Skip to main content
NCBI home page
Reproductive Sciences logoLink to Reproductive Sciences
. 2018 Oct 8;26(4):551–559. doi: 10.1177/1933719118804420

Block of Granulocyte-Macrophage Colony-Stimulating Factor Prevents Inflammation-Induced Preterm Birth in a Mouse Model for Parturition

Christopher Nold 1,2,✉, Julie Stone 2, Kathleen O’Hara 2, Patricia Davis 2, Vladislav Kiveliyk 2, Vanessa Blanchard 3, Steven M Yellon 3, Anthony T Vella 2
PMCID: PMC6421621  PMID: 30296925

Abstract

Objective:

A multitude of factors promotes inflammation in the reproductive tract leading to preterm birth. Macrophages peak in the cervix prior to birth and their numbers are increased by the cytokine granulocyte-macrophage colony-stimulating factor (GM-CSF). We hypothesize GM-CSF is produced from multiple sites in the genital tract and is a key mediator in preterm birth.

Study Design:

Ectocervical, endocervical, and amniotic fluid mesenchymal stem cells were treated with lipopolysaccharide (LPS), and the concentration and expression of GM-CSF was measured. Pregnant CD-1 mice on gestational day 17 received LPS and an intravenous injection of either anti-mouse GM-CSF or control antibody. After 6 hours, the preterm birth rate was recorded.

Results:

Treatment with LPS increased the GM-CSF concentration and messenger RNA expression after 24 hours in all 3 cell lines (P < .01). Mice treated with LPS and the GM-CSF antibody had a preterm birth rate of 25%, compared to a 66.7% preterm birth rate in controls, within 6 hours (P < .05, χ2). Treatment with the anti-mouse GM-CSF antibody decreased the concentration of GM-CSF in the mouse serum (P < .01) but did not alter the number of macrophages or collagen content in the cervix.

Conclusion:

These studies demonstrate that GM-CSF is produced from multiple sites in the genital tract and that treatment with an antibody to GM-CSF prevents preterm birth. Curiously, the anti-mouse GM-CSF antibody did not decrease the number of macrophages in the cervix. Further research is needed to determine whether antibodies to GM-CSF can be utilized as a therapeutic agent to prevent preterm birth.

Keywords: preterm birth, cervical remodeling, inflammation

Background

In the United States, preterm birth remains the leading cause of neonatal morbidity and mortality,1 and worldwide complications from preterm birth result in 3.1 million neonatal deaths annually.2 Despite the short- and long-term health consequences of preterm birth and the sizeable health-care costs, research efforts thus far have resulted in marginal improvements in the preterm birth rate. This lack of progress reflects a poor understanding of the mechanisms and molecules that mediate preterm birth. Spontaneous preterm birth is considered a complex syndrome with various etiologies ranging from decidual hemorrhage to uterine distension, but the most frequently cited etiology appears related to a pro-inflammatory process from pregnancy-related reproductive tissues.3,4

Macrophages are innate immune cells that contribute to preterm labor. Macrophages play a role in the rupture of fetal membranes, as they have been shown to produce matrix metallopeptidases 9 (MMP-9), an enzyme increased in fetal membranes in preterm labor and preterm premature rupture of membranes (PPROM).5–8 Additionally, macrophages produce vasoactive molecules (nitric oxide and prostaglandins) and pro-inflammatory cytokines which lead to edema and the release of MMPs by the fetal membranes leading to PPROM and prepartum changes in the reproductive tract.9 Similarly, in rodents, the number of macrophages in the cervix peaks prior to birth, a period where the cervix softens and ripens and progesterone in the circulation is at or near peak concentrations.10–12 This increased presence or activity of macrophages leads to the increase in leukocyte collagenases, which are thought to promote prepartum extracellular matrix remodeling during the transition from a soft to ripe cervix.12 Therefore, based on the increased presence and activity of macrophages prior to birth during cervical ripening, macrophages may play an essential role in preterm birth.

Granulocyte-macrophage colony-stimulating factor (GM-CSF), a monomeric glycoprotein, may also be involved in the parturition process by stimulating stem cells to produce monocytes and promoting their maturation into macrophages in the tissue. In pathophysiologic circumstances, GM-CSF is secreted during a pro-inflammatory response by immune cells, endothelial cells, and fibroblasts.13 Mature tissue-resident macrophages also produce GM-CSF to amplify the local inflammatory process, promote the additional recruitment of immune cells and mature tissue-resident macrophages, and stimulate the maturation of precursor cells.14

Prior data have shown GM-CSF is linked with preterm birth. For example, serum GM-CSF concentrations elevated in women during the first trimester are at increased risk of preterm delivery.15 The GM-CSF concentrations are also increased in vaginal swabs obtained from patients with a short cervix, a risk factor for spontaneous preterm birth.16 Similarly, in an inflammation-induced mouse model for preterm birth, serum GM-CSF concentrations increase within 6 hours of treatment.17 Together, these studies suggest increased prepartum concentrations of GM-CSF are associated with spontaneous preterm birth.

