Stem Cells
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

First published online November 1, 2007
Stem Cells Vol. 26 No. 2 February 2008, pp. 550 -561
doi:10.1634/stemcells.2007-0406; www.StemCells.com
© 2008 AlphaMed Press

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
2007-0406v1
2007-0406v2
26/2/550    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Reprints/Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chen, C.-P.
Right arrow Articles by Yang, Y.-C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chen, C.-P.
Right arrow Articles by Yang, Y.-C.

TRANSLATIONAL AND CLINICAL RESEARCH

Trafficking of Multipotent Mesenchymal Stromal Cells from Maternal Circulation Through the Placenta Involves Vascular Endothelial Growth Factor Receptor-1 and Integrins

Chie-Pein Chena,b,c, Ming-Yi Leeb, Jian-Pei Huanga, John D. Aplind, Yi-Hsin Wub, Cing-Siang Hub, Pei-Chun Chenb, Hung Lie, Shiaw-Min Hwangf, Shu-Hsiang Liub, Yuh-Cheng Yangb

aDivision of High Risk Pregnancy and
bDepartment of Medical Research, Mackay Memorial Hospital, Taipei, Taiwan;
cMackay Medicine, Nursing and Management College, Taipei, Taiwan;
dDivision of Human Development, Medical School, University of Manchester, Manchester, United Kingdom;
eInstitute of Molecular Biology, Academia Sinica, Taipei, Taiwan;
fBioresource Collection and Research Center, Food Industry Research and Development Institute, Hsinchu, Taiwan

Key Words. Integrins • Multipotent mesenchymal stromal cell • Microchimerism • Placenta Vascular endothelial growth factor receptor

Correspondence: Correspondence: Chie-Pein Chen, M.D., Ph.D., Division of High Risk Pregnancy, Mackay Memorial Hospital, 92 Sec. 2 Chung San North Road, Taipei 104, Taiwan Telephone: 011-886-2-2543-3535; Fax: 011-886-2-2754-3769; e-mail: cpchen{at}ms2.mmh.org.tw

Received on May 24, 2007; accepted for publication on October 23, 2007.

First published online in STEM CELLS EXPRESS  November 1, 2007.

   ABSTRACT
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosure of Potential...
 Acknowledgments
 References
 
Maternal cells can become engrafted in various fetal organs during pregnancy. The nature of the cells and the mechanisms of maternofetal cell trafficking are not clear. We demonstrate that human lineage-negative, CD34-negative (Lin–CD34–) multipotent mesenchymal stromal cells express {alpha}2, {alpha}4, {alpha}5, and β1 integrins, which mediate their adhesion to endothelium, and vascular endothelial growth factor receptor-1 (VEGFR-1), which mediates their response to vascular endothelial growth factor A (VEGF-A). A maternal-fetal VEGF-A concentration gradient exists across the placental barrier, and cord blood plasma induces transendothelial and trans-Matrigel migration of stem cells in vitro. Migration is inhibited by a VEGF-A-neutralizing antibody or antibodies against VEGFR-1 or integrin {alpha}2, {alpha}4, {alpha}5, or β1. When Lin–CD34– multipotent mesenchymal stromal cells are transferred to rat maternal venous blood, they traffic through the placenta, engraft in various fetal organs, and persist in offspring for at least 12 weeks. Cell proliferation ability is retained in the xenogeneic placenta. Maternofetal trafficking is significantly reduced by blocking antibodies against integrins {alpha}2, {alpha}4, {alpha}5, and β1 or VEGFR-1. These results suggest that maternal microchimerism arises by the trafficking of multipotent mesenchymal stromal cells via VEGF-A- and integrin-dependent pathways across the hemochorial placenta to fetal tissues.

Disclosure of potential conflicts of interest is found at the end of this article.


   INTRODUCTION
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosure of Potential...
 Acknowledgments
 References
 
Placenta is not an absolute barrier: cells can pass from mother to fetus and vice versa [1–4]. Maternal cells have been demonstrated in fetal circulation [5] and in liver, spleen, thymus, thyroid, and skin of neonatal organs [6]. Long-term engraftment of maternal cells in the postnatal bone marrow and spleen has been detected in a rodent model [7]. Humans, rats, and mice have hemochorial placentas [8], but the nature of the specific cell types and the regulatory mechanisms that orchestrate the trafficking of maternal cells through the placenta are not known. Maternal blood contains progenitor and stem cells of hematopoietic or mesenchymal origin [9, 10], representing a potential source of cells that might be capable of transfer and engraftment in the fetus. Such trafficking, excluding maternal-fetal interface disruption such as abortion, placental abnormality, or preeclampsia [11, 12], would require a multistep cascade including initial cell adhesion to trophoblast or exposed placental extracellular matrix (ECM), transmigration, passage through the basement membrane and endothelium of the placenta, and finally migration and targeting to the fetal organ.

It has been reported that placenta expresses vascular endothelial growth factor A (VEGF-A), and a higher level of VEGF-A has been observed in fetal circulation than maternal circulation [13–16]. VEGF-A signals through two receptor tyrosine kinases, vascular endothelial growth factor receptor-1 (VEGFR-1)/flt-1 and VEGFR-2/flk-1 [17]. VEGF-A, acting through VEGFR-1, is known to stimulate the motility of a range of cell types, including progenitor and stem cells [18–20].

Integrins are heterodimeric cell surface receptors. Different {alpha} and β subunits are expressed in limited combinations to produce different ligand specificities [21]. Integrins mediate the homing of transplanted hematopoietic stem cells to the bone marrow [22], as well as the recruitment of inflammatory cells to sites of inflammation [23]. Thus, they may be involved in maternal cell mobilization across the placenta. We hypothesized that maternal stem cells can traffic through the placenta, resulting in long-term postnatal engraftment; that integrins are required for transplacental migration; and that VEGF-A in fetal circulation, acting through VEGFR-1 on maternal stem cells, may play a regulatory role in this process. We established a pregnant rat model with the maternal injection of human multipotent mesenchymal stromal cells (hMSCs) expressing integrin {alpha}2, {alpha}4, {alpha}5, and β1 subunits and VEGFR-1. The results suggest that maternal stem cells traffic through the placenta via the interaction of VEGF-A and VEGFR-1, with a requirement for cell surface integrins. Our findings may help to explain the mechanisms of maternal microchimerism in the fetus.


   MATERIALS AND METHODS
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosure of Potential...
 Acknowledgments
 References
 
hMSC Preparation and Labeling
A human bone marrow-derived immortalized multipotent mesenchymal stromal cell (MSC) line (hBM1) was a kind gift from Dr. Toguchida (Institute for Frontier Medical Sciences, Kyoto University, Kyoto, Japan) and was cultured in Dulbecco's modified Eagle's medium (DMEM) (Invitrogen, Grand Island, NY, http://www.invitrogen.com) with 10% fetal bovine serum (FBS) (HyClone, Logan, UT, http://www.hyclone.com) [24].

Frozen human bone marrow mononuclear cells (hBM2) from one 19-year-old male (Cambrex, Walkersville, MD, http://www.cambrex.com) and fresh human adult bone marrow mononuclear cells (hBM3) from one 37-year-old female who had undergone orthopedic surgery were collected and expanded in {alpha}-modified minimum essential medium (HyClone) containing 20% FBS (HyClone) and 4 ng/ml basic fibroblast growth factor (R&D Systems Inc., Minneapolis, http://www.rndsystems.com) as previously described [25, 26]. Tissue and blood were obtained after informed consent of the patients, and all experiments were approved by the Institutional Review Board of Mackay Memorial Hospital. To track cell migration and protein expression in vivo, cells were incubated with 1 µg/ml bis-benzimide (Hoechst 33342; Sigma-Aldrich, St. Louis, http://www.sigmaaldrich.com) for 24 hours at 37°C and were trypsinized before transplantation.

Flow Cytometry Analysis
The cell surface phenotype of hMSCs was characterized by a panel of phycoerythrin- or fluorescein isothiocyanate (FITC)-conjugated antibodies purchased from AbD Serotec (Oxford, U.K., http://www.ab-direct.com), Biolegend (San Diego, http://www.biolegend.com), or BD Pharmingen (Franklin Lakes, NJ, http://www.bdbiosciences.com/index_us.shtml) using standard fluorescence-activated cell sorting analysis.

