Sample collection procedures for this study were approved by the Binzhou Medical University Ethics Committee (Shandong, China). All subjects provided written informed consent for the collection of samples and subsequent analysis. The Ethics Committees approved this consent procedure. This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of Binzhou Medical University. The protocol was approved by the Committee on the Ethics of Animal Experiments of Binzhou Medical University. All procedures were performed under sodium pentobarbital anesthesia, and all efforts were made to minimize suffering of the animals.

C57BL/6 wild-type mice were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China). LILRB4-deficient (LILRB4^−/−) C57BL/6 mice were obtained from Riken BioResource Center (Tsukuba, Japan). Subsequently, 6- to 8-week-old female mice were housed five per cage, and 8- to 10-week-old male mice were housed one per cage. All mice were maintained in the specific pathogen-free animal house of Binzhou Medical University at 22–26°C with 50–60% humidity and a 12 h light/12 h dark cycle, with abundant sterilized water and food (Jiangsu Biological Engineering Co., Ltd., China). One day after cohabitation with males (at a ratio of 2 females:1 male), female mice with vaginal plugs [gestational day (gd) 0] were segregated and randomized into uninfected, infected with T. gondii, and LILRB4^−/− infected with T. gondii groups.

Genomic DNA was extracted from mouse tails using a tissue DNA extraction kits (Generay, China). Polymerase chain reaction (PCR) was used to synthesize cDNA. After initial denaturation (3 min at 95°C), PCR was performed with 35 amplification cycles of denaturation for 30 s at 95°C, annealing for 30 s at 55°C, and extension for 60 s at 72°C, followed by a final extension for 5 min at 72°C. PCR products were separated by electrophoresis in 2% agarose gels, and sizes were estimated using Trans DNA marker I (100–700 bp; Transgene, France). Gels were stained with GelStain (10,000×; Transgene, France) to visualize DNA. Primers for PCR amplification were LILRB4 P1, 5′-ACCGGTGGATGTGGAATGTGTG-3′; LILRB4 P2, 5′-GTCCTGGGTTCCAGAATAAGAC-3′; and LILRB4 P3, 5′-TCTGCTCTTAGGAAATTACAGAA-3′.

The expected PCR product sizes were 260 bp (mutant), 371 and 260 bp (heterozygote), and 371 bp (wild-type). Homozygous LILRB4^−/− mice were continually bred for the duration of the study.

Toxoplasma gondii tachyzoites were maintained in HEp-2 cells in minimum essential media (MEM) (Hyclone, USA), 5% fetal bovine serum (FBS; Gibco, USA), and 100 IU/ml penicillin/streptomycin, which were purchased from Sigma-Aldrich (USA). After culture, tachyzoites were centrifuged at 1,500 rpm (433 × g) for 10 min, and purified tachyzoites were resuspended in MEM and counted using a Neubauer chamber.

Pregnant mice (either LILRB4^−/− infected or wide-type infected groups of mice) were inoculated intraperitoneally (i.p.) with 400 tachyzoites in 200 µl of sterile phosphate buffer (PBS) on gd 8. Uninfected groups of mice were inoculated with 200 µl of sterile PBS. All mice were sacrificed at 6 days post-infection (dpi), uteri were removed, and the total number of implantation and resorption sites counted. Resorption sites were identified by their small size and the necrotic and hemorrhagic appearance of the embryos. Placenta were compared with those of uninfected mice. Abortion rates were calculated as the ratio of resorption sites to total implantation sites (resorptions plus normal implantation sites).

Embryos and placenta were dissected and removed from mice at gd 14. Dispersed cells were prepared from small pieces of placenta and uterine tissue using a GentleMACS dissociator (Miltenyi, Germany). Single cell suspensions were then obtained by filtration through 48 µm sterile nets. After Ficoll density gradient centrifugation in mouse lymphocyte separation medium (TBD Science, China), mononuclear cells were collected from the white film layer and assessed by flow cytometry.

Decidual tissues were taken from women without evidence of clinical genital infections during pregnancy. All voluntary abortions occurred in the Department of Obstetrics and Gynecology, Yantai Hospital of Traditional Chinese Medicine and Yantai Affiliated Hospital of Binzhou Medical University between 8 and 10 weeks after conception. Sample collection protocols were approved by the ethics Committee of Binzhou Medical University. Tissues were rinsed with sterile saline solution 5–8 times, and decidual tissues were picked and cultured in Dulbecco’s modified eagle medium/high glucose medium (Hyclone, USA) supplemented with 100 IU/ml penicillin and 100 IU/ml streptomycin (Sigma-Aldrich, USA).

