Toll-like receptor 2/4 inhibitors can reduce preterm birth in mice.

OBJECTIVES: Preterm birth (PTB) occurs in 5% to 18% of newborns. However, the underlying inflammatory mechanisms have not been elucidated.
METHODS: We established a mouse model of infection-associated PTB. Physical signs in pregnant mice with or without lipopolysaccharide (LPS) treatment were observed, and the frequencies of Toll-like receptor (TLR)2- and TLR4-positive CD11b+ cells were analyzed. Cytokine levels in plasma and pathological changes were assessed following LPS treatment. A rescue experiment was used to probe potential immunologic mechanisms underlying PTB.
RESULTS: Lymphocyte infiltration could be observed in the placentas of mice following intrauterine injection with LPS. The percentage of inflammatory cells decreased 12 hours after treatment. Moreover, TLR2 and TLR4 expression in peripheral blood cells was significantly increased 4 hours after intraperitoneal injection of LPS. Peak TLR2 and TLR4 expression in peripheral blood cells occurred 8 hours post-treatment. TLR4 and TLR-2/4 inhibitors reduced levels of interleukin-10, interferon-γ, and tumor necrosis factor-α in peripheral blood and delayed PTB.
CONCLUSIONS: TLR2 and TLR4 inhibition could play important roles in PTB.

Introduction

Preterm birth (PTB) in humans is defined as parturition before 37 weeks of gestation. According to the World Health Organization, PTB occurs in 5% to 18% of neonates.[1] PTB is one of the most common adverse pregnancy outcomes and is closely associated with neonatal morbidity and mortality.[2,3] Despite great advances in obstetrics and neonatology, the rate of PTB has not decreased over the past 20 to 30 years. There are still no effective therapies to prevent PTB and/or improve the physical condition of mothers and fetuses.

Decreasing PTB is a major goal in obstetrics and gynecology. However, lack of information regarding the detailed mechanisms of PTB is a significant impediment. Studies have identified various factors related to PTB including intrauterine infection, inflammation, malnutrition, and injury (4). Intrauterine infection and inflammation are considered the primary contributors to PTB: almost 40% of PTBs were attributable to intrauterine infection.[4] The pregnant mouse model can be used to investigate the mechanisms underlying PTB as well as to explore the potential interventions to avoid this outcome.[5] Toll-like receptors (TLRs) are closely related to inflammation and play important roles in innate immunity as pattern recognition receptors. TLRs play critical roles in innate immunity against pathogenic microorganisms, and can affect pregnancy outcomes directly or indirectly. In previous studies, collectins and inflammation cytokines, which are the products of TLR-related signaling pathways, were demonstrated to play critical roles in infection-induced PTB and preterm labor.[6,7] However, the detailed immunologic mechanisms related to parturition remain unknown. Therefore, understanding the immunologic mechanisms of PTB and the roles of TLRs could provide new directions for therapy and prevention of PTB.

To further elucidate the immunologic mechanisms in a mouse model of PTB, we investigated changes in lymphocyte subsets. C57BL/J6 pregnant mice were used to establish the PTB model. We examined changes in TLR2/4 expression in CD11b positive cells in peripheral blood. We also studied the functions of TLR2/4 inhibitors in this model. Our results provide valuable information toward prevention of PTB in humans.

Materials and methods

Animals

This study was approved by the Committee on the Ethics of Animal Experiments of the First Affiliated Hospital of Kunming Medical University (permit number: KMAE-081920). The Research Ethics Committee of the First Affiliated Hospital of Kunming Medical University approved the collection of tissue samples (2017/25XY82). All procedures were conducted in accordance with the principles laid out in the 1964 Helsinki Declaration.

