Molecular characterization and expression analysis of annexin B3 and B38 as secretory proteins in Echinococcus granulosus

Background Cystic echinococcosis is a parasitic zoonotic disease, which poses a threat to public health and animal husbandry, and causes significant economic losses. Annexins are a family of phospholipid-binding proteins with calcium ion-binding activity, which have many functions. Methods Two annexin protein family genes [Echinococcus granulosus annexin B3 (EgAnxB3) and EgAnxB38] were cloned and molecularly characterized using bioinformatic analysis. The immunoreactivity of recombinant EgAnxB3 (rEgAnxB3) and rEgAnxB38 was investigated using western blotting. The distribution of EgAnxB3 and EgAnxB38 in protoscoleces (PSCs), the germinal layer, 18-day strobilated worms and 45-day adult worms was analyzed by immunofluorescence localization, and their secretory characteristics were analyzed preliminarily; in addition, quantitative real-time reverse transcription polymerase chain reaction was used to analyze their transcript levels in PSCs and 28-day strobilated worms stages. The phospholipid-binding activities of rEgAnxB3 and rEgAnxB38 were also analyzed. Results EgAnxB3 and EgAnxB38 are conserved and contain calcium-binding sites. Both rEgAnxB3 and rEgAnxB38 could be specifically recognized by the serum samples from E. granulosus-infected sheep, indicating that they had strong immunoreactivity. EgAnxB3 and EgAnxB38 were distributed in all stages of E. granulosus, and their transcript levels were high in the 28-day strobilated worms. They were found in liver tissues near the cysts. In addition, rEgAnxB3 has Ca2+-dependent phospholipid-binding properties. Conclusions EgAnxB3 and EgAnxB38 contain calcium-binding sites, and rEgAnxB3 has Ca2+-dependent phospholipid-binding properties. EgAnxB3 and EgAnxB38 were transcribed in PSCs and 28-day strobilated worms. They were expressed in all stages of E. granulosus, and distributed in the liver tissues near the hydatid cyst, indicating that they are secreted proteins that play a crucial role in the development of E. granulosus.

industry each year [3]. It has been listed as one of the neglected tropical diseases by the World Health Organization [4,5]. E. granulosus pathogenic research is of great significance, and studying the mechanism of the interaction between E. granulosus and its host is also necessary for the prevention and control of CE.
Proteomics analysis revealed that annexins were present in E. granulosus hydatid fluid, protoscoleces (PSCs) culture fluid and exosomes [6][7][8]. Annexins are a family of phospholipid-binding proteins with calcium ionbinding activity, which function in calcium signaling, and are widely distributed in eukaryotic cells [9]. According to differences in gene location and protein structure, annexins are divided into five categories [10]. Annexins have multiple functions, such as cell anti-inflammatory activity, membrane repair and membrane transport, and probably participate in cell proliferation, differentiation, and apoptosis [11][12][13][14][15].
Parasite annexins have been found to regulate the host immune response and maintain the structural integrity of the cell [16,17]. Parasite annexins have also been considered as potential targets for the development of drug and vaccine candidates [16,18]. Leishmania promastigotes could combine annexin V to ensure normal function in the absence of phosphatidylserine [19]. Annexin 2 of Schistosoma mansoni is involved in epidermal development [20]. Recombinant annexin B30 of S. mansoni gave no significant protection against the parasite, which suggested it may not be suitable as a vaccine candidate [21]. Annexin B1 of Taenia solium can downregulate the immune response of the host [22,23]. In humans, E. granulosus annexins have only been found in the fluid of pulmonary hydatids, not in the fluid of vertebral or paravertebral hydatids, which suggests that they play an important role in the calcification process of cysts [7]. Up until now, only E. granulosus annexin B33 (EgAnxB33) has been studied [24]. In the present study, cloning, expression, bioinformatic analysis, western blotting, relative fluorescence quantitative polymerase chain reaction (PCR), immunofluorescence localization, and phospholipid-binding bioactivity analyses of EgAnxB3 and EgAnxB38 were performed to provide basic information and new research directions for the interaction between E. granulosus and its hosts.

Animals
Four female New Zealand rabbits (9 weeks old) were purchased from Dashuo Experimental Animal (Chengdu, China; experimental animal production license no. SYXK2019-189). Four male beagles (6 months old) were provided by the Dujiangyan Beagle Breeding Center of the Sichuan Institute of Musk Deer Breeding.
Albendazole and levamisole were used for deworming in the first 3 months.

