INTRODUCTION
Taenia solium is a cestode parasite that causes intestinal taeniasis and is widely distributed in areas where raw or insufficiently cooked pork is consumed. T. solium thrives in the human small intestine, where it produces numerous eggs that are passed out with gravid segments in the feces. Pigs are the typical intermediate host and are infected with the eggs, which develop into T. solium metacestode (TsM). Humans also serve as an intermediate host. Dissemination of TsMs in the human body produces the same situation as in pigs. TsM preferentially infects subcutaneous tissue, but may invade the central nervous system (CNS) and results in neurocysticercosis (NC). NC is one of the most important aetiologies of late-onset seizures and has been definitively associated with significant mortality and morbidity (del Brutto et al. Reference del Brutto, Sotelo and Roman1998). NC is also being increasingly recognized in developed countries owing to a rapid increase in immigration from countries where NC is endemic (Esquivel et al. Reference Esquivel, Diaz-Otero and Gimenez-Roldan2005; Sorvillo et al. Reference Sorvillo, Wilkins, Shafir and Eberhard2011). Recent epidemiological studies have indicated that the most common source of infective eggs is the home and surroundings of symptom-free carriers (Flisser et al. Reference Flisser, Rodriguez-Canul and Willingham2006; Lescano et al. Reference Lescano, Garcia, Gilman, Gavidia, Tsang, Rodriguez, Moulton, Villaran, Montan and Gonzalezet2009). T. solium infection is intimately associated with the development of human cysticercosis (Gilman et al. Reference Gilman, del Brutto, Garcia and Martinez2000; Garcia et al. Reference Garcia, Gilman, Gonzalez, Rodriguez, Gavidia, Tsang, Falcon, Lescano, Moulton, Bernal and Tovar2003).
Parasitic helminths have adapted to survive in their host. The inadequate ability to synthesize lipid molecules de novo, particularly the long-chain fatty acids (FAs), retinols and sterols, forces these organisms to rely on uptake from the host (Smyth and McManus, Reference Smyth and McManus1989; McDermott et al. Reference McDermott, Cooper and Kennedy1999). These lipidic molecules play important roles in cellular homeostasis and in the formation of macromolecular structures of helminths (Chunchob et al. Reference Chunchob, Grams, Viyanant, Smooker and Vichasri-Grams2010). The parasite should have an efficient biological system for the uptake and transport of FAs/hydrophobic molecules provided by their host. Hydrophobic-ligand-binding-proteins (HLBPs) are a group of small lipid binding proteins (7–10 kDa), which possess α-helical rich domains with hydrophobic binding sites (Barrett, Reference Barrett2009). HLBPs might participate in detoxification, transport and sequestration of lipid or hydrophobic molecules in the intracellular/extracellular phase (Glatz and van der Vusse, Reference Glatz and van der Vusse1996; Barrett, Reference Barrett2009). These proteins appear to be confined to the cestode parasites including Moniezia expansa (MeHLBP; Janssen and Barrett, Reference Janssen and Barrett1995; Barrett et al. Reference Barrett, Saghir, Timanova, Clarke and Brophy1997), Hymenolepis diminuta (HdHLBP; Saghir et al. Reference Saghir, Conde, Brophy and Barrett2001) and TsM (Lee et al. Reference Lee, Kim, Bae, Chung, Suh, Na, Kim, Kang, Ma and Kong2007; Kim et al. Reference Kim, Bae, Yang, Hong and Kong2011) since no homologous proteins have been identified in other animal taxa.
T. solium is a highly specialized flatworm and is well-adapted in the human small intestine, in which oxygen tension is extremely low. The large body surface area of T. solium involved in absorption of nutrients from the host, as well as maintenance of free FA concentration below toxic levels, strongly suggested that certain lipid-transport systems might be proficiently operational in this parasite. However, information regarding the T. solium hydrophobic ligand transporter(s) is not currently available. Identification of such molecule(s) might be valuable for a better understanding of host-parasite interaction of this clinically important parasite.
In the present study, we identified T. solium HLBPs and subsequently characterized 2 novel genes (designated as TsHLBP1 and 2) that putatively code for these proteins. The proteins were abundantly expressed in the adult stage but not in the metacestode stage. Both TsHLBP1 and 2 bound to 1-anilinonapthalene 8-sulfonic acid (1,8-ANS), cis-parinaric acid (cPnA) and 16-(9-anthroyloxy) palmitic acid (16-AP). The extracellular nature of the proteins might suggest their possible roles in sequestration and/or elimination of toxic hydrophobic molecules exposed to the complex environments within the human intestine, in addition to their role in the uptake of host derived lipid/FAs.
MATERIALS AND METHODS
Parasite samples
T. solium was collected from patients in Guangxi Province, China. The worms were identified by morphological characteristics of the mature and gravid proglottids and by observing a diagnostic 474 bp-long band that appeared in polymerase chain reaction (PCR) employing primers for T. solium-specific mitochondrial DNA (5′-CTAGGCCACTTAGTAGTTTAGTTA-3′ and 5′-CATAAAACACTCAAACCTTATAGA-3′) (Jeon et al. Reference Jeon, Chai, Kong, Waikagul, Insisiengmay, Rim and Eom2009). TsMs were collected from naturally infected pigs as previously described (Lee et al. Reference Lee, Kim, Bae, Chung, Suh, Na, Kim, Kang, Ma and Kong2007). Three adult worms and TsMs were separately incubated at room temperature in 50 ml of RPMI 1640 medium in the presence of a protease inhibitor cocktail (1 tablet/25 ml; Complete, Roche, Basel, Switzerland). The incubation medium was harvested and centrifuged at 20 000 g for 30 min, after which the supernatants were collected. The excretory-secretory products of adult (TsESP) and metacestode (TsMESP) were dialysed against phosphate-buffered saline (PBS; 100 mM, pH 7·4) for 4 h and filtered through a 0·45 μm filter (Millipore, Billerica, MA, USA). TsM cyst fluid (CF) was obtained via puncture of individual intact cysts in the presence of a protease inhibitor cocktail (Roche). The scolex and neck, and immature, mature and gravid proglottids of adult worm and the scolex, neck and bladder wall of TsM were separately homogenized with a Teflon-pestle homogenizer (Schleicher & Schuell, Dassel, Germany) in PBS (100 mM, pH 7·4) supplemented with a protease inhibitor cocktail. The homogenates and CF were centrifuged at 20 000 g for 1 h and the resulting supernatants were used as respective soluble proteins. Protein concentrations were determined by Bio-Rad protein assay kit (Hercules, CA, USA). All procedures were done at 4°C unless otherwise specified and stored at −80°C until used. All experiments involving humans and animals were approved by the Ethics Committees of the Parasitology Institute, Guangxi Centers for Disease Control and Prevention, Nanning, Guangxi, China. Informed consent was obtained from individual patients.
