Comparative Study of Immunological and Structural Properties of Two ...

8 downloads 0 Views 2MB Size Report
Background: Recently, botulinum neurotoxin (BoNT)-derived recombinant proteins have been suggested as potential botulism vaccines. Here, with ...
Iranian Biomedical Journal 16 (4): 185-192 (October 2012) DOI: 10.6091/ibj.1076.2012

Comparative Study of Immunological and Structural Properties of Two Recombinant Vaccine Candidates against Botulinum Neurotoxin Type E Mosayeb Rostamian1, Seyed Jafar Mousavy*1, Firouz Ebrahimi1, Seyyed Abolghasem Ghadami2,3, Nader Sheibani4, Mohammad Ebrahim Minaei1 and Mohammad Ali Arefpour Torabi1 1

Dept. of Biology, Faculty of Basic Sciences, Imam Hussein University, Tehran, Iran; 2Dept. of Biology, Faculty of Sciences, Razi University, Kermanshah, Iran; 3Medical Biology Research Center, Kermanshah University of Medical Science, Kermanshah, Iran; 4Dept. of Ophthalmology and Visual Sciences and Pharmacology, University of Wisconsin School of Medicine and Public Health, Madison, WI, USA Received 16 May 2012; revised 6 August 2012; accepted 8 August 2012

ABSTRACT Background: Recently, botulinum neurotoxin (BoNT)-derived recombinant proteins have been suggested as potential botulism vaccines. Here, with concentrating on BoNT type E (BoNT/E), we studied two of these binding domain-based recombinant proteins: a multivalent chimer protein, which is composed of BoNT serotypes A, B and E binding subdomains, and a monovalent recombinant protein, which contains 93 amino acid residues from recombinant C-terminal heavy chain of BoNT/E (rBoNT/E-HCC). Both proteins have an identical region (48 aa) that contains one of the most important BoNT/E epitopes (YLTHMRD sequence). Methods: The recombinant protein efficiency in antibody production, their structural differences, and their BoNT/E-epitope location were compared by using ELISA, circular dichroism, computational modeling, and hydrophobicity predictions. Results: Immunological studies indicated that the antibody yield against rBoNT/E-HCC was higher than chimer protein. Cross ELISA confirmed that the antibodies against the chimer protein recognized rBoNT/E-HCC more efficiently. However, both antibody groups (anti-chimer and anti-rBoNT/E-HCC antibodies) were able to recognize other proteins. Structural studies with circular dichroism showed that chimer proteins have slightly more secondary structures than rBoNT/E-HCC. Conclusion: The immunological results suggested that the above-mentioned identical region in rBoNT/E-HCC is more exposed. Circular dichroism, computational protein modeling and hydrophobicity predictions indicated a more exposed location for the identical region in rBoNT/E-HCC than the chimer protein, which is strongly in agreement with immunological results. Iran. Biomed. J. 16 (4): 185-192, 2012 Keywords: Botulinum neurotoxin type E (BoNT/E), Cross ELISA, circular dichroism, Computational modeling, recombinant vaccine-candidates

INTRODUCTION

B

otulism is a dangerous neuroparalytic syndrome caused by blocking acetylcholine release at the neuromuscular junctions [1, 2]. Botulinum neurotoxins (BoNT), which cause botulism, are produced in seven different serotypes (A, B, C, D, E, F and G) by Clostridium botulinum strain [3]. Botulism syndrome is classified into three forms: the food-born, wound, and infant (intestinal) botulism [1, 3]. BoNT are initially produced as a stable complex of approximately 900 kDa and then divided into a 150kDa neurotoxin and non-toxic components [1]. The 150-kDa neurotoxin consists of two polypeptide

chains: a light chain (50 kDa) and a heavy chain (100 kDa), which are linked through a disulfide bond [4]. The light chain is organized as an N-terminal catalytic domain, while the heavy chain comprises an internal translocation domain and a C-terminal receptor binding domain. The heavy chain, via receptor-mediated endocytosis, mediates translocation of light chain across the endosomal membrane into the cytosol. BoNT recognize nerve membranes by binding to two components: a group of membrane glycophospholipids called gangliosides and specific protein receptors such as synaptotagmin (for BoNT/D and G) or synaptic vesicle membrane protein, SV2 (for BoNT/A and E)

*Corresponding author; Tel.: (+98-21) 7710 4934; Fax: (+98-21)7710 4935; E-mail: [email protected]

1866

Rosstamian et al.

