The anti-inflammatory limb of the immune response in preterm labor, intra-amniotic infection/inflammation, and spontaneous parturition at term: A role for interleukin-10

Abstract

Objective. The anti-inflammatory limb of the immune response is crucial for dampening inflammation. Spontaneous parturition at term and preterm labor (PTL) are mediated by inflammation in the cervix, membranes, and myometrium. This study focuses on the changes in the amniotic fluid concentrations of the anti-inflammatory cytokine interleukin (IL)- 10. The objectives of this study were to determine whether there is a relationship between amniotic fluid concentrations of IL-10 and gestational age, parturition (at term and preterm), and intra-amniotic infection/inflammation (IAI).

Study design. A cross-sectional study was conducted including 301 pregnant women in the following groups: (1) mid-trimester of pregnancy who delivered at term (n = 112); (2) mid-trimester who delivered preterm neonates (n = 30); (3) term not in labor without IAI (n = 40); (4) term in labor without IAI (n = 24); (5) term in labor with IAI (n = 20); (6) PTL without IAI who delivered at term (n = 31); (7) PTL without IAI who delivered preterm (n = 30); (8) PTL with IAI who delivered preterm (n = 14). IL-10 concentrations in amniotic fluid were determined by a specific and sensitive immunoassay. Non-parametric statistics were used for analysis.

Results. (1) IL-10 was detectable in amniotic fluid and its median concentration did not change with gestational age from mid-trimester to term. (2) Patients in labor at term had a significantly higher median amniotic fluid IL-10 concentration than that of patients at term not in labor (p = 0.04). (3) Women at term in labor with IAI had a significantly higher median amniotic fluid IL-10 concentration than that of patients at term in labor without IAI (p = 0.02). (4) Women with PTL and IAI who delivered preterm had a significantly higher median amniotic fluid concentration of IL-10 than those without IAI who delivered preterm and than those who delivered at term (p = 0.009 and p < 0.001, respectively). (5) Among patients with preterm labor without IAI, those who delivered preterm had a significantly higher median amniotic fluid IL-10 concentration than those who delivered at term (p = 0.03).

Conclusions. The anti-inflammatory cytokine IL-10 is detectable in the amniotic fluid of normal pregnant women. Spontaneous parturition at term and in preterm gestation is associated with increased amniotic fluid concentrations of IL-10. IAI (preterm and at term) is also associated with increased amniotic fluid concentrations of IL-10. We propose that IL-10 has a role in the regulation of the immune response in vivo by initiating actions that dampen inflammation.

Keywords:

• Pregnancy
• inflammation
• anti-inflammation
• IL-10
• cytokine
• amniotic fluid
• mid-trimester

Introduction

Pro-inflammatory cytokines, such as interleukin (IL)-1 [1-8], IL-6 [6-12], tumor necrosis factor (TNF) [5],[6],[13-18], IL-18 [19], IL-16 [20], and chemokines such as IL-8 [7],[8],[21-23], monocyte chemotactic protein-1 (MCP-1) [24],[25], macrophage inflammatory protein-1 alpha (MIP-1α) [26], RANTES [27], and epithelial cell-derived neutrophil-activating peptide-78 [28], are involved in the mechanisms responsible for preterm labor (PTL). Indeed, strong evidence suggests that these cytokines are produced by gestational tissues in response to microbial products [1],[29-33], are present in high concentrations in the amniotic fluid of patients with intra-amniotic infection/inflammation (IAI) [1],[3],[6],[9-11],[13],[14],[16],[19-22],[24],[26],[27],[34-38], stimulate uterine contractility by inducing the production of prostaglandins [39-50], and can induce PTL and delivery [2],[5],[51-53].

IL-10 is an anti-inflammatory cytokine that participates in a negative feedback loop to dampen inflammation [54]. Lymphocytes [55],[56], macrophages [57],[58], and dendritic cells [58] are the major sources of this cytokine, however, cytotrophoblast, syncytiotrophoblast [59],[60], and decidual mononuclear cells [61],[62] are additional sources during pregnancy.

A role for IL-10 in pregnancy is supported by the observations that IL-10 null mice (−/−) are more susceptible to lipopolysaccharide (LPS)-induced fetal loss [63], and that the administration of recombinant IL-10 to pregnant mice, rats, and rhesus monkeys challenged with LPS, Escherichia coli, or IL-1β, reduces the intra-amniotic concentrations of pro-inflammatory cytokines and prostaglandins, as well as the proportion of preterm fetal losses, preterm deliveries, and the severity of fetal white matter lesions when compared to non IL-10 treated controls [63-66]. In addition, some polymorphisms in the IL-10 gene appeared to confer susceptibility to preterm birth [67-69] and to neonatal inflammation-associated adverse outcome [70],[71].

IL-10 administration has been proposed as an anti-inflammatory agent in the treatment of bacterial sepsis [72], PTL [63],[65], and in the prevention of the neonatal brain injury in the context of IAI [73]. The objectives of this study were to determine whether there is a relationship between amniotic fluid concentrations of IL-10 and gestational age, parturition (at term and preterm), and IAI.

Materials and methods

Study design

This was a cross-sectional study designed to examine the relationship between amniotic fluid concentrations of IL-10 and gestational age, spontaneous labor at term, preterm parturition, and IAI.

Amniotic fluid was collected by trans-abdominal amniocentesis from 301 women with singleton pregnancies in the following groups: (1) Patients undergoing mid-trimester amniocentesis for clinical indications who delivered at term (n = 112); (2) women who had a mid-trimester amniocentesis and subsequently delivered a preterm neonate (n = 30); (3) women at term not in labor without IAI (n = 40); (4) patients at term in labor without evidence of IAI (n = 24); (5) women at term in labor with IAI (n = 20); (6) women with PTL and intact membranes without IAI at the time of amniocentesis who delivered at term (n = 31); (7) women in PTL without IAI who delivered preterm neonates (n = 30); (8) women in PTL with IAI who delivered preterm neonates (n = 14).

