INTRODUCTION
Heat shock proteins (HSPs) are some of the phylogenetically conserved and ubiquitously expressed protein families in bacteria, mammals and plants. These proteins play essential roles in stress tolerance and the folding, activation and assemblage of many proteins. According to their homology, function and size, they can be divided into different families, e.g. HSP110, HSP90, HSP70, HSP60, HSP40 and small HSP (Buchanan, Reference Buchanan2000; Gullo and Teoh, Reference Gullo and Teoh2004). The molecular chaperone HSP90 is important in the folding and functioning of many proteins involved in cell survival, especially those participating in cell cycle regulation and signal transduction (Pearl and Prodromou, Reference Pearl and Prodromou2006). Due to its broad functions, it is highly abundant in both stressed and non-stressed cells, and constitutes 2·8% of the total cellular protein (Brandau et al. Reference Brandau, Dresel and Clos1995). It has also been implicated as molecular marker for identifying and differentiating parasite species and genotypes, such as Cryptosporidium and Ditylenchus (Feng et al. Reference Feng, Dearen, Cama and Xiao2009; Vovlas et al. Reference Vovlas, Troccoli, Palomares-Rius, De Luca, Cantalapiedra-Navarrete, Liébanas, Landa, Subbotin and Castillo2015). Furthermore its contributions to the immune response have led to encouraging studies of its use as an antigen or adjuvant in vaccine, especially for cancer and get the promising results (Tosti et al. Reference Tosti, di Pietro, Ferrucci and Testori2009; Reitsma and Combest, Reference Reitsma and Combest2012; Crane et al. Reference Crane, Han, Ahn, Oehlke, Kivett, Fedoroff, Butowski, Chang, Clarke, Berger, McDermott, Prados and Parsa2013).
However, little is known about the Hsp90 gene of piroplasms. Gerhards et al. (Reference Gerhards, Ebel, Dobbelaere, Morzaria, Musoke, Williams and Lipp1994) showed that an 87 kDa HSP90 protein was expressed by Theileria parva during both sporozoite and schizont but not in the piroplasm phases, although the corresponding transcript was detected. Unlike that of other microbial pathogens, T. parva and Theileria annulata-induced IкB kinase activity does not require functional HSP90 in the schizont stage (Hermann and Dobbelaere, Reference Hermann and Dobbelaere2006). Ruef et al. (Reference Ruef, Ward, Oxner, Conley, Brown and Rice-Ficht2000) based on the phylogenetic analysis of HSP70 and Hsp90 genes, showed that Babesia microti was paraphyly with Babesia bovis and Theileria, which supports B. microti should be a basal group to Babesia and Theileria rather than Theileria. Khan et al. (Reference Khan, He, Zhang, Wang, Tao, Song, Sajid, Yu, Hu, Fang, Hu, Zhou and Zhao2014) revealed that there were 2 HSP90 proteins in Babesia orientalis, BoHSP90-A and BoHSP90-B. Sera from buffalo infected by B. orientalis react with recombinant protein BoHSP90-A and BoHSP90-B. To date, nothing has been reported about HSP90 from ovine Babesia species.
Ovine babesiosis is one of the most important tick-borne hemoparasite diseases and is responsible for economic losses for small ruminant production. This disease is mainly due to B. bovis and Babesia motasi. In China, ovine babesiosis has been reported since 1980s (Chen, Reference Chen1982; Zhao et al. Reference Zhao, Li and Min1986) and is a common disease in the north of China. Several geographic strains have been collected in our institute, such as Babesia sp. BQ1 (Lintan), Babesia sp. BQ1 (Ningxian), Babesia sp. Tianzhu, Babesia sp. Madang, Babesia sp. Hebei and Babesia sp. Xinjiang (Yin et al. Reference Yin, Lu, Luo, Zhang, Lu and Dou1997; Guan et al. Reference Guan, Yin, Luo, Lu, Zhang, Ma, Yuan, Lu, Wang and Muhe2001, Reference Guan, Yin, Luo, Lu, Zhang, Gao and Lu2002; Bai et al. Reference Bai, Liu, Liu, Ren and Li2002). Phylogenetic analysis of these strains based on the 18S RNA gene or ribosomal DNA internal transcribed spacer (ITS) sequences suggest that they could be separated into 2 clusters, Babesia sp. Xinjiang and B. motasi-like clusters (Liu et al. Reference Liu, Yin, Guan, Schnittger, Liu, Ma, Dang, Liu, Ren, Bai, Ahmed and Luo2007; Niu et al. Reference Niu, Luo, Guan, Liu, Ma, Liu, Gao, Ren, Li, Qiu and Yin2009). In the present study, a full-length Hsp90 cDNA was cloned and characterized from Babesia sp. BQ1 (Lintan), and designed as BQHsp90. Five Chinese ovine Babesia isolates together with T. annulata, T. parva, B. bovis and other apicomplexa parasite species were subjected to phylogenetic analysis based on the HSP90 amino acid sequences. Furthermore, the antigenicity of recombinant BQHSP90 (rBQHSP90) expressed in the prokaryotic system was evaluated by western blot and enzyme-linked immunosorbent assay (ELISA) to investigate the potentiality as diagnostic antigen.