Adhesion Assay
The 24-well plates were precoated with 1 µg/ml fibronectin (Upstate, Lake Placid, NY, http://www.upstate.com), laminin (Chemicon, Temecula, CA, http://www.chemicon.com), or 2.5% bovine serum albumin (Sigma-Aldrich) as controls at 4°C for overnight. hMSCs were preincubated (30 minutes at 37°C) in Hanks' balanced saline solution (Invitrogen) containing either 3 µg/ml mouse nonspecific immunoglobulin (IgG; Dako, Carpinteria, CA, http://www.dako.com) or a mouse blocking monoclonal antibody specific to integrin {alpha}2 (P1E6; 1:200), {alpha}4 (P4G9; 1:1,000), {alpha}5 (SAM-1; 1:100), and β1 (JB1A; 1:1,000) subunits (Chemicon). After being washed with phosphate-buffered saline (PBS), hMSCs were pretreated with various concentration of human fetal umbilical vein blood plasma (1:10 to 1:1,000 dilution; 1 hour), recombinant human VEGF-A protein (rhVEGF-A; 0.5–50 ng/ml; Chemicon; 1 hour), or serum-free DMEM (Invitrogen) as a control. The use of 1:100 diluted fetal plasma or 50 ng/ml VEGF-A caused a maximal change in hMSC adhesion, and these concentrations were chosen for subsequent experiments. A total of 2–6 x 104 cells were seeded on precoated wells and allowed to adhere for 30 minutes at 37°C. Adherent cells were quantified by counting in six randomly selected fields per well (magnification, x100; Axiovert 200; Carl Zeiss, Müchen-Hallbergmoos, Germany, http://www.zeiss.com).

The hMSC-endothelium adhesion assay was performed on 24-well plates containing confluent monolayers of human umbilical vein endothelial cells (1 x 105 cells per well). The endothelial cells were activated by overnight incubation with or without tumor necrosis factor (TNF)-{alpha} (2 ng/ml; R&D Systems).

3-(4,5-Dimethylthiazol-2-yl)-5-(3-Carboxymethoxyphenyl)-2-(4-Sulfophenyl)-2H-Tetrazolium Inner Salt Proliferation Assay
Cell proliferation was assessed by a 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium inner salt (MTS) assay (Cell Titer96 Aqueous; Promega, Madison, WI, http://www.promega.com).

Transendothelial Migration
Endothelial cells (2 x 104 cells per well) were layered on 8-µm-pore Transwell® (Corning Costar, Cambridge, MA, http://www.corning.com/lifesciences) precoated with 10 µg/ml fibronectin (Upstate) and cultivated at 37°C for 48 hours [27]. hMSCs were stained with Cell Tracker green-5-chloromethylfluorescein diacetate (Molecular Probes, Eugene, OR, http://probes.invitrogen.com) for 30 minutes. After being washed with PBS, the cells were preincubated with either a blocking antibody to VEGFR-1 (Flt-1/EWC; 1:1,000; Abcam, Cambridge, U.K., http://www.abcam.com) or VEGFR-2 (clone 89106; 0.25 µg/ml; R&D Systems) or with various blocking antibodies as described above at 4°C for 30 minutes. A total of 6–8 x 104 hMSCs were added to the upper chamber of the Transwell® and allowed to migrate for 4 hours with endothelial cell growth medium (Promocell, Heidelberg, Germany, http://www.promocell.com) in the absence or presence of rhVEGF-A (50 ng/ml) or human fetal umbilical vein plasma (1:100) with or without VEGF-A neutralizing antibody (26503; 0.5 µg/ml; R&D Systems) in the lower chamber.

Trans-Matrigel Migration
The 8-µm-pore Transwell® (Corning Costar) were precoated with 25 µl of 1 mg/ml Matrigel (BD Pharmingen) overnight. hMSCs were preincubated with either a nonspecific mouse IgG or a blocking antibody specific to integrin {alpha}2 or β1 subunit, as described above. A total of 1–2 x 105 hMSCs were added to the upper chamber of the Transwell® and allowed to migrate for 24 hours with serum-free DMEM in the presence or absence of rhVEGF-A (50 ng/ml; Chemicon) or human fetal umbilical vein plasma (1:100) in the lower compartment.

Gelatin Zymography
The hMSCs (3 x 105) were incubated in serum-free medium overnight, and rhVEGF-A (0–50 ng/ml; Chemicon) was added to the medium for 24–48 hours. Cell-conditioned medium was then collected for zymographic analysis of matrix metalloproteinase 2 (MMP-2) and MMP-9 activity [28].

VEGF-A Enzyme-Linked Immunosorbent Assay
Human plasma was collected from caesarean section of term delivery. Human plasma VEGF-A levels were evaluated by the Quantikine VEGF immunoassay (analysis sensitivity, <9 pg/ml; R&D Systems). A rat-specific enzyme-linked immunosorbent assay (sensitivity limit, 3 pg/ml; RayBiotech, Norcross, GA, http://www.raybiotech.com) was used for rat plasma.

Reverse Transcription-Polymerase Chain Reaction
The primers included human VEGFR-1 sense, 5'-GAAGGCATGAGGATGAGAGC-3'; antisense, 5'-CAGGCTCATGAACTTGAAAGC-3'; human VEGFR-2 sense, 5'-GGCCAAGTGATTGAAGCAGATG-3'; antisense, 5'-TTCAGATCCACAGGGATT-GCTC-3'; 18S sense, 5'-TAGAGCTAATACATGCCGACGG-3'; antisense, 5'-GGGCCTCGAAAGAGTCCTGTATT-3'. The reverse transcription-polymerase chain reactions (PCRs) were performed according to the prior reports [29, 30].

Animal Preparation
The care of animals complied with the institutional guidelines. All dated pregnant Sprague-Dawley rats were anesthetized (intraperitoneal injection, 0.35 g/kg chloral hydrate) and catheterized through the right jugular vein using a polyethylene tube (PE-50) catheter (BD Pharmingen) for cell transfusion (1 x 107 cells per pregnant rat) on embryonic day (E) 17. On E21, the embryo and placenta were removed from the uterus and dissected. The placenta, fetal blood, and various fetal organs, including brain, heart, lung, liver, and spleen, were collected. For the adult organ study, the pups of the same litter of pregnant rat with cell transfusion were sacrificed after timed breeding (3 and 12 weeks, respectively). For in vivo blocking experiments, the hMSCs were first preincubated with either 3 µg/ml mouse nonspecific IgG or a mouse blocking antibody specific to either integrin {alpha}2, {alpha}4, {alpha}5, β1, VEGFR-1, or VEGFR-2 as that used in vitro before transfusing into pregnant rats.

Immunohistochemistry
Immunohistochemistry was performed as previously described [31] or demonstrated by horseradish peroxidase (HRP)-conjugated secondary antibody (EnVisionTM +/HRP Dual Link Rabbit/Mouse; Dako) and 3,3'-diaminobenzidine tetrahydrochloride staining. Primary antibodies included monoclonal antibodies against β-III-tubulin (TU-20; 1:200; Chemicon), troponin I (1:200; Santa Cruz Biotechnology Inc., Santa Cruz, CA, http://www.scbt.com), surfactant protein D (1:100; Chemicon), albumin (188835; 1:10; R&D Systems), human hepatocyte (OCH1E5; 1:25; Dako), CD45 (MRC OX-1; 1:200; Chemicon), VEGFR-1 (Flt-1/EWC; 1:1,000; Abcam), VEGFR-2 (KDR-1; 1:500; Sigma-Aldrich), cellular fibronectin (FN-3E2; 1:400; Sigma-Aldrich), vascular cell adhesion molecule-1 (VCAM-1) (BBIG-V1; 1:1,000; R&D Systems), and Ki-67 (MIB-1; 1:100; Dako) and polyclonal antibody against von Willebrand factor (1:200; Sigma-Aldrich) and VEGF-A (1:100; Santa Cruz Biotechnology). For the cytokeratin staining, the monoclonal antibody against rat cytokeratin 18 (RGE53; 1:100; Chemicon) and keratin 8 (1:100; Chemicon) was mixed in a 1:1 dilution. Nonspecific rabbit IgG (1:100) and mouse IgG (1:100) were purchased from Dako.

Fluorescent In Situ Hybridization
Cryosections of rat tissue were probed by a rhodamine-labeled X chromosome probe (CEP X, Xp11.1-q11.1, locus DXZ1; Vysis, Bergisch-Gladbach, Germany, http://www.vysis.com) as recommended by the manufacturer. Fluorescent in situ hybridization (FISH) in combination with immunohistochemistry was performed as prior report [32]. After the posthybridization washes, sections were incubated with the antibody against human and rat surfactant protein D (1:100; Chemicon), or human-specific antibody against human hepatocytes (OCH1E5; 1:25; Dako) at room temperature for 1 hour, followed by an FITC-conjugated secondary antibody (Dako) for 1 hour.