Clinical definitions

Patients were considered to have a normal pregnancy if they did not have obstetrical, medical, or surgical complications of pregnancy, and delivered a term (≥37 weeks) neonate with a birth weight above the 10^th percentile for gestational age. PTL was defined in the presence of regular uterine contractions occurring at a frequency of at least two every 10 minutes combined with documented cervical change, prior to 37 weeks of gestation. Intra-amniotic infection was defined by a positive amniotic fluid culture for microorganisms. Intra-amniotic inflammation was defined in the presence of an amniotic fluid white blood cell (WBC) count >100 cells/mm³.

Amniotic fluid sample collection and analysis

Amniotic fluid was collected by trans-abdominal amniocentesis and fluid not required for clinical purposes was centrifuged to remove cellular and particulate matter and the supernatant stored at − 70°C until analysis. Amniotic fluid WBC count, Gram stain, and glucose concentrations were used in the management of patients with PTL. Amniotic fluid samples from the term and preterm groups were cultured for aerobic and anaerobic species as well as genital mycoplasmas (Ureaplasma urealyticum and Mycoplasma hominis). Pregnant patients were enrolled at Hutzel Hospital, Detroit, MI, USA, Pennsylvania Hospital, Philadelphia, Pennsylvania, USA, and Sotero del Rio Hospital, Puente Alto, Chile. The collection of samples for research was approved by the Institutional Review Boards of Wayne State University, Pennsylvania Hospital, and Sotero del Rio Hospital, as well as the Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Department of Health and Human Services (NIH/DHHS). All participants provided written informed consent for collection of clinical data and biological materials. Many of these samples have been previously employed to study the biology of inflammation, hemostasis, angiogenesis regulation, and growth factor concentrations in non-pregnant women, normal pregnant women, and those with complications.

Human interleukin-10 (IL-10) immunoassay

Amniotic fluid concentrations of IL-10 were determined by specific and sensitive enzyme-linked immunoassays (Endogen Inc., Pierce Biotechnology, Rockford, IL, USA). Validation included spike and recovery experiments, which produced parallel curves indicating that amniotic fluid matrix constituents did not interfere with antigen–antibody binding in this assay system. The calculated inter- and intra-assay coefficients of variation for the IL-10 immunoassay in our laboratory were 2.8% and 4.4%, respectively. The lower limit of detection (sensitivity) was 2.6 pg/mL.

Statistical analysis

Data were tested for normality using the Kolmogorov–Smirnov test. Kruskal–Wallis test was used for analysis of variance. Comparisons between two groups were performed using Mann–Whitney rank sum tests. A p value < 0.05 was considered statistically significant. SPSS v.14.0 (SPSS Inc., Chicago, IL, USA) was used for analysis.

Results

Three hundred and one patients were included in this study. The demographic and clinical characteristics of patients in the different study groups are displayed in Table I.

Table I.  Clinical and demographic characteristics and amniotic fluid concentrations of IL-10 in the study groups.

Diagnostic group	Number of cases	GA amniocentesis (weeks)	GA delivery (weeks)	Maternal age (years)	Birth weight (grams)	AF IL-10 concentrations (pg/mL)
Mid-trimester amniocentesis, term delivery	112	16 (16–17)	39 (38–40)	36 (35–38)	3362 (3096–3632)	11.7 (10.7–13.4)
Mid-trimester amniocentesis, preterm delivery	30	16 (15.9–17)	33.5 (26.8–35.3)	36.5 (35–39)	2166 (1281–2846)	11.5 (10.2–12.9)
Term not in labor, no IAI	40	39 (38–40)	39 (38.4–40)	24 (21–30)	3260 (3023–3500)	10.4 (3.5–19.5)
Term in labor, no IAI	24	40 (39–40)	40 (39–40)	22 (18.3–27)	3560 (3173–3760)	19.3 (5.5–59.7)
Term in labor, IAI	20	39 (38–40)	39 (38.1–40)	25 (21.8–29)	3245 (3100–3600)	58.1 (25.3–151.0)
PTL, no IAI, delivered at term	31	31.9 (30.3–33)	39 (38–40)	21 (19–26)	3060 (2807–3350)	40.0 (32.9–50.9)
PTL, no IAI, delivered preterm	30	28.4 (25.8–31.6)	31 (29–33.2)	20.5 (18–23.5)	1618 (1200–2077)	59.9 (30.8–160.4)
PTL, IAI, delivered preterm	14	29.5 (25.8–32.6)	29 (25.8–33.2)	25 (20–29.3)	1250 (843–2049)	301.3 (68.6–504.8)

Values are expressed in median and inter-quartile ranges. GA, gestational age; AF, amniotic fluid; IL-10, interleukin-10; IAI, intra-amniotic infection/inflammation; PTL, preterm labor.

IL-10 is a physiologic constituent of amniotic fluid

IL-10 was detectable in 95% (286/301) of cases. Non-detectable IL-10 was found in 15 patients. One had PTL without IAI, six patients were at term and were not in labor (all without IAI), and eight patients were at term in labor (five of whom did not have IAI).

Amniotic fluid IL-10 concentration as a function of gestational age

There was no significant difference in the median amniotic fluid concentration of IL-10 between patients in the mid-trimester of pregnancy who delivered at term and patients at term not in labor at the time of amniocentesis (mid-trimester: median 11.7 pg/mL, range 8.6–22.9 vs. term not in labor: median 10.4 pg/mL, range 0–33.1; p = 0.1) (Table I and Figure 1).

