Research Article [Araştırma Makalesi]

Türk Biyokimya Dergisi [Turkish Journal of Biochemistry–Turk J Biochem] 2015; 40(2):181–187doi: 10.5505/tjb.2015.69335

Molecular cloning and sequence analysis of serine protease cDNA from the venom of the centipede Scolopendra subspinipes dehaani

[Scolopendra subspinipes dehaani kırkayak zehrinden elde edilen serin proteaz cDNA’sının moleküler klonlaması ve dizi analizi]

ABSTRACTObjective: Proteolytic enzymes are one of the well-known components of animal venom. The aim of this study was to identify protease from the venom gland of Scolopendra subspinipes dehaani.Methods: After total RNA extraction, a cDNA encoding a serine protease was identified by RT-PCR and was subsequently sequenced. The sequence was analysed by multiple alignment and homology modelling.Results: A cDNA encoding the precursor of a centipede venom serine protease was identified. The full-length cDNA was 1,014 nucleotides with 780 nucleotides of open reading frame. The precursor nucleotide sequence encoded a signal peptide of 19 residues and a mature protein of 260 residues. It belonged to clan PA of serine proteases in the MEROPS database classification. Catalytic residues were identified by multiple alignments, including an unusual hydrogen bond network in the catalytic site. Secondary structure prediction and homology modelling revealed a chymotrypsin-like fold, which is characteristic of PA clan proteases.Conclusion: The sequence of the identified serine protease showed unusual features, making this centipede venom serine protease an interesting candidate to investigate. A better understanding of the molecular mechanism of venom components will be useful for further applications and help to improve envenomation treatment.Key Words: Scolopendra subspinipes dehaani, cDNA, serine proteaseConflict of Interest: The authors have no conflict of interest.

ÖZETAmaç: Proteolitik enzimler hayvan zehirlerinin en çok bilinen bileşenlerinden biridir. Bu ça-lışmanın amacı Scolopendra subspinipes dehaani zehir salgı bezinden proteaz enziminin ta-nımlanmasıdır.Metod: Toplam RNA eldesinin ardından, RT-PCR tekniği kullanılarak serin proteaz proteinini kodlayan gene ait cDNA tanımlanmış ve dizi analizine tabi tutulmuştur. Elde edilen dizi çoklu dizi hizalama ve homoloji modelleme ile analiz edilmiştir.Bulgular: Kırkayak zehirinde bulunan serin proteaz öncülünü kodlayan bir cDNA tanımlan-mıştır. cDNA, 780 nükleotit açık okuma çerçevesinde olmak üzere, toplam 1.014 nükleotit uzunluğundadır. Öncül nükleotit dizisi 19 amino asitlik bir sinyal peptidini ve 260 amino asit-lik olgun proteini kodlamaktadır. Bu protein MEROPS veritabanı sınıflandırmasına göre serin proteazlardan PA grubuna aittir. Çoklu dizi hizalama kullanılarak alışılmadık bir hidrojen bağı örgüsüne sahip katalitik bölge ve bu bölgedeki katalitik aminoasitler tanımlanmıştır. İkincil yapı tahmin ve homoloji modelleme çalışmaları PA grup proteazlara özgün olan kimotiripsin benzeri katlanmayı göstermiştir. Sonuç: Tanımlanan serin proteaz dizisi alışılmadık özellikler göstermesi sebebiyle kırkayak zehrindeki serin proteaz enzimini ileri çalışmalar için aday yapmıştır. Venom bileşenlerinin moleküler mekanizmalarının daha iyi anlaşılması sonraki çalışmalar için yardımcı olabileceği gibi zehirlenmelerin tedavilerine de yardımcı olacaktır.Anahtar Kelimeler: Scolopendra subspinipes dehaani, cDNA, serin proteazÇıkar Çatışması: Yazarların çıkar çatışması yoktur.

http://www.TurkJBiochem.com 181 ISSN 1303-829X (electronic) 0250-4685 (printed)

Muchalin Meunchan1,Sompong Thammasirirak1, Jureerat Daduang2,Teerasak Somdee3,Sakda Daduang1

1Khon Kaen University, Faculty of Science, Department of Biochemistry, Protein and Proteomics Research Center for Commercial and Industrial Purposes (ProCCI), Khon Kaen, Thailand2Khon Kaen University, Faculty of Associated Medical Sciences, Department of Clinical Chemistry, Khon Kaen, Thailand3Khon Kaen University, Faculty of Science, Department of Microbiology, Khon Kaen, Thailand

Correspondence Address[Yazışma Adresi]

Sakda Daduang, PhD.