MATERIALS AND METHODS
Parasites and sera
A clonal line (G7) of Babesia sp. BQ1 (Lintan) was grown in vitro in sheep erythrocytes as described by Guan et al. (Reference Guan, Moreau, Brisseau, Luo, Yin and Chauvin2010b ) and infected sheep blood was cryopreserved in liquid nitrogen in the Vectors and Vector-borne Diseases (VVBD) Laboratory, LVRI, China.
Positive sera of Babesia sp. BQ1 (Lintan) (16 sera from 2 sheep), Babesia sp. BQ1 (Ningxian) (11 sera, 1 sheep), Babesia sp. Tianzhu (10 sera, 1 sheep), Babesia sp. Hebei (9 sera, 1 sheep), Babesia sp. Xinjiang (2 sera, 2 sheep), Theileria luwenshuni (3 sera, 1 sheep) and Theileria uilenbergi (1 serum, 1 sheep) collected from infected sheep, were provided by VVBD. Sera from the 9 sheep of pre-infection were considered as the negative sera (Guan et al. Reference Guan, Ma, Liu, Ren, Wang, Yang, Li, Liu, Du, Li, Liu, Zhu, Yin and Luo2012b ). And sera collected from 3 sheep (numbers 2007, 3216 and 3533) (Guan et al. Reference Guan, Chauvin, Rogniaux, Luo, Yin and Moreau2010a ) during 84 days post-infection (dpi) with Babesia sp. BQ1 (Lintan) were used to evaluate antibody kinetics. All sera were pre-absorbed against lysates of Escherichia coli BL21 as previously described by Guan et al. (Reference Guan, Moreau, Liu, Ma, Rogniaux, Liu, Niu, Li, Ren, Luo, Chauvin and Yin2012a ).
Construction and immunoscreening of cDNA expression library
Purification of merozoites, construction and immunoscreening of Babesia sp. BQ1 (Lintan) cDNA expression library refer to the previous description (Guan et al. Reference Guan, Moreau, Liu, Ma, Rogniaux, Liu, Niu, Li, Ren, Luo, Chauvin and Yin2012a ). Briefly, merozoites were purified from the in vitro culture when parasitemia reached 8–10%. The purified mRNA was used to construct cDNA library (NOVAGEN, USA). The Babesia sp. BQ1 (Lintan) merozoite cDNA expression library was immunoscreened using immune sera collected from sheep infected with Babesia sp. BQ1 (Lintan) in 6th week post-infection. Ninety-three positive plaques were revealed by primary screening of the library on plates. Phage plugs were removed from the plates according to the sites of positive signal on the membrane and subjected to re-screening until all the signals on the membranes were positive. Pure phage stock was converted to plasmid by using the in vivo auto-subcloning capabilities of the loxP-cre system of λ screen vector in host strain BM25·8. Recombinant plasmid isolated from BM25·8 was transformed into the host strain JM109 and sent to TaKaRa Company (China) for sequencing. Sequence analysis was done using the Lasergene software package for Windows (DNASTAR, Madison, WI) and the National Center for Biotechnology Information (NCBI) database. The nucleotide sequence of an Expressed Sequence Tag (EST) fragment showed high homology with the Hsp90 gene of B. bovis based on the BLASTn in GenBank and was designated as BQHsp90.
Amplification of the full-length cDNA of BQHsp90
5′ end of BQHsp90 was amplified using SMART™ rapid amplification of cDNA ends (RACE) cDNA amplification kit (Clontech Laboratories, USA) from cDNA of Babesia sp. BQ1 (Lintan) with Gene specific primer 1 (5′- ACGCTCAGTCCACCTCCTCCATCTT -3′) designed from 3′ end of BQHsp90 EST fragment according to the manufacturer's instructions (Guan et al. Reference Guan, Moreau, Liu, Ma, Rogniaux, Liu, Niu, Li, Ren, Luo, Chauvin and Yin2012a ). Amplified polymerase chain reaction (PCR) fragments were routinely cloned into pGEM-T easy vector (Promega, USA) and nucleotide sequences determined by the TakaRa Company (China). The full-length cDNA sequence of BQHsp90 was assembled using the Lasergene software package for Windows (DNASTAR, Madison, WI) and the open reading frame (ORF) was determined using ORF Finder (www.ncbi.nlm.nih.gov/gorf).
