Human First-Trimester Trophoblast Cells Recruit CD56brightCD16– NK Cells into Decidua by Way of Expressing and Secreting of CXCL12/Stromal C
http://www.100md.com
免疫学杂志 2005年第13期
Abstract
More than 70% of decidual lymphocytes are NK cells characterized by CD56brightCD16– phenotype, but the mechanisms by which these NK cells are recruited in the decidua are still almost unrevealed. In this study, we first analyzed the transcription of 18 chemokine receptors in the first-trimester decidual CD56brightCD16– NK cells. Among these receptors, CXCR4 and CXCR3 were found highly transcribed, and the expression of CXCR4 was verified in most of the decidual CD56brightCD16– NK cells by flow cytometry. The first-trimester human trophoblasts were found expressing CXCL12/stromal cell-derived factor 1, the specific ligand of CXCR4, by way of in situ hybridization and immunohistochemistry. The primary cultured trophoblast cells were also found to secrete stromal cell-derived factor 1 spontaneously, and its concentration was 384.6 ± 90.7 pg/ml after the trophoblast cells had been cultured for 60 h. All of the ligands for CXCR3 were below the minimal detectable concentration when trophoblast cells were cultured for up to 48 h. Both recombinant human SDF-1 and supernatants of the cultured trophoblast cells exhibited chemotactic activity on decidual CD56brightCD16– NK cells. Our findings suggest that human first-trimester trophoblast cells produce CXCL12, which in turn chemoattracts decidual CD56brightCD16– NK cells. This activity could contribute to the recruitment mechanism of decidual lymphocytes, especially CD56brightCD16– NK cells, in decidua, and may be used at a local level to modulate the immune milieu at the materno-fetal interface.
Introduction
The mechanisms by which the human hemiallogeneic feto-placental unit is not rejected by the maternal immune system have been under intense attention, and it has now become clear that decidual lymphocytes have unique functions in local cytokine production, endovascular invasion, and placental development so as to keep pregnancy going smoothly (1, 2, 3, 4). More than 70% of decidual lymphocytes are NK cells characterized by the CD56brightCD16– phenotype (5, 6). In comparison to peripheral NK cells, decidual NK cells display different functional behaviors. These decidual NK cells have decreased killing activity against class I MHC-negative target cells and are also more susceptible to the inhibition mediated by MHC class I molecules (classical and nonclassical) expressed by extravillous cytotrophoblasts (7, 8, 9, 10). The enrichment in the decidua with this specific NK subset is outstanding compared with the 10% composition of NK cells in the peripheral blood, and the CD56brightCD16– population composes only 10–15% of the total peripheral NK population.
The origin of these decidual CD56brightCD16–NK cells is unknown, but because they closely resemble the minor agranular CD56bright NK cell population in the blood, one possibility is that this blood NK cell population might move into the uterus to proliferate, differentiate, enlarge, and acquire cytoplasmic granules in the hormone-rich mucosal microenvironment (11). Recently, uterine segment transplantation in mice has shown that uterine NK precursor cells do not self-renew in the uterus. Transplantable uterine NK precursor cells are found in spleen, lymph nodes, bone marrow, thymus, and liver (12), but the mechanisms by which the CD56brightCD16– NK cells are accumulated in the decidua are still almost unknown.
Chemokines constitute a family of closely related chemoattractant cytokines. Several groups of chemokines have been described and two major groups can be distinguished based on the presence (CXC) or absence (CC) of a single amino acid between two cysteine residues near the N terminus. Chemokine activity is mediated through cell surface receptors related to the 7-transmembrane domain receptor family. Chemokines were initially recognized to play a pivotal role in leukocyte communication and migration, both in physiological and pathological contexts. These molecules control immune cell trafficking and recirculation of the leukocyte population among the blood vessels, lymph, lymphoid organs, and tissues, an important process in host immune surveillance and in acute and chronic inflammatory responses (13, 14, 15). Now it has become evident that chemokines not only play a fundamental role in development, homeostasis, and immune system function, but also are important in the angiogenesis and angiostatic processes, CNS development, and HIV pathogenesis (16, 17, 18, 19).
Since chemokines and their receptors are key players in leukocyte communication and migration, it is very reasonable to speculate that chemokine receptors and their ligands are involved in the preferential accumulation of the CD56brightCD16– NK cells in the decidua: the chemokines at the materno-fetal interface act with the chemokine receptors expressed by CD56brightCD16– NK cells, then attract CD56brightCD16– NK cells to the decidua. In this study, we first analyzed the transcription of 18 chemokine receptors in first-trimester decidual CD56brightCD16– NK cells. Among these receptors, CXCR4 was found to be highly transcribed. The decidual CD56brightCD16– NK cells were then verified expressing CXCR4 by flow cytometric analysis. After that we detected the expression of CXCL12/stromal cell-derived factor 1 (SDF-1),2 the specific ligand of CXCR4, in the first-trimester human placentas. Finally, we analyzed the chemotactic activity of CXCL12 on decidual CD56brightCD16– NK cells. Our findings suggest that the first-trimester human trophoblast cells produce CXCL12, which in turn endows the trophoblast cells with the capacity to attract decidual CD56brightCD16– NK cells into the decidua.
Materials and Methods
Placental samples
The first-trimester human placentas and deciduas (gestational age, 5–10 wk) were obtained from clinically normal pregnancies, which were terminated for nonmedical reasons, at the Obstetrics and Gynecology Hospital of Fudan University. Each patient completed a signed, written consent form. The Fudan University Human Investigation Committee approved the form and the use of the placental tissue.
Isolation of decidual CD56brightCD16– NK cells
The pooled decidual tissues were trimmed into 1-mm pieces and enzymatically digested for 20 min, using vigorous shaking, with 1.5 mg type I DNase and 24 mg type IV collagenase (Sigma-Aldrich) present in 15 ml of RPMI 1640 medium. This procedure was repeated three times. After an additional 5-min incubation at room temperature without shaking, the supernatants were collected and loaded onto Ficoll density gradients to purify the lymphocyte population. Decidual CD56brightCD16– NK cells were further purified by incubation with microbeads of conjugated anti-human CD56 mAb (Miltenyi Biotec). Separation was performed with the AutoMACS instrument (Miltenyi Biotec).
Flow cytometric analysis
The evaluation of cell surface expression of CD3/CD56/CD16 was performed on immunopurified decidual NK cells by direct cell surface labeling after presaturation of cells at 4°C for 20 min with goat normal serum diluted 1/20 in PBS. The cells were then washed twice and incubated at 4°C for 20 min with FITC-conjugated mouse anti-human CD3, tricolor-conjugated mouse anti-human CD16, R-PE-conjugated mouse anti-human CD56 (Caltag Laboratories), or the relevant isotype control (Sino-America). The cell surface expression of CXCR4 was also assessed using PE-conjugated mouse anti-human CXCR4 (BD Pharmingen). All analysis was conducted on a BD Biosciences FACScan flow cytometer using CellQuest software. Three independent experiments were performed, each in triplicate. The purified decidual NK cells were derived from 15 decidual samples.