123 Mittapap Road, Department of Biochemistry, Faculty of Science, Khon Kaen University 40002 Khon Kaen, ThailandPhone: +66 817 686580E-mail: sakdad@kku.ac.th

Registered: 21 August 2014; Accepted: 01 December 2014[Kayıt Tarihi: 21 Ağustos 2014; Kabul Tarihi: 01 Aralık 2014]

Yayın tarihi 5 Mart 2015 © TurkJBiochem.com

[Published online March 5, 2015]

serine protease (CVSP) was first identified by PCR us-ing a couple of specific primers. These primers were de-signed according to the conserved catalytic amino acids for serine proteases: histidine (named primerH) and ser-ine (primerS) [11]. The gene was amplified using a PCR master mix reagent kit (Thermo Scientific, USA). PCR was performed for 30 cycles of 94°C (30 s); 55°C (30 s); 72°C (1 min) and a final extension at 72°C for 7 min. The purified amplicon was inserted into the pGEM®-T Easy Vector (Promega, USA) according to the manufacturer’s instructions and transformed into Escherichia coli, strain DH5α. Positive transformants were verified by blue/white colony selection and colony PCR. The plasmid construct was extracted using the IllustraTM Plasmid Prep Mini Spin Kit (Qiagen, Netherlands). DNA sequencing from both directions was performed by the 1st BASE Compa-ny (Malaysia) using the BigDye® Terminator v3.1 Cycle Sequencing Kit. The similarity of CVSP was identified using the Basic Local Alignment Search Tools (BLAST) program. The full sequence of CVSP was determined by 3′ and 5′- rapid amplification of cDNA ends (RACE) sys-tems (Invitrogen Life Technologies, USA). Primer H and a Universal Amplification Primer (UAP) were used for 3′ RACE, and PrimerS and an Abridged Anchor Primer (AAP) were used for the first round of 5′ RACE. The PCR product obtained from the first round of 5′ RACE was used as a template for nested PCR using the UAP and SpaseD primers. The full-length CVSP was completed as a single fragment using the designed primers: dehaF1 for precursor CVSP, dehaF2 for mature CVSP and the com-mon reverse primer dehaR. The design of specific primers was based on the sequences that were obtained from the RACE technique. All of the primers are listed in Table 1. The PCR reactions were performed as described above. The PCR products were visualised by 1.5% agarose gel electrophoresis and then cloned and sequenced using the methods described aboveSequence analyses and multiple alignmentSequence analysis was performed using BioEdit soft-ware. The Basic Local Alignment Search Tools (BLAST) (http://www.ncbi.nlm.nih.gov/) program was used to re-trieve similar sequences from Nr/Nt, Nr, SwissProt and the Protein Data Bank (PDB). The blastn algorithm with the Nr/Nt and Nr databases was used for nucleotide se-

Turk J Biochem 2015; 40(2):181-187 182 Meunchan et al.

IntroductionSerine proteases are widely distributed in all kingdoms of cellular life [1]. They are well-known proteolytic enzymes that are named due to the role played by a nucleophilic serine residue in the catalysis of peptide bond hydroly-sis along with two other amino acids: a basic residue, as an histidine, and an acidic residue, as an aspartate [2,3]. Proteases using this catalytic triad exhibit at least four dis-tinct folds, as illustrated by trypsin, subtilisin, prolyloligo-peptidase and Clp peptidase, and correspond to clan PA, SB, SC and SK, respectively from the MEROPS protease database [4]. The majority of serine proteases uses the ca-nonical catalytic triad and follow the chymotrypsin-like fold of clan PA [5]. Proteolytic enzymes are a major com-ponent of venom. They mainly affect the haemostatic sys-tem of the prey or victim [6]. Indeed, a number of snake venom serine proteases that affect haemostasis have been reported, including procoagulant, anticoagulant, platelet aggregating or fibrinolytic proteases [7-9].Although stings and envenomation by centipede are fairly frequent, they result more often in benign and lo-cal symptoms with spontaneous healing, and treatment is mainly supportive. Thus, knowledge on centipede venom is limited, because centipedes are regarded as a neglected group of venomous animals due to their limited venom quantity [10]. Nevertheless, centipede venom is a com-plex mixture and may be a potential source of bioactive molecules. In this study, we report the molecular cloning of a cDNA from the S. subspinipes dehaani venom gland that encodes a serine protease. The sequence was anal-ysed in comparison to the known sequence.

