Intra-bone marrow co-transplantation of umbilical cord mesenchymal stem cells promotes the engraftment of umbilical cord blood stem cells in iron overload NOD/SCID mice CURRENT STATUS: POSTED

Background: Iron overload aggravates the difficulty of umbilical cord blood stem cell engraftment and the survival undergoing transplantation. Mesenchymal stem cells (MSC) have been implicated in playing a significant role in HSC engraftment. This study aimed to determine the effect of intra-bone marrow (IBM) co-transplantation of umbilical cord blood mononuclear cells (UCB-MNC) and mesenchymal stem cells (UC-MSC) on the engraftment and hematopoietic recovery in an iron overload hematopoietic microenvironment. Methods: The iron overload model was established by dose-escalation intraperitoneal injection of iron dextran in NOD/SCID mice. Iron deposition in the bone marrow, heart, and liver was examined using H&E staining. Serum levels of ferritin and iron status in the liver were measured. The iron overload NOD/SCID mice were sublethally irradiated and divided into four groups for transplantation: (1) control group, (2) IBM MSC+ group: IBM injection of combined UCB-MNC/UC-MSC, (3) IBM group: IBM injection of only UCB-MNC, and (4) IV group: intravenous injection of UCB-MNC. Six weeks after transplantation, the human CD45 + cells in the bone marrow were analyzed by flow cytometry. The semi-quantitative analysis of vascular endothelial growth factor (VEGF-a), osteopontin (OPN), and stromal cell-derived factor-1a (SDF-1a) were examined by immunohistochemistry staining (IHC). Results: The survival rate and the percentages of human CD45 + cells in bone marrow were highest in the IBM MSC+ group with statistical significance. In addition, the levels of VEGF-a, OPN, and SDF-1a in bone marrow were all significantly higher in the IBM MSC+ group than the other groups. Conclusion: IBM co-transplantation of UC-MSC might improve the engraftment of UCB-MNC in iron overload NOD/SCID mice. The anesthetized i.p. injection hydrate. separated by centrifugation at × g at Serum levels of ferritin were determined by using ELISA kit according to the manufacturer’s instructions. The fresh liver was harvested and 100 mg of the liver per mouse was used for the measurement of ion levels using an iron assay kit according to the manufacturer’s instructions. DiR-labeled intravenous (IV) that homing HSC.


Background
Hematopoietic stem cell transplantation (HSCT) is currently the only method that can cure severethalassemia and myelodysplastic syndrome [1]. Umbilical cord blood-derived hematopoietic stem cells (UCB-HSC) are emerging as alternative sources of HSC for allogenic HSCT due to their weak immunogenicity and low requirement for HLA matching [2]. However, the clinical application of UCB-HSC is limited due to their small cell numbers and lower implantation rate than other sources of HSC [3]. The engraftment time might be significantly delayed, which increases the risk of infection and death of patients [4]. Repeated blood transfusions also cause tissue iron overload which destroys the hematopoietic microenvironment in the bone marrow and further dampens the efficacy of HSCT [5].
How to promote the implantation of UCB-HSC in the iron-overloaded hematopoietic microenvironment is a challenge to be conquered.
Umbilical cord mesenchymal stem cells (UC-MSC) have been shown to not only have multi-directional differentiation functions but also improve the hematopoietic microenvironment in the bone marrow [6,7]. This hematopoietic microenvironment is critical for stable and successful implantation of HSC [8].
We hypothesized that co-transplantation of UC-MSC might improve the iron overload-damaged hematopoietic niche to benefit the engraftment of HSCT. The low homing rate of traditional intravenous infusion is another reason causing low engraftment of UCB-HSC [9]. Attempts have been made to use intra-bone marrow (IBM) injection to reduce the loss of cord blood stem cells in nonhematopoietic organs and promote implantation [10]. Therefore, we established an iron-overload murine model to test the effects of IBM co-transplantation of UC-MSC on the engraftment of UCB-HSC.
In this study, we demonstrated that combining UC-MSC transplantation for the IBM HSCT greatly increases the expression of vascular endothelial growth factor (VEGF-a), osteopontin (OPN), and stromal cell-derived factor-1a (SDF-1a) in the iron overloaded bone marrow. Notably, the engraftment of UCB-HSC is significantly enhanced. Our results provide an experimental basis for the feasibility of IBM co-transplantation of MSC and HSC to overcome an iron overload compromised bone marrow hematopoietic niche.