RT-PCR analysis
Total cellular RNA was extracted from freshly purified decidual CD56brightCD16– NK cells (cells derived from 15 decidual samples) using TRIzol reagent (Invitrogen Life Technologies). One microgram of total RNA was denatured, and reverse transcription was performed for 1 h at 42°C with 0.5 μg oligo(dT)18, 1.0 mM 4dNTP, 20 U RNasin RNase inhibitor (Promega), 200 U Moloney virus-reverse transcriptase (Superscript II; Invitrogen Life Technologies), and 5x reaction buffer in a total of 20 μl. Amplification was performed with 5 μl of cDNA, 0.2 mM dNTP, 1 mM MgCl2, 2.5 U AmpliTaq DNA polymerase (PerkinElmer), 0.8 mM specific sense and antisense primers, and 10x reaction buffer in a 50-μl reaction volume. The primers used for the detection of chemokine receptors and housekeeping gene GAPDH are indicated in Table 1. After a 5-min precycle at 95°C, the reaction was followed by 30 cycles of 1 min at 94°C, 1 min at 55°C, and 1 min at 72°C. When the final cycle was over, samples were kept at 72°C for 15 min to complete the synthesis. The PCR products were analyzed on 1.5% agarose gels and ethidium bromide-stained bands were photographed. Authenticity of all of the amplified PCR products was confirmed by direct sequencing. Three independent experiments were performed, each in triplicate.
In situ hybridization
Human CXCL12 in situ hybridization kits were obtained from Boshide. The oligonucleotide probes correspond to three sequences in CXCL12 mRNA (5'- AGGTCGTGGTCGTGCTGGTCCTCGTGCTGA-TCAAGCATCTCAAAATTCTCAACACTCCAA-CAGGAGTACCTGGAGAAAGCTTTAAACAAG-3'). Five first-trimester placentas collected for in situ hybridization were immediately fixed with 4% paraformaldehyde for 1 h at 4°C. They were then permeated with 12% sucrose and passed through an increasing concentration gradient of sucrose up to 18%. The specimens for frozen section were embedded with OCT compound, frozen at –80°C, and stored until further analysis. The frozen sections (8-μm thick) were prepared by a cryostat and treated with 10 μg/ml proteinase K. The slides were incubated for 2 h at 37°C in a prehybridization buffer. Hybridization was performed under coverslips overnight at 40°C. Following hybridization, the coverslips were removed and the slides were rinsed three times for 10 min with 2x SSC, then treated with 20 μg/ml RNase in 0.5 M sodium chloride and 10 mM Tris (pH 8.0) for 30 min at 37°C to digest the nonhybridization RNA probe. They were then washed three times for 20 min in 0.1x SSC at 42°C. Bound mRNA was detected with alkaline phosphatase-labeled anti-digoxigenin goat Ab and developed with 5-bromo-4-chloro-3-indolyl phosphate and nitroblue tetrazolium color development substrate.
Immunohistochemical staining
Five first-trimester placentas were collected for immunohistochemical staining. From the frozen specimens, serial sections of 8-μm thickness were prepared in a cryostat and fixed with cold acetone for 5 min. They were blocked sequentially with methanol containing 3% H2O2 and 20% human serum and then incubated with 10 μg/ml anti-CXCL12 mAb (R&D Systems) or 1:100 anti-human cytokeratin-7 mAb (DakoCytomation) or isotype-matched irrelevant IgG (Sino-America) overnight at 4°C. They were then sequentially treated with a biotinylated rabbit anti-mouse Ab and an HRP-labeled streptavidin. Development was performed by treating the sections with a liquid 3,3'-diaminobenzidine plus substrate kit (Sino-America). After counterstaining with hematoxylin, immunostaining of CXCL12 or cytokeratin-7 on the tissue sections was evaluated by light microscope.
Trophoblast cell culture
The first-trimester human placenta was separated carefully from the decidua under a stereomicroscope, minced into small fragments, and incubated in HBSS containing 0.25% trypsin (Invitrogen Life Technologies, 4.2 mM MgSO4, 25 mM HEPES, and 50 Kunitz/ml DNase type IV (Invitrogen Life Technologies) for 10 min at 37°C with gentle agitation. Then the suspension was taken and filtered through a stainless steel mesh of 100-μm pores. The trypsin digestion was stopped with 10% FCS (HyClone). The remaining placental tissues were subject to another three cycles of 10-min trypsinization. The four resultant cell suspensions were pooled, centrifuged at 300 x g for 10 min, and resuspended in 4 ml of DMEM (Invitrogen Life Technologies). This suspension was layered over a preformed Percoll gradient (Pharmacia) made up in HBSS. The gradient was made from 70% to 5% Percoll (v/v) in 5% steps of 2 ml each by dilutions of 90% Percoll with HBSS. The gradient was centrifuged at 1000 x g for 20 min. The middle layer (density, 1.048–1.062 g/ml) was removed and washed with DMEM. Cells were diluted with DMEM supplemented with 2 mM glutamine, 10% heat-inactivated FCS, 25 mM HEPES, 100 IU/ml penicillin, and 100 μg/ml streptomycin.
The isolated human trophoblast cells were plated in plastic petri dishes and incubated at 37°C for 40 min to allow the contaminating macrophages to adhere to the plastic. The nonadherent trophoblast cells were transferred to fresh plates as well as eight-chambered LabTek slides at a concentration of 8 x 105 cells/ml. Cells were cultured in DMEM supplemented with 2 mM glutamine, 10% heat-inactivated FCS, 25 mM HEPES, 100 IU/ml penicillin, and 100 μg/ml streptomycin at 37°C in 95% air and 5% CO2.
Immunocytochemistry
After 24 h of culture, trophoblast cells were fixed in 4% paraformaldehyde for 20 min at room temperature, washed in PBS, and permeabilized for 4 min in 0.3% Triton X-100-PBS. The cells were incubated with 20% human serum in PBS for 30 min to reduce nonspecific binding. Primary Abs diluted in PBS containing 1% BSA were added. Anti-human cytokeratin-7 and anti-human vimentin mAb (DakoCytomation) were used as markers for identification of trophoblast lineage and nontrophoblast lineage, respectively. The dilutions of anti-cytokeratin-7 and anti-vimentin were both 1/100. Isotype-matched irrelevant IgG (Sino-America) was used as a control. After incubation with primary Ab overnight at 4°C, the cells were washed in PBS-0.1% Tween 20, and then incubated with HRP-labeled secondary Ab (Sino-America) for 2 h at room temperature. The slides were stained with 3,3'-diaminobenzidine and counterstained with hematoxylin. The experiments were repeated five times.
ELISA
The supernatants of trophoblast cell cultures were harvested after 12, 24, 36, 48, and 60 h of culture. Each supernatant was centrifuged at 2000 x g and stored at –80°C. ELISAs were performed with human SDF-1, CXCL9, CXCL10, and CXCL11 kits (R&D Systems.) according to the instructions of the manufacturer. The minimal detectable concentrations for SDF-1, CXCL9, CXCL10, and CXCL11 were 18, 5, 5, 15 pg/ml, respectively. Five individual placental samples were tested for each chemokine. At least three wells’ supernatants were collected at each time point for each placental sample.
In vitro cell migration assay
The purified decidual CD56brightCD16– NK cells were resuspended in DMEM at a concentration of 3 x 106 cells/ml and incubated for 4 h with 100 ng/ml anti-human CXCR4 mAb (clone 12G5; BD Pharmingen) or 100 ng/ml isotype-matched irrelevant IgG (Sino-America). After that the decidual CD56brightCD16[minus ]NK cells of 3 x 105 in 100 μl of DMEM were loaded into each Transwell filter (5-μm pore filter Transwell, 24-well cell clusters; Corning). The filters were then plated in each well containing 600 μl of supernatant medium of trophoblast cell cultures (choose the supernatant medium in which the concentration of SDF-1 is >400 pg/ml) or DMEM supplemented with differing concentrations of recombinant human SDF-1 (rhSDF-1; R&D Systems). Three wells were used per condition per experiment. After a 3-h incubation at 37°C in 5% CO2, the upper chambers were removed and cells in the bottom chamber were collected, counted, and analyzed by flow cytometry. Results are expressed as the percentage of migrating cells (number of migrated cells/total number of input cells x 100). The experiments were repeated four times and the purified decidual NK cells were derived from 20 decidual samples.
