2009; Lal et al. 2009; Roy et al. 2010). Inthe present study, more cytoskeleton-related protein expres-sions were observed in L3 than L5. In the host brain, L3develop to L5. The parasite cytoskeleton and cell membranereassemble. This change may lead to parasite growth. Inaddition, the parasite requires cytoskeleton-related proteinsto generate signal transduction.

After the host infected with L3, these larvae migrate to thehost brain. The migration requires cell movement-related pro-teins and may cause the expression of cytoskeleton or cellmembrane-related proteins. We found that connexin 40,prohibitin, and MHC class I-related chain are cytoskeleton-related proteins. Moreover, the movement-related protein my-osin is highly expressed in L5, whereas actin is highlyexpressed in L3.

Nucleotide-binding proteins, protein-binding proteins, andprotein transport proteins are biologically essential ones. The-se proteins have been found to be highly expressed inT. vaginalis and E. tenella (Bente et al. 2003; Lal et al.2009). In this study, we found the association betweenpreproalbumin and albumin with protein-binding and proteintransport proteins in L5. Metabolism-related proteins werefound to be highly expressed in rodent harboringTrypanosoma evansi (Roy et al. 2010). They are importantfor surviving. In this study, we found that polo kinase is highlyexpressed. Moreover, nucleotide-binding proteins, protein-binding proteins, and protein transport proteins also have thesame expression pattern.

In the A. cantonensis EST database, more metabolism-related mRNAs were found in L3 than L5. In the presentstudy, this phenomenon was confirmed at the proteomicslevel. A total of 35 protein spots were identified to be associ-ated with metabolism in biological process (4 of 54 in L5 and3 of 7 in L3). This result shows that metabolism-relatedproteins were found in L3 (42.9 %) than L5 (7.4 %).

Albendazole, an anticytoskeletal drugs, is a member of thebenzimidazole compounds used as a drug indicated for thetreatment of a variety of worm infections. This drug maycause tegument and intestine degenerative alterations in para-sites by binding to the colchicine-sensitive site of tubulin andthen inhibiting tubulin polymerization and assembly into mi-crotubules. InA. cantonensis, this drug has been reported to bethe drug of choice against larvae and was effective within15 days postinfection (Wang et al. 2006). Among cytoskeletonand plasma membrane in this proteome database, higher ex-pressions of connexin 40 and prohibitin were found in L5 andactin in L3. Cell motility, growth, and morphological changesrequire remodeling of cytoskeleton in response to intracellularand extracellular signals (Martínez-Valladares et al. 2012;O’Hagan et al. 2011; Huang et al. 2009), and reorganization

of the microtubule network is important for the survival ofnematodes (Banora et al. 2011). The identification of thesecytoskeleton proteins provides a rationale for the treatmentagainst A. cantonensis.

The pathogenesis caused by A. cantonensis remains un-clear. In addition, diagnosis and treatment of the disease arealso in debate. This proteomic study revealed different proteinexpressions in L3 and L5 of A. cantonensis. These changesmay reflect the development of L3 from the poikilothermicsnails to L5 in the homoeothemic rats. In the previous prote-omic study (Huang et al. 2013), only 37 proteins were iden-tified and 29 of them were cytoskeleton-associated proteins.However, we have identified 100 protein spots in the presentstudy. Among these spots, 33 were from L3 and 67 from L5.By MALDI-TOF/TOF, HSP60 was confirmed to be highlyexpressed in L5. This heat-shock protein may be a suitabletarget for further investigations. Moreover, the results of thepresent study may also provide information for the finding ofstage-specific proteins and biomarker for diagnosis ofangiostrongyliasis.

Acknowledgments This work was supported in part by grants from theNational Science Council, Executive Yuan, ROC (NSC100-2320-B-182-013 and NSC101-2320-B-182-045) and Chang Gung Memorial HospitalResearch Grant CMRPD1B0351. We thank Chang Gung MolecularMedicine Research Center for technical supports.

