The lipoproteins of C. trachomatis were predicted using DOLOP (A database of bacterial lipo-proteins). The result can be find at http://www.mrc-lmb.cam.ac.uk/genomes/dolop/predicted/ct.shtml.

All chlamydial organisms were either purchased from ATCC (Manassas, VA) or acquired from Dr. Harlan Caldwell at the Rocky Mountain Laboratory, NIAID/NIH and Dr. Ted Kou at the University of Washington (Seattle, WA). The chlamydial organisms were propagated, purified, aliquoted, and stored as previously described (Zhong et al., 2001). All chlamydial organisms were routinely checked for mycoplasma contamination.

The ORFs of D381, D541, D067, and D775 from C. trachomatis serovar D were cloned into pGEX-6P-2 vectors (Amersham Pharmacia Biotech, Inc., Piscataway, NJ). According the prediction site of the lipobox, the target amino acid mutagenesis primers were designed and constructed into pGEX-6P-2 vector according the protocol of quick change ^R II site directed mutagenesis Kit. The following primers were used for cloning the ORFs:

The ORF was expressed as a fusion protein with glutathione- S- transferase (GST) fused to the N-terminuses as previously described (Wu et al., 2011). Expression of the fusion protein was induced with isopropyl-beta-D-thiogalactoside (IPTG; Invitrogen, Waltham, MA) and the fusion proteins were extracted by lysing the bacteria via sonication in a Triton-X100 lysis buffer (1% TritonX-100, 1 mM PMSF, 75 units/ml Aprotinin, 20 μM Leupeptin, and 1.6 μM Pepstatin; Sigma-Aldrich, St. Louis, MO). After removing cell debris by high-speed centrifugation, the fusion protein was purified using glutathione-conjugated agarose beads (Pfizer Inc., New York, NY).

TLR4^−/−, myD88^−/−, Tirap/Mal,^−/− Trif^−/−, and female Balb/c mice at the age of 3–4 weeks old were purchased from Charles River Laboratories, Inc. (Wilmington, MA). The isolated monocytes were seeded in 96-well plates (100 cells/μl) in DMEM supplemented with GlutaMAX-I (2 mM), streptomycin (100 μg/ml), penicillin (100 U/ml), and heat inactivated 10% (v/v) FBS (Invitrogen, Waltham, MA). After the cells were stimulated for 24 h at 37°C, the supernatants were collected by centrifugation (400 × g, 4°C, 10 min). The LPS (E. coli serotype O55:B5) supplemented in cell culture and triacylated control lipopeptide, racemic Pam3, were purchased from Sigma-Aldrich (St. Louis, MO). Affinity chromatography on polymyxin B agarose (Sigma-Aldrich, St. Louis, MO) was used to remove LPS from recombinant proteins.

Transfected HEK-293 cell lines stably expressing human TLR1/2, TLR2, TLR2/6, or TLR2/CD14 genes were gifts from Dr. Yan Xiang (University of Texas, Health Science Center at San Antonio). The cells carrying an empty plasmid, TLR1/2 plasmids, or TLR2/6 plasmids were cultured in high glucose DMEM (Dulbecco's Modified Eagle's Medium; Invitrogen, Gaithersburg, MD) with 10% FBS, streptomycin (100 μg/ml), penicillin (100 U/ml), and blasticidin S (10 mg/ml; InvivoGen, San Diego, CA). The cells carrying TLR2/CD14 plasmids were cultured in in high glucose DMEM supplemented with 50 mg/ml Hygromycin B Gold (InvivoGen, San Diego, CA). After 24 h culture in a 96-well microplate with an initial concentration of 500 cells/μl, the cells were co-cultured with E. coli LPS (1 μg/ml) or lipoproteins (1 to 0.001 μg/ml) for another 24 h. The lipoproteins were diluted in D10 medium (with 20 μg/ml gentamycin and with or without polymycin B 50 μg/ml) and incubated at 37°C for 1 h. The culture supernatants were collected for the ELISA analysis of IL-8 or mMIP-2 levels (DuoSet ELISA; R&D Systems, Inc. Minneapolis, MN). Results were obtained from cultures performed in triplicates.

Proteinase K sensitivity was tested by incubating D775 or D541 with 100 μg/ml proteinase K (Promega) in 100 mM Tris-HCl (pH 8.0) at 37°C for 2 h, followed by addition of 200 mM PMSF (phenylmethylsulfonyl fluoride). Lipoproteins D775 and D541 in a buffer with 50 mM HEPES (pH 7.5) and 10 mM CaCl₂ were incubated with1000 U/ml pig pancreas lipase (type VI-S lipase; Sigma-Aldrich) at 37°C for 16 h, followed by heating at 100°C before beginning the assay (Bas et al., 2008). The lipoproteins were also incubated in enzyme free buffer as controls. TLR4^−/− macrophage were incubated with lipoprotein with lipase, lipoprotein with protease, lipase alone, or protease alone. Cell-free supernatants were harvested after 4 h incubation for cytokine measurement.

The lipoprotein structures (D067, D381, D541, and D775) were initially built by homologous modeling using SWISS-MODEL (swissmodel.expasy.org). The initial models were then solvated in a water box and run 10,000 steps of energy minimization and 10 ns molecular dynamics using NAMD (NAnoscale Molecular Dynamics) v 2.9. The modeled lipoprotein-TLR complex structures were built by manually aligning the lipobox cysteine to the Pam3-cysteine based on the crystal structure of TLR2/TLR6- Pam2CSK4 complex (PDB code: 3A79).