Electrophoresis and protein identification by mass spectrometry analysis
The crude TsESP and respective extracts were separated by 12 or 15% SDS-PAGE under reducing conditions. The gels were visualized by colloidal Coomassie blue G-250 (CBB) staining or further processed with immunoblotting. For 2-dimensional electrophoresis (2-DE), samples were precipitated with an equal volume of ice-chilled 20% trichloroacetic acid and washed in cold acetone prior to being air-dried. The dried samples were mixed with immobilized pH gradient [IPG] buffer (0·5%) supplemented with urea (7 M), thiourea (2 M), 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS, 4%) and dithiothreitol (DTT, 1%) (lysis buffer). The TsESP (30 μg) were loaded onto an IPG strip (pH 3–10 or 6–11) with a cup-loading instrument and focused for a total of 32 kVh. After equilibration, the IPG strip was processed by 15% SDS-PAGE (160×160×1 mm3). The proteins were visualized with CBB staining. Protein spots of interest were excised, in-gel tryptic digested and analysed with a Voyager-DE STR matrix-assisted laser desorption/ionization-time of flight mass spectrometry (MALDI-TOF MS) (PerSeptive Biosystems, Framingham, MA, USA) and Ultraflex MALDI-TOF/TOF MS system (Bruker Daltonics, Bremen, Germany). The solution-phase nitrocellulose method was used for MALDI-TOF MS analysis to the target samples. Mass spectra were obtained in the reflectron/delayed extraction mode, with a 20 kV accelerating voltage and 120 ns delay time. The strongest peptide fragment abundant in the spots obtained from MS scan was isolated and fragmented by collision-induced dissociation, which was subjected to tandem TOF/TOF MS/MS analysis. Monoisotopic peptide masses were selected in a range between 700 and 3500 Da and the proteins were identified by peptide mass fingerprinting (PMF) and tandem MS, using the Matrix Science MASCOT (Matrix Science, Boston, MA, USA) and the National Center for Biotechnology Information (NCBI) database (http://www.ncbi.nlm.nih.gov).
In silico analysis of expressed sequence tags (ESTs) and isolation of TsHLBP genes from a T. solium cDNA library
The EST sequences of T. solium registered in NCBI database were screened with the fragmental EST clones detected during MASCOT analysis using the BLASTn program. Two contigs were constructed using T. solium EST sequences matched to these fragments and the CAP3 Sequence Assembly Program (http://pbil.univ-lyonl.fr/cap3.php). A T. solium cDNA library, which was constructed in ZAP Express cDNA synthesis kit and ZAP Express cDNA Gigapack III Gold Cloning Kit (Stratagene, La Jolla, CA, USA) using mRNA extracted from adult worm (10 pieces of neck and immature proglottids, and 5 pieces of mature and gravid proglottids were pooled and subjected to RNA extraction), was screened by PCR using T3 and T7 promoter primers and the contig-specific primers. The contig-specific primer sets included a same sense primer of 5′-AAAAACCCAGTTCGTGTTATCCGA-3′ since the 2 contigs showed 100% homology within their 5′-region and 2 anti-sense primers of 5′-TCAGTCGTCACCACCATCCTCCAAC-3′ and 5′-CAGTTCTCACTCTTCAACTCCTTCA-3′, which were designed based on the nucleotide sequences of 3′- region of each contig. The thermal cycler profile included pre-heating at 94°C for 4 min, and 35 cycles at 94°C for 50 sec, 58°C for 30 sec, 72 °C for 1 min and a final extension at 72°C for 5 min. The resulting PCR products were ligated into the pGEM-T Easy vector (Promega, Madison, WI, USA) for DNA sequencing. Two full-length cDNA sequences were obtained by overlapping the 5′- and 3′-region sequences. The isolated cDNAs were further confirmed by PCR using primers matched to each terminus of the full length sequences. The coding profile and deduced amino acid sequence were analysed with the open reading frame (ORF) Finder and the BLAST programs.
Sequence and phylogenetic analyses
The PSORT (http://psort.nibb.ac.jp/) and SignalP (http://www.cbs.dtu.dk/services/SignalP/) programs were used for the prediction of signal peptides. The isoelectric points (pI) and theoretical molecular masses were calculated employing the Compute pI/Mw tool (http://us.expasy.org/tools/pi_tool.html). The secondary structure was theoretically predicted employing the PredictionProtein software (http://cubic.bioc.columbia.edu/predictprotein) and ExPASy molecular biology server (http://www.expasy.ch/tools/).