Characterization of BQHSP90
Multiple sequence alignment was performed on the deduced BQHSP90 amino acid sequence and those of B. bovis (AAF61428), T. annulata (XP_952473), T. parva (AAA30132) and Plasmodium falciparum (CAA82765) using MEGA 4. The putative signal peptide was predicted using SignalP (www.cbi.dtu.dk/services/SignalP). The molecular mass (Mw) and theoretical isolectric point (pI) were calculated on line (www.expasy.org/tools.pi_tool.html). Motif scan and transmembrane topology prediction were done using MyHits (http://myhits.isb-sib.ch/cgi-bin/motif_scan). The structure prediction of BQHSP90 was performed using the I-TASSER online services (http://zhanglab.ccmb.med.umich.edu/services/) and SWISS-MODEL (http://www.expasy.org/swissmod/SWISS-MODEL.html).
The structure of the BQHsp90 genomic sequence was confirmed by performing PCR analysis with BQHsp90-G primers (F10-34: AGTATCTACCCAGCGACATCTTTCT, R2200-2221: TCAGTCCACCTCCTCCATCTTA) designed from the sequence data of the BQHsp90 cDNA clone with initial denaturation at 94 °C for 1 min, then 94 °C for 30 s; 55 °C for 30 s and 72 °C for 2 min, for 35 cycles, then at 72 °C for 10 min. Sequencing was performed as described above. Locations and exon–intron structures of the BQHsp90 genomic sequences were predicted using the GENSCAN programme (http://genes.mit.edu/GENSCAN.html) and ClustalW.
Phylogenetic analysis
Hsp90 genomic DNA (gDNA) was amplified with primers UHSP90 (F436–460: CGGTGTCGGTTTCTACTCGGCTTAC, R2200–2224: CGCTCAGTCCACCTCCTCCATCTTA) designed on the conserved region of Hsp90 from Chinese Babesia isolates infective for small ruminants, Babesia sp. BQ1 (Lintan), Babesia sp. BQ1 (Ningxian), Babesia sp. Hebei, Babesia sp. Tianzhu and Babesia sp. Xinjiang (at 94 °C for 1 min as initial denaturation, then 94 °C for 30 s; 63 °C for 30 s and 72 °C for 2 min, for 35 cycles, then at 72 °C for 10 min). The PCR products were processed as described above for sequencing by the TakaRa Company. The prediction of introns and splicing sequences, and the deduction of amino acid sequences were performed with GENSCAN (http://genes.mit.edu/GENSCAN.html) and the ClustalW programme in MEGA 4 software. Several HSP90 protein sequences of apicomplexa species that are important for humans and domestic animals were identified, i.e. B. bovis, T. annulata, T. parva, Toxoplasma gondii, Eimeria acervulina, Eimeria tenella, Crytosporidium parvum, Cryptosporidium muris, Plasmodium vivax and P. falciparum and downloaded from the Blast server (www.ncbi.nlm.nih.gov/Blast) (Table 1). Multiple sequence alignment and evaluation of the phylogenetic relationships were performed using the ClustalW programme in MEGA 4 and DNAStar.
Table 1. Information about the apicomplexa species included in this study
a Derived from cDNA.
b Derived from gDNA.
Prokaryotic expression of the BQHsp90 gene
The BQHsp90 containing the entire ORF was amplified from cDNA of Babesia sp. BQ1 (Lintan) using BQHsp90 primers and the PCR product was ligated into pGEM-T-easy (pGEM-BQHsp90) for sequencing. The Champion™ pET Directional TOPO® Expression Kit (Invitrogen, USA) was used for the expression of BQHSP90. The ORF of BQHsp90 was amplified from recombinant plasmid pGEM-BQHsp90 with a pair of expression primers BQHsp90-TOPO: F68–88: CACCATGGCGACGGAGAGTCAGGAG, R2197–2221: TCAGTCCACCTCCTCCATCTTAGGG) as follows: initial denaturation at 94 °C for 1 min, then 94 °C for 30 s; 65 °C for 30 s and 72 °C for 2 min, for 30 cycles, then at 72 °C for 10 min. The PCR product was gel-purified with Agarose Gel DNA Extraction Kit (TakaRa, China). The pET200/D-TOPO® cloning reaction was set up and recombinant plasmid was constructed into One Shot® TOP10 Chemically Competent E. coli for characterization following the user manual. After sequencing, the recombinant plasmid extracted from One Shot® TOP10 was transformed into BL21 Star™ (DE3) One Shot® Chemically Competent E. coli for expression. To induce recombinant BQHSP90 (rBQHSP90) expression, isopropyl β-D-1-thiogalactopyranoside (IPTG) was added to a final concentration of 1 mm, and expression was induced for 4 h at 37 °C. The rBQHSP90 was purified from supernatants of the lysates by Ni affinity chromatography according to the manufacturer's protocol (Invitrogen, USA).