Real-Time PCR
Numbers of hMSCs in rat tissue were estimated in a modification of a prior report [33]. The rat organs were weighed, and the genomic DNA of total heart, spleen, and 100 mg of other organs was extracted using a Qiagen kit (Hilden, Germany, http://www1.qiagen.com) and pooled. Human-specific β2-microglobulin primers (sense, 5'-ACCCCCACTGAAAAAGATGAGTATG-3'; antisense, 5'-ACTATCTTGGGCTGTGACAAAGTC-3') and an internal TaqMan detection probe (5'-CCTGCCGTGTGAACCA-3'; Applied Biosystems, Foster City, CA, http://www.appliedbiosystems.com) were used. All sequences were amplified by a first step of 15 seconds at 95°C, followed by 1 minute at 60°C for 50 cycles. No signal amplification was observed in rat tissue without human genomic DNA.

Primers and probes for 18S were obtained from Applied Biosystems (catalog number 18S, Hs99999901_s1; MOL1980000, 198000). Relative quantification of target gene expression was calculated by the comparative CT method. When indicated, the β2-microglobulin signal was normalized against the relative quantity of 18S and expressed as {Delta}CT = CT β2-microglobulin – CT 18S.

A standard curve was generated through quantitative amplification of genomic DNA extracted from a serial dilution of hBM1 cells mixed with the DNA of individual rat placenta homogenates. Total human cell DNA was prepared from 1 x 106 hBM1 cells. Serial dilutions of 25 ng of this DNA preparation in distilled water were used, and the corresponding cell numbers were calculated. This calibration curve was created by plotting the number of hBM1 cells in rat tissue DNA corresponding to the amount of each standard DNA versus the value of its {Delta}CT. The number of transplanted human β2-microglbulin-positive cells for all of the tested samples was expressed as hBM1 cell equivalents per milligram of total rat tissue, as determined by the standard calibration curve.

Statistics
The data are described as means ± SEM. Differences were assessed by using the independent-samples t test, paired t test, Mann-Whitney U test, or Wilcoxon signed rank test when appropriate. p < .05 was considered significant.


   RESULTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosure of Potential...
 Acknowledgments
 References
 
Characterization of hMSCs
Three types of hMSCs originating from human bone marrow were used in this study. The immortalized MSC line (hBM1; passages 30–50) has been reported to express CD29, CD44, CD105, and CD166 but not CD14, CD34, or CD45 [24]. MSCs (passages 6–15) were either derived from frozen bone marrow mononuclear cells supplied by Cambrex [26] (hBM2) or freshly collected by Ficoll-Paque density centrifugation (hBM3) (GE Healthcare Life Sciences, Uppsala, Sweden, http://www4.gelifesciences.com). They have multilineage potential and can be induced to differentiate into adipogenic, osteogenic, and chondrogenic lineages [24, 26]. We further characterized the immunophenotype of these cells by flow cytometry analysis to show expression of CD13, CD54, CD73, CD90, and human leukocyte antigen (HLA)-ABC but not HLA class II antigen. CXCR4 and c-Kit were absent, indicating the lack of receptors for stromal-derived factor-1 (SDF-1) or stem cell factor (SCF) (Fig. 1A).


Figure 1
View larger version (52K):
[in this window]
[in a new window]

  		Figure 1. Characterization of human multipotent mesenchymal stromal cells (hMSCs) and endothelial cells. (A): Flow cytometric analysis of surface-marker expression by hMSCs. The cells expressed CD13, CD44, CD54, CD73, CD90, CD105, CD166, and human leukocyte antigen (HLA)-ABC but were negative for HLA-DP, HLA-DQ, HLA-DR, CD34, CD45, CXCR4, and c-Kit. The shaded areas show the profile of the negative control. Ten thousand cells were counted from each sample. The data shown are representative of three different experiments from hBM1 cells. (B): The expression of adhesion molecules on hMSCs analyzed by flow cytometry. The hBM1 cells expressed high levels of integrin {alpha}2, {alpha}5, and β1 subunits but very low levels of integrin {alpha}4 and β2 subunits. The cells did not express β4 integrin. Integrin {alpha}4 inset: The hBM2 cells expressed a high level of integrin {alpha}4. (C–F): Immunofluorescence of adhesion molecules on hBM1cells. (C): {alpha}2 integrin. (D): {alpha}4 integrin. Very few hBM1 cells were stained. (E): {alpha}5 integrin. (F): β1 integrin. The cell nuclei were counterstained by 4,6-diamidino-2-phenylindole (DAPI). Scale bar = 30 µm. (G, H): A few hBM1 cells were positive for VEGFR-1 immunostaining (G) and negative for VEGFR-2 (H). Scale bar = 60 µm. (I): The hBM2 cells expressed weak VEGFR-2 immunostaining. Scale bar = 60 µm. (J): The three types of hMSCs all expressed VEGFR-1 mRNA, but only hBM2 cells expressed a very low level of VEGFR-2 mRNA. HUVEC acted as a positive control. (K, L): HUVEC expressed high levels of fibronectin (K) and vascular endothelial growth factor A (L). The cell nuclei were counterstained by DAPI. Scale bar = 60 µm. Abbreviations: bp, base pair; HUVEC, human umbilical vein endothelial cell; VEGFR, vascular endothelial growth factor receptor.

 
The cells were further observed to express high levels of integrin subunits {alpha}2, {alpha}5, and β1, very low levels of integrin β2, and no integrin β4 (Fig. 1B). A low level of integrin {alpha}4 was observed in hBM1 and hBM3 cells (Fig. 1B; supplemental online Fig. 1B), but its expression was high in hBM2 cells (Fig. 1B, inset of integrin {alpha}4; supplemental online Fig. 1A). In immunofluorescence, the hBM1 cells showed strong staining for integrins {alpha}2, {alpha}5, and β1, but only a few cells expressed integrin {alpha}4 (Fig. 1C–1F). A few hBM1 cells were positive by immunostaining for VEGFR-1 (Fig. 1G). The hBM1 and hBM3 cells were negative for VEGFR-2 (Fig. 1H, 1J), whereas the hBM2 cells expressed a low level of mRNA encoding VEGFR-2 with very weak immunostaining (Fig. 1I, 1J). Because the phenotypes of hBM1 and hBM3 were similar, hBM1 and hBM2 were used for the following in vitro and in vivo experiments.

Fibronectin Is Highly Expressed at the Maternal-Fetal Interface
The results suggest that integrins {alpha}2β1 and {alpha}5β1 are consistently present on these hMSCs. Integrin {alpha}2β1 is a ligand for laminin and collagen, and integrin {alpha}5β1 is a ligand for fibronectin [21]. In the placental barrier, endothelial cells provide a platform for cell adhesion at the maternal-fetal interface. As shown in Figure 1K and 1L, endothelial cells express high levels of fibronectin and VEGF-A, which may interact with integrin {alpha}5β1 and VEGFR-1 on the cell surface of hMSCs.

Fibronectin was found to be abundantly expressed in human placental endothelial cells, villous stroma, and decidual cells. Similarly, fibronectin was expressed in the lining cells of rat placental labyrinth, decidua, and uterine wall vessels (Fig. 2A–2D). Furthermore, VEGF-A was identified in the syncytiotrophoblastic layer, with staining also being apparent in some stromal cells and vessels of term placenta (Fig. 2E), which was consistent with prior reports [13, 14]. hBM2 cells expressed a high level of integrins {alpha}4 and β1, indicating that integrin {alpha}4β1 is present in these cells. VCAM-1 and connecting segment-1 of cellular fibronectin are reported to be the ligands for integrin {alpha}4β1 [34, 35]. However, VCAM-1 was not detected in trophoblast or endothelial cells in the placental vasculature (Fig. 2F) [36, 37].


Figure 2
View larger version (64K):
[in this window]
[in a new window]

  		Figure 2. Fibronectin and vascular endothelial growth factor A (VEGF-A) at the maternal-fetal interface. (A): Abundant fibronectin expression in human placenta, including villous vessel walls and endothelial cells (arrow). (B): Fibronectin expression in the lining trophoblast and endothelial cells of the rat placental labyrinth obtained at embryonic day 21 (arrow). (C): Strong fibronectin staining in human decidua. (D): Strong fibronectin staining in rat decidua and attached myometrium. (E): VEGF-A expression in the syncytiotrophoblastic layer of human placenta (large arrow). Staining was also observed in the villous stroma and villous vessel (small arrow). Inset: Placental villus was immunostained by rabbit nonspecific IgG as a control. (F): Vascular cell adhesion molecule-1 was negative in the vessel and trophoblasts of human placental villus. The tissues were counterstained with Mayer's hematoxylin. Scale bar = 50 µm. (G): VEGF-A levels in human maternal and umbilical vein blood plasma examined by enzyme-linked immunosorbent assay. VEGF-A concentration in fetal umbilical vein blood plasma was significantly greater than that in maternal plasma (10.7 ± 4.3 pg/ml vs. 1.3 ± 0.2 pg/ml; p < .001). (H): There was no significant difference in the VEGF-A level of the rat samples from maternal artery and vein (25.7 ± 7.7 pg/ml vs. 21.3 ± 6.8 pg/ml; p = .128). There was significantly higher VEGF-A in rat fetal plasma (35.1 ± 11.0 pg/ml; p = .018). Maternal rat arterial and venous blood were collected from femoral vessels, and fetal blood was collected after decapitation. Abbreviations: V, vessel of uterine wall; W, uterine wall.