References

Alicata JE (1965) Biology and distribution of the rat lungworm,Angiostrongylus cantonensis, and its relationship to eosinophilicmeningoencephalitis and other neurological disorders of man andanimals. Adv Parasitol 3:223–248

Alicata JE (1988) Angiostrongylus cantonensis (eosinophilic meningitis):historical events in its recognition as a new parasitic disease of man.J Wash Acad Sci 78:38–46

Asahi H, Izumiyama S, Tolba ME, Kwansa-Bentum B (2011)Plasmodium falciparum: differing effects of non-esterified fattyacids and phospholipids on intraerythrocytic growth in serum-freemedium. Exp Parasitol 127:708–713

BanoraMY, Rodiuc N, Baldacci-Cresp F, Smertenko A, Bleve-Zacheo T,Mellilo MT, Karimi M, Hilson P, Evrard JL, Favery B, Engler G,Abad P, de Almeida Engler J (2011) Feeding cells induced byphytoparasitic nematodes require γ-tubulin ring complex for micro-tubule reorganization. PLoS Pathog 7:e1002343

Bennuru S, Semnani R, Meng Z, Ribeiro JM, Veenstra TD, Nutman TB(2009) Brugia malayi excreted/secreted proteins at the host/parasiteinterface: stage- and gender-specific proteomic profiling. PLoSNeglTrop Dis 3:e410

Bente M, Harder S, Wiesgigl M, Heukeshoven J, Gelhaus C, Krause E,Clos J, Bruchhaus I (2003) Developmentally induced changes of theproteome in the protozoan parasite Leishmania donovani.Proteomics 3:1811–1829

Chang SH, Tang P, Wang LC (2011) A transcriptomic study on thepepsin-activated infective larvae of Angiostrongylus cantonensis.Mol Biochem Parasitol 179:47–50

Parasitol Res (2014) 113:3591–3600 3599

Chang SH, Tang P, Yen CM, Chow KP, Wang LC (2014) Atranscriptomic analysis on gene expressions in the infective thirdand pathogenic fifth larval stages of Angiostrongylus cantonensis.Parasitol Int 63:42–48

Fang W, Xu S, Wang Y, Ni F, Zhang S, Liu J, Chen X, Luo D (2010) ESproteins analysis of Angiostrongylus cantonensis: products of thepotential parasitism genes? Parasitol Res 106:1027–1032

Huang HC, Yao LL, Song ZM, Li XP, Hua QQ, Li Q, Pan CW, Xia CM(2013) Development-specific differences in the proteomics ofAngiostrongylus cantonensis. PLoS ONE 8:e76982

Huang KY, Chien KY, Lin YC, HsuWM, Fong IK, Huang PJ, Yueh YM,Gan RR, Tang P (2009) A proteome reference map of Trichomonasvaginalis. Parasitol Res 104:927–933

Lal K, Bromley E, Oakes R, Prieto JH, Sanderson SJ, Kurian D, Hunt L,Yates JR 3rd, Wastling JM, Sinden RE, Tomley FM (2009)Proteomic comparison of four Eimeria tenella life-cycle stages:unsporulated oocyst, sporulated oocyst, sporozoite and second-generation merozoite. Proteomics 9:4566–4576

Leon-Felix J, Ortega-Lopez J, Orozco-Solis R, Arroyo R (2004) Twonovel asparaginyl endopeptidase-like cysteine proteinases from theprotist Trichomonas vaginalis: their evolutionary relationship withinthe clan CD cysteine proteinases. Gene 335:25–35

Liu F, Lu J, HuW,Wang SY, Cui SJ, Chi M, YanQ,Wang XR, Song HD,XuXN,Wang JJ, ZhangXL, ZhangX,Wang ZQ,Xue CL, BrindleyPJ, McManus DP, Yang PY, Feng Z, Chen Z, Han ZG (2006) Newperspectives on host-parasite interplay by comparativetranscriptomic and proteomic analyses of Schistosoma japonicum.PLoS Pathog 2:e29

Ma GY, Zhang JZ, Yin GR, Zhang JH, Meng XL, Zhao F (2009)Toxoplasma gondii: proteomic analysis of antigenicity of solubletachyzoite antigen. Exp Parasitol 122:41–46