Statistical analysis
All values are shown as mean ± SD. Data from the in vitro cell migration assay were performed using one-way ANOVA, with application of the Dunnett test. Differences were accepted as significant at p < 0.05.
Results
The decidual NK cells, being CD56brightCD16–, differ phenotypically from the majority of classical circulating NK cells. Using immunofluorescent labeling for the Ags CD56, CD16, and CD3 and subsequent flow cytometric analysis, it was confirmed that the purity of the decidual NK cells isolated in this study was >95% (Fig. 1).
Transcription of chemokine receptors in CD56brightCD16– NK cells
Using PCR primers specific for human chemokine receptors, we have analyzed the transcription of chemokine receptors in freshly isolated decidual CD56brightCD16– NK cells. Three independent experiments (each in triplicate) were done and the results were similar. The decidual CD56brightCD16– NK cells transcribed CCR1, CCR3, CCR5–CCR8, CXCR1–CXCR4, CXCR6, and CX3CR1 and the ratios of the band intensities of chemokine receptors to that of GAPDH were 0.81 ± 0.14, 0.78 ± 0.11, 0.59 ± 0.09, 0.71 ± 0.12, 0.89 ± 0.14, 0.53 ± 0.12, 0.91 ± 0.17, 0.78 ± 0.13, 1.09 ± 0.21, 1.47 ± 0.18, 0.92 ± 0.16 and 0.71 ± 0.16, respectively. Among these chemokine receptors, CXCR4 and CXCR3 were highly transcribed in CD56brightCD16– NK cells (the mean band intensities of these two chemokine receptors were higher than that of GAPDH). The mRNA of CCR2, CCR4, CCR9, CCR10, CXCR5, and XCR1 was never detected in CD56brightCD16– NK cells by RT-PCR (Fig. 2).
Having known that the first-trimester human trophoblast cells produce CXCL12 spontaneously, we assessed the chemotaxis of CXCL12 to the decidual CD56brightCD16– NK cells. The results showed that within a suitable range of concentration rhSDF-1 indeed induced the chemotaxis of decidual CD56brightCD16– NK cells, which had been preincubated with isotype-matched irrelevant IgG in a dose-dependent manner. A total of 21.60 ± 4.19% and 16.60 ± 2.78% of the input cells migrated to the bottom chamber containing 10 ng/ml and 100 ng/ml rhSDF-1, respectively (p < 0.001 compared with cells migrating toward DMEM only). In addition, the supernatant of trophoblast cells also showed chemotactic activity on the decidual CD56brightCD16– NK cells (p < 0.001 compared with cells migrating toward DMEM only). When the cells were preincubated with anti-CXCR4, they did not specifically migrate to the bottom chamber containing rhSDF-1 or the supernatant of trophoblast cells, which clearly demonstrates that the interaction between the CXCL12 secreted by human trophoblast cells and the CXCR4 expressed by the CD56brightCD16– NK cells directs the migration of the CD56brightCD16– NK cells (Fig. 7).
Discussion
It is very meaningful to explore how and why the decidual tissue is vigorously enriched with CD56brightCD16– NK cells since the unique composition of decidual lymphocyte population probably holds the key to proper feto-maternal immune tolerance. Chemokines and chemokine receptors are pivotal players in immune cells trafficking, recirculation, and homing; therefore, the specific preferential accumulation of CD56brightCD16– NK cells in the decidua may be attributable to the interaction between chemokines and their receptors. In human and mouse uteri, constant expression of mRNA and protein for MCP-1, MIP-1, and RANTES have been documented during the estrous cycle. The increased expression of these molecules occurs post implantation in both species (20, 21, 22, 23, 24). These observations suggest potential roles of MCP-1, MIP-1, and RANTES in leukocyte recruitment to the pregnant uterus. However, it is very surprising that normal recruitment of peripheral NK cells was observed in the decidua of mice in which CCR2 (sole signaling receptor for MCP-1), CCR5 (a receptor for both MIP-1 and RANTES), and MIP-1 were knocked out (25); thus, these molecules probably play a pivotal role in intradecidual organization and the retention of various cell types in the decidua, but are less important in the preferential migration of the decidual NK cells. Therefore, there may be other chemokines and their ligands involved in the preferential accumulation of CD56brightCD16– NK cells in the decidua.
Recently, a report has documented different expression patterns of some chemokine receptors on peripheral CD16+ and CD16– NK cells. Peripheral CD16+ NK cells uniformly expressed CXCR1 and CX3CR1 at high levels. CXCR4 was also present at high levels on these cells. CXCR2 and CXCR3 were present at lower levels. CCR1–7 and CCR9 were absent on this population, as well as CXCR5 and the orphan chemokine receptor Bonzo. As for peripheral CD16– NK cells, CCR5, CCR7, CXCR3, and CXCR4 were uniformly expressed at high levels. CX3CR1 was expressed at detectable, but very low levels. CCR1–4, 6, and 9 were not expressed, nor were CXCR1, CXCR2, and CXCR5 on this population (26). This information suggests that analysis of the chemokine receptors repertoire in the decidual CD56brightCD16– NK cells may give some useful clues to exploring the mechanisms of these cells’ specific accumulation. In this study, we first analyzed the transcription of 18 chemokine receptors in the decidual CD56brightCD16– NK cells and then found CXCR4 and CXCR3 mRNA highly expressed. Using flow cytometry, we confirmed the translation of CXCR4 in >83% of the purified decidual CD56brightCD16– NK cells.
CXCR4 is an extraordinary chemokine receptor. Apart from its important impact on HIV-1 infection, CXCR4 and its specific ligand CXCL12 play a key role in lymphocyte trafficking and recruitment at sites of inflammation and in hematopoiesis and development processes such as organogenesis, vascularization, and embryogenesis (27, 28). It also has been demonstrated that CXCL12 induces both the chemotaxis and calcium mobilization in NK cells (29, 30, 31). CXCR4 binds to the heterotrimeric G proteins, and is coupled to Go, Gq, and Gs but not to Gi, Gz, G12, or G13. Both Go and Gq are involved in CXCL12-induced chemotaxis and calcium mobilization in NK cells, since Abs to these G protein subfamilies inhibit both activities. In addition, Lck is required for SDF-1-induced lymphoid cell chemotaxis, and the Src homology 3 domain is important for the function of Lck in SDF-1-mediated chemotaxis (32).
Since CXCL12 is a chemoattractant for NK cells and in our experiments CXCR4 was expressed in >83% of the decidual CD56brightCD16– NK cells, we speculate that CXCL12 might exist at the maternal-fetal interface prominently, and that the interaction between CXCL12 at the maternal-fetal interface and CXCR4 on the CD56brightCD16– NK cells’ surface might contribute to the recruitment of these cells into decidua. Using in situ hybridization and immunohistochemistry, we found that the first-trimester trophoblast cells constitutively expressed CXCL12. In addition, the primarily cultured first-trimester trophoblast cells secreted continually soluble SDF-1 into the culture medium. The SDF-1 accumulated in the supernatant over time and its concentration was 384.6 ± 90.7 pg/ml after trophoblast cells had been cultured for 60 h.