Antigenicity and specificity analysis of rBQHSP90 by western blot and ELISA
The details could refer to previous description (Guan et al. Reference Guan, Moreau, Liu, Ma, Rogniaux, Liu, Niu, Li, Ren, Luo, Chauvin and Yin2012a ). Briefly, for western blot, the purified rBQHSP90 proteins were electrophoresed and transferred to nitrocellulose (NC) membranes. The NC sheets were blocked with 10% skimmed milk powder in 0·1 m Tris-buffered saline (pH 7·6) and 0·1% Tween (TBST) overnight at 4 °C, and then incubated for 1 h with sera diluted at 1/20 in TBST. The sheets were incubated with monoclonal anti-goat/sheep IgG-alkaline phosphatase conjugates (Sigma) diluted at 1/1000 in TBST for 1 h. Positive signals were revealed using 5-bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium (BCIP/NBT) liquid substrate system (Sigma). For ELISA, microplates (Nunc) were coated with 2 µg mL−1 of rBQHSP90 in 0·1 m pH 9·6 carbonate buffer at 37 °C for 1 h and then at 4 °C overnight. The plates were blocked with 150 µL of 2% gelatin in carbonate buffer at 37 °C for 30 min. After drying the plate, blank (phoshphate-buffered saline containing 0.1% Tween 20, PBST) and sera (dilution of 1:20) were distributed in duplicate and the plates were incubated at 37 °C for 1 h. Peroxidase conjugate of monoclonal anti-goat/sheep IgG clone GT-34 (Sigma) diluted at 1:1000 was added to each well and the plates were again incubated at 37 °C for 1 h. 50 µL 3, 3′, 5, 5′ - Tetramethylbenzidine (TMB) (Sigma) were added to each well and incubated at room temperature for 15 min. The reaction was stopped by adding 50 µL of 0·1 m H2SO4 and the plates were then read at 450 nm with an ELISA automat (Bio-RAD, USA).
RESULTS
Construction and immunoscreening of cDNA expression library
Analysis of the cDNA expression library of Babesia sp. BQ1 (Lintan) merozoites revealed sequences more than 200 bp in length involved in 10 EST with high homology to B. bovis genes. Except for 2 hypothetical proteins, these included gliding-associated protein 45 (GAP45), p200, Rab1b, histone H2A protein, cyclophilin, RNA recognition motif containing protein, membrane protein and HSP90 by sequence alignment. The clone 45, a 699 bp length containing a 94 bp poly (A) tail but no entire ORF was found. It showed 81% identity with the Hsp90 gene of B. bovis (accession numbers: XM_001611504 and AF136649) when aligned by BLASTn (http://blast.ncbi.nlm.nih.gov/Blast.cgi) and was designated BQHsp90.
Amplification and characteristics of BQHsp90 full-length
A complete cDNA sequence of BQHsp90 was obtained based on the 5′ RACE amplification. BQHsp90 is a 2392 bp cDNA containing a 2154 bp ORF which encodes a predicted 717 amino acid residues protein with a theoretical pI of 5·0 and mass of 83 kDa. The sequence was deposited in GenBank under the accession number GQ397856. Multiple sequence alignment of the deduced BQHSP90 amino acid sequence with the HSP90 of B. bovis, T. annulata, T. parva and P. falciparum revealed identities of 92·0, 81·9, 81·7 and 70·6%, respectively.
No signal sequence was identified in the transcript using the SignalP software. MyHits analysis revealed that BQHSP90 contains a HSP90 family signature at position 27–36 and a tetratricopeptide repeat (TPR) binding site at the COOH-terminus. In addition, a nuclear localization signal (NLS) was also detected from 356 to 359 residues in BQHSP90. SMART and MyHits analysis revealed a conserved adenosine triphosphate (ATP) binding domain between positions 29 and 183, including a conserved GxxGxG motif from 121 to 126 residues (essential for ATP binding) and HATPase_c domain from 31 to 144 residues (essential for ATPase activity) (Fig. 1). And simple three-dimensional (3D) structure analysis using the crystal structure of yeast HSP90 (Protein Data Bank (PDB) code: 2cg9) as a template showed that BQHSP90 comprised 3 domains including N-terminal, middle domain and C-terminal domain, and the binding sites for ATP located in N-terminal. All these domains/subdomains, essential for client protein binding and ATP hydrolysis confirmed that BQHSP90 should belong to HSP90 family.
Fig. 1. Multiple sequence alignment of HSP90 proteins from Babesia sp. BQ1 (Lintan), B. bovis, T. annulata, T. parva and P. falciparum. The dots and short lines in the sequence represent identical residues and non-existent residues, respectively. Abbreviations: Hsp90, heat shock protein 90.
gDNA and cDNA of BQHsp90 were amplified with BQHsp90-G primers and sequenced. Exon–intron structure analysis with GENSCAN revealed that the BQHsp90 gDNA consisted of 2 introns and 3 exons, and the introns were 403 and 142 bp in length and located at sites 1423/1424 and 1924/1925, respectively. The predicted splice sites of the introns conformed to the guanine thymine - adenine guanine (GT-AG) rule (online Fig. S1).