BLASTP searches of the NCBI database and the Hidden Markov Model analyses (InterProScan; http://www.ebi.ac.uk/Inter-ProScan/) using the amino acid sequences of the TsHLBPs resulted in the retrieval of >50 sequences. After the removal of the redundant sequences, 9 representative sequences (>30% sequence identity at the amino acid level) were aligned with the ClustalX and optimized using GeneDoc (http://www.psc.edu/biomed/genedoc/). Phylogenetic analysis was conducted with the Neighbour-joining algorithm equipped in the MEGA software package (ver. 4.0). The sequence divergence was calculated with the Jones-Taylor-Thoronton (JTT) substitution model. Indels between pairs of sequences were regarded as missing data. A phylogenetic tree was displayed with TreeView and the statistical significance of each branching point was evaluated by a bootstrapping analysis of 1000 replicates of the input alignment (SEQBOOT).
Expression and purification of recombinant TsHLBPs (rTsHLBPs)
The mature portion of TsHLBP1 and 2 were amplified from the cDNA library using 3 nucleotide primers. The sense primer contained a SalI site (underlined) was 5′-CAGTCGACCAAAAAACCCAGTTCGTG-3′. Two gene-specific anti-sense primers designed on the basis of the coding sequences of 3′-region of each cDNA sequence containing NotI site were 5′-AAGCGGCCGCTTCAGTCGTCACC-3′ and 5′-AAGCGGCCGCTTCAGTTCTCACTC-3′. The PCR conditions were identical as described above. The amplicons were digested with SalI and NotI and cloned into the pET-28a-c(+) expression vector (Amersham-Pharmacia Biotech, Uppsala, Sweden), which expresses the recombinant protein in fusion with His-tag. The orientation of the inserted DNA was verified by DNA sequencing. The plasmids were transformed into Escherichia coli BL21 (DE3). Upon induction with 0·5 mM isopropyl-β-D-thiogalactoside (IPTG) for 4 h, the cells were harvested and sonicated in native-condition lysis buffer (50 mM NaH2PO4, 300 mM NaCl and 10 mM imidazole). The supernatants were adsorbed to a nickel-nitrilotriacetic acid (Ni-NTA) affinity column (Qiagen, Valencia, CA, USA) and the fusion proteins were eluted with lysis buffer containing 250 mM imidazole. The recombinant proteins (rTsHLBP1 and 2) were dialysed against PBS overnight at 4°C and analysed by 15% SDS-PAGE under reducing conditions.
Generation of antibody and immunoblotting
The antisera against rTsHLBP1 and 2 were raised in 6-week-old, specific pathogen-free female BALB/c mice by subcutaneous injection of 50 μg of purified rTsHLBP1 and 2 mixed with Freund's adjuvants (Sigma-Aldrich, St Louis, MO, USA) 3 times at 2-week intervals. Two weeks after the final boost, 10 μg of each protein was injected in the tail vein. Ten days later, the blood was collected by heart puncture and centrifuged at 3000 g for 5 min. The anti-sera (anti-rTsHLBP1 and 2) were stored at −80°C until used.
For immunoblotting, ESPs and respective proteins extracted from adult and metacestode were separated by 12 or 15% reducing SDS-PAGE or 2-DE employing pH 6–11 IPG strip, after which the proteins were transferred to a nitrocellulose membrane (Schleicher & Schuell, Dassel, Germany). The membrane was incubated with 1:2000 diluted anti-rTsHLBP1 or 2 in PBS containing 0·05% Tween 20 and subsequently with horseradish peroxidase (HRP)-conjugated anti-mouse IgG (Cappel, West Chester, PA, USA) diluted at 1:5000 in the same buffer. Immune reactions were developed by an enhanced chemiluminescence (ECL) detection kit (Amersham Biosciences, Buckinghamshire, UK). All immunoblot images were developed after 1 min exposure.
Assay of hydrophobic ligand binding activity and steady-state kinetics of rTsHLBPs
The crude TsESP (native proteins with HLBP activity), TsM 150 kDa HLBP (positive control) (Lee et al. Reference Lee, Kim, Bae, Chung, Suh, Na, Kim, Kang, Ma and Kong2007), TsM 120 kDa protein (negative control) (Lee et al. Reference Lee, Bae, Jeong, Chung, Je, Kim, Na, Ju, Kim, Ma, Cho and Kong2005) and rTsHLBP1 and 2 were delipidated for 2 h using Sephadex-LH (Sigma-Aldrich) prior to subjection to lipid binding assay. The FA binding property of these proteins was detected using the fluorescent analogues 1,8-ANS, 16-AP, 11-([5-dimethylaminonaphthalene-1-sulfonyl]amino) undecannoic acid (DAUDA) and dansyl-DL-α-aminocaprylic acid (DACA), and the naturally fluorescent cPnA (Molecular Probes, Eugene, OR, USA). The excitation wavelengths used for 1,8-ANS, 16-AP, DAUDA, DACA and cPnA were 370, 360, 350, 350 and 320 nm, respectively. Competitive displacement by saturated and unsaturated FAs was done employing palmitic acid (PA), myristic acid (MA), oleic acid (OA), linoleic acid (LA) and arachidonic acid (AA) (Sigma-Aldrich). All of these chemicals were stored as 10 mM stock solutions in ethanol at −20°C in the dark and were freshly diluted to 0·1 mM with PBS just prior to use. Fixed concentrations (5 or 10 μM) of fluorescent ligands were added to the TsESP and increasing concentrations of rTsHLBPs. The reactions were equilibrated for 2 min and the fluorescence emission spectra were recorded at 25°C (200 μl/well) employing black 96-well Microfluor 1 plates (Thermo Electron Corporation, Marietta, OH, USA) and an Infinite M-200 automated multi-detector (Tecan, GmbH, Austria). Competition assays were conducted by monitoring the change in fluorescence intensity at the peak transmission wavelength measured for either the rTsHLBPs:1,8-ANS and TsESP:1,8-ANS complexes or with other ligands:protein complex in the presence of different concentrations of unlabelled FAs.