 
A Maternofetal VEGF-A Concentration Gradient
There were no significant differences observed between samples from human brachial artery and vein or mesenteric/uterine vein [38]. Therefore, we simultaneously analyzed the plasma from brachial veins of pregnant women and umbilical veins of their fetuses for VEGF-A. As shown in Figure 2G, the human maternal plasma VEGF-A levels were almost undetectable, perhaps because of the presence of high molecular weight VEGF-binding protein in pregnant women [39]. VEGF-A concentration in human umbilical vein plasma was low but was significantly greater than that in maternal plasma (10.7 ± 4.3 pg/ml vs. 1.3 ± 0.2 pg/ml; p < .001). The VEGF-A concentration in maternal rat venous blood plasma was consistent with a prior report [40] and was similar in the samples from maternal rat artery. However, VEGF-A was significantly higher in fetal rat plasma (35.1 ± 11.0 pg/ml; p = .018; Fig. 2H). Thus, there is a gradient of VEGF-A from maternal blood to the fetus. High concentration of VEGF-A in the fetal blood may play a role in the mobilization of maternal circulating stem cells.

VEGF-A and Cord Blood Plasma Enhance hMSC Adhesion to Fibronectin and Laminin Through Integrins {alpha}4β1, {alpha}5β1, and {alpha}2β1
The hMSCs spontaneously adhered to fibronectin and laminin, and adhesion was significantly increased as VEGF-A concentration increased (supplemental online Fig. 2). VEGF-A did not affect hMSC proliferation (supplemental online Fig. 3). Cell adhesion was maximum upon stimulation with VEGF-A (50 ng/ml) or fetal umbilical vein plasma (1:100; Fig. 3A–3D). VEGF-A or fetal plasma-enhanced cell adhesion to fibronectin was significantly blocked by antibodies against integrin {alpha}5 or β1 subunit in hBM1 and was blocked by antibodies against integrin {alpha}4, {alpha}5, or β1 subunit in hBM2, indicating that VEGF-A or fetal plasma-mediated hMSC adhesion on fibronectin is dependent on integrin {alpha}5β1 in hBM1 and on both {alpha}4β1 and {alpha}5β1 in hBM2 (Fig. 3A, 3C). VEGF-A or fetal plasma-enhanced hMSC adhesion to laminin was significantly blocked by antibodies against integrin {alpha}2 or β1 subunit, indicating VEGF-A or fetal plasma-mediated hMSC adhesion on laminin is dependent on integrin {alpha}2β1. There was no significant alteration of hMSC adhesion to laminin in the presence of antibody against integrin {alpha}5 (Fig. 3B, 3D).


Figure 3
View larger version (51K):
[in this window]
[in a new window]

  		Figure 3. Adhesion of human multipotent mesenchymal stromal cells (hMSCs) to fibronectin, laminin, endothelial cells and proliferation assay. VEGF (50 ng/ml) and HP (1:100) enhanced hMSC adhesion to fibronectin (A, C) and laminin (B, D). The adhesion of hBM1 and hBM2 cells to fibronectin and laminin was significantly increased by VEGF or HP. (A): The adhesion of hBM1 to fibronectin was significantly inhibited by blocking antibody against β1 integrin (1:1,000) or {alpha}5 integrin (1:100) but not by nonspecific IgG (3 µg/ml) or antibody against {alpha}4 integrin (1:1,000). (B): The adhesion of hBM1 to laminin was significantly inhibited by blocking antibody against β1 integrin (1:1,000) or {alpha}2 integrin (1:100) but not by nonspecific IgG (3 µg/ml) or antibody against {alpha}5 integrin (1:100). (C): The adhesion of hBM2 to fibronectin was significantly inhibited by blocking antibodies as indicated in (A). (D): The adhesion of hBM2 to laminin was significantly inhibited by blocking antibodies as indicated in (B). (E): VEGF (50 ng/ml) and HP enhanced hMSC adhesion to endothelial cells. The endothelial cell was induced to express VCAM-1 by TNF-{alpha}. (Immunostaining figures in lower panel: cell nuclei were counterstained by 4,6-diamidino-2-phenylindole. Scale bar = 30 µm) The expression of VCAM-1 did not enhance the hBM2 cell adhesion to endothelial cells. (F): hMSCs were seeded into a 96-well culture plate (5 x 103 cells per well) after preincubating with a blocking antibody to VEGFR-1 (1:1,000) and various blocking antibodies as described above and cultured for 24–48 hours. 3-(4,5-Dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt assay revealed that the hBM2 proliferation was not affected by blocking antibody treatment. The data shown are the mean ± SEM of three separate experiments. **, p < .01; ***, p < .001, compared with NC of each category. Abbreviations: DMEM, Dulbecco's modified Eagle's medium; HP, human umbilical vein plasma; NC, control; TNF, tumor necrosis factor; VCAM, vascular cell adhesion molecule; VEGF, vascular endothelial growth factor A; VEGFR, vascular endothelial growth factor receptor.

 
VEGF-A (50 ng/ml) and human fetal plasma (1:100) enhance hMSC adhesion to endothelial cells (Fig. 3E). Since VCAM-1 is a ligand for integrin {alpha}4β1 [35], we further examined the effect of VCAM-1 expression on hBM2 cell adhesion. VCAM-1 expression was stimulated in endothelial cells by treatment with TNF-{alpha}, but this did not enhance hBM2 cell adhesion to endothelial cells (Fig. 3E). VCAM-1 was not detected in the vascular system of normal term placenta (Fig. 2F). High levels of fibronectin are expressed by endothelial cells (Fig. 1K), and this may play a role in mediating hMSC transplacental migration via its binding to integrins {alpha}4β1 and {alpha}5β1, activated by VEGF-A or fetal plasma. Treating hMSCs with various antibodies against integrin subunits and VEGFR-1 did not affect cell proliferation (Fig. 3F). The number of viable cells in these cells cultured for 2 days, directly counted by trypan blue exclusion, was not different (not shown), a result that was confirmed by the MTS assay for cell proliferation. This finding was further verified using hMSCs transfected with integrin β1 small interfering RNA (siRNA) (supplemental online Methods). Integrin β1 mRNA and protein were significantly diminished in hMSCs transfected with integrin β1 siRNA after 24, 48, and 72 hours. However, hMSC proliferation was not affected (supplemental online Fig. 4A–4C).

VEGF-A Induces hMSC Transendothelial Cell Migration, an Effect Mediated Through VEGFR-1
We further observed that VEGF-A had a dose-dependent (0.5–50 ng/ml) chemotactic effect on transendothelial cell migration that was maximum at 50 ng/ml (not shown). VEGF-A (50 ng/ml) and fetal plasma (1:100) significantly induced transendothelial migration (Fig. 4A–4D). This effect was significantly inhibited by anti-rhVEGF-A or anti-VEGFR-1 antibody but not by anti-VEGFR-2 antibody (Fig. 4A–4D). This indicated that the low level of VEGFR-2 in hBM2 cells did not have a role in transendothelial migration. These effects were also abrogated by application of blocking antibodies against integrin {alpha}5 or β1 on hBM1 and integrin {alpha}4, {alpha}5, or β1 on hBM2 (Fig. 4A–4D). Therefore, VEGF-A and integrins {alpha}4β1 and {alpha}5β1 synergize to mobilize hMSC transendothelial cell migration.


Figure 4
View larger version (36K):
[in this window]
[in a new window]

  		Figure 4. VEGF (50 ng/ml) and HP (1:100) induced human multipotent mesenchymal stromal cell transendothelial and trans-Matrigel migration. (A, B): VEGF (A) and HP (B) significantly induced the transendothelial migration of hBM1 cells. The transendothelial migration effect was inhibited by treatment with blocking antibody to VEGFR-1 (1:1,000), integrin β1 (1:1,000), and integrin {alpha}5 (1:100) but not anti-VEGFR-2 antibody (0.25 µg/ml), anti-integrin {alpha}4 (1:1,000), or mouse nonspecific IgG (3 µg/ml). The HP-induced hBM1 cell transendothelial migration was further inhibited by neutralizing antibody against recombinant human VEGF-A (0.5 µg/ml). (C, D): VEGF (C) and HP (D) significantly induced the transendothelial migration of hBM2 cells. The transendothelial migration effect was inhibited by blocking antibodies as indicated in (A) and (B). The VEGF- and HP-induced hBM2 cell transendothelial migration was further inhibited by blocking antibody against integrin {alpha}4 (1:1,000). (E, F): VEGF and HP induced hBM1 (E) and hBM2 (F) cells to migrate across Matrigel. Migration was significantly inhibited by blocking antibody against integrin {alpha}2 (1:200) or β1 (1:1,000) but not mouse nonspecific IgG (3 µg/ml). (G): Gelatin zymography showed that hBM1 secrete MMP-2 but not MMP-9 (92 kD) into culture medium. Pro-MMP-2 was visible at approximately 72 kD and active MMP-2 at 62 kD. MMP-2 was higher at day 2 than at day 1. The data shown are the mean ± SEM of three separate experiments. **, p < .01; ***, p < .001, compared with NC of each category. Abbreviations: EGM, endothelial cell growth medium; HP, human umbilical vein plasma; kD, kilodaltons; MMP, matrix metalloproteinase; NC, control; VEGF, vascular endothelial growth factor A; VEGFR, vascular endothelial growth factor receptor.