Parasites
Cysts of E. granulosus were obtained from naturally infected sheep in Sichuan Province, China. PSCs and the germinal layer were separated aseptically as described previously [25,26]. The 28-day strobilated worms were acquired by artificially infecting beagles. Each beagle was given 50,000 PSCs orally and then euthanized after 28 days. The 18-day strobilated worms and 45-day adult worms were provided by the Department of Parasitology of Sichuan Agricultural University.

Sera
Sera against E. granulosus were isolated from naturally infected sheep. Negative sera were collected from cestode-free sheep. Corresponding sera were obtained in Sichuan Province, China and infection was determined by autopsy.
The preparation process of the polyclonal antibodies of recombinant EgAnxB3 (rEgAnxB3) and rEgAnxB38 was the same as previously reported [26]. Briefly, rabbit sera were collected as the negative control before immunization. Then each rabbit was immunized with 200 μg recombinant protein emulsified with Freund's complete adjuvant (Sigma-Aldrich, St. Louis, MO), followed by three boosters using Freund's incomplete adjuvant. Two weeks after the final immunization, antisera were isolated and immunoglobulin G (IgG) was extracted.

Cloning of EgAnxB3 and EgAnxB38
An RNA extraction kit (Tiangen, Beijing, China) was used to extract total RNA from PSCs, and the Reverse Transcription System Kit (TaKaRa, Dalian, China) was used to synthesize cDNA. Specific primers were designed at GenBank using the sequences of EgAnxB3 and EgAnxB38 (accession numbers XM_024495323.1 and XM_024494485.1, respectively). The sequence of EgAnxB3 was amplified using the primers 5′-CGGA TCC ATG GCG ACT GTC AAG CCT TGCTG-3′ (BamHI) and 5′-CCG GAA TTC TTA CTC CAG TAT GGC AAG C-3′ (EcoRI). The sequence encoding EgAnxB38 was amplified using primers 5′-CGGA TCC ATG GCT GGC TAT CCT CCA CC-3′ (BamHI) and 5′-CGGAA TTC TCA GGG TCC AAC CAA AGC CAC-3′ (EcoRI). Target fragments were amplified and cloned. Single colonies were selected for PCR identification, and the plasmid from the bacterial solution that tested positive by PCR was sequenced.

Western blotting
Western blotting was performed as described previously [28]. Briefly, the crude protein extracts of PSCs and the purified recombinant protein were transferred onto a polyvinylidene difluoride membrane. The membrane was incubated with sera [1:200 volume/volume (v/v) dilution] from CE-infected sheep or polyclonal antibodies (1:200 v/v dilution) for 12 h at 4 °C. After four washes, horseradish peroxidase (HRP)-conjugated sheep anti-rabbit IgG or rabbit anti-sheep IgG (1:2000 v/v dilution; Boster, Wuhan, China) was added and incubated for 1 h at 37 °C. The immunoreactive protein signals were visualized using an Enhanced HRP-DAB Chromogenic Substrate Kit (Tiangen).

Quantitative real-time reverse transcription PCR
Total RNA and cDNA of PSCs and 28-day strobilated worms were obtained as described above. Quantitative real-time reverse transcription PCR was used to analyze expression profiles of EgAnxB3 and EgAnxB38 in PSCs and 28-day strobilated worms. The primers for EgAnxB3 were 5′-TGC CAA CAC GGA TGC CCA AAC-3′ and 5′-CTG GTG CGG TGT GCG AGA AC-3′. The primers for EgAnxB38 were 5′-CGC TAC GCA GAG GAC AAG AACG-3′ and 5′-CTC GCA TCT ACC CAG CAC CAAC-3′. Expression of the actin gene was detected for use as an internal control for normalization. Primers specific to E. granulosus actin were 5′-ATG GTT GGT ATG GGA CAA AAGG-3′ and 5′-TTC GTC ACA ATA CCG TGC TC-3′. The data were analyzed using the 2 − ΔΔCT method [29].