The equilibrium dissociation constant (K d) for rTsHLBP1 and 2 bound to 1,8-ANS was estimated by adding increasing doses (0·1–80 μM) of 1,8-ANS to a 200 μl solution of each recombinant protein (0·25 μM). Fluorescence data (Exmax 370 nm and Emmax 460 nm) were normalized to the peak fluorescence intensity and corrected for background fluorescence of the ligand alone at each concentration. Corrected data were analysed using the one-site saturation model and best fit algorithm (y=[Bmax X]/[K d + X]) contained within the SigmaPlot10 software (Systat, San Diego, CA, USA). The K d value for binding of increasing concentrations of cPnA (0·5–20 μM) was also determined at Exmax 320 nm and Emmax 420 nm following the same procedures described above. Scatchard-plot analysis was performed using the same data as for the saturation experiments. All experiments were independently done in triplicate.
Determination of affinity to specific FAs
The relative binding affinity of saturated FAs (MA and PA) and unsaturated FAs (OA, LA and AA) towards rTsHLBPs was determined by displacement assay employing specific FAs (1 μM) binding to the 1,8-ANS:protein complex. The rTsHLBPs (each 8 μg, equivalent to 1 μM) were incubated for 5 min with different FAs (each 1 μM) at 25°C followed by equilibration with 1,8-ANS. Fluorescent data (Exmax 370 nm and Emmax 460 nm) were corrected for background fluorescence and then analysed by non-linear regression using hyperbolic one-site saturation model and best fit algorithm contained in the SigmaPlot10 software (Systat). The data are presented as the mean±S.D. (n=3). A P value <0·05 by paired t-test was considered statistically significant.
RESULTS
Identification of TsHLBPs through protein array of TsESP
We separately incubated 3 adult worms for 30 min, 1 h and 2 h to obtain TsESP. Figure 1a showed SDS-PAGE analysis of the respective ESP, together with whole worm extracts. The protein profiles of the TsESP showed a fairly similar pattern irrespective of their incubation times (lanes a–c); some minor difference was observed, which might result from incubation of different worms. However, the protein profile of the whole worm extracts showed a clearly distinct banding pattern (lane d). We also observed microscopic degeneration of the tegument of the 2-h incubated worms, but no significant change was recognized (data not shown). These results demonstrated that the excretion-secretion behaviour of T. solium was stable and suitable for use as crude TsESP. TsESP revealed a complicated mixture of protein bands varying from 7 to over 100 kDa, among which approximately 10 kDa proteins constituted the major component. The low molecular weight proteins below 20 kDa were selected (shown by dotted box) for further characterization because several previous studies indicated that the cestode HLBPs have relatively small molecular masses of 7–10 kDa (Janssen and Barrett, Reference Janssen and Barrett1995; Barrett et al. Reference Barrett, Saghir, Timanova, Clarke and Brophy1997; Saghir et al. Reference Saghir, Conde, Brophy and Barrett2001; Lee et al. Reference Lee, Kim, Bae, Chung, Suh, Na, Kim, Kang, Ma and Kong2007). We applied IPG strips ranging in pH from 6 to 11 and detected at least five 10 kDa protein spots between pI 9·05 and 10·2 (spot nos. 1–5; upper panel of Fig. 1b). These spots were processed by proteome analysis. PMF of spots 1, 2 and 3 matched with 26 T. solium adult ESTs available in the GenBank database including EL760933 and EL757812 (Table 1). Tandem mass spectra of m/z 1195·61 (Fig. 1c) also matched with over 100 ESTs of the parasite database. Spot 6, which positioned at approximately 5·5 kDa with pI 9·18, was closely related to TsM RS1 (AAD51765). Protein spots 4 and 5 could not be identified. The protein identification by MALDI-TOF/TOF MS analysis is shown in Table 1.
Identification of hydrophobic-ligand-binding proteins of Taenia solium ESP. (a) Three different batches of TsESPs, which were prepared by incubating 3 different adult worms at room temperature for 30 min, 1 h and 2 h (lanes a–c), were analysed concomitantly with the whole worm extracts (lane d) by 12% SDS-PAGE under reducing conditions. Low molecular weight proteins under 20 kDa in the ESP were selected for protein array (dotted box). (b) The crude TsESP were isoelectrically focused using IPG strip (pH 6–11), after which resolved by 15% SDS-PAGE (highlight view between M r 5–20 kDa). The protein spots that migrated to ca. 10 kDa were selected for mass spectrometry (upper panel). The gel was further processed with immunoblotting probed with anti-rTsHLBP2 antibody (lower panel). (c) MALDI-TOF/TOF MS analysis of the peptides generated by in-gel trypsinization of the protein spot 3. Amino acid sequence m/z=1195·646 (KQGLPDEDFF) was determined from mass differences in the y and b-fragment ion series and matched residues by database search. The results of tandem mass spectra of spots 1 and 2 were also exactly the same as that of spot 3.
Identification of putative TsHLBPs from TsESP by MALDI-TOF/TOF MS and database searching
* Protein name was determined after gene cloning.
Isolation and molecular properties of 2 cDNAs putatively coding for TsHLBP1 and 2
Comparison of the T. solium EST sequences retrieved during MASCOT analysis by the CAP3 Sequence Assembly Program revealed 2 distinct consensus contigs (EL760933 and EL757812). The amino acid sequences showed 94% sequence identities with each other and 30–47% with other HLBP members. Based on these contig sequences, we designed gene-specific primers and amplified the 5′- and 3′-regions of the corresponding cDNAs from a T. solium cDNA library. By overlapping these sequences, we obtained 362- and 352-bp long full-length cDNAs (designated TsHLBP1 and TsHLBP2). Both the coding regions were flanked by 11- and 62-bp of 5′- and 3′-untranslated regions. The polyadenylation signals (AATCAA and AATAAA) were recognized at 34- and 39-bp downstream from the stop codons of both genes.
