IrML– a gene encoding a new member of the ML protein family from the hard tick, Ixodes ricinus

Authors

  • Jana Horáčková,

    1. University of South Bohemia, Faculty of Science, České Budějovice, Czech Republic
    Search for more papers by this author
  • Nataliia Rudenko,

    1. University of South Bohemia, Faculty of Science, České Budějovice, Czech Republic
    2. Biology Centre, v.v.i., Institute of Parasitology AS CR, České Budějovice, Czech Republic
    Search for more papers by this author
  • Maryna Golovchenko,

    1. University of South Bohemia, Faculty of Science, České Budějovice, Czech Republic
    2. Biology Centre, v.v.i., Institute of Parasitology AS CR, České Budějovice, Czech Republic
    Search for more papers by this author
  • Sabína Havlíková,

    1. Slovak Academy of Sciences, Institute of Virology, Bratislava, Slovakia
    Search for more papers by this author
  • Libor Grubhoffer

    1. University of South Bohemia, Faculty of Science, České Budějovice, Czech Republic
    2. Biology Centre, v.v.i., Institute of Parasitology AS CR, České Budějovice, Czech Republic
    Search for more papers by this author

ABSTRACT:

Blood intake causes significant changes in ticks, triggering vital physiological processes including differential gene expression. A gene encoding Ixodes ricinus ML-domain containing protein (IrML) is one of the set of the genes that are strongly induced by blood meals. IrML belongs to the ML protein family that commonly occurs in diverse organisms and is involved in lipid binding and transport, pathogen recognition or in immune response. An IrML gene was amplified from cDNA of engorged I. ricinus females using the gene-specific primers designed on a basis of partial sequences of related genes for ML domain protein. IrML was shown to be expressed mainly in the gut, but also in salivary glands and hemolymph of all tick developmental stages. Using in situ hybridization, IrML transcripts were detected in type II and III salivary glands acini. Analysis of the predicted structure of I. ricinus ML-domain containing protein and its localization in the tick body could suggest that IrML is a secreted protein and is possibly involved in tick innate immunity.

INTRODUCTION

The European hard tick Ixodes ricinus is a hematophagous parasite of a variety of vertebrates. Blood-feeding activates many genes in different tick tissues, triggering the expression and synthesis of constitutive and inducible proteins (Mans and Neitz 2004). Expression of genes encoding proteins that affect the internal tick environment, as well as genes encoding proteins secreted by the tick directly into the site of a tick bite on the host and modulating host immune response, are induced simultaneously. A gene encoding ML domain-containing protein (AY323234) is assigned to the group of genes involved in pathogen recognition and has been detected in the castor bean tick by means of subtractive hybridization, proving it to be strongly induced by blood-feeding (Rudenko et al. 2005).

Members of the ML-domain protein family were found in several groups of organisms, such as fungi, plants, and animals. More than 150 proteins have been included in this protein family based on the presence of a single ML (MD-2-related lipid-recognition) domain that contains an N-terminal signal peptide and four cysteine residues at conservative positions forming disulfide bridges (Inohara and Nuñez 2002), important for protein functional activities (Ichikawa et al. 1998, Wright et al. 2000, Friedland et al. 2003). ML-domain containing proteins structurally consist of multiple β-strands which create β-sheets. It was confirmed that ML-domain containing proteins are secreted or luminal proteins that are able to bind lipids and lipid molecules due to the presence of the ML-domain (Kirchhoff et al. 1996, Mullen et al. 2003).

According to the degree of sequence similarity, ML proteins can be divided into four groups (Inohara and Nuñez 2002). Human MD-1, MD-2 (Myeloid Differentiation-1, 2) proteins, and their orthologs are the main representatives of group I. MD-2 proteins bind bacterial lipopolysaccharide (LPS) and interact with the cell receptor TLR4 (Toll-like receptor 4), necessary for an innate immune response to LPS (Gruber et al. 2004, Hyakushima et al. 2004). Toll-like receptors are involved in pathogen recognition through the recognition of pathogen-associated molecules (McGuinness et al. 2003, Akira 2004). MD-1 proteins interact with a similar receptor RP105 (RadioProtective) on the surface of antigen-presenting cells (Miura et al. 1998, Divanovic et al. 2005).