Thus, to determine whether GM-CSF has a role in the mechanism of inflammation-initiated preterm labor and preterm birth, a series of experiments using multiple different reproductive tissue–derived cell lines were used to determine potential sites of GM-CSF production. To determine whether GM-CSF plays a direct mechanistic role in controlling preterm birth, an inflammatory mouse model was used to test if a therapeutic blockade of GM-CSF inhibits critical features of cervical remodeling and prevents preterm birth.

Methods

Cervical Cell Culture and Preparation

Immortalized HPV 16/E6E7 ectocervical (ATCC CRL-2614) and endocervical (ATCC CRL-2615) cells (American Type Culture Collection, Manassas, Virginia) were maintained in keratinocyte serum-free medium (GIBCO BRL Life Technologies, Gaithersburg, Maryland) supplemented with 50 μg/mL bovine pituitary extract (BPE), 0.1 ng/mL epidermal growth factor (EGF), 100 U/mL penicillin, and 100 μg/mL of streptomycin (Life Technologies, Grand Island, New York) at 37°C in a 5% CO2 humidified incubator.

Amniotic Fluid Mesenchymal Stem Cell Culture and Preparation

Human amniotic fluid mesenchymal stem cells (AF-MSCs) were studied to determine whether cells within the amniotic cavity have the potential to produce GM-CSF. Briefly, AF-MSCs were obtained from patients undergoing amniocentesis at Hartford Hospital (Hartford Hospital IRB #FINC003364HU). The discarded sample (∼5 mL) was collected and AF-MSCs were manually isolated by the published starter cell method.18

The AF-MSCs were maintained in AF media, which consists of minimal essential medium with alpha modifications (Life Technologies) containing 18% Chang B medium (Irvine Scientific, Santa Ana, California), 2% Chang C medium (Irvine Scientific), 20% fetal bovine serum (Atlanta Biologicals, Lawrenceville, Georgia), 2 mM l-glutamine (Life Technologies), 100 U/mL penicillin, and 100 μg/mL streptomycin (Life Technologies) at 37°C in a 5% CO2 humidified incubator. The AF-MSCs utilized were less than passage 10.

Detection of Soluble Immunobiological Mediators

Ectocervical and endocervical cells were plated at a concentration of 9.6 × 104 cells/well (1.07 × 104 cells/cm2) on 6-well plates for 24 hours in keratinocyte serum-free medium containing BPE, EGF, penicillin, and streptomycin. Cells were then treated with keratinocyte serum-free media containing only penicillin and streptomycin for an additional 24 hours prior to lipopolysaccharide (LPS) treatment. The AF-MSCs were plated at a concentration 1 × 104 cells/cm2 in AF media for 24 hours. All 3 cell lines were treated with 25 μg/mL of LPS strain Escherichia coli 055:B5 (LPS; Sigma Chemical Company, St Louis, Missouri; N = 3-4 per treatment group in all 3 cell lines) for 24 hours. Media were collected and assessed for GM-CSF using an enzyme-linked immunosorbent assay (ELISA; R&D Systems, Minneapolis, Minnesota) and analyzed according to the manufacture’s protocol. RNA was collected for quantitative real-time polymerase chain reaction (qRT-PCR) for expression of GM-CSF.

Mouse Model of Intrauterine Inflammation

All procedures were performed with Institutional Animal Care and Use Committee approval from the University of Connecticut School of Medicine. As previously reported, this model mimics the clinical scenario of an inflammatory response in the uterus leading to a spontaneous preterm birth and does not result in maternal mortality.19 CD-1 timed pregnant mice (Charles River Laboratories, Wilmington, Massachusetts) received an intrauterine injections of LPS (E coli, 055:B5, Sigma Chemical Co) at a dose of 250 µg in 100 µL of phosphate-buffered saline (PBS) on embryonic day 17 (E17). The treatment day of E17 was chosen to mimic the typical clinical presentation of intrauterine inflammation associated with presentation of cervix ripening and preterm labor. The survival surgeries were performed by placing the pregnant mouse under a mask maintaining a continuous flow of isoflurane to obtain adequate deep anesthesia. A mini-laparotomy was then performed in the lower abdomen; LPS or saline was infused into the right uterine horn between the first and second gestational sacs (just cephalad from the cervix). Peritoneal closure was performed with sutures and the incision closed with staples. Mice were monitored until either the birth of the first pup or up to 6 hours, after which the dams were euthanized with carbon dioxide. For these experiments, preterm birth was defined as delivering at least one mouse pup through the cervix within 6 hours of the intrauterine injection. Each cervix and uterus was harvested and flash frozen with liquid nitrogen for messenger RNA expression. Samples were stored at −80°C until processed for qPCR analysis as described subsequently. Sixteen animals received an intrauterine injection either LPS or saline.