Figure 1. Amniotic fluid concentrations of IL-10 in normal pregnancies. No significant differences were found in amniotic fluid median IL-10 concentrations between patients in the mid-trimester of pregnancy who delivered at term and patients at term not in labor at the time of amniocentesis (mid-trimester who delivered at term: median 11.7 pg/mL, range 8.6–22.9 vs. term no labor, no IAI: median 10.4 pg/mL, range 0–33.1; p = 0.1). (IAI = Intra-amniotic infection/inflammation; LOD = Lower Limit of Detection).

Spontaneous labor at term is associated with higher concentrations of IL-10

The median amniotic fluid IL-10 concentration was significantly higher in women in spontaneous labor at term than in those not in labor at term (term in labor: median 19.3 pg/mL, range 0–115.6 vs. term not in labor: median 10.4 pg/mL, range 0–33.1; p = 0.04) (Table I and Figure 2). These results suggest that spontaneous labor at term is associated with increased availability of IL-10. It is important to stress that patients in labor did not have evidence of intra-amniotic infection (all had negative amniotic fluid cultures for aerobic and anaerobic bacteria and low white blood cell counts).

Figure 2. Amniotic fluid concentrations of IL-10 in pregnant women at term. The median amniotic fluid IL-10 concentration was significantly higher in women in spontaneous labor at term than in those not in labor at term (term in labor: median 19.3 pg/mL, range 0–115.6 vs. term not in labor: median 10.4 pg/mL, range 0–33.1; p = 0.04). Women with spontaneous labor at term and evidence of intra-amniotic infection and/or intra-amniotic inflammation had a significantly higher median amniotic fluid concentration of IL-10 than women in spontaneous labor at term without IAI (term in labor, IAI: median 58.1 pg/mL, range 0–426.7 vs. term in labor, no IAI: median 19.3 pg/mL, range 0–115.6; p = 0.02). (IAI = Intra-amniotic infection/inflammation; LOD = Lower Limit of Detection).

IL-10 is increased in the amniotic fluid of women with intra-amniotic infection/inflammation

Women with spontaneous labor at term and evidence of intra-amniotic infection and/or inflammation had a significantly higher median amniotic fluid concentration of IL-10 than women in spontaneous labor at term without IAI (term in labor, IAI: median 58.1 pg/mL, range 0–426.7 vs. term in labor, no IAI: median 19.3 pg/mL, range 0–115.6; p = 0.02) (Table I and Figure 2).

Women with PTL and IAI had a significantly higher median amniotic fluid concentration of IL-10 than patients with PTL and intact membranes without IAI who delivered a preterm neonate (PTL with IAI: median 301.3 pg/mL, range 11.3–672.3 vs. PTL without IAI resulting in a preterm delivery: median 59.9 pg/mL, range 0–580.9; p = 0.009) (Table I and Figure 3).

Figure 3. Amniotic fluid IL-10 concentrations in patients with preterm labor (PTL). The median amniotic fluid IL-10 concentration was significantly higher in women with spontaneous preterm labor, without IAI and who subsequently delivered a preterm neonate than in those who delivered at term (PTL without IAI resulting in a preterm delivery: median 59.9 pg/mL, range 0–580.9 vs. PTL resulting in term delivery: median 40.0 pg/mL, range 13.2–139.8; p < 0.03). Women with PTL and IAI had a significantly higher median amniotic fluid concentration of IL-10 than patients with preterm labor and intact membranes without IAI who delivered a preterm neonate (PTL with IAI: median 301.3 pg/mL, range 11.3–672.3 vs. PTL without IAI resulting in a preterm delivery: median 59.9 pg/mL, range 0–580.9; p = 0.009). (IAI = Intra-amniotic infection/inflammation; LOD = Lower Limit of Detection).

IL-10 is elevated in women with preterm labor in the absence of intra-amniotic infection/inflammation

The median amniotic fluid IL-10 concentration was significantly higher in women with spontaneous PTL without IAI and who subsequently delivered a preterm neonate than in those who delivered at term (PTL without IAI resulting in a preterm delivery: median 59.9 pg/mL, range 0–580.9 vs. PTL resulting in term delivery: median 40.0 pg/mL, range 13.2–139.8; p = 0.03) (Table I and Figure 3).

Amniotic fluid IL-10 in the mid-trimester of pregnancy did not differ between women who had a preterm delivery and those who had a term delivery

There was no significant difference between the median amniotic fluid concentration of IL-10 in the mid-trimester of women who subsequently delivered preterm and those who delivered at term (mid-trimester amniocentesis who delivered a preterm neonate: median 11.5 pg/mL, range 8.8–18.8 vs. mid-trimester amniocentesis who delivered at term: median 11.7 pg/mL, range 8.6–22.9; p = 0.3) (Table I and Figure 4).

Figure 4. Amniotic fluid IL-10 concentrations in women in the mid-trimester of pregnancy. There was no significant difference between the median amniotic fluid concentration of IL-10 in the mid-trimester of women who subsequently delivered preterm and those who delivered at term (mid-trimester amniocentesis who delivered a preterm neonate: median 11.5 pg/mL, range 8.8–18.8 vs. mid-trimester amniocentesis who delivered at term: median 11.7 pg/mL, range 8.6–22.9; p = 0.3). (LOD = Lower Limit of Detection).

Discussion

Principal findings of this study

(1) IL-10 is detectable in amniotic fluid and its median concentration does not change with gestational age from mid-trimester to term. (2) Spontaneous labor at term is associated with increased amniotic fluid concentrations of IL-10. (3) Similarly, spontaneous PTL in the absence of IAI, is associated with an increased amniotic fluid concentration of IL-10. (4) IAI (at term or in PTL) is also associated with a significantly higher amniotic fluid IL-10 concentration than spontaneous preterm or term labor without IAI. Collectively, our findings suggest that IL-10 participates in the mechanisms of labor (term and preterm) as well as in the host response to intrauterine infection.

What is IL-10?