Immunofluorescence localization
Immunofluorescence localization was performed as described previously [26]. The sections (PSCs, germinal layer from fertile/infertile cysts, 18-day strobilated worms and 45-day adult worms) were incubated with purified anti-rEgAnxB3/anti-rEgAnxB38 rabbit IgG or negative rabbit sera (1:200 v/v dilutions) for about 14 h at 4 °C. Fluorescein isothiocyanate-conjugated sheep anti-rabbit IgG (1:100 dilution in 0.1% Evans blue solution) was incubated with sections for 1 h at 37 °C in the dark. The fluorescence signals were observed under a fluorescence microscope. Meanwhile, to detect the possible secretion of EgAnxB3 and EgAnxB38, the liver tissues away from the cysts and near the cyst wall were also evaluated [24].

Phospholipid-binding bioactivity assay
The preparation of liposomes was based on previous reports with some modifications [24,30]. Soybean lecithin (0.9 g; Sangon, Shanghai, China) and 0.3 g cholesterol (Sangon) were mixed and dissolved in anhydrous ethanol in a small beaker; the beaker was placed in a 65-70 °C water bath with stirring in order that the contents could dissolve completely, after which the beaker was rotated to remove the ethanol. Thirty milliliters of preheated phosphate-buffered saline was added to the beaker containing lecithin and cholesterol lipid membrane, and the contents stirred and heated in a water bath at 65-70 °C for 10 min. Finally, the beaker was placed on a magnetic stirrer and the contents stirred for 30-60 min. The phospholipid-binding assay was performed as described previously [24,31]. There were three experimental groups for each recombinant protein: A, B, and C. Twenty microliters of liposomes, 30 μl rEgAnxB3/ rEgAnxB38, and 30 μl of 1 mM CaCl 2 (except for group C) was added to each group, and 50 mM Tris-HCl added as a supplement to a total volume of 100 μl. All the groups were incubated in ice water and centrifuged to separate the supernatant from the precipitate. The precipitate in group B was washed with Tris-HCl. Thirty microliters of 1 mM ethylenediaminetetraacetic acid (EDTA) and 70 μl of Tris-HCl was added to the precipitate of group B and the mixture incubated in ice water for 30 min. The supernatant and precipitate were separated by centrifugation. All the supernatant and precipitate samples were analyzed using 12% sodium dodecyl sulphate-polyacrylamide gel electrophoresis.

Gene amplification and bioinformatics analysis
The results of the bioinformatics analysis of EgAnxB3 and EgAnxB38 are shown in Table 1. Both EgAnxB3 and EgAnxB38 were predicted to be extracellular proteins, but without a signal peptide.
Homology modeling was carried out on EgAnxB3 and EgAnxB38 in the SWISS-MODEL database, and two proteins with high homology were found, whose Protein Data Bank numbers were 1ala.1.A and 1aii.1.A. Using 1ala.1.A and 1aii.1.A as templates, three-dimensional models of EgAnxB3 and EgAnxB38 were constructed, respectively. The similarity between EgAnxB3 and the template sequence was 45.75%, while the similarity was 39.31% between EgAnxB38 and the template sequence. Both EgAnxB3 and EgAnxB38 have calciumbinding sites, but EgAnxB3 contains two types of calcium ion-binding domain, whereas EgAnxB38 only contains one calcium ion-binding domain (Fig. 1).

Expression, purification, and western blotting of rEgAnxB3 and rEgAnxB38
The molecular mass of EgAnxB3 was about 52 kDa, which was close to the expected value [note: the pET-32a(+) tag proteins weigh about 18 kDa]. Solubility   analysis showed that EgAnxB3 was largely expressed as a soluble form and partly in inclusion bodies (Fig. 4a). The molecular mass of EgAnxB38 was about 67 kDa. Solubility analysis showed that EgAnxB38 was largely expressed in inclusion bodies (Fig. 4b). Western blotting showed that rEgAnxB3 and rEgAnxB38 reacted with rabbit anti-rEgAnxB3 antibody and rabbit anti-rEgAnxB38 antibody, respectively, and with CE-positive sheep sera. Moreover, the native EgAnxB3 protein and EgAnxB38 protein in crude protein from PSCs were recognized using the rabbit anti-rEgAnxB3 antibody and rabbit anti-rEgAnxB38 antibody, respectively (Fig. 4).

Transcriptional profiles of EgAnxB3 and EgAnxB38
EgAnxB3 and EgAnxB38 genes were transcribed in both PSCs and 28-day strobilated worms, and the transcript levels of EgAnxB3 and EgAnxB38 in 28-day strobilated worms were both significantly higher (P < 0.05) than those in the PSCs (Fig. 5).