Phylogenetic analysis
Hsp90 gene fragments were amplified from all gDNA of Babesia sp. BQ1 (Lintan), Babesia sp. BQ1 (Ningxian), Babesia sp. Hebei, Babesia sp. Tianzhu and Babesia sp. Xinjiang. The sequence sizes were 2333, 2339, 2340, 2332 and 2166 bp, corresponding to the accession numbers GQ443608, GQ443604, GQ443605, GQ443606 and GQ443607, respectively. Exon–intron analyses of these genomic Hsp90 with GENSCAN revealed that the variety of sequence lengths resulted in different sized introns (in GenBank database). Multiple sequence alignments based on the gDNA and deduced amino acid sequences of HSP90 in 5 Chinese ovine Babesia isolates revealed that the major varieties are present in the charged linker region (online Fig. S2). The identities of the 5 Chinese ovine Babesia isolates were determined on the basis of the nucleotide and amino acid sequences of HSP90. The minimum identities of nucleotide and amino acid sequences in 5 Chinese Babesia isolates were present in Babesia sp. Xinjiang (less than 80 and 90%). The identities of the amino acid and nucleotide sequences were more than 97·8 and 93·2%, respectively, between Babesia sp. BQ1 (Lintan), Babesia sp. BQ1 (Ningxian), Babesia sp. Hebei and Babesia sp. Tianzhu. In addition, the maximum identities were found for Babesia sp. BQ1 (Lintan) and Babesia sp. Tianzhu (99·3 and 99·4% for the amino acid and nucleotide sequence) and for Babesia sp. BQ1 (Ningxian) and Babesia sp. Hebei (99·5 and 98·8% for the amino acid and nucleotide sequence) (online Table S1). The bootstrap test of phylogeny for the apicomplexa HSP90 amino acid dataset generated identical tree topologies for neighbour-joining (NJ) analyses, using the ClustalW programme in software MEGA 4. The relationships were in agreement with the traditional taxonomic classification as all Babesia, Theileria, Cryptosporidium, Eimeria, Toxoplasma and Plasmodium were classified into separate branches. The Babesia group could be divided into 3 clades: Babesia sp. Xinjiang, B. bovis and the 4 Chinese B. motasi-like isolates. However, these 4 Chinese B. motasi-like isolates could be further divided into 2 subclades, one including Babesia sp. BQ1 (Lintan) and Babesia sp. Tianzhu and the other Babesia sp. BQ1 (Ningxian) and Babesia sp. Hebei (Fig. 2).
Fig. 2. Phylogenetic relationships between 5 Chinese Babesia isolates and 10 other apicomplexa species based on the amino acid dataset of Hsp90 genes. The evolutionary history was inferred using the NJ method. The percentages of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) are shown next to the branches. The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. There were a total of 564 positions in the final dataset. Phylogenetic analyses were conducted in MEGA4. Abbreviations: NJ, neighbour-joining; Hsp90, heat shock protein 90.
Expression and antigenic analysis of rBQHSP90
An entire BQHsp90 ORF was successfully inserted into the pET200/D-TOPO® vector and expressed in BL21 Star™ (DE3). The recombinant BQHSP90 protein (rBQHSP90) was expressed in 2 forms; soluble protein and inclusion bodies. The soluble rBQHSP90 protein was purified from supernatant of recombinant BL21 (Fig. 3A). Western blot analysis showed that Babesia sp. BQ1 (Lintan) infected sheep serum could specifically recognize the rBQHSP90. No reactions from rBQHSP90 were detected with negative sera, or from the pET200/D-TOPO® vector control with either positive or negative sera (Fig. 3B), which indicates BQHSP90 could induce antibodies production when animal was infected by Babesia sp. BQ1 (Lintan).
Fig. 3. Sodium dodecyl sulfate polyacrylamide gel electropheresis (SDS-PAGE) (A) and western blot (B) of rBQHSP90. (A) M is the marker of mass (kDa); 1–4 represent lysates of recombinant BL21 (DE3) before induction, lysates of recombinant BL21 (DE3) after induction (LRAI), soluble proteins in LRAI and purified rBQHSP90, respectively. (B) M is the protein marker (kDa). 1 and 3 are rBQHSP90 protein reacted each with positive serum from sheep infected by Babesia sp. BQ1 (Lintan) and negative serum; 2 and 4 are pET200/D/lacZ control reacted with positive serum from sheep infected by Babesia sp. BQ1 (Lintan) and negative serum, respectively. The right and left numbers represent the mass of each band in the protein marker.
