The Fate of Mesenchymal Stem Cells Transplanted ... - CyberLeninka

ARTICLE

doi:10.1016/j.ymthe.2004.02.022

The Fate of Mesenchymal Stem Cells Transplanted into Immunocompetent Neonatal Mice: Implications for Skeletal Gene Therapy via Stem Cells Christopher Niyibizi,1,* Sujing Wang,2 Zhibao Mi,2 and Paul D. Robbins2 1

Department of Orthopaedics and Rehabilitation, Pennsylvania State University College of Medicine, 500 University Drive, Hershey, PA 17033, USA 2 Department of Molecular Genetics and Biochemistry, University of Pittsburgh School of Medicine, Pittsburgh, PA 15213, USA

*Corresponding author. Fax: (717) 531-0672. E-mail: [email protected].

To explore the feasibility of skeletal gene and cell therapies, we transduced murine bone marrow-derived mesenchymal stem cells (MSCs) with a retrovirus carrying the enhanced green fluorescent protein and zeocin-resistance genes prior to transplantation into 2-day-old immunocompetent neonatal mice. Whole-body imaging of the recipient mice at 7 days postsystemic cell injection demonstrated a wide distribution of the cells in vivo. Twenty-five days posttransplantation, most of the infused cells were present in the lung as assessed by examination of the cells cultured from the lungs of the recipient mice. The cells persisted in lung and maintained a high level of gene expression and could be recovered from the recipient mice at 150 days after cell transplantation. A significant number of GFP-positive cells were also present in the bones of the recipient mice at 35 days post-cell transplantation. Recycling of the cells recovered from femurs of the recipient mice at 25 days posttransplantation by repeated injections into different neonatal mice resulted in the isolation of a clone of cells that was detected in bone and cartilage, but not in lung and liver after systemic injection. These data demonstrate that MSCs persist in immunocompetent neonatal mice, maintain a high level of gene expression, and may participate in skeletal growth and development of the recipient animals.

INTRODUCTION The use of stem cells for the treatment of a variety of diseases is currently an active area of investigation [1 – 5]. Research on the use of human embryonic stem cells is limited due to ethical concerns, but research on the use, biology, and engraftment characteristics of adult derived stem cells is an area of active investigation. There are many sources of adult stem cells; these include bone marrow, muscle, and fat [6 – 8]. Cells isolated from these sources have been shown to be multipotent in that they can give rise to different differentiated cell phenotypes, for example, osteoblasts, chondrocytes, and adipocytes [9 – 11]. The cells isolated from these tissue sources are of great interest because of their potential use as delivery vehicles for therapeutic genes for musculoskeletal repair and regeneration [12 – 15]. In addition, these cells have potential for use in the treatment of genetic skeletal disorders [16 – 21]. Clinical trials using whole marrow in children with severe forms of osteogenesis imperfecta (OI), a brittle bone disease resulting from type I collagen gene mutations, have in fact been attempted, and the results

MOLECULAR THERAPY Vol. 9, No. 6, June 2004 Copyright B The American Society of Gene Therapy 1525-0016/$30.00

showed that the cells hold promise for the treatment of OI [19,20]. These and other related studies have demonstrated that the whole marrow contains cells capable of engrafting and differentiating into osteoblasts in vivo [36,38]. Although numerous studies have been performed on stem cell engraftment using different animal models, there is still a lack of understanding regarding the engraftment characteristics of the cells, the number of cells that are detected in different tissues and organs of the cell recipients, the location of the cells in vivo, and the nature and phenotype of the cells that engraft in different tissues. In addition most studies have been performed using mature or SCID animal models [17,21,22]. Treatment of genetic diseases using adult stem cells will, however, require early intervention before manifestation of serious abnormalities. In addition, use of the stem cells for disease treatment will require a clear understanding of the engraftment characteristics of the stem cells in different tissues and differentiation of the cells into the tissue-specific cell phenotypes in vivo.

955

ARTICLE

In the present communication, we have used adherent murine bone marrow-derived mesenchymal stem cells to determine the fate of the cells infused into neonatal mice. In addition, we evaluated the effects of repeated systemic delivery of the cells recovered from bone on the homing and differentiation of the cells to different cell lineages in vivo.