[5, 6]. The lighht chain is a protease p that cleaves target a synaptosom mal-associateed prooteins in nervve cells such as prootein of 25--kDa and vesicle v mem mbrane proteiin synnaptobrevin. Cleavage off these proteiins causes thhe bloockage of acetylcholinne release and finallly neuuroparalysis [7, [ 8]. V Vaccination a against botullism by toxooids has som me lim mitations, inclluding the neeed for speciffic equipmennts whhich leads too high cost,, the low yield y of toxiin prooduction by Clostridium C b botulinum straain, the dangeer of C. botulinum m handling, annd the potenttial side effeccts andd unexpectedd immunologgical reactionns. To prevennt bottulism, reseaarchers have been recentlyy interested in i usiing recombinnant BoNT-baased proteins as vaccine [9911]]. These typpes of vacccine have reesolved manny preevious conceerns related to use off toxoids. An A exaample of theese recombinnant proteinss is based on o BooNT-binding domains with mulltivalent annd moonovalent anntigenic prooperties [12]. Antibodiees agaainst these reecombinant vaccines v are proven to be b efffective in neutralizing n BoNT effeccts [12]. Thhe muultivalent vaaccines are more preeferable thaan rBoNT/E-HCC chimer

Irann. Biomed. J., October O 2012

monov valent vaccinnes due to their ability to o immunize againsst multiple neurotoxin serootypes. Heree, we study two t of thesee binding dom main-based recomb binant proteiins whose BooNT neutralizzing ability has beeen previouslyy reported. T These proteins include a multiv valent chimerr protein (1877 amino acid d) which is compo osed of serotyypes A, B andd E binding subdomains [13] an nd a monovallent recombinnant protein (259 ( amino acids) which contaains 93 aminno acid resid dues of Cnal heavy chaain of BoNT ttype E (rBoN NT/E-HCC) termin [14]. Both B of these proteins haave an identical region (48 aa) that conttains one oof the most important D sequence) [12]. The BoNT/E epitopes (YLTHMRD n sequences and their homology have h been protein depicteed in Figurre 1. In this study, thee scale of antibody production against two above--mentioned binant proteeins in rabbbits was compared. recomb Furtheermore, we characterized some featurees of these vaccin nes as a criterion of multiivalent and monovalent m vaccin ne comparisoon by ELISA A. Finally, we w further confirm med the resuults of otherr studies usin ng circular dichro oism and moleecular modeliing [15]. Trx tag

rBoNT/E-HCC chimer

rBoNT/E-HCC chimer

rBoNT/E-HCC chimer

rBoNT/E-HCC chimer F 1. Sequence alignment of recombinant C-terminal heavyy chain of BoN Fig. NT/E (rBoNT/E--HCC) and chim mer protein. Th he amino acid seqquences are num mbered from thhe aminoterminaal of proteins. Different D parts of the proteinss have been deppicted with diffferent colors. Chiimer protein is composed of a tag of 58 aa from BoNT/A (B BoNT/A subdom main), 61 aa froom BoNT/B (B BoNT/B subdom main), and 48 aa from BoNT/E (BoNT/E subddomain) from N- to C-terminnal, respectivelly. rBoNT/E-H HCC is compossed of a tag, and a 93-aa of BoN NT/E from N- to C-terminal, respectively. r Thhe red arrow (448 aa) depicts a region of chim mer protein, whhich is exactly similar s to 48aa sequence s of rBoNT/E-HCC (tthe identical reggion). The purpple box shows the t active BoNT T/E-epitope (YLTHMRD sequ uence) of the prooteins. Trx tag shows s a Trx•Taag™ thioredoxiin protein in pE ET 32 [16]. Forr interpretationn of the referencces to color in this text, the reader is referred to t the web version of the articlle. http://IIBJ.pasteur.ac.iir