A second series of experiments were performed to determine whether systemic treatment with an antibody to GM-CSF would prevent preterm birth or impact cervical remodeling. Mice were treated as described previously, and immediately after the intrauterine injection of LPS while still anesthetized, a 200 μg/100 μL retro-orbital injection was administered containing either an anti-mouse GM-CSF antibody (Bio X Cell, West Lebanon, New Hampshire, catalog #BE0259) or a nonspecific isotype immunoglobulin G (IgG) control antibody (Bio X Cell, catalog #BE0089). The particular IgG control antibody was chosen to match the isotype and species of the GM-CSF, thus to mimic any nonspecific xenogeneic reaction to treatment. Mice were again monitored until either the birth of the first pup or up to 6 hours, after which the dams were euthanized with carbon dioxide. The cervix was harvested, flash frozen with liquid nitrogen, and shipped to Loma Linda University for histology and analyses (see below). In addition, the AF and serum were collected to determine the concentration of GM-CSF by ELISA (BioLegend, San Diego, California) and analyzed according to the manufacture’s protocol.

A power analysis was performed to determine whether the GM-CSF antibody statistically decreases the rate of preterm birth compared to the isotype control antibody in mice receiving an intrauterine injection of LPS. We estimated the preterm birth rate within 6 hours would be 75% in mice receiving LPS and the control antibody and 20% in mice receiving LPS and the GM-CSF antibody. Assuming an α of .05 and a power of 80%, we estimated we would need 12 animals per treatment group.

Cervix Histology and Analyses

Cervix was immersion fixed in fresh 4% paraformaldehyde, transferred the next day to 70% ethanol, paraffin-embedded, sectioned (6 µm), and stained with a picrosirius red to identify cross-linked collagen.20 Optical density of polarized light birefringence from picrosirius red stain reflects cross-linked collagen fibers in the extracellular matrix of the stroma.12 Other sections were stained to identify mature Mϕs (F4/80 antibody 1:800 dilution, T-2006; Bachem, Torrance, California) and counterstained with hematoxylin to visualize cell nuclei as previously described.10,11,21 Sections were imaged with an Aperio Scanscope microscope and 8-16 photomicrographs (300 × 417 µm) taken from each of 2 cross sections of the cervix for each mouse. Cell nuclei and Mϕs in stroma were counted using NIH ImageJ (version 1.50) with care taken to exclude blood vessels, lumen, epithelium, and other atypical structures. As before, Mϕs were defined as brown stain within a cell boundary in proximity to hematoxylin-stained cell nuclei. Collagen and Mϕ numbers/area were normalized to the total number of cells/area for each animal to adjust for heterogeneity of anatomy that occurs within and among sections, as well as individuals. All cervices analyzed from mice receiving LPS only or LPS and the GM-CSF control antibody delivered within 6 hours. Cervices analyzed from mice receiving LPS and the GM-CSF neutralizing antibody did not deliver within 6 hours.

Quantitative Polymerase Chain Reaction

Total RNA was extracted from both the immortalized ectocervical and endocervical cells after 24 hours of LPS exposure and the mouse cervix and uterus using Qiagen RNeasy (Hilden, Germany) mini kits according to product protocol. Complementary DNA (cDNA) was generated from 1 µg of RNA/sample using a cDNA reverse transcription kit (Bio-Rad Laboratories, Hercules, California). To assess expression, qPCR was performed using equivalent dilutions of each sample on a Bio-Rad CFX qPCR instrument (N = 3-4 per treatment group). The CSF2 primer assays Hs00929873_m1 and Mm01290062_m1 conjugated to Taqman MGB probes were utilized in this analysis (Applied Biosystems, Foster City, California). The relative abundance of GM-CSF was divided by the relative abundance of 18S in each sample to generate a normalized abundance. All samples were analyzed in triplicate, and each experiment or N was analyzed separately and divided by the values of LPS and multiplied by 100 in order to present the results as percent LPS.

For AF-MSCs, RNA was extracted using Qiagen RNeasy mini kits according to product protocol. Complementary DNA was generated from 1 µg of RNA/sample using the Qiagen miScript II RT kit. To assess expression, qPCR was performed using equivalent dilutions of each sample (N = 4 per treatment group) using RT2 Profiler PCR custom array plates in combination with RT2 SYBR Green Mastermixes on a Bio-Rad CFX qPCR instrument. The relative abundance of GM-CSF was determined using the ΔΔCT method.