The deduced proteins comprised 87 and 85 amino acids and had predicted molecular masses of 9·96 and 9·89 kDa (pIs 9·41 and 9·73). Ala19 in both polypeptides might constitute a signal peptidase recognition site (arrow; Fig. 2a), which resulted in mature proteins of 7·84 and 7·77 kDa (pI 9·55 and 9·84). The deduced amino acid sequences of these genes were mostly identical (94%), but differed in 6 amino acids near the C-terminal region. Prediction of the secondary structure by the PredictionProtein (http://cubic.bioc.columbia.edu/predictprotein) and ExPASy (http://www.expasy.ch/tools/) revealed that these 2 proteins contained 2 α-helical domains and a strong coiled-coil structure, which might be involved in diffusion-mediated transfer of FAs to phospholipid membranes (Corsico et al. Reference Corsico, Liou and Storch2004). When we determined the genomic organization of these genes employing T. solium genomic DNA, the chromosomal segments of TsHLBP1 and TsHLBP2 appeared to be 510 and 506 bp, respectively. Both genes harboured 4 exons with 3 intervening introns. These genomic sequences were registered in the GenBank database with the Accession numbers JF732995 and JF732996.
Comparison of primary structure and phylogenetic relationship of TsHLBPs with other homologues. (a) Multiple alignment of the amino acid sequences of HLBP-related members. Two predicted α-helices are indicated by black bars below the alignment. The Trp residues are indicated by dotted boxes. The arrow indicates signal peptidase recognition site. Closed circles on the top indicated the hydrophobic residues of TsHLBPs matched with MeHLBP. (b) A phylogenetic tree was constructed with the neighbour-joining algorithm of MEGA software, using the alignment shown in (a). Statistical significance of each branching point was evaluated by a bootstrapping analysis of 1000 samplings of original input. (c) Comparison of TsHLBP genomic structures with other homologues. Open reading frames and untranslated regions are marked with grey and open boxes. Introns are shown by solid lines. The lengths of exon and intron were also presented. The phase of the each intron was indicated in parenthesis. The introns, which were shown to be orthologous among the family members examined, and those conserved in the 7 kDa genes (RS1, Ts14 and Ts18) and TsHLBPs are marked with vertical dotted lines. Gene names were adapted from the GenBank. Accession numbers in parenthesis indicate the nucleotide sequence numbers.
Phylogenetic analysis
We retrieved 9 representative amino acid sequences of the cestode HLBP-related members, which included 1 of each from T. asiatica (TaHC9; a clone identified in T. asiatica, whose information is not currently available), M. expansa (MeHLBP), H. diminuta (HdHLBP) and Echinococcus granulosus (EgAgB), and 5 from T. solium metacestode which included excretion-secretion antigen m13 h (m13 h), excretion-secretion antigen b1 (b1), diagnostic glycoprotein (RS1), 14 kDa diagnostic antigen (Ts14) and 18 kDa glycoprotein (Ts18). The amino acid sequence encoded by the TaHC9 clone showed the highest sequence identity (90%). TsHLBP1 and 2 demonstrated a maximum score of 61·6 and 61·2 (query coverage of 94 and 82%) with e-values of 3e-08 and 5e-08 with MeHLBP and HdHLBP. The incorporation of amino acid sequences of the cloned genes with other cestode HLBPs and T. solium homologues into a phylogenetic tree separated the molecules into 3 distinct clades (Fig. 2b). TsHLBP1 and 2 formed a distinct clade with MeHLBP, HdHLBP and TaHC9 clone. The TsM m13 h and b1 were positioned into the same clade with EgAgB, whereas RS1 formed a different clade with TsM Ts14 and Ts18.
The genomic structures of the TsHLBPs and other related members retrieved from GenBank were compared to determine their overall genomic organization as well as to observe the evolutionary status of TsHLBPs. As shown in Fig. 2c, the position and phase of the first intron among the members, together with the third intron of TsHLBPs and second introns of RS1, Ts14 and Ts18, were tightly conserved, strongly suggesting that these introns were orthologues. Conversely, the second intron in TsHLBPs and m13 h/b1 were independent of each other as their position was different, although they shared the same phase. The amplification of TsHLBPs from the genomic DNA of a single worm demonstrated the paralogous nature of these 2 genes.
Expression of rTsHLBPs
We amplified the mature segments of TsHLBP1 and 2 (each of 204-bp and 198-bp) and cloned them into the pET-28a-c(+) vector. The rTsHLBPs expressed as soluble forms were purified by Ni-NTA affinity chromatography. Both the rTsHLBP1 and 2 migrated to approximately 11 kDa in SDS-PAGE analysis; the molecular masses appeared to be slightly higher than those of native proteins owing to His-tag (Supplementary Fig. 1, Online version only). Immunoblot analysis of the rTsHLBP1 and 2 probed with anti-rTsHLBP2 revealed a strong recognition pattern. However, these mouse antibodies did not differentially recognize the respective protein probably due to their high sequence identity (94%). Native TsHLBPs also showed a strong immune signal against both of the antibodies. Although we could not identify the proteins by proteome analysis, spots 4 and 5 also showed a positive reaction and suggested their possible relationship with TsHLBPs. Conversely, spot 6, which was highly related to TsM RS1 revealed no observable reaction (lower panel of Fig. 1b).
Expression pattern of the TsHLBPs in developmental and different anatomical compartments
We comparatively observed the expression pattern of TsHLBPs in metacestode and adult stages via immunoblotting using mouse antibodies generated against rTsHLBPs. As shown in Fig. 3, TsHLBPs were not expressed during the metacestode stage, but were significantly expressed in the adult stage throughout the entire strobila. The same blot probed with pre-immune mouse serum did not show any detectable reaction (data not shown).
