Peptidoglycan Mediates Loa22 and Toll-like Receptor 2 Interactions in Pathogenic Leptospira

Leptospirosis is an overlooked zoonotic disease caused by pathogenic Leptospira. The kidney is the major organ infected by Leptospira which causes tubulointerstitial nephritis. Leptospira outer membrane components contain several virulence factors that play important roles in the pathogenesis of leptospirosis. Among them, OmpA-like protein Loa22 is essential for leptospiral virulence. However, the pathogenic mechanisms of tubulointerstitial nephritis involving this virulence factor are still unclear and need further investigation. In this study, pull-down assays suggested that Toll-like receptor 2 (TLR2) proteins interacted with Loa22 from Leptospira outer membrane extractions. Combination of Atomic force microscopy (AFM) and side-directed mutagenesis suggested that Loa22 exhibited high affinity for Leptospira peptidoglycan (LPGN) and the residues of Loa22 were involved in LPGN interaction. Mutation of two key residues within the OmpA-like domain of Loa22, Asp122 and Arg143, significantly attenuated their relative affinities for LPGN indicating that these two residues were responsible for LPGN binding. Thus Loa22 OmpA domain was responsible for interacting with LPGN and the two indicated residues may participate in binding to LPGN. Recombinant Loa22 (rLoa22) protein was further complexed with LPGN and incubated with HEK293-TLR2 cells to monitor inflammatory responses. Inflammatory responses were provoked by rLoa22-LPGN complexes, but not rLoa22 alone, involved CXCL8/IL8, hCCL2/MCP-1, and hTNF-α activation. Confocal microscopy further identified the co-localization of Loa22-LPGN complexes and TLR2 receptors on HEK293-TLR2 cell surface. The affinity between Loa22-LPGN complexes and TLR2 were further confirmed and measured by AFM and ELISA. Downstream signals from TLR2 including p38, ERK, and JNK were observed by western blotting induced by Loa22-LPGN complexes. In summary, this study identified LPGN in leptospira mediates interactions between Loa22 and TLR2 and induces downstream signals to trigger inflammatory responses. Interactions between Loa22-LPGN-TLR2 reveal a novel binding mechanism for the innate immune system and infection induced by leptospira. Author summary Leptospirosis is one of the most overlooked zoonotic diseases caused by pathogenic Leptospira in warm and humid regions worldwide. With the infection by Leptospira, many organs are invaded and can result in multiple-organ failure (Weil’s syndrome). Kidney is the major organ infected by pathogenic Leptospira, which would manifest as tubulointerstitial nephritis. In this study, we focused on the outer membrane lipoprotein Loa22 (Leptospiral OmpA-like domain 22) from pathogenic Leptospira which triggers inflammatory responses on renal tubular cell. Protein domain prediction indicated that Loa22 contains an important domain termed OmpA-like domain and the function of this domain is peptidoglycan (PGN) binding. From sequence alignments of Loa22 with other OmpA proteins, two important amino acids, Asp122 and Arg143, were found to be highly conserved. The role of the two conserved residues in AbOmpA (OmpA protein in A. baumannii) and Pal (peptidoglycan-associated lipoprotein in E. coli) proteins are important for PGN binding. These two residues in Loa22 were altered by site-directed mutagenesis to obtain D122A and R143A variants. In pull-down and AFM analysis, the binding capacities of Loa22 variants to Leptospira PGN (LPGN) were significantly decreased as compared to rLoa22WT, indicating that the two residues are involved in LPGN binding. Furthermore, recombinant Loa22 and its variants in the absence or presence of LPGN, were incubated with HEK293-TLR2 cells, to confirm the role of LPGN in triggering inflammatory responses involving CXCL8/IL8, hCCL2/MCP-1, and hTNF-α. These factors are involved in downstream signaling of inflammatory responses through Toll-like receptor 2 (TLR2). In addition, confocal microscopy was employed to observe the co-localization of Loa22-LPGN complexes and TLR2 receptors on HEK293-TLR2 cell surfaces. Finally, the interaction forces between rTLR2 and rLoa22-LPGN complexes were measured by AFM and ELISA to conclude the necessary role of LPGN in rLoa22-TLR2 complex formation. In summary, these results demonstrate that the interaction of Loa22 protein with the important cell wall component, PGN, concomitantly triggered inflammatory responses of host cells through interaction with TLR2.

