- Email: [email protected]
Systemic increase in human maternal circulating CD14+CD16− MCP-1+ monocytes as a marker of labor
Research
www.AJOG .org
OBSTETRICS
Systemic increase in human maternal circulating CD14DCD16L MCP-1D monocytes as a marker of labor Marc Bardou, MD, PhD; Tarik Hadi, PhD; Guillaume Mace, MD; Matthieu Pesant, PhD; Julie Debermont; Marina Barrichon, MsC; Maeva Wendremaire, PharmD, PhD; Nicole Laurent, MD; Paul Sagot, MD; Fre´de´ric Lirussi, PharmD, PhD OBJECTIVES: To study the influence of pregnancy and labor on the proportion and level of activation of monocyte subpopulations in human pregnancy. STUDY DESIGN: Peripheral blood samples were obtained from healthy
nonpregnant women (n ¼ 6); women in the third-trimester of healthy pregnancies (n ¼ 18) and women with preterm premature rupture of membranes (n ¼ 46), just before delivery for the last 2 groups. Monocyte subpopulations were characterized by flow cytometry using CD14, CD16, and activation level using macrophage chemoattractant protein-1 (MCP-1) and CCR2 antibodies. RESULTS: The relative proportion of each monocyte subset in
nonpregnant women was similar to that in women with healthy or complicated pregnancies. However, pregnancy was associated with a significant decrease in MCP-1 expressing monocytes (79.5% 19.8% vs 9.3% 6.8% and 11.9% 8.3% for nonpregnant, healthy pregnancy, and preterm premature rupture of membranes
(respectively, P < .05). Spontaneous labor was associated with a return to nonpregnant values for the proportion of MCP-1 expressing monocytes in both normal (74.4% 16.9) and preterm premature rupture of membranes pregnancy (68.4% 35.6), irrespective of the mode of delivery (vaginal or cesarean section). This was not observed in women who delivered without spontaneous labor onset. CCR-2 (MCP-1 receptor) expression was not modified in monocytes at the time of labor, but was significantly increased in granulocytes (3646 1080 vs 7338 2718 for nonlaboring and laboring preterm premature rupture of membranes, respectively, P < .05) CONCLUSION: In light of previous reports of a role for MCP-1 in labor,
our results suggest the downregulation of activation levels of monocytes, via MCP-1 expression might be involved in maternofetal immune tolerance. Monocyte reactivation might be associated with labor. Key words: human pregnancy, monocytes, preterm labor, preterm rupture of membrane, prospective study
Cite this article as: Bardou M, Hadi T, Mace G, et al. Systemic increase in human maternal circulating CD14DCD16L MCP-1D monocytes as a marker of labor. Am J Obstet Gynecol 2014;210:70.e1-9.
P
regnancy and delivery are both complex immune situations involving numerous processes. During pregnancy, maternal immunity is reduced, with both neutrophils1 and monocytes2 displaying decreased functionality and
reactivity compared with those in nonpregnant women. This decrease in maternal immune response is necessary to prevent the fetal allograft from rejection and permits fetal cells to invade the uterine mucosa to establish the
From Centre d’Investigations Cliniques Plurithématique 803 (INSERM CIC-P 803) (Drs Bardou, Mr Hadi and Ms Debermont); Center for Lipid Research, INSERM (Institut National de la Santé et de la Recherche Médicale) (Drs Bardou, Wendremaire, and Lirussi, Mr Hadi, and Ms Barrichon); CHU de Dijon (Drs Bardou, Wendremaire, Sagot, and Lirussi, Mr Hadi, and Ms Barrichon); Service de Gynécologie et d’Obstétrique (Drs Mace and Sagot); Laboratoire de Pharmacologie-Toxicologie (Drs Wendremaire and Lirussi); and Service d’Anatomie Pathologique (Dr Laurent), Dijon, France, and Humanitas Clinical and Research Center, Leukocyte Biology Laboratory (Dr Pesant), University of Milan, Milan, Italy. Received May 2, 2013; revised July 27, 2013; accepted Aug. 27, 2013. This work was supported by a grant from the French Ministry of Health (PHRC interregional 2010) and from INSERM and Direction Générale de l’Offre de Soins (DGOS) (AAP-RCT 2011). The authors report no conflict of interest. Presented at the 58th annual meeting of the Society for Gynecologic Investigation, Miami, FL, March 16-19, 2011. Reprints will not be available. 0002-9378/$36.00 ª 2014 Mosby, Inc. All rights reserved. http://dx.doi.org/10.1016/j.ajog.2013.08.031
70.e1 American Journal of Obstetrics & Gynecology JANUARY 2014
placenta.3,4 The role of endometrial monocytes/macrophages and dendritic cells has recently been suggested as pivotal for successful pregnancy establishment in cattle.5 Samstein and colleagues recently suggested that maternalefetal tolerance to paternal alloantigens is an active process in which plasmatic Tregs specifically respond to paternal antigens to induce tolerance.6 In contrast, human parturition, either at term or preterm, is increasingly recognized as a sterile inflammatory event.7 Several studies have shown a massive influx of both neutrophils and monocytes/macrophages from maternal blood into the major effector tissues of parturition, ie, myometrium, cervix, and fetal membranes during labor.8-11 Human parturition is a complex phenomenon with several factors, such as hormonal changes, increased production of prostaglandins and release of inflammatory cytokine, interacting together.12
Obstetrics
www.AJOG.org This recruitment of maternal monocytes and granulocytes from maternal blood is induced by the secretion of chemotactic proinflammatory cytokines, such as interleukin-8 (IL-8), MIP-1a or macrophage chemotracant protein-1 (MCP-1).13 MCP-1, also known as C-C motif chemokine ligand 2 (CCL-2), is a major mediator of monocyte/macrophage infiltration at inflammatory sites both under physiologic and pathologic conditions. The role of MCP-1 during delivery was suggested by its presence in the myometrium and gestational membranes during full-term labor and in amniotic fluid during preterm labor.14 Experimental studies conducted in pregnant mice demonstrated that the administration of lipopolysaccharide (LPS) induces parturition within 12 to 24 hours and that this event is preceded by a peak in MCP-1 expression in the maternal bloodstream, occurring 6 hours after LPS injection.15 Beyond the contribution of cytokines to parturition, Yuan and colleagues16 demonstrated that full-term and preterm labor are also regulated by modifications in the proportion or number and activation status of peripheral blood monocytes and granulocytes. They concluded that monocyte and granulocyte priming precedes labor onset. Recent studies have highlighted the heterogeneity of human peripheral blood monocytes. Based on the recently proposed nomenclature, monocyte subsets can be discriminated by their differential expression of CD14 and CD16.17,18 Monocytes are thus characterized as CD14high CD16 classical (a major subset accounting for 70% to 90% of peripheral blood monocytes), intermediate CD14high CD16þ, and nonclassical CD14low CD16þ monocytes.18 Monocytes also differ phenotypically by chemokine-receptor expression such as CCR2 (receptor of MCP-1) and CX3CR1 (fractalkine receptor). Classical CD14high CD16 monocytes mainly express CCR2, whereas nonclassical CD14low CD16þ mainly express CX3CR1.17,18 Although the proportion of each monocyte subset has only been scarcely studied in pregnancy, it may vary with the onset of inflammatory events.19
Monocytes appear to play a central although complex role during pregnancy, as they are involved in both tolerance to the fetal allograph and in delivery. Thus this study aimed to investigate (1) the changes induced by pregnancy in the relative proportions and activation level of each monocyte subset, (2) the influence of preterm premature rupture of membranes (pPROM) on this proportion and activation level, and (3) the influence of intrauterine infection and labor onset in women with pPROM on the activation status of monocytes via MCP-1 and CCR-2 expression and that of granulocytes via CCR-2 expression.
M ATERIALS
AND
M ETHODS
Study design and patient recruitment The study was designed as a prospective, single center, observational study. Peripheral blood samples were obtained between January 2009 and April 2010, from healthy nonpregnant female volunteers (nonpregnant, NP, n ¼ 6), in the third-trimester of a healthy pregnancy (healthy pregnancies, HP, n ¼ 18), and from women with preterm premature rupture of membranes (pPROM, n ¼ 46). Inclusion criteria were as follows: pPROM defined by either vaginal pooling of amniotic fluid during the speculum examination or a positive Actim PROM TEST (Medix Biochemica, Kauniainen, Finland). Normal pregnancy: women admitted to our institution for full-term delivery following a healthy pregnancy. Nonpregnant women with childbearing potential: age between 18 and 40, regular cycle, during the first phase of the cycle, and without oral contraceptive use at the time of blood sampling. Exclusion criteria were as follows: for the pPROM group: hospitalization before 24 or after 34 weeks of gestation, delivery within 12 hours after admission, hemorrhagic placenta previa, HIV, HBV, or HCV infection, consent refusal. Diagnosis of chorioamnionitis: chorioamnionitis was assessed in all of the pregnancies including pPROM in the present study by a positive culture of the placenta and/or a histologic
Research
assessment of the placenta by a single pathologist (Dr Laurent) using validated criteria.20 All clinical and biologic data were collected prospectively and reported in the case report form. This study was approved by the French Medicines Agency (registration no. 2008-A00119-46), and local ethics committee (CPP-Est1, Dijon, France). Written informed consent was obtained from all volunteers.
Blood sampling and fluorescenceactivated cell sorting for the analysis of blood monocyte subsets Fresh blood samples were collected by venipuncture, either during the first phase of the cycle for nonpregnant women or before delivery during spontaneous labor or before labor induction or cesarean delivery for pregnant women. Women who had their blood sampling performed before labor induction were logically included in the nonlaboring group irrespective of the final mode of delivery (ie, vaginal or cesarean delivery). Blood was collected in EDTA separator tubes and promptly subjected to antigen characterization. Monocyte subpopulations were analyzed with monoclonal antibodies against CD14-FITC, CD16-APC-Cy7, CCR2-Alexa 647, and MCP1-PE (all from BD Bioscience). Whole blood (100 mL) was first subjected to erythrocyte lysis with ACK lysing buffer (ACK; BioWhittaker, Walkersville, MD) followed by centrifugation at 240 g, 4 C for 10 minutes. Pelleted leukocytes were then washed with phosphate buffered saline (PBS), centrifuged at 240 g, 4 C for 10 minutes and stained for 20 minutes at 4 C with CD14-FITC, CD16-APC-Cy7, and CCR2-Alexa 647 antibodies. Leukocytes were then fixed in PBS0.1% paraformaldehyde for 10 minutes at room temperature, washed with PBS and subjected to MCP1-PE intracellular staining in PBS-Tween 0.1%-BSA 1% for 1 hour at room temperature. Staining with the
JANUARY 2014 American Journal of Obstetrics & Gynecology
70.e2
Research
Obstetrics
corresponding isotype control was performed for each staining. Cytometric analysis was performed on an LSRII cytometer (Becton Dickinson) using fluorescence-activated cell sorting Diva 6.1.2 software (Becton Dickinson). Monocytes were gated in a forward scatter/side scatter dot plot and 4-color staining was analyzed within this primary gate. Monocyte subsets were defined according to their CD14/CD16 pattern and classified into classical, intermediates, and nonclassical as described above.18 The proportions of monocyte subsets were expressed as percentage of total monocytes. The percentage of each subset expressing CCR2 and MCP-1 were obtained by appropriate gating and expressed as a percentage of CCR2þ and MCP-1þ monocytes within these gates. CCR2 and MCP-1 expression in each monocyte subset was also determined by mean fluorescence intensity minus the respective isotype control (MFIMFI isotype).