Human IL-10 is an 18 kDa cytokine [74],[75] encoded by a gene on chromosome 1 [76], originally described in 1989 as a factor produced by mouse Th2 clones (in response to concavalin A or antigen stimulation), which is capable of inhibiting the production of pro-inflammatory cytokines by activated Th1 cells [77]. For this reason it was termed ‘cytokine synthesis inhibitor’ or ‘cytokine inhibitory factor’[77].

IL-10 was subsequently isolated, expressed, and cloned [78],[79]. The mouse and human IL-10 genes are 90% homologous, and both proteins are non-covalent homodimers. Human IL-10 is minimally or not glycosylated, which is in contrast to its murine counterpart [77],[80].

The IL-10 cytokines family

It is now recognized that IL-10 belongs to a family of cytokines, which includes IL-19, IL-20, IL-22, IL-24, IL-26, and a series of herpesviral and poxviral proteins [81]. The IL-10-related cytokines share limited homology with IL-10 (25% identity and 50% similarity on average) [81], share several receptor subunits, and are expected to display similar functions [82], however further research is required to address their biological activities [82].

Sources of IL-10

Activated Th2 cells are the major source of IL-10 production [74],[77], and IL-10 has been classically considered a Th2 cytokine marker [83]. This was based upon both the observation that IL-10 was a product of Th2 mouse clones [77], and the classic Th-1 and Th-2 paradigm [84] with a mutually exclusive activation and cytokine production [85],[86]. According to this line of thought, Th1 cells produce interferon (IFN)-γ, IL-2, and other cytokines involved in the control of cell-mediated immunity (i.e., macrophage activation and delayed-type hypersensitivity responses) and Th2 cells produce IL-4, IL-5, IL-6, IL-9, IL-10, and IL-13 involved in humoral-antibody mediated responses and in the induction of B cell growth and differentiation into plasma cells [84],[87]. Recent evidence, however, suggests that IL-10 is produced by both Th1 and Th2 populations [55],[88],[89], stimulated B cells [56],[90-93], LPS-activated monocytes [57], macrophages [58], dendritic cells [58],[94-96], as well as regulatory T cells (T_reg) [97-102], which are involved in controlling both the innate and the adaptive immune responses [103-105] and in the prevention of immunopathology (self-antigens reactivity and/or exaggerated immune responses to pathogens) [99],[103],[104],[106],[107]. Of note, T_reg cells producing IL-10 can be developed under different regimens of antigen stimulation, both in vitro[97],[108] and in vivo[98],[109], particularly upon prolonged stimulation, which induces the appearance of ‘anergic’ T-cells, which fail to proliferate and express significantly higher levels of IL-10 [109]. There is evidence suggesting that IL-10 produced by T_reg cells is involved in controlling some immune responses in vivo[98],[110], including allergic and autoimmune pathologies [111-113], and in dampening inflammation in different organs such as the brain [111],[113], airways [112], and gut [110],[114].

Homology between IL-10 with viral products

It is of interest that some viruses, such as the Epstein–Barr virus (EBV) [78],[115] and the equine herpes virus type 2 [116], contain in their genome nucleotide sequences that are highly homologous to the human and mouse IL-10 genes [115],[116]. In particular, the recombinant protein of the gene ‘BamHI C fragment rightward reading frame’ (BRCF1), the open reading frame of the EBV genome, is similar in size to human IL-10, and mimics its activity on both macrophages and B cells [115]. These proteins (in addition to the human IL-10 expression induced by EBV-transformed B cells [56],[92]) could represent a mechanism through which viruses interact with the host's immune system to suppress the local antiviral immune response [56],[92], and favor viral survival by inhibiting the synthesis of macrophages and T cell cytokines with anti-viral properties [75].

IL-10 production by gestational tissues

During pregnancy, sources of IL-10 include cytotrophoblasts [59],[60], syncytiotrophoblasts [59], and chorion [117],[118], as well as decidual mononuclear cells/macrophages [61],[62],[119],[120] and decidual natural killer (NK) cells [120],[121]. IL-10 secretion from decidual mononuclear cells/macrophages has been reported to begin in early pregnancy [61],[120] and has been detected at term gestation [62]. The higher IL-10 secretion by decidual cells, when compared to the non-pregnant endometrium, may contribute to the suppression of the local cell-mediated immunity, to enhance embryonic implantation and favor pregnancy maintenance [122]. Hanna et al. [59] examined the expression profile of IL-10, of its receptor, as well as the expression profile of other cytokines (IL-2, IL-4, and IFN-γ) in placental explants and cytotrophoblasts. They reported that IL-10 is the predominant cytokine expressed in first and second trimester placental explants in normal pregnancies, and that its expression is gestational age-dependent, followed by a transcriptional attenuation as term gestation approaches [59].

IL-10: A cytokine that dampens inflammation

The resolution of the inflammatory response is crucial for survival. Although it was believed for many years that the inflammatory response would resolve passively, it is now clear that this is an active process requiring: (1) removal of the inflammatory stimulus, (2) decrease in the bioavailability of pro-inflammatory mediators and a switch to an anti-inflammatory cytokine milieu, and (3) removal of the inflammatory cells and debris allowing for tissue repair.

IL-10 is a pleiotropic anti-inflammatory cytokine involved in limiting and terminating inflammatory responses [54], recently defined as a ‘governor’ of inflammation [54],[83]. Its production by various T cell subsets after prolonged activation is important to prevent excessive inflammation [123]. Cytokines with pro-inflammatory properties such as IL-12, IL-27, and IL-6, in cooperation with transforming growth factor (TGF)-β, induce IL-10 production in various T helper cell subsets [124-128]. Of note, stimulation of macrophages and dendritic cells with LPS and toll-like receptor (TLR)-2 agonists, activates signaling pathways that converge on the IL-10 gene and induce its expression [83]. The production of IL-10 by stimulated human monocytes is relatively late, as compared to the production of IL-1α, IL-1β, IL-6, IL-8, tumor necrosis factor alpha (TNF-α), and granulocyte colony-stimulating factor (G-CSF) [57].