Introduction 7 order to identify their function as PRRs (19). In humans, there are 10 TLR members (TLR1-TLR10) and 12 (TLR1-TLR9, TLR11-TLR13) in mouse. TLRs are further divided into two subfamilies; cell surface TLRs and intracellular TLRs, according to their localization . Cell surface TLRs include   TLR1, TLR2, TLR4, TLR5, TLR6, and TLR10, whereas intracellular TLRs include TLR3, TLR7, TLR8, TLR9, TLR11, TLR12, and TLR13 (18). TLR family proteins play a pivotal role in innate immunity by recognizing conserved patterns in diverse microbial molecules (20). Among these TLRs, TLR2 in association with TLR1 or TLR6 is essential for sensing bacterial lipoproteins and lipopeptides (21,22). TLRs containing leucine-rich repeats (LRRs) are responsible for pattern recognition at the extracellular portion and a Toll/IL-1 receptor (TIR) domain is responsible for signal transduction at the cytoplasmic portion (18). The crystal structure of TLR2/1 complex with its ligand, Pam 3 CSK 4 , revealed that two fatty esters of the glycerol moiety are embedded in a hydrophobic pocket of TLR2, while the amide-bound lipid chain is fitted into the hydrophobic channel in TLR1 (23). A lipid molecule of the lipoprotein presumably supported the binding domain of TLR2. In addition to the lipidation domain of lipoprotein, several known structures and motifs of the TLR2-binding protein have been reported. The PorB protein from N. meningitides has been suggested as a TLR2 ligand and the binding mechanism was hypothesized to involve electrostatic interactions contributing to ligand/receptor interactions (24). The BspA protein from T. forsythia with a LRR domain was also suggested as the TLR2 a ligand (25). The pentameric B subunit of type IIb E.
coli enterotoxin (LT-IIb-B5) uses its hydrophobic upper pore region to directly bind TLR2 and the positive controls, LipL32, was observed by the proteins identification and anti-LipL32 recognition ( Figure 1B) (27,29,30). This result demonstrated that the method used for identification of TLR2-binding candidates searching and identification is suitable and valid. An interesting TLR2 binding candidate, Loa22, was observed in the MALDI-TOF analysis and protein identification after co-immunoprecipitanting with TLR2. In order to confirm the MALDI-TOF results from co-immunoprecipitantition, an anti-Loa22 antibody was used to verify the interaction of Loa22 and TLR2. The western blot clearly demonstrated the interaction of Loa22 and TLR2 after co-immunoprecipitantion ( Figure 1B). Loa22 was present in pathogenic Leptospires but not in non-pathogenic Leptospires, indicating that Loa22 protein is probably a virulence factor (14). Loa22 is anchored to the outer membrane of pathogenic Leptospires and contains a large OmpA domain, known as a peptidoglycan-binding domain ( Figure 1A). Therefore, recombinant Loa22 (rLoa22) was constructed and expressed in E. coli to obtain purified rLoa22.

Protein Purification and Mutagenesis
The Loa22 protein contains 195 amino acids, and domain prediction indicated the N-terminal signal peptide and C-terminal OmpA domain ( Figure 1A). Sequence alignments of Loa22 with other OmpA domain proteins (Pal protein from E. coli and OmpA protein from A. baumannii) indicated that sequence similarity is low; however, two important PGN-binding residues, Asp 122 and Arg 143 , were highly conserved in these OmpA domain proteins ( Figure 1C). Therefore, these two residues were mutated to Ala using site-directed mutagenesis to generate D122A and R143A variants. The rLoa22 protein was expressed in E. coli ClearColi TM BL21 (DE3) pLys (Lucigen, Middleton, WI) and further purified by Ni 2+ -NTA affinity column and size exclusion chromatography ( Figure 1D-F). In order to remove E. coli endotoxin contamination, the MonoQ column and polymyxin B resin were used to further remove endotoxin from purified rLoa22 protein. In the Limulus amebocyte lysate (LAL) assay, rLoa22 from E. coli Clearcoli TM contained negligible endotoxin and was suitable for inflammation assays ( Figure 2G) (27). In addition, the N-terminal His 6 -tag tail was removed by Enterokinase and size exclusion chromatography, according to the previous report (27), in order to observe the obvious effects of rLoa22 protein without the His tag. In addition, rLoa22 protein and its related variants were treated at 100 o C for 30 min or digested by proteinease K and served as a negative control under several conditions.