Iran. Biomed. J., October 2012

Comparison of Two Recombinant Vaccine Candidates for Botulism

MATERIALS AND METHODS Bacteria, chemicals and media. All molecular biology grade chemicals and bacterial culture media were from Merck (Germany). Chemical agents for nickel nitrilotriacetic acid agarose (Ni-NTA) resin were from Qiagen (USA). LB powder was obtained from Difco (Sparkes, MD, USA). The pET-contained E. coli strains (pET32a encoding the sequence for rBoNT/EHCC protein and pET28a, encoding the sequence for chimer protein were used to express recombinant proteins [13, 14]. Amino acid sequence alignments. Comparison of amino acid sequences of rBoNT/E-HCC and chimer protein (Fig. 1) was carried out using the ClustalW program (http://www.ebi.ac.uk/Tools/clustalw2/index. html) [17]. Recombinant protein expression and purification. The expression process was conducted as reported elsewhere [13, 14, 18]. After final step of protein expression process, the chimer protein was sufficiently pure, while the rBoNT/E-HCC solution was followed by further purification as mentioned below. The all separated protein fractions were run on 12% SDSPAGE and stained with Coomassie blue. A 50% NiNTA bead suspension was used for purification of rBoNT/E-HCC protein [18]. Buffer E (100 mM NaH2PO4, 10 mM Tris-HCl and 8 M urea, pH 4.5) was applied to elute the expressed protein from Ni-NTA column. The purity of the sample was determined by SDS-PAGE analysis and a single protein band (31 kDa) was obtained. Antigenicity testing. In order to compare the recombinant protein antigenicity, purified recombinant proteins were mixed with an equal volume of complete Freund's adjuvant for initial injection and incomplete Freund's adjuvant for subsequent injections into 3 rabbits intradermally. Each rabbit received 100 µg of antigen at weeks 0, 2 and 4. Two rabbits were used as controls receiving only adjuvant. A week after second and third injections, the animals were bled and sera were restored for ELISA experiments. ELISA plates were coated with an optimal concentration of purified proteins (3.5 µg ml-1 per well) in a coating buffer (15 mM Na2CO3 and 36 mM NaHCO3, pH 9.8) and allowed to adhere at 4°C overnight. Also, one row was incubated with coating buffer alone (no-antigen) as control. Plates were washed four times with 400 µl PBST and blocked with 100 µl per well of skim milk (50 mg ml-1) at 37°C for 45 min. After washing, a serial two-fold dilutions in PBST, starting at 1:200, of rabbit serum samples were then added (100 ml per well) and plates were incubated at 37°C for 30 min.

187

Following a washing step, plates were incubated at 37°C for 30 min with anti-rabbit immunoglobulin Ghorseradish peroxidase conjugate (1:2,000; 100 µl per well) in PBST. Plates were washed four times with 400 µl PBST and then incubated with 100 µl per well of ortho-phenylenediamine (Sigma, USA) and H2O2 as substrate. The reaction was terminated by addition of 2.5 M H2SO4 (100 µl/well) and the absorbance was read at 490 nm using an ELISA plate reader (Anthos2020, Eugendorf, Austria). Antibody purification. Protein-G-Sepharose columns (Sigma, USA) were used to purify IgG from sera. After washing the columns with 100 mM and then 10 mM Tris (pH 7.8), the sera were mixed 1:10 with Tris (1 M) and applied to the columns. The columns were washed again as indicated above and purified IgG was eluted using 100 mM glycine, pH 3.0. Circular dichroism spectropolarimetry. In order to obtain information about the secondary structure of the recombinant proteins, circular dichroic spectra were gathered in far-UV regions (195-260 nm) using a JASCO model J-810 spectropolarimeter at 25°C. The cell volume was approximately 0.5 ml with a path length of 0.1 mm. The recombinant proteins were at a concentration of 0.25 mg ml-1 in 100 mM NaH2PO4, 10 mM Tris-HCl, pH 8.0. The secondary structures of proteins were determined using CDNN program, version 2.1.0.223 (obtained from Institute of Biophysics and Biochemistry of Tehran University, Dr. Moosavi-Movahedi Laboratory). Computational molecular modeling of the rBoNT/E-HCC and chimer protein. The tertiary structures of rBoNT/E-HCC and chimer protein are unknown. Here, we obtained de novo protein structure predictions at the Robetta [19, 20] and Lomets [21] servers. The evaluation of the quality of the models was performed with Qmean server [22]. The molecular dynamic simulations were carried out using HyperChem Professional version 8.0 Molecular Modeling System [23]. Charmm27 force field [24] was used for the simulations according to the procedure described by Fendri et al. [25]. The energy minimization of these proteins was carried out under implicit solvent conditions using the conformational analysis programs. The lowest energy conformations were then solvated with TIP3P water explicitly, and finally the overall system was energy minimized using the Polak-Ribiere conjugate gradient method until the convergence of the gradient (0.01 kJ mol-1) [23]. PyMol version 0.99 beta 06 (http://www.pymol.org), UCSF Chimera and MSMS were used to produce molecular graphics images and other representations [26, 27].

http://IBJ.pasteur.ac.ir

1888

Rosstamian et al.