Specificity analysis of rBQHSP90 for Babesia sp. BQ1 (Lintan) infection
Specificity of rBQHSP90 for Babesia sp. BQ1 (Lintan) infection was evaluated via testing sera from individually infected sheep by Babesia sp. BQ1 (Lintan), Babesia sp. BQ1 (Ningxian), Babesia sp. Tianzhu, Babesia sp. Hebei, Babesia sp. Xinjiang, T. luwenshuni, T. uilenbergi, together with negative sera, using ELISA and western blot. The results indicated that rBQHSP90 could not specifically differentiate Babesia sp. BQ1 (Lintan) from other Babesia species/strains infection. However, rBQHSP90 had no reaction with positive sera of T. luwenshuni and T. uilenbergi (Fig. 4). Thus, BQHSP90 could be used to detect Babesia infection but not to distinguish different Babesia species infection. In addition, the antibody kinetics of BQHSP90 were evaluated using sera from 3 sheep during 84 dpi with Babesia sp. BQ1 (Lintan). BQHSP90 specific antibodies produced from 1st week post-infection, and peaked in 3rd–4th weeks post-infection. In 5th week, it returned to the level of 1st week and later, showed a fluctuation (Fig. 5).
Fig. 4. Specificity analysis of rBQHSP90 for ovine piroplasms infection. (A) Level of BQHSP90 protein specific antibodies in sera from sheep of pre- and post-infection in ELISA. (B) Western blot analysis of rBQHSP90 specificity using sera from sheep of pre- and post-infection. M, standard molecular weight markers (kDa); 1, negative sera from sheep of pre-infection; 2–8, positive sera from sheep infected by Babesia sp. BQ1 (Lintan), Babesia sp. Tianzhu, Babesia sp. BQ1 (Ningxian), Babesia sp. Hebei, Babesia sp. Xinjiang, T. luwenshuni and T. uilenbergi, respectively.
Fig. 5. Kinetics of humoral response against rBQHSP90 of 3 sheep experimentally infected by Babesia sp. BQ1 (Lintan).
DISCUSSION
A Hsp90 gene from a B. motasi-like parasite was firstly cloned and characterized in the present study. The protein encoded by the BQHsp90 gene is an 83 kDa member of the HSP90 family, sharing high homology with B. bovis, T. annulata and T. parva HSP90 around the N- and C-terminal domains. In silico, several HSP90 conserved signatures, such as the ATP-binding domain, HSP90 signature, TPR binding site, GXXGXG motif and NLS, were identified on BQHSP90 protein. Comparison of the gDNA and cDNA sequences of BQHSP90 showed that the BQHsp90 gene contains 2 introns. This is in agreement with reports that all Hsp90s contain introns but that the number of introns differs from one to another (Girvitz et al. Reference Girvitz, Ouimet and Kapoor2000; De Luca et al. Reference De Luca, Di Vito, Fanelli, Reyes, Greco and De Giorgi2009; Khan et al. Reference Khan, He, Zhang, Wang, Tao, Song, Sajid, Yu, Hu, Fang, Hu, Zhou and Zhao2014). The structural analysis of BQHSP90 also showed high similarity with the 3D structure of other HSP90 proteins, i.e. presence of N-terminal domain and COOH-terminal domain linked by a charged linker region. As the BQHSP90 protein exhibits several structural characteristics common to the HSP90 family, it probably shares some important functions such as ATPase activity. For example, Kumar et al. (Reference Kumar, Pavithra and Tatu2007) and Zhang et al. (Reference Zhang, Botër, Li, Kadota, Panaretou, Prodromou, Shirasu and Pearl2008) used 3D structure analysis to show that conserved domains of HSP90 were involved in nucleotide binding, ATPase activity, co-chaperone binding and intersubunit interactions. It might therefore be possible to use inhibitors of BQHSP90 to develop therapeutic treatments against Babesia sp. BQ1 (Lintan) infection similar to these of toxoplasmosis and malaria previously described by Echeverria et al. (Reference Echeverria, Matrajt, Harb, Zappia, Costas, Roos, Dubremetz and Angel2005) and Kumar et al. (Reference Kumar, Musiyenko and Barik2003, Reference Kumar, Pavithra and Tatu2007).