Chemotaxis assay revealed that, within a proper range of concentration, rhSDF-1 indeed induced the chemotaxis of the decidual CD56brightCD16– NK cells in a dose-dependent manner. In addition, the supernatant of trophoblast cell cultures also showed chemotactic activity on the decidual CD56brightCD16– NK cells. When cells were preincubated with anti-CXCR4, they did not specifically migrate to the culture supernatants of trophoblast cells, which clearly indicates that the chemotactic property of the trophoblast supernatant is attributable to the presence of CXCL12.
The homing of leukocytes from the blood to the tissues is a multistep process, which involves both chemokines and adhesion molecules. Chemokine signaling directs cell movement toward tissues, whereas egress of circulating cells from vessels requires lymphocyte-endothelial cell interactions. Although Peripheral CD16+ NK cells also express CXCR4, as reported by Campbell et al. (26), the intensity of CXCR4 on the peripheral CD16+ and CD16– NK cell surface, and the ability of these two NK cell subsets to migrate in response to CXCL12 are very likely different. More important, in humans circulating CD56brightCD16– NK cells express functionally active L-selectin at extremely high levels, compared with CD56dimCD16+ NK cells (33), and L-selectin is necessary in the adhesion of CD56brightCD16– NK cells to the decidualizing uterus (12, 34). Thus, under the dual control of chemokines and adhesion molecules, CD56brightCD16– NK cells preferentially accumulate in the deciduas.
This report has demonstrated that trophoblast cells contribute to the recruitment of the decidual CD56brightCD16– NK cells. It should be noted that many adhesion molecules are expressed in uterine stroma and uterine stroma itself is capable of attracting circulating leukocytes, as distinct populations of immune cells migrate into the uterus during the course of the menstrual cycle (5, 35). Furthermore, the decidual leukocytes can be found in the decidua in the event of an ectopic pregnancy, when the placenta has implanted outside of the uterus (36). Therefore, the decidual leukocyte infiltration is also regulated by the decidual vasculature as is suggested by the highly specific combinatorial expression of adhesion molecules in pregnant murine uterine blood vessels (37, 38). These adhesion molecules are integral to a multistep homing process that also requires chemokines. In addition, secondary lymphoid organ chemokine, which is reported to preferentially attract CD56bright over CD56dim NK cells (39), is expressed on cells that compose the lumen of uterine vessels (40), which may also serve as a selective step allowing CD56brightCD16– NK cells to dominate in the decidua.
Our study also revealed that decidual CD56brightCD16– NK cells highly transcribe CXCR3 as well as CXCR4. Using ELISA we measured the ligands for CXCR3, i.e., CXCL9, CXCL10, and CXCL11 in supernatants of trophoblast cell cultures, which were derived from 15 irrelevant first-trimester placentas. Three CXCR3 ligands were all below the minimal detectable concentration when the trophoblast cells have been cultured for 12, 24, 36, or 48 h. After 60 h, CXCL9 was detected in one of the five samples at a low level, and CXCL10 was detected in two of the five samples also at low levels. Therefore, first-trimester trophoblast cells do not produce CXCR3 ligands at a remarkable level. A recent study shows that CXCL9 and CXCL10 are found in the surface epithelia, glandular epithelia, and stroma in human endometrium. The concentrations of CXCL9 and CXCL10 are higher in the secretory phase than in the proliferative phase and correlate with the number of endometrial NK cells (41). Furthermore, estradiol and progesterone can induce expression of CXCL10 and/or CXCL11 within human endometrium (42). We think it is very likely that CXCL12 and all of the ligands for CXCR3 are involved in CD56brightCD16– NK cell recruitment or localization. CXCL12 produced by trophoblast cells may be responsible for recruitment of CD56brightCD16– NK cells to sites adjacent to the trophoblast cells, whereas CXCR3 ligands in endometrium may be important for maintaining CD56brightCD16– NK cells within deciduas.
In summary, we have analyzed the transcription of all identified chemokine receptors in CD56brightCD16– NK cells isolated from the first-trimester human decidua. Among these chemokine receptors, CXCR4 mRNA was highly transcribed and most of the decidual CD56brightCD16– NK cells expressed CXCR4 protein. This study also confirms that the first-trimester human trophoblast cells produce CXCL12, which endows the trophoblast cells with the capacity to direct the migration of CD56brightCD16– NK cells into the decidua. This could contribute to the recruitment of the decidual leukocytes or may be used to modulate the immune milieu at the maternal-fetal interface so as to keep the pregnancy going smoothly.
Acknowledgments
We thank Zhengli Gu for his helpful comments on this manuscript.
Disclosures
The authors have no financial conflict of interest.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 Address correspondence and reprint requests to Dr. Dajin Li, Laboratory of Reproductive Immunology, Institute of Obstetrics and Gynecology, Fudan University, Shanghai 200011, China. E-mail address: djli@shmu.edu.cn
2 Abbreviations used in this paper: SDF-1, stromal cell-derived factor 1; TC, tricolor; rhSDF-1, recombinant human SDF-1.
Received for publication July 20, 2004. Accepted for publication March 25, 2005.
References
Croy, B. A., S. Chantakru, S. Esadeg, A. A. Ashkar, Q. Wei. 2002. Decidual natural killer cells: key regulators of placental development. J. Reprod. Immunol. 57: 151-168.
Boyson, J. E., B. Rybalov, L. A. Koopman, M. Exley, S. P. Balk, F. K. Racke, F. Schatz, R. Masch, S. B. Wilson, S. B. , J. L. Strominger. 2002. CD1d and invariant NKT cells at the human maternal-fetal interface. Proc. Natl. Acad. Sci. USA 99: 13741-13746.
Croy, B. A., A. A. Ashkar, K. Minhas, J. D. Greenwood. 2000. Can murine uterine natural killer cells give insights into the pathogenesis of preeclampsia?. J. Soc. Gynecol. Invest. 7: 12-20.
Guimond, M. J., B. Wang, B. A. Croy. 1998. Engraftment of bone marrow from severe combined immunodeficient (SCID) mice reverses the reproductive deficits in natural killer cell-deficient tg26 mice. J. Exp. Med. 187: 217-223.
Bulmer, J. N., L. Morrison, M. Longfellow, A. Ritson, D. Pace. 1991. Granulated lymphocytes in human endometrium: histochemical and immunohistochemical studies. Hum. Reprod. 6: 791-798.
Norwitz, E. R., D. J. Schust, S. J. Fisher. 2001. Implantation and the survival of early pregnancy. N. Engl. J. Med. 345: 1400-1408.
Ponte, M., C. Cantoni, R. Biassoni, A. Tradori-Cappai, G. Bentivoglio, C. Vitale, S. Bertone, A. Moretta, L. Moretta, M. C. Mingari. 1999. Inhibitory receptors sensing HLA-G1 molecules in pregnancy: decidua-associated natural killer cells express LIR-1 and CD94/NKG2A and acquire p49, an HLA-G1-specific receptor. Proc. Natl. Acad. Sci. USA 96: 5674-5679.
Verma, S., A. King, Y. W. Loke. 1997. Expression of killer cell inhibitory receptors on human uterine natural killer cells. Eur. J. Immunol. 27: 979-983.
