Maternal circulating leukocytes display early chemotactic responsiveness during late gestation
© Gomez-Lopez et al; licensee BioMed Central Ltd. 2013
Published: 31 January 2013
Skip to main content
© Gomez-Lopez et al; licensee BioMed Central Ltd. 2013
Published: 31 January 2013
Parturition has been widely described as an immunological response; however, it is unknown how this is triggered. We hypothesized that an early event in parturition is an increased responsiveness of peripheral leukocytes to chemotactic stimuli expressed by reproductive tissues, and this precedes expression of tissue chemotactic activity, uterine activation and the systemic progesterone/estradiol shift.
Tissues and blood were collected from pregnant Long-Evans rats on gestational days (GD) 17, 20 and 22 (term gestation). We employed a validated Boyden chamber assay, flow cytometry, quantitative real time-polymerase chain reaction, and enzyme-linked immunosorbent assays.
We found that GD20 maternal peripheral leukocytes migrated more than those from GD17 when these were tested with GD22 uterus and cervix extracts. Leukocytes on GD20 also displayed a significant increase in chemokine (C-C motif) ligand 2 (Ccl2) gene expression and this correlated with an increase in peripheral granulocyte proportions and a decrease in B cell and monocyte proportions. Tissue chemotactic activity and specific chemokines (CCL2, chemokine (C-X-C motif) ligand 1/CXCL1, and CXCL10) were mostly unchanged from GD17 to GD20 and increased only on GD22. CXCL10 peaked on GD20 in cervical tissues. As expected, prostaglandin F2α receptor and oxytocin receptor gene expression increased dramatically between GD20 and 22. Progesterone concentrations fell and estradiol-17β concentrations increased in peripheral serum, cervical and uterine tissue extracts between GD20 and 22.
Maternal circulating leukocytes display early chemotactic responsiveness, which leads to their infiltration into the uterus where they may participate in the process of parturition.
Parturition resembles an immunological response, which is characterized by the infiltration of immunological cells and secretion of immunological mediators (i.e. cytokines) in the maternal-fetal tissues [1–3]. A significant role for this immunological response in most (if not all) labors has been confirmed, regardless of the presence of infection or timing of delivery [4–6]. Given that inflammatory processes seem to activate parturition and are intertwined with early (or upstream), mid (or midstream) and late (or downstream) phases of the birth cascade, there are many research groups, including our group, focused on understanding how they trigger the termination of gestation. This knowledge will reveal mechanisms and possibly prevent pregnancy complications such as preterm birth.
Accordingly, it is essential to understand how this inflammatory stage is created in the maternal-fetal tissues. Leukocyte recruitment appears as the first step in the conditioning of this stage . There is evidence that activated leukocytes extravasate from the local circulation into these tissues [8, 9] by chemotactic processes and expression of specific chemokines [10–12]. Several chemokines have been studied in the maternal fetal tissues. Among the most relevant we will mention three of them: chemokine (C-C motif) ligand 2 (CCL2) or monocyte chemotactic protein-1 (MCP-1) that participates in the monocyte recruitment, chemokine (C-X-C motif) ligand 1 (CXCL1) or neutrophil-activating protein 3 (NAP-3) that participates in the neutrophil recruitment, and CXCL10 or interferon gamma-induced protein 10 (IP-10) that participates in the T-cell recruitment [7, 11]. Infiltrated leukocytes create this particular inflammatory stage, and along with maternal-fetal tissues, secrete other inflammatory mediators, including pro-inflammatory cytokines, prostaglandins and corticotrophin releasing hormone, that stimulate expression of uterine activation proteins (UAPs) . There are at least nine UAPs  and likely many more, including matrix metalloproteinases (MMPs) that are particularly important in converting the uterus of pregnancy to the uterus of labor [15, 16]. These actions are feed-forward and iterative thereby amplifying the inflammatory stage and birth cascade, which leads to the downstream events of parturition: membrane rupture, cervical dilatation, myometrial contractility, placental separation, and uterine involution . Concomitantly, the influence of progesterone (P4) in maintaining pregnancy and suppressing inflammation wanes and estradiol-17β (E2) levels rise [18, 19], thereby permitting and accelerating the changes leading to parturition.