 
Integrin {alpha}2β1 Mediates Transmigration of hMSCs Through a Reconstituted Basement Membrane Barrier
VEGF-A and human umbilical vein plasma stimulated migration of hMSCs across a reconstituted basement membrane (Matrigel) barrier (Fig. 4E, 4F). Transmigration was inhibited by blocking antibody to either integrin {alpha}2 or integrin β1. This result suggests that VEGF-A and integrin {alpha}2β1 act cooperatively to stimulate hMSC migration across basement membrane.

Migration across basement membranes may also be dependent on MMP production. Culture medium from either hBM1 or hBM2 showed a time-dependent accumulation of pro-MMP-2 (72 kD) and active MMP-2 (62 kD). hBM3 secreted MMP-9 but not MMP-2 (Fig. 4G; supplemental online Fig. 1C). VEGF-A did not enhance MMP-2 or MMP-9 expression by hMSCs.

Engraftment of hMSCs in Fetal Organs
The ability of MSCs of maternal origin to traffic through the placenta was studied using a rat model. Bis-benzimide-labeled hMSCs (1 x 107 cells) were transplanted intravenously into pregnant rats at E17. There were neither transplant-related mortalities nor morbidities associated with immunorejection.

To show that hMSCs migrated from placenta to the various fetal organs by way of the bloodstream, we collected fetal blood and various organ samples at E21 (term is E21 ± 1 day), 5 days after hMSCs were transplanted. Organ specimens were also collected 3 and 12 weeks postnatally (supplemental online Table 1). Most fetal tissues had demonstrable human cell engraftment at E21. Although the distribution pattern and number of hMSCs in individual rat fetuses differed, hBM1 cells were detectable in more than 70% of fetal rats (supplemental online Table 1), with no evidence of rejection of xenogeneic cells. Similarly to hBM1, hBM2 cells could migrate transplacentally from the maternal blood stream into the fetus and engraft in various fetal organs. hBM1 engraftment in fetal organs is shown in Figures 5 and 6.


Figure 5
View larger version (69K):
[in this window]
[in a new window]

  		Figure 5. Immunofluorescence of fetal rat organs at embryonic day 21, 5 days after transplantation of bis-benzimide-labeled human multipotent mesenchymal stromal cells (hMSCs). The hMSCs (blue fluorescence, arrows) were found to be present in various fetal rat tissues, which were immunostained using fluorescein isothiocyanate-conjugated antibodies. (A): Several hMSCs (arrow) were seen in the placental labyrinth, which was immunostained for cytokeratins 8 and 18. (B): A large maternal blood vessel at its point of entry into the labyrinth, stained by antibody against von Willebrand factor. An hMSC is seen interacting closely with, and possibly migrating through, the endothelial layer (arrow). (C): Several blue-stained hMSC nuclei in an area of rat placental labyrinth. Some were clearly Ki67-positive (arrow). (D): Rat brain was immunostained by an antibody against β-III-tubulin. (E): Rat heart was immunostained for troponin I. (F): Rat lung was immunostained for surfactant protein D. (G): Rat liver was immunostained for albumin. (H): Rat spleen was immunostained for CD45. Small arrow, CD45-positive cells. (I): Fluorescent in situ hybridization of rat liver. The cell nuclei were counterstained by 4,6-diamidino-2-phenylindole. The presence of the hMSC (arrow) was revealed by the human-specific rhodamine-labeled X chromosome probe. (J, K): Smear of rat blood under a bright field (J) and a matched image under fluorescence illumination (K). Two bis-benzimide-labeled hMSCs were observed (arrow). (L): Rat spleen was immunostained for mouse nonspecific IgG as a negative control. (A, B, D, F–I): Insets show a twofold magnification of the indicated cells (arrows). Scale bars = 20 µm (C, F, I) and 45 µm (A, B, D, E, G, H, J, K, L).

 


Figure 6
View larger version (89K):
[in this window]
[in a new window]

  		Figure 6. Fluorescent in situ hybridization (FISH) with or without immunohistochemistry of rat tissues probed with a human-specific rhodamine-labeled X chromosome probe. The cell nuclei were stained blue, and the X chromosome centromeres were stained red (arrow). Female human cells are clearly seen within rat tissues: placenta (A), brain (B), heart (C), lung (D), liver (E), and spleen (F); Scale bar = 20 µm. (G): Female hBM1 cells cultured in vitro were negative for human-specific hepatocyte staining. (H): Combining FISH with immunohistochemistry to show the engrafted hBM1 differentiation in the alveoli of rat lung. Two female cells bearing two X chromosomes labeled with rhodamine (arrow) were located in the cell stained with surfactant protein D. This antibody immunoreacts with both human and rat surfactant protein D. Nuclei were counterstained with 4,6-diamidino-2-phenylindole (DAPI). (I): One female cell bearing two X chromosomes labeled with rhodamine was located in the cell stained with specific antibody against human hepatocyte. Nuclei were counterstained with DAPI. The microchimeric cell was in a sinusoid area of the liver. (J): A positive control for the human-specific antibody against human hepatocytes of human hepatic tissue that was obtained from a patient who underwent major surgery for hepatoma. Scale bars = 25 µm. (A–D, F–I): Insets show a twofold magnification of the indicated cells (arrows).

 
Bis-benzimide-labeled hMSCs were identified in the placental labyrinth, which was immunostained for cytokeratins 8 and 18 (Fig. 5A). Cells adhered to and migrated through the wall of a large placental vessel carrying maternal blood into the labyrinth (Fig. 5B). In areas of placental labyrinth containing abundant hMSCs, we observed nuclei of bis-benzimide-labeled hMSCs that were also positive for Ki67, indicating that these cells have proliferation activity (Fig. 5C). We assessed the presence of hMSCs in the rat brain cortex using immunohistochemistry with an antibody against β-III-tubulin. Myocardiac fibers were immunostained for troponin I. Pulmonary alveoli were immunostained for surfactant protein D. The hMSCs were present in these organs. The bis-benzimide-labeled hMSCs seemed to colocalize with the immunoreactivity of surfactant protein D in the alveoli of rat lung. Bis-benzimide-labeled hMSCs in the sinusoid of fetal liver showed albumin immunoreactivity. hMSCs were present in the parenchyma of fetal spleen that was immunostained for CD45, a marker for leukocytes (Fig. 5D–5H).

The identities of hMSCs present in fetal tissue were further confirmed by FISH with a human X chromosome-specific probe. A representative image of liver is shown (Fig. 5I). The hMSCs could also be detected in a fetal blood smear collected 5 days after maternal transplantation (Fig. 5J, 5K).

hMSCs persisted in postnatal (3 and 12 weeks) tissues, including brain, lung, heart, liver, and spleen (supplemental online Table 1). Again, FISH was used to study various organ samples. hMSCs showed two positive signals in each cell, revealing that they contained two X chromosomes in vitro (Fig. 6G). Analysis of organs from rats whose mothers had been transplanted with hMSCs yielded a surprisingly extensive distribution repertoire, including brain, heart, lung, liver, and spleen (Fig. 6A–6F). The cells persisted postnatally for more than 3 months after transplantation without any graft-versus-host reaction or graft rejection, consistent with maternal microchimerism [41].

FISH and immunofluorescence were combined to determine whether hMSCs underwent differentiation after engraftment. Several human X chromosome-positive cells stained for surfactant protein D in the alveoli of rat lung. However, the antibody reacts with both human and rat surfactant protein D (Fig. 6H). Very low numbers (one or two) of human X chromosome-positive cells stained with a human-specific anti-hepatocyte antibody in liver sinusoids 3 weeks after birth (Fig. 6I), but this suggests that the engrafted hMSCs may differentiate into liver cells. Staining in individual color channels is shown in supplemental online Figure 5.