Group II contains several major mite allergen proteins: Drosophila melanogaster proteins, Caenorhabditis elegans proteins, and Niemann-Pick protein type C2 (Npc2). Npc2 (or HE1) and its homologs bind cholesterol and similar sterols (Storch and Xu 2009). It is not clear whether Npc2 directly participates in cholesterol metabolism, but it is involved in cholesterol transport within the cell (Okamura et al. 1999, Friedland et al. 2003, Infante et al. 2008). Mite proteins often act as allergens and are able to evoke immune reactions. Several groups of allergens interact with IgE antibodies or work as cysteine or other substrate-specific proteases (Thomas et al. 2002, Johannessen at al. 2005).

Group III is composed of plant and fungal proteins, including several Arabidopsis thaliana proteins, phosphatidylglycerol/phosphatidylinositol transfer protein (PG/PI-TP) of Aspergillus oryzae, and its orthologs and paralogs (Record et al. 1999, Inohara and Nuñez 2002).

Group IV originally consisted of the human GM2 activator protein (GM2-AP) and its orthologs. GM2-AP is able to bind and transport many lipid molecules (glycolipids, gangliosides, and at least one phosphoacylglycerol) (Mahuran 1998, Inohara and Nuñez 2002). A newly-identified protein group related to GM2-AP in Drosophila melanogaster, CheB proteins, is involved in pheromone perception (Starostina et al. 2009).

ML-domain containing proteins were identified in the soft ticks Ornithodoros coriaceus and O. parkeri, and in the hard ticks Rhipicephalus microplus, R. appendiculatus, Dermacentor variabilis, Ixodes scapularis, and I. ricinus (Francischetti et al. 2009). Three genes encoding ML-domain containing proteins that are mainly expressed in the tick gut were identified in the European tick I. ricinus, including the gene for the ML-domain containing protein (AY323234), alerg1, the gene for allergen-like protein (AJ547805), and the gene encoding Der-p2-like allergen (EF116564) (Rudenko et al. 2005, Horáčková et al. 2010). The function of the gut-expressed ML proteins in ticks is unknown, but it is possible that they might be involved in the tick immune response to pathogens and mediate defensive reactions, or in lipid metabolism.

Here we present the description and analysis of a novel gene encoding ML-domain containing protein (IrML) in I. ricinus (EU034645). Tissue and stage specific pattern of IrML gene expression was determined, protein structural analysis and similarity search were conducted, and the localization of IrML transcripts in the salivary glands of females on different feeding states was revealed by in situ hybridization.

MATERIALS AND METHODS

Ticks

I. ricinus ticks were provided by the Biological Centre, Institute of Parasitology, Academy of Sciences of the Czech Republic. Uninfected ticks were reared for several generations in the animal facilities of the institute and fed on guinea pigs that were raised to be free of infection under the strict hygiene regulations of the Central Commission for the Protection of Animals (§21, section 3e, law 246/1992). Blood-fed larvae, nymphs, and females were obtained by attaching a container (cell) with ticks to a shaved area on the back of an adult uninfected guinea pig and left on the animal until the specified engorgement stage. After the separation of the fed ticks from the guinea pig, RNA was isolated from them using TRI Reagent (Sigma, U.S.A.). The fully engorged females were kept at 4° C for another three days for blood digestion prior to RNA isolation.

First strand cDNA synthesis and polymerase chain reaction (PCR)

First strand cDNA was synthesized using RevertAid™ H Minus First Strand cDNA Synthesis Kit (MBI Fermentas, Lithuania) according to the protocol provided by the manufacturer. Primers ML1F and ML1R were designed according to the known partial nucleotide sequence of the gene encoding I. ricinus ML-domain containing protein (AY323234) used for amplification (ML1F 5′-ATGGCCGGTTCTATGGTATTC-3′/ML1R 5-AAATTCTCCGGCTTCAGTTGTC-3′, Generi Biotech, Czech Republic).