The purpose of this initial, observational study was to clarify the timing of key relationships in the birth cascade to one another in a rat model. We hypothesized that an early event in parturition is due to an increased responsiveness in peripheral leukocytes to chemotactic stimuli expressed by the gestational tissues. This event precedes the expression of UAPs and the systemic progesterone/estradiol shift in rodents.
Animal protocols were approved by the University of Alberta Health Sciences Animal Policy and Welfare Committee (#625/03/11/C), and the experiments were conducted in accordance with the Guidelines and Policies of Canadian Council on Animal Care (SOP.RES:RA:0001.0). Timed-pregnant Long-Evans rats weighing 250-350 g were obtained from Charles River (Sherbrooke, PQ) on gestational day (GD) 11-12. Rats were housed in the University of Alberta Health Sciences Laboratory Animal Services.
Rats were euthanized according to approved local policy with a lethal dose of isoflurane (Halocarbon Products Corporation, USA) by inhalation in a large beaker on GD17, 20 or 22 (n=5 each). Parturition occurs at GD22.5. Blood by heart puncture (10 mL) and maternal-fetal tissues were collected in sterile conditions from post-mortem rats. Maternal-fetal tissues included uterus, cervix, fetal membranes and placenta. Tissues were washed in saline solution and placed immediately in RNALater (Ambion, Applied Biosystems, USA) or in liquid nitrogen (and then kept at -80ºC) until they were assayed.
Blood samples were immediately divided into two tubes: 5 mL into a heparinized vacuum tube (BD Vacutainer, USA) and 5 mL into a 15 mL centrifuge tube (Corning Incorporated, USA). Serum was isolated from the centrifuge tube and stored at -20ºC until assayed. Polymorphonuclear and mononuclear leukocytes were isolated from the heparinized tube using a Ficoll gradient (Polymorphoprep, Axis-Shield, Norton, USA) following the manufacturer’s instructions. Total leukocytes were then washed in 1X PBS and the pellet was resuspended and stored in 500 µL of RNALater or resuspended and incubated in supplemented RPMI medium 1640 (1% antibiotics [5,000 units of penicillin and 5,000 µg of streptomycin/mL] and 10% fetal bovine serum) (Invitrogen, USA) at 37ºC for 24 h. Following the incubation time these leukocytes were used in chemotaxis assays.
Tissues previously placed in liquid nitrogen were gradually defrosted. Tissues were cut in fragments of ~1-cm diameter and homogenized in 1 mL of DMEM (High Glucose 1X and 1% antibiotics) (Invitrogen) using a Polytron (Polytron PRO200 homogenizer, PRO Scientific, USA). These extracts were centrifuged at 4ºC, 12,000 x g for 30 min, repeating this last step until a clear supernatant was obtained. Protein extracts then were stored at -20ºC until use.
The chemotaxis assay was performed using a modified validated Boyden chamber assay (AP48, Neuro Probe, USA) . First, we investigated the onset of peripheral leukocyte migration or chemotaxis during gestation in response to uterine and cervical extracts obtained at term (GD22). Chemotaxis assays were performed using extracts from uterine and cervical tissues obtained on GD22 and maternal peripheral leukocytes obtained on GD17, 20 and 22. Second, we tested the migratory activity of leukocytes at term gestation (GD22) against uterine, cervical, fetal membrane and placental extracts obtained from GD17, 20 and 22. According to the above, 50 μL of leukocyte suspension containing 105 leukocytes was placed on top of the polycarbonate membrane (5µm pore size, #PFB5, Neuro Probe) with either extracts from uterine or cervical tissues as chemoattractant in the lower compartment. Chambers were incubated for 120 min at 37°C in humidified air containing 5% CO2. Incubation time was previously established by performing incubation time curves . Afterwards, chemoattracted leukocytes were removed from the lower compartment and centrifuged at 500 × g for 5 min at room temperature. The pellet was fixed with 500 µL of OptiLyse® B Lysing Solution (Beckman Coulter, USA) and counted by flow cytometry. The flow cytometer (FACSCANTO II, http://flowcytometry.med.ualberta.ca/ ) was set to analyze the samples for 30 s, which involved 104–105 events. The coefficient of variance of this method (inter and intra assay) is < 5%. Leukocytes attracted by the negative control (DMEM medium) were subtracted in all cases.
