Biodistribution of locally or systemically transplanted osteoblast-like cells

Article focus

To ensure safety of the cell-based therapy for bone regeneration, we examined biodistribution of locally or systemically transplanted osteoblast-like cells.

Key messages

Locally transplanted osteoblast-like cells resided within the transplanted site and were not detected in other organs at two weeks after transplantation.

Intravenously infused cells were cleared from various organ tissues within three days of transplantation, without showing any adverse effects in the transplanted rats.

Strengths and limitations

Strengh: This is the first validation of biodistribution of locally or systemically transplanted osteoblast-like cells. We clearly demonstrated the safety of cell therapies using culture-expanded osteoblast-like cells.

Limitation: Transplanted cells were heterogeneous and the cellular localisation has not been analysed in vascular or other organs including heart, pancreas, and intestine.

Introduction

Distraction osteogenesis (DO) is a well-established therapeutic technique used for limb lengthening and bone regeneration with an external device following osteotomy.^1-5 Long-term treatment, however, is essential to extend the bone by DO, which may lead to a risk of adjacent joint contractures and pin track infection. In order to minimise these DO-related complications, several trials have been carried out such as ultrasonic stimulation,^6,7 electronic stimulation,⁸ hyperbaric oxygen exposure,⁹ transplantation of fresh bone marrow (BM) cells or administration of growth hormone or cytokines to shorten the treatment period and enhance osteogenesis in DO.^10-13 We have developed a novel DO technique to shorten the consolidation period by performing transplantation of culture-expanded osteoblast-like cells together with platelet rich plasma (PRP), thrombin and calcium.^14-16 Our cell therapy accelerated bone regeneration and decreased the associated adverse effects, which indicated the utility of combining osteoblast-like cells and biomaterials for DO.

Systemic adverse effects regarding BM cell transplantation still remain controversial. Several studies have reported the safety of transplanting both fresh and cultured BM cells,^17,18 On the other hand, some investigators have found that autologous transplantation of BM cells developed unexpected adverse effects, including unfavourable angiogenesis and even death.^19,20 It is important to assess the systemic biodistribution of transplanted BM cells in the recipient to ensure secure use of BM cells in regeneration medicine.

In this study, local and systemic biodistribution of transplanted osteoblast-like cells was analysed in a rat model. For local biodistribution of transplanted cells, we used a segmental bone defect model instead of a DO model to simplify the experimental procedure. Donor BM cells were harvested from green fluorescent protein (GFP)-transgenic (GFP-Tg) rats in order to trace the transplanted cells in the recipient rat. Flow cytometric and histological analyses were performed to assess the biodistribution of donor cells after transplantation.

Materials and Methods

Transplantation of culture-expanded osteoblastic cells

A segmental bone defect rat model wasprepared as described previously.²¹ Briefly, rats were anesthetised with an intraperitoneal infusion of Nembutal (30 mg/kg to 40 mg/kg: diluted with saline solution) (Dainippon Pharmaceutical Co, Ltd, Osaka, Japan) and a mini external fixator-distractor (Nagoya Screw Manufacturing Co., Ltd, Nagoya, Japan) was secured to the rat’s right femoral bone using four threaded pins. The mid-third of the bone shaft was osteotomised. After fixation of the distraction device, a bone defect 5 mm in length was prepared by lengthening the femur and collagen sponge (Koken Co., Ltd., Tokyo Japan) containing 5 x 10⁶ of culture BM cells, which were put on the site of the bone defect (Fig. 1a). The animals were allowed to move freely in cages after recovery from anesthesia. For intravenously infusion experiments, five million osteoblast-like cells were systemically infused into the normal SD rats via puncture of the femoral vein (Fig. 1b). Infusion of PBS without osteoblast-like cells was used as a negative control. To analyse donor-cell contribution, recipient-rat tissues were periodically assessed at one and two weeks after transplantation in bone defect rats, and at two hours, one and three days after intravenous infusion.