A growing body of evidence supports a role for IL-10 in the down-regulation of the immune response to protect the host from an exaggerated, even if effective, inflammatory response. Such evidence includes the following: (1) IL-10 knock-out mice infected with Trypanosoma cruzi and Toxoplasma gondii die rapidly because of a massive and sustained inflammatory response [129],[130], and are predisposed to have an exaggerated Th1 immune response in the context of Plasmodium chabaudi chabaudi malaria [131] and Listeria monocytogenes meningoencephalitis [132]. (2) In contrast, elevated concentrations of IL-10, or an imbalance between IL-10 and IL-12 in favor of IL-10, have been associated with persistent infection-induced inflammation, resulting in the development of invasive and chronic disease, such as human visceral leishmaniasis [133], chronic human Candida albicans infections [134], as well as paracoccidioidomycosis [135-140]. (3) Upon LPS challenge, IL-10 deficient mice display an uncontrolled production of TNF-α[141], are extremely vulnerable to the generalized Shwartzman reaction [141], and are more sensitive to lethal acute endotoxin shock (which occurs with a 20-fold lower dose of LPS) [141]. In contrast, mice receiving IL-10 (recombinant IL-10 injection) [72] and transgenic mice over-expressing IL-10 in their macrophages [142], are protected from a lethal intraperitoneal injection of endotoxin [72], and display a decreased production of pro-inflammatory cytokines [72],[142] (TNF-α[72],[142] and IL-12 [142]) when compared to controls.

The IL-10 cell of origin appears to have biological importance. IL-10 produced by T-cells is involved in the regulation of T cell responses, whereas IL-10 production from other cell types (i.e., macrophages, dendritic cells) is required to dampen the acute inflammatory response (normal innate responses to LPS or skin irritation) [143].

Of interest, IL-10 may participate in mediating immune-regulatory functions of galectin-1, a β-galactoside binding protein that appears to play a pivotal role in feto-maternal tolerance [144],[145]. Blois et al. [144] have recently reported that galectin-1 promotes the generation of tolerogenic uterine dendritic cells [144], and skews the balance toward a Th2 dominant cytokine profile at the feto-maternal interface [144]. The immunomodulatory effects elicited by galectin-1 are partly mediated by induction of the expression of IL-10. Evidence in support of this includes: (1) recombinant galectin-1 induces the production of IL-10 (mRNA and protein) by CD4 + and CD8 + cells [146]. (2) Incubation of peripheral blood mononuclear cells with stable galectin-1 homodimers (dGal) induces high production of IL-10, particularly from monocytes and T cells, but not by B cells [147]. (3) Administration of recombinant galectin-1 to null mutant mice prevents fetal loss and restores maternal–fetal tolerance through multiple mechanisms including the in vivo expansion of IL-10 secreting regulatory T cells [144]. (4) In IL-10 null mice, administration of galectin-1 does not significantly lower the number of stress-triggered fetal losses [144].

Galectin-1 expression has been previously demonstrated in the endometrium, decidua, and human placenta (villous stroma, endothelial cells, as well as syncytio- and cyto-trophoblast) [148-152]. Our group has recently demonstrated, by in situ hybridization and immunostaining, that galectin-1 mRNA and protein are abundantly expressed in all the cell types of fetal membranes (amnion epithelial cells, chorionic trophoblasts, mesenchymal cells) and decidual stromal cells in the third trimester [153]. Recently we reported a significantly higher galectin-1 mRNA expression (two-fold, p = 0.002) in the chorioamniotic membranes of patients with preterm prelabor rupture of the membranes (PPROM) and histological chorioamnionitis, compared to those without [153]. Thus, we propose that galectin-1 and IL-10 have a role in the resolution of the inflammatory response in chorioamnionitis.

Biological activities of IL-10

The anti-inflammatory properties of IL-10 are related to its down-regulation of T cell functions, which occurs mainly through an indirect mechanism involving the antigen presenting cells [154], as well as through a potent suppression of the effector functions of macrophages and NK cells, as well as T cells [80].

Some of the cellular targets and biological activities of IL-10 include the following:

Inactivation of monocytes/macrophages

IL-10 inhibits the production of pro-inflammatory cytokines by LPS-activated macrophages (IL-1α and IL-1β, granulocyte macrophage-CSF (GM-CSF), G-CSF, TNF-α, IL-6, IL-8, IL-10, and IL-12) [57],[155], and reduces their intracellular microorganisms killing potential with several mechanisms including the inhibition of TNF production [54],[129],[156], and preventing the release of reactive oxygen intermediates [77],[156]. IL-10 down-regulates the expression of major histocompatibility complex (MHC) class II antigens on LPS-activated monocytes [57], inhibits the macrophage co-stimulatory activity by selectively inhibiting the up-regulation of B7 expression [157], and decreases the expression of cell-adhesion molecules such as intercellular adhesion molecule (ICAM)-1 and B7 on monocytes [158].

Effects on maturation and functions of dendritic cells

IL-10 inhibits the maturation of dendritic cells from monocyte precursors [159],[160]. In addition, dendritic cells cultured with IL-10 are less efficient as antigen-presenting cells to CD4 + T cell clones [161] and have an increased capacity to capture antigens [161] but a concomitant decreased stimulatory activity [161]. Among the proposed mechanisms is the observation that IL-12 production by dendritic cells upon interaction with CD4 + T cells is down-regulated by IL-10 [162].

