Introduction

Babesia spp. taxonomically belongs to the parasitic protozoa of the phylum Apicomplexa (He et al. 2017). It can infect a variety of animals, including humans and mice, with severe acute disease characterized by sequestration of mature parasites; the attributes of the disease are high fever, hemoglobinuria, and a respiratory distress syndrome-like condition (Novacco et al. 2019; Wright et al. 1988). Ticks including Boophilus microplus, Boophilus annulatus, Ixodes persulcatus, and Rhipicephalus haemaphysaloides are main vectors of Babesia spp. (Klompen et al. 1996; Young and Morzaria 1986). Generally, after a larva or nymph feeds on the blood of infected animals, Babesia spp. can be transmitted by nymphs or adult ticks during their next stage of development (Berdyev 1985). Hemoparasites pose a serious threat to the livestock population worldwide, mostly affecting livestock industry in subtropical and tropical regions throughout the world. Babesia orientalis, a Babesia species, has been identified in water buffalo in southern and central China, which lead to a massive economic loss to the Chinese buffalo industry (Kivaria 2006; Ristic 1988).

Microbes such as Plasmodium falciparum, African trypanosomes, Candida albicans, and Borrelia hermsii establish long-term existence in host cells via antigenic variation mechanism which is a rapid alteration in the structure and antigenicity of essential components (Arnarson et al. 2017; Biswal et al. 2019; Deitsch et al. 1997; Langsley and Roth 1987; Palmer et al. 2016). The pathogen-synthesized components which appeared during the course of a single infection have been demonstrated to serve a multi-functional role; similar mechanism is being shown by the bovine hemoparasite, Babesia bovis (Allred 2001; Allred et al. 1994). At least two interrelated mechanisms have been proposed to establish persistent infections in cattle: cytoadhesion of infected erythrocytes containing mature parasites and rapid antigenic variation of the cytoadhesion ligands (Allred et al. 1994; O’Connor et al. 1999). The second mechanism is supposed to protect the cytoadhesion from ongoing and adaptive immune responses (Allred et al. 1994). The variant erythrocyte surface antigen-1 (VESA1), which is a size-polymorphic heterodimeric protein composed of two subunits 1a and 1b, is responsible for mediation of antigenic variation and cytoadhesion (Allred et al. 2000, 1993). The ves1α and ves1β genes, from clearly related but structurally distinct branches of the ves gene family, encode the two subunits VESA1a and 1b, respectively (Allred et al. 2000; Xiao et al. 2010). The ves gene family is conserved among different Babesia species (B. bigemina, B. divergens, B. ovata, and B. bovis) but greatly diversified in protein structure and gene repertoire, involving persistent challenges of host-parasite interactions for millions of years. B. bovis VESA1 proteins possess cysteine-rich CKRD and VDCS domains, which are lacking in VESA1 proteins in B. divergens, B. bigemina, and B. ovata (Jackson et al. 2014).

This study is the first to identify and characterize VESA1 protein of B. orientalis (BoVESA1) with molecular weight ~ 28 kDa. The immunogenicity and localization of BoVESA1 were identified by western blotting and immunofluorescent assay (IFA). The results indicate that BoVESA1 is a promising antigen for diagnostic usage and vaccine development against B. orientalis; the results are constructive to ensure the promotion of relevant research.

Materials and methods

Cloning and sequencing of the BoVESA1 gene

The genome of B. orientalis (unpublished data) was screened for the reported VESA1 amino acid sequences of B. bovis (T2Bo) (GenBank: XP_001609533.1) using TBLASTN. The gene with a remarkable similarity was designated as the BoVESA1 gene. Complete recombinant plasmids were constructed using the homologous recombination method. Specific primers for the BoVESA1 gene and pGEX-6p-1 vector (Table 1) were designed using Clone Manager software and then synthesized (Tianyi Huiyuan Biological Technology, Wuhan, China).