Integrin {alpha}2, {alpha}4, {alpha}5, and β1 Subunits and VEGFR-1 Mediate hMSC Transplacental Trafficking In Vivo
To determine whether integrin {alpha}2, {alpha}4, {alpha}5, and β1 subunits and VEGFR-1 mediate hMSC attachment to the endothelial cells and transmigration through the ECM barrier of basal lamina and placental stroma in vivo, hMSCs were injected into pregnant rats in the presence of different blocking antibodies against integrin subunits and VEGFR-1. A real-time quantitative PCR assay for human β2-microglbulin was developed to evaluate hMSC numbers in the treated animals. DNA extracted from rat tissue was mixed with DNA extracted from hBM1 cells (Fig. 7A–7D) to produce a highly precise and reproducible standard curve over a wide linear range down to five hBM1 cells in 250 ng of genomic DNA.


Figure 7
View larger version (36K):
[in this window]
[in a new window]

  		Figure 7. In vivo study of maternofetal cell trafficking by quantitative real-time polymerase chain reaction (PCR) analysis of human multipotent mesenchymal stromal cell (hMSC) number in rat tissues collected at embryonic day 21 (5 days after transplant). (A): Quantitative real-time PCR for human-specific β2-microglobulin sequences was performed on 25 ng of genomic DNA extracted from human stem cells (hBM1) serially diluted into individual rat placenta homogenates. (B): Real-time PCR of genomic DNA from (A), targeting sequences that encode 18S rRNA. (C): Genomic DNA of rat placenta was mixed with the DNA equivalent of 5–11,068 hBM1 cells. The amount of rat DNA equivalent to 1 mg of rat tissue was measured. (D): Real-time quantitative β2-microglobulin PCR was used to generate a standard curve. The {Delta}CT values are plotted as a function of human cell number equivalent added per milligram of rat tissue. The human cell number in a particular rat sample was determined by interpolation from a standard curve of {Delta}CT values generated from the hBM1 DNA dilution series. (E–K): The hMSCs were treated with blocking antibodies or mouse nonspecific IgG as in the in vitro study. Postmortem analysis of organs from fetal rats confirmed that anti-{alpha}2, -{alpha}4, -{alpha}5, and -β1 integrin and -VEGFR-1 antibodies all inhibited transplacental trafficking of hMSCs. The mean numbers of hBM1 cells (numbers above the x-axes) in the placenta (E), brain (F), heart (G), lung (H), liver (I), and spleen (J) of fetal rat were all significantly decreased. (K): The mean numbers of hBM2 cells in various fetal rat organs with or without anti-{alpha}4 antibody blocking are shown above the x-axis. (L): The mean numbers of hBM2 cells in various fetal rat organs with or without anti-VEGFR-2 antibody blocking are shown above the x-axis. The anti-VEGFR-2 antibody did not reduce the number of hBM2 crossing the placenta. n indicates number of fetuses from different dams. The Mann-Whitney U test was used in the data analysis. Abbreviations: IgG, mouse nonspecific immunoglobulin; NC, control; VEGFR, vascular endothelial growth factor receptor.

 
Postmortem analysis of the organs from E21 fetal rats confirmed that anti-{alpha}2, anti-{alpha}5, and anti-β1 integrin and anti-VEGFR-1 antibodies inhibit hBM1 cell transplacental trafficking (Fig. 7E–7J). Anti-{alpha}4 antibody but not anti-VEGFR-2 antibody further inhibited transplacental trafficking of hBM2 cells (Fig. 7K, 7L). These blocking antibodies substantially reduced the number of hMSCs crossing the placenta and engraftment in the organs of fetal rat at E21 as estimated by real-time PCR for human-specific β2-microglbulin sequences (Fig. 7E–7K). These observations indicate that integrins {alpha}2β1, {alpha}4β1, and {alpha}5β1 and VEGFR-1 are involved in hMSC trafficking through the placenta.


   DISCUSSION
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosure of Potential...
 Acknowledgments
 References
 
Previous data indicate the presence of a small but significant population of mesenchymal stem cells in the blood circulation of adult women [10, 42]. Trafficking of maternal MSCs to the fetus during pregnancy has not been demonstrated. Transplantation of large quantities of hMSCs may not be an accurate reflection of a physiological process. However, we have used cell tracking [43] and FISH to demonstrate conclusively that these cells are capable of migration during pregnancy from the maternal circulation across the placenta and into fetal blood and that they colonize multiple fetal tissues (brain, heart, lung, liver, and spleen) and survive for as long as 12 postnatal weeks.

Maternal cell transmission to offspring might occur by transplacental passage in utero or postnatally by breast-feeding [44]. To explore the mechanism of maternal stem cell transmission to conceptuses in utero, we delivered cells intravenously at E17 and assessed fetal organs collected at E21, as well as 3–12 weeks after delivery. Marleau et al. [7] reported maternal cells first appearing in fetal thymus at E12.5, coincident with the maturation of the hemochorial interface [1]. Morphologically heterogeneous maternal cells were present predominantly in bone marrow and thymus in more than 90% of late-gestation fetuses. Others have reported maternal cells in nonlymphoid organs [44].

After transfer of undifferentiated hMSCs into pregnant rat, these cells showed proliferative activity in the placenta, suggesting that a maternal cell population may expand in the fetal environment. MSCs have been reported to have local immunosuppressive properties that permit them to escape the immune system and survive transplantation in an allogeneic setting [45]. It has been shown that maternal microchimerism can give rise to differentiated, tissue-specific cell types, including pancreatic islet β cells [46], cardiomyocytes [47], and keratinocytes [48], in offspring. Human cord blood-derived cells can differentiate into hepatocytes in the mouse liver [49], and hMSCs colonized multiple fetal sheep tissues for as long as 13 months after in utero transplantation [50]. Differences observed in cell numbers may be due to colonization efficiency in different tissue environments or the rate of cell turnover in each organ [51]. In the present study, albumin staining of engrafted cells may suggest hepatocytic differentiation. By the use of FISH and immunostaining, we have been able to confirm the differentiation of a few maternally derived cells into hepatocytes. The developmental environment of the fetus may enable a wider spectrum of cell fates than in an adult tissue, but further work will be required to investigate other cell lineages and their biological significance.

hMSCs do not express CXCR4 and c-Kit, thus excluding the involvement of SDF-1 or SCF in stem cell mobilization [52–54]. Several observations support the hypothesis that VEGF-A may be a key factor in transplacental migration of hMSCs. The placenta expresses VEGF-A, which may be secreted into fetal blood; VEGF-A shows a concentration gradient from maternal to fetal circulations. hMSC migration across basement membrane matrix and endothelial monolayers is stimulated by VEGF-A in vitro, and this is VEGFR-1-dependent. Finally, antibodies that block VEGF-receptor interaction inhibit the engraftment of fetal tissues in vivo. This observation is supported by the report that plasma elevation of VEGF-A promotes mobilization of hematopoietic stem cells [18]. Bone marrow-derived circulating cells can be summoned by VEGF-A and participate in adult neovascularization [55]. It is known that VEGFR-1 signaling is involved in the reconstitution of hematopoiesis by recruiting hematopoietic stem cells [56]. VEGF-A stimulates chemotactic migration of human mesenchymal progenitor cells via VEGFR-1 [20]. Furthermore, VEGF-A can activate β1 integrins or increase avidity, and this may induce hMSC adhesion and migration [57]. Since fetal MSCs traffic to the maternal side (and become engrafted) more efficiently then the reverse process [4], it would appear that fetal MSCs must respond to signals other than VEGF-A. Distinct adhesion and chemokine/growth factor systems may be involved.

The route of stem cell migration through the placenta remains to be determined. In hemochorial species, including rat and human, passage across the trophoblast barrier, or discontinuities therein, must occur. Cells must then traverse the fetal endothelium. Placental endothelial cells are fibronectin-positive, and maternal MSCs can pass across endothelial monolayers in vitro in an integrin {alpha}4β1-/{alpha}5β1-dependent mechanism. Furthermore, antibodies to this integrin inhibit transplacental passage of stem cells in vivo. Integrin {alpha}4β1 has been found to promote the homing of circulating bone marrow-derived progenitor cells to VCAM-1 and cellular fibronectin during neovascularization [58]. However, VCAM-1 is expressed in neither trophoblasts nor endothelial cells of normal term placenta. Since integrin {alpha}5β1 is a high-affinity fibronectin receptor, a role for fibronectin is strongly indicated [59]. Reagents blocking integrin {alpha}2β1 powerfully block hMSC adhesion on laminin in vitro and fetal rat engraftment, perhaps reflecting the ability of this integrin to bind basement membrane ligands (collagens and laminins) found between the trophoblast and endothelial cell layers of the placenta [31, 60]. MMP-2 and MMP-9 are expressed by hMSCs, can digest fibronectin, laminin, and collagen types I, IV, VII, and X [61], and may facilitate transplacental migration [62].