Materials And Methods Animals
NOD/SCID mice (aged 6 weeks, body weight 20-22 g) were obtained from Beijing Cyagen Biosciences.
The mice were housed under specific-pathogen-free (SPF) condition in the Southern Hospital Animal Center. All animal care and experimental procedures were conducted according to ethical standards of animal use and were approved by the Institutional Committee of Animal Care and Use of Southern Medical University.

Iron overload mouse model
Fifty-eight female NOD/SCID mice were randomly grouped to control (n = 8), low dose (12.5 mg/ml, n = 42), and high dose (50 mg/ml, n = 8) groups. The control mice were intraperitoneally (i.p.) injected with normal saline (0.2 ml/injection). The mice of low and high dose groups were i.p. injected with iron dextran (0.2 ml/injection) at indicated concentration, once per every 3 days for 4 weeks. Identification of the iron overload model was assessed by histopathological examination (H&E staining) of iron deposits in the bone marrow, heart, and liver, as well as by measurement of serum ferritin and liver iron.

Measurement of serum ferritin and liver iron
After the establishment of the iron overload model for 4 weeks, the mice in each group were anesthetized by i.p. injection of 0.3% chloral hydrate. Blood collected from the inner canthus vein and the serum was then separated by centrifugation at 1000 × g at 4 °C for 20 minutes. Serum levels of ferritin were determined by using ELISA kit according to the manufacturer's instructions. The fresh liver was harvested and 100 mg of the liver per mouse was used for the measurement of ion levels using an iron assay kit according to the manufacturer's instructions.

Immunohistochemistry (IHC) staining and semi-quantitative scoring
Six weeks after the transplantation, femur samples were harvested, fixed with 10% formaldehyde, and embedded in paraffin. Serial sections (2-µm-thick) of each sample were processed immunohistochemically for the expression of OPN, VEGF-a, and SDF-1a using IHC kits according to the manufacturer's instruction. Briefly, tissue sections were deparaffinized, rehydrated, and blocked for endogenous hydrogen peroxide activity. Antigen retrieval was performed in EDTA buffer with a microwave oven. The primary and secondary antibodies were sequentially applied and incubated in a humidity box at 37 °C for 30 min and 40 min, respectively, followed by developing in DAB solution for 1-2 minutes. Sections were counterstained with hematoxylin solution for 4 min. The positive expression of OPN, VEGF-a, and SDF-1a were found in the cytoplasm and cell membrane and shown as brownish staining, which was semi-quantitatively assessed by scoring the proportion and intensity of stained cells according to published methods of T Padro et al. [11]. Briefly, the percentage of positive cells stained with each specific antibody within cellular areas was estimated according to a three-grade scale (1 = < 10% positive cells, 2 = 10-50% positive cells, 3 = > 50% positive cells). The degree of cellular staining intensity was defined in the following manner: 0 = faint or negative, 1 = light, 2 = moderate, 3 = intense. The score of each sample was obtained by observing 10 fields at high power (400×) magnification using an inverted microscope. The mean of cellular staining intensity was multiplied with the number obtained when estimating the percentage of positive cells according to the three grade scales. The results were expressed as arbitrary units (AU).

Statistical analysis
All experiments were performed at least three times. Differences were compared using student's ttests. The statistical analyses were performed with SPSS software (version 10.0).