Table 1 Primers used for construction of pGEX-6p-1-BoVESA1 recombinant plasmids in this study
Full size table

The BoVESA1 gene was amplified from both B. orientalis cDNA and gDNA obtained from previous studies (Guo et al. 2018a) with Phanta Max Super-Fidelity DNA Polymerase (Vazyme, Nanjing, China). The thermal cycling parameters were as follows: activation at 95 °C for 3 min; 35 cycles of denaturation at 95 °C for 15 s, annealing at 54 °C for 15 s, extension at 72 °C for 1 min, and a final extension at 72 °C for 5 min. The vector pGEX-6p-1 plasmid DNA was linearized for homologous recombination. The thermal cycling parameters were as follows: initial denaturation at 95 °C for 3 min; 35 cycles of denaturation at 95 °C for 15 s, annealing at 61 °C for 15 s, extension at 72 °C for 4 min, and a final extension at 72 °C for 5 min.

The PCR products were electrophoresed using 1.5% ethidium bromide-stained agarose gel and purified using a DNA Gel Extraction Kit (TSINGKE Biological Technology, Beijing, China). The purified BoVESA1 gene was ligated into the extracted product of the pGEX-6p-1 vector. All the constructs were confirmed by DNA sequencing (Tianyi Huiyuan Biological Technology, Wuhan, China).

Phylogenetic and bioinformatics analysis

To attain a holistic comprehension of the evolutionary genetics analysis of the BoVESA1 gene, the amino acid sequence of BoVESA1 was aligned with VESA1 sequences in the database by BLASTP. Phylogenetic analysis was performed by the neighbor-joining method, owing to its superiority to classical programs based on multiple alignments, using MEGA7 software (Kumar et al. 2016). The amino acid (aa) sequence of BoVESA1 was analyzed using the ExPASY online tool (http://www.expasy.org/translate/). Open reading frames (ORFs) were searched using ORF finder (https://www.ncbi.nlm.nih.gov/orf finder/). The putative signal peptide and potential subsequence motifs were predicted using SignalP 5.0 Server (http://www.cbs.dtu.dk/services/SignalP/) and Motif Scan Server (https://myhits.isb-sib.ch/cgi-bin/motif_scan/). The grand average of hydropathicity index (GRAVY) of BoVESA1 was calculated using the ExPASY ProtParam tool (https://web.expasy.org/protparam/). Subsequently, protein sequences were analyzed in detail by predicting the putative glycosylphosphatidylinositol (GPI) anchors with GPI Prediction Server 3.0 (http://mendel.imp.ac.at/sat/gpi/gpi_server.html) and predicting the transmembrane (TM) regions using https://embnet.vital-it.ch/software/TMPRED_form.html#opennewwindow. The secondary structure of BoVESA1, surface probability, hydrophilicity, flexible region, and antigenic index were analyzed and predicted by the PROTEAN program of the Lasergene® software package in DNAstar.

The complete BoVESA1 amino acid sequence was submitted to the I-TASSER server (https://zhanglab.ccmb.med.umich.edu/) for predicting its three-dimensional structure. There were approximately three sessions to predict the three-dimensional structure of BoVESA1. First, I-TASSER modeling started from the structure templates identified by the local meta-threading server (LOMETS) from the PDB library. The 10 best threading templates with higher structure accuracy of the conserved regions, owing to their highest significance in the threading alignments and hence a higher normalized Z-score, were selected from each threading program (Yang and Zhang 2015; Roy et al. 2012). Second, based on the significance of threading template alignments and the convergence parameters of the structure assembly simulations, five models corresponding to the five largest structure clusters were recorded using the SPICKER program (Yang et al. 2015; Zhang 2009). The model with the best quality among five selected models was regarded as the preliminary tertiary structure of BoVESA1 for prediction. Eventually, TM-align, a structural alignment program, was performed to match the first I-TASSER model to all structures in the PDB library, and the top 10 proteins were reported to have the closest structural similarity.

Expression and purification of the recombinant proteins

The recombinant plasmid (pGEX-6p-1-BoVESA1) was transformed into E. coli BL21 (DE3) strain. Small-scale culture was performed under isopropyl-β-D-thiogalactopyranoside (IPTG) at concentration of 1 mM and the induction was continued at 37 °C for 5 h to have the capacity and form of proteins certified by gel electrophoresis. BoVESA1 was expressed as a glutathione‐S‐transferase (GST)-fusion protein in E. coli as an inclusion body. The transformed E. coli were then lysed by ultrasonication in PBS (pH = 7.4) containing phenylmethylsulfonyl fluoride (PMSF) and centrifuged further at 12,000 × g for 30 min at 4 °C. Precipitates were treated with PBS containing 0.2 g/mL sodium N-dodecanoyl sarcosinate (SKL) for 2 h at 25 °C and then centrifuged at 15,000 × g for 10 min at 4 °C. Supernatant containing rBoVESA1 was treated with a refolding buffer (0.2 g/mL PEG-4000, 1 mM oxidized glutathione, and 2 mM reduced glutathione) overnight at 4 °C to allow complete refolding of the protein and thus to obtain purified protein.