Immunolocalization of EgAnxB3 and EgAnxB38
EgAnxB3 and EgAnxB38 were distributed in the germinal layer of fertile/infertile cysts, the parenchyma and tegument of 18-day strobilated worms and 45-day adult worms, and the calcareous corpuscles and hooks of PSCs. Compared with EgAnxB3, EgAnxB38 was not only distributed in the calcareous corpuscles and hooks of PSCs, but also in the tegument of PSCs. Meanwhile, the distribution range of EgAnxB38 in fertile/infertile cysts, adult worms, and PSCs was significantly wider than that of EgAnxB3, showing a strong fluorescence signal (Fig. 6).
EgAnxB3 and EgAnxB38 were distributed in hepatic sinusoids of liver tissue close to hydatid cysts. In addition, EgAnxB38 was also found in hepatic sinusoids in the liver distant from the hydatid cysts, whereas EgAnxB3 was not found in this area (Fig. 7).

Phospholipid-binding bioactivity analysis
Recombinant EgAnxB3 and EgAnxB38 were reacted with lipidosomes in the presence or absence of Ca 2+ to examine their calcium-dependent phospholipid-binding properties. The mixtures were centrifuged, and the bound protein was pelleted with the precipitate and the unbound protein remained in the supernatant. In the presence of Ca 2+ , rEgAnxB3 was observed in the precipitate rather than in the supernatant (group A). When Ca 2+ was removed from the recombinant protein using EDTA, rEgAnxB3 largely remained in the supernatant (group B). rEgAnxB3 was mainly observed in the supernatant in the absence of Ca 2+ (group C). However, rEgAnxB38 was observed in both the precipitate and supernatant, regardless of the presence of Ca 2+ (Fig. 8).