Lactate Dehydrogenase Cytotoxicity Assay

The viability of AF-MSCs after treatment with 25 μg/mL of LPS was determined by measuring lactate dehydrogenase (LDH) leaking into the medium using the Pierce LDH cytotoxicity assay (Thermo Scientific, Waltham, Massachusetts; N = 4 per treatment group). After collecting cell culture medium, the cells were lysed with reaction mixture containing substrate mix (lyophilizate) and assay buffer. Extracellular LDH in the media was quantified using a coupled enzymatic reaction in which LDH catalyzes the conversion of lactate to pyruvate via NAD+ reduction to NADH. Diaphorase then uses NADH to reduce a tetrazolium salt (INT) to a red formazan product that can be measured at 490 nm. The level of formazan formation is directly proportional to the amount of LDH released into the medium, which is indicative of cytotoxicity. We have previously published 25 μg/mL of LPS is not cytotoxic to ectocervical and endocervical cells, and therefore, the cytotoxicity assay was not performed on these cell lines.22

Statistical Analysis

Statistical analyses were performed comparing means or medians depending on whether the data were parametric (normally distributed P > .05 Levene test followed by Student t test or 1-way analysis of variance [ANOVA]) or nonparametric (Mann-Whitney test or ANOVA on ranks). If significance was met for multiple comparisons, pairwise comparison was then performed using Student-Newman-Keuls test. Analysis was performed by running GraphPad Prism software (GraphPad, San Diego, California).

Results

Lipopolysaccharide Activates Release and Upregulates GM-CSF in Cervical Epithelial Cells and AF-MSCs

Treatment with LPS resulted in an increase in GM-CSF concentrations after 24 hours in ectocervical cells (P < .0001), endocervical cells (P < .0005), and AF-MSCs (P < .01) as measured by ELISA. Expression of GM-CSF by qRT-PCR increased in ectocervical cells (P < .0005), endocervical cells (P < .01), and AF-MSCs (P < .0001; Figure 1). Thus, LPS increases the concentration and gene expression of GM-CSF from cervical epithelial and AF-MSCs.

Figure 1.

Figure 1.

Open in a new tab

Bar graph representing the mean and SEM demonstrating the effect of 25 μg/mL of lipopolysaccharide for 24 hours increased the concentration of GM-CSF from (A) ectocervical cells (*P < .001), (B) endocervical cells (*P < .0005), and (C) AF-MSCs (*P < .001) compared to saline (sal) controls by t test. Lipopolysaccharide also increased the expression of GM-CSF from (D) ectocervical (*P < .0005), (E) endocervical (*P < .01), and (F) AF-MSCs (*P < .0001) compared to sal controls using qPCR by t test (N = 3-4 per treatment group for each cell line. AF-MSCs indicates amniotic fluid mesenchymal stem cells; GM-CSF, granulocyte-macrophage colony-stimulating factor; qPCR, quantitative polymerase chain reaction; SEM, standard error of the mean.

Treatment With LPS Did Not Increase LDH

While exposure to LPS leads to an increase in GM-CSF concentration, LPS had no effect on the concentration of LDH compared to controls in AF-MSCs after 24 hours (data not shown) using the LDH assay (P = .61). This finding is consistent with previous evidence that LPS does not induce cell death in ectocervical and endocervical cells.22

Lipopolysaccharide Increases Expression of GM-CSF in the Mouse Cervix and Uterus

The GM-CSF expression was increased in the cervix and uterus of mice treated with an intrauterine injection of LPS compared to control mice that received saline (P < .005, Figure 2). Thus, LPS is a potent promotor of GM-CSF expression.

Figure 2.

Figure 2.

Open in a new tab

Bar graph depicting the mean and SEM demonstrating the effect of LPS on the expression of GM-CSF 6 hours after an intrauterine injection of 250 μg of LPS. Lipopolysaccharide increased the expression of GM-CSF from the mouse cervix (*P < .005) and uterus (*P < .0001) compared to mice receiving an intrauterine injection of sal by t test. GM-CSF, granulocyte-macrophage colony-stimulating factor; LPS, lipopolysaccharide; sal, saline; SEM, standard error of the mean.

Granulocyte-Macrophage Colony-Stimulating Factor Blockade Therapy Prevents LPS-Induced Preterm Birth and Decreases the GM-CSF Concentration in the Serum and AF

Inflammation-induced preterm birth, defined as delivering a mouse pup through the cervix within 6 hours, was reduced with GM-CSF antibody treatment. Within 6 hours of an intrauterine LPS injection, 87.5% (14 of 16) of mice were in the process of delivering at least one pup compared to 0 of 16 mice receiving an intrauterine injection of saline (Figure 3). In mice administered both LPS and a control isotype IgG antibody, 66.67% (8 of 12) were in the process of delivering preterm within 6 hours. Of the 12 mice receiving LPS and the GM-CSF antibody, 75% (9 of 12) had not delivered within 6 hours, for a preterm birth rate of 25% (P < .05, χ2).

Figure 3.

Figure 3.