Figs. 1a - 1b

Figure 1a – Schematic diagrams of the experimental procedure of a rat segmental bone defect model. BM cells harvested from GFP-Tg rat were cultured in a medium for osteoblastic differentiation (OS medium) for three weeks. Three passaged (P3) 5 x 10⁶ cells were embedded in a collagen sponge and transplanted into bone defect rats. Figure 1b – The experimental procedure for intravenous infusion of osteoblast-like cells into the healthy SD rats. BM cells harvested from GFP-Tg rat were cultured in an OS medium for three weeks. Three passaged (P3) 5 x 10⁶ cells were systemically infused into the SD rats via femoral vein.

Results

Transplanted collagen scaffolds still remained and a majority of GFP-positive osteoblast-like cells survived in the scaffolds at two weeks after donor cell transplantation (Figs 2a and 2b).

Figs. 2a - 2b

Figure 2a – Histological analyses of the femur from a rat bone defect model. Left panel indicates the wild-type rat as a negative control (n = 4), middle panel indicates the GFP-Tg rat as a positive control (n = 1), and right panel indicates the bone defect rat transplanted with osteoblast-like cells after two weeks of transplantation (Tp-2w, n = 4). Images (x 20) are merged with GFP field and bright field. Figure 2b – Low magnification images of the femur from bone defect rats transplanted with osteoblast-like cells (x2). Histological analysis of the bone defect rats was performed after two weeks of transplantation (Tp-2w). The scaffolds contain osteoblast-like cells were located on the defected site. The images on the right are merged with GFP field and bright field (BF).

In order to analyse several cell populations, the gate of flowcytometric analysis was set at a wide area. As shown in Fig. 3a, none of GFP-positive cells was detected in the contralateral BM of the operated femur by flow cytometric analysis. Not only BM, but also other tissues including PB, lung, liver, spleen and brain did not have GFP-positive cells at one or two weeks after transplantation (Fig. 3b). Consistent with flow cytometric analysis, GFP-positive cells were not observed in lung, liver, spleen and brain by histological analysis (data not shown). In addition, there were no histological abnormalities including vascular occlusion, cell invasion or cell division of transplanted cells in all organs tested.

Figs. 3a - 3b

Figure 3a – Representative flow cytometric plots showing bone marrow (BM) from the wild-type (WT) rats and GFP-Tg rats (GFP-Tg), and BM and peripheral blood (PB) from bone defect rats that were transplanted with osteoblast-like cells after two weeks of transplantation (Tp-2w). Numbers shown in the figure indicate the frequency of GFP-positive cells. Figure 3b – Graph showing percentage of GFP-positive cells in BM, PB, lung, liver, spleen and brain at one and two weeks after transplantation. WT rat was used as a negative control, whereas BM of GFP-Tg rat was used as a positive control (n = 3 or 4 each). Horizontal lines represent means.

Histological examinations of various organs demonstrated that none of GFP-positive osteoblast-like cells were detected in BM, PB, lung, liver, spleen and brain both at one and two weeks after intravenous infusion (Fig. 4, and data not shown). By the clearance kinetic experiments, we detected approximately 0.03% and 0.004% of GFP-positive cells in the lung at two hours and one day after intravenous infusion, respectively. However, these cells were cleared after three days of intravenous infusion (Fig. 5a). GFP-positive cells were rarely detected in liver (0.001%), PB (0.001%), BM (< 0.001%) and spleen (< 0.001%) at two hours after infusion. These cells were no longer detectable after one day of intravenous infusion (Fig. 5b). It should be noted that no GFP-positive cells were detected in the brain. Histological examinations of the lung and liver revealed that only a few GFP-positive cells were seen in the lung at two hours after intravenous infusion, but they completely disappeared at three days after infusion (Fig. 5c). No vascular occlusion was detected in lung, liver, spleen and brain for three days post-infusion (data not shown). Quantitative real-time PCR analysis demonstrated that the GFP gene was not detected in any tissue of a recipient rat, including an ovary (Fig. 5d). Because qPCR assays require more than 100 cells to detect the target gene, this result indicates that GFP-positive cells residing in several tissues at two hours after transplantation were at less than 100.