RT-PCR analysis of stage specific and tissue specific patterns and time expression of the gene encoding I. ricinus ML domain-containing protein

RT-PCR was performed using the Enhanced Avian HS RT-PCR kit (Sigma, U.S.A.). An equal amount of total RNA isolated from unfed and engorged larvae, nymphs and females of I. ricinus and different tissues (salivary glands, gut, ovaries, Malpighian tubules, and hemolymph) of partially engorged females (five days) was used as a template in RT-PCR in order to reveal the IrML gene expression in different developmental and feeding stages. To determine time expression of the gene, equal amounts of total RNA from unfed, 1, 3, 5, 7 days-fed, and totally engorged adult females, were used as templates in RT-PCR. Also, an equal amount of RNA from salivary glands and gut from females unfed, 1, 2, 4 and 6 days after feeding, were used as templates in other PCRs. The primers, ML2F 5′-CACCATGGCCGGTTCTATGGTA-3′, and ML2R 5′- CTATTGCTTCAGCTCTACCG-3′ were used for the amplification with an annealing temperature of 55° C. A constitutively-expressed actin gene (AY333957) was chosen as a control. The control amplification was performed at the annealing temperature of 55° C with the primers Act-F1 5’-CGTCTGGATCGGCGGCTCTAT-3’ and Act-R1 5’-ACGCGCACTCTTTTCCACAATCTC-3’. All RNAs used were first checked for DNA contamination by adding them directly into the PCR reaction with the specific primers.

Bioinformatics/sequence analysis

Sequences were compared against the GenBank non-redundant (NR) protein database using the BlastX program from the executable package at ftp://ftp.ncbi.nlm.nih.gov/blast/executables (Altschul et al. 1997) and searched against the Conserved Domains Database (CDD) (ftp://ftp.ncbi.nlm.nih.gov/pub/mmdb/cdd), which includes all Pfam (Bateman et al. 2000) and Smart (Schultz et al. 1998) protein domains, as well as against the VectorBase (Lawson et al. 2009) (http://www.vectorbase.org) and the Gene Index Project (Quackenbush et al. 2001) (http://compbio.dfci.harvard.edu/tgi/cgi-bin/tgi/Blast/index.cgi). Protein secondary structure prediction was performed by the PSIpred program (http://bioinf.cs.ucl.ac.uk/psipred) (McGuffin et al. 2000). Putative signal peptide, possible sites of O-glycosylation, N-glycosylation, and phosphorylation were determined with the prediction servers Signal P 3.0 (http://www.cbs.dtu.dk/services/SignalP) (Bendtsen et al. 2004), NetOGlyc 3.1 (http://www.cbs.dtu.dk/services/NetOGlyc) (Julenius et al. 2005), NetNGlyc 1.0 (http://www.cbs.dtu.dk/services/NetNGlyc), and NetPhos 2.0 (http://www.cbs.dtu.dk/services/NetPhos) (Blom et al. 1999), respectively.

Tertiary structure prediction

For prediction of the 3D structure of the IrML protein, CPHmodel 2.0 Server was used (http://www.cbs.dtu.dk/services/CPHmodels) to find the homologues for the I. ricinus molecule. A pdb file of the molecule (according to the similar sequences of structures from the protein databank) was downloaded and used in Polyview-3D for protein visualization (http://polyview.cchmc.org/polyview3d.html) (Porollo and Meller 2007).

Preparation of in situ hybridization (ISH) probes

A specific antisense DNA probe was prepared using the PCR digoxigenin (Dig) Probe Synthesis Kit (Roche Applied Science, Germany). The cDNA amplified using ML2F and ML2R primers served as template for probe synthesis. The Dig labeled probe was produced by asymmetric PCR from 300 ng of the obtained cDNA and gene-specific reverse primer (ML2R) in a final reaction volume of 25 μl. The initial denaturation step (95° C for 2 min) was followed by ten cycles of 95° C for 30 s, 52° C for 30 s, and 72° C for 1.5 min, and subsequently by 11 cycles of 94° C for 30 s, 52° C for 30 s, and 72° C for 2.5 min. The Dig-labeled probe was gel purified and stored at –20° C. The specific sense probe was synthesized using the same protocol but with the ML2F primer.

In situ hybridization (ISH)

Dissected tick organs were fixed in 4% paraformaldehyde at 4° C overnight and subjected to ISH as previously described by Kim et al. (2006). The color reaction was observed under a binocular microscope and stopped by repeated washes in PBS (phosphate buffered saline) with 0.2% Tween 20, pH 7.2 (PBST) followed by a wash in PBS:glycerol (1:1) for 20 min.