Open in a new tab

Mice treated with an intrauterine injection of sal and LPS had a preterm birth rate (process of delivering) within 6 hours of 0% and 87.5%, respectively. Mice treated with LPS and the isotype control antibody (LPS + IgG) had a preterm birth rate of 66.7% compared to a 25% preterm birth rate in mice receiving LPS and the anti-mouse GM-CSF antibody (LPS + GM-CSFab; *P < .05, χ2). GM-CSF, granulocyte-macrophage colony-stimulating factor; IgG, immunoglobulin G; LPS, lipopolysaccharide; sal, saline.

Treatment with the GM-CSF antibody also decreased the concentration of GM-CSF in the serum. In mice receiving LPS only or LPS along with the IgG control antibody, serum concentrations of GM-CSF were significantly increased compared to saline controls and LPS + GM-CSF antibody mice (P < .001). Additionally, GM-CSF antibody treatment decreased the concentration of GM-CSF in the AF compared to the 3 other groups (Figure 4). Thus, suppressing GM-CSF in serum was associated with a reduced incidence of preterm birth.

Figure 4.

Figure 4.

Open in a new tab

Bar graph depicting the mean and SEM demonstrating the concentration of GM-CSF in the serum and amniotic fluid of mice treated with sal, LPS, LPS and the GM-CSF antibody (LPS + GM-CSFAb), and LPS and the control antibody (LPS + IgG). The concentration of GM-CSF was increased with LPS and LPS + IgG compared to mice treated with sal (*P < .0001) in the serum, but not in the amniotic fluid. Treatment with the GM-CSF antibody decreases the concentration of GM-CSF in the serum (**P < .001) and amniotic fluid (*P < .01), by 1-way ANOVA, Student-Newman-Keuls test. ANOVA indicates analysis of variance; GM-CSF, granulocyte-macrophage colony-stimulating factor; IgG, immunoglobulin G; LPS, lipopolysaccharide; sal, saline; SEM, standard error of the mean.

Granulocyte-Macrophage Colony-Stimulating Factor Antibody Effects on Characteristics of Cervix Remodeling

Images of the cervix stroma contained an abundance of purple-stained cell nuclei, a fraction of which were stained brown as an indication of the presence of mature macrophages. Abundance of macrophages, rather than morphology of cell nuclei or macrophages, appeared to differ across sections with respect to treatments (Figure 5). In response to treatment, neither the density of cell nuclei nor optical density of collagen birefringence in cervix stroma sections changed compared to mice in the saline group (Figure 6, panels A and B). However, macrophage density was reduced in cervix stroma in mice treated with the GM-CSF or IgG control antibody compared to those treated with either PBS or LPS.

Figure 5.

Figure 5.

Open in a new tab

Photographs of cervix from day 17 pregnant mice injected with treatments 6 hours earlier. Left to right: sal, LPS, LPS + GM-CSFab, or LPS+ isotype control antibody (LPS + IgG)-treated cervix (E17) stained by immunohistochemistry for macrophages (brown) with hematoxylin counterstain. Scale bar = 25 μm. E17 indicates embryonic day 17; GM-CSFab, granulocyte-macrophage colony-stimulating factor antibody; IgG, immunoglobulin G; LPS, lipopolysaccharide; sal, saline.

Figure 6.

Figure 6.

Open in a new tab

Density of cell nuclei/area, optical density of picrosirius red-stained collagen, and density of macrophage in the cervix stroma on E17 of pregnancy treated with sal, LPS, LPS + GM-CSFab, or LPS + isotype control antibody (LPS + IgG). Data are mean ± SE. *P < .05 versus sal and LPS groups by ANOVA. ANOVA indicates analysis of variance; E17, embryonic day 17; GM-CSFab, granulocyte-macrophage colony-stimulating factor antibody; IgG, immunoglobulin G; LPS, lipopolysaccharide; sal, saline; SE, standard error.

Discussion

Using in vitro models, our results demonstrated GM-CSF is produced from ectocervical cells, endocervical cells, and AF-MSCs when challenged with LPS. The GM-CSF is also upregulated in the cervix and uterus in a mouse model of preterm birth. Of critical importance, neutralization of systemic GM-CSF was associated with a decrease in the rate of inflammation-induced preterm birth, suggesting that GM-CSF activity is a critical mediator in this animal model of inflammation-induced preterm parturition. Therefore, these findings raise the possibility that systemic suppression of GM-CSF may be an approach to prevent preterm birth.