Fig. 4

Representative macroscopic images of the GFP expression in the lung (left) and liver (right) from the recipient rats at one week after intravenous infusion (i.v.-1w) of GFP-positive cells (upper panel), and from the GFP-Tg rats as a positive control (lower panel). Nuclei were stained with DAPI.

Figs. 5a - 5d

Figure 5a – Graph showing percentage of GFP-positive cells in lung at two hours, one and three days after intravenous infusion (n = 4 or 5 each). Values are presented as standard error of the mean (SEM). Figure 5b – Graph showing percentage of GFP-positive cells in BM, PB, liver, spleen and brain at two hours, one and three days after intravenous infusion (n = 4 or 5 each). Values are presented as standard error of the mean (SEM). Figure 5c – Representative macroscopic images of the GFP expression in lung (left) and liver (right) at two hours (i.v.-2h) and three days (i.v.-3d) after intravenous infusion. PBS without osteoblast-like cells was used as a negative control, whereas GFP-Tg rat was used as a positive control. Nuclei were stained with DAPI. Figure 5d – Bar graph showing quantification of the GFP gene by real-time PCR. Copy numbers of GFP were normal to that of GAPDH. PBS was used as a negative control (NC), whereas GFP-Tg rat was used as a positive control (PC). Values are presented as standard error of the mean (SEM).

Discussion

This study presents evidence for the clinical safety of cultured osteoblast-like cells derived from BM cells in regenerative medicine. To our knowledge, this is the first report to show the systemic biodistribution of transplanted osteoblast-like cells in the recipient using the rat model of a segmental bone defect. We found that the majority of GFP-positive cells were localised at the transplanted site, without migrating to PB or other tissues. Transplanted cells survived in the transplanted site because the healing process occurs in the bone defect area rich in cytokines and growth factors, where the transplanted collagen scaffold is located. In addition, the atelocollagen sponge has a structure which facilitates the ready supply of nutrients to the cells. Unfortunately, we did not demonstrate the final fate of these transplanted cells; however, based on the in vitro data, we assume they differentiate into mature osteoblasts. Transplanted cells, which were differentiated into osteoblast-like cells for three weeks, might be suitable to reside in the bone, although the atelocollagen scaffold could play a role in preventing diffusion of embedded cells. Even though we did not quantify the viability of osteoblasts derived from transplanted cells, we presume that these differentiated, rather than un-differentiated cells, have more capability to differentiate into the osteoblasts and calcify.

We demonstrated that a small number of transplanted GFP-positive cells was detected in the lung, liver, PB, BM and spleen at two hours after intravenous injection. Previous reports have shown that cultured BM cells infused intravenously were particularly retained in reticular tissues.^22,23 Contrary to these reports, Eggenhofer et al²⁴ demonstrated the short-term survival of infused cultured BM cells and a lack of distribution of viable BM cells beyond the lung. We showed that transplanted osteoblast-like cells were completely cleared within three days after systemic infusion, even in the reticular tissues. We assume that undifferentiated mesenchymal stem cells (MSCs) used in previous studies tend to move to other organs, rather than differentiated cells. Since the differentiated cells were transplanted, the number of MSCs contained in transplanted cells was less than others and this might provide favourable results. Notably, no GFP- positive cells were detected in genital tissues such as those within the ovary, indicating that transplanted osteoblast-like cells may not affect offspring. Our results suggest the safety of transplantation of osteoblast-like cells in bone regenerative medicine.

There are several limitations to our study. First, osteoblast-like cells obtained from BM after three weeks of cultivation are heterogeneous, without sorting MSCs. Second, we did not analyse the cell localisation in the vascular or other organs including the heart, pancreas and intestine. Further studies for characterisation of transplanted cells and detailed examinations of other important organs are needed to prove the safety of cell therapies using cultured BM cells.

In the current study, we showed that osteoblast-like cells locally transplanted into bone defects were not detected in other tissues at two weeks after transplantation and that systemically infused osteoblast-like cells mainly resided in the lung at one day of intravenous infusion, but were cleared within three days of infusion. Consistent with other previous reports concerning cultured BM cell transplantation studies, donor cells do not migrate to other tissues or organs and do not show any negative effects to the recipients, although there are some differences in duration of clearance.^18,25,26