Development of polyclonal antibodies

Kunming mice (specific pathogen-free) were subcutaneously injected with 100 μg of purified rBoVESA1 emulsified in Freund’s complete adjuvant (FCA, Sigma, San Francisco, CA, USA). Mice were injected with the same antigen (concentrated dose) in Freund’s incomplete adjuvant (FIA, Sigma) on days 14, 28, and 42 after the first immunization. Serum samples from immunized mice and one healthy mouse were collected 14 days after the last immunization. The collected samples were stored at − 20 °C for further use.

SDS-PAGE and western blot analysis

rBoVESA1 was subjected to 12% sodium dodecyl-sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to a nitrocellulose membrane (Merck Millipore, Kenny, NJ, USA), which was subsequently blocked with 5% (w/v) skimmed milk in Tris-buffered saline Tween-20 (TBST). The membranes probed with serum (1:50 dilution) from the buffalo infected with B. orientalis or healthy water buffalo were incubated at 25 °C for 1 h with goat anti-bovine IgG/HRP (1:2000) as secondary antibodies (Bioss, Beijing, China) and visualized using the electrochemiluminescence (ECL) method.

To determine the native BoVESA1 in merozoites of B. orientalis, proteins from the lysates of B. orientalis-infected buffalo erythrocytes and uninfected buffalo erythrocytes were first electrophoresed through 12% SDS-PAGE gel, transferred to a nitrocellulose membrane, and probed with anti-rBoVESA1 mice serum (1:50 dilution) or the naive mouse serum at the same dilution as controls. After three washes with Tris-buffered saline Tween-20 (TBST), the membranes were incubated with goat anti-mouse IgG/HRP (1:4000) as secondary antibodies (Bioss) and finally detected using the ECL approach.

Indirect immunofluorescence assay

To determine the cellular localization of BoVESA1, water buffalo RBCs infected with B. orientalis at merozoite stages (4% parasitemia) were smeared on glass slides, air-dried, and then fixed in cold acetone: methanol (50% v/v) solution for 20 min at − 20 °C. After three washes with cold 1 × PBS, smears of blood cells were permeabilized with 0.1% Triton X-100 and blocked with 1% BSA and then incubated with the anti-rBoVESA1 mouse serum (1:100) and the naive mouse serum (1:100, negative control) for 1 h at 37 °C. The secondary antibody was goat anti-mouse IgG (Alexa Fluor 594) diluted 1000 times with 1 × PBS, followed by parasite nucleus staining with Hoechst stain for 1 h at 37 °C. Finally, 10 μL of anti-fluorescence quenching agent was used to prevent the fluorescence quenching effect, and the images were obtained using a fluorescence microscope.

Results

Characterizations of BoVESA1 gene

The BoVESA1 gene contained a 753-bp open reading frame without introns and encoded a polypeptide of 250 aa (Fig. 1). The nucleotide sequence was submitted to GenBank with the accession number MT721826. In addition, the BoVESA1 gene was expressed as a GST-fusion protein with a predicted molecular weight of 53 kDa in the SDS-PAGE. It was also demonstrated that the expressed proteins were insoluble and existed in the form of inclusion bodies (Fig. 2). Primary purified rBoVESA1 was obtained, collected, and dialyzed to immunize mice.