Blood taken by heart puncture was labeled with the following fluorochrome-conjugated anti-rat mAb: CD45-Alexa Fluor 488 for total leukocytes (clone OX-1, #202205, BioLegend, USA), OX43-PE for monocytes (sc-53109, Santa Cruz Biotechnology, USA), CD45R-PE for B cells (clone HIS24, #554881, BD Pharmingen, USA), CD161-Alexa Fluor 647 for NK cells (10/78, #203110, BioLegend) and CD3-FITC/CD4-PC7/CD8-APC for T cells (clones 1F4, OX-38, OX-8, #PN A32909; Beckman Coulter). Additionally, we identified mast cells using a primary purified mouse anti-rat Mast cell (clone AR32AA4, #551770, BD Pharmigen) and a secondary rabbit anti-mouse IgG2a-PE (sc-3765, Santa Cruz Biotechnology). Leukocytes were then fixed using 500 µL of OptiLyse B (Beckman Coulter), washed, and resuspended in 500 µL of 1X PBS to be analyzed by flow cytometry. The flow cytometer was set to analyze 105 events. The phenotype of leukocytes was analyzed within the CD45+ and CD3+ region, respectively. Granulocytes were identified as CD45+OX43-CD45R-CD161-CD3- by using the parameters of FS, SS and CD45 .
Primers used for RT-PCR
Optimal Temperature (ºC)
NCBIa Reference Sequences
f: 5’-gca CCC AAA CCG AAG TCA-3’
r: 5’-AAG CCA GCG TTC ACC AGA-3’
f: 5’- GGT GAG CCA AAG AAG GTC-3’
r: 5’-ACA CTG GGT AAA GGG AGG-3’
f: 5’-GCA GGT GTC CCA AAG AAG-3’
r: 5’-TCA AAG GTC CTG AAG TCC-3’
ELISAs for rat CXCL1 (#RCN100, R&D Systems, USA; sensitivity = 0.7 - 1.3 pg/mL; Intra- and inter-assay precision = 6 and 5.7%), CXCL10 (#E90371Ra, USCN Life Science, USA; sensitivity < 12.5 pg/mL; Intra- and inter-assay precision = 10 and 12%) and CCL2 (#KRC1011, Invitrogen; sensitivity < 8 pg/mL; Intra- and inter-assay precision = 5.7 and 8.4%) were performed in protein extracts of tissues following manufacturer’s instructions. ELISAs for P4 (#PG129S-100, Calbiotech, USA; sensitivity < 0.22 ng/ml; Intra- and inter-assay precision = 2.6 and 8.2 %) and E2 (#ES180S-100, Calbiotech; sensitivity < 3.94 pg/ml; Intra- and inter-assay precision = 8.1 and 5.8%) were performed in both serum and protein extracts of tissues following the manufacturer’s instructions. The protein concentration of protein extracts were measured with Protein Assay Reagent (Precision RedTM, Cytoskelton, USA) at UV-vis 600 nm by the ND-1000. Chemokine and hormonal concentrations were normalized with protein concentrations.
The data were examined initially by the Shapiro-Wilk test for normal distribution. The Kruskal-Wallis and Mann–Whitney U tests were used since the data were not normally distributed. Statistical analyses were performed using SPSS (SPSS Inc, USA), version 18.0. A P value of ≤ 0.05 was considered statistically significant.
While maternal-fetal tissue leukocyte chemotaxis represents the dynamic relationship between total tissue chemotactic capacity and overall leukocyte responsiveness, it is of considerable interest to examine the expression of individual chemokines in several tissues such as cervix, uterus, fetal membranes and placenta during late rat gestation. We therefore examined the mRNA and protein expression of CCL2, chemokine for monocytes, CXCL1 for neutrophils, and CXCL10 for T cells.