Expression pattern of TsHLBPs in Taenia solium adult and metacestodes. The crude extracts of different body segments (each 20 μg) and ESP/CF (5 μg) were resolved by 15% reducing SDS-PAGE, and then transblotted to a nitrocellulose membrane. The blot was probed with anti-rTsHLBP2 (1:2000 dilution) followed by HRP-conjugated anti-mouse IgG. The reactions were developed by ECL. The recombinant proteins (rTsHLBP1 and 2; 100 ng) were also included. SN, scolex and neck; Im, immature proglottid; Mat, mature proglottid; Gra, gravid proglottid; ESP, excretory-secretory products; SC, scolex; BW, bladder wall; CF, cyst fluid. Molecular masses (M r) are shown by kDa.
Hydrophobic ligand-binding characteristics of rTsHLBP1 and 2
The hydrophobic ligand-binding activity of rTsHLBP1 and 2 were assayed employing FA analogues. The binding of 1,8-ANS to the native protein (TsESP) and rTsHLBPs significantly augmented the fluorescence emission spectra with a blueshift from 530 to 460 nm (Fig. 4a, b and c). TsM 150 kDa HLBP (2 μg) which was employed as the positive control also showed high binding activity. TsM 120 kDa protein (2–5 μg), which was used as the the negative control did not show any binding activity (part of data not shown). The FA binding specificity of TsESP, rTsHLBP1 and 2 was further evaluated by measuring the degree of displacement of 1,8-ANS in the presence of unlabelled lipid ligand, in which OA dose-dependently reduced fluorescence intensity. As shown in Fig. 4d-f, addition of 2 and 5 μM OA to the TsESP and rTsHLBPs resulted in approximately 80 and 100% reduction of fluorescence emission by 1,8-ANS. The rTsHLBP1 and 2 also showed dose-dependent binding activity towards cPnA (Fig. 5, panels b and c) and 16-AP, respectively, while no binding activity was observed against DAUDA and DACA (Supplementary Fig. 2, Online version only, and data not shown). However, the displacement of cPnA and 16-AP was inconsistent; no gradual decrease of fluorescent intensity was observed when increasing concentrations of OA were added in the reaction mixture (Fig. 5, panels b, d and f). The equilibrium dissociation constant (K d) of 1,8-ANS bound to rTsHLBP1 and 2 was 11·4±1·64 and 13·5±1·28 μM, respectively. In addition, K d of cPnA toward rTsHLBP 1 and 2 was 3·2±0·26 and 3·7±0·48 μM (Fig. 6, panels a–d). Scatchard-plot analysis using the data obtained from titration experiments revealed a straight line indicating a single binding site (Fig. 6, insets).
In vitro binding of 1,8-ANS to the TsESP and rTsHLBPs, and competition assay with oleic acid. (a, b and c) Binding of 1,8-ANS (5 μM) to TsESP, rTsHLBP1 and rTsHLBP2 is shown by a dose-dependent increase of fluorescence emission spectra (Exmax 370 and Emmax 460 nm). The reaction mixtures were equilibrated for 2 min and the fluorescence emission spectra were recorded at 25°C using black 96-well plates. The TsM 150 kDa protein was used as the positive control. (d, e and f) The competitive binding of oleic acid (2 and 5 μM) resulted in a reduced fluorescence intensity of 1,8-ANS (5 μM) bound to TsESP, rTsHLBP1 and 2 (each of 2 μg). Competition assay was monitored by the changes in relative fluorescence intensity at the peak transmission wavelength. ANS, 1,8-ANS; OA, oleic acids.
Binding of cPnA to the TsESP and rTsHLBPs, and displacement of cPnA from TsESP/rTsHLBPs:cPnA complex. (a, c and e) Increasing doses of TsESP and rTsHLBPs were incubated with cPnA (5 μM) for 2 min and the fluorescence emission spectra (Exmax 320 and Emmax 420 nm) were recorded at 25°C using black 96-well plates. The TsM 150 kDa protein was used as the positive control. (b, d and f) The competitive binding of oleic acid (2–20 μM) and the respective proteins (doses are indicated in parentheses) did not invoke significant changes of relative fluorescence intensity of cPnA. The displacement effects of oleic acid are shown in the histogram (n=3, mean±S.D.). cPnA, cis-parinaric acid.
Steady-state kinetics of 1,8-ANS (a and b) and cPnA (c and d) bound to rTsHLBPs. Change in fluorescence intensity of reaction mixtures containing rTsHLBP1 and 2 (each 2 μg, equivalent to 0·25 μM) and increasing doses of 1,8-ANS (0·1–80 μM) and cPnA (0·05–20 μM) were monitored. The curve was used to derive the equilibrium dissociation constant (K d) for 1,8-ANS:rTsHLBPs and cPnA:rTsHLBPs interactions. The curved lines represent the theoretical binding with dissociation constant determined by fitting the experimental curves to a single-site binding model.
The binding affinities of different FAs to rTsHLBPs were further determined by a competition assay, in which 1,8-ANS bound to the proteins was displaced by non-fluorescent saturated and unsaturated FAs. As shown in Fig. 7, displacement of 1,8-ANS from the protein binding site was 31·6, 27·5, 52·7 and 87·3% for rTsHLBP1 and 24·6, 19·6, 58·1 and 92·5% for rTsHLBP2 when MA, PA, OA and LA (each 1 μM) were added to the 1,8-ANS: protein complex. These results demonstrated the significantly different effects of saturated (PA and MA) and unsaturated FAs (LA and OA) on the 1,8-ANS:protein complex, except for AA (P<0·01).
Effects of non-fluorescent FAs on binding between 1,8-ANS and rTsHLBPs. The rTsHLBPs (1 μM each) were incubated for 5 min with different FAs (each 1 μM) at 25°C followed by equilibration with 1,8-ANS. The inhibitory effects of FAs were determined by measuring fluorescence emission (Exmax 370 and Emmax 460 nm) (n=3). * and #, P<0·05; ** and ##; P<0·01 compared to control. MA, myristic acid; PA, palmitic acid; OA, oleic acid; LA, linoleic acid; AA, arachidonic acid. The numbers of carbon atom and carbon-carbon double bond are indicated in parenthesis.