A pro-inflammatory immune response arising in reproductive tissue is believed to be a major component of preterm birth.23,24 Prior work on this topic has shown statins will decrease the preterm birth rate and the concentration of GM-CSF in the serum of mice treated with LPS.25,26 In addition, treatment with a broad-spectrum chemokine inhibitor, which reduced serum concentration of several cytokines including GM-CSF, decreased the preterm birth rate in a mouse model of preterm birth.27 Although the function of GM-CSF is to increase the influx monocytes which differentiate into macrophages, the presence of macrophages in the cervix does not correlate with cervical function due to their heterogeneity.12 Activated macrophages will produce a number of cytokines and chemokines, including tumor necrosis factor, interleukin (IL)-1, IL-6, IL-8, and IL-12.28 Activated macrophages will also activate complement,29 and an increase in complement levels has been shown to mediate preterm birth.30 Thus, GM-CSF may decrease the number of activated differentiated macrophages in the cervix, resulting in a decrease in cytokines and complement level, thus decreasing preterm birth.

Treatment with the GM-CSF-blocking antibody reduces the number of macrophages in the cervix within 6 hours after LPS to the same extent as the control isotype antibody. Although the effects of an inflammatory challenge on tissue-resident immune cells have received little attention, this finding raises the possibility that reduced macrophage density in the cervix and perhaps other tissues may be in response to a xenogenic immune challenge. Nevertheless, the phenotype of macrophages and their local activity within 6 hours of LPS treatment and leading up to labor is likely different between mice receiving the blocking GM-CSF versus isotype control antibody, as mice receiving the GM-CSF antibody are less likely to deliver preterm. This focus on macrophage density and activity seems warranted as the neutralization of GM-CSF did not affect the cell nuclei density or extracellular cross-linked collagen structure in the peripartum cervix. This lack of GM-CSF effect was not expected, given that these end points of cervical remodeling (cellular hypertrophy, edema, and biomechanical compliance) are at or near plateau in a ripened cervix on E17.11,12 Therefore, the mechanism by which the GM-CSF antibody prevents preterm birth is likely due to its ability to suppress macrophage activity and signaling rather than decreasing the number and density of macrophages in the cervix.

Of further interest is the counterintuitive finding that macrophages in the cervix did not increase in mice receiving LPS alone despite preterm birth. This result is consistent with a previous study31 and may reflect the complexity of mature macrophage functions, that is, inflammatory or regulatory, that are guided by local signals.32 These activities reflect macrophage phenotypes that would not be differentiated by staining for the F4/80 epitopes which reflects a role in phagocyte-related matrix interactions.33 Moreover, the time course of a changed census of residency by macrophages in the cervix may be overshadowed by effects of LPS to act directly upon Toll-like receptors on cervical stromal cells within the 6-hour period leading up to preterm birth.34 Although assessment of macrophage phenotypes and activities was beyond the scope in these studies, these findings raise the possibility that inflammation-induced ripening and preterm birth may accelerate pro-inflammatory processes in ways that differ from those at term parturition.

The use of immunomodulators has been widely done to treat a host of medical conditions ranging from cancers to autoimmune disorders,35,36 and specifically a monoclonal antibody to GM-CSF has been used to treat a variety of diseases, including pediatric neuroblastomas, respiratory diseases, autoimmune disorders, and gastrointestinal conditions.37,38 However, to our knowledge, this is the first use of a GM-CSF antibody to prevent preterm birth. As the etiology of preterm birth and cervical remodeling is initiated by an upregulation of pro-inflammatory processes, and due to the increase in GM-CSF from multiple sites in the genital tract, our results were expected. Although counterintuitive, LPS did not alter the concentration of GM-CSF in the AF, raising the possibility that inflammation-induced preterm birth is not acting through a mechanism in the AF. However, these results increase our understanding of the molecular pathways leading to preterm birth and may open up a new area of treatment modalities to decrease preterm birth and improve neonatal outcomes.

Our study is not without limitations. A lack of human cervical biopsies prior to term necessitates the use of animal models to test mechanistic questions. We realize these findings using an animal model may not address the multitude of factors or their convergent association on preterm birth and may not be directly extrapolated to humans.39,40 However, a number of therapeutic strategies in humans were originally developed or tested using animal models, such as the use of corticosteroids in lambs to improve neonatal outcomes.41 Furthermore, our findings using an antibody to GM-CSF in pregnancy to decrease the rate of preterm birth hold great promise for studies in higher mammalian species.

In summary, our results show the inhibition of GM-CSF during a pro-inflammatory immune response in the reproductive tract will prevent preterm birth. Our results also showed treatment with the GM-CSF antibody decreases the concentration of GM-CSF in the serum but does not alter the number of macrophages in the cervix, suggesting that the GM-CSF antibody may be impacting macrophage function and activity. Further research is needed to determine whether an antibody to GM-CSF could be utilized as a therapeutic agent to prevent spontaneous preterm birth.

Acknowledgments

The authors would like to thank the Department of Women’s Health at Hartford Hospital for supporting this work. The authors would also like to thank Anne Heureman for her assistance with preparing the figures for publication and data analysis.

Authors’ Note: This research was conducted at the University of Connecticut School of Medicine in Farmington, Connecticut, and Loma Linda University in Loma Linda, California.