Irann. Biomed. J., October O 2012

aracterization n of antiibody responses to Cha recom mbinant protteins. ELISA A was perfformed to determ mine antiboddy titers oof sera from m rabbits vaccin nated againstt the recom mbinant proteins. Two weeks after last vacccination, bloood samples were taken SA. Produceed antibody titers t were and used for ELIS ved in rabbitts immunizedd by rBoNT//E-HCC or observ chimerr protein (Figg. 3).

F Fig. 2. SDS--PAGE analyysis of recom mbinant proteein exppression and purification. p Thhe separated protein p fractionns werre run on 12% % SDS-PAGE gel g and stained with Coomasssie bluue. Lanes 1 and a 2, cell lyysate of E. cooli BL21 (DE3) conntaining pET28a-chimer gene before and afteer induction wiith isoppropyl-1-thio-bb-D galactopyrranoside (IPTG G), respectivelly; Lannes 5 and 6, cell c lysate of E. E coli BL21 (DE3) containinng pET T32a-rBoNT/E-HCC (recombinant C-terminaal heavy chain of BoN NT/E) gene after and beefore inductioon with IPTG G, resppectively. The expressed recom mbinant proteinns were observeed at approximately a 2 and 31 kDa; Lane 3, purifieed chimer proteein 20 obttained after finnal ultracentrifuugation in bufffer II (200 mM M NaC Cl, 50 mM Triss, 8 M Urea annd 2.5 M glycerrol) [13]; Lane 4, purrified rBoNT/E E-HCC by Ni--NTA agarose affinity colum mn afteer adding washh buffer E (1000 mM NaH2PO O4, 10 mM triisbasse, 8 M urea, pH H 4.5) [18]. Lanne Mw shows protein p molecullar weiight marker.

Reco ombinant prootein and purrified IgG in nteractions. The ch himer proteinn and anti-chhimer protein n IgG, and rBoNT T/E-HCC and anti--rBoNT/E-HC CC IgG interacctions were evaluated e by ELISA. App proximately 8 ng of anti-rBooNT/E-HCC IgG and anti-chimer a n IgG were able to reacct with their respective protein antigen n (Fig. 4). Since S both prroteins had an a identical region n, in order to test crooss reactivity y between rBoNT T/E-HCC andd anti-chimer protein serum m/IgG, and chimerr protein annd anti-rBoN NT/E-HCC serum/IgG, s ELISA A was applied. Once rBoN NT/E-HCC was w coated on EL LISA plate wells, w the antti-chimer protein serum and pu urified IgG dilutions d weree added separrately (Fig. 5), an nd alternativeely chimer pprotein was coated c and anti-rB BoNT/E-HCC C serum annd purified IgG were added separately (Fig. ( 5). Theese studies sh howed that mbinant proteeins were gen nerated and antibodies to recom nized their respective aantigen. How wever, the recogn antibodies affinity and titers were significan ntly higher oNT/E-HCC compared too chimer protein. in rBo It may refer r to the recombinaant protein structural 4.0 -♦- rBoNT/E-HC CC -□- chimera proteein -∆- control of rBoNT/E-HCC -×- control of chiimera protein

3.5

SULTS AND D DISCUSSIO ON RES

3.0 2.5

OD 490 nm

Expression and purifi E fication of recombinan nt prooteins. Thee recombinnant expresssion vectors (enncoding the desired recoombinant prooteins) with a Hiss6 tag were prepared p in thhe T7-promooter-based pE ET vecctors (pET322a (+) for rB BoNT/E-HCC C and pET288a (+)) for chimer protein) p in E. E coli BL21 (DE3) cells as a preeviously desccribed [13, 144]. The cells were w grown in i LB B broth and expression waas induced ussing isopropyyl1-tthio-b-D galaactopyranosidde. Each cultuure sample waas evaaluated for protein p expression by SDS-PAGE (Fig. 2, Lanes 1, 2, 5 and 6). The rBoNT/E-HCC in bacterria s biind to Ni-NTA suppernatant wass allowed to selectively agaarose gel thrrough its Hiss6 tag and eluuted using thhe eluution buffers (100 mM NaH N 2PO4, 10 mM Tris-HC Cl andd 8 M urea,). The purifieed recombinannt protein waas anaalyzed by SDS-PAGE S showing a 31-kDa bannd corrresponding to t the rBoNT T/E-HCC (Fiig. 2, Lane 4). 4 Thhe purified reccombinant chhimer protein also showedd a sinngle band (~220 kDa) on SD DS-PAGE (Fig. 2, Lane 3)).