RESULTS Murine Mesenchymal Stem Cells (MSCs) and Cell Surface Markers We characterized the adherent murine MSCs harvested from 8-week-old mice as stem cells by evaluating the cell surface marker expression and expression of alkaline phosphatase activity when exposed to BMP-2. Table 1 summarizes the results of the cell surface marker expression and alkaline phosphatase activity of the cells before and after BMP-2 treatment. Previous studies have shown that mesenchymal stem cells express CD13, CD90 (Thy-1), and CD105 (endoglin) cell surface markers [27,30]. The results showed that the murine adherent cells isolated here were CD13, CD45, CD90, and CD105 positive. The cells were, however, CD34 and CD117 negative (Table 1). Although the cells were CD34 negative, the presence of hematopoietic cell contamination cannot be ruled out. The cells, however, exhibited very high levels of alkaline phosphatase activity when stimulated by BMP-2. We used the cells as such for the subsequent experiments without further purification. Transduction of the Murine MSCs To aid in cell tracking, we transduced the adherent MSCs with a retrovirus carrying the enhanced green fluorescent protein (eGFP) cDNA. To increase the efficiency of cell transduction, we transduced the cells repeatedly with the retroviral vector. After four rounds of retroviral vector treatment, at 18-h intervals, the transduction efficiency was about 70%. After selection in a medium containing zeocin, 100% of the cells were GFP+ (Fig. 1A). Infusion of GFP-Positive Cells into Neonatal Mice We infused the GFP+ cells into neonatal mice via the superficial temporal vein. The temporal vein is visible in newborn mice until 4 days after birth; we therefore injected the cells into the neonates at 2 or 3 days after birth. At 5 days after cell infusion, whole-body imaging of the recipient mice demonstrated a wide expression of GFP within the neonates; however, most of the GFP was observed in the upper extremities (Fig. 1B). This may have been due to cell leakage during injection. By 7 days after cell infusion, we observed GFP expression throughout the whole body of the neonates, including the internal organs and the paws (Fig. 1C). Beyond 7 days

956

doi:10.1016/j.ymthe.2004.02.022

post-cell infusion, it was not possible to follow the distribution of GFP+ cells in vivo because of excessive hair that interfered with whole-body imaging. Isolation of the GFP-Positive Cells from the Recipient Mice To follow further the distribution of the GFP+ cells in vivo, we harvested tissues from the recipient mice and isolated the cells therein. Fig. 2 shows the morphological appearance of GFP-positive cells recovered from the lung at 7, 60, and 90 days and the liver at 7 days post-cell infusion. We recovered a large number of GFP+ cells from the lungs of the recipient mice at 7 days after cell infusion. This is consistent with previous findings indicating that the first tissue the cells encounter is the lung [37]. A larger number of cells could still be recovered from the lungs even after 150 days post-cell infusion. The morphological appearance of the GFP+ cells recovered from the lungs and liver of the recipient mice was distinct at each time point, suggesting that the cells may have undergone differentiation into different cell types in vivo. However, the cells recovered from the lungs and liver expressed very high levels of alkaline phosphatase (ALP) activity when stimulated with BMP-2 (Table 1). There were very few GFP+ cells detected in the bones harvested from the recipient mice before 14 days postcell infusion. At 25 days post-cell infusion, we observed larger numbers of GFP+ cells growing in and out of bone chips prepared from the femurs of the recipient mice (Fig. 3A). Surprisingly, we recovered the majority of the GFP+ cells from the right femurs. This observation may, however, be due to problems with cell harvesting and recovery from the left femurs. We expanded the GFP+ cells isolated from bone in culture in the presence of zeocin. The cells exhibited a spindle-shaped morphology and expressed high levels of ALP activity when stimulated by BMPs (Fig. 3B and Table 1). GFP+ cells were also present in the bone marrow of the recipient mice. The recovered cells expressed high levels of alkaline phosphatase activity when they were stimulated with BMP-2. Cell Recycling in Bone and Engraftment We reinjected the cells recovered from the recipient mice at 25 days postinfusion into neonatal mice and, after 35 days, recovered the cells and expanded them in culture. Reinjection of the day 35 cells into neonatal mice demonstrated that the cells were present in bone and cartilage but not in the lung and liver. These cells were smaller in size than the original cells, exhibited a spindleshaped morphology, and expressed very high levels of alkaline phosphatase activity (Figs. 4A and 4B and Table 1). Analysis of cell surface markers showed that the cells recovered from bone with a propensity to mig-

MOLECULAR THERAPY Vol. 9, No. 6, June 2004 Copyright B The American Society of Gene Therapy

doi:10.1016/j.ymthe.2004.02.022

ARTICLE

FIG. 1. Whole-body image of a neonatal mouse that was infused with GFP+ cells at 5 and 7 days after cell infusion. GFP+ cells were infused into neonatal mice at 2 days after birth. The mice were euthanized and examined under a fluorescence microscope. (A) MSCs transduced with DFG-eGFP-Zeor and selected for GFP+ cells; 100% of the cells are GFP positive. (B) Distribution of the GFP+ cells infused in a neonatal mouse at 5 days postinfusion and (C) at 7 days postinfusion. At 5 days most of the cells are located in the upper extremities and at 7 days, the GFP+ cells appear to be distributed throughout the entire mouse. (A) Original magnification 100, (B and C) original magnification 10.

rate to bone and cartilage expressed CD90, CD13, and high levels of CD105 cell surface markers (Fig. 4B). The data suggest that these cells exhibited stem cell characte-

ristics [27,31,34]. The cells also expressed low levels of CD45, presumably due to some endothelial cell contamination. When we reinjected these cells into neonatal

FIG. 2. Morphological appearance of GFP+ cells recovered from lungs at 7, 60, and 90 days and liver at 7 days post-cell infusion. The cells recovered from liver exhibit a morphological appearance distinct from those recovered from lung. Cells recovered from lung appear to possess a distinct morphological appearance at each different time point. It was not determined in the present study if the cells underwent differentiation. Original magnification 100.