Plasma levels of MCP-1 MCP-1 was measured in the plasma of women who by ELISA test using Quantikine ELISA kit (R&D systems, ref. DCP00). Analyses were performed in duplicate, according to manufacturer’s instruction. Quantification was done according to internal standard, without normalization to protein level. Statistical analysis Sample size calculation: as no data were available on the proportion of MCP-1 expressing monocyte at the time of study design, preliminary results were used to assess sample size. With a type 1 error of 0.025, 7 patients per group gave us a 0.80 power to detect a 65% decrease in the proportion of MCP-1 expressing monocytes in pregnant women. This MCP-1 expressing monocyte assessment took also part of a prospective study of women with pPROM, and it explain why the number of pPROM pregnancy is much higher than 7. Results were expressed as the percentage ( standard error to the mean, SEM) of each subpopulation
www.AJOG.org (classical, CD4highCD6; intermediates CD4highCD6þ; and nonclassical, CD4lowCD6þþ) among total monocytes and as median fluorescence intensity (MFI) for MCP-1 and CCR-2 in arbitrary units. Several analyses were performed according to our study objectives. First, the different subsets of monocytes in nonpregnant and women with healthy pregnancies were compared. They were then compared in healthy and pPROM pregnancies. Thereafter, the proportion and MFI of MCP-1 expressing monocytes was compared in women with healthy or with pPROM pregnancies, and then in those who delivered spontaneously after labor onset and those who underwent cesarean section performed before labor onset or labor induction. Mean and SEM for percentages of each subpopulation were generated on each of the variables. Tests were performed using both parametric and nonparametric methods (t tests and Mann-Whitney U tests for continuous variables, and Fisher exact tests for ordinal variables). All
analyses were performed using GraphPad Instat 3 (GraphPad Software Inc., La Jolla, CA).
R ESULTS Baseline characteristics of the population are presented in the Table. Mean maternal age was similar in the different groups. The average gestational age at delivery and subsequently birthweight were significantly lower in the pPROM group than in the normal pregnancy group. Most of the women with pPROM (18/46, 56.5%), and almost all of the women with an uncomplicated pregnancy (17/18, 94.4%) had a vaginal delivery. Among the women with a healthy pregnancy, 14 delivered after labor onset (14 natural vaginal delivery, no cesarean section during labor), 4 delivered before spontaneous labor onset (3 with induced vaginal delivery, and 1 with cesarean section not during labor). In the pPROM group, 21 women delivered after labor onset (16 natural
TABLE
Clinical characteristics of study participants Characteristic Number of women Age, y
Healthy Nonpregnant pregnancy 6 32 ( 5.9)
Birthweight, g
pPROM
18
46
29.71 ( 5.38)
30.04 ( 6.10)
3138.24 ( 404.80) 1803.4 ( 760.46)
Parity ( SD)
0.5 ( 1.22)
2.30 ( 1.30)
Gravidity ( SD)
1.5 ( 1.10)
1.15 ( 1.37)
Tobacco
3/6
Gestational age at delivery, wks
3/18
10/46
39.3 ( 1.3)
31.2 ( 2.46)
Vaginal (natural)
14 (77.7)
16 (34.8)
Vaginal (induced)
3 (16.7)
10 (21.7)
Cesarean section, during labor
0 (0)
Cesarean section, not during labor
1 (5.6)
15 (32.6)
0 (0)
22 (47.8)
Mode of delivery, n (%)
Confirmed chorioamnionitis pPROM, preterm premature rupture of membranes.
Bardou. MCP-1 expressing monocytes and labor. Am J Obstet Gynecol 2014.
70.e3 American Journal of Obstetrics & Gynecology JANUARY 2014
P value
5 (10.8)
.049
Obstetrics
www.AJOG.org
FIGURE 1
Gating strategy assessing the proportion of monocyte subsets
A and B, Gating strategy used to determine the proportion of each monocyte subset in nonpregnant and pregnant women. Monocytes were gated in a forward scatter/side scatter dot plot and CD14/ CD16 stainings were used to identify monocytes within this primary gate. Monocyte subsets were defined according to their CD14/CD16 pattern and we defined: classical monocytes as CD14high CD16, intermediate monocytes as CD14high CD16þ, and nonclassical monocytes as CD14low CD16þ. B and C, Relative proportion of each monocyte subset in nonpregnant women (NP), women with normal pregnancies before the onset of labor, A, in pPROM women and B, the influence of chorioamnionitis, on the relative proportion of each monocyte subset. Proportions of each monocytes subset are represented as mean percentage SEM. Bardou. MCP-1 expressing monocytes and labor. Am J Obstet Gynecol 2014.
vaginal deliveries and 5 cesarean sections during labor), and 25 delivered before spontaneous labor onset (10
induced vaginal deliveries and 15 cesarean sections before labor onset). Twenty-two (47.8%) women in the
Research
pPROM group had a confirmed chorioamnionitis vs none in the healthy pregnancy group. Influence of pregnancy and pregnancyrelated complications on the relative proportion of each monocyte subset. Figure 1, A, shows the gating strategy used to determine the relative proportion of each monocyte subpopulation presented in Figure 1, B and C. Figure 1, B, shows that nonpregnant women (NP), and pregnant women, with either a healthy (HP) or pPROM pregnancy had similar relative proportions of the classical (CD14highCD16), intermediate (CD14highCD16þ), and nonclassical monocytes (CD14lowCD16þþ) (classical: 95.0 3.2%, 92.4 5.3% and 94.0 3.3%; intermediate: 3.2 0.9%, 5.4 3.7% and 4.4 2.5%; and nonclassical; 1.7 1%, 1.6 1.7% and 1.5 1.2% for NP, HP, and pPROM women, respectively, P ¼ NS). Figure 1, C, shows that in women with pPROM, the presence of confirmed (see Method section) chorioamnionitis (CA) did not affect the relative proportion of each monocyte subset (classical: 93.5 3.2% vs 94.8 3.3%; intermediate: 4.8 2.6% vs 3.5 2.4%; and nonclassical: 1.7 1.5% vs 1.2 0.8%, respectively, in women with pPROM with or without CA, respectively, Figure 1, C).
MCP-1 expression in monocytes is reduced in normal and pPROM pregnancies Figure 2 shows that, compared with nonpregnant women (Figure 2, A), nonlaboring pregnant women, with either a healthy or pPROM pregnancy (Figure 2, B), had a significantly lower proportion of MCP-1 expressing monocytes (MCP-1þ) (79.5% 19.75, n ¼ 6; 9.3% 6.8, n ¼ 4 and n ¼ 25 for NP, nonlaboring HP, and nonlaboring pPROM, respectively; P < .001) (Figure 2, D, black bars). When blood sampling was performed in pregnant women before the onset of spontaneous labor, mean fluorescence intensity was also decreased, compared with nonpregnant women (Figure 2, D, red, green and blue lines).
JANUARY 2014 American Journal of Obstetrics & Gynecology
70.e4
Research
Obstetrics
Labor induces MCP-1 expression in both classical and intermediate monocytes Moreover, Figure 2, C and D, also shows that spontaneous labor was associated with a marked and statistically significant increase in the percentage of MCP-1-expressing monocytes; proportions returned to a nonpregnant state in both HP women (73.4% 23.3% vs 9.3% 6.8%, for laboring and nonlaboring HP, respectively, P < .001) and women with pPROM (68.4% 35.6% vs 11.9% 8.3% for laboring and nonlaboring pPROM, respectively, P < .001). In women with pPROM, the proportion of MCP-1 expressing monocytes, according to the presence or absence of spontaneous labor, was not statistically different according to the mode of delivery (nonlaboring women: 13.41% 8.03 vs 10.86% 5.63 [P ¼ .429]; and laboring women 50.8% 31.04 vs 75.15% 33.63 [P ¼ .276] for cesarean or vaginal delivery, respectively). No such comparison was feasible for women with healthy pregnancies as all of them had a vaginal delivery. In laboring (n ¼ 21) and nonlaboring (n ¼ 25) women with pPROM, we further investigated the contribution of each monocyte subset to MCP-1 expression. Figure 3, B, shows that labor onset was associated with an increase in the proportion of MCP-1eexpressing intermediates (73.7% 16.8% vs 10.6% 8.0% in laboring and nonlaboring pPROM, respectively, P < .001) and classical monocytes (74.4% 19.9% vs 9.3% 6.8% in laboring and nonlaboring pPROM, respectively, P < .001). No significant differences between laboring and nonlaboring women with pPROM were observed in the very low proportions of nonclassical MCP-1expressing monocytes (0.8% 0.5%, vs 0.3% 0.2%, in laboring and nonlaboring pPROM, respectively, P ¼ NS). As shown in Figure 3, C, the increased proportion of MCP-1 expressing intermediate and classical monocytes in women in labor was paralleled by a significant increase in the median MCP-1
www.AJOG.org
FIGURE 2
Influence of pregnancy and labor on the percentage of monocytes expressing MCP-1, in normal and pPROM pregnancies
MCP-1 staining: presented granularity/MCP-1 dot plots are representative of the MCP-1 staining obtained in monocytes for A, nonpregnant women and B, pregnant woman not in labor. C, Considered significant are indicated. Nonpregnant (n ¼ 6), normal pregnancy (n ¼ 18, with 14 women with spontaneous labor and 4 without), pPROM (n ¼ 46 with 27 women with spontaneous labor, and 19 without). D, shows a representative overlay of the MCP-1 stainings obtained in monocytes from nonpregnant, healthy pregnant, women with pPROM (not in labor and in labor). MCP-1, macrophage chemoattractant protein-1; pPROM, preterm premature rupture of membranes. **P < .01 between nonpregnant and pregnant nonlaboring women, black bars; yP < .05; between spontaneous labor, and the absence of labor, grey and black bars, respectively. Bardou. MCP-1 expressing monocytes and labor. Am J Obstet Gynecol 2014.
fluorescence intensity, (expressed in arbitrary density unit, AU; intermediate: 1148 AU 370 vs 299 AU 177, and classical monocytes: 1083 AU 346 vs 241 AU 172, for laboring and nonlaboring pPROM women, respectively, P < .001), suggesting that these 2 monocyte subpopulations had an
70.e5 American Journal of Obstetrics & Gynecology JANUARY 2014
activated status. No significant difference was observed in the nonclassical monocyte subpopulation (137.7 AU 44.7 vs 80.3 AU 38.2 for laboring and nonlaboring pPROM women, respectively, P ¼ NS). In contrast plasma levels of MCP-1 in laboring and nonlaboring women were
Obstetrics
www.AJOG.org
Research
expression in any of the 3 subsets (data not shown).