PGN binding assay
Loa22 is a lipoprotein with a C-terminal OmpA domain, which is speculated to bind the essential cell wall component, PGN. To verify the PGN-binding activity of rLoa22, AFM was used to investigate the interaction between rLoa22 and LPGN. The Leptospira was immobilized on a mica surface and washed three times with PBS buffer containing 0.1% (w/v) Triton X-114 to remove the outer membrane and expose the PGN layer ( Figure S1A,B). The rLoa22-modified AFM tip was used to measure the affinity between rLoa22 and Leptospira cell wall. AFM force-distance curves were recorded to distinguish specific and non-specific interactions (Figure 2A). The specific interaction force-distance curves were selected to analyze interactions between rLoa22 and LPGN. In contrast, the tip only was used to measure the Leptospiral surface, as well as the rLoa22-modified AFM tip was used to measure affinity for the mica surface as a negative control. The interaction forces of the two controls were calculated as 26.3 ± 5.1 and 31.2 ± 4.7 pN, respectively (Supplemental S1C,D and Figure 2B). The interaction force between rLoa22WT and PGN from Leptospira was calculated as 58.2 ± 5.6 pN (Supplemental S1E and Figure 2B). In addition, the binding frequency between rLoa22WT and LPGN was calculated as 15.8% as compared to the mica surface as 2.1% ( Figure 2B).
This result clearly demonstrated the LPGN binding activity of rLoa22 from pathogenic Leptospira.
In addition, the rLoa22 mutation variants were prepared on AFM tip for PGN binding activity measurement. As expected, the interaction forces and binding frequency of D122A and R143A variants displayed low basal levels of LPGN binding activity that indicated the two residues played vital roles in LPGN binding (Supplemental S1F-G and Figure 2B). The two rLoa22 variants exhibited gross impairment in LPGN binding ability as compared to rLoa22WT, suggesting their crucial roles in maintaining of LPGN binding by rLoa22. To investigate the PGN binding activity of rLoa22 to other different PGN molecules, commercially available PGN molecules including those from E. coli (EPGN), S. aureus (SPGN), and B. subtilis (BPGN) were selected for incubation with rLoa22 protein at 37 o C for 30 min. In addition, the PGN molecule from pathogenic L.santarosai serovor Shermani (LPGN) was isolated as described in Materials and Methods to test its affinity for rLoa22 protein. After three centrifugation steps and washes, the pellets were subjected to SDS-PAGE and western blot analysis. The results indicated that rLoa22 showed high affinity for LPGN ( Figure   2C). The mutated variants of rLoa22, D122A and R143A, showed low affinity for the four types of PGN molecules, indicating that the two residues are important for PGN binding. Taken together, these results demonstrated that PGN molecules from pathogenic Leptospira with high affinity for rLoa22, while other PGN molecules exhibited relatively low affinity for rLoa22 protein ( Figure 2C). Therefore, LPGN was selected for subsequent studies.