Decidual tissues were immediately washed 5–6 times in Roswell Park Memorial Institute (RPMI) medium, cut into pieces, and then digested in 0.1% collagenase type IV (Sigma-Aldrich, USA) solution containing 25 IU/ml DNase-I (Sigma-Aldrich, USA) at 37°C for 30 min. Single cell suspensions were obtained using a GentleMACS dissociator (Miltenyi Biotech, Germany) and were filtered through 48 µm nylon mesh filters. Density gradient centrifugation was then performed using human lymphocyte separation medium (TBD Science, China) at 2,000 rpm (771 × g) for 20 min at 20°C according to the manufacturer’s instructions for mononuclear cell collection. Decidual macrophages were then purified using a human CD14 positive selection kit (Stem Cell Science, USA) according to the manufacturer’s instructions with purity of >95% for all experiments. Human decidual macrophages were counted, and aliquots containing more than 1.5 × 10⁶ cells were allocated to uninfected, infected, and LILRB4-neutralized infected groups. Cells of the LILRB4-neutralized infected group were treated with neutralizing anti-LILRB4 antibody (10 µg/mL) for 1 h prior to T. gondii infection, which was performed at a ratio of 2:1, T. gondii to decidual macrophages. Samples containing approximately 5.0 × 10⁵ decidual macrophages were cultured in RPMI medium supplemented with 10% FBS (Gibco, USA), and 100 IU/ml penicillin and 100 IU/ml streptomycin (Sigma-Aldrich, USA) for 24 h at 37°C in a humidified 5% CO₂ incubator.

The following fluorochrome-conjugated, mouse-specific monoclonal antibodies (mAbs) were used for assessment: Pe-cy7-conjugated anti-F4/80 (marker of mouse-decidual macrophage), PE-conjugated anti-LILRB4, FITC-conjugated anti-CD206, FITC-conjugated anti-CD86, FITC-conjugated anti-CD80, APC-conjugated anti-TNF-α (all from BioLegend, USA), APC-conjugated anti-IL-10 (Becton Dickinson, BD, USA), APC-conjugated anti-iNOS (eBioscience, USA), and APC-conjugated anti-Arg-I (RD, USA). Isolated mononuclear cells were incubated with anti-F4/80, anti-LILRB4, and anti-CD206; or anti-F4/80, anti-LILRB4, and anti-CD80; or anti-F4/80, anti-LILRB4, and anti-CD86 mAbs at 4°C in the dark for 30 min and were then washed twice. To stain intracellular Arg-I and iNOS, cells were first incubated with anti-F4/80 and anti-LILRB4 mAbs at 4°C in the dark for 30 min and then washed. Subsequently, cells were fixed and permeabilized in 1× Fix/Perm buffer (eBioscience, USA) for 30 min according to the manufacturer’s instructions. After washing twice, cells were incubated with anti-Arg-I and anti-iNOS mAbs at 4°C in the dark for 45 min and were then washed twice. For analysis of the intracellular cytokines IL-10 and TNF-α, cells were initially stimulated for 4 h with leukocyte activation cocktail, with BD GolgiPlug (BD, USA). Cells were then collected and incubated with anti-F4/80 and anti-LILRB4 mAbs at 4°C in the dark for 30 min and then washed. The cells were then fixed and permeabilized in 1× Fix/Perm buffer (eBioscience, USA) at 4°C for 30 min according to the manufacturer’s instructions. After washing twice, cells were incubated with anti-IL-10 or anti-TNF-α mAbs at 4°C in the dark for 45 min and washed.

Assays were performed with the following fluorochrome-conjugated, human-specific mAbs: Pe-cy7-conjugated anti-CD14, APC-conjugated anti-LILRB4, and PE-conjugated anti-TNF-α, all from eBioscience, USA. FITC-conjugated anti-CD206, FITC-conjugated anti-CD163, FITC-conjugated anti-CD209, PE-conjugated anti-CD80, PE-conjugated anti-CD86, and PE-conjugated anti-IL-10 were from BD, USA. Similarly, decidual macrophages were incubated with anti-CD14, anti-LILRB4, and anti-CD206; or anti-CD14, anti-LILRB4, and anti-209, or anti-CD14, anti-LILRB4, and anti-CD163, or anti-CD14, anti-LILRB4, and anti-CD80, or anti-CD14, anti-LILRB4, and anti-CD86 mAbs at 4°C in the dark for 30 min and were then washed once. Decidual macrophages were also collected and incubated with anti-CD14, anti-LILRB4, and anti-IL-10, or anti-TNF-α mAbs (as described above for intracellular cytokine staining). Analysis was performed with a FACScanto™ II instrument (Becton Dickinson, USA).