Szekeres-Bartho, J.. 2002. Immunological relationship between the mother and the fetus. Int. Rev. Immunol. 21: 471-495.(Xia Wu, Li-Ping Jin, Min-)
More than 70% of decidual lymphocytes are NK cells characterized by CD56brightCD16– phenotype, but the mechanisms by which these NK cells are recruited in the decidua are still almost unrevealed. In this study, we first analyzed the transcription of 18 chemokine receptors in the first-trimester decidual CD56brightCD16– NK cells. Among these receptors, CXCR4 and CXCR3 were found highly transcribed, and the expression of CXCR4 was verified in most of the decidual CD56brightCD16– NK cells by flow cytometry. The first-trimester human trophoblasts were found expressing CXCL12/stromal cell-derived factor 1, the specific ligand of CXCR4, by way of in situ hybridization and immunohistochemistry. The primary cultured trophoblast cells were also found to secrete stromal cell-derived factor 1 spontaneously, and its concentration was 384.6 ± 90.7 pg/ml after the trophoblast cells had been cultured for 60 h. All of the ligands for CXCR3 were below the minimal detectable concentration when trophoblast cells were cultured for up to 48 h. Both recombinant human SDF-1 and supernatants of the cultured trophoblast cells exhibited chemotactic activity on decidual CD56brightCD16– NK cells. Our findings suggest that human first-trimester trophoblast cells produce CXCL12, which in turn chemoattracts decidual CD56brightCD16– NK cells. This activity could contribute to the recruitment mechanism of decidual lymphocytes, especially CD56brightCD16– NK cells, in decidua, and may be used at a local level to modulate the immune milieu at the materno-fetal interface.
Introduction
The mechanisms by which the human hemiallogeneic feto-placental unit is not rejected by the maternal immune system have been under intense attention, and it has now become clear that decidual lymphocytes have unique functions in local cytokine production, endovascular invasion, and placental development so as to keep pregnancy going smoothly (1, 2, 3, 4). More than 70% of decidual lymphocytes are NK cells characterized by the CD56brightCD16– phenotype (5, 6). In comparison to peripheral NK cells, decidual NK cells display different functional behaviors. These decidual NK cells have decreased killing activity against class I MHC-negative target cells and are also more susceptible to the inhibition mediated by MHC class I molecules (classical and nonclassical) expressed by extravillous cytotrophoblasts (7, 8, 9, 10). The enrichment in the decidua with this specific NK subset is outstanding compared with the 10% composition of NK cells in the peripheral blood, and the CD56brightCD16– population composes only 10–15% of the total peripheral NK population.
The origin of these decidual CD56brightCD16–NK cells is unknown, but because they closely resemble the minor agranular CD56bright NK cell population in the blood, one possibility is that this blood NK cell population might move into the uterus to proliferate, differentiate, enlarge, and acquire cytoplasmic granules in the hormone-rich mucosal microenvironment (11). Recently, uterine segment transplantation in mice has shown that uterine NK precursor cells do not self-renew in the uterus. Transplantable uterine NK precursor cells are found in spleen, lymph nodes, bone marrow, thymus, and liver (12), but the mechanisms by which the CD56brightCD16– NK cells are accumulated in the decidua are still almost unknown.
Chemokines constitute a family of closely related chemoattractant cytokines. Several groups of chemokines have been described and two major groups can be distinguished based on the presence (CXC) or absence (CC) of a single amino acid between two cysteine residues near the N terminus. Chemokine activity is mediated through cell surface receptors related to the 7-transmembrane domain receptor family. Chemokines were initially recognized to play a pivotal role in leukocyte communication and migration, both in physiological and pathological contexts. These molecules control immune cell trafficking and recirculation of the leukocyte population among the blood vessels, lymph, lymphoid organs, and tissues, an important process in host immune surveillance and in acute and chronic inflammatory responses (13, 14, 15). Now it has become evident that chemokines not only play a fundamental role in development, homeostasis, and immune system function, but also are important in the angiogenesis and angiostatic processes, CNS development, and HIV pathogenesis (16, 17, 18, 19).
Since chemokines and their receptors are key players in leukocyte communication and migration, it is very reasonable to speculate that chemokine receptors and their ligands are involved in the preferential accumulation of the CD56brightCD16– NK cells in the decidua: the chemokines at the materno-fetal interface act with the chemokine receptors expressed by CD56brightCD16– NK cells, then attract CD56brightCD16– NK cells to the decidua. In this study, we first analyzed the transcription of 18 chemokine receptors in first-trimester decidual CD56brightCD16– NK cells. Among these receptors, CXCR4 was found to be highly transcribed. The decidual CD56brightCD16– NK cells were then verified expressing CXCR4 by flow cytometric analysis. After that we detected the expression of CXCL12/stromal cell-derived factor 1 (SDF-1),2 the specific ligand of CXCR4, in the first-trimester human placentas. Finally, we analyzed the chemotactic activity of CXCL12 on decidual CD56brightCD16– NK cells. Our findings suggest that the first-trimester human trophoblast cells produce CXCL12, which in turn endows the trophoblast cells with the capacity to attract decidual CD56brightCD16– NK cells into the decidua.
Materials and Methods
Placental samples
The first-trimester human placentas and deciduas (gestational age, 5–10 wk) were obtained from clinically normal pregnancies, which were terminated for nonmedical reasons, at the Obstetrics and Gynecology Hospital of Fudan University. Each patient completed a signed, written consent form. The Fudan University Human Investigation Committee approved the form and the use of the placental tissue.
Isolation of decidual CD56brightCD16– NK cells
The pooled decidual tissues were trimmed into 1-mm pieces and enzymatically digested for 20 min, using vigorous shaking, with 1.5 mg type I DNase and 24 mg type IV collagenase (Sigma-Aldrich) present in 15 ml of RPMI 1640 medium. This procedure was repeated three times. After an additional 5-min incubation at room temperature without shaking, the supernatants were collected and loaded onto Ficoll density gradients to purify the lymphocyte population. Decidual CD56brightCD16– NK cells were further purified by incubation with microbeads of conjugated anti-human CD56 mAb (Miltenyi Biotec). Separation was performed with the AutoMACS instrument (Miltenyi Biotec).
Flow cytometric analysis
The evaluation of cell surface expression of CD3/CD56/CD16 was performed on immunopurified decidual NK cells by direct cell surface labeling after presaturation of cells at 4°C for 20 min with goat normal serum diluted 1/20 in PBS. The cells were then washed twice and incubated at 4°C for 20 min with FITC-conjugated mouse anti-human CD3, tricolor-conjugated mouse anti-human CD16, R-PE-conjugated mouse anti-human CD56 (Caltag Laboratories), or the relevant isotype control (Sino-America). The cell surface expression of CXCR4 was also assessed using PE-conjugated mouse anti-human CXCR4 (BD Pharmingen). All analysis was conducted on a BD Biosciences FACScan flow cytometer using CellQuest software. Three independent experiments were performed, each in triplicate. The purified decidual NK cells were derived from 15 decidual samples.