Twelve-week-old pregnant mice (C57/BL6J strain) were purchased from the Institute of Laboratory Animal Science, Chinese Academy of Medical Sciences (Beijing, China) (certificate of conformity number: SCXK BJ 2014∼0002). Eighteen pregnant mice were divided into the placebo group (n = 3) and the lipopolysaccharide (LPS) treatment group (n = 15). We did not use non-pregnant mice in the control group because the innate immune responses of pregnant and non-pregnant women differ. In the LPS treatment group, the 15 mice were divided into five subgroups (three mice per group). Each subgroup was injected with LPS at different time points (4, 8, 12, 16, and 20 hours). Mice were maintained under pathogen-free conditions with a relative humidity of 40% to 50% and a temperature of 22 to 25°C in the Laboratory Animal Facility of the First Affiliated Hospital of Kunming Medical University. Mice were ad libitum access to food and water. One week of adaptive feeding was performed before the study. All animal procedures followed the guidelines of the Chinese Council for Animal Care.

LPS-induced mouse PTB model

A LPS-induced mouse PTB model was established according to a previous study.[8] On day 15 of gestation, the pregnant mice were injected with 100 µL of LPS solution (500 µg/kg) (Sigma-Aldrich, St. Louis, MO, USA) prepared in sterile phosphate-buffered saline (PBS) (Sigma-Aldrich).[9] The pregnant mice in the placebo group were received intrauterine injections with 100 µL of PBS. The pregnant mice were observed for eating habits, piloerection, decreased movement, vaginal bleeding, and PTB. Mice undergoing preterm labor or labor at term were separated when fetuses were found in the cage. The rates of PTB and vaginal bleeding were recorded. The PTB rate was determined using the following formula: PTB rate = total PTB mice/total pregnant mice × 100%.

Flow cytometry

Peripheral blood mononuclear cells (PBMCs) were isolated from 3 mL of venous maternal blood from different groups of mice using Ficoll-Paque PLUS centrifugation as described previously.[10] After centrifugation, PBMCs were collected from the interphase layer and washed four times with RPMI 1640 medium. PBMCs (1 × 107 cells/mL) were suspended in RPMI 1640 supplemented with 10% (v/v) fetal bovine serum. The number of PBMCs was counted using an Automated Cell Counter (Thermo Fisher, Waltham, MA, USA). About 106 PBMCs were resuspended in 100 µL of flow cytometry staining buffer (eBioscience, San Diego, CA, USA). The cells were incubated with fluorescein isothiocyanate-conjugated anti-mouse CD11b (eBioscience), phycoerythrin (PE)-conjugated anti-mouse TLR2 (eBioscience) and allophycocyanin-conjugated anti-mouse TLR4 (eBioscience) antibodies for 20 minutes at 4°C. After washing twice with staining buffer (eBioscience), stained cells were analyzed on a FACS AriaIII flow cytometer with FACS Diva software (BD Biosciences, La Jolla, CA, USA). Fifty thousand cells were analyzed for each sample. Fluorescently labeled isotype control antibodies were used to assess non-specific staining. All experimental protocols followed the manufacturer’s instructions.

