Objective
To determine whether proangiogenic immature myeloid cells are present in human placentas.
Study Design
Biopsies were obtained from 61 placentas of term pregnancies. Percentage of CD45 + CD33 + LIN2 − HLADR − immature myeloid cells of total CD45 + hematopoietic cells was determined by flow cytometry. Location of immature myeloid cells in the placenta was identified using confocal microscopy. The proangiogenic potential of immature myeloid cells was analyzed by endothelial tube formation.
Results
Immature myeloid cells comprise ∼25% of human placental CD45 + hematopoietic cells and infiltrate placentas in proximity of blood vessels. The percentage of immature myeloid cells correlated positively with placental weight (r 2 = 0.108, P = .01) and birthweight (r 2 = 0.087, P = .02). Endothelial tube formation was increased in the presence of immature myeloid cells as compared with the presence of CD45 + LIN2 + control cells.
Conclusion
Human placentas are populated by immature myeloid cells in the proximity of blood vessels. Consistent with their involvement in angiogenesis, immature myeloid cells accelerated endothelial tube formation. The presence of immature myeloid cells in pathologic pregnancies warrants further studies.
Accumulating evidence suggest that the proliferative, invasive, and immune tolerance mechanisms that malignant tumors use to establish a nutrient supply and evade or edit the host immune response are similar to those used by the developing placenta during normal pregnancy. In addition to the shared capacity for invading through normal tissues, both cancer cells and cells of the developing placenta create a microenvironment supportive of both immunologic privilege and angiogenesis.
Immune aberrations have been demonstrated in tumorigenesis, and myeloid-derived suppressor cells (MDSCs), also known as immature myloid cells (IMCs) have been shown to play a pivotal role in mediating immune suppression in animal models of human tumors. IMCs represent a heterogeneous population of cells that accumulate in tumor-bearing hosts and as a result of tumor-induced alterations in myelopoiesis they have been found in peripheral blood, lymphoid organs, and the tumor tissue itself. An increased population of IMCs was identified in patients with non-small-cell lung cancers. Accordingly, IMCs detected in the peripheral blood of patients bearing several tumor types express the common myeloid marker CD33 but lack markers of mature myeloid cells such as the MHC class II molecule HLADR.
In our previous study, comparing immune cell populations present in mouse tumors with those present in mouse placentas, we found a similar population of CD11b + Gr1 + IMCs in both models that express similar pronagiogenic factors. Furthermore, when isolated from the mouse placenta or tumor, we have shown that these cells can induce angiogenesis in matrigel plugs in vivo. Little is known about the presence or role of placental IMCs in either normal or aberrant human pregnancies.
The goal of our study was to investigate whether IMCs observed in mouse placentas are also present in human pregnancies. We also studied the localization of these cells in human placentas and the functional role of these immune cells as proangiogenic cells in the placenta.
Materials and Methods
Placental biopsies (n = 61) were collected from third trimester deliveries of normal pregnancies from June to Dec. 2010, at the delivery room of our medical center. A 2-cm square biopsy was taken from the maternal side to the fetal side of the placenta. The study was approved by the local institutional review board.
Flow cytometry
Placental biopsies were digested with an enzyme mixture including: 25 μg/mL DNase (Sigma-Aldrich, St. Louis, MO) and 1 μg/mL collagenase (Sigma-Aldrich) dissolved in phosphate-buffered saline (PBS), at 37 ° C for 60 minutes. Digested specimens were filtered through a 40 μm cell strainer and resuspended in an erythrocyte lysis buffer. Single cells were then suspended in fluorescence-activated cell sorting (FACS) buffer (PBS, 5 nmol/L EDTA, 1% bovine serum albumin [BSA]). Immunostaining was performed by incubating the cells with monoclonal antibodies for 30 minutes on ice. Staining reagents included fluorochrome (allophycocyarin [APC], phycoerythrin [PE], or fluorescein-iso-thiocyanate [FITC]) labeled anti-CD45, CD33, LIN2, HLA-DR [all purchased from BD Biosciences, San Diego, CA]). The percentage of IMCs per total CD45 + hematopoietic cells in the placental tissue was analyzed. Flow cytometry was performed with a FACS Calibur (Becton Dickinson, Mountain View, CA). Flow cytometry based cell sorting was performed using FACS Aria (Becton Dickinson).