957

ARTICLE

doi:10.1016/j.ymthe.2004.02.022

FIG. 3. Morphological appearance of the GFP+ cells recovered from bone. Femurs were harvested from the recipient mice at 25 days postinfusion, cut into small pieces, and placed in culture. (A) A large number of GFP+ cells are growing out of the bone chips recovered from the recipient mice. (B) GFP+ cells recovered from bone and selected in a medium supplemented with zeocin. The recovered cells exhibit a morphological appearance distinct from those recovered from the lung and liver. Original magnification 100.

mice, 14 days postinfusion, the cells were detected in bone and cartilage but not in the lungs or liver of the recipient mice. Fig. 5 shows the morphological appearance of the cells recovered from the cartilage of the recipient mice. The cells exhibited a chondrocyte-like morphology and expressed high levels of GFP (Figs. 5A and 5B). The GFP+ chondrocyte-like cells were present in the ribs and knee cartilages of the recipient mice. We recovered no chon-

drocyte-like cells from the cartilages of the mice transplanted with the nonrecycled cells. These data suggest that cell recycling in bone either induced cell commitment to osteoblastic and chondrocytic cell lineages or separated out a clone of cells with a distinct morphology that allowed them to migrate to bone and cartilage without being trapped in the lungs. The cells, however, once they differentiated into chondrocytes in vivo, no longer expressed the above cell surface markers. We analyzed the

TABLE 1: Cell surface markers and alkaline phosphatase activity (ALP) of the murine MSCs and the GFP+ cells recovered from various tissues CD13 +

GFP original cells Recovered cells from bone – day 35 GFP+ cells from cartilage GFP+ cells from lung – day 7 GFP+ cells from liver – day 7

CD34

+ + ND ND

ND ND

CD45

CD90

CD105

+ +

++ +

+++ +++

ND ND

ND ND

ND ND

CD117

ND ND

ALP activity BMP-2

ALP activity+BMP-2

+ +

+++ +++

+ +

+++ +++

The cell surface markers were identified by using specific antibodies against each cell surface marker. Alkaline phosphatase activity was determined before and after treatment with BMP-2 as described previously [13]. The original cells represent the cells isolated from mice femurs that were subsequently transduced with DFG-eGFP. +, low expression of a cell surface marker or low ALP activity; ++, moderate cell surface marker expression or ALP activity; +++, very high expression of cell surface marker or ALP activity; , absence of a cell surface marker expression or ALP activity.

958


ARTICLE

doi:10.1016/j.ymthe.2004.02.022

gen chains purified from bovine bone (lane 4). These data clearly demonstrate that the cells that were recycled in bone and injected into neonatal mice differentiate into chondrocytes in vivo and become part of the growing cartilage, suggesting that the cells may contribute to skeletal development.

DISCUSSION The data presented in this communication have clearly demonstrated that the murine mesenchymal stem cells infused in neonatal mice lodged into most of the mouse tissues within 5 – 7 days after cell infusion. Most of the transplanted cells by day 7 were in the lung and liver. At later time points the cells were also detected in the bones of the recipient mice. The neonatal mice that were transplanted with the GFP+ cells did not undergo any marrow ablation or any other conditioning prior to cell transplan-

FIG. 4. Morphological appearance of the GFP+ cells recovered from bone at 35 days postinfusion and after two rounds of recycling in bone. (A) The cells exhibit a spindle-shaped morphology and are smaller than the cells recovered from the lung. (B) All the cells were CD105 positive. These cells were found in bone and cartilage only when they were reinjected into neonatal mice. Original magnification 100.

GFP+ cells isolated from cartilages for collagen synthesis to confirm that these cells were indeed chondrocytes. Immunofluorescence Localization and Western Blotting for Type II Collagen To demonstrate that the GFP+ cells harvested from cartilages were chondrocytes, we determined synthesis of type II collagen by the cells recovered from the cartilages of recipient mice by immunofluorescence localization and Western blotting. Immunofluorescence localization for type II collagen demonstrated clear staining for type II collagen by the GFP+ cells (Figs. 6A and 6B). We confirmed the immunofluorescence data by Western blotting of the collagens synthesized by the GFP+ cells recovered from cartilages of the recipient mice (Fig. 6C). The antibody immunoreacted with type II collagen purified from bovine cartilage (Fig. 6C, lane 1, from left), a protein band with similar migration on SDS – PAGE from the collagens synthesized by the GFP+ cells (lane 2), and an identical protein band from the GFP+ cells cultured in alginate (lane 3). The antibody used in the present studies did not show any immunoreactivities with type I colla-


FIG. 5. Morphological appearance of the GFP+ cells recovered from cartilage. Rib and knee joint cartilage was harvested from the recipient mince, minced into small pieces, and placed in tissue culture in a medium supplemented with zeocin. (A) Morphological appearance of the cells observed under phase contrast and (B) identical cells observed under fluorescence. The cells exhibit a chondrocytic morphological appearance and express high levels of GFP. Original magnification 100.