Cytokine levels in plasma

Following different treatments, mice were euthanized by decapitation. Plasma was collected by eyeball extraction at 4, 8, 12, 16, 20 hours post-LPS injection. The plasma was stored at −80°C. Levels of seven cytokines [interleukin (IL)-2, interferon (IFN)-γ, tumor necrosis factor (TNF)-α, IL-4, IL-6, IL-17A, and IL-10) were analyzed using a cytometric bead array (CBA) mouse Th1/Th2/Th17 cytokine kit (Catalog No. 560485, BD Biosciences) and a BD Accuri C6 flow cytometer following the manufacturer’s instructions. The limits of detection provided by the manufacturer for the seven cytokines (IL-2, IFN-γ, TNFα, IL-4, IL-6, IL-17A, and IL-10) were 0.1 pg/mL, 0.5 pg/mL, 0.9 pg/mL, 0.03 pg/mL, 1.4 pg/mL, 0.8 pg/mL, and 16.8 pg/mL, respectively. In brief, plasma samples were thawed at room temperature and then placed on ice for the duration of the analysis. One vial of mixed standards was freshly reconstituted in 2.0 mL of assay diluent and then serially diluted. The concentrations of standards for each cytokine were 0, 20, 40, 80, 156, 312.5, 625, 1250, 2500, and 5000 pg/mL. Seven types of cytokine capture beads were freshly mixed in equal amounts (10 μL of beads per assay tube) in a master tube. To perform the assay, 50 μL of mixed beads were incubated with 50 μL of standards or samples along with 50 μL of PE Detection Reagent in a MultiScreen filter plate (1.2-μm pore size, EMD Millipore, Darmstadt, Germany) at room temperature for 2 hours. The plate was drained on a vacuum manifold. The beads in each of the individual wells of the plate were resuspended in 120 μL of wash buffer and analyzed on a BD Accuri C6 flow cytometer. The seven distinct fluorescent beads were sorted via signals captured in the FL4 channel. PE intensity of individual beads was captured in the FL2 channel. Approximately 200 events for each bead group were acquired (based on experience from previous experiments). The acquired data were subsequently analyzed to determine individual cytokine concentrations in each sample using FCAP Array software (BD Biosciences).

Immunohistochemistry

Mice uteri were stored in 10% neutral formalin before paraffin embedding (Sigma). Sections were deparaffinized in xylene (Sigma) and rehydrated in a series of decreasing ethanol concentrations [75%, 85%, 95%, 100% gradient in ethanol (Sigma)]. Non-specific binding sites were blocked using 2% goat, 2% horse, and 2% donkey serum (v/v) in PBS. A commercially available blocking serum was used for the detection of mouse primary antibody in mouse tissue (Mkb-2213, Vector Laboratories, Burlingame, CA, USA). Antigen retrieval was performed using 10 mM sodium citrate buffer, pH 6.0 (Sigma). Sections were stained using anti-CD25 (1:100, Abcam, ab128955, Cambridge, MA, USA) and anti-FOXP3 (1:1000, Santa Cruz Biotechnology, 2A11G9, Dallas, TX, USA) primary antibodies overnight at 4°C. Mouse anti-human CD45 (1: 100, Dako, 00087397, Santa Clara, CA, USA) antibody was used to stain lymphocytes. For FOXP3/CD25, horseradish peroxidase-conjugated anti-mouse and alkaline phosphatase-conjugated anti-rabbit (Biocare Medical, MRCT525H, Pacheco, CA, USA) secondary antibodies were used. For CD45, the secondary antibody was the Vector ABC R-T-U universal kit (Vector Labs, PK-7200, Burlingame, CA, USA). All experimental steps were performed strictly according to the manufacturers’ instructions.

Rescue experiment

For the rescue experiment, 27 mice were divided into the placebo group (n = 3), the TLR4 inhibitor group (n = 12), and the TLR2 + 4 inhibitor group (n = 12). In the TLR4 and TLR2 + 4 inhibitor groups, each subgroup (n = 3) was injected with the relevant inhibitor at different time points (0, 4, 12, and 24 hours). To block the functions of TLR2 and TLR4, 100 µL of TLR4 inhibitor (InvivoGen, San Diego, CA, USA) and 100 µL of TLR2+TLR4 inhibitor (Novus Biologicals, Littleton, CO, USA) were injected. An equivalent volume of PBS as placebo was injected in the placebo group. The time of preterm labor was observed and recorded. Meanwhile, mouse peripheral blood was collected for flow cytometry and CBA tests. All experimental procedures followed protocols provided by the manufacturers.

Statistical analysis

The PTB rates of different groups were analyzed using the log-rank test (survival analysis). Differences in cell percentages were assessed using the Student’s t-test. Results were presented as means ± standard deviations. Statistical analyses were performed using GraphPad Prism 5 (GraphPad Software Inc., La Jolla, CA, USA). Values of P<0.05 were considered statistically significant.

