IrAE: an asparaginyl endopeptidase (legumain) in the gut of the hard tick Ixodes ricinus

Abstract

Ticks are ectoparasitic blood-feeders and important vectors for pathogens including arboviruses, rickettsiae, spirochetes and protozoa. As obligate blood-feeders, one possible strategy to retard disease transmission is disruption of the parasite's ability to digest host proteins. However, the constituent peptidases in the parasite gut and their potential interplay in the digestion of the blood meal are poorly understood. We have characterised a novel asparaginyl endopeptidase (legumain) from the hard tick Ixodes ricinus (termed IrAE), which we believe is the first such characterisation of a clan CD family C13 cysteine peptidase (protease) in arthropods. By RT-PCR of different tissues, IrAE mRNA was only expressed in the tick gut. Indirect immunofluorescence and EM localised IrAE in the digestive vesicles of gut cells and within the peritrophic matrix. IrAE was functionally expressed in Pichia pastoris and reacted with a specific peptidyl fluorogenic substrate, and acyloxymethyl ketone and aza-asparagine Michael acceptor inhibitors. IrAE activity was unstable at pH > or = 6.0 and was shown to have a strict specificity for asparagine at P1 using a positional scanning synthetic combinatorial library. The enzyme hydrolyzed protein substrates with a pH optimum of 4.5, consistent with the pH of gut cell digestive vesicles. Thus, IrAE cleaved the major protein of the blood meal, hemoglobin, to a predominant peptide of 4kDa. Also, IrAE trans-processed and activated the zymogen form of Schistosoma mansoni cathepsin B1 -- an enzyme contributing to hemoglobin digestion in the gut of that bloodfluke. The possible functions of IrAE in the gut digestive processes of I. ricinus are compared with those suggested for other hematophagous parasites.

Figure 1. Multiple alignment of IrAE with S. mansoni and human asparaginyl-endopeptidases

I. ricinus: the hard tick Ixodes ricinus, this work (GenBank AY584752); S. mansoni: the blood fluke Schistosoma mansoni (Caffrey et al., 2000; GenBank AJ250582); H. sapiens: human (Chen et al., 1997; GenBank Y09862). The conserved His and Cys residues forming the catalytic dyad of AE are marked with an asterisk. Space symbol indicates the possible N-glycosylation sites. The arrows depict the cleavage sites of the N- and C-terminal pro-domains experimentally determined for human AE (Chen et al., 2000; Li et al., 2003). The bold dot marks the predicted Asn residue cleavage of the IrAE C-terminal domain. The lower-case partial sequence shows the experimental N-terminal sequence of an enriched IrAE fragment.

Figure 2. Phylogenetic comparison of IrAE with selected AEs

Trees were reconstructed using the Neighbor Joining method (Saitou and Nei, 1987) using amino-acid sequences spanning across the putative mature enzymes and C-terminal extensions. Mammal - Mus musculus (O89017), Rattus norvegicus (Q9R0J8); Homo sapiens (Q99538); amphibian - Xenopus laevis (AAH56842); tick - Ixodes ricinus (AAS94231); plant - Canavalia ensiformis (P49046), Oryza sativa (Q8GS39); Arabidopsis thaliana (P49047), Vigna radiata (Q9AUD9); helminth - Haemonchus contortus (CAJ45481), Schistosoma mansoni (Q9NFY9), Schistosoma japonicum (CAA50304), Fasciola hepatica (CAC85636); protist - Trichomonas vaginalis 1 and 2 (GenBank AAQ93039, AAQ93040, respectively). The horizontal bar represents a distance of 0.1 substitutions per site. Numbers at the branches represent bootstrap support. The sequence obtained in this study is marked in boldface.

Figure 3. Tissue expression profile of IrAE in female I. ricinus

Messenger RNA levels were determined by semi-quantitative two-step RT PCR. Tissues were dissected and pooled from 10 semi-engorged females fed for five days on guinea pigs. I. ricinus ferritin mRNA was used as a loading control. For details, see Material and Methods.

Figure 4. Localization of IrAE in the gut of female I. ricinus by indirect immunofluorescence microscopy and immunogold electron microscopy

Sections were prepared from guts dissected from semi-engorged I. ricinus females (5 days of feeding). Panel A – semi-thin sections stained with toluidine blue - general structure of the tick gut showing the boundary between the gut epithelium (GE) digestive vesicles and the gut lumen (GL), containing large hemoglobin crystals (Hb) and digestive gut cells (dGC); Nc–nuclei; Vs – digestive vesicles. Panel B – semi-thin section labeled with Ra×IrAE serum (1:25) and FITC-conjugated anti-rabbit antibody merged with Höchst 33–258 staining (blue). Note, IrAE-specific signal was markedly enriched within the peritrophic matrix (PM) and also present intracellularly in the digestive vesicles. Panel C – a detailed image of a digestive gut cell in the phase of detachment from the gut epithelium, same staining as in Panel B. Panel D – electron microscopy of ultrathin-sections labeled with Ra×IrAE serum (1:25) and protein A conjugated with immunogold particles. IrAE-specific labeling within the peritrophic matrix showing the association of IrAE with microvilli (Mv).