2.0 1.5 1.0 0.5 0

1:1600

1:3200

1:64000

1:12800

1:25600

Sera d dilutions ody titration Fig. 3. ELISA ressults of rabbitt serum antibo C and chimer prroteins. Both an nti-rBoNT/Eagainst rBoNT/E-HCC a anti-chimerr protein sera from rabbits bind b to their HCC and antigen ns up to 1:25,600; how wever, anti-rBoNT/E-HCC (recomb binant C-terminal heavy chaain of BoNT/E) serum had higher affinity. a Serum m from non-imm munized rabbit did not react with eiither rBoNT/E--HCC or chim mer protein. Th he error bars present high reproduciibility of doublee repeats of the experiment.

http://IIBJ.pasteur.ac.iir

Iran. Biomed. J., October 2012

Comparison of Two Recombinant Vaccine Candidates for Botulism

conformational epitope determinations. An ensemble of tertiary structures for each protein was obtained using the modeling programs Robetta [19, 20] and Lomets [21]. These programs were used to generate de novo protein structure prediction models. A set of 15 three-dimensional models was generated for each protein (10 from Lomets and 5 from Robetta). To

4.0 -♦- rBoNT/E-HCC -□- chimera protein

3.5 3.0 2.5

OD 490 nm

189

s

2.0

2.5

(A)

1.5 2.0

1.0

0.0 140

120

100

80

60

40

20

0

OD 490 nm

0.5

1.5

1.0

IgG (ng) Fig. 4. Comparison of ELISA results of anti-rBoNT/E-HCC (recombinant C-terminal heavy chain of BoNT/E) and antichimer protein IgG reactions with their antigens (rBoNT/E-HCC and chimer protein, respectively). Approximately 8 ng of antirBoNT/E-HCC and anti-chimer protein IgG could signify and bind to their antigens. In comparison with anti-chimer protein, IgG and chimer protein reaction, anti-rBoNT/E-HCC IgG reacted with its target protein (rBoNT/E-HCC) with higher affinity. The error bars present high reproducibility of double repeats of the experiment.

0.5

0 1:200

1:400

1:800

1:1600

1.8

(B)

1.6 1.4

features, which lead to different identical region (i.e. BoNT/E-epitope containing region) locations.

Theoretical studies. After achieving some insights into the proteins compactness by circular dichroism, computational studies were used to characterize and evaluate the mentioned hypothesis. The sequences of rBoNT/E-HCC and chimer protein were subjected to de novo tertiary structure predictions for

1.2

OD 490 nm

Circular dichroism. ELISA results showed that the different structural features of the recombinant proteins may lead to more exposure of the identical region in rBoNT/E-HCC. To verify this hypothesis and also gain insight into structural properties of the recombinant proteins, circular dichroism was applied. Circular dichroism spectra of both rBoNT/E-HCC and chimer protein were taken at 25°C to determine their secondary structures. The amount of secondary structures was higher in chimer protein in comparison with rBoNT/E-HCC (Fig. 6). Circular dichroism spectra analysis using CDNN program, version 2.1.0.223, confirmed these results (Table 1). These results support our hypothesis and indicate in rBoNT/E-HCC, where the amount of secondary structures is less, the protein tends to be less ordered and may affect BoNT-epitope location to be more exposed.

1:3200 1:6400

Sera dilutions

1.0 0.8 0.6 0.4 0.2 0 1000

500

250

125

62.5

31.25

IgG (ng) Fig. 5. ELISA results of cross reaction between rBoNT/EHCC and anti-chimer protein serum/IgG, and chimer protein and anti-rBoNT/E-HCC (recombinant C-terminal heavy chain of BoNT/E) serum/IgG. Assuming OD 0.5 as a reliable answer: (A) both anti-rBoNT/E-HCC and anti-chimer protein sera could bind to the other proteins up to appropriately 1:1,600 dilution. However, anti-rBoNT/E-HCC interacts more strongly. (B) Approximately 125 ng of both anti-rBoNT/E-HCC and antichimer protein IgG could interact with the other recombinant antigens. However, anti-chimer IgG react with rBoNT/E-HCC more efficiently that may be due to the more exposed location of epitopes in rBoNT/E-HCC. The dash line depicts the interaction of rBoNT/E-HCC with chimer serum (panel A) or IgG (panel B); the solid lines depict the interaction of chimer protein with rBoNT/E-HCC serum (panel A) or IgG (panel B). The error bars represent high reproducibility of double repeats of the experiments.

http://IBJ.pasteur.ac.ir

1900

Rosstamian et al.