Materials and MethodsTotal RNA isolation and cDNA synthesisThe centipedes, S. subspinipes dehaani, were collected from a suburban area of Khon Kaen City, Thailand. Their forcipules were dissected, and the venom glands were then immediately ground in liquid nitrogen. Fifty mg of tissue was homogenised in TRIzol® reagent. Total RNA was isolated using TRIzol® (Invitrogen Life Technolo-gies, USA) according to the manufacturer’s instructions. The quality of the RNA was verified by 1.2% agarose gel electrophoresis. RNA concentration was determined us-ing a nanodrop spectrophotometer (Thermo Scientific, USA). Then, the RNA was directly used for cDNA syn-thesis. First-strand cDNA was generated using the Rever-tAidTM First Strand cDNA Synthesis Kit (Thermo Scien-tific, USA). Five µg of total RNA was used in the reaction buffer, which contained 1 µl oligo (dT)18 primer, 1 µl Ri-boLock RNase inhibitor, 2 µl dNTP mix and 200 units of RevertAid M-MuLV reverse transcriptase. The cDNA was used as the template for PCR reactions.Gene construction and sequencingThe cDNA sequence encoding a partial centipede venom

Table 1. List and sequence of primers

Primer Sequence 5’-3’

primer H GTCGTTACTGCTGCTCATTG

primer S GGACCACCTGAATCACC

SPaseD TCCAATTTCAACAAAGCTATGTCG

dehaF1 ATGATCGTTCTTTTGAATATATTGATC

dehaF2 ATTATCGGCGGAACAATTGTCG

dehaR GGAATGAGTAGTAATAATGTTATT

were then aligned using the “K*Sync” alignment method [17]. From this alignment, a template was generated and the variable regions were modelled in the context of the fixed template using Rosetta fragment assembly. Finally, the side chains of the final model were repacked using a Monte-Carlo algorithm with a back-bone dependent side chain rotamer library. Pymol was used to visualise and generate pictures of three-dimensional models (http://py-mol.org/).

Results and DiscussionFull-length cDNA determinationA partial sequence of the cDNA encoding a centipede venom serine protease (CVSP) was identified from total RNA (Fig. 1a) by RT-PCR using specific primers (Table 1) that were designed according to the conserved catalytic amino acids of serine protease: histidine (primer H) and serine (primer S) [11]. The expected PCR product (443 bp) showed a single sharp band for the partial CVSP se-quence (Fig. 1b). The full-length cDNA was obtained by RACE system using specific primers (primer H and SPas-eD), designed on the basis of the partial sequence and the universal primers provided from the RACE kit. The full-length cDNA consists of 1,014 bp with 780 bp of an open reading frame (ORF), a 5′-untranslated region (5′-UTR) of 101 bp upstream from the first ATG and a 3′-UTR of 133 bp downstream from the stop codon. The ORF en-codes a 260 amino acid residue protein with a 19-amino-acid signal peptide sequence (Fig. 2a). The precursor (780 bp) and the mature serine protease (723 bp) were cloned using the reverse primer dehaR and the forward primer dehaF1 or dehaF2, respectively (Fig. 1c).Alignment and amino acid sequence analysis of CVSPThe mature CVSP has a calculated MW of 26940.61 Da and a pI of 5.74. The N-terminal amino acid sequence