Results

LPS injection induced physical signs in pregnant mice

To investigate whether LPS injection induced physical signs, we observed pregnant mice injected with LPS. Various physical signs and changes were observed in LPS-treated mice. The fur of LPS-treated pregnant mice turned upside down and became scattered. A disturbed mental state, slowed movement, and cessation of eating and drinking were typical characteristics observed in LPS-treated mice. None of these signs were detected in the placebo group injected with PBS. In the LPS-treated group, 10% of pregnant mice had bloody show and membrane rupture 4 hours post-injection of LPS, and 80% of pregnant mice had bloody show and membrane rupture 8 hours post-injection of LPS. Bloody show and membrane rupture were present in all pregnant mice 12 hours post-injection of LPS. In addition, 60%, 80% and 100% of pregnant mice started labor 8, 12 and 16 hours post-injection of LPS, respectively (Table 1). These results suggested that LPS could induce physical signs in pregnant mice.

Table 1.

Physical signs in pregnant mice treated with LPS at different time points (N=30).

Physical sign	4 hours	8 hours	12 hours	16 hours	20 hours
Fur	Smooth	Lackluster	Scattered	Upside down	Scattered
Mental state	Move quickly	Move slowly	Apathy	Numb and huddle	Numb and huddle
Body temperature	36 ± 2°C	37 ± 3°C	37.5 ± 2°C	38 ± 3°C	37 ± 2°C
Drinking water	Decreased	None	None	None	Drinking water
Food	Decreased	Decreased	None	None	Taking food
Bloody show	10%	80%	100%	/	/
Membrane rupture	10%	80%	100%	/	/
Labor	0%	60%	80%	90%	100%

Expression of TLR2 and TLR4 in CD11b+ cells was altered in LPS-treated PTB mice

To study the expression of TLR2 and TLR4, PBMCs from LPS-treated PTB mice were analyzed by flow cytometry. The results of gating analysis are shown in Figure 1a. All available cells were gated for identification of CD11b+ cells. TLR2 and TLR4 positive cells were gated from the CD11b+ cells (Figure 1a, 1c). Pregnant mice treated with LPS at 4, 8, 12, 16, and 20 hours were analyzed. The percentage of CD11b+ cells in blood was significantly increased in LPS-treated mice 4 (P = 0.038), 8 (P<0.01), 12 (P = 0.035), 16 (P<0.001) and 20 hours (P<0.001) post-treatment compared with the placebo group (Figure 1b, Figure 2a). The percentage of CD11b+TLR2+ cells in blood was significantly higher in LPS-treated mice 4 (P<0.001), 8 (P<0.001), 12 (P = 0.011), 16 (P<0.001) and 20 (P<0.01) hours post-treatment compared with the placebo group (Figure 2b). The percentage of CD11b+TLR4+ cells in blood was significantly higher in LPS-treated mice 4 (P<0.001) and 8 (P<0.001) hours post-treatment compared with the placebo group. However, there were no differences in the percentages of CD11b+TLR4+ cells in blood 12, 16 and 20 hours post-LPS treatment compared with the placebo group (Figure 2c). These data suggested that the composition of immune cells was altered by LPS treatment in pregnant mice.

Figure 1.

Proportions of CD11b+, TLR2+CD11b+ and TLR4+CD11b+ cells in LPS-treated PTB mice and placebo mice. a: CD11b+ myeloid cells, CD11b+TLR2+ cells and CD11b+TLR4+ cells in the blood of PTB mice were analyzed by flow cytometry. b: Percentages of CD11b+ myeloid cells in the placebo group and LPS-treated group at different time points (4, 8, 12, 16, and 20 hours). FITC, fluorescein isothiocyanate. c: Expression of TLR2 and TLR4 by CD11b+ myeloid cells in the placebo group and LPS-treated group at different time points (4, 8, and 12 hours) (*P<0.05, **P<0.01, ***P<0.001). Each treatment group consisted of three experimental mice. Pregnant mice (n = 18) were divided into the placebo group (n = 3) and the LPS treatment group (n = 15).