Stained tissues were mounted in 100% glycerol on glass slides and observed under a fluorescence microscope Olympus BX51 with Nomarski DIC optics and attached Olympus DP70 digital camera (Olympus, Japan). Tissues from five to ten ticks were studied for each feeding stage and used for hybridization. Two negative controls were performed. The hybridization step was omitted in one of the controls, and the specific sense probe instead of the antisense probe was used in the protocol. Images were processed using with the program Paint.NET v3.5.1.

RESULTS

We identified a fourth I. ricinus gene from the ML-domain family encoding ML- (MD-2-related lipid recognition) domain containing protein (IrML). The isolated cDNA sequence of IrML is 557 bp long and contains an ORF of 157 amino acids. The preprotein carries a 21 amino acids-long N-terminal signal peptide. Six conserved cysteine residues, characteristic for group II of ML proteins and forming three disulfide bridges in order 1–6, 2–3, 4–5 in preprotein were detected at positions 33, 49, 54, 101, 108, 148 (Figure 1). The predicted molecular weight of the mature protein is 14.83 kDa and has an isoelectric point of 7.98. IrML has a predicted helix in the region of the signal peptide that also occurs in other unmatured ML proteins and β-structures are typical for the remaining amino acid chain (Inohara and Nuñez 2002).

Figure 1.

Nucleotide and protein sequence of IrML. A conserved motif is framed and marked in bold. Conserved cysteine residues are marked by asterisks, signal peptide is indicated in bold. Residues of putative lipid binding cavity (30, 41, 44, 46, 56, 62, 64, 66, 79, 81, 83, 100, 110, 118, 134, 136, 138, 140, 146, 149, 151, 153) are in italics. Possible sites of O-glycosylation and phosphorylation are marked by shading and framed, respectively.

IrML contains an Npc2-like domain and belongs to the ML protein family (Marchler-Bauer and Bryant 2004). Alignment of the IrML protein sequence with several other members from different groups of the ML protein family revealed that the highest level of similarity was to those from group II (Figure 2). BLASTP, GenBank, and Gene Index Project analyses of the amino acid sequence revealed its relatedness to several proteins similar to Niemann-Pick disease type C2 protein (Npc2) from several tick species (I. scapularis, D. variabilis, and R. microplus) (Table 1).

Figure 2.

Alignment of IrML and several other members from group II of ML protein family. IrML –I. ricinus ML-domain containing protein (ABU43149), MmNpc2 –Mus musculus Npc2 protein (AAH03471), HsNpc2 –Homo sapiens Npc2 protein (AAH02532), Derp2 –Dermatophagoides pteronyssinus Der p 2 allergen (ABY53034), Derf2 –D. farinae Der f 2 allergen (CAI05849), Tyrp2 –Tyrophagus putrescentiae Tyr p 2 allergen (CAI05849). Conserved cysteine residues are marked by asterisks, signal peptides are in italics. Black boxes with white letters show identical residues, and gray boxes with black letters show chemically similar residues.

Table 1.  Tick proteins with the highest sequence identity to IrML (ABU43149).
OrganismProteinNumber% of identity% of similarity
I. ricinusML-domain containing proteinAAP840989798
I. scapularisputative secreted proteinEEC162209598
I. scapularisputative ML-domain containing proteinEEC073239397
D. variabilisputative ML-domain containing proteinACF355466274
D. variabilisputative ML-domain containing proteinACF355386274
R. microplussimilar to ML-domain containing proteinTC159875982

Analysis of the tertiary structure of IrML showed the highest structural homology to the Npc2 protein (Niemann-Pick disease type 2 protein) from Bos taurus (PDB 2hka), with 24.8% similarity. It was predicted that IrML would have an immunoglobulin fold, a β-sandwich characterized by seven to nine antiparallel β-strands forming two β-sheets (Figure 3).

Figure 3.

Tertiary structure of bovine Npc2 protein – 2hka (A – three chains of Npc2 protein with ligands: a – N-acetyl-D-glucosamine, b – acetic acid, c – cholesteryl sulfate, d – sulfuric acid, e – glycerol) and predicted tertiary structure of ML-domain containing protein of I. ricinus (B) with conserved cysteine residues marked.