The data of this study demonstrate that maternal peripheral leukocytes display some of the earliest changes (by GD20) that precede parturition in the Long-Evans rat. The most telling of these is their ability to increase their migration through a Boyden chamber in response to a term uterine or cervical chemotactic stimulus. Next is the ability to increase their expression of Ccl2, and early changes in the relative proportions of monocytes, B cells and granulocytes. Later, between GD20 and 22, further changes occur in leukocytes, serum, uterus, cervix, fetal membranes and placenta. These include increased activity of chemotactic factors, changes in tissue and circulating levels of steroids, and the up-regulation of UAPs. These data suggest that an increased responsiveness of maternal peripheral leukocytes may be an early step in the birth cascade. Subsequent steps could include the invasion of uterine tissues by responsive leukocytes where they contribute to creating local inflammatory microenvironments that promote parturition through stimulation of chemokines and chemokine activity and the stimulation of UAP expression.
Long-Evans rats may be an appropriate model for immunological studies such as exploring the relationships between leukocyte migration responsiveness, chemokine activity, and uterine activation. Previous studies have demonstrated that Long-Evans rats alter their gestation length when treated with pro-inflammatory agents , whilst other rat strains such as the Sprague-Dawley does not appear to respond accordingly . Another advantage over the Sprague-Dawley rat model is that the Long-Evans dam is 50% or more larger, thereby facilitating collection of blood and tissues.
The Boyden chamber migration assay is a very sensitive technique for assessing leukocyte responsiveness to chemotactic signals . To our knowledge, this is the first time it has been used in an animal study with the purpose of delineating temporal events in the birth cascade. Previously, we used it to demonstrate that the human maternal-fetal interface, fetal membranes, and choriodecidua exhibit chemotaxis of maternal peripheral leukocytes and this increases during term labor [10, 12]. This chemotaxis correlated with specific chemokine expression and infiltration of neutrophils, macrophages and T cells [7, 11]. Here we used term chemotactic activity extracted from various tissues and tested it against leukocytes collected at different times in late gestation to demonstrate the changing responsiveness of leukocytes to the same chemotactic signals as gestation progressed.
First, we demonstrated that before term (GD20) maternal circulating leukocytes expressed very high levels of Ccl2, which correlates with their high responsiveness to be attracted by the uterine and cervical extracts. Increased migratory responsiveness is an indication of increased leukocyte activation as activated leukocytes often constitutively express high-affinity forms of integrins and chemokine receptors, which increase their ability to migrate in response to chemokines [25, 26]. Further studies will be required to prove that they are actually activated and therefore invade the uterus. The important message is that the timing of the increased migratory responsiveness occurs before term pregnancy, which means that this event precedes parturition.
CCL2 is produced by many cell types and is a potent chemoattractant for monocytes and basophils that express its receptors, CCR2 and CCR4. Besides acting as chemokine, CCL2 induces the expression of integrins required for chemotaxis . We suggest that Ccl2 is expressed by maternal circulating responsive leukocytes to facilitate the recruitment of monocytes and other leukocytes expressing its receptors into the reproductive tissues (e.g. uterus and cervix). In addition, our data showed that the proportion of monocytes in the maternal circulating blood decreases at this time. This last may mean that monocytes are recruited massively towards these tissues and therefore their proportions decrease in the maternal peripheral blood before term pregnancy.
The uterus and the cervix are interesting not only in their respective roles in parturition (contraction vs. softening and effacement), but each produced term chemotactic activity that attracted more leukocytes on GD20 than GD17. This suggests that these tissues also play a chemotactic role during late gestation.
Our data show that the uterus at term exhibits increased leukocyte chemotaxis and this coincides with high expression of Ccl2. CCL2 is largely identified with the recruitment of monocytes into the uterus and its levels increase throughout late gestation . Because our results showed that the protein levels of CCL2 do not increase on term pregnancy, we suggest that other chemokines (and other chemotactic factors) may be participating in this increased leukocyte chemotaxis that we observed. Future studies will perform protein microarrays in uterine extracts to investigate which chemotactic factors participate in this event during term pregnancy.
Although mRNA levels of Ccl2 and Cxcl1 in placenta and fetal membranes tended to increase at term gestation, these tissue extracts did not contain the protein nor exhibit differential chemotactic activity throughout late gestation (data not shown). This finding suggests that these tissues can express these chemokines and therefore they contribute in the creation of chemotactic gradients that promote leukocyte infiltration into the maternal-fetal interface (i.e. uterus and decidua); however, unlike the uterus they do not participate in the active protein-related chemotaxis of leukocytes. Besides contributing to Cxcl1 expression, fetal membranes may also contain external chemokines (e.g. uterine chemokines) because these tissues have a high content of extracellular matrix, and it has been established that chemokines have high affinity for extracellular matrix proteins .