Suppression of Th1 cell responses

IL-10 inhibits the synthesis of the pro-inflammatory cytokine IL-12 by macrophages and dendritic cells [54],[162],[163]. Of note, IL-12 plays a central role in the ontogeny of a Th1 type immune response, directing IFN-γ production (a Th1-associated cytokine) rather than IL-4 production (Th2-associated cytokine) from T cells [75]. In addition, IL-10 is an inhibitor of the Th1 cell cytokine production (IL-2, IL-3, TNF, IFN-γ, and GM-CSF) [77]. It has been reported that the inhibition of Th1 cytokine synthesis by IL-10 is not complete, since it does not affect the early cytokine synthesis, whereas later synthesis is strongly inhibited [77].

Suppression of the effector function of natural killer (NK) cells

IL-10 inhibits the production of IFN-γ by NK cells in response to IL-2 and in the presence of accessory cells [164].

Regulation of the growth and differentiation of mast cells

IL-10, in synergy with other cytokines such as IL-3 and/or IL-4, enhances growth and proliferation of mast cells [165], and activates transcription of genes for two mast cell proteases, mucosal mast cell protease (MMCP)-1 and MMCP-2[166],[167].

Generation of T regulatory cells 1 (Tr1) from CD4 + T cells chronically activated

These regulatory cells are characterized by a low proliferative capacity, high production of IL-10, low levels of IL-2 and no IL-4 [97], and may, in turn, play a role in controlling immune responses and tolerance in vivo[54].

Protection of the functional integrity of inflamed tissues

IL-10 protects the epithelial monolayer integrity otherwise disrupted by IFN-γ treatment. This is accomplished by maintaining the size selectivity of the epithelial tight junction [168].

Amniotic fluid IL-10 in normal pregnancy

Previous studies of amniotic fluid IL-10 found no changes in the IL-10 concentration with advancing gestational age [169],[170]. In contrast, Greig et al. [171] reported a significantly higher median IL-10 amniotic fluid concentration in women at term than those in the second trimester (p < 0.001) [171]. We found no significant difference between the median IL-10 amniotic fluid concentration in women in the mid-trimester and that of women at term not in labor.

Of note, no gestational age-related changes in IL-10 concentrations have been described, not in maternal plasma (longitudinal study with sampling at 12, 20, and 35 weeks of gestation) [172], or in maternal serum (first and second trimester uncomplicated pregnancies) [173], despite a significantly higher concentration in pregnant than in non-pregnant controls [172].

The sources of amniotic fluid IL-10 are still unclear, and may include the mother, the fetus (i.e., keratinocytes, upon stimulation, can produce IL-10 [174],[175]), as well as a local production by gestational tissues (cytotrophoblasts [59],[60], syncytiotrophoblasts [59], chorion [117],[118], decidual stromal cells [176], mononuclear cells/macrophages [61],[62],[119],[120], and NK cells [120],[121]).

Amniotic fluid IL-10 in the mid-trimester

Among patients in the mid-trimester, there were no significant differences in the median amniotic fluid concentration of IL-10 between those who subsequently delivered at term and those who delivered preterm. Apuzzio et al. [177], in contrast to our findings, have previously reported higher concentrations of both IL-10 and IL-6 in second trimester amniotic fluid from patients who subsequently delivered preterm, as well as an inverse correlation between amniotic fluid IL-10 concentration at the time of mid-trimester genetic amniocentesis and gestational age at delivery [177].

IL-10 in spontaneous labor at term

Women in spontaneous labor at term have a significantly higher median amniotic fluid concentration of IL-10 than women at term not in labor. This finding is consistent with the view that labor at term is an event characterized by activation of the inflammatory cascade [178-180], as suggested by the evidence of inflammatory cells and a higher production of pro-inflammatory cytokines in the cervix, myometrium, chorioamniotic membranes, and amniotic cavity [179],[181], as well as by an up-regulation of inflammatory associated genes [182].

It is unlikely that chorion and decidua contribute to the additional pool of IL-10 that is detectable in amniotic fluid during labor at term. Indeed, previous studies focusing on gestational tissues have reported either no significant differences [59],[62],[183] or a significant reduction in IL-10 expression in the presence of labor when compared to controls [184-186]. Of note, Hanna et al. [59] reported a labor-associated up-regulation of TNF-α and IL-1 expression in placental explants and freshly isolated cytotrophoblasts in contrast to an absence of appreciable changes in IL-10 expression [59]. In contrast to our findings, two previous studies have reported no significant differences in amniotic fluid IL-10 concentrations between women at term not in labor and those in labor [170],[171]. Perhaps the larger sample size in our study is an explanation for the differences.

IL-10 in preterm labor

Among women with PTL who delivered preterm, those with evidence of IAI had a significantly higher median amniotic fluid concentration of IL-10 than those without IAI. However, PTL leading to preterm delivery, even in the absence of infection and inflammation, was associated with higher bioavailability of IL-10 in amniotic fluid. This suggests that an increase in amniotic fluid IL-10 is part of the common pathway of parturition because such a finding was also observed in spontaneous labor at term. Our interpretation is consistent with the view that labor is an inflammatory process [182],[187-189].

The sources and the mechanisms responsible for the higher IL-10 amniotic fluid pool in the presence of infection/inflammation-associated preterm birth are still unclear. There is evidence that IL-10 production is induced in gestational tissues in the presence of inflammation and that IL-10 may participate in dampening inflammation. Such evidence includes the following: (1) LPS injection into the cervix of timed-pregnant rats is followed by an increase in Th1 cytokines such as TNF-α, but also by a mild, although significant increase in placental IL-10 concentrations [190]. (2) Gestational tissues, stimulated with IL-1β, express higher levels of IL-10, both mRNA and protein [191], with the strongest response in decidua [191]. (3) IL-10 is involved in suppressing TNF-α production by human term choriodecidua stimulated by LPS [192].