Although small-subunit (SSU) rRNA genes are commonly used to determine the molecular phylogenies of eukaryotes and prokaryotes and some new species were discovered based on SSU rRNA gene sequences (Inokuma et al. Reference Inokuma, Yoshizaki, Shimada, Sakata, Okuda and Onishi2003; Oosthuizen et al. Reference Oosthuizen, Zweygarth, Collins, Troskie and Penzhorn2008), several authors (Philippe et al. Reference Philippe, Lopez, Brinkmann, Budin, Germot, Laurent, Moreira, Müller and Le Guyader2000; Cavalier-Smith and Chao, Reference Cavalier-Smith and Chao2003; Stechmann and Cavalier-Smith, Reference Stechmann and Cavalier-Smith2003) have demonstrated that, due to the extremely variable rate and mode of rRNA evolution, unsound methods or artefactual grouping may also produce phylogenies if we just used SSU rRNA genes in phylogenetic analysis. Thus, together with conserved protein-coding genes, it makes phylogenetic analysis more sound and close to virtual evolutional relationship of organisms. For instance, Fukuda and Endoh (Reference Fukuda and Endoh2008) used both the Hsp90 and β-tubulin genes to determine the phylogeny of the dinoflagellate Noctiluca scintillans and proposed a possible evolutionary position between the diploid dinoflagellates and haploid core dinoflagellates. To date, 18S rRNA gene was the primarily molecular marker used to understand phylogenetic relationships of piroplasms (Gubbels et al. Reference Gubbels, Yin, van der Weide, Bai, Nijman, Liu and Jongejan2000; Ahmed et al. Reference Ahmed, Luo, Schnittger, Seitzer, Jongejan and Yin2006). However, the phylogenetic analysis based on single gene marker cannot reflect the relationships of species clearly (Schnittger et al. Reference Schnittger, Rodriguez, Florin-Christensen and Morrison2012). Thus, more gene targets have been introduced for the phylogenetic analysis of piroplasms, including ITS, 28S rRNA, HSP70 and mitochondrial genes cytochrome b (cob), cytochrome c oxidase subunit I (cox I) and III (cox III) (Tian et al. Reference Tian, Liu, Yin, Luo, Guan, Xie, Luo, Zheng, Tian, Yuan, Wang, Chen and Wang2013a , Reference Tian, Luo, Zheng, Xie, Shen, Yin, Luo, Tian, Yuan, Wang and Liu b ; Yamasaki et al. Reference Yamasaki, Inokuma, Sugimoto, Shaw, Aktas, Yabsley, Yamato and Maede2007; Gou et al. Reference Gou, Guan, Liu, Ma, Chen, Liu, Ren, Li, Yang, Yin and Luo2013a , Reference Gou, Guan, Ma, Liu, Liu, Ren, Li, Yang, Chen, Yin and Luo b ).
In this study, we used the Hsp90 gene to carry out a phylogenetic analysis within the apicomplexa species and to clarify the phylogeny of Chinese ovine Babesia isolates. Multiple sequence alignment of the deduced BQHSP90 amino acid sequence showed that BQHSP90 shared more than 70% identity with the HSP90 of other apicomplexa. The Babesia, Theileria, Eimeria, Cryptosporidium, Plasmodium and Toxoplasma genera were apparent as clearly separate clusters as in traditional taxonomy. In China, several geographical strains of ovine large Babesia have been isolated from field-collected blood or ticks in the past 2 decades. Liu et al. (Reference Liu, Yin, Guan, Schnittger, Liu, Ma, Dang, Liu, Ren, Bai, Ahmed and Luo2007) subjected these Chinese ovine Babesia isolates to a phylogenetic analysis based on the 18S rRNA gene sequences, and showed that 7 Chinese isolates could be separated into 2 clusters, Babesia sp. Xinjiang and B. motasi-like (Babesia sp. BQ1 (Lintan), Babesia sp. BQ1 (Ningxian), Babesia sp. Tianzhu and Babesia sp. Hebei together with European B. motasi) clusters. Phylogenetic analyses, based on ribosomal DNA ITS sequences, suggest that all these strains, except Babesia sp. Xinjiang, should be considered as B. motasi (Niu et al. Reference Niu, Luo, Guan, Liu, Ma, Liu, Gao, Ren, Li, Qiu and Yin2009). In our phylogeny analysis, based on the amino acid sequences of HSP90, the results were similar to those of analyses based on ribosomal genes, 18S rRNA and ITS genes in that Babesia sp. Xinjiang, B. bovis and 4 Chinese B. motasi-like isolates were separated into 3 distinct clusters on the phylogenetic tree. This provides further evidence that Babesia sp. Xinjiang appears to be a new Babesia species infective for small ruminants. Interestingly, the 4 Chinese B. motasi-like isolates could be further divided into 2 subclades, one containing Babesia sp. BQ1 (Lintan) and Babesia sp. Tianzhu and the other Babesia sp. BQ1 (Ningxian) and Babesia sp. Hebei. Multiple sequence alignment indicated that Babesia sp. BQ1 (Lintan) and Babesia sp. Tianzhu were more closely related, as were Babesia sp. BQ1 (Ningxian) and Babesia sp. Hebei. Uilenberg (Reference Uilenberg2006) provided a detailed description of the taxonomic state of B. motasi, declaring that it could be separated into at least 2 species or subspecies based on the differences in pathogenicity (low virulence in northern Europe and high virulence in southern Europe and the Mediterranean basin), infectivity to sheep and goats, serology and even morphology. The taxonomy of the Chinese B. motasi-like isolates, with one group of low virulence (Babesia sp. BQ1 (Lintan) subclade) (Guan et al. Reference Guan, Yin, Luo, Lu, Zhang, Gao and Lu2002) and another of high virulence (Babesia sp. BQ1 (Ningxian) subclade) (Bai et al. Reference Bai, Liu, Liu, Ren and Li2002), seems to align with this viewpoint. In addition, Guan et al. (Reference Guan, Chauvin, Rogniaux, Luo, Yin and Moreau2010a ) showed that a soluble merozoite antigen of Babesia sp. BQ1 (Lintan) from in vitro culture cross-reacted with sera from Babesia sp. Tianzhu infected sheep, but not with those from Babesia sp. BQ1 (Ningxian) and Babesia sp. Hebei. Nevertheless, before these Chinese ovine B. motasi-like isolates can be classified as 2 species or subspecies, further evidence from biological studies of the tick vector, virulence or antigenicity will be required.