Fig. 1
figure 1

PCR amplification of BoVESA1 gene from B. orientalis gDNA and cDNA. Lane M: marker; lane 1: amplicon from gDNA; lane 2: amplicon from B. orientalis cDNA; lane 3: amplicon from distilled water as blank control; lane 4: amplicon from uninfected buffalo erythrocytes cDNA as negative control

Full size image
Fig. 2
figure 2

SDS-PAGE expression and purification of rBoVESA1. Lane M: molecular weight marker; lane 1: induced pGEX-6p-1-BoVESA1; lane 2: non-induced pGEX-6p-1-BoVESA1 as control; lane 3: supernatant of pGEX-6p-1-BoVESA1; lane 4: pellet of pGEX-6p-1-BoVESA1; lane 5: purified rBoVESA1. The corresponding bands are indicated by arrows

Full size image

Phylogenetic and bioinformatics analysis

The amino acid sequence of BoVESA1 was aligned with VESA1 sequences in the database. Among all the amino acid sequences producing significant alignments, BoVESA1 was most closely related to VESA1 protein in B. bovis (XP_001611438.1) with an identity of 34% and a query cover of 92%. Additionally, BoVESA1 showed 30% and 21% identity to VESA1 of Babesia sp. Xinjiang (XP_028870229.1) and B. bigemina (XP_012769198.1), respectively. The evolutionary relationships between VESA1 amino acid sequences from Babesia bovis, Babesia orientalis, Babesia bigemina, Babesia ovata, Babesia microti, and Babesia sp. Xinjiang were analyzed using the neighbor-joining method in the MEGA7 software, using P. falciparum erythrocyte membrane protein 1 (PfEMP-1) as an outlier (Fig. 3). According to the phylogenetic analysis, the results showed that BoVESA1 has a closer relationship with VESA1 of B. bovis than those of other Babesia species.

Fig. 3
figure 3

Molecular phylogenetic analysis of BoVESA1 amino acid sequences by the neighbor-joining method. The tree is drawn to scale, with branch lengths measured in the number of substitutions per site. The analysis involved 24 amino acid sequences. All positions containing gaps and missing data were eliminated. There were a total of 172 positions in the final dataset. The sequence newly identified in this study is presented with a bold font. GenBank accession numbers are indicated

Full size image

Four important indexes, including hydrophilicity, flexible regions, antigenic index, and surface probability, were predicted on the basis of some relevant algorithms that were inclusive of Kyte-Doolittle, Karplus Dchulz, Jameson-Wolf, and Emini, through the PROTEAN program of the Lasergene® software package in DNAstar. Hydrophilic regions were predicted to take up more of the amino acid sequence than hydrophobic regions; the grand average of hydropathicity (GRAVY) index of BoVESA1 was − 0.284, implying a high hydrophilicity of BoVESA1. The structures of turn and coil had certain flexibility and deformability, which may promote the binding process with antibodies and serve as a possible epitope, in contrast to the structures of α-helices and β-sheets, which are indeformable and uneasy to bind antibodies. Regions with high accessibility are distributed on the surface of the antigen to bind with antibodies more easily. High antigenic regions with a uniform distribution throughout the amino acid sequence focused mainly in the range of 7 to 62 and 112 to 163 aa. These four related indices are needed to evaluate potential B cell epitopes: good hydrophilicity and antigenicity as well as high accessibility and flexibility. Conclusively, three regions of 9 to 13 aa, 51 to 62 aa, and 147 to 152 aa are most likely to be potential B cell epitopes.

The BoVESA1 gene was predicted to have one exon and no intron with a lack of GPI anchor sites and N-terminal signal peptide. Consistent with the VESA1 proteins of other Babesia species, BoVESA1 was revealed to contain a transmembrane region (59–84 aa). The motif scanning result of the amino acid sequence of BoVESA1 was obtained from Motif Scan Server. It mainly consisted of an N-glycosylation site between 10 and 13 aa, four casein kinase II phosphorylation sites in the range of 5 to 238 aa, three protein kinase C phosphorylation sites in the range of 56 to 146 aa, seven N-myristoylation sites in the range of 35 to 245 aa, and one RGD cell attachment sequence (146–148 aa).

The amino acid sequence of BoVESA1 was processed using the I-TASSER software to predict the three-dimensional structure. A schematic illustration of this process is shown in Fig. 4. Among the proteins selected from the PDB library, a light-harvesting chlorophyll a/b-binding (LHC) protein (PDB code: 6ijj3) possessed the closest structural similarity to the BoVESA1 model, which is encoded by the lhca3 gene of Chlamydomonas reinhardtii (Cr); LHC being a membrane protein can collect and transfer light energy to the reaction centers of the photosystem I (PSI) (Su et al. 2019). Because of its similar structure to 6ijj3, BoVESA1 was speculated to be combined with some functional proteins to transmit energy and signal, which remains to be deeply studied in further research.