Of note, there are conflicting reports on the effects of IL-10 on fetal membranes (amniochorion). Indeed, there is evidence supporting both that IL-10 exerts an anti-inflammatory activity on this target (suppressing the production of LPS-induced [193] and that of matrix metalloproteinases [194],[195], as well as the expression of IL-6 and IL-8 [196],[197] in the amniochorion), as well as evidence that IL-10 induces pro-inflammatory actions (including stimulation of IL-8 production [198]). Thus, the high IL-10 amniotic fluid concentrations in the context of PTL, particularly in the presence of IAI, may reflect part of the immune response to a local inflammatory process.

IL-10 may also have a role in regulating prostaglandin (PG) biosynthesis within gestational tissues, with most of the evidence in support of a down-regulation of PGE2 production including: (1) IL-10 suppresses the LPS-stimulated production of PGE2 by human choriodecidua [192]. (2) IL-10 diminishes the basal, as well as the LPS-induced, PGE2 production and expression of COX-2 mRNA by human fetal membranes [199]. (3) IL-10 treatment of preterm placental explants significantly inhibits both the intrinsic and the LPS-induced expression of COX-2 and PGE2 release [184]. (4) IL-10 reverses the PGE2 output induced by IL-1β on villous and chorion trophoblasts [200]. In contrast, in another study it has been reported that treatment with IL-10 significantly increases the amnion PGE2 production [198].

In a previous study, no significant differences in IL-10 amniotic fluid concentrations were reported between women with PTL and chorioamnionitis, women with preterm contractions (undelivered within 7 days), and those with PTL (delivered within 7 days) [170]. Of note, the definition of infection according to clinical rather than the histological criteria may have contributed to these findings. Some patients with absence of clinical signs of inflammation, in the presence of a subclinical and undetected histological inflammation, may have been classified as ‘not infection/not inflammation’.

Our findings of higher amniotic fluid concentrations of IL-10 in the presence of infection/inflammation-associated preterm birth are consistent with the observations of Greig et al. [171] who reported that patients with intra-amniotic infection (defined in the presence of a positive amniotic fluid culture or, alternatively, chorioamnionitis), have significantly higher amniotic fluid IL-10 concentrations than those of patients in PTL without infection (p < 0.001). In addition, the authors [171] reported that IL-10 amniotic fluid concentrations, in the infection-associated PTL, were significantly (p = 0.014) higher if this event occurred prior to the 30^th week of gestation than afterward. This finding has been interpreted as an attempt from the mother, or fetus, to prevent preterm birth when the fetal survival rate is low. Greig et al. [171], however, did not include in his study patients with isolated intra-uterine inflammation in the absence of infection, who are, in contrast, included in our study population.

Of note, IL-10 amniotic fluid concentrations in women experiencing an episode of PTL have been investigated as a potential tool to predict the interval to delivery. However, no significant differences in IL-10 amniotic fluid concentration (by enzyme-linked immunosorbent assay) have been reported between women with preterm contractions who delivered within 7 days of presentation and those women who delivered after this timeframe [170].

IL-10 polymorphisms and the risk of preterm birth and neonatal complications

PTL and PPROM are syndromes that result from genetic and environmental interactions. In particular, it has been hypothesized that the presence of polymorphisms in immunoregulatory genes may influence the susceptibility to PTL and delivery [67], as well as to fetal and neonatal complications [71]. This could be the case for the IL-10 gene; indeed, upon LPS stimulation, there are striking inter-individual differences in IL-10 production, and part of this variability is genetically determined [201],[202]. Of interest, the 5′-flanking region of the IL-10 gene contains several polymorphisms including microsatellite polymorphisms upstream of the transcription start site, and three single nucleotide polymorphisms (SNPs) at positions -1082 (G/A), -819 (C/T), and -592 (C/A). Only three haplotypes are found in the Caucasian population (GCC, ACC, and ATA) [201],[203-205].

Two haplotypes, in particular, have been the subject of intense investigation in the context of preterm birth, the IL-10 -1082A/-819T/-592A (IL-10 ATA)-haplotype [67], associated with low IL-10 production [201],[205] and the IL-10 -1082G/-819C/-592C (IL-10 GCC)-haplotype, which has been associated with a higher protein production [201],[205].

Annells et al. [67], in a study including 387 white pregnant women, examined the relationship between preterm birth and 22 SNPs in genes encoding cytokines, apoptosis mediators, and involved in the host defense. The IL-10 ATA-haplotype was independently associated with early preterm birth (<29 weeks of gestation; multivariable odds ratio (OR) 2.1; p = 0.4), as was homozygosity for the IL-10 590C allele (OR 3.4; p = 0.02). In addition, homozygosity for the IL-10 GCC-haplotype was more common in women with PPROM [67]. After Bonferroni correction for multiple comparisons, none of the genetic associations in the study were significant [67], but this has been possibly attributed to an over-correction.

Another polymorphism that has been the subject of investigation particularly in the context of infant risk of inflammatory-related perinatal complications is the IL-10 (-1082 G/A) polymorphism, in which a single nucleotide substitution of adenine for guanine at position 1082, in the promoter region, is associated with a higher production of IL-10 [205]. The presence of the IL-10 (-1082 G/A) polymorphism has been proposed to contribute to a reduced inflammatory response in fetuses and infants, thus decreasing the risk for adverse neonatal inflammation-associated outcomes [71]. Premature infants (gestational age at birth < 32 weeks) homozygous for the high IL-10 producer (-1082 G/A) allele: (1) were significantly less likely to develop ultrasound-defined periventricular echodensities than their heterozygous and non-carrier peers; (2) were less likely to suffer from prematurity-associated disorders such as bronchopulmonary dysplasia, high grade retinopathy, cerebral palsy, and developmental delay at age 2 + years. It has been proposed that such a polymorphism might contribute to a reduced inflammatory response in the fetus and infant, and subsequently decrease the risk for adverse neonatal inflammation-associated outcomes [71]. In another study including 294 patients, however, no major influences of the presence of this polymorphism were reported on prematurity-associated diseases [70].