RT-PCR analysis
Total cellular RNA was extracted from freshly purified decidual CD56brightCD16– NK cells (cells derived from 15 decidual samples) using TRIzol reagent (Invitrogen Life Technologies). One microgram of total RNA was denatured, and reverse transcription was performed for 1 h at 42°C with 0.5 μg oligo(dT)18, 1.0 mM 4dNTP, 20 U RNasin RNase inhibitor (Promega), 200 U Moloney virus-reverse transcriptase (Superscript II; Invitrogen Life Technologies), and 5x reaction buffer in a total of 20 μl. Amplification was performed with 5 μl of cDNA, 0.2 mM dNTP, 1 mM MgCl2, 2.5 U AmpliTaq DNA polymerase (PerkinElmer), 0.8 mM specific sense and antisense primers, and 10x reaction buffer in a 50-μl reaction volume. The primers used for the detection of chemokine receptors and housekeeping gene GAPDH are indicated in Table 1. After a 5-min precycle at 95°C, the reaction was followed by 30 cycles of 1 min at 94°C, 1 min at 55°C, and 1 min at 72°C. When the final cycle was over, samples were kept at 72°C for 15 min to complete the synthesis. The PCR products were analyzed on 1.5% agarose gels and ethidium bromide-stained bands were photographed. Authenticity of all of the amplified PCR products was confirmed by direct sequencing. Three independent experiments were performed, each in triplicate.
In situ hybridization
Human CXCL12 in situ hybridization kits were obtained from Boshide. The oligonucleotide probes correspond to three sequences in CXCL12 mRNA (5'- AGGTCGTGGTCGTGCTGGTCCTCGTGCTGA-TCAAGCATCTCAAAATTCTCAACACTCCAA-CAGGAGTACCTGGAGAAAGCTTTAAACAAG-3'). Five first-trimester placentas collected for in situ hybridization were immediately fixed with 4% paraformaldehyde for 1 h at 4°C. They were then permeated with 12% sucrose and passed through an increasing concentration gradient of sucrose up to 18%. The specimens for frozen section were embedded with OCT compound, frozen at –80°C, and stored until further analysis. The frozen sections (8-μm thick) were prepared by a cryostat and treated with 10 μg/ml proteinase K. The slides were incubated for 2 h at 37°C in a prehybridization buffer. Hybridization was performed under coverslips overnight at 40°C. Following hybridization, the coverslips were removed and the slides were rinsed three times for 10 min with 2x SSC, then treated with 20 μg/ml RNase in 0.5 M sodium chloride and 10 mM Tris (pH 8.0) for 30 min at 37°C to digest the nonhybridization RNA probe. They were then washed three times for 20 min in 0.1x SSC at 42°C. Bound mRNA was detected with alkaline phosphatase-labeled anti-digoxigenin goat Ab and developed with 5-bromo-4-chloro-3-indolyl phosphate and nitroblue tetrazolium color development substrate.
Immunohistochemical staining
Five first-trimester placentas were collected for immunohistochemical staining. From the frozen specimens, serial sections of 8-μm thickness were prepared in a cryostat and fixed with cold acetone for 5 min. They were blocked sequentially with methanol containing 3% H2O2 and 20% human serum and then incubated with 10 μg/ml anti-CXCL12 mAb (R&D Systems) or 1:100 anti-human cytokeratin-7 mAb (DakoCytomation) or isotype-matched irrelevant IgG (Sino-America) overnight at 4°C. They were then sequentially treated with a biotinylated rabbit anti-mouse Ab and an HRP-labeled streptavidin. Development was performed by treating the sections with a liquid 3,3'-diaminobenzidine plus substrate kit (Sino-America). After counterstaining with hematoxylin, immunostaining of CXCL12 or cytokeratin-7 on the tissue sections was evaluated by light microscope.
Trophoblast cell culture
The first-trimester human placenta was separated carefully from the decidua under a stereomicroscope, minced into small fragments, and incubated in HBSS containing 0.25% trypsin (Invitrogen Life Technologies, 4.2 mM MgSO4, 25 mM HEPES, and 50 Kunitz/ml DNase type IV (Invitrogen Life Technologies) for 10 min at 37°C with gentle agitation. Then the suspension was taken and filtered through a stainless steel mesh of 100-μm pores. The trypsin digestion was stopped with 10% FCS (HyClone). The remaining placental tissues were subject to another three cycles of 10-min trypsinization. The four resultant cell suspensions were pooled, centrifuged at 300 x g for 10 min, and resuspended in 4 ml of DMEM (Invitrogen Life Technologies). This suspension was layered over a preformed Percoll gradient (Pharmacia) made up in HBSS. The gradient was made from 70% to 5% Percoll (v/v) in 5% steps of 2 ml each by dilutions of 90% Percoll with HBSS. The gradient was centrifuged at 1000 x g for 20 min. The middle layer (density, 1.048–1.062 g/ml) was removed and washed with DMEM. Cells were diluted with DMEM supplemented with 2 mM glutamine, 10% heat-inactivated FCS, 25 mM HEPES, 100 IU/ml penicillin, and 100 μg/ml streptomycin.
The isolated human trophoblast cells were plated in plastic petri dishes and incubated at 37°C for 40 min to allow the contaminating macrophages to adhere to the plastic. The nonadherent trophoblast cells were transferred to fresh plates as well as eight-chambered LabTek slides at a concentration of 8 x 105 cells/ml. Cells were cultured in DMEM supplemented with 2 mM glutamine, 10% heat-inactivated FCS, 25 mM HEPES, 100 IU/ml penicillin, and 100 μg/ml streptomycin at 37°C in 95% air and 5% CO2.
Immunocytochemistry
After 24 h of culture, trophoblast cells were fixed in 4% paraformaldehyde for 20 min at room temperature, washed in PBS, and permeabilized for 4 min in 0.3% Triton X-100-PBS. The cells were incubated with 20% human serum in PBS for 30 min to reduce nonspecific binding. Primary Abs diluted in PBS containing 1% BSA were added. Anti-human cytokeratin-7 and anti-human vimentin mAb (DakoCytomation) were used as markers for identification of trophoblast lineage and nontrophoblast lineage, respectively. The dilutions of anti-cytokeratin-7 and anti-vimentin were both 1/100. Isotype-matched irrelevant IgG (Sino-America) was used as a control. After incubation with primary Ab overnight at 4°C, the cells were washed in PBS-0.1% Tween 20, and then incubated with HRP-labeled secondary Ab (Sino-America) for 2 h at room temperature. The slides were stained with 3,3'-diaminobenzidine and counterstained with hematoxylin. The experiments were repeated five times.
ELISA
The supernatants of trophoblast cell cultures were harvested after 12, 24, 36, 48, and 60 h of culture. Each supernatant was centrifuged at 2000 x g and stored at –80°C. ELISAs were performed with human SDF-1, CXCL9, CXCL10, and CXCL11 kits (R&D Systems.) according to the instructions of the manufacturer. The minimal detectable concentrations for SDF-1, CXCL9, CXCL10, and CXCL11 were 18, 5, 5, 15 pg/ml, respectively. Five individual placental samples were tested for each chemokine. At least three wells’ supernatants were collected at each time point for each placental sample.
In vitro cell migration assay
The purified decidual CD56brightCD16– NK cells were resuspended in DMEM at a concentration of 3 x 106 cells/ml and incubated for 4 h with 100 ng/ml anti-human CXCR4 mAb (clone 12G5; BD Pharmingen) or 100 ng/ml isotype-matched irrelevant IgG (Sino-America). After that the decidual CD56brightCD16[minus ]NK cells of 3 x 105 in 100 μl of DMEM were loaded into each Transwell filter (5-μm pore filter Transwell, 24-well cell clusters; Corning). The filters were then plated in each well containing 600 μl of supernatant medium of trophoblast cell cultures (choose the supernatant medium in which the concentration of SDF-1 is >400 pg/ml) or DMEM supplemented with differing concentrations of recombinant human SDF-1 (rhSDF-1; R&D Systems). Three wells were used per condition per experiment. After a 3-h incubation at 37°C in 5% CO2, the upper chambers were removed and cells in the bottom chamber were collected, counted, and analyzed by flow cytometry. Results are expressed as the percentage of migrating cells (number of migrated cells/total number of input cells x 100). The experiments were repeated four times and the purified decidual NK cells were derived from 20 decidual samples.