Figure 2.

Temporal analysis of the percentages of CD11b+ myeloid cells, CD11b+TLR2+ cells, and CD11b+TLR4+ cells in PTB mice and placebo mice.

a: Proportion of CD11b+ cells in PTB mice 4, 8, 12, 16, and 20 hours post-LPS treatment. Peripheral blood in the placebo group was collected 4 hours post-PBS treatment. b: Proportion of TLR2+ cells among CD11b+ cells in the peripheral blood of PTB mice 4, 8, 12, 16, and 20 hours post-LPS treatment. Peripheral blood in the placebo group was collected 4 hours post-PBS treatment. c: Proportion of TLR4+ cells among CD11b+ cells in the peripheral blood of PTB mice 4, 8,12,16, and 20 hours post-LPS treatment. Peripheral blood in the placebo group was collected 4 hours post-PBS treatment. (*P<0.05, **P<0.01, ***P<0.001). Each treatment group consistent of three experimental mice. Pregnant mice (n = 18) were divided into the placebo group (n = 3) and the LPS treatment group (n = 15).

Cytokine concentrations were altered in LPS-induced PTB mice

To probe whether LPS treatment changed the pathological features of the mouse placenta, we used an immunohistochemical method. Figure 3a shows that leukocytes and regulatory T cells accumulated under the chorion of the placenta in LPS-treated mice after 4 hours. Meanwhile, the number of leukocytes was dramatically increased in LPS-treated after 8 and 12 hours. New capillaries were formed in the placenta with severe interstitial edema (Figure 3a). Nevertheless, the number of leukocytes returned to a normal level after preterm labor (20 hours following LPS treatment). We further invested the concentrations of IFN-γ, TNF-α, IL-6, IL-10, and IL-17A in blood. The concentrations of IFN-γ, TNF-α, IL-6, IL-10, and IL-17A 4 hours after LPS treatment were significantly higher in all PTB mice compared with the placebo group (P<0.01) (Figure 3b). Plasma levels of IL-10 could be only examined 4 hours post-LPS treatment because of kit detection limitations (detection limit for IL-10: 16.8 pg/mL). Levels of IFN-γ, TNF-α, IL-6, and IL-17A gradually decreased 8, 12, 16, and 20 hours post-LPS treatment. However, IL-2 concentrations in PTB mice were slightly lower 4 hours post-LPS treatment compared with the placebo group. No differences in IL-2 concentrations were detected in PTB mice 4, 8, 12, 16, and 20 hours post-LPS treatment. Moreover, 80% of pregnant mice underwent preterm delivery following LPS injection (Figure 3c, P = 0.002). By contrast, no PTB occurred in placebo group mice, all of which underwent normal labor (Figure 3c). These results revealed that cytokine concentrations were altered in LPS-treated PTB mice. In addition, LPS treatment was closely related to PTB.

Figure 3.

Uterine pathological sections, inflammatory cytokines and preterm labor time in LPS-treated PTB mice and placebo mice.

a: Uterine pathological sections of LPS-treated PTB mice and placebo mice at different time points (4, 8, 12, and 20 hours post-LPS treatment). FOXP3 (brown nuclear stain) and CD25 (red cytoplasm/membrane stain) staining is shown in the LPS-treated PTB mice and placebo mice (400×). Black arrows represent endometrial walls in the placebo group 12 hours post-treatment. Black arrows represent uterine villi in the LPS-treated PTB mice 12 hours post-treatment. b: Cytokine levels in the peripheral blood of PTB mice at different time points (4, 8, 12, 16, and 20 hours post- LPS treatment). Peripheral blood in the placebo group was collected 4 hours post-PBS treatment. c: Timing of preterm labor in PTB mice injected with LPS (*P<0.05, **P<0.01, ***P<0.001). Each treatment group consisted three experimental mice. Pregnant mice (n = 18) were divided into the placebo group (n = 3) and the LPS treatment group (n = 15).