quences while the blastp algorithm with SwissProt and the Protein Data Bank (PDB) was used for protein se-quences. The parameters used for blastn were an expected threshold of 10, a word size of 11, 2 and -3 as Match/Mismatch Scores and a Gap Cost of 5 for existence and 2 for extension. For blastp, an expected threshold of 10 was used with a word size of 3, Gap Costs of 11 for exis-tence and 1 for extension and a BLOSUM62 matrix was selected as the substitution matrix.The deduced amino acid sequence was generated using the ExPASy server (http://www.expasy.org/) [12]. The signal peptide sequence was predicted with the SignalP 4.1 program with “eukaryote” selected as organism group parameter, “Input sequences do not include TM regions” to use the SignalP-noTM method and a default D-cutoff value of 0.45 for SignalP-noTM networks (http://www.cbs.dtu.dk/services/SignalP/) [13]. Disulfide bonds were predicted with the DiANNA 1.1 server and the “Disul-fide bond connectivity prediction” parameter (http://cla-vius.bc.edu/clotelab/DiANNA/) [14]. Multiple sequence alignment was carried out with MUSCLE and “aligned” as output order parameter (http://www.ebi.ac.uk/Tools/msa/muscle/) [15], while the alignment rendering was performed with ESPript (http://espript.ibcp.fr) [16]. The secondary structure of the S. subspinipes dehaani serine protease was predicted using PSIPRED with “Mask low complexity regions” as a sequence filter parameter (http://bioinf.cs.ucl.ac.uk/psipred/).Homology modelling Homology modelling was performed with the Robetta server (http://robetta.bakerlab.org/) [17]. The first step involved screening the query sequence with successive methods, BLAST, PSI-BLAST, FFAS03 and 3D-Jury, to identify regions that possessed homology with an experi-mentally characterised structure. The parent and the query

Turk J Biochem 2015; 40(2):181-187 183 Meunchan et al.

Figure 1. Agarose gel electrophoresis. (a) Total RNA extraction from the S. subspinipes dehaani venom gland, 1: 1 kb DNA ladder, 2: total RNA extract. (b) Amplification by RT-PCR, 3 and 5: 100 bp DNA ladder, 4: PCR product (443 bp) of the partial CVSP sequence (c) Amplification by RACE, 6: PCR product (780 bp) of the CVSP precursor, 7: PCR product (723 bp) of mature CVSP.

bp bp bp1 3 5

6,0003,000

3,0003,0002,5002,0001,5001,2001,000

900800700600500400

300

200

100

2,5002,0001,5001,2001,000

900800700600500400300

200

100

2,5002,0001,500

1,000

750

500

250

2 4 6 7(a) (b) (c)

serine proteases, CVSP can be classified in the S1 family of clan PA [4,18]. Substrate specificity of chymotrypsin-like proteases of the PA clan is mainly determined by the interaction between the S1 pocket and the P1 amino acid side chain of the substrate [3]. In CVSP, the S1 pocket is negatively charged due to the conserved Asp202 (Fig. 3). This suggested a substrate specificity similar to trypsin for peptides with positively charged amino acids in position P1, such as arginine or lysine.According to PSIPRED, CVSP is composed of 12 beta sheets and 2 alpha helices (Fig. 3). The predicted second-

Turk J Biochem 2015; 40(2):181-187 184 Meunchan et al.

is similar to other serine proteases with two aliphatic amino acids, isoleucines, followed by two glycines [I(V)I(V)GG] (Fig. 3). The amino acids His61, Asp108 and Ser208 that form the catalytic triad of CVSP are strictly conserved among all analysed sequences. The oxyanion hole from CVSP is formed by Gly206 and Ser208. As the catalytic residue, Ser208 is strictly conserved whereas Gly206 is conserved among centipede and snake serine protease sequences, but not with hymenoptera. According to the comparison of the sequences surrounding the cata-lytic amino acids of CVSP and evolutionary markers of

(a)

C46 C62 C141 C180 C195 C204 C214 C234

(b)

Figure 2. The sequence and predicted disulfide bonds of CVSP. (a) The nucleotide and deduced amino acid sequence for CVSP. The puta-tive signal peptide sequence is underlined. The conserved His, Asp and Ser catalytic triad are shown in bold characters, whereas the terminal codon is represented by a star (*). The polyadenylation signal is underlined with a dashed line. (b) Representation of the predicted disulfide bonds on the CVSP sequence.