HSP90-specific antibodies have been used as a diagnostic marker of disease progression in some tumours and psychiatric disorders (Shen et al. Reference Shen, Liu, Yang, Lin, Chen, Yeh and Leu2006; McCarthy et al. Reference McCarthy, Pick, Kluger, Gould-Rothberg, Lazova, Camp, Rimm and Kluger2008). They are also found in patients with various autoimmune disorders such as rheumatoid arthritis, ankylosing spondylitis, systemic lupus erythematosus and inflammatory bowel diseases. HSP90 has rarely been considered for use in the diagnosis of diseases caused by pathogen invasion, due to the presence of anti-HSP90 antibodies in the normal IgG repertoire and cross-reaction between antibodies and HSP90 from different organisms (Pashov et al. Reference Pashov, Kenderov, Kyurkchiev, Kehayov, Hristova, Lacroix-Desmazes, Giltiay, Varamballi, Kazatchkine and Kaveri2002; Virdi et al. Reference Virdi, Thakur, Dutt, Kumar and Singh2009). Nevertheless, the level of anti-HSP90 specific antibodies significantly increases when hosts are infected by pathogens. De Andrade et al. (Reference De Andrade, Kirchhoff, Donelson and Otsu1992) showed that recombinant HSP90 from Leishmania donovani donovani did not cross-react with sera from Trypanosoma cruzi and T. gondii patients in either western blot or ELISA. HSP90 from Trichinella spiralis was also specific and showed no reaction with irrelevant immune rat sera by western blot (Martinez et al. Reference Martinez, Pérez-Serrano, Bernadina and Rodríguez-Caabeiro2001). In the present study, no reaction was detected between recombinant BQHSP90 and sera recovered from sheep of pre-infection and infected by T. luwenshuni and T. uilenbergi either in western blot or ELISA. Sera positive for Babesia sp. BQ1 (Lintan) had strong reactions with rBQHSP90, following by Babesia sp. Tianzhu, Babesia sp. BQ1 (Ningxian), Babesia sp. Hebei and Babesia sp. Xinjiang. Thus, it had cross-reaction with sera from other Chinese ovine Babesia isolates and cannot be used to differentiate infection of these Babesia parasites infections. However, rBQHSP90 can be used to distinguish the infection of ovine Babesia from those of Theileria. Antibody kinetics of BQHSP90 in 3 sheep during 84 dpi with Babesia sp. BQ1 (Lintan) revealed that specific antibody levels of BQHSP90 drastically increased from 1st week post-infection, and peaked in 3rd–4th weeks post-infection. In 5th week, it returned and later, showed a fluctuation, which suggests rBQHSP90 may be a potential sero-diagnostic antigen to detect early infection of ovine Babesia.
In summary, there is a member of HSP90 family in ovine Babesia and it has similar characteristics and predictive function involved in chaperone protein, which suggests possibly it could be used as adjuvant in vaccine during the control babesiosis. Recombinant protein rBQHSP90 can specifically recognize antibodies of Babesia HSP90 post-infection. It indicates that the molecule may have potential as an antigen for detecting parasites of Babesia genus infection but not differentiating species infection. Finally, HSP90 can be considered as molecular marker for clarifying phylogenic relationships of piroplasms.
SUPPLEMENTARY MATERIAL
To view supplementary material for this article, please visit http://dx.doi.org/ 10·1017/S0031182015000797.
ACKNOWLEDGEMENTS
We are grateful for the help of Jingming Wang and Shuaiyang Zhao from Lanzhou Veterinary Research Institute for their assistance with purification of rBQHSP90.
FINANCIAL SUPPORT
This study was financially supported by the National Basic Science Research Program (973 programme) of China (no. 2015CB150300) (JX Luo); the NSFC (no. 31372432 (JX Luo), no. 31272556 (YL), no. 31402189 (JL Liu)); ASTIP, FRIP (2014ZL010), CAAS (HY); ‘948’ (2014-S05) (HY); NBCIS CARS-38 (HY); Jiangsu Co-innovation Center Programme for Prevention and Control of Important Animal Infectious Diseases and Zoonoses (HY), State Key Laboratory of Veterinary Etiological Biology Project (HY). The research was also facilitated by research funds from the French National Institute for Agricultural Research (INRA) and the École Nationale Vétérinaire, Agroalimentaire et de l'Alimentation Nantes – Atlantique (Oniris) (A. C. and E. M.).