Irann. Biomed. J., October O 2012

10

[θ].deg cm2 dmol-1 × 103

5

0

-5

-10

-15

-20 210

220

230 0

2 240

250

260

Wavvelength (nm) F 6. Circular dichroism specctrum of recom Fig. mbinant proteinss. The CD specttrum was recordded as describeed in materials and a methods. Chiimer protein haas more seconddary structures.. The dash andd solid lines sho ow the circularr dichroism patttern of rBoNT T/E/HCC and chim mer protein, resspectively. Erroor bars of spectrra for duplicatee repeating expeeriments and higgh reproducibillity are shown in i the inset.

sellect the best three-dimenssional model, the Qmeaan serrver [22] waas used for model qualiity estimationn. Baased on the global score of the whoole model, thhe higghest predicteed model reliiability was measured m from m 0 to t 1 (1 as thhe highest quuality estimattion). For botth prooteins, one of the modelss generated by b the Robettta serrver yielded the t highest vaalue of 0.781 and 0.666 foor rBooNT/E-HCC and chimer protein, p respeectively. Thesse moodels weree used thhereafter for f mappinng connformational epitopes. PyMol P and UCSF U Chimeer weere used to produce p moleecular graphicc images. Thhe ressulting modells have been shown s in Figgure 7. Panel A shoows a moleecular surfacce representaation of eacch prootein, and pannel B shows the hydrophoobicity surfacce of the corresponnding molecuules in the saame orientatioon of panel A. Figgure 7 indicattes that the iddentical regioon mer protein is in a moree hydrophobic loccation in chim reggion which leads l to burrying the epitope into thhe prootein. Altogeether, Figuree 7 shows that identical reggion locationn in rBoNT/E E-HCC seem ms to be morre expposed. These results are inn full agreem ment with thosse of immunologiccal and circullar dichroism studies. P Previous worrks by others have focusedd on design of o reccombinant vaaccines againnst a single BoNT B serotyppe (m monovalent) or o against a group g of them m (multivalennt) [100]. Here, we focused on two recombinant vaccinees agaainst BoNT whose w abilitiees to neutralizze BoNT werre preeviously repoorted [13, 14]: 1 a multivalent chimeer prootein (187 am mino acid ressidues) whichh is composeed of types A, B and E binnding subdoomains, and a

valent recom mbinant proteein (259 am mino acid) monov which contains 93 amino acidd residues off rBoNT/EHCC. The monovvalent vaccine is able to neutralize e [13, 14]. Since both BoNT/E more efficiently binant proteins have a 48 aa-identiical region recomb (that contains c onee of the moost importan nt BoNT/E epitopes [YLTHMR RD sequencee, Fig. 1]), we w designed some experimentss to characcterize theirr antibody interacction with thhe opposite protein (i.e. rBoNT/EHCC with w anti-chimer protein serum/IgG, and a chimer protein n with anti- rBoNT/E-HC r CC serum/IgG G). We also determ mined their structural as well as other immun nological featture differencces. Imm munological studies s show wed that botth proteins were able a to evoke high antiboddy titers in rab bbits (as an animall model); hoowever, the antibody titeers against rBoNT T/E-HCC werre higher. Cross ELISA ex xperiments confirm med that antibodies a aggainst chimer protein recogn nize rBoNT/E E-HCC more efficiently, th hough both antibodies were able to recognize the other binant antigeen. These reesults suggesst that the recomb epitope sequence against the identical region r (i.e. Tablle 1. Circular dichroism sppectra analysis by CDNN program m version 2.1.0.223. 2nd sttructure r rBoNT/E-HCC C (%) Chiimer (%) Helix 38.51 39.80 strand d and turn 29.35 29.56 Rndm m. coil 32.14 30.64 Total sum 100.00 100.00