234 form disulfide bonds 1, 2 and 4, respectively, and are strictly conserved, while cysteines 180 and 195 that form disulfide bond 3 are only missing in the hymenop-tera serine protease sequence (Fig. 3). Residue Ile229 of CVSP corresponds to Ser214 in bovine trypsin, which is known to form a hydrogen bond with the catalytic histi-dine and aspartate by its carbonyl and its OH side chain, respectively. This last hydrogen bond is present in most

ary structure is consistent with the secondary structure of trypsin from Bos taurus (PDB: 4GUX) (Fig. 3). However, the loop between beta strands 7 and 8, which corresponds to amino acids 146–167, is longer in CVSP than other ser-ine protease in the alignment, including serine protease of S. mutilans. The CVSP sequence also includes 8 cysteine residues that could form a maximum of 4 disulfide bonds (Fig. 2b). Cysteines 46 and 62, 141 and 214 and 204 and

Turk J Biochem 2015; 40(2):181-187 185 Meunchan et al.

Figure 3. Multiple sequence alignment of the S. subspinipes dehaani serine protease CVSP, and serine proteases from the venom of centi-pede (Scolopendra subspinipes mutilans [AAD00320]), hymenoptera (Polistes dominula [AAP37412], Vespa magnifica [ABY78898], Apis mellifera [NP_001011584]) and snake (Gloydius saxatilis [Q7SZE1], Bothrops jararacussu [Q2PQJ3], Gloydius halys [AFM29142]) and trypsin (Bos taurus [NP_001107199]). Conserved catalytic residues are indicated by a black star (), oxyanion residues by a black circle (●), Ile229 participating in the hydrogen bond network with the catalytic triad by a black triangle (▲), Asp202 participating in the S1 pocket by a black square (■) and predicted disulfide bonds by number. The secondary structure of trypsin (Bos taurus [4GUX]) and the PSIPRED predicted secondary structure of CVSP are indicated on the top.

Conflict of InterestThere are no conflicts of interest among the authors.

References[1] Rawlings ND, Barrett AJ. Evolutionary families of peptidases.

Biochem J 1993; 290( Pt 1):205-18.

[2] Polgár L. The catalytic triad of serine peptidases. Cell Mol Life Sci 2005; 62(19-20):2161-72.

[3] Hedstrom L. Serine protease mechanism and specificity. Chem Rev 2002; 102(12):4501-24.

[4] Rawlings ND, Waller M, Barrett AJ, Bateman A. MEROPS: the database of proteolytic enzymes, their substrates and inhibitors. Nucleic Acids Res 2014; 42(Database issue):D503-9.

[5] Page MJ, Di Cera E. Serine peptidases: classification, structure and function. Cell Mol Life Sci. 2008; 65(7-8):1220-36.

[6] Serrano SM. The long road of research on snake venom serine pro-teinases. Toxicon 2013; 62:19-26.

[7] Matsui T, Fujimura Y, Titani K. Snake venom proteases affecting hemostasis and thrombosis. Biochim Biophys Acta 2000; 1477(1-2):146-56.

[8] Kini RM. Serine proteases affecting blood coagulation and fi-brinolysis from snake venoms. Pathophysiol Haemost Thromb 2005; 34(4-5):200-4.

[9] Serrano SM, Maroun RC. Snake venom serine proteinases: se-quence homology vs. substrate specificity, a paradox to be solved. Toxicon 2005; 45(8):1115-32.

[10] Rates B, Bemquerer MP, Richardson M, Borges MH, Morales RA, et al. Venomic analyses of Scolopendra viridicornis nigra and Scol-opendra angulata (Centipede, Scolopendromorpha): shedding light on venoms from a neglected group. Toxicon 2007; 49(6):810-26.

[11] You WK, Sohn YD, Kim KY, Park DH, Jang Y, et al. Purification and molecular cloning of a novel serine protease from the centi-pede, Scolopendra subspinipes mutilans. Insect Biochem Mol Biol 2004; 34(3):239-50.

[12] Artimo P, Jonnalagedda M, Arnold K, Baratin D, Csardi G, et al. ExPASy: SIB bioinformatics resource portal. Nucleic Acids Res 2012; 40:597-603.