Figure 5. Enzymatic characteristics of recombinant IrAE expressed in Pichia pastoris

IrAE was expressed in P. pastoris, desalted and concentrated by ultrafiltration (30 kDa cutoff). All activity assays were performed at room temperature in triplicate using the fluorogenic substrate Z-AAN-AMC. Panel A – Activation of IrAE in the presence of DTT. IrAE was transferred to citrate-phosphate-salt buffer (CPS) pH 4.5 and the activation started by addition of 4 mM DTT. IrAE was fully activated in 3 hrs and the activity was stable for at least 20 hrs. Panel B – pH optimum of activated IrAE. Activated IrAE (20 h activation) was transferred to CPS of specified pH and the rate of AMC production was measured. Notice the loss of activity at pH > 6.0. Panel C – IrAE stability at different pH values. An aliquot of activated IrAE was pre-incubated for 4 h in CPS buffer of specified pH, then transferred to the CPS buffer of pH 5.5 with 4 mM DTT and the rate of AMC production was measured. The activated IrAE was stable only at pH < 6.0. Panel D – Inhibition of activated IrAE with a legumain-specific inhibitor. The activated IrAE was pre-incubated with two-fold serial dilutions of the inhibitor Aza-Asn-11A in CPS buffer, pH 5.5 for 30 minutes and then the rate of AMC production was measured. The IC₅₀ concentration of the inhibitor was in the range of about 500 nM. The open symbol represents the IrAE activity without inhibitor.

Figure 6. Visualization of active recombinant IrAE by specific antiserum and the activity-based probe, Fhx-PD-AOMK

Desalted and concentrated IrAE in P. pastoris medium (see above) was resolved by reducing, gradient SDS-PAGE. Parts of the gel were silver stained or electroblotted onto PVDF membrane and visualized with either Ra×IrAE antibodies or the activity-based fluorescent probe, Fhx-PD-AOMK. Lane 1 – Silver stain of desalted and concentrated IrAE in P. pastoris medium; Lane 2 – Western blot using Ra×IrAE serum (1 :100), swine×Ra-IgG – peroxidase conjugate and diaminobenzidine as substrate. Lanes 3, 4 – IrAE in P. pastoris medium pre-incubated with Fhx-PD-AOMK in the absence and presence of a legumain-specific inhibitor Aza-Asn-11a, respectively. For details, see Material and Methods.

Figure 7. Specificity profile of IrAE using a positional scanning synthetic combinatorial library

P1-P4 specificities were determined with a synthetic peptidyl library in which randomized positions were incorporated by addition of the isokinetic mixture of 20 amino acids (Choe et al., 2006).

Figure 8. Trans-processing of S. mansoni pro-cathepsin B1 by IrAE

Trans-processing of the SmCB1 zymogen (expressed in P. pastoris) was performed as described using the endogenous S. mansoni AE (Sajid et al., 2003). Graph – pro-SmCB1 was pre-incubated with or without activated IrAE at the specified pH and the SmCB1 activity assayed with the fluorogenic substrate Z-Phe-Arg-AMC. Inset – processing of pro-SmCB1. Pro-SmCB1 was visualized using a radio-iodinated version of the DCG-04 activity-based probe (Sajid et al., 2003). Processing could be inhibited by prior incubation with the legumain-specific inhibitor, Aza-Asn-11a. SmCB1pm, SmCB1pmⁱ, SmCB1m refer to the zymogen, partially processed SmCB1 zymogen and the mature form of SmCB1, respectively (Sajid et al., 2003).

Figure 9. Digestion of hemoglobin by activated IrAE

Human hemoglobin (10 µg) was incubated with activated IrAE in CPS buffers of specified pH values in the presence (+) or absence (−) of the legumain-specific inhibitor, Aza-Asn-11a. The reaction mixture was resolved on reducing, gradient SDS-PAGE and stained with Coomassie Brilliant Blue. Digestion of hemoglobin was most efficient at pH 4.5 and was inhibited by Aza-Asn-11a. The scale on right of panel indicates the positions of pre-stained molecular weight standards.