Martínez-Valladares M, Donnan A, Geldhof P, Jackson F, Rojo-VázquezFA, Skuce P (2012) Pyrosequencing analysis of the beta-tubulingene in Spanish Teladorsagia circumcincta field isolates. VetParasitol 184:371–376

O’Hagan R, Piasecki BP, Silva M, Phirke P, Nguyen KC, Hall DH,Swoboda P, Barr MM (2011) The tubulin deglutamylase CCPP-1regulates the function and stability of sensory cilia in C. elegans.Curr Biol 21:1685–1694

Oura CA, McKellar S, Swan DG, Okan E, Shiels BR (2006) Infection ofbovine cells by the protozoan parasite Theileria annulatamodulatesexpression of the ISGylation system. Cell Microbiol 8:276–288

PembertonAD,Knight PA,Wright SH,MillerHR (2004) Proteomic analysisof mouse jejunal epithelium and its response to infection with theintestinal nematode, Trichinella spiralis. Proteomics 4:1101–1108

Roy N, Nageshan RK, Pallavi R, Chakravarthy H, Chandran S, Kumar R,Gupta AK, Singh RK, Yadav SC, Tatu U (2010) Proteomics ofTrypanosoma evansi infection in rodents. PLoS ONE 22:e9796

Shim YH, Paik YK (2010) Caenorhabditis elegans proteomics comes ofage. Proteomics 10:846–857

Song Z, Huang H, Tan F, Zhang E, Hu J, Pan C (2012) Differentialproteomics analysis of female and male adults of Angiostrongyluscantonensis. Exp Parasitol 131:169–174

Sotillo J, Valero ML, Sánchez Del Pino MM, Fried B, Esteban JG,Marcilla A, Toledo R (2010) Excretory/secretory proteome of theadult stage of Echinostoma caproni. Parasitol Res 107:691–697

Tazir Y, Steisslinger V, Soblik H, Younis AE, Beckmann S, GreveldingCG, Steen H, Brattig NW, Erttmann KD (2009) Molecular andfunctional characterisation of the heat shock protein 10 ofStrongyloides ratti. Mol Biochem Parasitol 168:149–157

Walker J, Vasquez JJ, Gomez MA, Drummelsmith J, Burchmore R,Girard I, Ouellette M (2006) Identification of developmentally-regulated proteins in Leishmania panamensis by proteome profilingof promastigotes and axenic amastigotes. Mol Biochem Parasitol147:64–73

Wang LC, Chao D, Chen ER (1989) Acquired immunity in ratsagainst Angiostrongylus cantonensis infection. Int J Parasitol19:617–620

Wang LC, Chao D, Chen ER (1991) Experimental infection routes ofAngiostrongylus cantonensis in mice. J Helminthol 65:296–300

Wang LC, Jung SM, Chen CC, Wong HF, Wan DP, Wan YL (2006)Pathological changes in the brains of rabbits experimentally infectedwith Angiostrongylus cantonensis after albendazole treatment: his-topathological and magnetic resonance imaging studies. JAntimicrob Chemother 57:294–300

Wang QP, Lai DH, Zhu XQ, Chen XG, Lun ZR (2008) HumanAngiostrongyliasis. Lancet Infect Dis 8:621–630

Wang QP, Wu ZD, Wei J, Owen RL, Lun ZR (2012) HumanAngiostrongylus cantonensis: an update. Eur J Clin MicrobiolInfect Dis 31:389–395

Wu Y, Craig A (2006) Comparative proteomic analysis of metabolicallylabelled proteins from Plasmodium falciparum isolates with differ-ent adhesion properties. Malar J 5:67–80

Xu SS, Ni F, Luo DM (2009) Expressed sequence tags (ESTs) analysis ofAngiostrongylus cantonensis. Chin J Parasitol Parasitic Dis 27:248–250

Zhou XW, Kafsack BF, Cole RN, Beckett P, Shen RF, Carruthers VB(2005) The opportunistic pathogen Toxoplasma gondii deploys adiverse legion of invasion and survival proteins. J Biol Chem 280:34233–34244

3600 Parasitol Res (2014) 113:3591–3600