Endothelial tube formation assay
To determine proangiogenic potential of the IMCs in vitro, equal concentrations of IMCs and control cells from the same placenta were mixed with primary human umbilical vein endothelial cells (HUVEC) culture (ATCC, VA). Specifically, 24-well plates were precoated with growth factor reduced matrigel (BD Biosciences, Bedford, MA). A total of 40,000 CD45 + CD33 + LIN2 − HLADR − IMCs or an equal number of CD45 + LIN2 + placental cells that served as controls, were mixed with 30,000 HUVEC cells in EGM-2 medium (Lonza, Basel, Switzerland) and were plated in each well. After 6 hours, at 37°C, endothelial tubes that formed were quantified under an inverted microscope. Specifically, the number of endothelial cell tubes per low power field was counted for each sample. The mean quantity of endothelial tubes formed in the presence of CD45 + CD33 + LIN2 − HLADR − IMCs was compared with that in the presence of control CD45 + LIN2 + bone marrow-derived cells.
Confocal microscopy
Frozen sections (12 μM) of placental tissue were fixed with acetone (at −20°C), washed in PBS and blocked with 20% horse serum in PBS. Immunostaining was performed with the following antibodies: for IMCs—mouse anti-human CD33 (BD Pharmingen, Franklin Lakes, NJ) and Cy3-labeled anti-mouse antibody (Jackson ImmunoResearch, West Grove, PA) for secondary staining; for endothelial cells—goat anti-human CD31 (Santa Cruz Biotechnology, Santa Cruz, CA) and Cy2-labeled anti-goat antibody (Jackson ImmunoResearch) for secondary staining. Mounting was performed on glass slide with Vectashield HardSet Mounting Media with 4′,6-diamidino-2-phenylindole (DAPI) (Vector Laboratories, Burlingame, CA). Optical sections were scanned using a Zeiss LSM 510 Meta Duoscan confocal system (Carl Zeiss, Jena, Germany).
Statistical analysis
Statistical analysis of continues parameters was performed using 2-tailed Student t tests. Correlations were analyzed using 2-tailed R Pearson test. P < .05 was considered statistically significant.
Results
We first sought to determine whether IMCs are present in human placentas. Placental single-cell suspensions were stained and analyzed by flow cytometry. CD45 + CD33 + LIN2 − HLADR − IMCs comprised 25.2 ± 7.7 (range, 10–40%) of total placental CD45 + bone marrow-derived hematopoietic cells ( Figure 1 ). The demographic characteristics of the study population are demonstrated in Table 1 .
| Age, y | BMI | Gestational age, wk | Birthweight, g | Placenta weight, g | |
|---|---|---|---|---|---|
| Mean | 28 | 27.4 | 39 | 3244 | 544 |
| SD | 5.94 | 4.35 | 1.28 | 325 | 83 |
| Minimum | 18 | 19.7 | 35.3 | 2370 | 350 |
| Maximum | 42 | 37.5 | 42.4 | 4285 | 710 |
When analyzing their abundance in all study population (n = 61) the percentage of IMCs correlated positively with placental weight ( r 2 = 0.108, P = .01) and birthweight ( r 2 = 0.087, P = .02) ( Figure 2 ), we found no significant correlations with parity ( Table 2 ), nor gestational age, maternal age, or body mass index (BMI) (data not shown).
| n | Mean | SD | Minimum | Maximum | |
|---|---|---|---|---|---|
| Primips | 18 | 25.4 | 10.4 | 9.8 | 51.1 |
| Multips | 43 | 25.2 | 6.3 | 14.9 | 41.4 |
To determine the location of IMCs within the placenta, we examined frozen placental sections by confocal microscopy. IMCs expressing CD33 were found to be localized adjacent to endothelial cells (identified by the expression of CD31) ( Figure 3 ) in the perivascular space.