However, the role of IL-10 polymorphisms in determining the predisposition to preterm birth is still the subject of debate. Indeed, in a study on the role of SNPs of inflammation-associated genes (IL-10, IL-1, TNF-α, and TLR-4), no overall differences in genotype and allele frequencies were detected between mothers who delivered preterm and at term [68]. Only in women with clinical chorioamnionitis and carriers of the IL-10 (1082)*G allele, was the risk for delivery before 29 weeks markedly increased (OR 22, 95% confidence interval (CI) 2.5–191) [68].

In another study, no differences in genotype frequencies of IL-10 polymorphisms were detected between women with a history of PTL < 37 weeks and women with successful pregnancies [69]. In addition, when multifetal gestations were evaluated, no relationships between the maternal IL-10 genotype and spontaneous preterm birth or PPROM were observed [206].

A key role for IL-10 in preterm labor

Endogenous IL-10 synthesis in gestational tissues may be of physiological importance in infection-induced PTL because of its anti-inflammatory role in deactivating macrophages and inhibiting their LPS-induced synthesis of inflammatory cytokines [63]. Evidence in support of a key role for IL-10 includes: (1) LPS administration to IL-10 null mutant mice resulted in an elevation of serum pro-inflammatory cytokines (including TNF-α and IL-6) and lower concentrations of the soluble TNF Receptor II compared to controls [63]. (2) IL-10 null mutant mice are more sensitive to LPS-induced PTL [207], and a 10-fold lower dose of LPS is required to elicit 50% preterm fetal loss when compared to the wild-type mice [63]. (3) Surviving fetuses in IL-10 null mice are more likely to display fetal growth restriction [63]. (4) Recombinant IL-10 administration, prior to LPS challenge, reduces serum concentrations of pro-inflammatory cytokines and the proportion of fetal loss and preterm delivered fetuses [63]. (5) IL-10 administration rescues pregnancy in LPS treated IL-10 (−/−) mice [207]. (6) Co-administration of IL-10 to pregnant rats, either at the time of LPS challenge or 24 hours later, resulted in term deliveries with no differences in litter size and birth weight compared with the controls [64].

The mechanisms through which IL-10 may delay the onset of PTL are still the subject of debate and may include a reduction of the local production of pro-inflammatory mediators and prostaglandin synthesis. This is supported by the following evidence: (1) IL-10 administration to pregnant rhesus monkeys receiving an intra-amniotic infusion of IL-1β, reduces the IL-1β induced uterine contractility, and is accompanied by a significant reduction in the amniotic fluid leukocyte count, prostaglandins, as well as TNF-α concentrations [65]. (2) IL-10 inhibits the IL-1β and TNF-α-induced production of PGE-2 by term human placental cells [208]. (3) IL-10 reverses the effects of IL-1β on prostaglandin synthesis and metabolism in purified cultures of villous trophoblast and chorion trophoblast cells from human term placentas [200].

IL-10 administration: Possible implications in the prevention of preterm birth-associated complications

Clinical trials in which IL-10 was administrated systemically to healthy volunteers have demonstrated that, at doses associated with biological activities, this cytokine has minimal side effects [209-213], including a mild flu-like syndrome [210], transient mild to moderate neutrophilia, monocytosis, lymphopenia (dramatic reductions of 40–70% in circulating lymphocytes expressing the T cell markers CD2, CD3, and CD7), and a delayed decrease of platelet count [209],[210]. Thus, IL-10 has been proposed as a candidate for treatment of bacterial sepsis, and more generally as an effective anti-inflammatory agent [72] and clinical trials have shown that its systemic administration is followed by modest but significant improvement of psoriasis [214],[215], Crohn's disease [216], rheumatoid arthritis [217], and chronic hepatitis C infection [218].

Infection during pregnancy and the perinatal period has been linked to cerebral white matter damage and cerebral palsy. The reader is referred to the review from Wolfberg et al. on the anti-inflammatory and immunomodulatory strategies to protect the perinatal brain [73]. A growing body of evidence suggests that IL-10 administration may modulate the inflammatory response leading to neonatal brain injury in the context of intra-uterine infection/inflammation. Indeed, brains of pups born to infected dams (intrauterine inoculation of E. coli) treated with IL-10 showed no evidence of severe white matter injury (frank cavitation, massive white matter necrosis) and a reduced rate of mild (evidence of histiocytes and rare apoptotic nuclei) and moderate (small vacuole formation in addition to histiocytes and rare apoptotic nuclei) lesion formation compared to controls not IL-10 treated [66].

When administered to pregnant rhesus monkeys, this cytokine gains access to the amniotic cavity where it harbors with a half-life of 13.2 hours [65], which is higher than the half-life of 2–4 hours in the circulation of volunteers following subcutaneous administration [210]. The longer half-life of IL-10 within the amniotic cavity indicates that the amniotic cavity may provide a depot-like effect prolonging the effectiveness of the treatment [65]. Treatment with IL-10 did not cause a change in the baseline fetal heart rate or in periodic fetal heart rate patterns and variability, as well as no change in maternal and/or fetal arterial pH and PO₂[65].

Suppression of macrophage/microglial activation has been proposed as a potential mechanism involved in the protective effect of IL-10 against maternal E. coli-induced neonatal white matter damage [219]. Alternatively, Sadowsky et al. [65] proposed that the intravenous administration of IL-10, despite its rapid decay, may be important in targeting decidual macrophages.

In conclusion, IL-10 concentration in amniotic fluid is elevated in spontaneous term and preterm labor (with and without infection/inflammation) and intra-amniotic infection/inflammation. This cytokine plays an important role in the control of labor and fetal and maternal inflammation.

Acknowledgement

This research was supported by the Intramural Research Program of the Eunice Kennedy Shriver National Institute of Child Health and Human Development, NIH, DHHS.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.