rLoa22 Co-localizes with TLR2 on HEK293-TLR2 Cells
Attachment of Leptospira outer membrane proteins to host cell membrane is the first step invading the host during Leptospira infection. Previous studies showed that the Loa22 is up-regulation when host infection induces high levels of antibody production in the infected patient's serum (14, 31). However, the receptor on host cell membrane which recognizes Loa22 protein is still unknown and needs further investigation. The results mentioned above suggested that TLR2 is a possible receptor on host cell membranes for Loa22. In order to demonstrate the co-localization of rLoa22 and TLR2, purified rLoa22 protein and its variants were incubated with HEK293-TLR2 cells for 4h and the cells were then washed, fixed, and incubated with conjugated antibodies for confocal microscopy analysis (Figure 3). rLoa22 and rTLR2 proteins were stained with rabbit polyclonal anti-Loa22 and mouse monoclonal anti-V5 primary antibodies follow by Alexa594 (red) conjugated anti-rabbit and Alexa488 (green) conjugated anti-mouse secondary antibodies, respectively. HEK293 cells lacking TLR2 expression were used as negative controls, with very little or no Alexa 488 fluorescence ( Figure 3A). Co-localization of rLoa22WT-LPGN complexes and TLR2 receptors on HEK293-TLR2 cell was shown in Figure 3B. The results indicated that TLR2 receptor and the rLoa22 protein were mostly present on the cell surface, with consistent partial localization in the cytosol. The merged colors in several portions indicated that the two proteins were co-localized on HEK293-TLR2 cells ( Figure 3B). In contrast, the two mutated variants reduced the cell binding ability, and the red color was absent in the confocal images ( Figure 3C,D). The results from confocal microscopy clearly showed rLoa22-LPGN complexes directly interacted with TLR2 on HEK293-TLR2 cell surface, while the mutated variants, rLoa22D122A-LPGN and rLoa22R143A-LPGN, of rLoa22 significantly decreased co-localization with TLR2 on the cell surface.
To exclude any His 6 -tag effect on Loa22, the His 6 -tag tail at the N-terminus was removed by Enterokinase, and the non-tagged rLoa22-LPGN complexes also showed significantly increased expression levels of CXCL8/IL8, hCCL2/MCP-1, and hTNF-α in HEK293-TLR2 cells as compared to PBS control. The results indicated that rLoa22-LPGN complexes, with or without the N-terminal For functionalization, the AFM tips were incubated in a 1% (v/v) solution of glutaraldehyde (Grade II, Sigma) in PBS for 1 h at room temperature. For anchoring proteins, the tips were rinsed with PBS to remove the glutaraldehyde remainders and subsequently incubated with proteins, including wild type Loa22, its variants, and BSA, respectively, for 1 h at room temperature. The functionalized AFM tips were then rinsed three times with PBS to remove the unbound proteins (29, 30). The mica surface was modified for better deposition of proteins interested according to previous report (43,44).
In brief, the mica surface (1 cm 2 ) was functionalized with 0.1% (v/v) APTES (Sigma) (AP-mica) followed subsequently by 1% (v/v) glutaraldehyde (Grade II, Sigma) treatment. The proteins (100 ng) were then deposited covalently on the functionalized mica surface for 2 h at room temperature. The unbound proteins were removed and the proteins fixed on the mica were ready for later AFM measurements.

AFM Force-distance Curves Measurement and Analysis
The functionalized cantilever tips used in this work had a spring constant (k) in the range of 0.02-0.08 N/m as determined from the amplitude of their thermal vibrations. A commercial atomic force microscope (Nanoscope III, Digital Instruments, Santa Barbara, CA) with a J type scanner was employed throughout this study. The force volume software takes a force curve at each of the 999 points during a two dimensional scan over a sample surface. The X-Y scan size was 150 μm, and the Z scan distance was 5 μm at a rate of 1 Hz. The force applied to the protein modified mica surface was kept below 400 pN. The distance-force curves and force parameters were obtained according to the methods previously described (29,42). For antibody neutralization analysis, the protein modified mica was treated with anti-TLR2 antibodies (diluted 1:1,000 in PBS) for 1 h followed by five washes with PBS to remove unbound antibodies. The antibody against TLR2 (AbTLR2) was purchased from GeneTex (Irvine, CA). For the binding events, force curves from at least 10 areas on each surface were independently selected for analysis. All the measurements described above were performed with modified tips and showed repeatedly similar results. The force mapping data were transformed into force extension curves for each subsection, and the area under the force extension curvewas then calculated. For dynamic force analysis, the rupture speed was controlled within the range from 14 to 1400 nm/s. SPIP (ImageMetrology A/S,Hørsholm,Denmark)was used for data analysis, and values of extraction force were thus determined with SigmaPlot version 10.0 (SPSS, Chicago, IL). The maximum distribution force was calculated by a Gaussian fit to the force distribution curve. The dynamic profile for maximum distribution force versus the loading rate was obtained from equation where F* is the most probable rupture force, k B is the Boltzmann constant, T is the temperature, Δx is the potential width of the energy barrier along the direction of the applied force, k off is the natural dissociation rate at zero force, and F is the loading rate (45).

Ethics Statement
All animal procedures and experimental protocols were approved by the Institutional Animal Care

Founding Information.
This work was supported by the grants from CGMH-NTHU Joint Research CMRPG3E0331 to