959

ARTICLE

doi:10.1016/j.ymthe.2004.02.022

FIG. 6. Collagen synthesis by the GFP+ cells recovered from cartilages of recipient mice. (A) GFP+ cells. (B) Immunofluorescence localization for type II collagen in GFP+ cells using a secondary antibody conjugated with Cy3. (C) Immunoblot for type II collagen synthesized by the GFP+ cells from cartilage. The antibody immunoreacts with type II collagen standard, a protein band with migration identical to that of the standard type II collagen synthesized by the GFP+ cells recovered from the cartilage. The antibody does not cross-react with type I collagen chains (I). The data demonstrate that the GFP+ cells recovered from cartilages of recipient mice are chondrocytes.

tation; however, the cells persisted in the tissues of the recipient mice. One hundred fifty days after cell transplantation, the GFP+ cells were recovered from the lungs of the recipient mice; however, there were no GFP+ cells recovered from bone at this time point. These data suggest that either the immune system in the neonate is not fully developed and therefore the mice are able to accept allogeneic cells or the cells themselves are immunoprivileged. Most of the transplanted cells remained lodged in the lung. It is not clear whether the cells were trapped there for such an extended period or they established residence in the lung because they differentiated into lung-specific cells. Previous studies have shown that mesenchymal stem cells that engraft in the lung exhibited epitheliumlike morphology, suggesting that the GFP+ cells that persist in the lung may differentiate into lung epithelial cells [31]. In the present study we did not attempt to determine the phenotype of the cells that persisted in lung. Some of the GFP+ cells recovered from the lungs of the recipient mice, however, exhibited a morphological appearance distinct from that of the original transplanted cells. These data suggest that some of the transplanted cells that persist in the lung may differentiate into lungspecific cells or other yet unidentified cell types. The surprising finding in the present communication is that after three rounds of cell recycling in bone, a clone of cells was identified that, after systemic injection, was

960

detected in bone and cartilage of the recipient mice but not in the lung. These data can be explained in several different ways. First, the cells used in the present study did not represent a single population. However, in this mixture of cells, large numbers of cells were CD13, CD90, CD105, and CD45 positive. These cell surface markers, with the exception of CD45, though not specific, are believed to be characteristic of mesenchymal stem cells [27,30]. The cells were CD117 and CD34 negative; however, the presence of hematopoietic cells cannot be ruled out. Therefore since the cells did not comprise one cell population, the lung may have acted as a filter. Nevertheless the cells that were infused in neonatal mice were enriched in MSCs. Cells that were too large to pass through the lung capillaries may have remained lodged in the lung, other cells perhaps because of different shape and size were able to escape the lung entrapment and ended up in the bone. Studies establishing models of breast cancer metastasis to bone have shown that the tumor cells rapidly take residence in the lungs because this organ is the first to be encountered by the cells after intravenous injection. The lodging of the cells in lungs is partly due to the trapping of the cells by the small capillary lumen that restricts cell migration. The tumor cells are also believed to attach specifically to the endothelium of pulmonary capillaries to form tumor colonies [33]. Thus in the present study, most of the cells are


doi:10.1016/j.ymthe.2004.02.022

trapped in the lungs and the few cells that escape entrapment by the lungs end up in the bones and perhaps other tissues. The cells that migrate to bone may be morphologically distinct from those that take up residence in the lungs. Therefore, this process may selectively favor migration of cells with specific morphological characteristics from lung to bone and cartilage. Second, the bone microenvironment may play a role in influencing the cells that migrate to bone by supplying microenvironmental cues that imbibe the recycled cells with a predisposition to return to bone. This latter hypothesis is supported by the observation that repeated passing of breast carcinoma cells in bone resulted in a subclone of cells with a unique predilection to metastasize to bone when injected via the tail vein [28, 32]. It remains to be established if indeed the same mechanism applies to the murine normal cells that have the propensity to migrate to bone and to differentiate toward osteoblastic and chondrogenic cell lineages. The model system used in the present study for MSC engraftment is neonatal mice. These mice are rapidly growing and, therefore, the transplanted MSCs may be differentiating into chondrocytes that contribute to the growth and development of the mice. Indeed, a large number of GFP+ cells were recovered from the cartilages of the recipient mice and these cells were shown to have differentiated into chondrocytes in vivo. The data suggest that the cells may participate in the growth and development of the growing mice. These observations are supported by the data of Sands and colleagues, who have shown that marrow-derived endothelial progenitor cells engraft into vasculature of nonablated newborn mice but not in adult mice [35]. Their hypothesis is that early neonatal development, which represents a period of accelerated tissue growth, provides a milieu for vessel growth and development. The present findings that a larger number of transplanted cells in neonatal mice differentiated into chondrocytes in vivo may be the result of accelerated growth that allows transplanted cells to participate in the growth and development of the recipient mice. In summary the present data offer an opportunity to investigate further the effects of mesenchymal stem cell recycling in host tissues and to understand the role played by the microenvironment in inducing cell commitment. In addition, recycled cells may provide information as to the nature of the cells and the type of cell surface markers and molecules that they secrete that allow them to migrate to bone and to differentiate into chondrocytes in vivo. This information will be useful for cell targeting to specific tissues.