Indeed, the level of cervical chemotactic activity far exceeded that from any other uterine tissue. Our data also showed that the cervix has a peak of CXCL10 on GD20. CXCL10 helps to recruit effector T cells , and it is also produced by granulocytes in an extremely pro-inflammatory microenvironment (e.g. labor). Granulocytes, monocytes and some T cells infiltrate the human cervix during late gestation . Monocytes seem to participate in the cervical remodeling before the onset of labor , granulocytes seem to play a key role at the postpartum period [34, 35]; however, the T cell function is unclear. Recently, we demonstrated that T cells infiltrate the cervical zone of the fetal membranes during normal labor and that these cells are associated with MMP-9 [36, 37], which also participates in surrounding tissue remodeling . We suggest that CXCL10 on GD20 may participate in the recruitment of T cells into the cervix, which could participate in the cervical softening before term pregnancy .
The fall in tissue and circulating P4 concentrations and the rise in E2 concentrations is expected and occurs as a consequence of luteolysis. P4 withdrawal contributes to the recruitment of immune cells into the rat uterus and cervix , and it is associated with the increase in UAPs, especially OTR and FP [14, 41].
Currently, we cannot explain the discrepancies between the mRNA levels/protein levels of chemokines and chemotactic activity in the maternal-fetal tissues. We can only speculate that chemotactic factors released by specific reproductive tissues may participate in the leukocyte recruitment into the maternal-fetal interface. However, given the specific properties of uterine chemotactic factors in contrast to those from other tissues in terms of the graded and increasing attraction of peripheral leukocytes as pregnancy wanes, it is obvious that this is a complicated area requiring considerable further study. Characterizing and mechanistic studies are needed to identify all key chemokines that exist, their specific receptors and the precise cellular adhesion molecules that actively participate in these events during pregnancy and term labor.
In conclusion we demonstrated that the maternal circulating leukocytes increase their responsiveness to migrate to term chemotactic stimuli, and this coincides with their high expression of Ccl2, and changes in relative proportions in peripheral blood. These events precede uterine tissue chemotactic activity expression, expression UAPs, and the changes in steroid hormone concentrations typical of the termination of pregnancy. We surmise that peripheral leukocyte responsiveness precedes leukocyte recruitment to specific gestational tissues, including the cervix and uterus . These create a tissue specific immunological microenvironment that promotes tissue activation leading to labor at term of pregnancy. Knowing these events in rats and other animal models and eventually translating these outcomes to women, will assist in identifying points of convergence of the birth cascade and new therapeutic targets to treat problems such as preterm delivery.
Chemokine (C-C motif) ligand 2*
Cluster of differentiation
Chemokine (C-X-C motif) ligand 1*
Chemokine (C-X-C motif) ligand 10*
Dulbecco's Modified Eagle Medium
Enzyme-linked immunosorbent assay
Prostaglandin F receptor*
Interferon gamma-induced protein 10
monocyte chemotactic protein-1
Matrix metallopeptidase 9
Neutrophil-activating protein 3
Quantitative real time-polymerase chain reaction
Uterine activation Proteins
Lower case italics were used for gene names and upper case normal fonts were used for protein names
This work was supported by the Alberta Innovates – Health Solutions Interdisciplinary Preterm Birth and Healthy Outcomes Team award (#200700595). NGL is sponsored by the Molly Towell Foundation for Perinatal Research.
This article has been published as part of BMC Pregnancy and Childbirth Volume 13 Supplement 1, 2013: Preterm Birth: Interdisciplinary Research from the Preterm Birth and Healthy Outcomes Team (PreHOT). The full contents of the supplement are available online at http://www.biomedcentral.com/bmcpregnancychildbirth/supplements/13/S1.
All of the publication fees will be funded by the Preterm Birth and Healthy Outcomes Team Interdisciplinary Team Grant (#200700595) from Alberta Innovates - Health Solutions, formerly the Alberta Heritage Foundation for Medical Research.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.