Turk J Biochem 2015; 40(2):181-187 186 Meunchan et al.

chymotrypsin-like proteases [3,19]. The polar character of position 229 is well conserved in the alignment; how-ever, only CVSP presents a non-polar residue for this po-sition (Fig. 3). The presence of a non-polar isoleucine at position 229 suggests an unusual hydrogen bond network in the catalytic site of CVSP.Homology modelling of CVSPThe homology model was built using the Robetta server [17] on the structure 1YC0, the human hepatocyte growth factor activator, which has a 33.6% identity with CVSP. The model exhibited the classical fold of clan PA prote-ases, referred to as a chymotrypsin-like fold and composed of two β-barrels with a classical Greek-key architecture for the β strand topology of each β-barrel (Fig. 4) [5]. The catalytic triad, His61, Asp108 and Ser208, is found in close proximity in the active site cleft between the two β-barrels. Ile229 is part of the amino acids surrounding the catalytic triad that form an extensive hydrogen bond network (Fig. 4) [20]. However, the non-polar properties of the Ile229 side chain suggest an unusual organisation of this hydrogen bond network, making CVSP an interesting candidate to investigate. A better understanding of the molecular mech-anism of venom components will be useful for further ap-plications and help to improve envenomation treatment.AcknowledgementsThis research was supported by the Thailand Research Fund (TRF) through the Royal Golden Jubilee Ph.D. pro-gram (PHD/ 0282/2549) and the “TRF-CHE Research Grant for Mid-Career University Faculty” and was jointly funded by the TRF and the Office of the Higher Education Commission (OHEC), Ministry of Education, Thailand for fiscal years 2007-2009 and additionally supported by the Khon Kaen University (KKU) Research Fund for fis-cal years 2007-2010.

(a) (b) (c)

Figure 4. Homology modelling of the S. subspinipes dehaani serine protease. (a) The homology model shows the two β-barrels at the N- and C-terminus, the C-terminal α helix and the loop from amino acids 146–167. The side chains of the catalytic triad residues His61, Asp108, and Ser208 lie in a cleft between the two β-barrels and are shown as sticks, while the amide nitrogen of the backbone for oxyanion residues Gly206 and Ser208 are indicated as blue spheres. Ile229 is represented as a stick, and the methyl group that is usually replaced by the OH of a serine in the homologous serine protease is indicated as a yellow sphere. (b) Homo sapiens hepatocyte growth factor activator, PDB 1YC0, was used as template for homology modelling. The OH of the serine 245 side chain is indicated as a red sphere. (c) Superposition of both structures, with the model in blue and the template in green.

[17] Kim DE, Chivian D, Baker D. Protein structure prediction and anal-ysis using the Robetta server. Nucleic Acids Res 2004; 32:526-31.

[18] Krem MM, Di Cera E. Molecular markers of serine protease evolu-tion. EMBO J 2001; 20(12):3036-45.

[19] McGrath ME, Vásquez JR, Craik CS, Yang AS, Honig B, et al. Per-turbing the polar environment of Asp102 in trypsin: consequences of replacing conserved Ser214. Biochemistry 1992; 31(12):3059-64.

[20] Meyer E. Internal water molecules and H-bonding in biological macromolecules: a review of structural features with functional implications. Protein Sci 1992; 1(12):1543-62.

[13] Petersen TN, Brunak S, von Heijne G, Nielsen H. SignalP 4.0: discriminating signal peptides from transmembrane regions. Nat Methods 2011; 8(10):785-6.

[14] Ferrè F, Clote P. DiANNA: a web server for disulfide connectivity prediction. Nucleic Acids Res 2005; 33:230-2.

[15] Edgar RC. MUSCLE: multiple sequence alignment with high accu-racy and high throughput. Nucleic Acids Res 2004; 32(5):1792-7.

[16] Gouet P, Robert X, Courcelle E. ESPript/ENDscript: Extracting and rendering sequence and 3D information from atomic structures of proteins. Nucleic Acids Res 2003; 31(13):3320-3.

Turk J Biochem 2015; 40(2):181-187 187 Meunchan et al.

Supplementary Figure 1. Figure 1. Pairwise alignment of CVSP (centipede venom serine protease, Scol-opendra subspinipes dehaani [AIJ10719]) and HGFA (hepatocyte growth factor activator, Homo sapiens [1YC0_A]). Conserved catalytic residues are indicated by a black star (), oxyanion residues by a black circle (●), Ile229 participating in the hydrogen bond network with the catalytic triad by a black triangle (▲), Asp202 participating in the S1 pocket by a black square (■) and predicted disulfide bonds by number. The secondary structure of HGFA (Homo sapiens [1YC0_A] is indicated on the top.

Supplementary File 1 [Ek 1]