Discussion
The classification and nomenclature of annexin B group are confusing. Generally, parasites are invertebrates, and their annexins should be named "annexin B. " However, when we screened EgAnxB3 and EgAnxB38 at the National Center for Biotechnology Information based on Zheng's research in 2013 [32], they were both called "AnxA7, " which is obviously incorrect. We found that EgAnxB3 had high similarity with T. solium annexin B3 via sequence alignment, so we named it "EgAnxB3. " However, we did not find an appropriate named parasite sequence with high homology to EgAnxB38. However, we found an article that named the annexins of many parasites by sequence alignment and phylogenetic tree analysis, including the annexins of E. granulosus [33]. We searched for E. granulosus annexins on the Hofmann Laboratory website (http://www.struc tural chemi stry.org/ annex ins/seq/searc h.php), which showed that our naming of EgAnxB3 is correct, and that the name EgAnxB33 is also consistent with that previously published [24]. Therefore, we confirm here that, for E. granulosus, "EgAnxB38" should be used rather than "AnxA7. " Annexin proteins generally have four homologous repeat domains, and most of the domains contain a typical K-G-X-G-T sequence. It has been reported that the calcium-binding capacity of annexin proteins is based on their K-G-X-G-T sequence [33][34][35]. The amino acid sequence analysis showed that both EgAnxB3 and EgAnxB38 have four repeat domains and contain a K-X-G-T sequence. The tertiary structure prediction showed that both EgAnxB3 and EgAnxB38 have Ca 2+ -binding sites. To verify the calcium-binding ability of rEgAnxB3 and rEgAnxB38, we conducted a Ca 2+ -dependent phospholipid-binding assay and found that, in the absence of Ca 2+ , rEgAnxB3 could not bind to liposomes and was found in the supernatant, while in the presence of Ca 2+ , rEgAnxB3 bound with liposomes and was found in the precipitate. When EDTA was added to combine with calcium ions, EgAnxB3 was again found in the supernatant, indicating that rEgAnxB3 has Ca 2+ -dependent phospholipid-binding properties. However, the Ca 2+ -dependent phospholipid-binding properties of rEgAnxB38 were relatively weak. rEgAnxB38 was observed in both the precipitate and supernatant, regardless of the presence of Ca 2+ . There are two possible reasons for this. First, in this study, EgAnxB38 was expressed as an inclusion body protein, and thus possibly did not fold correctly, resulting in a lack of calcium-binding capacity. Second, regardless of the amino acid structure analysis or tertiary structure prediction, EgAnxB38 had fewer Ca 2+ -binding sites, resulting in an insufficient calcium-binding capacity.
The transcript levels of annexins are different at various stages of parasite development. Taenia multiceps AnxB2 and AnxB12 show high levels of transcription in the oncosphere. The transcript levels of AnxB2 and AnxB12 decreased from oncosphere to adult, while that of AnxB3 increased [34]. The transcript level of C. sinensis AnxB30 is higher in the metacercaria stage, and lower in adult worms and eggs [36]. The transcript levels of EgAnxB3 and EgAnxB38 in the 28-day strobilated worms were higher than those of PSCs in this study. Considering that annexins can protect the parasite from the host immune system during parasitism, and that PSCs develop into adult worms in the small intestine of the definitive host [16,17], it is suggested that EgAnxB3 and EgAnxB38 might play a crucial role in the process of PSC invasion of the definitive host. The parasite communicates with the host through the tegument, and molecules distributed in the tegument participate in the host-parasite interaction, which includes excretion, nutrient absorption, and interaction with the host immune system [37,38]. In the present study, both EgAnxB3 and EgAnxB38 were found to be located in the tegument and parenchyma of immature and gravid proglottids on the basis of immunofluorescence localization analysis, but EgAnxB38 had a wider distribution and stronger fluorescence intensity. These results indicate that both these proteins might have a paramount role in the process of PSC invasion of the definitive host and development of the parasite, but that EgAnxB38 may possibly play a larger role than EgAnxB3 in this process. EgAnxB38 was distributed in the hooks and tegument of the PSCs, while EgAnxB3 was either not distributed in the hooks and tegument at all, or at a much lower level. Considering that PSCs require the participation of the hook during development in the definitive host's intestines, it is speculated that EgAnxB38 participates in the interaction between the PSCs and the definitive host. PSCs form and mature in the germinal layer of E. granulosus [39]. EgAnxB3 and EgAnxB38 were distributed in the germinal layer, and their distribution range in fertile cysts was larger than that in infertile cysts, suggesting that they might be involved in the growth and development of PSCs.
Most annexins are cytoplasmic proteins or cytoskeletal proteins [40]; however, traces of annexins have also been found in the extracellular regions, although these annexins lack the signal peptide sequences required for extracellular secretion [41][42][43]. This interesting phenomenon suggests that some annexins are secreted. C. sinensis AnxB30 is a secretory protein that is involved in the interaction between the parasite and host, and affects the host's autoimmune response [36]. As a secreted protein, T. solium AnxB1 is involved in the interaction with the host inflammatory cells [16,22,23]. Annexins were also found to be present in the excretory-secretory product and hydatid cyst fluid of E. granulosus [6,44]. EgAnxB33 could be detected in the hydatid fluid of E. granulosus and was located on the inflammatory cells and fibroblasts of the host-derived layer, suggesting that it might play a paramount role in the interaction between E. granulosus and the host [24]. Although EgAnxB3 and EgAnxB38 have no signal peptide, in this study, they were found to be located in the hepatic sinusoids of liver tissue near hydatid cysts, indicating that they can be secreted into the extracellular areas to exert their physiological effects through non-classical secretion pathways, such as transfer via extracellular vesicles [45]. EgAnxB38 was also found to be distributed in hepatic sinusoids in the liver, which were far away from the cysts. The fact that EgAnxB38 had a wider distribution in different life cycle stages of E. granulosus suggests that it plays a more important role in the development and interaction with the host. Certain excretory or secretory products of the parasite can regulate the host immune system through interaction with the host [46,47]. The results of this study indicate that further investigation is needed to show if EgAnxB3 and EgAnxB38 play a role as secreted proteins in the immune evasion of E. granulosus. In addition, parasite annexins have potential as drug and vaccine targets, so it is also worth exploring if EgAnxB3 and EgAnxB38 can be used as drug targets and vaccine candidates.

Conclusions
In conclusion, we found that both EgAnxB3 and EgAnxB38 contain calcium-binding sites, and rEgAnxB3 has Ca 2+ -dependent phospholipid-binding properties. Both rEgAnxB3 and rEgAnxB38 could be specifically recognized by CE-positive sheep sera, indicating that they had strong immunoreactivity. The transcription level of EgAnxB3 and EgAnxB38 in 28-day strobilated worms was higher than that in PSCs. EgAnxB3 and EgAnxB38 were distributed in all stages of E. granulosus, and were also distributed in the liver tissues near hydatid cysts, indicating that they are secreted proteins that might play a crucial role in the development of E. granulosus.