The RT-PCR profiling of IrML mRNA levels revealed that its expression is strongly induced by blood meals in all developmental stages, including larvae, nymphs, and adult females. This was confirmed by the time expression using total RNA from unfed and differently engorged females. The expression of IrML, analyzed in the whole tick body, starts on the first day of feeding and remains constant (data not shown) after the third day of feeding. Tissue specific patterns confirm that the IrML gene was expressed in the gut, salivary glands, and hemolymph. An expression level of the actin gene, used as a control, was constant through the whole experiment (Figure 4). RT-PCR analysis of IrML gene expression in tissues with the highest level of expression showed interesting results. While IrML expression in the gut acts in the same way as the revealed analysis of the expression in the whole body, its expression in salivary glands starts after the third day of feeding and then rapidly increases (Figure 5).

Figure 4.

Stage and tissue specific IrML gene expression analysis by RT-PCR. IrML – ML-domain containing protein, ACT – actin, NF – unfed ticks, BF – engorged ticks (8 days), tissues of partially engorged females (5 days): OV – ovaries, G – gut, SG – salivary glands, MT – Malphigian tubes, HL – hemolymph.

Figure 5.

Tissue specific IrML gene expression analysis by RT-PCR in the gut and salivary glands during tick feeding. 0, 1, 2, 4, 6 days of blood feeding, IrML – ML-domain containing protein, ACT – actin.

Localization of IrML transcripts in the salivary glands of females on different feeding states was determined by in situ hybridization. The strongest positive staining in salivary glands was observed in five-day fed females. No staining was visible in the salivary glands of unfed females, a nonspecific signal was detected in the salivary ducts of unfed ticks, but the same observation was also made in unstained salivary glands (negative controls) of unfed females (Figure 6). Expression of IrML mRNA in female ticks was observed in granular cells of types II and III acini of the salivary glands. A slight staining of type II acini appeared in the salivary glands of three-day fed ticks, while rather expressive staining of type III acini was observed in five-day fed females. The mRNA of IrML was not detected in either agranular cells of type I acini nor in ducts of the salivary glands. Although RT-PCR profiling of IrML expression showed that while the gene is strongly induced in the midgut of fed ticks, detection of transcripts by ISH was not successful (data not shown).

Figure 6.

Localization of IrML gene expression in tick salivary glands by in situ hybridization. A – gene expression in salivary glands of females on different feeding states, B – negative control – omitted hybridization, C – negative control – hybridization with specific sense probe. sd – main salivary duct, I – type I acinus, II – type II acinus, III – type III acinus. Scale bars: 200 μm.

DISCUSSION

A gene encoding a protein from the ML-domain protein family was identified and isolated from the castor bean tick I. ricinus. Members of this protein family have already been found in several species of hard and soft ticks (Francischetti et al. 2009). The described IrML gene (EU034645) is similar to another I. ricinus gene encoding an ML-domain containing protein with a sequence of 465 bp that could be an isoform of the studied gene (AY323234). The gene for the ML-domain containing protein was identified as a differentially expressed gene in the subtractive cDNA library of fully engorged and Borrelia burgdorferi-infected I. ricinus (Rudenko et al. 2005). Here, it was assigned to a group of genes connected to pathogen recognition and defense. While the first described ML-domain containing protein was strongly induced only in the tick midgut, the novel gene encoding IrML is also additionally expressed in salivary glands and hemolymph. It has already been shown in the example of other I. ricinus proteins that the site of protein expression might determine or affect the protein function (Rudenko et al. 2007, Chrudimská et al. 2010). Analysis of midgut transcripts of D. variabilis revealed that transcripts of the genes encoding ML-domain containing proteins are the third most abundant (specifically, allergen-like proteins) in this tick species (Anderson et al. 2008). Transcripts of ML protein genes were also detected in the midgut of R. microplus (Kongsuwan et al. 2010) and in salivary glands of the soft tick O. parkeri (Francischetti et al. 2008) or in the body of I. scapularis (de la Fuente et al. 2008). Thus, ML-domain containing proteins are expressed in different tick tissues that imply their potential participation in variable processes.