Statistical analysis
All values are shown as mean ± SD. Data from the in vitro cell migration assay were performed using one-way ANOVA, with application of the Dunnett test. Differences were accepted as significant at p < 0.05.
Results
The decidual NK cells, being CD56brightCD16–, differ phenotypically from the majority of classical circulating NK cells. Using immunofluorescent labeling for the Ags CD56, CD16, and CD3 and subsequent flow cytometric analysis, it was confirmed that the purity of the decidual NK cells isolated in this study was >95% (Fig. 1).
Transcription of chemokine receptors in CD56brightCD16– NK cells
Using PCR primers specific for human chemokine receptors, we have analyzed the transcription of chemokine receptors in freshly isolated decidual CD56brightCD16– NK cells. Three independent experiments (each in triplicate) were done and the results were similar. The decidual CD56brightCD16– NK cells transcribed CCR1, CCR3, CCR5–CCR8, CXCR1–CXCR4, CXCR6, and CX3CR1 and the ratios of the band intensities of chemokine receptors to that of GAPDH were 0.81 ± 0.14, 0.78 ± 0.11, 0.59 ± 0.09, 0.71 ± 0.12, 0.89 ± 0.14, 0.53 ± 0.12, 0.91 ± 0.17, 0.78 ± 0.13, 1.09 ± 0.21, 1.47 ± 0.18, 0.92 ± 0.16 and 0.71 ± 0.16, respectively. Among these chemokine receptors, CXCR4 and CXCR3 were highly transcribed in CD56brightCD16– NK cells (the mean band intensities of these two chemokine receptors were higher than that of GAPDH). The mRNA of CCR2, CCR4, CCR9, CCR10, CXCR5, and XCR1 was never detected in CD56brightCD16– NK cells by RT-PCR (Fig. 2).
Having known that the first-trimester human trophoblast cells produce CXCL12 spontaneously, we assessed the chemotaxis of CXCL12 to the decidual CD56brightCD16– NK cells. The results showed that within a suitable range of concentration rhSDF-1 indeed induced the chemotaxis of decidual CD56brightCD16– NK cells, which had been preincubated with isotype-matched irrelevant IgG in a dose-dependent manner. A total of 21.60 ± 4.19% and 16.60 ± 2.78% of the input cells migrated to the bottom chamber containing 10 ng/ml and 100 ng/ml rhSDF-1, respectively (p < 0.001 compared with cells migrating toward DMEM only). In addition, the supernatant of trophoblast cells also showed chemotactic activity on the decidual CD56brightCD16– NK cells (p < 0.001 compared with cells migrating toward DMEM only). When the cells were preincubated with anti-CXCR4, they did not specifically migrate to the bottom chamber containing rhSDF-1 or the supernatant of trophoblast cells, which clearly demonstrates that the interaction between the CXCL12 secreted by human trophoblast cells and the CXCR4 expressed by the CD56brightCD16– NK cells directs the migration of the CD56brightCD16– NK cells (Fig. 7).
Discussion
It is very meaningful to explore how and why the decidual tissue is vigorously enriched with CD56brightCD16– NK cells since the unique composition of decidual lymphocyte population probably holds the key to proper feto-maternal immune tolerance. Chemokines and chemokine receptors are pivotal players in immune cells trafficking, recirculation, and homing; therefore, the specific preferential accumulation of CD56brightCD16– NK cells in the decidua may be attributable to the interaction between chemokines and their receptors. In human and mouse uteri, constant expression of mRNA and protein for MCP-1, MIP-1, and RANTES have been documented during the estrous cycle. The increased expression of these molecules occurs post implantation in both species (20, 21, 22, 23, 24). These observations suggest potential roles of MCP-1, MIP-1, and RANTES in leukocyte recruitment to the pregnant uterus. However, it is very surprising that normal recruitment of peripheral NK cells was observed in the decidua of mice in which CCR2 (sole signaling receptor for MCP-1), CCR5 (a receptor for both MIP-1 and RANTES), and MIP-1 were knocked out (25); thus, these molecules probably play a pivotal role in intradecidual organization and the retention of various cell types in the decidua, but are less important in the preferential migration of the decidual NK cells. Therefore, there may be other chemokines and their ligands involved in the preferential accumulation of CD56brightCD16– NK cells in the decidua.
Recently, a report has documented different expression patterns of some chemokine receptors on peripheral CD16+ and CD16– NK cells. Peripheral CD16+ NK cells uniformly expressed CXCR1 and CX3CR1 at high levels. CXCR4 was also present at high levels on these cells. CXCR2 and CXCR3 were present at lower levels. CCR1–7 and CCR9 were absent on this population, as well as CXCR5 and the orphan chemokine receptor Bonzo. As for peripheral CD16– NK cells, CCR5, CCR7, CXCR3, and CXCR4 were uniformly expressed at high levels. CX3CR1 was expressed at detectable, but very low levels. CCR1–4, 6, and 9 were not expressed, nor were CXCR1, CXCR2, and CXCR5 on this population (26). This information suggests that analysis of the chemokine receptors repertoire in the decidual CD56brightCD16– NK cells may give some useful clues to exploring the mechanisms of these cells’ specific accumulation. In this study, we first analyzed the transcription of 18 chemokine receptors in the decidual CD56brightCD16– NK cells and then found CXCR4 and CXCR3 mRNA highly expressed. Using flow cytometry, we confirmed the translation of CXCR4 in >83% of the purified decidual CD56brightCD16– NK cells.
CXCR4 is an extraordinary chemokine receptor. Apart from its important impact on HIV-1 infection, CXCR4 and its specific ligand CXCL12 play a key role in lymphocyte trafficking and recruitment at sites of inflammation and in hematopoiesis and development processes such as organogenesis, vascularization, and embryogenesis (27, 28). It also has been demonstrated that CXCL12 induces both the chemotaxis and calcium mobilization in NK cells (29, 30, 31). CXCR4 binds to the heterotrimeric G proteins, and is coupled to Go, Gq, and Gs but not to Gi, Gz, G12, or G13. Both Go and Gq are involved in CXCL12-induced chemotaxis and calcium mobilization in NK cells, since Abs to these G protein subfamilies inhibit both activities. In addition, Lck is required for SDF-1-induced lymphoid cell chemotaxis, and the Src homology 3 domain is important for the function of Lck in SDF-1-mediated chemotaxis (32).
Since CXCL12 is a chemoattractant for NK cells and in our experiments CXCR4 was expressed in >83% of the decidual CD56brightCD16– NK cells, we speculate that CXCL12 might exist at the maternal-fetal interface prominently, and that the interaction between CXCL12 at the maternal-fetal interface and CXCR4 on the CD56brightCD16– NK cells’ surface might contribute to the recruitment of these cells into decidua. Using in situ hybridization and immunohistochemistry, we found that the first-trimester trophoblast cells constitutively expressed CXCL12. In addition, the primarily cultured first-trimester trophoblast cells secreted continually soluble SDF-1 into the culture medium. The SDF-1 accumulated in the supernatant over time and its concentration was 384.6 ± 90.7 pg/ml after trophoblast cells had been cultured for 60 h.