We suggest that maternal stem cells are responsible for fetal microchimerism in normal pregnancy. Other adhesion molecules or other stem/progenitor cells (e.g., hematopoietic stem cells) may also participate in maternal-to-fetal cell migration through signaling pathways triggered by fetal chemokines or growth factors. There are multiple clinical implications. Since stem cells can proliferate in fetal niches [63], maternal venous administration of hMSCs with the capacity to differentiate into a repertoire of mature mesenchymal cells might prove useful in the treatment of intrauterine fetal disease [64]. For example, mononuclear cell from human umbilical cord blood can be delivered prenatally into the {alpha}-N-acetylglucosaminidase enzyme-deficient fetus in the Sanfilippo type B disease model [65]. On the other hand, circulating maternal stem cells may be responsible for vertical transmission of maternally borne viruses to the fetus, demanding fresh consideration of perinatal infectious disease and its prevention. It remains to be determined whether other maternal cell types also colonize fetal tissues and whether maternofetal cell trafficking is enhanced as a result of pathological conditions.


   DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosure of Potential...
Acknowledgments
 References
 
The authors indicate no potential conflicts of interest.


   ACKNOWLEDGMENTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosure of Potential...
 Acknowledgments
References
 
This work was supported by Grant MMH-E 95001 from Mackay Memorial Hospital.


   REFERENCES
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosure of Potential...
 Acknowledgments
 References
 

  1. Piotrowski P, Croy BA. Maternal cells are widely distributed in murine fetuses in utero. Biol Reprod 1996;54:1103–1110.[Abstract]

  2. Maloney S, Smith A, Furst DE et al. Microchimerism of maternal origin persists into adult life. J Clin Invest 1999;104:41–47.[Medline]

  3. Barinaga M. Cells exchanged during pregnancy live on. Science 2002;296:2169–2172.[Abstract/Free Full Text]

  4. O'Donoghue K, Chan J, de la Fuente J et al. Microchimerism in female bone marrow and bone decades after fetal mesenchymal stem-cell trafficking in pregnancy. Lancet 2004;364:179–182.[CrossRef][Medline]

  5. Lo ES, Lo YM, Hjelm NM et al. Transfer of nucleated maternal cells into fetal circulation during the second trimester of pregnancy. Br J Haematol 1998;100:605–606.[CrossRef][Medline]

  6. Srivatsa B, Srivatsa S, Johnson KL et al. Maternal cell microchimerism in newborn tissues. J Pediatr 2003;142:31–35.[CrossRef][Medline]

  7. Marleau AM, Greenwood JD, Wei Q et al. Chimerism of murine fetal bone marrow by maternal cells occurs in late gestation and persists into adulthood. Lab Invest 2003;83:673–681.[Medline]

  8. Adamson SL, Lu Y, Whiteley KJ et al. Interactions between trophoblast cells and the maternal and fetal circulation in the mouse placenta. Dev Biol 2002;250:358–373.[CrossRef][Medline]

  9. Rowley SD, Yu J, Gooley T et al. Trafficking of CD34+ cells into the peripheral circulation during collection of peripheral blood stem cells by apheresis. Bone Marrow Transplant 2001;28:649–656.[CrossRef][Medline]

  10. Huss R, Lange C, Weissinger EM et al. Evidence of peripheral blood-derived, plastic-adherent CD34(-/low) hematopoietic stem cell clones with mesenchymal stem cell characteristics. STEM CELLS 2000;18:252–260.[Abstract/Free Full Text]

  11. Bianchi DW. Fetomaternal cell trafficking: A new cause of disease? Am J Med Genet 2000;91:22–28.[CrossRef][Medline]

  12. Bianchi DW, Farina A, Weber W et al. Significant fetal-maternal hemorrhage after termination of pregnancy: Implications for development of fetal cell microchimerism. Am J Obstet Gynecol 2001;184:703–706.[CrossRef][Medline]

  13. Lyall F, Young A, Boswell F et al. Placental expression of vascular endothelial growth factor in placentae from pregnancies complicated by pre-eclampsia and intrauterine growth restriction does not support placental hypoxia at delivery. Placenta 1997;18:269–276.[CrossRef][Medline]

  14. Vuorela P, Hatva E, Lymboussaki A et al. Expression of vascular endothelial growth factor and placenta growth factor in human placenta. Biol Reprod 1997;56:489–494.[Abstract]

  15. Baker PN, Krasnow J, Roberts JM et al. Elevated serum levels of vascular endothelial growth factor in patients with preeclampsia. Obstet Gynecol 1995;86:815–821.[CrossRef][Medline]

  16. Galazios G, Papazoglou D, Giagloglou K et al. Umbilical cord serum vascular endothelial growth factor (VEGF) levels in normal pregnancies and in pregnancies complicated by preterm delivery or pre-eclampsia. Int J Gynaecol Obstet 2004;85:6–11.[CrossRef][Medline]

  17. Takahashi H, Shibuya M. The vascular endothelial growth factor (VEGF)/VEGF receptor system and its role under physiological and pathological conditions. Clin Sci (Lond) 2005;109:227–241.[Medline]

  18. Hattori K, Dias S, Heissig B et al. Vascular endothelial growth factor and angiopoietin-1 stimulate postnatal hematopoiesis by recruitment of vasculogenic and hematopoietic stem cells. J Exp Med 2001;193:1005–1014.[Abstract/Free Full Text]

  19. Krysiak O, Bretschneider A, Zhong E et al. Soluble vascular endothelial growth factor receptor-1 (sFLT-1) mediates downregulation of FLT-1 and prevents activated neutrophils from women with preeclampsia from additional migration by VEGF. Circ Res 2005;97:1253–1261.[Abstract/Free Full Text]

  20. Fiedler J, Leucht F, Waltenberger J et al. VEGF-A and PlGF-1 stimulate chemotactic migration of human mesenchymal progenitor cells. Biochem Biophys Res Commun 2005;334:561–568.[CrossRef][Medline]

  21. Stupack DG, Cheresh DA. Get a ligand, get a life: Integrins, signaling and cell survival. J Cell Sci 2002;115:3729–3738.[Abstract/Free Full Text]

  22. Papayannopoulou T, Priestley GV, Nakamoto B et al. Molecular pathways in bone marrow homing: Dominant role of alpha(4)beta(1) over beta(2)-integrins and selectins. Blood 2001;98:2403–2411.[Abstract/Free Full Text]

  23. Premack BA, Schall TJ. Chemokine receptors: Gateways to inflammation and infection. Nat Med 1996;2:1174–1178.[CrossRef][Medline]

  24. Okamoto T, Aoyama T, Nakayama T et al. Clonal heterogeneity in differentiation potential of immortalized human mesenchymal stem cells. Biochem Biophys Res Commun 2002;295:354–361.[CrossRef][Medline]

  25. Pittenger MF, Mackay AM, Beck SC et al. Multilineage potential of adult human mesenchymal stem cells. Science 1999;284:143–147.[Abstract/Free Full Text]

  26. Chang YJ, Shih DT, Tseng CP et al. Disparate mesenchyme-lineage tendencies in mesenchymal stem cells from human bone marrow and umbilical cord blood. STEM CELLS 2006;24:679–685.[Abstract/Free Full Text]

  27. Peled A, Kollet O, Ponomaryov T et al. The chemokine SDF-1 activates the integrins LFA-1, VLA-4, and VLA-5 on immature human CD34(+) cells: Role in transendothelial/stromal migration and engraftment of NOD/SCID mice. Blood 2000;95:3289–3296.[Abstract/Free Full Text]

  28. Johnson DW, Saunders HJ, Baxter RC et al. Paracrine stimulation of human renal fibroblasts by proximal tubule cells. Kidney Int 1998;54:747–757.[CrossRef][Medline]

  29. Bortoloso E, Del Prete D, Dalla Vestra M et al. Quantitative and qualitative changes in vascular endothelial growth factor gene expression in glomeruli of patients with type 2 diabetes. Eur J Endocrinol 2004;150:799–807.[Abstract]

  30. Feistritzer C, Kaneider NC, Sturn DH et al. Expression and function of the vascular endothelial growth factor receptor FLT-1 in human eosinophils. Am J Respir Cell Mol Biol 2004;30:729–735.[Abstract/Free Full Text]

  31. Chen CP, Aplin JD. Placental extracellular matrix: Gene expression, deposition by placental fibroblasts and the effect of oxygen. Placenta 2003;24:316–325.[CrossRef][Medline]

  32. Donadoni C, Corti S, Locatelli F et al. Improvement of combined FISH and immunofluorescence to trace the fate of somatic stem cells after transplantation. J Histochem Cytochem 2004;52:1333–1339.[Abstract/Free Full Text]

  33. Zijlstra A, Mellor R, Panzarella G et al. A quantitative analysis of rate-limiting steps in the metastatic cascade using human-specific real-time polymerase chain reaction. Cancer Res 2002;62:7083–7092.[Abstract/Free Full Text]