Fourteen chlamydial lipoproteins were predicted from its genome. They are D067, D175, D253, D381, D386, D444, D541, D548, D567, D600, D734, D770, D775, and D780. The possible function and putative lipobox were also predicted (Table 1). Our previous animal experiments data suggested that CT381 (ABC transporter, solute binding pr; R-binding periplasmic pr 2), CT567 (Hypothetical pr), CT541 (MIP-like pr), CT067 (ABC transporter, solute binding pr; adhesin), CT775 (possible sn-Glycerol-3-P Acyltransferase) were related to the inflammatory reaction. The modeled structures of D067, D381, D541, and D775 were displayed in Supplementary Figure 1. Five proteins were chosen for expression and mutation to explore their roles in host innate immunity.

The mouse macrophage were stimulated using C. trachomatis recombinant lipoproteins, including D541, D541 mutant (D541C20A), D381, D381 mutant (D381C24A), D067, D067 mutant (D067C22A), D775, D775 mutants (D775C13A, D775C20A, and D775C13AC20A) as well as Hsp60, E. coli LPS, Pam3, and GST. Compared to GST and blank control groups, very high levels of mMIP-2 secretion were detected in treatment groups, especially in Hsp60 and D381 treatment groups (Figure 1A). The stimulation effect of LPS decreased by the addition of polymyxin B (PMB), as shown in Figure 1A. The results demonstrated that 50 μg/ml polymycin B could prevent E. coli LPS contamination. These data also indicated that the predicted lipoproteins induced proinflammatory cytokines in mouse macrophages.

When the assay were repeated using TLR4^−/− mouse macrophage, very few mMIP-2 was detected in LPS and Hsp60 treatment groups while high levels of mMIP-2 were detected in Pam3, D541, D381, D067, and D775 groups (Figure 1B). These result was consistent with previously reported data. However, in the groups of lipobox mutated proteins (D381C24A, D541C20A, D775C13A, D775C20A, and D775C13AC20A), proinflammatory cytokine levels were much lower than the wild type protein groups. It suggested that the stimulation by D541, D381, and D775 required the lipid modification on lipobox cysteine, which was consistent with the lipoprotein prediction results. Therefore, these data indicated that D541, D381, D067, D775 could induce proinflammatory cytokines in macrophages, which might not be only through TLR4 pathway. The stimulation by D541 and D381 required lipid modification. In the presence of lipoproteins, the secretion level of proinflammatory cytokines was correlated with the stimulator concentrations (Figure 1B).

Cytokine productions were measured for the HEK-293 cells expressing human TLR1/2, TLR2, TLR2/6, or TLR2/CD14 in response to C. trachomatis D541, D541 mutant, D381, D381 mutant, D067, D067 mutant, D775, and D775 mutants, as well as Hsp60, E. coli LPS, Pam3, and GST. As showed in Figure 2B, high levels of IL-8 were stimulated in TLR1/2 expressing cell line. However, much lower IL-8 levels were detected in their lipobox cysteine mutated proteins compared with that stimulated by wild-type proteins (Figure 2B), suggesting that TLR1/2 signal pathway was involved in Pam3, D541, D381, and D775 stimulations and the stimulation required the lipobox cysteine. There were very low cytokine levels detected in stimulation assays of TLR2 and TLR2/6 expressing cells (Figures 2A,B). In addition, TLR2/CD14 could be stimulated by D381 and Pam3. The results indicated that most of the C. trachomatis lipoproteins stimulated proinflammatory cytokines through the TLR1/2 pathway and D381 could also use TLR2/CD14 pathway.

To explore whether the lipid or protein part contributes to the proinflammatory activity of recombinant lipoproteins, D775 and D541 were first treated with proteinase K. Results showed that proinflammatory activity of D775 and D541 was maintained after proteinase digestion (Figure 3), which was similar to the result of a macrophage-activating lipopeptide from Mycoplasma fermentans (Mühlradt et al., 1996). However, lipase treatment of Chlamydial lipoprotein D775 led to dramatic decrease of mMIP-2 release after cell stimulation. These results confirmed that D775 required lipid modification to stimulate proinflammatory cytokine production. But our data indicated that D541 did not significantly influenced by lipase which was inconsistent with a previous report (Bas et al., 2008).

To determine whether the ability of proinflammatory cytokine production induced by D541 and D775 was in a MyD88-dependent manner, cytokine productions in MyD88 KO, Tirap/Mal KO, and Trif KO macrophages were stimulated by C. trachomatis recombinant lipoproteins (D541 and D775), LPS, and Pam3. When MyD88 were knocked out, mMIP-2 production decreased to an extreme low level in the stimulation assay of LPS, Pam3, D541, and D775 (Figure 4). It showed that the proinflammatory cytokines induced by D775 and D541 were through MyD88-dependent pathway. When Tirap/Mal were knocked out, mMIP-2 release remained high level after stimulation by Pam3 and D775 while the stimulation of D541 relied on both MyD88 and Tirap/Mal. The knock-out of Trif substantially decreased the stimulation effects of D541.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.