DISCUSSION
The HLBPs of cestode parasites are distinguishable from other types of lipid-binding proteins such as nematode polyprotein allergens (NPAs), FA-binding proteins (FABPs) and FA and retinol-binding proteins (FARs) (Kennedy et al. Reference Kennedy, Brass, McCruden, Price, Kelly and Cooper1995, Reference Kennedy, Garside, Goodrick, McDermott, Brass, Price, Kelly, Cooper and Bradley1997; Garofalo et al. Reference Garofalo, Rowlinson, Amambua, Hughes, Kelly, Price, Cooper, Watson, Kennedy and Bradley2003; Solovyova et al. Reference Solovyova, Meenan, McDermott, Garofalo, Bradley, Kennedy and Byron2003; Meenan et al. Reference Meenan, Ball, Bromek, Uhrín, Cooper, Kennedy and Smith2011). The HLBPs may, because of their abundance, be crucial for the uptake and/or transport of hydrophobic ligands and contribute to the maintenance of parasitic homeostasis especially in cestode parasites.
In the present study, we identified excretory-secretory TsHLBPs by proteomic analysis. The 10 kDa proteins with different pIs demonstrated similar PMF and the same amino acid fragments compatible with the HLBPs found in other cestode parasites. We subsequently isolated 2 novel genes through mining of the T. solium EST database and cDNA library screening. The genes showed variable degrees of sequence identity (31–90%), but revealed high similarities of amino acid type among the sequences and contained 2 α-helical domains. These molecules conserved hydrophobic residues, which might be responsible for maintenance of the portal region and hydrophobic ligand-binding capacity (Barrett et al. Reference Barrett, Saghir, Timanova, Clarke and Brophy1997; Kim et al. Reference Kim, Bae, Yang, Hong and Kong2011). The recombinant proteins demonstrated typical hydrophobic ligand binding within the sizeable range of dissociation constants with other hydrophobic molecule transporters (Kennedy et al. Reference Kennedy, Brass, McCruden, Price, Kelly and Cooper1995, Reference Kennedy, Garside, Goodrick, McDermott, Brass, Price, Kelly, Cooper and Bradley1997; Basavaraju et al. Reference Basavaraju, Zhan, Kennedy, Liu, Hawdon and Hotez2003; Pastukhov and Ropson, Reference Pastukhov and Ropson2003; Lee et al. Reference Lee, Kim, Bae, Chung, Suh, Na, Kim, Kang, Ma and Kong2007; Fairfax et al. Reference Fairfax, Vermeirea, Harrisona, Bungirod, Grant, Husain and Cappello2009). All of these findings prompt to us to conclude that the cDNAs cloned are intimately associated with the genuine HLBPs expressed in T. solium adult stage.
Cladistic analysis of TsHLBPs revealed that these proteins comprise the cestode-specific HLBPs together with MeHLBP and HdHLBP, which might have originated from a common ancestor (Bae et al. Reference Bae, Xue, Lee, Kim and Kong2010; Zhang et al. Reference Zhang, Li, Jones, Zhang, Zhao, Blair and McManus2010). The ancestral gene is known to have been highly expanded by gene duplication, after speciation of the respective donor organisms (Zhang et al. Reference Zhang, Li, Jones, Zhang, Zhao, Blair and McManus2010). In the T. solium genome, the multiplied genes appear to have evolved into 2 large lineages, the 7- (RS1, Ts14 and Ts18) and the 10-kDa (CyDA, b1 and m13 h) gene families, each of which has further diverged into multiple sublineages (Bae et al. Reference Bae, Xue, Lee, Kim and Kong2010). The paralogous T. solium genes exhibited genomic structures, which are conserved among members of each family (Kim et al. Reference Kim, Bae, Yang, Hong and Kong2011). TsHLBP1 and 2 were considered to be the members of 10-kDa gene family, as judged by their molecular masses; however, their genomic organizations were found to be more closely related to those of 7-kDa gene members. The 7-kDa and TsHLBP genes possessed 2 conserved orthologous introns, while the 10-kDa genes shared only the first intron with TsHLBPs. The second intron of 10-kDa and TsHLBP genes appeared to be integrated into the respective positions independently. Therefore, it might be concluded that the 10-kDa genes first diverged from the common ancestral gene(s) prior to the separation of 7-kDa and TsHLBP lineage gene(s). This would suggest a very recent duplication, although divergence segregated at the C-termini is unusual. After the diversification, these paralogous entities might acquire the gene-specific spatiotemporal expression patterns. The ancestral gene of cestode FABPs was also found to be duplicated to produce FABPs of the specific donor organisms (Alvite et al. Reference Alvite, Canclini, Corvo and Esteves2008). These collective data suggest that the genomic divergence and resulting diverse biochemical properties among the cestode HLBPs or within the T. solium homologues might reflect their roles in association with the host-specific adaptation of the parasites.