http://IIBJ.pasteur.ac.iir

Irann. Biomed. J., October O 2012

Comparisoon of Two Recoombinant Vaccin ne Candidates for f Botulism

191

epitope location, rB BoNT/E-HCC C seems to be b a better date for use as vaccine against BoN NT/E than candid chimerr protein, though t morre studies should s be perform med in this arrea. ACK KNOWLEDG GMENTS The supports of Research R Couuncils of Imaam Hussein Teh hran are Univerrsity and Universityy of acknow wledged. Thhe authors thhank Dr. Seeyed Latif Mousaavi (Shahed University, U Irran) for providing pETcontain ned E. coli bacteria. We aare also indeb bted to Mr. Abbass Hajizadeh and a Mr. Shahhram Nazariaan for their excelleent technical assistance. F 7. Three-ddimensional moolecular modelss of recombinaant Fig. rBooNT/E-HCC (recombinant ( C-terminal heavy h chain of BoN NT/E) and chhimer protein. Panel A shoows the surfacce reppresentation off rBoNT/E-HC CC (left) and chimer proteein (rigght), the structuure visualizationn by PyMol version 0.99 beta006. Redd parts show thhe BoNT/E-epitope location inn proteins. Pannel B depicts d the hydrophobicity suurfaces of eacch protein in thhe sam me orientation of o panel A. Thee hydrophobicity surface show ws the amino acid hyydrophobicity with w colors rangging from dodgger bluue for the most hydrophilic too white at 0.0 to t orange red for f the most hydrophhobic. The im mages were made with UCS SF Chiimer [27] and the surface was w calculated with w the MSM MS pacckage [26]. For interpretation of o the referencees to color in thhis textt, the reader is referred r to the web w version of the article.

RD) in rBoN NT/E-HCC is BooNT/E-epitoppe, YLTHMR moore exposed in i comparisoon with the chimer c proteiin andd it may referr to the differrences in theirr structure. T Therefore, strructural studdies were used to confirm m this hypothesiis. Structuraal studies with circulaar wed that chiimer protein have slightlly dicchroism show moore secondaary structurees than rB BoNT/E-HCC C. Diffferences of circular c dichrroism pattern between thesse twoo recombinant proteinss, which reefer to theeir struuctural differrences, may affect the iddentical regioon (BooNT/E-epitoppe) location. In rBoNT/E E-HCC, wherre thee amount of secondary s strructures is leess, the proteiin tennds to be less ordered and may affect on o the identical reggion (BoNT-epitope) locaation to be more m exposed. Prootein modelinng and calcuulations of hydrophobicit h ty shoowed a morre exposed location for the identical reggion (BoNT/E-epitope) inn rBoNT/E-H HCC than thhe chiimer protein. These results are in full agreement a witth thee immunologgical and ciircular dichrroism studiees. Furthermore, the t lower antibody a titeers of chimeer prootein (lower efficiency off the multivaalent vaccinee), andd lower affinnity betweenn chimer prootein and anttirBooNT/E-HCC antibodies were influeenced by thhe ideentical regionn (BoNT/E-eppitope) positiions in the tw wo prooteins. Altogeether, the results suggest that t due to thhe

REFEREN NCES 1. 2. 3. 4. 5.

6.

7.

8.

9. 10. 11. 12.

13.

Sim mpson LL. Bootulinum neurrotoxin and tetanus toxin. Accademic Press,, San Diego CA A; 1989. Haatheway CL. Toxigenic cllostridia. Clin n Microbiol Reev.1990 Jan;3((1):66-98. Sim mpson LL. Thhe origin, struccture, and pharmacological activity of botulinum toxinn. Pharmacol Rev.1981 ep;33(3):155-888. Sep Scchiavo G, Montecucco M C.. Tetanus an nd botulism neeurotoxins: isoolation and asssay. Meth En nzymol.1995 Feeb;248:643-652. Do ong M, Richaards DA, Gooodnough MC, Tepp WH, Johnson EA, Chapman ER. Synaptotagmin ns I and II o botulinum nneurotoxin B into i cells. J meediate entry of Ceell Biol.2003 Sep;162(7):129 S 93-303. Min Dong, Huiisheng Liu, W William H. Tepp, Eric A. Johnson, Rogerr Janz et al. Glycosylated SV2A and V2B mediate thhe entry of bootulinum neuro otoxin E into SV neeurons. Mol Biool Cell.2008 D Dec;19(12):522 26-37. Scchiavo G, Santtucci A, Dasguupta BR, Mehta PP, Jontes J, Benfenati F. Botulinum B neuurotoxins serottypes A and P-25 at distinctt COOH-term minal peptide E cleave SNAP bo onds. FEBS Lettt.1993 Nov;3335(1):99-103. Scchiavo G, Maalizio C, Trim mble WS, Po olverino de Laaureto P, Milan G et al. Botulinum G neurotoxin cleeaves VAMP P/synaptobrevinn at a sing gle Ala-Ala peeptide bond. J Biol B Chem.19994 Aug;269(32) 2):20213-6. Middlebrook JL. Protection strrategies againsst botulinum xin. Adv Exp Med M Biol.1995 Sep;383:93-8. tox By yrne MP, Sm mith LA. Deveelopment of vaccines v for preevention of botulism. Biochhimie.2000 Sep ep-Oct;82(910 0):955-66. Ro obinson RF, Nahata N MC. M Management of o botulism. An nn Pharmacothher 2003 Jan;337(1):127-31. Ku ubota T, Watannabe T, Yokossawa N, Tsuzu uki K, Indoh T, Moriishi K ett al. Epitope rregions in the heavy h chain b typee E neurotoxin n recognized off Clostridium botulinum by y monoclonal antibodies. A Appl Environ Microbiol. 19 997 Apr;63(4):1214-8. Eb brahimi F, Rassaee MJ, Mouusavi SL, Bab baeipour V. Prroduction andd characterizaation of a recombinant r ch himera antigeen consistingg botulinum neurotoxin