Fig. 4
figure 4

Schematic illustrations of predicted tertiary structures of BoVESA1. a Predicted tertiary structure of BoVESA1 showing α-helices (red) and coils (green). The tertiary structure contains nine α-helices (α1 and α9). b Tertiary structure of light-harvesting chlorophyll a/b-binding (LHC) protein (PDB code: 6ijj3). The tertiary structure in identical orientations contains eight α-helices (α1 and α8)

Full size image

Identification of the recombinant and native BoVESA1

To evaluate the immunoreactivity of BoVESA1, rBoVESA1 was reacted with the serum collected from B. orientalis-infected and uninfected buffalo and analyzed by western blotting. Only one specific band corresponding to rBoVESA1 (53 kDa) appeared in the serum from the infected water buffalo (Fig. 5a), but no band was observed in the serum of the control buffalo. In addition, BoVESA1 has been predicted to have good hydrophilicity and antigenicity; thus, BoVESA1 may be conceived as a good diagnostic antigen for the detection of B. orientalis.

Fig. 5
figure 5

Western blot identification of immunoreactivity and native form of BoVESA1. a Lane M: molecular weight marker; lane 1: rBoVESA1 reacted with B. orientalis positive serum; lane 2: rBoVESA1 probed with the serum of uninfected buffalo. b Identification of the native form of BoVESA1 in B. orientalis merozoite lysate. Lane 1: reaction of lysate of B. orientalis-infected buffalo erythrocytes with the serum against rBoVESA1; lane 2: reaction of lysate of uninfected buffalo erythrocytes with the serum against rBoVESA1; lane 3: lysate of B. orientalis-infected buffalo erythrocytes probed with the serum of naïve mice; lane 4: lysate of uninfected buffalo erythrocytes probed with the serum of naïve mice. The corresponding bands are indicated by arrows

Full size image

The identification of native BoVESA1 was also performed by incubating the lysates of B. orientalis-infected erythrocytes with anti-rBoVESA1 serum. The results displayed a ~ 28 kDa band, which was similar to the anticipated size of native BoVESA1 (Fig. 5b), while the control group showed no band. The presence of a ~ 28 kDa band demonstrated the excellent immunological properties of rBoVESA1 and confirmed the presence of native BoVESA1 in B. orientalis merozoite.

Localization of BoVESA1

IFA was performed to determine the localization of BoVESA1 by mouse anti-rBoVESA1 polyclonal antibodies serum and naive mouse serum. The results showed a strong signal in infected RBCs (Fig. 6), while no fluorescence was detected in the negative control.

Fig. 6
figure 6

Localization of BoVESA1. Immunofluorescence and electron microscopy analysis of BoVESA1 in the smears. PcAb-BoVESA1 (red) and nucleus staining of Hoechst (blue). a Reactivity of the anti-rBoVESA1 serum with intracellular parasite. b Negative control: the primary antibody was serum from a naïve mouse. Scale bars: 1 μm

Full size image

Discussion

Several studies have been conducted on B. orientalis including analyzing and sequencing the mitochondrial and apicoplast genomes of B. orientalis (He et al. 2017, 2014; Huang et al. 2015). However, it still remains unclear about the mechanism of invasion of parasites into host erythrocytes and avoiding host immune responses.

Persistent infections in B. bovis could be established with the involvement of at least two mechanisms including endothelial cytoadhesion and antigenic variation. Sequestration of infected erythrocytes (David et al. 1983; Wright 1972) results from cytoadhesion of the infected red blood cells to the capillary along the post-capillary venous endothelium via some knob-like structures emerging from its surface (Aikawa et al. 1985; Hutchings et al. 2007; O’Connor and Allred 2000; O’Connor et al. 1999; Schmidt et al. 1982). For antigenic variation, it was proposed to make sure adhesion function escaped from adaptive host immune responses (Biggs et al. 1992; Leech et al. 1984; Magowan et al. 1988; O’Connor et al. 1997). Both of these functions are related to the VESA1 on the surface of infected erythrocytes (Allred et al. 2000, 1994; O’Connor and Allred 2000; O’Connor et al. 1997). Usually, it exists in two different types, VESA1a and VESA1b, encoded by ves1α and ves1β, respectively; a small third branch of only three copies, ves1γ, has been observed as well but is of unknown function (Allred 2019). The transcription of B. bovis ves gene occurs at a “locus for active transcription” (LAT) (Al-Khedery and Allred 2006), including ves1α/ves1β or ves1α/ves1α gene pairs with divergent orientations. The pathogens such as Plasmodium falciparum, African trypanosomes, and Candida albicans have the ability to escape immune responses due to continuous generation of variant antigens. The ves genes in Babesia spp. also follow similar mechanism as the var genes in the malaria parasite Plasmodium falciparum (Bopp et al. 2013) and the vsg genes in Trypanosoma brucei (Jackson et al. 2012).