Declaration of Conflicting Interests: The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding: The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was funded by the Department of Women’s Health at Hartford Hospital (C.N.) and NIH HD054931 (S.M.Y.).

References

  • 1. Goldenbertg R, Culhane J, Iams J, Romero R. Epidemiology and causes of preterm birth. Lancet. 2008;371(9606):75–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Blencowe H, Cousens S, Oestergaard M, et al. National, regional and worldwide estimates of preterm birth. Lancet. 2012;379(9832):2162–2172. [DOI] [PubMed] [Google Scholar]
  • 3. Romero R, Dey S, Fisher S. Preterm labor: one syndrome, many causes. Science. 2014;345(6198):760–765. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Kemp W, Saito M, Newnham JP, Nitsos I, Okamura K, Kallapur SG. Preterm birth, infection, and inflammation advances from the study of animal models. Reprod Sci. 2010;17(7):619–628. [DOI] [PubMed] [Google Scholar]
  • 5. Huang WC, Sala-Newby GB, Susana A, Johnson JL, Newby AC. Classical macrophage activation up-regulates several matrix metalloproteinases through mitogen activated protein kinases and nuclear factor-kappaB. PLoS One. 2012;7(8):e42507. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Vadillo-Ortega F, Hernandez A, Gonzalez-Avila G, Bermejo L, Iwata K, Strauss J. Increased matrix metalloproteinase activity and reduced tissue inhibitor of metalloproteinases-1 levels in amniotic fluids from pregnancies complicated by premature rupture of membranes. Am J Obstet Gynecol. 1996;174(4):1371–1376. [DOI] [PubMed] [Google Scholar]
  • 7. Athayde N, Edwin SS, Romero R, et al. A role for matrix metalloproteinase-9 in spontaneous rupture of the fetal membranes. Am J Obstet Gynecol. 1998;179(5):1248–1253. [DOI] [PubMed] [Google Scholar]
  • 8. Xu P, Alfaidy N, Challis JR. Expression of matrix metalloproteinase (MMP)-2 and MMP-9 in human placenta and fetal membranes in relation to preterm and term labor. J Clin Endocrinol Metab. 2002;87(3):1353–1361. [DOI] [PubMed] [Google Scholar]
  • 9. Gomez-Lopez N, StLouis D, Lehr M, Sanchez-Rodriguez E, Arenas-Hernandez M. Immune cells in term and preterm labor. Cell Mol Immunol. 2014;11(6):571–581. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Yellon S, Dobyns A, Beck H, Kurtzman J, Garfield R, Kirby M. Loss of progesterone receptor-mediated actions induce preterm cellular and structural remodeling of the cervix and premature birth. PLoS One. 2013;8(12):e81340. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Kirby M, Heuerman A, Custer M, et al. Progesterone receptor-mediated actions regulate remodeling of the cervix in preparation for preterm parturition. Reprod Sci. 2016;23(11):1473–1483. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Yellon S. Contributions to the dynamics of cervix remodeling prior to term and preterm birth. Biol Reprod. 2017;96(1):13–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Cetean S, Căinap C, Constantin A, et al. The importance of the granulocyte-colony stimulating factor in oncology. Clujul Med. 2015;88(4):468–472. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Shi Y, Liu C, Roberts A, et al. Granulocyte-macrophage colony-stimulating factor (GM-CSF) and T-cell responses: what we do and don’t know. Cell Res. 2006;16(2):126–133. [DOI] [PubMed] [Google Scholar]
  • 15. Curry AE, Thorsen P, Drews C, et al. First-trimester maternal plasma cytokine levels, pre-pregnancy body mass index, and spontaneous preterm delivery. Acta Obstet Gynecol Scand. 2009;88(3):332–342. [DOI] [PubMed] [Google Scholar]
  • 16. Chandiramani M, Seed P, Orsi N, et al. Limited relationship between cervico-vaginal fluid cytokine profiles and cervical shortening in women at high risk of spontaneous preterm birth. PLoS One. 2012;7(12):e52412. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Basraon S, Menon R, Makhlouf M, et al. Can statins reduce the inflammatory response associated with preterm birth in an animal model? Am J Obstet Gynecol. 2012;207(3):224.e1–227.e1. [DOI] [PubMed] [Google Scholar]