MATERIALS

AND

METHODS

Cell isolation. Murine bone marrow-derived MSCs were isolated from 8week-old mice (B6C3Fe a/a) using a modified procedure described previously [23]. Briefly, femurs were harvested from 8-week old mice, stripped of periosteum, and cut at both ends. The bones were placed in petri dishes containing DMEM supplemented with 20% FBS, 1% penicillin/streptomy-


ARTICLE

cin (v/v), and 50 Ag/ml ascorbic acid. After 5 days in culture, the bones were removed and the medium in the petri dishes was removed and replaced with DMEM supplemented with the above factors. At confluence, the cells were trypsinized and replated in T-75 flasks. The nonadherent cells were removed after 4 days in culture and the adherent cells were maintained in culture in DMEM supplemented with 20% FBS, 10 ng/ml TGF-h1, and 50 Ag/ml ascorbic acid. The media were replaced every 3 days and after three media changes, when the cells reached confluence, they were trypsinized and passaged. The cells were maintained in DMEM supplemented with 20% FBS, 50 Ag/ml ascorbic acid without TGF-h1. Characterization of the isolated cells. The cells were characterized by evaluating the expression of the cell surface markers and expression of differentiated cell phenotypes under specific conditions. The cell surface markers evaluated were CD13, CD34, CD45, CD90, CD105, and CD117. The rat anti-mouse monoclonal antibodies against CD13, CD34, CD45, CD90, and CD117 conjugated to either FITC or Cy5 were purchased from BD Pharmingen (San Diego, CA, USA). The CD105 rat anti-mouse monoclonal antibody was purchased from Research Diagnostics, Inc. (Flanders, NJ, USA). The anti-mouse IgG whole molecule conjugated with Cy3 was purchased from Sigma Biochemicals (St. Louis, MO, USA). The cell surface marker immunolocalization was performed using the manufacturer’s protocol. Briefly, for cell surface marker immunocytochemistry, the MSCs harvested from mice were plated in eight-well chamber slides at a concentration of 1 104 cells per well, in DMEM, 10% FBS, supplemented with 50 Ag/ml ascorbic acid. The cells were then washed in PBS and fixed in methanol/acetic acid (3:1) for 10 min. The fixed cells were then washed and blocked with 1% BSA in PBS for 30 min. The cell surface marker antibodies conjugated to either FITC or Cy5 were added to the cells according to their respective cell surface markers at 5 Ag/ml in PBS and incubated for 1 h at room temperature. For CD105, the monoclonal antibody was added to the cells at a concentration of 1:10 in PBS. The secondary antibody, anti-mouse IgG, was added to the cells at a concentration of 1:200 in PBS after washing as above. The cells were then washed and examined under a fluorescence microscope. Transduction of the cells with a retrovirus carrying eGFP cDNA. For cell tracking in vivo, the cells were transduced with a high-titer retrovirus carrying the enhanced green fluorescent protein and zeocin-resistance genes (eGFP-Zeor). The recombinant vector used in this study originates from the retroviral vector DFG [24,25]. The eGFP gene was inserted into NcoI and NotI sites of DFG. The IRES-Zeo cassette was created by inserting a NotI and BamHI fragment. The recombinant eGFP retroviral vector was generated by transfection of DFG-eGFP-Zeo plasmid into a BOSC-derived packaging cell line (Phoenix), using the calcium phosphate method. Briefly, 3 Ag of DFG-eGFP-Zeo was transfected into Phoenix cells in 10cm cell culture dishes and the conditioned media were changed 5 h after the transfection. Viral supernatants were collected 72 h after transfection and used to infect the bone marrow-derived mesenchymal stem cells. For cell transduction, the cells were plated in six-well plates in DMEM until 60% confluency. The cells were then treated with 1 ml of a high-titer DFG-eGFP retrovirus in the presence of 10% FBS. After 24 h, the medium containing the retroviral vector was replaced with new medium and the retroviral vector was again added as above. After two more rounds of medium changes with addition of viral vector, the medium was discarded and new medium without the viral vector was added and the cells were incubated for a further 24 h in DMEM supplemented with 10% FBS. After 24 h, the medium was replaced with new medium supplemented with 25 Ag/ ml zeocin to select for the transduced cells. The cells were maintained in DMEM, 10% FBS, and 25 Ag/ml zeocin with medium changes every 3 days. The selected cells were expanded in culture in the presence of zeocin and were then used for the transplantation into neonatal mice. Transplantation of MSCs into neonatal mice. The GFP+ cells were trypsinized from the T-75 flasks and washed in sterile PBS and filtered through a nylon mesh. The cells were then suspended in PBS solution at 5 106 cells/ml. One hundred microliters of the cell suspension containing 5 104 filtered cells was then drawn up in a 0.5-ml syringe