As mentioned, IrML mRNA was found in three different tissues (gut, salivary glands, and hemolymph) with the maximum level of expression (analyzed on the whole body mRNA) determined on the third day of blood feeding. Interestingly, the strongest staining was detected by in situ hybridization in salivary glands from five-day blood-fed females. This contrast might indicate that the majority of the gene expression occurs in the gut during the first days of tick feeding and then moves to the salivary glands. This observation was then confirmed by the analyses of gene expression in the gut and salivary glands. The expression in the gut parallels the expression in the whole body, whereas expression in the salivary glands starts later. Results of ISH with a DIG-labeled probe on salivary glands showed that the transcript of the IrML gene is expressed mainly in granular type II and III of acini cells, which are responsible for secretion, especially cells of type III. This fact and the presence of an N-terminal signal peptide in the protein sequence might confirm the possibility that IrML is produced as a secreted protein. Although the ISH technique provides sensitivity and specificity of traditional nucleic acid hybridization with histological localization with minimal tissue disruption, the number of studies that use ISH for localization of expressed genes in ticks is rather limited. Nevertheless, this technique was successfully used in the case of localization of differentially expressed secreted cement protein from the tick R. appendiculatus (Havlíková et al. 2009). In this report, the transcripts of the gene were identified by real-time PCR in salivary glands and midguts of the tick, although their detection by ISH was successful only in salivary glands. The authors conclude that the obtained positive staining is not the correct signal due to the fact that the same signal was also observed in the controls. This phenomenon, as in our case, may be caused due to endogenous alkaline phosphatases and not the specific detection of transcripts.

ML proteins with defined functions are involved in transport, lipid metabolism, or in innate immunity processes. They are able to recognize lipid molecules and interact with them. This ability allows ML proteins (probably only several of them) to be involved in binding of PAMPs (pathogen-associated molecular patterns) that often include lipids (besides carbohydrates, proteins, or nucleic acid motifs). PAMPs are conserved molecules usually on the surface of the microorganisms that are necessary for their survival and are used for their recognition by components of innate immunity (that carry PRRs – pattern recognition receptors – for PAMPs) (McGuinness et al. 2003, Akira 2004). The predicted presence of a lipid-binding site in IrML indicates its possible role in processes that could include the elimination of microorganisms from ingested blood in the gut, or transport of lipids from blood through hemolymph to other organs (salivary glands). However, this speculation needs to be confirmed.

ML proteins in arthropods have been found in classes of Arachnida (mites, ticks), Insecta (Drosophila melanogaster, Bombyx mori, Apis mellifera, Anopheles gambiae), and Malacostraca (Litopenaeus vannamei) (Schultz et al. 1998, Letunic et al. 2008). Homologs of the Npc2 protein have been found previously in insects. In addition, ecdysteroid-regulated proteins have been identified in several insect species as well. In mites, proteins that belong to the ML protein family are allergens, specifically group 2 allergens. These molecules are responsible for causing severe allergies in humans. On the other hand, the Der f 2 allergen from Dermatophagoides farinae was shown to bind bacterial surfaces and is possibly involved in defense responses (Ichikawa et al. 1998). Allergens from ML protein family are homologues of HE1 epididymal protein (Thomas et al. 2002). However, their exact function has not yet been defined.

Occurrence of IrML gene expression in different tick tissues during and after a blood meal indicates the involvement of this protein in physiological processes in various parts of the tick body. The knowledge of the functional activities of other ML proteins might be used to predict the possible involvement of I. ricinus IrML in transport processes, pathogen recognition, or in immune responses.

Acknowledgments

We are grateful to Jan Erhart for his help during the work with ticks, to Dr. Boris Klempa (Institute of Virology, Slovak Academy of Sciences, Bratislava, Slovakia) for the utilization of facilities and teaching the technique of in situ hybridization in his laboratory, to Dr. Iva Dyková for her help with microscope work, and to Jason Dean for his help in linguistic editing of the manuscript. This work was financially supported by the Grant Agency of the Czech Republic (grant 524/06/1479), Ministry of Education of the Czech Republic (MSM 6007665801, LC06009) and grant Z60220518 (research project of the Institute of Parasitology AS CR).

Ancillary