Blocking TLR2 and/or TLR4 could delay PTB in LPS-treated mice

To study the relationships between TLR2 and/or TLR4 inhibitors and PTB, LPS-treated PTB mice were injected with a TLR4 inhibitor (CLI-095, InvivoGen) or a TLR2 + 4 inhibitor (OxPAPC, InvivoGen). The TLR4 and TLR2 + 4 inhibitors were injected at 0, 4, 12 and 24 hours. PTB rates in pregnant mice at different time points were calculated. We found that PTB was significantly delayed in the TLR2 + 4 inhibitor-treated group compared with the placebo group (treated with PBS containing dimethyl sulfoxide, P = 0.0476). However, there were no significant differences between the TLR4 inhibitor group and the placebo group (Figure 4a). Moreover, the proportion of CD11b+ cells in blood was assessed by flow cytometry. The proportion of CD11b+ cells was decreased in the TLR2 + 4 inhibitor-treated group compared with the placebo group (Figure 4b, P<0.01). Moreover, the percentages of CD11b+TLR2+ cells and CD11b+TLR4+ cells in the TLR2 + 4 inhibitor-treated group were significantly lower than those of the placebo group (Figure 4c, P<0.001). The percentage of CD11b+TLR4+ cells in the TLR4 inhibitor-treated group was significantly lower compared with the placebo group (Figure 4d, P<0.01). Histopathological evaluation indicated that the numbers of leukocytes and capillaries returned to normal levels in the placentas of TLR2 + 4 inhibitor-treated and TLR4 inhibitor-treated mice (Figure 4e).

Figure 4.

The effects of TLR4 and TRL2 + 4 inhibitors on CD11b+ cells.

a: Timing of PTB in LPS-treated mice administered TLR4 and TLR2 + 4 inhibitors; b: Proportion of CD11b+ cells in mice receiving different treatments; c: Proportion of CD11b+TLR2+ cells in mice receiving different treatments; d: Proportion of CD11b+TLR4+ cells in mice receiving different treatments; e: Histopatholo-gical evaluation of the numbers of leukocytes and capillary nets (400×). Each treatment subgroup consisted of three experimental mice. Pregnant mice (n = 27) were divided into the placebo group (n = 3) and the inhibitor treatment group (n = 24).

Discussion

Parturition involves multiple events, including inflammation.[11] Inflammation involves proinflammatory cytokine synthesis, leukocyte infiltration in fetal and maternal reproductive tissues,[12],[13] uterine contractions, and expulsion of the fetus. Therefore, devising interventions to prevent PTB requires an understanding of the physiological mechanisms of full-term birth. In late gestation, a previous study identified increased maternal systemic and local inflammation.[11] Related biological changes included phenotypic alterations of peripheral blood leukocytesand high expression of IL-1β, IL-6, TNF-α, and chemokines in the uterus, cervix, placenta, and fetal membranes.[13],[14] Studies in rodent and primate models indicated that the expression of inflammatory cytokines was elevated before active labor, suggesting that a shift to an inflammatory state occurred in the preparation phase of parturition.[15],[16] In mice, this shift is accompanied by the withdrawal of the protective anti-inflammatory actions of progesterone.[17] In this study, our results suggested that LPS treatment could induce physical signs in pregnant mice. In addition, the expression of TLR2 and TLR4 in CD11b+ cells was altered in LPS-treated PTB mice.