961

ARTICLE

equipped with a 30-gauge needle and the cells were injected into 2-dayold mice via the superficial temporal vein [26]. Five mice were analyzed at each time point of 5, 7, 25, 35, 60, 90, and 150 days after cell infusion. Tracking of the GFP-positive cells injected into neonatal mice. The neonatal mice that received the cells were examined at 5, 7, 25, 35, 60, 90, and 150 days after cell injection. The distribution of the GFP+ cells in the recipient neonatal mice was determined by whole-body imaging of the recipient mice at 5 and 7 days and by the isolation of the cells from the tissues and organs of the recipient mice thereafter. Whole-body imaging of the GFP-positive cells in recipient mice. At 5 and 7 days of cell infusion, the recipient mice were sacrificed by exposure to isoflurane and whole bodies were then examined under a dissecting stereomicroscope equipped with fluorescence detection (Olympus FSZ 12). Neonatal mice that received saline injection alone were used as negative controls.Recovery of the GFP-positive cells from the tissues of the recipient mice. Beyond 7 days of cell infusion, the recipient mice developed excessive hair and it was not possible to perform whole-body imaging to follow the fate of the cells in vivo. To confirm the whole-body imaging results and to follow further the fate of the cells in vivo beyond 7 days, the lung, liver, heart, spleen, kidney, bone, and cartilage were harvested from the recipient mice at the time points indicated above and the cells were isolated from these tissues and examined for GFP+ cells. For cell isolation, the harvested tissues were minced into small pieces and treated with trypsin in Hanks’ solution at 2 mg/ml for 2 h. This was followed by collagenase treatment, 2 mg/ml in Hanks’ solution (type 1; Sigma) for 2 h. After enzyme digestion, the digests were filtered through a nylon membrane, and the cells were plated in petri dishes. For bone, marrow was flushed from femurs and tibia, centrifuged, and plated in petri dishes. After marrow flush, the bones were cut into small pieces and placed in petri dishes as tissue explants. This same procedure was used for chondrocyte isolation from the cartilages of the recipient mice. The cells were then cultured in DMEM, 20% FBS, supplemented with 50 Ag/ml ascorbic acid, 1% v/v penicillin/streptomycin, and 25 Ag/ml zeocin to select for the GFP+ cells. The isolated cells were characterized for the expression of the cell surface markers and expression of alkaline phosphatase activity in the presence of BMP-2. Recycling of MSCs in neonatal mouse bones. The cells recovered from the recipient mice femurs at 25 days after cell transplantation were maintained in DMEM supplemented with 25 Ag/ml zeocin to select for the GFP+ cells. The GFP+ cells were reinjected into the 2-day-old neonatal mice via the temporal vein as described above. The injected mice were sacrificed at day 35 after cell infusion. Femurs were harvested from the recipient mice, minced into small pieces, and cultured in DMEM, 20% FBS. The cells that grew out of the bone chips were cultured in DMEM supplemented with zeocin to select for the GFP+ cells. The recovered GFP+ cells were reinjected into the 2-day old mice and the mice were sacrificed after 14 days of cell infusion. Femurs, cartilage from knee joints, and ribs were harvested, minced, and cultured as explants. The recovered cells were characterized in terms of cell surface markers, osteogenic potential, and collagen synthesis. Collagens synthesized by the GFP+ cells recovered from cartilage. To determine whether the GFP+ cells recovered from the cartilaginous tissues of the recipient mice were chondrocytes, the collagens synthesized by the recovered cells were determined. Briefly, the GFP+ cells were plated in sixwell plates and at confluency, the DMEM containing 20% FBS was replaced with serum-free medium supplemented with 50 Ag/ml ascorbic acid, 50 Ag/ ml h-aminopropionitrile, and, in some wells, 10 ACi of tritiated proline. After 24 h, the medium and the cell layers were combined and adjusted to 0.5 M in acetic acid. Pepsin was added to the samples at 100 Ag/ml and the samples were incubated in pepsin for 2 h at 4jC. After 2 h, the pepsinresistant proteins were precipitated with 3 M NaCl for 24 h at 4jC followed by centrifugation. The recovered pellets were analyzed by SDS – PAGE followed by autoradiography. In some experiments, the collagens were blotted onto polyvinylidene difluoride membrane and the membrane was