Purified human decidual macrophages from uninfected, infected, and LILRB4-neutralized infected groups were cultured for 24 h after T. gondii infection, at a ratio of T. gondii to macrophage of 2:1. Cells were harvested from suspensions, and IL-10 and TNF-α levels were analyzed by ELISA according to the manufacturer’s protocols. Standard curves were generated using standards for each assay, and all measurements of absorbance were performed in triplicate at 450 nm. Concentrations were calculated according to standard curves and respective formulas.

Purified human decidual macrophages were harvested after treatment and washed with PBS, then lysed for 40 min on ice in 100 µl of lysis buffer containing phenylmethanesulfonyl fluoride at 16:1 (Beyotime Biotech, China). Cell lysates were centrifuged at 12,000 × g for 20 min at 4°C, and supernatants were mixed with 5× sodium dodecyl sulfate-polyacrylamide (SDS-PAGE) loading buffer and boiled for 5 min. Aliquots containing 40 µg of protein were separated on 12% (for detection of Arg-I) or 8% (for detection of iNOS) SDS-PAGE gels. Proteins were then transferred onto polyvinylidene fluoride membranes (Millipore, USA) and were blocked with 7% non-fat dry milk in Tris-buffered saline containing 0.2% Tween-20 (TBS-T) at room temperature for at least 4 h. Membranes were then incubated with rabbit anti-human arginase-I and iNOS antibodies (1:1,000 dilution; Abcam, USA) at 4°C overnight. Membranes were then washed four times in 1× TBS-T for 10 min and incubated with horseradish peroxidase conjugated goat anti-rabbit IgG secondary antibody (1:2,000 dilution; Protech, USA) at room temperature for 1 h. After washing three times for 15 min, proteins on membranes were visualized using an enhanced chemiluminescence kit (ECL; F. Hoffmann-La Roche Ltd., Switzerland) at various exposure times. GAPDH was detected using rabbit anti-GAPDH polyclonal antibody (Protech, USA) as a loading control.

Mouse placentas and uteri were removed after treatments and were washed three times and fixed with 4% paraformaldehyde immediately. Tissues were then washed in running water and placed in a graded ethanol series of 30, 50, and 70% and were then paraffin embedded using standard methods. Paraffin sections were stained with hematoxylin and eosin dye (H&E; Shanghai Novland Co., Ltd., China) according to the manufacturer’s instructions. Images of paraffin embedded sections were recorded at 20× magnification and are presented with 5 µm scale bars.

Purified, human decidual macrophages from uninfected, infected, and LILRB4-neutralized infected groups were cultured for 24 h after T. gondii infection and then fixed in 4% paraformaldehyde for 15 min and blocked with goat serum for 1 h. Cells were then incubated overnight at 4°C with anti-LILRB4 (Santa Cruz, Germany) and anti-CD163 (BD, USA), or anti-LILRB4 and anti-CD86 (BD, USA) antibodies. DyLight 488-goat anti-rabbit IgG (Abbkine, USA) was used as a secondary antibody for anti-LILRB4 antibody, and DyLight 649-goat anti-mouse IgG (Abbkine, USA) was used as a secondary antibody for anti-CD163 and anti-CD86. Cells were incubated with appropriate concentrations of secondary antibodies at 37°C for 1 h and were subsequently stained with the nucleic acid stain 4′,6-diamidino-2-phenylindole for 15 min. Finally, cells were observed using a laser confocal microscope (Leica, Germany).

Data are presented as means ± SD. Statistical analyses were performed using the GraphPad prism 5 statistics software package. Differences were identified using unpaired t-tests and were considered significant when two-tailed p values were less than 0.05 or very significant when two-tailed p values were less than 0.01.

Infected pregnant mice were unkempt, had less mobility, erect fur, and placentas that were significantly inflamed with hyperemia. Absorbed fetuses and stillbirths were more prevalent in infected mice than in uninfected pregnant mice (Figures 1A,B). Infected mice had significantly lower placental and fetal weights as well as higher abnormal fetal ratios than uninfected mice (Figure 1D). Paraffin embedded sections of infected but not uninfected placentas showed damage as hemorrhage and lymphocyte infiltration, with spiral arteries (Figure 1E). In LILRB4^−/− infected mice, pregnancy outcomes were worse than outcomes observed in infected wild-type mice, which included more severe placental ischemia. Relative to wide-type infected mice, LILRB4^−/− infected fetuses were almost shapeless (Figures 1B,C), with reduced placental and fetal weights and abnormal fetal ratios (Figure 1D). Hemorrhage and lymphocyte infiltration were further exacerbated in the LILRB4^−/− infected mice compared with infected wide-type mice (Figure 1E). Moreover, the percentage of decidual macrophages in all infected groups of mice was significantly higher than in the uninfected mice but did not differ significantly from that of the LILRB4^−/− infected mice (Figure 1F).