  34. Elices MJ, Tsai V, Strahl D et al. Expression and functional significance of alternatively spliced CS1 fibronectin in rheumatoid arthritis microvasculature. J Clin Invest 1994;93:405–416.[Medline]

  35. Kilger G, Needham LA, Nielsen PJ et al. Differential regulation of alpha 4 integrin-dependent binding to domains 1 and 4 of vascular cell adhesion molecule-1. J Biol Chem 1995;270:5979–5984.[Abstract/Free Full Text]

  36. Steinborn A, Sohn C, Heger S et al. Labour-associated expression of intercellular adhesion molecule-1 (ICAM-1) in placental endothelial cells indicates participation of immunological processes in parturition. Placenta 1999;20:567–573.[CrossRef][Medline]

  37. Lakasing L, Campa JS, Parmar K et al. Normal expression of cell adhesion molecules in placentae from women with systemic lupus erythematosus and the antiphospholipid syndrome. Placenta 2000;21:142–149.[CrossRef][Medline]

  38. Salgado R, Benoy I, Weytjens R et al. Arterio-venous gradients of IL-6, plasma and serum VEGF and D-dimers in human cancer. Br J Cancer 2002;87:1437–1444.[CrossRef][Medline]

  39. Vuorela-Vepsalainen P, Alfthan H, Orpana A et al. Vascular endothelial growth factor is bound in amniotic fluid and maternal serum. Hum Reprod 1999;14:1346–1351.[Abstract/Free Full Text]

  40. Fadini GP, Sartore S, Schiavon M et al. Diabetes impairs progenitor cell mobilisation after hindlimb ischaemia-reperfusion injury in rats. Diabetologia 2006;49:3075–3084.[CrossRef][Medline]

  41. Liegeois A, Escourrou J, Ouvre E et al. Microchimerism: A stable state of low-ratio proliferation of allogeneic bone marrow. Transplant Proc 1977;9:273–276.[Medline]

  42. Cipriani P, Guiducci S, Miniati I et al. Impairment of endothelial cell differentiation from bone marrow-derived mesenchymal stem cells: New insight into the pathogenesis of systemic sclerosis. Arthritis Rheum 2007;56:1994–2004.[CrossRef][Medline]

  43. Frangioni JV, Hajjar RJ. In vivo tracking of stem cells for clinical trials in cardiovascular disease. Circulation 2004;110:3378–3383.[Abstract/Free Full Text]

  44. Zhou L, Yoshimura Y, Huang Y et al. Two independent pathways of maternal cell transmission to offspring: Through placenta during pregnancy and by breast-feeding after birth. Immunology 2000;101:570–580.[CrossRef][Medline]

  45. Le Blanc K, Ringden O. Immunobiology of human mesenchymal stem cells and future use in hematopoietic stem cell transplantation. Biol Blood Marrow Transplant 2005;11:321–334.[CrossRef][Medline]

  46. Nelson JL, Gillespie KM, Lambert NC et al. Maternal microchimerism in peripheral blood in type 1 diabetes and pancreatic islet beta cell microchimerism. Proc Natl Acad Sci U S A 2007;104:1637–1642.[Abstract/Free Full Text]

  47. Stevens AM, Hermes HM, Rutledge JC et al. Myocardial-tissue-specific phenotype of maternal microchimerism in neonatal lupus congenital heart block. Lancet 2003;362:1617–1623.[CrossRef][Medline]

  48. Khosrotehrani K, Guegan S, Fraitag S et al. Presence of chimeric maternally derived keratinocytes in cutaneous inflammatory diseases of children: The example of pityriasis lichenoides. J Invest Dermatol 2006;126:345–348.[CrossRef][Medline]

  49. Newsome PN, Johannessen I, Boyle S et al. Human cord blood-derived cells can differentiate into hepatocytes in the mouse liver with no evidence of cellular fusion. Gastroenterology 2003;124:1891–1900.[CrossRef][Medline]

  50. Liechty KW, MacKenzie TC, Shaaban AF et al. Human mesenchymal stem cells engraft and demonstrate site-specific differentiation after in utero transplantation in sheep. Nat Med 2000;6:1282–1286.[CrossRef][Medline]

  51. Krause DS, Theise ND, Collector MI et al. Multi-organ, multi-lineage engraftment by a single bone marrow-derived stem cell. Cell 2001;105:369–377.[CrossRef][Medline]

  52. Kucia M, Dawn B, Hunt G et al. Cells expressing early cardiac markers reside in the bone marrow and are mobilized into the peripheral blood after myocardial infarction. Circ Res 2004;95:1191–1199.[Abstract/Free Full Text]

  53. Orlic D, Kajstura J, Chimenti S et al. Mobilized bone marrow cells repair the infarcted heart, improving function and survival. Proc Natl Acad Sci U S A 2001;98:10344–10349.[Abstract/Free Full Text]

  54. Hattori K, Heissig B, Tashiro K et al. Plasma elevation of stromal cell-derived factor-1 induces mobilization of mature and immature hematopoietic progenitor and stem cells. Blood 2001;97:3354–3360.[Abstract/Free Full Text]

  55. Grunewald M, Avraham I, Dor Y et al. VEGF-induced adult neovascularization: Recruitment, retention, and role of accessory cells. Cell 2006;124:175–189.[CrossRef][Medline]

  56. Hattori K, Heissig B, Wu Y et al. Placental growth factor reconstitutes hematopoiesis by recruiting VEGFR1(+) stem cells from bone-marrow microenvironment. Nat Med 2002;8:841–849.[CrossRef][Medline]

  57. Podar K, Tai YT, Lin BK et al. Vascular endothelial growth factor-induced migration of multiple myeloma cells is associated with beta 1 integrin- and phosphatidylinositol 3-kinase-dependent PKC alpha activation. J Biol Chem 2002;277:7875–7881.[Abstract/Free Full Text]

  58. Jin H, Aiyer A, Su J et al. A homing mechanism for bone marrow-derived progenitor cell recruitment to the neovasculature. J Clin Invest 2006;116:652–662.[CrossRef][Medline]

  59. Hauzenberger D, Klominek J, Sundqvist KG. Functional specialization of fibronectin-binding beta 1-integrins in T lymphocyte migration. J Immunol 1994;153:960–971.[Abstract]

  60. de Fougerolles AR, Sprague AG, Nickerson-Nutter CL et al. Regulation of inflammation by collagen-binding integrins alpha1beta1 and alpha2beta1 in models of hypersensitivity and arthritis. J Clin Invest 2000;105:721–729.[Medline]

  61. Shapiro SD. Matrix metalloproteinase degradation of extracellular matrix: Biological consequences. Curr Opin Cell Biol 1998;10:602–608.[CrossRef][Medline]

  62. Son BR, Marquez-Curtis LA, Kucia M et al. Migration of bone marrow and cord blood mesenchymal stem cells in vitro is regulated by stromal-derived factor-1-CXCR4 and hepatocyte growth factor-c-met axes and involves matrix metalloproteinases. STEM CELLS 2006;24:1254–1264.[Abstract/Free Full Text]

  63. Zanjani ED, Almeida-Porada G, Livingston AG et al. Human bone marrow CD34- cells engraft in vivo and undergo multilineage expression that includes giving rise to CD34+ cells. Exp Hematol 1998;26:353–360.[Medline]

  64. Chan J, O'Donoghue K, de la Fuente J et al. Human fetal mesenchymal stem cells as vehicles for gene delivery. STEM CELLS 2005;23:93–102.[Abstract/Free Full Text]

  65. Garbuzova-Davis S, Gografe SJ, Sanberg CD et al. Maternal transplantation of human umbilical cord blood cells provides prenatal therapy in Sanfilippo type B mouse model. FASEB J 2006;20:485–487.[Abstract/Free Full Text]

  66. Shen X, Hong MS, Moss J et al. BIG1, a brefeldin A-inhibited guanine nucleotide-exchange protein, is required for correct glycosylation and function of integrin beta1. Proc Natl Acad Sci U S A 2007;104:1230–1235.[Abstract/Free Full Text]




This article has been cited by other articles:


Home page
 Hum ReprodHome page
C.-P. Chen, S.-H. Liu, J.-P. Huang, J. D. Aplin, Y.-H. Wu, P.-C. Chen, C.-S. Hu, C.-C. Ko, M.-Y. Lee, and C.-Y. Chen
Engraftment potential of human placenta-derived mesenchymal stem cells after in utero transplantation in rats
Hum. Reprod., January 1, 2009; 24(1): 154 - 165.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
2007-0406v1
2007-0406v2
26/2/550    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Reprints/Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chen, C.-P.
Right arrow Articles by Yang, Y.-C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chen, C.-P.
Right arrow Articles by Yang, Y.-C.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
STEM CELLS THE ONCOLOGIST CME ALPHAMED PRESS JOURNALS