962

doi:10.1016/j.ymthe.2004.02.022

probed with the type II collagen-specific antibody (Biodesign International, Kennebunk, ME, USA) using the methods described previously [29]. Immunofluorescence localization. Synthesis of type II collagen by the GFP+ cells recovered from the cartilages was also determined by immunofluorescence localization using a mouse type II collagen monoclonal antibody (Biodesign International). Briefly, GFP+ MSCs recovered from the cartilages of the recipient mice were plated in four-chamber culture slides until confluence. The cells were then fixed in methanol:acetic acid at (3:1 v/v) for 10 min. The fixed cells were washed in several changes of PBS and then blocked by addition of 1% BSA in PBS. After several washes in PBS, a mouse monoclonal antibody specific for type II collagen was added to the cells at a concentration of 2 Ag/ml. The immunofluorescence visualization was achieved by treatment of the cells with the goat antimouse IgG conjugated to Cy3 (Sigma). The slides were observed under a fluorescence microscope.

ACKNOWLEDGMENTS This work was supported in part by NIH Grant R01 AR049688 and a grant from the Children’s Brittle Bone Foundation to C.N. The authors thank Dr. Simon Watkins and Dr. Glenn Papworth of University of Pittsburgh for the help with the whole-body imaging. RECEIVED FOR PUBLICATION DECEMBER 20, 2003; ACCEPTED FEBRUARY 29, 2004.

REFERENCES 1. Lovell-Badge, R. (2001). The future for stem cell research. Nature 414: 88 – 91. 2. Bianco, P., Riminucci, M., Gronthos, S., Robey, P. G. (2003). Bone marrow stromal stem cells: nature, biology, and potential applications. Stem Cells 19: 180 – 192. 3. Robert, K. (2002). Stem cells on the way to restorative medicine. Immunol. Lett. 83: 1 – 12. 4. Park, K. I., et al. (2002). Global gene and cell replacement strategies via stem cells. Gene Ther. 9: 613 – 624. 5. Niyibizi, C., Wallach, C. J., Mi, Z., and Robbins, P. D. (2002). Approaches for skeletal gene therapy. Crit. Rev. Eukaryotic Gene Expression 12: 163 – 173. 6. Zuk, P. A., et al. (2002). Human adipose tissue is a source of multipotent stem cells. Mol. Biol. Cell 13: 4279 – 4295. 7. Lee, J. Y., et al. (2000). Clonal isolation of muscle-derived cells capable of enhancing muscle regeneration and bone healing. J. Cell Biol. 150: 1085 – 1100. 8. Jiang, Y., et al. (2002). Pluripotency of mesenchymal stem cells derived from adult marrow. Nature 418: 41 – 49. 9. Pittenger, M. F., et al. (1999). Multilineage potential of adult human mesenchymal stem cells. Science 284: 143 – 147. 10. Dennis, J. E., Merriam, A., Awadallah, A., Yoo, J. U., Johnstone, B., and Caplan, A. I. (1999). A quadripotential mesenchymal progenitor cell isolated from the marrow of an adult mouse. J. Bone Miner Res. 14: 700 – 709. 11. Jiang, Y., Vaessen, B., Lenvik, T., Blackstad, M., Reyes, M., and Verfaillie, C. M. (2000). Multipotent progenitor cells can be isolated from postnatal murine bone marrow, muscle, and brain. Exp. Hematol. 30: 896 – 904. 12. Lieberman, J. R., et al. (1999). The effect of regional gene therapy with bone morphogenetic protein-2-producing bone-marrow cells on the repair of segmental femoral defects in rats. J. Bone Joint Surg. Am. 81: 905 – 917. 13. Niyibizi, C., Smith, P., Mi, Z., Phillips, C. L., and Robbins, P. (2001). Transfer of proalpha2(I) cDNA into cells of a murine model of human osteogenesis imperfecta restores synthesis of type I collagen comprised of alpha1(I) and alpha2(I) heterotrimers in vitro and in vivo. J. Cell. Biochem. 83: 84 – 91. 14. Gelse, K., von der Mark, K., Aigner, T., Park, J., and Schneider, H. (2003). Articular cartilage repair by gene therapy using growth factor-producing mesenchymal cells. Arthritis Rheum. 48: 430 – 441. 15. Van Damme, A., Vanden Driessche, T., Collen, D., Chuah, M. K. (2002). Bone marrow stromal cells as targets for gene therapy. Curr. Gene Ther. 2: 195 – 209. 16. Prockop, D. J. (1997). Marrow stromal cells as stem cells for nonhematopoietic tissues. Science 276: 71 – 74. 17. Pereira, R.F. et al. (1998). Marrow stromal cells as a source of progenitor cells for nonhematopoietic tissues in transgenic mice with a phenotype of osteogenesis imperfecta. 18. Oyama, M., et al. (1999). Retrovirally transduced bone marrow stromal cells isolated from a mouse model of human osteogenesis imperfecta (oim) persist in bone and retain the ability to form cartilage and bone after extended passaging. Gene Ther. 6: 321 – 329.