FIGURE 3
MCP-1-expressing monocytes and MFI: influence of labor
C OMMENT
A, Gating strategy used to determine the percentage of MCP-1 expressing monocytes in each subset. Classical, intermediates, and nonclassical monocytes were characterized as described previously, B and C, MCP-1 staining overlay in each subpopulation: in women B, with pPROM B, not in labor and C, in labor. D and E, Percentage of MCP-1 expressing D, monocytes and E, MCP-1 MFI in classical, intermediates and nonclassical monocytes from women with pPROM (in labor vs not in labor). Results are represented as mean value SEM. The significance of differences between the groups in each subset was analyzed using a 1-way ANOVA followed by a Mann-Whitney U test. Values of P considered significant are indicated. Grey bars ¼ labor (n ¼ 27), black bars ¼ not in labor (n ¼ 19). ANOVA, analysis of variance; MCP-1, macrophage chemoattractant protein-1; MFI, mean fluorescence intensity; pPROM, preterm premature rupture of membranes. *P < .05; **P < .01. Bardou. MCP-1 expressing monocytes and labor. Am J Obstet Gynecol 2014.
not different as it is shown on Figure 4 (in pg/mL, 119.5 15.8 vs 133.5 22.9, respectively, P ¼ .613).
Influence of labor on the proportion of CCR2-expressing monocytes and granulocytes We finally investigated the impact of labor on the expression of CCR2 (MCP-1 receptor) in each monocyte subset and
in granulocytes. Figure 5, A, shows the gating strategy used to assess CCR2 expression in granulocytes. Figure 5, B, shows that labor was associated with a marked and statistically significant increase in granulocytes CCR2 MFI (7338 AU 2718 vs 3646 AU 1080 for laboring and nonlaboring pPROM, respectively, P < .05). In contrast, labor did not influence monocyte CCR-2
In humans, CD14high CD16 classical monocytes are predominant because they represent 70-90% of peripheral blood monocytes, whereas nonclassical monocytes (including patrollers and intermediates) account for 10-30% of peripheral blood monocytes.17 Even though the physiologic roles of the different monocyte subsets are not fully understood, at least 2 subsets (classical and nonclassical) might have different roles during homeostasis and the inflammatory response.21 Indeed, studies suggest that classical monocytes initiate inflammatory activities and are weak producers of inflammatory cytokines but strong producers of reactive oxygen species (ROS), whereas nonclassical monocytes express high levels of inflammatory cytokines but produce limited amounts of ROS.22,23 In this study, we have shown that whereas pregnancy does not affect the relative proportion of monocyte subsets, either classical (CD14highCD16), intermediates (CD14highCD16þ), and nonclassical (CD14lowCD16D), even in the case of pPROM with confirmed CA, it is associated with a marked decrease in classical and intermediate MCP-1 expressing monocytes, irrespective of gestational age. This is in apparent contradiction with the study by Melgert et al19 suggesting that the percentage of classical monocyte is decreased during pregnancy (91% [83-98%] vs 94% [90-98%] for pregnant and nonpregnant women, respectively), whereas the percentage of combined nonclassical/intermediate monocytes was higher in pregnant women (8.5% [2.3-16.6%] vs 5.6% [1.9-9.5%]). We have no formal explanation for this apparent discrepancy, but one is that blood sampling was done between week 28 and 32 of pregnancy in the Megert’s study, whereas in ours it was at 39.3 1.3 week of amenorrhea. In support of this assumption, Figure 1, B, of the present paper shows that classical monocytes are numerically lower in pPROM (31.8 2.46 weeks of amenorrhea) than in
JANUARY 2014 American Journal of Obstetrics & Gynecology
70.e6
Research
Obstetrics
healthy pregnancies, even if statistical significance was not reached. We also showed that the proportion of these monocytes returned to nonpregnant values at the time of spontaneous labor, whereas plasma levels of MCP-1 in women who delivered after spontaneous labor and those who delivered before labor did not differ. It has recently been reported that MCP-1/Ccl2 mRNA expression increases in pregnant rat leukocyte between day 17 and day 20 of pregnancy with a significant decrease close to delivery (day 22).24 This study suggests that MCP1eexpressing leukocytes might be involved earlier in rat than in human parturition. Comparison of both this study and our report must be conducted cautiously as human and rodent processes leading to parturition are not exactly the same and as Gomez-Lopez assessed mRNA expression, whereas we targeted intracellular expression of MCP1. Interestingly patterns of these changes were the same in healthy and pPROM pregnancies despite a significant difference in term at delivery (39.3 1.3 vs 31.8 2.46), suggesting spontaneous labor increases MCP-1expressing monocytes, irrespective of the term. Our results also suggest that labor, per se, is associated with granulocyte activation. The role of granulocyte and monocyte activation in the delivery process is controversial. Indeed, whereas rodent studies have suggested that neither neutrophils25 nor the chemokine receptor CCR226 are required for successful initiation of labor, human studies support a role for these cells for labor. Gervasi et al27 described specific phenotypic and metabolic changes in maternal granulocytes and monocytes in women with preterm labor with intact membranes, and Osman et al9 showed that leukocytes (predominantly neutrophils and macrophages) infiltrate the uterine cervix coincident with the onset of labor. In our study, we found no significant differences in MCP-1expressing monocytes according to the type of pregnancy, either normal or complicated with pPROM, suggesting that this proportion is not affected by this specific complication, whereas Melgert et al19 reported
www.AJOG.org that this proportion was significantly decreased in preeclamptic pregnancies compared with normal pregnancies or nonpregnant women, suggesting disease specific changes. On the other hand, Naruse et al28 reported that maternal plasma levels of MCP-1 were significantly higher in pregnancies complicated with severe preeclampsia (PE) than in normal pregnancies, but Melgert et al19 found a same percentage of MCP-1 expressing monocytes in normal and PE pregnancies. We have no formal explanations for these discrepancies between the Melghert et al19 and the Naruse28 studies, except that they did not measure the same factor, plasmatic or intracellular MCP-1. We also found that CA was not associated with significant changes in either the relative proportion of monocyte subsets, MCP-1expressing monocytes, or with variations in maternal plasma levels of MCP-1. This is not an unexpected finding as it has been shown that the consequences of fetal exposure to intraamniotic infection is associated with phenotypic and metabolic changes in granulocytes and monocytes, consistent with activation in fetal but not in maternal immune cells.29 As MCP-1 initiates cellular migration and the activation of monocytes/macrophages and granulocytes, the reduced expression observed in classical and intermediate monocytes (CD14high cells) during pregnancy might reduce the ability of these cells to recruit other leukocytes from maternal blood to gestational tissues and contribute to fetal protection. Phagocytosis of trophoblast debris might be involved in maternofetal tolerance as it reduces the expression of several factors involved in the immune response, including cell-surface expression of MHC-II molecules and MCP-1, and induces the secretion of antiinflammatory factors and decreased the secretion of proinflammatory factors by macrophages.30 Nevertheless, some authors suggested that intraamniotic fluid infection (IAFI) was associated with higher levels of amniotic MCP-1 than in women without infection.31 In our study, we assessed plasma, and not amniotic fluid, levels of MCP-1, which may have been too low to reveal any differences.
70.e7 American Journal of Obstetrics & Gynecology JANUARY 2014
FIGURE 4
Circulating MCP-1 levels in laboring and nonlaboring women
MCP-1, macrophage chemoattractant protein-1. Bardou. MCP-1 expressing monocytes and labor. Am J Obstet Gynecol 2014.
Plasma levels of MCP-1 were no different according to the mode of delivery, whereas studies conducted in mice demonstrated that the administration of lipopolysaccharide (LPS) induced parturition within 12 to 24 hours and that this event was preceded by a peak in MCP-1 protein expression in the maternal blood, occurring 6 hours after the injection of LPS.15 The study by Esplin et al,31 cited above, showed that amniotic fluid levels of MCP-1 were higher in women who delivered preterm than in women who delivered at term after preterm labor, and even higher in those who delivered preterm with IAFI (median in pg/mL: 1543, 764, and 17,170, respectively, P < .001) suggesting that MCP-1 is involved in preterm labor even without infection. The implication of MCP-1 in human labor has been recently confirmed in a study showing a 3.9-fold increase in decidual Ccl2 mRNA expression in term laboring vs term nonlaboring women.32 But in support of our results these authors did not find significant differences in Ccl2 maternal plasma levels in term laboring and nonlaboring women. The apparent discrepancy with our results may be explained by the fact that the MCP-1 found in CA may not have been secreted by maternal monocytes. Indeed, recent results suggested that fetal macrophages, but not maternal macrophages, recruited by MCP-1 secreted by local fibroblasts, contribute to placental infiltration in cases of CA.33
Obstetrics
www.AJOG.org
FIGURE 5
CCR2 MFI in granulocytes from pPROM women in labor compared with those without labor
A, Gating strategy used to determine CCR2 expression in granulocytes. The granulocyte population was gated on a forward/side scatter. B, Representative CCR2 staining in granulocytes from healthy pregnant woman not in labor (red) and in labor (green). C, Results are presented as mean value SEM. The significance of differences between the groups in each subset was analyzed using a 1-way ANOVA followed by a Mann-Whitney U test. Values of P considered significant are indicated. Black line ¼ not in labor (n ¼ 27), grey line ¼ in labor (n ¼ 19). ANOVA, analysis of variance; MFI, mean fluorescence intensity; pPROM, preterm premature rupture of membranes. *P < .05; **P < .01; and ***P < .001. Bardou. MCP-1 expressing monocytes and labor. Am J Obstet Gynecol 2014.