By immunofluorescence (Figures 2A,B) and flow cytometry (Figures 2C,D), LILRB4 expression levels were significantly decreased on human decidual macrophages after T. gondii infection, in comparison with uninfected macrophages. As judged by flow cytometry, murine decidual macrophage levels of LILRB4 were downregulated in mice infected with T. gondii (Figures 2E,F).

Flow cytometry demonstrated LILRB4 to be significantly downregulated on human decidual macrophages by T. gondii infection. M1 membrane-functional molecules CD80 (Figures 3A,B) and CD86 (Figures 3C,D) were significantly upregulated. In macrophages in which LILRB4 was neutralized, M1 membrane-functional molecules were further upregulated relative to infected cells (Figures 3A–D). M2 membrane-functional molecules CD163, CD209, and CD206 were significantly downregulated on human decidual macrophages after T. gondii infection and were slightly increased in the LILRB4-neutralized macrophages compared with infected macrophages (Figures 3E,F for CD163, Figures 3G,H for CD209, and Figures 3I,J for CD206). Furthermore, M1/M2 ratios (CD80/CD163, CD80/CD209, CD80/CD206, CD86/CD163, CD86/CD209, and CD86/CD206) were lowest in uninfected, greater in infected, and greatest in LILRB4-neutralized and infected macrophages (Figure 3K).

By immunofluorescence, infected macrophages had decreased LILRB4 and CD163 and increased CD86 expression when compared with uninfected macrophages. In the infected, LILRB4-neutralized macrophages, LILRB4 expression level was very low, whereas CD163 was slightly upregulated and CD86 was significantly upregulated in comparison with infected macrophages (Figures 4A,B). Similar to these in vitro studies, M1 membrane-functional molecules CD80 and CD86 were both significantly increased after T. gondii infection and increased further in LILRB4^−/− infected mice (Figures 5A,D for CD80, Figures 5B,E for CD86). In contrast, the M2 membrane-functional molecule CD206 was significantly decreased after T. gondii infection and was slightly increased in LILRB4^−/− infected mice when compared with infected mice (Figures 5C,F). CD80/CD206 and CD86/CD206 ratios were lowest in uninfected, greater in infected, and greatest in LILRB4^−/− infected mice (Figure 5G).

By western blot analysis, the arginine catabolism enzyme iNOS was not detected in uninfected human decidual macrophages. After T. gondii infection the enzyme increased. The increase was even greater in infected LILRB4-neutralized macrophages. Arg-I synthesis in human decidual macrophages was significantly reduced following T. gondii infection and was further reduced in infected LILRB4-neutralized macrophages (Figures 6A,B). In vivo flow cytometric analysis demonstrated iNOS in decidual macrophages from T. gondii infected mice, with greater detection in decidual macrophages from infected LILRB4^−/− mice (Figures 6C,D). Arg-I expression was decreased in decidual macrophages from infected mice and further decreased in decidual macrophages from LILRB4^−/− infected mice (Figures 6E,F).

By flow cytometry, TNF-α levels within human decidual macrophages were increased following T. gondii infection and were further increased in infected LILRB4-neutralized cells (Figures 7A,C). IL-10 levels within human decidual macrophages were increased by T. gondii infection and reduced in infected LILRB4-neutralized cells (Figures 7B,C). TNF-α/IL-10 ratios were greater in infected cells than in uninfected cells and were further increased in infected LILRB4-neutralized cells (Figure 7D). ELISA analysis showed increased TNF-α secretion in supernatants from infected human decidual macrophages, which was further increased in supernatants from infected LILRB4-neutralized cells. IL-10 in supernatants from human decidual macrophages was increased after T. gondii infection and decreased in supernatants from infected human LILRB4-neutralized macrophages, when compared with the infected cells (Figure 7E). TNF-α/IL-10 ratios were increased in infected cells compared with uninfected cells and were further increased in infected LILRB4-neutralized cells (Figure 7F) By flow cytometry, TNF-α was increased in decidual macrophages from infected mice and was further increased in decidual macrophages from infected LILRB4^−/− mice (Figures 8A,C). IL-10 secretion by mouse-decidual macrophages was increased by T. gondii infection, while IL-10 secretion was decreased in decidual macrophages from infected LILRB4^−/− mice (Figures 8B,C). Finally, TNF-α/IL-10 ratios were increased in infected decidual macrophages compared with uninfected cells. A further increase was observed in decidual macrophages from infected LILRB4^−/− mice (Figure 8D).