The cestode HLBPs revealed characteristic features according to different parasite species. The TsHLBPs bound to 1,8-ANS, 16-AP and cPnA, but not to DAUDA and DACA. However, those of M. expensa and H. diminuta showed binding affinity to DAUDA and DACA (Janssen and Barrett, Reference Janssen and Barrett1995; Barrett et al. Reference Barrett, Saghir, Timanova, Clarke and Brophy1997; Saghir et al. Reference Saghir, Conde, Brophy and Barrett2001). EgAgB and TsM m13 h/b1 exhibited binding affinity only to 16-AP whereas TsM RS1 displayed strong activity against DAUDA and DACA (Chemale et al. Reference Chemale, Ferreiraa, Barrett, Brophy and Zaha2005; Kim et al. Reference Kim, Bae, Yang, Hong and Kong2011). The displacement assay revealed that 1,8-ANS was consistently displaced by unlabelled FAs, while 16-AP and cPnA were not. These apparent discrepancies are difficult to properly explain at this moment. The diverged primary structures observed among these homologous proteins might be related to the diversifications in their biochemical properties (Veerkamp et al. Reference Veerkamp, van Moerkerk, Prinsen and van Kuppevelt1999). However, 1,8-ANS bound non-specifically to exposed hydrophobic regions of proteins, especially those with apolar binding pockets. Therefore, binding and displacement of 1,8-ANS might result from conformational changes of respective proteins (Ahnström et al. Reference Ahnström, Faber, Axler and Dahlbäck2007). Similarly, cPnA and/or 16-AP might not bind to a specific binding pocket of the protein; they did not actually compete with OA targeted into the specific binding pocket of the rTsHLBPs. Alternatively, 1,8-ANS might bind weakly to the TsHLBPs, whose interaction could be relatively easily displaced in the presence of OA. Further studies are warranted to investigate this anomalous effect.
The rTsHLBPs demonstrated high affinity to long-chain unsaturated FAs and relatively low affinity to saturated FAs. This difference in binding affinities might reflect the characteristic feature of the FA composition of T. solium, which contains an unusually high content (approaching 80%) of unsaturated FAs, including LA and OA (Jacobsen and Fairbarin, Reference Jacobsen and Fairbarin1967). In this study, TsHLBPs did not bind to AA. Since the platyhelminth is able to convert LA to AA (Fusco et al. Reference Fusco, Salafsky and Kevin1985), T. solium might not uptake a host's AA. However, EgAgB (Alvite et al. Reference Alvite, Di Pietro, Santomé, Ehrlich and Esteves2001; Chemale et al. Reference Chemale, Ferreiraa, Barrett, Brophy and Zaha2005) and Ancylostoma ceylanicum FAR protein (Fairfax et al. Reference Fairfax, Vermeirea, Harrisona, Bungirod, Grant, Husain and Cappello2009) effectively bind to AA. The possibility that T. solium might uptake/transport AA by an as-yet-unknown mechanism, i.e., the use of FABP or FAR protein, cannot be excluded and needs further elucidation.
TsHLBPs appeared to be temporally regulated and might be expressed after being stimulated by certain signals from the definitive host. During the development in the human intestine, pepsin and bile fluid are critical for evagination of the scolex, followed by subsequent induction of tegumental re-organization, activation of cell proliferation and differentiation to adult worms (Rabiela et al. Reference Rabiela, Hornelas, Garcia-Allan, Rodríguez-del Rosal and Flisser2000; Espinoza et al. Reference Espinoza, Galindo, Bizarro, Ferreira, Zaha and Galanti2005; Cabrera et al. Reference Cabrera, Espinoza, Kemmerling and Galanti2010). These host factors might be responsible for expression of specific molecules including HLBPs. Furthermore, expression of the 150 kDa TsM HLBP complex is significantly downregulated in the adult stage (Lee et al. Reference Lee, Kim, Bae, Chung, Suh, Na, Kim, Kang, Ma and Kong2007), which would coincide with the expression of different types of HLBPs in the adult worm to convene its specialized requirements. The excretory-secretory nature of the TsHLBPs also suggests that these proteins might be expressed in the tegumental surface or subtegumental area, and raises an interesting issue regarding their biological roles. T. solium absorbs nutrients through the tegument from the human intestine, where micelle/FAs are enriched. The secreted TsHLBP(s) might bind to free FAs/micelles and are involved in the uptake of these molecules. In addition, TsHLBP(s) might participate in sequestration of toxic hydrophobic molecules and act as a buffer system or component of the detoxification mechanism since T. solium is directly exposed to host-derived hydrophobic toxic molecules within the intestine. The HLBPs or HLBP-related molecules secreted by other cestode and nematode parasites were also shown to mediate the detoxification and transportation (Janssen and Barrett, Reference Janssen and Barrett1995; Barrett et al. Reference Barrett, Saghir, Timanova, Clarke and Brophy1997; Kennedy, Reference Kennedy2000; Chemale et al. Reference Chemale, Ferreiraa, Barrett, Brophy and Zaha2005).
In this study, we could not differentially identify TsHLBP1 and 2 in TsESP, since these proteins revealed the same PMF spectrum, which were matched with middle portions of genes that showed 100% sequence identity to each other. These 2 TsHLBPs differed by 6 amino acids near the C-terminus. In accordance with this high sequence identity, antibodies generated against rTsHLBP1 and 2 were cross-reactive which further hampered efforts to immunologically differentiate these closely related molecules. However, isolation of distinct genes occupying different genomic loci and immunoblotting data strongly suggested that multiple genes paralogous to HLBP1 or 2 might exist within a single T. solium genome, as is shown by Echinococcus proteins (Zhang et al. Reference Zhang, Li, Jones, Zhang, Zhao, Blair and McManus2010). The last divergent exon of TsHLBP1 or TsHLBP2 would be obtained by exon shuffling and homologous recombination (Patthy, Reference Patthy1999).
In conclusion, we characterized 2 novel excretory-secretory TsHLBPs. The predicted structural similarities, close clustering in the phylogenetic tree and presence of an orthologous intron, as well as high binding affinity to FAs and their analogues, indicate that the TsHLBPs are members of the cestode HLBP family. Adult stage-specific expression and wide distribution throughout the entire strobila suggest that the TsHLBPs might be involved in adaptation of the adult worm into the specific host environment and in lipid shuttling from the host intestine. Further understanding in regard to the pathophysiological role of TsHLBPs would be warranted to unveil the molecular mechanism of the hydrophobic ligand-binding system inherent to the host-parasite interaction.
ACKNOWLEDGEMENTS
M. Rahman and E.-G. Lee contributed equally to the work.
FINANCIAL SUPPORT
This work was supported by the grant from the Samsung Biomedical Research Center (#SBRI BB1101-1).