  • 18. Phermthai T, Odglun Y, Julavijitphong S, et al. A novel method to derive amniotic fluid stem cells for therapeutic purposes. BMC Cell Biol. 2010;11:79 doi:10.1186/1471-2121-11-79 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Elovitz MA, Wang Z, Chien EK, Rychlik DF, Phillippe M. A new model for inflammation-induced preterm birth: the role of platelet-activating factor and toll-like receptor-4. Am J Pathol. 2003;163(5):2103–2111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Kirby M, Heuerman A, Custer M. et al. progesterone receptor-mediated actions regulate remodeling of the cervix in preparation for preterm parturition. Reprod Sci. 2016;23(11):1473–1483. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Nold C, Maubert M, Anton L, Yellon S, Elovitz M. Prevention of preterm birth by progestational agents: what are the molecular mechanisms? Am J Obstet Gynecol. 2013;208(3):223.e1–223.e7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Nold C, Anton L, Brown A, Elovitz M. Inflammation promotes a cytokine response and disrupts the cervical epithelial barrier: a possible mechanism of premature cervical remodeling and preterm birth. Am J Obstet Gynecol. 2012;206(3):208.e1–208.e7. [DOI] [PubMed] [Google Scholar]
  • 23. Bastek J, Gómez L, Elovitz M. The role of inflammation and infection in preterm birth. Clin Perinatol. 2011;38(3):385–406. [DOI] [PubMed] [Google Scholar]
  • 24. Menon R, Fortunato S. Infection and the role of inflammation in preterm premature rupture of the membranes. Best Pract Res Clin Obstet Gynaecol. 2007;21(3):467–478. [DOI] [PubMed] [Google Scholar]
  • 25. Basraon S, Costantine M, Saade G, Menon R. The effect of simvastatin on infection-induced inflammatory response of human fetal membranes. Am J Reprod Immunol. 2015;74(1):54–61. [DOI] [PubMed] [Google Scholar]
  • 26. Gonzalez J, Pedroni S, Girardi G. Statins prevent cervical remodeling, myometrial contractions and preterm labor through a mechanism that involves hemoxygenase-1 and complement inhibition. Mol Hum Reprod. 2014;20(6):579–589. [DOI] [PubMed] [Google Scholar]
  • 27. Shynlova O, Dorogin A, Li Y, Lye S. Inhibition of infection-mediated preterm birth by administration of broad spectrum chemokine inhibitor in mice. J Cell Mol Med. 2014;18(9):1816–1829. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Arango Duque G, Descoteaux A. Macrophage cytokines: involvement in immunity and infectious diseases. Front Immunol. 2014;5:491. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Gonzalez JM, Romero R, Girardi G. Comparison of the mechanisms responsible for cervical remodeling in preterm and term labor. J Reprod Immunol. 2013;97(1):112–119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Gonzalez J, Franzke C, Yang F, Romero R, Girardi G. Complement activation triggers metalloproteinases release inducing cervical remodeling and preterm birth in mice. Am J Pathol. 2011;179(2):838–849. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Yellon S, Ebner CA, Elovitz M. Medroxyprogesterone acetate modulates remodeling, immune cell census, and nerve fibers in the cervix of a mouse model for inflammation-induced preterm birth. Reprod Sci. 2009;16(3):257–264. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Parisi L, Gini E, Baci D, et al. Macrophage polarization in chronic inflammatory diseases: killers or builders? J Immunol Res. 2018;2018:8917804. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Dos Anjos Cassado A. F4/80 as a major macrophage marker: the case of the peritoneum and spleen. Results Probl Cell Differ. 2017;62:161–179. [DOI] [PubMed] [Google Scholar]
  • 34. Wang Y, Weng Y, Shi Y, Xia X, Wang S, Duan H. Expression and functional analysis of Toll-like receptor 4 in human cervical carcinoma. J Membr Biol. 2014;247(7):591–599. [DOI] [PubMed] [Google Scholar]
  • 35. Schioppo T, Ingegnoli F. Current perspective on rituximab in rheumatic diseases. Drug Des Devel Ther. 2017;11:2891–2904. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Marin-Acevedo J, Soyano A, Dholaria B, Knutson K, Lou Y. Cancer immunotherapy beyond immune checkpoint inhibitors. J Hematol Oncol. 2018;11(1):8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. McGinty L, Kolesar J. Dinutuximab for maintenance therapy in pediatric neuroblastoma. Am J Health Syst Pharm. 2017;74(8):563–567. [DOI] [PubMed] [Google Scholar]
  • 38. Shiomi A, Usui T, Mimori T. GM-CSF as a therapeutic target in autoimmune diseases. Inflamm Regen. 2016;36:8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Elovitz M, Mrinalini C. Animal models of preterm birth. Trends Endocrinol Metab. 2004;15(10):479–487. [DOI] [PubMed] [Google Scholar]
  • 40. Mitchell B, Taggart M. Are animal models relevant to key aspects of human parturition? Am J Physiol Regul Integr Comp Physiol. 2009;297(3):R525–R545. [DOI] [PubMed] [Google Scholar]
  • 41. Liggins G. Premature delivery of foetal lambs infused with glucocorticoids. J Endocrinol. 1969;45(4):515–523. [DOI] [PubMed] [Google Scholar]

Articles from Reproductive Sciences are provided here courtesy of Society for Reproductive Investigation

RESOURCES