Few human studies have focused on maternal serum or intracellular MCP-1 concentrations, but Yuan et al16 recently demonstrated that both full-term and preterm labor were associated with an increase in the proportion of monocytes and that expression of MCP-1 mRNA in leukocytes was also increased. Unfortunately, the authors did not investigate which type of cells contributed to this
increase. Thus, our work builds on the observations of Yuan et al16 because we observed an increase in MCP-1 expression at the protein level and defined which monocyte subsets were involved in this increase. MCP-1 (CCL-2) binds to its receptor, CCR2, mostly expressed on polymorphonuclear leukocytes (PMNs) and monocytes/macrophages, leading to the
Research
recruitment and extravasation of these cells from the peripheral blood stream to specific tissues. Even though it has been demonstrated that there are high levels of granulocytes and macrophages in uterine tissues during pregnancy until just before the onset of labor,34 the sequence of activation of these cells is poorly known. It has been suggested, however, that decidual inflammation precedes labor. Indeed, it has been shown that in human decidua, macrophage numbers were 4 times higher in full-term labor and 2.5 times higher in noninfection-associated PTL than in full-term nonlaboring samples.35 In this study, we observed that CCR-2 expression was induced in granulocytes but not in monocytes as usually described for classical monocytes, in contradiction with a recently published study suggesting that pregnancy reduces percentages of CCR2þ monocytes of both classical (CD16) and inflammatory (CD16þ) subsets.36 As monocytes are precursors of macrophages and as in our study blood sampling was performed either in women who were not in labor or who had already engaged the delivery process, it could thus be speculated that during labor, the activation/recruitment of classical and intermediate monocytes precedes the activation of granulocytes. As it has been suggested in mice,37 this late activation of granulocyte might also be required for postpartum decidual involution. Even though the origin of cell types implicated in the secretion of MCP-1 by gestational tissue in parturition remains unclear, the accumulation of intracellular MCP-1 in peripheral blood monocytes may contribute to the immune blast that occurs at delivery. Indeed, after recruitment from blood and differentiation into macrophages, monocytes/macrophages may release accumulated intracellular MCP-1 leading to the massive infiltration of leukocytes, resulting in dilation of the cervix, rupture of the fetal membranes and the initiation of myometrial contractions, and labor onset. To summarize our results, we suggest that the downregulation of the proportion of MCP-1 observed during pregnancy might be necessary for maternofetal tolerance, and that the increase in this
JANUARY 2014 American Journal of Obstetrics & Gynecology
70.e8
Research
Obstetrics
proportion might be a marker of spontaneous labor. In addition, labor is associated with a significant increase in CCR2-expressing granulocytes. Rather than combining several markers, including plasmatic MCP-1 concentration to determine the true onset of labor, as described,38 we propose the proportion of MCP-1expressing monocytes as a useful tool, for example, in women with preterm labor, and we have an ongoing clinical trial (NCT01340222) that aims to investigate this hypothesis. REFERENCES 1. Kindzelskii AL, Ueki T, Michibata H, Chaiworapongsa T, Romero R, Petty HR. 6-phosphogluconate dehydrogenase and glucose-6-phosphate dehydrogenase form a supramolecular complex in human neutrophils that undergoes retrograde trafficking during pregnancy. J Immunol 2004;172:6373-81. 2. Luft BJ, Remington JS. The adverse effect of pregnancy on macrophage activation. Cell Immunol 1984;85:94-9. 3. Arruvito L, Sanz M, Banham AH, Fainboim L. Expansion of CD4þCD25þand FOXP3þ regulatory T cells during the follicular phase of the menstrual cycle: implications for human reproduction. J Immunol 2007;178:2572-8. 4. Winger EE, Reed JL. Low circulating CD4(þ) CD25(þ) Foxp3(þ) T regulatory cell levels predict miscarriage risk in newly pregnant women with a history of failure. Am J Reprod Immunol 2011;66:320-8. 5. Mansouri-Attia N, Oliveira LJ, Forde N, et al. Pivotal role for monocytes/macrophages and dendritic cells in maternal immune response to the developing embryo in cattle. Biol Reprod 2012;87:123. 6. Samstein RM, Josefowicz SZ, Arvey A, Treuting PM, Rudensky AY. Extrathymic generation of regulatory T cells in placental mammals mitigates maternal-fetal conflict. Cell 2012;150: 29-38. 7. Rinaldi SF, Hutchinson JL, Rossi AG, Norman JE. Anti-inflammatory mediators as physiological and pharmacological regulators of parturition. Expert Rev Clin Immunol 2011;7: 675-96. 8. Hunt JS. Immunologically relevant cells in the uterus. Biol Reprod 1994;50:461-6. 9. Osman I, Young A, Ledingham MA, et al. Leukocyte density and pro-inflammatory cytokine expression in human fetal membranes, decidua, cervix and myometrium before and during labour at term. Mol Hum Reprod 2003;9:41-5. 10. Sakamoto Y, Moran P, Bulmer JN, Searle RF, Robson SC. Macrophages and not granulocytes are involved in cervical ripening. J Reprod Immunol 2005;66:161-73. 11. Shynlova O, Tsui P, Dorogin A, Lye SJ. Monocyte chemoattractant protein-1 (CCL-2)
www.AJOG.org integrates mechanical and endocrine signals that mediate term and preterm labor. J Immunol 2008;181:1470-9. 12. Kamel RM. The onset of human parturition. Arch Gynecol Obstet 2010;281:975-82. 13. Gomez-Lopez N, Estrada-Gutierrez G, Jimenez-Zamudio L, Vega-Sanchez R, VadilloOrtega F. Fetal membranes exhibit selective leukocyte chemotaxic activity during human labor. J Reprod Immunol 2009;80:122-31. 14. Esplin MS, Peltier MR, Hamblin S, et al. Monocyte chemotactic protein-1 expression is increased in human gestational tissues during term and preterm labor. Placenta 2005;26: 661-71. 15. Diamond AK, Sweet LM, Oppenheimer KH, Bradley DF, Phillippe M. Modulation of monocyte chemotactic protein-1 expression during lipopolysaccharide-induced preterm delivery in the pregnant mouse. Reprod Sci 2007;14: 548-59. 16. Yuan M, Jordan F, McInnes IB, Harnett MM, Norman JE. Leukocytes are primed in peripheral blood for activation during term and preterm labour. Mol Hum Reprod 2009;15:713-24. 17. Auffray C, Sieweke MH, Geissmann F. Blood monocytes: development, heterogeneity, and relationship with dendritic cells. Annu Rev Immunol 2009;27:669-92. 18. Ziegler-Heitbrock L, Ancuta P, Crowe S, et al. Nomenclature of monocytes and dendritic cells in blood. Blood 2010;116:e74-80. 19. Melgert BN, Spaans F, Borghuis T, et al. Pregnancy and preeclampsia affect monocyte subsets in humans and rats. PLoS One 2012;7: e45229. 20. Redline RW, Faye-Petersen O, Heller D, Qureshi F, Savell V, Vogler C. Amniotic infection syndrome: nosology and reproducibility of placental reaction patterns. Pediatr Dev Pathol 2003;6:435-48. 21. Soehnlein O, Lindbom L. Phagocyte partnership during the onset and resolution of inflammation. Nat Rev Immunol 2010;10: 427-39. 22. Grage-Griebenow E, Flad HD, Ernst M. Heterogeneity of human peripheral blood monocyte subsets. J Leukoc Biol 2001;69:11-20. 23. Grage-Griebenow E, Zawatzky R, Kahlert H, Brade L, Flad H, Ernst M. Identification of a novel dendritic cell-like subset of CD64(þ)/CD16(þ) blood monocytes. Eur J Immunol 2001;31: 48-56. 24. Gomez-Lopez N, Tanaka S, Zaeem Z, Metz GA, Olson DM. Maternal circulating leukocytes display early chemotactic responsiveness during late gestation. BMC Pregnancy Childbirth 2013;13 Suppl 1:S8. 25. Timmons BC, Mahendroo MS. Timing of neutrophil activation and expression of proinflammatory markers do not support a role for neutrophils in cervical ripening in the mouse. Biol Reprod 2006;74:236-45. 26. Menzies FM, Khan AH, Higgins CA, Nelson SM, Nibbs RJ. The chemokine receptor CCR2 is not required for successful initiation of labor in mice. Biol Reprod 2012;86:118.
70.e9 American Journal of Obstetrics & Gynecology JANUARY 2014
27. Gervasi MT, Chaiworapongsa T, Naccasha N, et al. Phenotypic and metabolic characteristics of maternal monocytes and granulocytes in preterm labor with intact membranes. Am J Obstet Gynecol 2001;185:1124-9. 28. Naruse K, Noguchi T, Sado T, et al. Chemokine and free fatty acid levels in insulinresistant state of successful pregnancy: a preliminary observation. Mediators Inflamm 2012;2012:432575. 29. Kim SK, Romero R, Chaiworapongsa T, et al. Evidence of changes in the immunophenotype and metabolic characteristics (intracellular reactive oxygen radicals) of fetal, but not maternal, monocytes and granulocytes in the fetal inflammatory response syndrome. J Perinat Med 2009;37:543-52. 30. Abumaree MH, Chamley LW, Badri M, El-Muzaini MF. Trophoblast debris modulates the expression of immune proteins in macrophages: a key to maternal tolerance of the fetal allograft? J Reprod Immunol 2012;94: 131-41. 31. Esplin MS, Romero R, Chaiworapongsa T, et al. Monocyte chemotactic protein-1 is increased in the amniotic fluid of women who deliver preterm in the presence or absence of intra-amniotic infection. J Matern Fetal Neonatal Med 2005;17:365-73. 32. Hamilton SA, Tower CL, Jones RL. Identification of chemokines associated with the recruitment of decidual leukocytes in human labour: potential novel targets for preterm labour. PLoS One 2013;8:e56946. 33. Toti P, Arcuri F, Tang Z, et al. Focal increases of fetal macrophages in placentas from pregnancies with histological chorioamnionitis: potential role of fibroblast monocyte chemotactic protein-1. Am J Reprod Immunol 2011;65:470-9. 34. Houser BL. Decidual macrophages and their roles at the maternal-fetal interface. Yale J Biol Med 2012;85:105-18. 35. Hamilton S, Oomomian Y, Stephen G, et al. Macrophages infiltrate the human and rat decidua during term and preterm labor: evidence that decidual inflammation precedes labor. Biol Reprod 2012;86:39. 36. Bjorkander S, Heidari-Hamedani G, Bremme K, Gunnarsson I, Holmlund U. Peripheral monocyte expression of the chemokine receptors CCR2, CCR5 and CXCR3 is altered at parturition in healthy women and in women with systemic lupus erythematosus. Scand J Immunol 2013;77:200-12. 37. Shynlova O, Nedd-Roderique T, Li Y, Dorogin A, Nguyen T, Lye SJ. Infiltration of myeloid cells into decidua is a critical early event in the labour cascade and post-partum uterine remodelling. J Cell Molecul Med 2013;17: 311-24. 38. Holst RM, Hagberg H, Wennerholm UB, Skogstrand K, Thorsen P, Jacobsson B. Prediction of spontaneous preterm delivery in women with preterm labor: analysis of multiple proteins in amniotic and cervical fluids. Obstet Gynecol 2009;114:268-77.