The ves gene family is relatively conserved for Babesia spp. but highly diversified in protein structure and gene repertoire, owing to gene duplication and expansion over the years. The ves1α/ves1β branches in B. bovis do not show significant orthologous relationships with the ves1a/ves1b branches in B. bigemina or B. ovata (Jackson et al. 2014). Moreover, cysteine-rich CKRD and VDCS domains of B. bovis are lacking from VESA1 proteins encoded by ves1a/ves1b genes in B. bigemina and B. ovata (Jackson et al. 2014; Yamagishi et al. 2017), which may be responsible for the lack of cytoadhesion and knob-like structures in these species. Furthermore, ves-like short genes appeared after rapid divergence of ves1 structures (Jackson et al. 2014), implying its specific function and evolutionary dynamics.

Although plenty of studies have been conducted on the identification of VESA1 proteins in Babesia spp., but none of them focused on B. orientalis. In the present study, the VESA1 protein of B. orientalis was identified and characterized for the first time. We cloned and sequenced an intronless VESA1 gene with an open reading frame of 753 bp in B. orientalis. Aiming to obtain a detailed understanding of the genetic relationships among different species, the VESA1 amino acid sequences were aligned to each other in BLAST program. The amino acid homology of BoVESA1 to VESA1 protein of B. bovis and Babesia sp. Xinjiang was 34% and 30%, respectively, suggesting that BoVESA1 has a closer relationship with B. bovis. SBP3, SBP4, RON5, and thrombospondin-related anonymous protein 2 (TRAP2) amino acid sequences of B. orientalis have also been reported closer to those of B. bovis (Guo et al. 2018a, b; Tian et al. 2020; Zhan et al. 2019). BoVESA1 was also predicted to have one RGD cell attachment region (146–148 aa), which was possible to facilitate potential cytoadhesion and transmit signals. Moreover, a ~ 28 kDa band was detected in the lysates of B. orientalis-infected buffalo erythrocytes reacted with mouse anti-rBoVESA1 serum, indicating the existence of native VESA1 in B. orientalis merozoite. The IFA results showed a strong fluorescence signal of BoVESA1 (captured in the intracellular stage) in infected RBCs using anti-rBoVESA1 serum; however, the secretion of BoVESA1 into host RBCs might be related to the presence of a transmembrane region close to the N-terminal. Furthermore, in B. bovis, VESA1 protein was confirmed to be localized on the surface of infected erythrocytes, but the mechanism by which VESA1 proteins were trafficked to erythrocyte membranes remains to be further investigated. So far, researchers have discovered that overexpression of spherical body protein (SBP2t11) can cause reduced cytoadhesion of infected erythrocytes (Gallego-Lopez et al. 2019). Since VESA1 in B. bovis was mainly responsible for cytoadhesion, upregulation of the SBP2t11 protein may impede the export of VESA1 proteins to the surface of infected erythrocytes, implying the spherical body compartment was possible to be involved in the export pathway (Allred 2019). Despite vital importance of the invasion stage of parasites, deeper investigations on the post-invasion stage are still necessary because there can be far longer time of being exposed to the host immune system during post-invasion stage than invasion stage and may cause potential risks to animals.

Conclusion

In this study, BoVESA1 was cloned, sequenced, and expressed as a 53 kDa GST-fusion protein with the native form being identified as ~ 28 kDa by western blot and confirmed to have excellent immunogenicity that indicates that BoVESA1 can serve as a good candidate diagnostic antigen for the detection of B. orientalis infection. The IFA results showed a strong signal in infected RBCs. The identification of VESA1 in B. orientalis in the current study may facilitate further investigation on the ves gene and boost the development of effective therapy for B. orientalis infection.