Chemotaxis assay revealed that, within a proper range of concentration, rhSDF-1 indeed induced the chemotaxis of the decidual CD56brightCD16– NK cells in a dose-dependent manner. In addition, the supernatant of trophoblast cell cultures also showed chemotactic activity on the decidual CD56brightCD16– NK cells. When cells were preincubated with anti-CXCR4, they did not specifically migrate to the culture supernatants of trophoblast cells, which clearly indicates that the chemotactic property of the trophoblast supernatant is attributable to the presence of CXCL12.
The homing of leukocytes from the blood to the tissues is a multistep process, which involves both chemokines and adhesion molecules. Chemokine signaling directs cell movement toward tissues, whereas egress of circulating cells from vessels requires lymphocyte-endothelial cell interactions. Although Peripheral CD16+ NK cells also express CXCR4, as reported by Campbell et al. (26), the intensity of CXCR4 on the peripheral CD16+ and CD16– NK cell surface, and the ability of these two NK cell subsets to migrate in response to CXCL12 are very likely different. More important, in humans circulating CD56brightCD16– NK cells express functionally active L-selectin at extremely high levels, compared with CD56dimCD16+ NK cells (33), and L-selectin is necessary in the adhesion of CD56brightCD16– NK cells to the decidualizing uterus (12, 34). Thus, under the dual control of chemokines and adhesion molecules, CD56brightCD16– NK cells preferentially accumulate in the deciduas.
This report has demonstrated that trophoblast cells contribute to the recruitment of the decidual CD56brightCD16– NK cells. It should be noted that many adhesion molecules are expressed in uterine stroma and uterine stroma itself is capable of attracting circulating leukocytes, as distinct populations of immune cells migrate into the uterus during the course of the menstrual cycle (5, 35). Furthermore, the decidual leukocytes can be found in the decidua in the event of an ectopic pregnancy, when the placenta has implanted outside of the uterus (36). Therefore, the decidual leukocyte infiltration is also regulated by the decidual vasculature as is suggested by the highly specific combinatorial expression of adhesion molecules in pregnant murine uterine blood vessels (37, 38). These adhesion molecules are integral to a multistep homing process that also requires chemokines. In addition, secondary lymphoid organ chemokine, which is reported to preferentially attract CD56bright over CD56dim NK cells (39), is expressed on cells that compose the lumen of uterine vessels (40), which may also serve as a selective step allowing CD56brightCD16– NK cells to dominate in the decidua.
Our study also revealed that decidual CD56brightCD16– NK cells highly transcribe CXCR3 as well as CXCR4. Using ELISA we measured the ligands for CXCR3, i.e., CXCL9, CXCL10, and CXCL11 in supernatants of trophoblast cell cultures, which were derived from 15 irrelevant first-trimester placentas. Three CXCR3 ligands were all below the minimal detectable concentration when the trophoblast cells have been cultured for 12, 24, 36, or 48 h. After 60 h, CXCL9 was detected in one of the five samples at a low level, and CXCL10 was detected in two of the five samples also at low levels. Therefore, first-trimester trophoblast cells do not produce CXCR3 ligands at a remarkable level. A recent study shows that CXCL9 and CXCL10 are found in the surface epithelia, glandular epithelia, and stroma in human endometrium. The concentrations of CXCL9 and CXCL10 are higher in the secretory phase than in the proliferative phase and correlate with the number of endometrial NK cells (41). Furthermore, estradiol and progesterone can induce expression of CXCL10 and/or CXCL11 within human endometrium (42). We think it is very likely that CXCL12 and all of the ligands for CXCR3 are involved in CD56brightCD16– NK cell recruitment or localization. CXCL12 produced by trophoblast cells may be responsible for recruitment of CD56brightCD16– NK cells to sites adjacent to the trophoblast cells, whereas CXCR3 ligands in endometrium may be important for maintaining CD56brightCD16– NK cells within deciduas.
In summary, we have analyzed the transcription of all identified chemokine receptors in CD56brightCD16– NK cells isolated from the first-trimester human decidua. Among these chemokine receptors, CXCR4 mRNA was highly transcribed and most of the decidual CD56brightCD16– NK cells expressed CXCR4 protein. This study also confirms that the first-trimester human trophoblast cells produce CXCL12, which endows the trophoblast cells with the capacity to direct the migration of CD56brightCD16– NK cells into the decidua. This could contribute to the recruitment of the decidual leukocytes or may be used to modulate the immune milieu at the maternal-fetal interface so as to keep the pregnancy going smoothly.
Acknowledgments
We thank Zhengli Gu for his helpful comments on this manuscript.
Disclosures
The authors have no financial conflict of interest.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 Address correspondence and reprint requests to Dr. Dajin Li, Laboratory of Reproductive Immunology, Institute of Obstetrics and Gynecology, Fudan University, Shanghai 200011, China. E-mail address: djli@shmu.edu.cn
2 Abbreviations used in this paper: SDF-1, stromal cell-derived factor 1; TC, tricolor; rhSDF-1, recombinant human SDF-1.
Received for publication July 20, 2004. Accepted for publication March 25, 2005.
References
Croy, B. A., S. Chantakru, S. Esadeg, A. A. Ashkar, Q. Wei. 2002. Decidual natural killer cells: key regulators of placental development. J. Reprod. Immunol. 57: 151-168.
Boyson, J. E., B. Rybalov, L. A. Koopman, M. Exley, S. P. Balk, F. K. Racke, F. Schatz, R. Masch, S. B. Wilson, S. B. , J. L. Strominger. 2002. CD1d and invariant NKT cells at the human maternal-fetal interface. Proc. Natl. Acad. Sci. USA 99: 13741-13746.
Croy, B. A., A. A. Ashkar, K. Minhas, J. D. Greenwood. 2000. Can murine uterine natural killer cells give insights into the pathogenesis of preeclampsia?. J. Soc. Gynecol. Invest. 7: 12-20.
Guimond, M. J., B. Wang, B. A. Croy. 1998. Engraftment of bone marrow from severe combined immunodeficient (SCID) mice reverses the reproductive deficits in natural killer cell-deficient tg26 mice. J. Exp. Med. 187: 217-223.
Bulmer, J. N., L. Morrison, M. Longfellow, A. Ritson, D. Pace. 1991. Granulated lymphocytes in human endometrium: histochemical and immunohistochemical studies. Hum. Reprod. 6: 791-798.
Norwitz, E. R., D. J. Schust, S. J. Fisher. 2001. Implantation and the survival of early pregnancy. N. Engl. J. Med. 345: 1400-1408.
Ponte, M., C. Cantoni, R. Biassoni, A. Tradori-Cappai, G. Bentivoglio, C. Vitale, S. Bertone, A. Moretta, L. Moretta, M. C. Mingari. 1999. Inhibitory receptors sensing HLA-G1 molecules in pregnancy: decidua-associated natural killer cells express LIR-1 and CD94/NKG2A and acquire p49, an HLA-G1-specific receptor. Proc. Natl. Acad. Sci. USA 96: 5674-5679.
Verma, S., A. King, Y. W. Loke. 1997. Expression of killer cell inhibitory receptors on human uterine natural killer cells. Eur. J. Immunol. 27: 979-983.
Szekeres-Bartho, J.. 2002. Immunological relationship between the mother and the fetus. Int. Rev. Immunol. 21: 471-495.(Xia Wu, Li-Ping Jin, Min-)