www.AJOG .org
OBSTETRICS
Systemic increase in human maternal circulating CD14DCD16L MCP-1D monocytes as a marker of labor Marc Bardou, MD, PhD; Tarik Hadi, PhD; Guillaume Mace, MD; Matthieu Pesant, PhD; Julie Debermont; Marina Barrichon, MsC; Maeva Wendremaire, PharmD, PhD; Nicole Laurent, MD; Paul Sagot, MD; Fre´de´ric Lirussi, PharmD, PhD OBJECTIVES: To study the influence of pregnancy and labor on the proportion and level of activation of monocyte subpopulations in human pregnancy. STUDY DESIGN: Peripheral blood samples were obtained from healthy
nonpregnant women (n ¼ 6); women in the third-trimester of healthy pregnancies (n ¼ 18) and women with preterm premature rupture of membranes (n ¼ 46), just before delivery for the last 2 groups. Monocyte subpopulations were characterized by flow cytometry using CD14, CD16, and activation level using macrophage chemoattractant protein-1 (MCP-1) and CCR2 antibodies. RESULTS: The relative proportion of each monocyte subset in
nonpregnant women was similar to that in women with healthy or complicated pregnancies. However, pregnancy was associated with a significant decrease in MCP-1 expressing monocytes (79.5% 19.8% vs 9.3% 6.8% and 11.9% 8.3% for nonpregnant, healthy pregnancy, and preterm premature rupture of membranes
(respectively, P < .05). Spontaneous labor was associated with a return to nonpregnant values for the proportion of MCP-1 expressing monocytes in both normal (74.4% 16.9) and preterm premature rupture of membranes pregnancy (68.4% 35.6), irrespective of the mode of delivery (vaginal or cesarean section). This was not observed in women who delivered without spontaneous labor onset. CCR-2 (MCP-1 receptor) expression was not modified in monocytes at the time of labor, but was significantly increased in granulocytes (3646 1080 vs 7338 2718 for nonlaboring and laboring preterm premature rupture of membranes, respectively, P < .05) CONCLUSION: In light of previous reports of a role for MCP-1 in labor,
our results suggest the downregulation of activation levels of monocytes, via MCP-1 expression might be involved in maternofetal immune tolerance. Monocyte reactivation might be associated with labor. Key words: human pregnancy, monocytes, preterm labor, preterm rupture of membrane, prospective study
Cite this article as: Bardou M, Hadi T, Mace G, et al. Systemic increase in human maternal circulating CD14DCD16L MCP-1D monocytes as a marker of labor. Am J Obstet Gynecol 2014;210:70.e1-9.
P
regnancy and delivery are both complex immune situations involving numerous processes. During pregnancy, maternal immunity is reduced, with both neutrophils1 and monocytes2 displaying decreased functionality and
reactivity compared with those in nonpregnant women. This decrease in maternal immune response is necessary to prevent the fetal allograft from rejection and permits fetal cells to invade the uterine mucosa to establish the
From Centre d’Investigations Cliniques Plurithématique 803 (INSERM CIC-P 803) (Drs Bardou, Mr Hadi and Ms Debermont); Center for Lipid Research, INSERM (Institut National de la Santé et de la Recherche Médicale) (Drs Bardou, Wendremaire, and Lirussi, Mr Hadi, and Ms Barrichon); CHU de Dijon (Drs Bardou, Wendremaire, Sagot, and Lirussi, Mr Hadi, and Ms Barrichon); Service de Gynécologie et d’Obstétrique (Drs Mace and Sagot); Laboratoire de Pharmacologie-Toxicologie (Drs Wendremaire and Lirussi); and Service d’Anatomie Pathologique (Dr Laurent), Dijon, France, and Humanitas Clinical and Research Center, Leukocyte Biology Laboratory (Dr Pesant), University of Milan, Milan, Italy. Received May 2, 2013; revised July 27, 2013; accepted Aug. 27, 2013. This work was supported by a grant from the French Ministry of Health (PHRC interregional 2010) and from INSERM and Direction Générale de l’Offre de Soins (DGOS) (AAP-RCT 2011). The authors report no conflict of interest. Presented at the 58th annual meeting of the Society for Gynecologic Investigation, Miami, FL, March 16-19, 2011. Reprints will not be available. 0002-9378/$36.00 ª 2014 Mosby, Inc. All rights reserved. http://dx.doi.org/10.1016/j.ajog.2013.08.031
70.e1 American Journal of Obstetrics & Gynecology JANUARY 2014
placenta.3,4 The role of endometrial monocytes/macrophages and dendritic cells has recently been suggested as pivotal for successful pregnancy establishment in cattle.5 Samstein and colleagues recently suggested that maternalefetal tolerance to paternal alloantigens is an active process in which plasmatic Tregs specifically respond to paternal antigens to induce tolerance.6 In contrast, human parturition, either at term or preterm, is increasingly recognized as a sterile inflammatory event.7 Several studies have shown a massive influx of both neutrophils and monocytes/macrophages from maternal blood into the major effector tissues of parturition, ie, myometrium, cervix, and fetal membranes during labor.8-11 Human parturition is a complex phenomenon with several factors, such as hormonal changes, increased production of prostaglandins and release of inflammatory cytokine, interacting together.12
Obstetrics
www.AJOG.org This recruitment of maternal monocytes and granulocytes from maternal blood is induced by the secretion of chemotactic proinflammatory cytokines, such as interleukin-8 (IL-8), MIP-1a or macrophage chemotracant protein-1 (MCP-1).13 MCP-1, also known as C-C motif chemokine ligand 2 (CCL-2), is a major mediator of monocyte/macrophage infiltration at inflammatory sites both under physiologic and pathologic conditions. The role of MCP-1 during delivery was suggested by its presence in the myometrium and gestational membranes during full-term labor and in amniotic fluid during preterm labor.14 Experimental studies conducted in pregnant mice demonstrated that the administration of lipopolysaccharide (LPS) induces parturition within 12 to 24 hours and that this event is preceded by a peak in MCP-1 expression in the maternal bloodstream, occurring 6 hours after LPS injection.15 Beyond the contribution of cytokines to parturition, Yuan and colleagues16 demonstrated that full-term and preterm labor are also regulated by modifications in the proportion or number and activation status of peripheral blood monocytes and granulocytes. They concluded that monocyte and granulocyte priming precedes labor onset. Recent studies have highlighted the heterogeneity of human peripheral blood monocytes. Based on the recently proposed nomenclature, monocyte subsets can be discriminated by their differential expression of CD14 and CD16.17,18 Monocytes are thus characterized as CD14high CD16 classical (a major subset accounting for 70% to 90% of peripheral blood monocytes), intermediate CD14high CD16þ, and nonclassical CD14low CD16þ monocytes.18 Monocytes also differ phenotypically by chemokine-receptor expression such as CCR2 (receptor of MCP-1) and CX3CR1 (fractalkine receptor). Classical CD14high CD16 monocytes mainly express CCR2, whereas nonclassical CD14low CD16þ mainly express CX3CR1.17,18 Although the proportion of each monocyte subset has only been scarcely studied in pregnancy, it may vary with the onset of inflammatory events.19
Monocytes appear to play a central although complex role during pregnancy, as they are involved in both tolerance to the fetal allograph and in delivery. Thus this study aimed to investigate (1) the changes induced by pregnancy in the relative proportions and activation level of each monocyte subset, (2) the influence of preterm premature rupture of membranes (pPROM) on this proportion and activation level, and (3) the influence of intrauterine infection and labor onset in women with pPROM on the activation status of monocytes via MCP-1 and CCR-2 expression and that of granulocytes via CCR-2 expression.
M ATERIALS
AND
M ETHODS
Study design and patient recruitment The study was designed as a prospective, single center, observational study. Peripheral blood samples were obtained between January 2009 and April 2010, from healthy nonpregnant female volunteers (nonpregnant, NP, n ¼ 6), in the third-trimester of a healthy pregnancy (healthy pregnancies, HP, n ¼ 18), and from women with preterm premature rupture of membranes (pPROM, n ¼ 46). Inclusion criteria were as follows: pPROM defined by either vaginal pooling of amniotic fluid during the speculum examination or a positive Actim PROM TEST (Medix Biochemica, Kauniainen, Finland). Normal pregnancy: women admitted to our institution for full-term delivery following a healthy pregnancy. Nonpregnant women with childbearing potential: age between 18 and 40, regular cycle, during the first phase of the cycle, and without oral contraceptive use at the time of blood sampling. Exclusion criteria were as follows: for the pPROM group: hospitalization before 24 or after 34 weeks of gestation, delivery within 12 hours after admission, hemorrhagic placenta previa, HIV, HBV, or HCV infection, consent refusal. Diagnosis of chorioamnionitis: chorioamnionitis was assessed in all of the pregnancies including pPROM in the present study by a positive culture of the placenta and/or a histologic
Research
assessment of the placenta by a single pathologist (Dr Laurent) using validated criteria.20 All clinical and biologic data were collected prospectively and reported in the case report form. This study was approved by the French Medicines Agency (registration no. 2008-A00119-46), and local ethics committee (CPP-Est1, Dijon, France). Written informed consent was obtained from all volunteers.
Blood sampling and fluorescenceactivated cell sorting for the analysis of blood monocyte subsets Fresh blood samples were collected by venipuncture, either during the first phase of the cycle for nonpregnant women or before delivery during spontaneous labor or before labor induction or cesarean delivery for pregnant women. Women who had their blood sampling performed before labor induction were logically included in the nonlaboring group irrespective of the final mode of delivery (ie, vaginal or cesarean delivery). Blood was collected in EDTA separator tubes and promptly subjected to antigen characterization. Monocyte subpopulations were analyzed with monoclonal antibodies against CD14-FITC, CD16-APC-Cy7, CCR2-Alexa 647, and MCP1-PE (all from BD Bioscience). Whole blood (100 mL) was first subjected to erythrocyte lysis with ACK lysing buffer (ACK; BioWhittaker, Walkersville, MD) followed by centrifugation at 240 g, 4 C for 10 minutes. Pelleted leukocytes were then washed with phosphate buffered saline (PBS), centrifuged at 240 g, 4 C for 10 minutes and stained for 20 minutes at 4 C with CD14-FITC, CD16-APC-Cy7, and CCR2-Alexa 647 antibodies. Leukocytes were then fixed in PBS0.1% paraformaldehyde for 10 minutes at room temperature, washed with PBS and subjected to MCP1-PE intracellular staining in PBS-Tween 0.1%-BSA 1% for 1 hour at room temperature. Staining with the
JANUARY 2014 American Journal of Obstetrics & Gynecology
70.e2
Research
Obstetrics
corresponding isotype control was performed for each staining. Cytometric analysis was performed on an LSRII cytometer (Becton Dickinson) using fluorescence-activated cell sorting Diva 6.1.2 software (Becton Dickinson). Monocytes were gated in a forward scatter/side scatter dot plot and 4-color staining was analyzed within this primary gate. Monocyte subsets were defined according to their CD14/CD16 pattern and classified into classical, intermediates, and nonclassical as described above.18 The proportions of monocyte subsets were expressed as percentage of total monocytes. The percentage of each subset expressing CCR2 and MCP-1 were obtained by appropriate gating and expressed as a percentage of CCR2þ and MCP-1þ monocytes within these gates. CCR2 and MCP-1 expression in each monocyte subset was also determined by mean fluorescence intensity minus the respective isotype control (MFIMFI isotype).