ARTICLE

doi:10.1016/j.ymthe.2004.02.022

19. Horwitz, E. M., et al. (2002). Isolated allergenic bone marrow-derived mesenchymal cells engraft and stimulate growth in children with osteogenesis imperfecta: implications for cell therapy of bone. Proc. Natl. Acad. Sci. USA 99: 8932 – 8937. 20. Horwitz, E. M., et al. (1999). Transplantability and therapeutic effects of bone marrow-derived mesenchymal cells in children with osteogenesis imperfecta. Nat. Med. 5: 309 – 313. 21. Pereira, R. F., et al. (1995). Cultured adherent cells from marrow can serve as longlasting precursor cells for bone, cartilage and lung in irradiated mice. Proc. Natl. Acad. Sci. USA 92: 4857 – 4861. 22. Anklesaria, P., et al. (1987). Engraftment of a clonal bone marrow stromal cell line in vivo stimulates hematopoietic recovery from total body irradiation. Proc. Natl. Acad. Sci. USA 84: 7681 – 7685. 23. Balk, M. L., et al. (1997). Effect of rhBMP-2 on the osteogenic potential of bone marrow stromal cells from an osteogenesis imperfecta mouse (oim). Bone 21: 7 – 15. 24. Zitvogel, L., et al. (1994). Construction and characterization of retroviral vectors expressing biologically active human interleukin-12. Hum. Gene Ther. 5: 1493 – 14506. 25. Gambotto, A., et al. (2000). Immunogenicity of enhanced green fluorescent protein (EGFP) in BALB/c mice: identification of an H2-Kd-restricted CTL epitope. Gene Ther. 7: 2036 – 2040. 26. Sands, M. S., Barker, J. E. (1999). Percutaneous intravenous injection in neonatal mice. Lab. Anim. Sci. 49: 328 – 330. 27. Jones, E. A., et al. (2002). Isolation and characterization of bone marrow multipotential mesenchymal progenitor cells. Arthritis Rheum. 46: 3349 – 33460. 28. Peyruchaud, O., Winding, B., Pecheur, I., Serre, C. M., Delmas, P., and Clezardin, P. (2001). Early detection of bone metastases in a murine model using fluorescent human breast cancer cells: application to the use of the bisphosphonate zoledronic acid in the treatment of osteolytic lesions. J. Bone Miner. Res. 16: 2027 – 2034. 29. Niyibizi, C., Sagarriga Visconti, C., Gibson, G., and Kavalkovich, K. (1996). Identifica-


30.

31.

32.

33.

34.

35.

36.

37.

38.

tion and immunolocalization of type X collagen at the ligament – bone interface. Biochem. Biophys. Res. Commun. 222: 584 – 589. De Ugarte, D. A., et al. (2003). Differential expression of stem cell mobilization-associated molecules on multi-lineage cells from adipose tissue and bone marrow. Immunol. Lett. 289: 267 – 2670. Ortiz, L. A., et al. (2003). Mesenchymal stem cell engraftment in lung is enhanced in response to bleomycin exposure and ameliorates its fibrotic effects. Proc. Natl. Acad. Sci. USA 100: 8407 – 8411. Oku, N., Koike, C., Sugawara, M., Tsukada, H., Irimura, T., and Okada, S. (1994). Positron emission tomography analysis of metastatic tumor cell trafficking. Cancer Res. 54: 2573 – 2576. Al-Mehdi, A. B., Tozawa, K., Fisher, A. B., Shientag, L., Lee, A., and Muschel, R. J. (2000). Intravascular origin of metastasis from the proliferation of endothelium-attached tumor cells: a new model for metastasis. Nat. Med. 6: 100 – 102. Majumdar, M. K., Banks, V., Peluso, D. P., and Morris, E. A. (2000). Isolation, characterization, and chondrogenic potential of human bone marrow-derived multipotential stromal cells. J. Cell. Physiol. 185: 98 – 106. Young, P. P., Hofling, A. A., and Sands, M. S. (2002). VEGF increases engraftment of bone marrow-derived endothelial progenitor cells (EPCs) into vasculature of newborn murine recipients. Proc. Natl. Acad. Sci. USA 99: 11951 – 11956. Hou, Z., et al. (1999). Osteoblast-specific gene expression after transplantation of marrow cells: implications for skeletal gene therapy. Proc. Natl. Acad. Sci. USA 96: 7294 – 7299. Gao, J., Dennis, J. E., Muzic, R. F., Lundberg, M., and Caplan, A. I. (2001). The dynamic in vivo distribution of bone marrow-derived mesenchymal stem cells after infusion. Cells Tissues Organs 169: 12 – 20. Nilsson, S. K., et al. (1999). Cells capable of bone production engraft from whole bone marrow transplants in nonablated mice. J. Exp. Med. 189: 729 – 734.

963