http:://IBJ.pasteur.a ac.ir

192

14.

15.

16.

17.

18.

19.

Rostamian et al.

serotypes A, B and E binding subdomains. J Toxicol Sci.2010 Feb;35(1):9-19. AgheliMansour AA, Mousavi SL, Rasooli I, Nazarian S, Amani J, Farhadi N. Cloning, high level expression and immunogenicity of 1163-1256 residues of C-terminal heavy chain of C. botulinum neurotoxin type E. Biologicals.2010 Mar;38(2):260-4. Shone C, Agostini H, Clancy J, Gu M, Yang HH, Chu Y et al. Bivalent recombinant vaccine for botulinum neurotoxin types A and B based on a polypeptide comprising their effector and translocation domains that is protective against the predominant A and B subtypes. Infect Immun. 2009 Jul;77(7):2795-801. LaVallie ER, DiBlasio EA, Kovacic S, Grant KL, Schendel PF, McCoy JM. A thioredoxin gene fusion expression system that circumvents inclusion body formation in the E. coli cytoplasm. Biotechnology (NY).1993 Feb; 1(2):187-93. Thompson JD, Higgins DG, Gibson TJ. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res.1994 Nov;22(22):4673-80. Wingfield PT, Palmer I, Liang SM. Folding and purification of insoluble (inclusion-body) proteins from Escherichia coli. Curr Protoc Protein Sci.2001 May;6: Unit 6.5. Chivian D, Kim DE, Malmström L, Bradley P, Robertson T, Murphy P. Automated Prediction of CASP-5 Structures Using the Robetta Server. Proteins. 2003 Oct;53:524-33.

Iran. Biomed. J., October 2012

20. Kim DE, Chivian D, Baker D. Protein structure prediction and analysis using the Robetta server. Nucleic Acids Res.2004 Jul;32 (Web Server issue): W526-531. 21. Wu S, Zhang Y. LOMETS: a local meta-threadingserver for protein structure prediction. Nucleic Acids Res.2007 May;35(10):3375-82. 22. Benkert P, Künzli M, Schwede T. QMEAN Server for Protein Model Quality Estimation. Nucleic Acids Res. 2009 Jul;37(Web Server issue):W510-4. 23. Froimowitz M. HyperChem: a software package for computational chemistry and molecular modeling. Biotechniques.1993 Jun;14(6):1010-3. 24. Brooks BR, Bruccoleri RE, Olafson BD, States DJ, Swaminathan S, Karplus M. CHARMM: A Program for macromolecular energy, minimization, and dynamics calculations. J Comput Chem.1983 Sep;4(2): 187-217. 25. Fendri A, Frikha F, Miled N, Ben Bacha A, Gargouri Y. Modulating the activity of avian pancreatic lipases by an alkyl chain reacting with an accessible sulfhydryl group. Biochem Biophys Res Commun.2007 Sep;360(4):76577. 26. Sanner MF, Olson AJ, Spehner JC. Reduced surface: an efficient way to compute molecular surfaces. Biopolymers.1996 Mar;38(3):305-20. 27. Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, Meng EC, Ferrin TE. UCSF chimera-a visualization system for exploratory research and analysis. J Comput Chem.2004 Oct;25(13):1605-12.

http://IBJ.pasteur.ac.ir