Plasma levels of MCP-1 MCP-1 was measured in the plasma of women who by ELISA test using Quantikine ELISA kit (R&D systems, ref. DCP00). Analyses were performed in duplicate, according to manufacturer’s instruction. Quantification was done according to internal standard, without normalization to protein level. Statistical analysis Sample size calculation: as no data were available on the proportion of MCP-1 expressing monocyte at the time of study design, preliminary results were used to assess sample size. With a type 1 error of 0.025, 7 patients per group gave us a 0.80 power to detect a 65% decrease in the proportion of MCP-1 expressing monocytes in pregnant women. This MCP-1 expressing monocyte assessment took also part of a prospective study of women with pPROM, and it explain why the number of pPROM pregnancy is much higher than 7. Results were expressed as the percentage ( standard error to the mean, SEM) of each subpopulation
www.AJOG.org (classical, CD4highCD6; intermediates CD4highCD6þ; and nonclassical, CD4lowCD6þþ) among total monocytes and as median fluorescence intensity (MFI) for MCP-1 and CCR-2 in arbitrary units. Several analyses were performed according to our study objectives. First, the different subsets of monocytes in nonpregnant and women with healthy pregnancies were compared. They were then compared in healthy and pPROM pregnancies. Thereafter, the proportion and MFI of MCP-1 expressing monocytes was compared in women with healthy or with pPROM pregnancies, and then in those who delivered spontaneously after labor onset and those who underwent cesarean section performed before labor onset or labor induction. Mean and SEM for percentages of each subpopulation were generated on each of the variables. Tests were performed using both parametric and nonparametric methods (t tests and Mann-Whitney U tests for continuous variables, and Fisher exact tests for ordinal variables). All
analyses were performed using GraphPad Instat 3 (GraphPad Software Inc., La Jolla, CA).
R ESULTS Baseline characteristics of the population are presented in the Table. Mean maternal age was similar in the different groups. The average gestational age at delivery and subsequently birthweight were significantly lower in the pPROM group than in the normal pregnancy group. Most of the women with pPROM (18/46, 56.5%), and almost all of the women with an uncomplicated pregnancy (17/18, 94.4%) had a vaginal delivery. Among the women with a healthy pregnancy, 14 delivered after labor onset (14 natural vaginal delivery, no cesarean section during labor), 4 delivered before spontaneous labor onset (3 with induced vaginal delivery, and 1 with cesarean section not during labor). In the pPROM group, 21 women delivered after labor onset (16 natural
TABLE
Clinical characteristics of study participants Characteristic Number of women Age, y
Healthy Nonpregnant pregnancy 6 32 ( 5.9)
Birthweight, g
pPROM
18
46
29.71 ( 5.38)
30.04 ( 6.10)
3138.24 ( 404.80) 1803.4 ( 760.46)
Parity ( SD)
0.5 ( 1.22)
2.30 ( 1.30)
Gravidity ( SD)
1.5 ( 1.10)
1.15 ( 1.37)
Tobacco
3/6
Gestational age at delivery, wks
3/18
10/46
39.3 ( 1.3)
31.2 ( 2.46)
Vaginal (natural)
14 (77.7)
16 (34.8)
Vaginal (induced)
3 (16.7)
10 (21.7)
Cesarean section, during labor
0 (0)
Cesarean section, not during labor
1 (5.6)
15 (32.6)
0 (0)
22 (47.8)
Mode of delivery, n (%)
Confirmed chorioamnionitis pPROM, preterm premature rupture of membranes.
Bardou. MCP-1 expressing monocytes and labor. Am J Obstet Gynecol 2014.
70.e3 American Journal of Obstetrics & Gynecology JANUARY 2014
P value
5 (10.8)
.049
Obstetrics
www.AJOG.org
FIGURE 1
Gating strategy assessing the proportion of monocyte subsets
A and B, Gating strategy used to determine the proportion of each monocyte subset in nonpregnant and pregnant women. Monocytes were gated in a forward scatter/side scatter dot plot and CD14/ CD16 stainings were used to identify monocytes within this primary gate. Monocyte subsets were defined according to their CD14/CD16 pattern and we defined: classical monocytes as CD14high CD16, intermediate monocytes as CD14high CD16þ, and nonclassical monocytes as CD14low CD16þ. B and C, Relative proportion of each monocyte subset in nonpregnant women (NP), women with normal pregnancies before the onset of labor, A, in pPROM women and B, the influence of chorioamnionitis, on the relative proportion of each monocyte subset. Proportions of each monocytes subset are represented as mean percentage SEM. Bardou. MCP-1 expressing monocytes and labor. Am J Obstet Gynecol 2014.
vaginal deliveries and 5 cesarean sections during labor), and 25 delivered before spontaneous labor onset (10
induced vaginal deliveries and 15 cesarean sections before labor onset). Twenty-two (47.8%) women in the
Research
pPROM group had a confirmed chorioamnionitis vs none in the healthy pregnancy group. Influence of pregnancy and pregnancyrelated complications on the relative proportion of each monocyte subset. Figure 1, A, shows the gating strategy used to determine the relative proportion of each monocyte subpopulation presented in Figure 1, B and C. Figure 1, B, shows that nonpregnant women (NP), and pregnant women, with either a healthy (HP) or pPROM pregnancy had similar relative proportions of the classical (CD14highCD16), intermediate (CD14highCD16þ), and nonclassical monocytes (CD14lowCD16þþ) (classical: 95.0 3.2%, 92.4 5.3% and 94.0 3.3%; intermediate: 3.2 0.9%, 5.4 3.7% and 4.4 2.5%; and nonclassical; 1.7 1%, 1.6 1.7% and 1.5 1.2% for NP, HP, and pPROM women, respectively, P ¼ NS). Figure 1, C, shows that in women with pPROM, the presence of confirmed (see Method section) chorioamnionitis (CA) did not affect the relative proportion of each monocyte subset (classical: 93.5 3.2% vs 94.8 3.3%; intermediate: 4.8 2.6% vs 3.5 2.4%; and nonclassical: 1.7 1.5% vs 1.2 0.8%, respectively, in women with pPROM with or without CA, respectively, Figure 1, C).
MCP-1 expression in monocytes is reduced in normal and pPROM pregnancies Figure 2 shows that, compared with nonpregnant women (Figure 2, A), nonlaboring pregnant women, with either a healthy or pPROM pregnancy (Figure 2, B), had a significantly lower proportion of MCP-1 expressing monocytes (MCP-1þ) (79.5% 19.75, n ¼ 6; 9.3% 6.8, n ¼ 4 and n ¼ 25 for NP, nonlaboring HP, and nonlaboring pPROM, respectively; P < .001) (Figure 2, D, black bars). When blood sampling was performed in pregnant women before the onset of spontaneous labor, mean fluorescence intensity was also decreased, compared with nonpregnant women (Figure 2, D, red, green and blue lines).
JANUARY 2014 American Journal of Obstetrics & Gynecology
70.e4
Research
Obstetrics
Labor induces MCP-1 expression in both classical and intermediate monocytes Moreover, Figure 2, C and D, also shows that spontaneous labor was associated with a marked and statistically significant increase in the percentage of MCP-1-expressing monocytes; proportions returned to a nonpregnant state in both HP women (73.4% 23.3% vs 9.3% 6.8%, for laboring and nonlaboring HP, respectively, P < .001) and women with pPROM (68.4% 35.6% vs 11.9% 8.3% for laboring and nonlaboring pPROM, respectively, P < .001). In women with pPROM, the proportion of MCP-1 expressing monocytes, according to the presence or absence of spontaneous labor, was not statistically different according to the mode of delivery (nonlaboring women: 13.41% 8.03 vs 10.86% 5.63 [P ¼ .429]; and laboring women 50.8% 31.04 vs 75.15% 33.63 [P ¼ .276] for cesarean or vaginal delivery, respectively). No such comparison was feasible for women with healthy pregnancies as all of them had a vaginal delivery. In laboring (n ¼ 21) and nonlaboring (n ¼ 25) women with pPROM, we further investigated the contribution of each monocyte subset to MCP-1 expression. Figure 3, B, shows that labor onset was associated with an increase in the proportion of MCP-1eexpressing intermediates (73.7% 16.8% vs 10.6% 8.0% in laboring and nonlaboring pPROM, respectively, P < .001) and classical monocytes (74.4% 19.9% vs 9.3% 6.8% in laboring and nonlaboring pPROM, respectively, P < .001). No significant differences between laboring and nonlaboring women with pPROM were observed in the very low proportions of nonclassical MCP-1expressing monocytes (0.8% 0.5%, vs 0.3% 0.2%, in laboring and nonlaboring pPROM, respectively, P ¼ NS). As shown in Figure 3, C, the increased proportion of MCP-1 expressing intermediate and classical monocytes in women in labor was paralleled by a significant increase in the median MCP-1
www.AJOG.org
FIGURE 2
Influence of pregnancy and labor on the percentage of monocytes expressing MCP-1, in normal and pPROM pregnancies
MCP-1 staining: presented granularity/MCP-1 dot plots are representative of the MCP-1 staining obtained in monocytes for A, nonpregnant women and B, pregnant woman not in labor. C, Considered significant are indicated. Nonpregnant (n ¼ 6), normal pregnancy (n ¼ 18, with 14 women with spontaneous labor and 4 without), pPROM (n ¼ 46 with 27 women with spontaneous labor, and 19 without). D, shows a representative overlay of the MCP-1 stainings obtained in monocytes from nonpregnant, healthy pregnant, women with pPROM (not in labor and in labor). MCP-1, macrophage chemoattractant protein-1; pPROM, preterm premature rupture of membranes. **P < .01 between nonpregnant and pregnant nonlaboring women, black bars; yP < .05; between spontaneous labor, and the absence of labor, grey and black bars, respectively. Bardou. MCP-1 expressing monocytes and labor. Am J Obstet Gynecol 2014.
fluorescence intensity, (expressed in arbitrary density unit, AU; intermediate: 1148 AU 370 vs 299 AU 177, and classical monocytes: 1083 AU 346 vs 241 AU 172, for laboring and nonlaboring pPROM women, respectively, P < .001), suggesting that these 2 monocyte subpopulations had an
70.e5 American Journal of Obstetrics & Gynecology JANUARY 2014
activated status. No significant difference was observed in the nonclassical monocyte subpopulation (137.7 AU 44.7 vs 80.3 AU 38.2 for laboring and nonlaboring pPROM women, respectively, P ¼ NS). In contrast plasma levels of MCP-1 in laboring and nonlaboring women were
Obstetrics
www.AJOG.org
Research
expression in any of the 3 subsets (data not shown).