TLRs are strong candidates for activating inflammation during both term and preterm labor.[11] A previous study revealed important roles for TLR2 (binding to bacterial lipoprotein) and TLR4 (binding to bacterial LPS).[18] High expression of TLR2 and TLR4 was evident in chorioamnionitis membranes in preterm labor.[19] Moreover, other evidence supported the notion that blocking TLR4 function could prevent infection-associated PTB. For example, blockade of the TLR4 signaling pathway with an anti-TLR4 monoclonal antibody could reduce leukocyte activation.[20] Therefore, TLR4 inhibitors have been reported to reduce the incidence of preterm labor induced by LPS.[21] For example, administration of a TLR4 antagonist (the lipid A mimetic CXR-526) could effectively reduce fetal loss. A previous study suggested that Fusobacterium nucleatum, a Gram-negative bacterium, was associated with PTB and premature rupture of membranes in women.[22] CXR-526 did not suppress bacterial colonization of the placenta. However, it reduced the extent of necrosis within placental tissue.[23] Therefore, decreased placental necrosis might be related to TLR4 antagonist-mediated suppression of pro-inflammatory cytokine production. Moreover, a previous study suggested that CXR-526 was effective against inflammatory bowel disease in a mouse model. However, cytokine changes were not studied.[24] Other studies demonstrated that downstream effectors of TLR4-driven inflammation were targets for inhibiting infection-induced PTB.[25] Experiments in IL-1 and IL-6 null mice showed that the absence of these cytokines could amplify and accelerate the birth cascade.[26],[27] Based on the evidence mentioned above, we selected TLR4 as a target to study its potential role in preterm labor induced by LPS. In addition to cytokines, immune regulators that attenuate TLR-mediated inflammation might affect LPS-induced PTB. For example, the cannabinoid receptor Cnr2 modulates dendritic cell production of IL-10 and IL-6. Therefore, mice deficient in Cnr2 are resistant to LPS-induced PTB.[28] Prostaglandins synthesized following Ptgs2 induction triggered by LPS administration contribute to cervical ripening and other endpoints of parturition. Therefore, suppression of Ptgs2 with specific inhibitors can modulate LPS-induced PTB.[29] Pre-treatment with rosiglitazone could promote placental synthesis of peroxisome proliferator-activated receptor-γ, reducing the synthesis of TNF-α, IL-1, IL-6 and other pro-inflammatory chemokines.[30] Our results suggested that cytokine concentrations were altered in LPS-induced PTB mice. Concentrations of IFN-γ, TNF-α, IL-6, IL-10, and IL-17A were significantly higher 4 hours or longer post-LPS treatment in all PTB mice compared with the placebo group, consistent with previous studies.[31] Moreover, TLRs play a role in responding to endogenous mediators of tissue damage or injury. Surfactant protein A has a similar effect on THP-1 cells but this is mediated via TLR-2.[32] Heat shock protein 70 (HSP70), gp96 and HSP60 increase the production of proinflammatory cytokines through TLR-2.[33] The ability of cells to respond to endogenous mediators of inflammation is consistent with the “danger” model of immunity.[34] Studies of pregnant mice revealed that TLR-2 and TLR-4 are expressed throughout the second half of pregnancy.[35],[36] The human term placenta also expresses TLR-2 and TLR-4 in the syncytiotrophoblast, which would be in direct contact with invading pathogens.[33] Although expression of TLR-2 and TLR-4 was not increased in the human placenta during labor, these receptors do seem to be functional because IL-8 and TNF-α secretion by placental explants was increased after incubation with zymosan (a TLR-2 agonist) or LPS (a TLR-4 agonist).[37] The above results are consistent with our data. Inhibitors of TLR-2 and/or TLR-4 can affect immune-related PTB.

Our results indicated that PTB in pregnant mice could be induced by LPS. The proportions of CD11b+ cells, CD11b+TLR2+ cells, and CD11b+TLR4+ cells in blood were significantly higher in LPS-treated mice compared with placebo-treated mice. When TLR4 or TLR2 + 4 inhibitors were used to block TLR2 and/or TLR4 signaling, the percentages of these immune cells were down-regulated and PTB was significantly reduced. Therefore, TLR2 and TLR4 may be potential therapeutic targets for improving adverse outcomes in pregnant women. In summary, our study demonstrates that inhibitors of TLR2 and TLR4 had notable effects on LPS-induced PTB in a mouse model.