FIGURE 3
MCP-1-expressing monocytes and MFI: influence of labor
C OMMENT
A, Gating strategy used to determine the percentage of MCP-1 expressing monocytes in each subset. Classical, intermediates, and nonclassical monocytes were characterized as described previously, B and C, MCP-1 staining overlay in each subpopulation: in women B, with pPROM B, not in labor and C, in labor. D and E, Percentage of MCP-1 expressing D, monocytes and E, MCP-1 MFI in classical, intermediates and nonclassical monocytes from women with pPROM (in labor vs not in labor). Results are represented as mean value SEM. The significance of differences between the groups in each subset was analyzed using a 1-way ANOVA followed by a Mann-Whitney U test. Values of P considered significant are indicated. Grey bars ¼ labor (n ¼ 27), black bars ¼ not in labor (n ¼ 19). ANOVA, analysis of variance; MCP-1, macrophage chemoattractant protein-1; MFI, mean fluorescence intensity; pPROM, preterm premature rupture of membranes. *P < .05; **P < .01. Bardou. MCP-1 expressing monocytes and labor. Am J Obstet Gynecol 2014.
not different as it is shown on Figure 4 (in pg/mL, 119.5 15.8 vs 133.5 22.9, respectively, P ¼ .613).
Influence of labor on the proportion of CCR2-expressing monocytes and granulocytes We finally investigated the impact of labor on the expression of CCR2 (MCP-1 receptor) in each monocyte subset and
in granulocytes. Figure 5, A, shows the gating strategy used to assess CCR2 expression in granulocytes. Figure 5, B, shows that labor was associated with a marked and statistically significant increase in granulocytes CCR2 MFI (7338 AU 2718 vs 3646 AU 1080 for laboring and nonlaboring pPROM, respectively, P < .05). In contrast, labor did not influence monocyte CCR-2
In humans, CD14high CD16 classical monocytes are predominant because they represent 70-90% of peripheral blood monocytes, whereas nonclassical monocytes (including patrollers and intermediates) account for 10-30% of peripheral blood monocytes.17 Even though the physiologic roles of the different monocyte subsets are not fully understood, at least 2 subsets (classical and nonclassical) might have different roles during homeostasis and the inflammatory response.21 Indeed, studies suggest that classical monocytes initiate inflammatory activities and are weak producers of inflammatory cytokines but strong producers of reactive oxygen species (ROS), whereas nonclassical monocytes express high levels of inflammatory cytokines but produce limited amounts of ROS.22,23 In this study, we have shown that whereas pregnancy does not affect the relative proportion of monocyte subsets, either classical (CD14highCD16), intermediates (CD14highCD16þ), and nonclassical (CD14lowCD16D), even in the case of pPROM with confirmed CA, it is associated with a marked decrease in classical and intermediate MCP-1 expressing monocytes, irrespective of gestational age. This is in apparent contradiction with the study by Melgert et al19 suggesting that the percentage of classical monocyte is decreased during pregnancy (91% [83-98%] vs 94% [90-98%] for pregnant and nonpregnant women, respectively), whereas the percentage of combined nonclassical/intermediate monocytes was higher in pregnant women (8.5% [2.3-16.6%] vs 5.6% [1.9-9.5%]). We have no formal explanation for this apparent discrepancy, but one is that blood sampling was done between week 28 and 32 of pregnancy in the Megert’s study, whereas in ours it was at 39.3 1.3 week of amenorrhea. In support of this assumption, Figure 1, B, of the present paper shows that classical monocytes are numerically lower in pPROM (31.8 2.46 weeks of amenorrhea) than in
JANUARY 2014 American Journal of Obstetrics & Gynecology
70.e6
Research
Obstetrics
healthy pregnancies, even if statistical significance was not reached. We also showed that the proportion of these monocytes returned to nonpregnant values at the time of spontaneous labor, whereas plasma levels of MCP-1 in women who delivered after spontaneous labor and those who delivered before labor did not differ. It has recently been reported that MCP-1/Ccl2 mRNA expression increases in pregnant rat leukocyte between day 17 and day 20 of pregnancy with a significant decrease close to delivery (day 22).24 This study suggests that MCP1eexpressing leukocytes might be involved earlier in rat than in human parturition. Comparison of both this study and our report must be conducted cautiously as human and rodent processes leading to parturition are not exactly the same and as Gomez-Lopez assessed mRNA expression, whereas we targeted intracellular expression of MCP1. Interestingly patterns of these changes were the same in healthy and pPROM pregnancies despite a significant difference in term at delivery (39.3 1.3 vs 31.8 2.46), suggesting spontaneous labor increases MCP-1expressing monocytes, irrespective of the term. Our results also suggest that labor, per se, is associated with granulocyte activation. The role of granulocyte and monocyte activation in the delivery process is controversial. Indeed, whereas rodent studies have suggested that neither neutrophils25 nor the chemokine receptor CCR226 are required for successful initiation of labor, human studies support a role for these cells for labor. Gervasi et al27 described specific phenotypic and metabolic changes in maternal granulocytes and monocytes in women with preterm labor with intact membranes, and Osman et al9 showed that leukocytes (predominantly neutrophils and macrophages) infiltrate the uterine cervix coincident with the onset of labor. In our study, we found no significant differences in MCP-1expressing monocytes according to the type of pregnancy, either normal or complicated with pPROM, suggesting that this proportion is not affected by this specific complication, whereas Melgert et al19 reported
www.AJOG.org that this proportion was significantly decreased in preeclamptic pregnancies compared with normal pregnancies or nonpregnant women, suggesting disease specific changes. On the other hand, Naruse et al28 reported that maternal plasma levels of MCP-1 were significantly higher in pregnancies complicated with severe preeclampsia (PE) than in normal pregnancies, but Melgert et al19 found a same percentage of MCP-1 expressing monocytes in normal and PE pregnancies. We have no formal explanations for these discrepancies between the Melghert et al19 and the Naruse28 studies, except that they did not measure the same factor, plasmatic or intracellular MCP-1. We also found that CA was not associated with significant changes in either the relative proportion of monocyte subsets, MCP-1expressing monocytes, or with variations in maternal plasma levels of MCP-1. This is not an unexpected finding as it has been shown that the consequences of fetal exposure to intraamniotic infection is associated with phenotypic and metabolic changes in granulocytes and monocytes, consistent with activation in fetal but not in maternal immune cells.29 As MCP-1 initiates cellular migration and the activation of monocytes/macrophages and granulocytes, the reduced expression observed in classical and intermediate monocytes (CD14high cells) during pregnancy might reduce the ability of these cells to recruit other leukocytes from maternal blood to gestational tissues and contribute to fetal protection. Phagocytosis of trophoblast debris might be involved in maternofetal tolerance as it reduces the expression of several factors involved in the immune response, including cell-surface expression of MHC-II molecules and MCP-1, and induces the secretion of antiinflammatory factors and decreased the secretion of proinflammatory factors by macrophages.30 Nevertheless, some authors suggested that intraamniotic fluid infection (IAFI) was associated with higher levels of amniotic MCP-1 than in women without infection.31 In our study, we assessed plasma, and not amniotic fluid, levels of MCP-1, which may have been too low to reveal any differences.
70.e7 American Journal of Obstetrics & Gynecology JANUARY 2014
FIGURE 4
Circulating MCP-1 levels in laboring and nonlaboring women
MCP-1, macrophage chemoattractant protein-1. Bardou. MCP-1 expressing monocytes and labor. Am J Obstet Gynecol 2014.
Plasma levels of MCP-1 were no different according to the mode of delivery, whereas studies conducted in mice demonstrated that the administration of lipopolysaccharide (LPS) induced parturition within 12 to 24 hours and that this event was preceded by a peak in MCP-1 protein expression in the maternal blood, occurring 6 hours after the injection of LPS.15 The study by Esplin et al,31 cited above, showed that amniotic fluid levels of MCP-1 were higher in women who delivered preterm than in women who delivered at term after preterm labor, and even higher in those who delivered preterm with IAFI (median in pg/mL: 1543, 764, and 17,170, respectively, P < .001) suggesting that MCP-1 is involved in preterm labor even without infection. The implication of MCP-1 in human labor has been recently confirmed in a study showing a 3.9-fold increase in decidual Ccl2 mRNA expression in term laboring vs term nonlaboring women.32 But in support of our results these authors did not find significant differences in Ccl2 maternal plasma levels in term laboring and nonlaboring women. The apparent discrepancy with our results may be explained by the fact that the MCP-1 found in CA may not have been secreted by maternal monocytes. Indeed, recent results suggested that fetal macrophages, but not maternal macrophages, recruited by MCP-1 secreted by local fibroblasts, contribute to placental infiltration in cases of CA.33
Obstetrics
www.AJOG.org
FIGURE 5
CCR2 MFI in granulocytes from pPROM women in labor compared with those without labor
A, Gating strategy used to determine CCR2 expression in granulocytes. The granulocyte population was gated on a forward/side scatter. B, Representative CCR2 staining in granulocytes from healthy pregnant woman not in labor (red) and in labor (green). C, Results are presented as mean value SEM. The significance of differences between the groups in each subset was analyzed using a 1-way ANOVA followed by a Mann-Whitney U test. Values of P considered significant are indicated. Black line ¼ not in labor (n ¼ 27), grey line ¼ in labor (n ¼ 19). ANOVA, analysis of variance; MFI, mean fluorescence intensity; pPROM, preterm premature rupture of membranes. *P < .05; **P < .01; and ***P < .001. Bardou. MCP-1 expressing monocytes and labor. Am J Obstet Gynecol 2014.
Few human studies have focused on maternal serum or intracellular MCP-1 concentrations, but Yuan et al16 recently demonstrated that both full-term and preterm labor were associated with an increase in the proportion of monocytes and that expression of MCP-1 mRNA in leukocytes was also increased. Unfortunately, the authors did not investigate which type of cells contributed to this
increase. Thus, our work builds on the observations of Yuan et al16 because we observed an increase in MCP-1 expression at the protein level and defined which monocyte subsets were involved in this increase. MCP-1 (CCL-2) binds to its receptor, CCR2, mostly expressed on polymorphonuclear leukocytes (PMNs) and monocytes/macrophages, leading to the
Research
recruitment and extravasation of these cells from the peripheral blood stream to specific tissues. Even though it has been demonstrated that there are high levels of granulocytes and macrophages in uterine tissues during pregnancy until just before the onset of labor,34 the sequence of activation of these cells is poorly known. It has been suggested, however, that decidual inflammation precedes labor. Indeed, it has been shown that in human decidua, macrophage numbers were 4 times higher in full-term labor and 2.5 times higher in noninfection-associated PTL than in full-term nonlaboring samples.35 In this study, we observed that CCR-2 expression was induced in granulocytes but not in monocytes as usually described for classical monocytes, in contradiction with a recently published study suggesting that pregnancy reduces percentages of CCR2þ monocytes of both classical (CD16) and inflammatory (CD16þ) subsets.36 As monocytes are precursors of macrophages and as in our study blood sampling was performed either in women who were not in labor or who had already engaged the delivery process, it could thus be speculated that during labor, the activation/recruitment of classical and intermediate monocytes precedes the activation of granulocytes. As it has been suggested in mice,37 this late activation of granulocyte might also be required for postpartum decidual involution. Even though the origin of cell types implicated in the secretion of MCP-1 by gestational tissue in parturition remains unclear, the accumulation of intracellular MCP-1 in peripheral blood monocytes may contribute to the immune blast that occurs at delivery. Indeed, after recruitment from blood and differentiation into macrophages, monocytes/macrophages may release accumulated intracellular MCP-1 leading to the massive infiltration of leukocytes, resulting in dilation of the cervix, rupture of the fetal membranes and the initiation of myometrial contractions, and labor onset. To summarize our results, we suggest that the downregulation of the proportion of MCP-1 observed during pregnancy might be necessary for maternofetal tolerance, and that the increase in this
JANUARY 2014 American Journal of Obstetrics & Gynecology
70.e8
Research
Obstetrics
proportion might be a marker of spontaneous labor. In addition, labor is associated with a significant increase in CCR2-expressing granulocytes. Rather than combining several markers, including plasmatic MCP-1 concentration to determine the true onset of labor, as described,38 we propose the proportion of MCP-1expressing monocytes as a useful tool, for example, in women with preterm labor, and we have an ongoing clinical trial (NCT01340222) that aims to investigate this hypothesis. REFERENCES 1. Kindzelskii AL, Ueki T, Michibata H, Chaiworapongsa T, Romero R, Petty HR. 6-phosphogluconate dehydrogenase and glucose-6-phosphate dehydrogenase form a supramolecular complex in human neutrophils that undergoes retrograde trafficking during pregnancy. J Immunol 2004;172:6373-81. 2. Luft BJ, Remington JS. The adverse effect of pregnancy on macrophage activation. Cell Immunol 1984;85:94-9. 3. Arruvito L, Sanz M, Banham AH, Fainboim L. Expansion of CD4þCD25þand FOXP3þ regulatory T cells during the follicular phase of the menstrual cycle: implications for human reproduction. J Immunol 2007;178:2572-8. 4. Winger EE, Reed JL. Low circulating CD4(þ) CD25(þ) Foxp3(þ) T regulatory cell levels predict miscarriage risk in newly pregnant women with a history of failure. Am J Reprod Immunol 2011;66:320-8. 5. Mansouri-Attia N, Oliveira LJ, Forde N, et al. Pivotal role for monocytes/macrophages and dendritic cells in maternal immune response to the developing embryo in cattle. Biol Reprod 2012;87:123. 6. Samstein RM, Josefowicz SZ, Arvey A, Treuting PM, Rudensky AY. Extrathymic generation of regulatory T cells in placental mammals mitigates maternal-fetal conflict. Cell 2012;150: 29-38. 7. Rinaldi SF, Hutchinson JL, Rossi AG, Norman JE. Anti-inflammatory mediators as physiological and pharmacological regulators of parturition. Expert Rev Clin Immunol 2011;7: 675-96. 8. Hunt JS. Immunologically relevant cells in the uterus. Biol Reprod 1994;50:461-6. 9. Osman I, Young A, Ledingham MA, et al. Leukocyte density and pro-inflammatory cytokine expression in human fetal membranes, decidua, cervix and myometrium before and during labour at term. Mol Hum Reprod 2003;9:41-5. 10. Sakamoto Y, Moran P, Bulmer JN, Searle RF, Robson SC. Macrophages and not granulocytes are involved in cervical ripening. J Reprod Immunol 2005;66:161-73. 11. Shynlova O, Tsui P, Dorogin A, Lye SJ. Monocyte chemoattractant protein-1 (CCL-2)
www.AJOG.org integrates mechanical and endocrine signals that mediate term and preterm labor. J Immunol 2008;181:1470-9. 12. Kamel RM. The onset of human parturition. Arch Gynecol Obstet 2010;281:975-82. 13. Gomez-Lopez N, Estrada-Gutierrez G, Jimenez-Zamudio L, Vega-Sanchez R, VadilloOrtega F. Fetal membranes exhibit selective leukocyte chemotaxic activity during human labor. J Reprod Immunol 2009;80:122-31. 14. Esplin MS, Peltier MR, Hamblin S, et al. Monocyte chemotactic protein-1 expression is increased in human gestational tissues during term and preterm labor. Placenta 2005;26: 661-71. 15. Diamond AK, Sweet LM, Oppenheimer KH, Bradley DF, Phillippe M. Modulation of monocyte chemotactic protein-1 expression during lipopolysaccharide-induced preterm delivery in the pregnant mouse. Reprod Sci 2007;14: 548-59. 16. Yuan M, Jordan F, McInnes IB, Harnett MM, Norman JE. Leukocytes are primed in peripheral blood for activation during term and preterm labour. Mol Hum Reprod 2009;15:713-24. 17. Auffray C, Sieweke MH, Geissmann F. Blood monocytes: development, heterogeneity, and relationship with dendritic cells. Annu Rev Immunol 2009;27:669-92. 18. Ziegler-Heitbrock L, Ancuta P, Crowe S, et al. Nomenclature of monocytes and dendritic cells in blood. Blood 2010;116:e74-80. 19. Melgert BN, Spaans F, Borghuis T, et al. Pregnancy and preeclampsia affect monocyte subsets in humans and rats. PLoS One 2012;7: e45229. 20. Redline RW, Faye-Petersen O, Heller D, Qureshi F, Savell V, Vogler C. Amniotic infection syndrome: nosology and reproducibility of placental reaction patterns. Pediatr Dev Pathol 2003;6:435-48. 21. Soehnlein O, Lindbom L. Phagocyte partnership during the onset and resolution of inflammation. Nat Rev Immunol 2010;10: 427-39. 22. Grage-Griebenow E, Flad HD, Ernst M. Heterogeneity of human peripheral blood monocyte subsets. J Leukoc Biol 2001;69:11-20. 23. Grage-Griebenow E, Zawatzky R, Kahlert H, Brade L, Flad H, Ernst M. Identification of a novel dendritic cell-like subset of CD64(þ)/CD16(þ) blood monocytes. Eur J Immunol 2001;31: 48-56. 24. Gomez-Lopez N, Tanaka S, Zaeem Z, Metz GA, Olson DM. Maternal circulating leukocytes display early chemotactic responsiveness during late gestation. BMC Pregnancy Childbirth 2013;13 Suppl 1:S8. 25. Timmons BC, Mahendroo MS. Timing of neutrophil activation and expression of proinflammatory markers do not support a role for neutrophils in cervical ripening in the mouse. Biol Reprod 2006;74:236-45. 26. Menzies FM, Khan AH, Higgins CA, Nelson SM, Nibbs RJ. The chemokine receptor CCR2 is not required for successful initiation of labor in mice. Biol Reprod 2012;86:118.
70.e9 American Journal of Obstetrics & Gynecology JANUARY 2014
27. Gervasi MT, Chaiworapongsa T, Naccasha N, et al. Phenotypic and metabolic characteristics of maternal monocytes and granulocytes in preterm labor with intact membranes. Am J Obstet Gynecol 2001;185:1124-9. 28. Naruse K, Noguchi T, Sado T, et al. Chemokine and free fatty acid levels in insulinresistant state of successful pregnancy: a preliminary observation. Mediators Inflamm 2012;2012:432575. 29. Kim SK, Romero R, Chaiworapongsa T, et al. Evidence of changes in the immunophenotype and metabolic characteristics (intracellular reactive oxygen radicals) of fetal, but not maternal, monocytes and granulocytes in the fetal inflammatory response syndrome. J Perinat Med 2009;37:543-52. 30. Abumaree MH, Chamley LW, Badri M, El-Muzaini MF. Trophoblast debris modulates the expression of immune proteins in macrophages: a key to maternal tolerance of the fetal allograft? J Reprod Immunol 2012;94: 131-41. 31. Esplin MS, Romero R, Chaiworapongsa T, et al. Monocyte chemotactic protein-1 is increased in the amniotic fluid of women who deliver preterm in the presence or absence of intra-amniotic infection. J Matern Fetal Neonatal Med 2005;17:365-73. 32. Hamilton SA, Tower CL, Jones RL. Identification of chemokines associated with the recruitment of decidual leukocytes in human labour: potential novel targets for preterm labour. PLoS One 2013;8:e56946. 33. Toti P, Arcuri F, Tang Z, et al. Focal increases of fetal macrophages in placentas from pregnancies with histological chorioamnionitis: potential role of fibroblast monocyte chemotactic protein-1. Am J Reprod Immunol 2011;65:470-9. 34. Houser BL. Decidual macrophages and their roles at the maternal-fetal interface. Yale J Biol Med 2012;85:105-18. 35. Hamilton S, Oomomian Y, Stephen G, et al. Macrophages infiltrate the human and rat decidua during term and preterm labor: evidence that decidual inflammation precedes labor. Biol Reprod 2012;86:39. 36. Bjorkander S, Heidari-Hamedani G, Bremme K, Gunnarsson I, Holmlund U. Peripheral monocyte expression of the chemokine receptors CCR2, CCR5 and CXCR3 is altered at parturition in healthy women and in women with systemic lupus erythematosus. Scand J Immunol 2013;77:200-12. 37. Shynlova O, Nedd-Roderique T, Li Y, Dorogin A, Nguyen T, Lye SJ. Infiltration of myeloid cells into decidua is a critical early event in the labour cascade and post-partum uterine remodelling. J Cell Molecul Med 2013;17: 311-24. 38. Holst RM, Hagberg H, Wennerholm UB, Skogstrand K, Thorsen P, Jacobsson B. Prediction of spontaneous preterm delivery in women with preterm labor: analysis of multiple proteins in amniotic and cervical fluids. Obstet Gynecol 2009;114:268-77.