Establishment and identification of iron-overload murine model
The iron overload murine model was established by intraperitoneal injection of escalating doses of iron dextran to NOD/SCID mice. The mortality rate of the high dose group reached 60% and no mortality was observed in the low dose group. Four weeks after iron dextran injection, dosedependent iron deposits could be easily observed in the bone marrow, heart, and liver by using H&E staining (Fig. 1A). Dose-dependent tissue destructions were also observed. Compared with the control group, the injection of iron dextran induced the increased lipid vacuoles in the bone marrow, disorientation of cardiac muscle fibers, and distortion of the hepatic sinus. Iron status in the liver

Survival of the recipient mice
Due to the high mortality rate of iron overload mice induced by a high dose of iron dextran, we therefore chose the low dose model to perform cell transplantation. After transplantation, all mice in the control group died due to bone marrow failure and internal bleeding (median survival time: 11 days). Six weeks after transplantation, the survival rates of control, IV, IBM, and IBM MSC + groups were 0% (0/8), 60% (6/10), 60% (6/10), and 83% (10/11), respectively. The survival rate of the IBM MSC + group was highest but the statistical difference to the IBM group (P = 0.091) or IV group (P = 0.087) was not reached. There was no statistical difference in the survival rate between IBM and IV group (P = 0.925).

Migration and distribution of UCB-MNC in vivo
To investigate the organ distribution of UCB-MNC after transplantation, DiR-labeled UCB-MNCs were injected and traced using in vivo fluorescence imaging. On the day of transplantation (Fig. 2A), both groups showed high fluorescence intensity in the injection sites, including the tail vein of the IV group and tibia of the IBM group. Six weeks after transplantation, the IV group showed the labeled UCB-MNCs were accumulated in the spleen but not in the bone marrow (Fig. 2B), whereas the IBM group showed accumulation of UCB-MNCs was mainly in the bone marrow.

Verification of engraftment efficacy
To confirm the efficacy of UCB-MNC engraftment, six weeks after transplantation, the bone marrow of 22 survived mice was harvested and examined by the content of human CD45 + cells. Flow cytometry results showed the % of human CD45 + cells in the bone marrow of MSC+, IBM, and IV groups was 9.7%±1.86 , 4.1 ± 1.6 , and 4.5 ± 1.26 , respectively ( Fig. 3A-C). The human CD45 + cells were significantly higher in the IBM MSC + group when compared to the other two groups (Fig. 3D). No significant difference was found between IBM and IV groups (Fig. 3D). The results indicated cotransplantation of UC-MSC could significantly enhance the engraftment rate of UCB-MNC in the iron overload microenvironment.
Expression of VEGF-a, OPN, and SDF-1a in the bone marrow Next, the expression of hematopoiesis cytokines in the bone marrow was investigated. After 6 weeks of transplantation, the femur of 22 survived mice was harvested and examined by IHC staining for the expression of VEGF-a, OPN, and SDF-1a (Fig. 4). The bone marrow of the control group showed increased lipid vacuoles and decreased hematopoietic cells. The expression of VEGF-a, OPN, and SDF-1a was very weak in the control group but notably higher in the IBM MSC + group. The semiquantitative results were shown in Table 1. The results indicated co-transplantation of UC-MSC might promote engraftment of UCB-MNC via upregulation of hematopoiesis cytokines in the hematopoietic microenvironment.   [10,12,13]. However, we found there was no difference between IBM and IV groups regarding the engraftment rate of UCB-MNC when using the iron overload model. The reason causing this discrepancy might be that the impairment of the hematopoietic niche by iron overload might affect the engraftment rate of HSCT conducted by different methods. Nevertheless, here we suggested even in an iron overloaded condition, IBM co-transplantation of MSCs might be a feasible approach to overcome the compromised hematopoietic microenvironment to achieve successful HSCT.

Discussion
The debilitating effect of iron overload has been indicated by the increasing clinical evidence that iron overload has a suppressive effect on hematopoiesis in myelodysplastic syndromes or anemia patients and that iron chelation therapy could improve this situation [14,15]. Iron overload markedly decreased the ratio and clonogenic function of murine HSC progenitors by the elevation of reactive oxygen species (ROS). Reducing the ROS level improved the long-term and multi-lineage engraftment of HSC after transplantation [16]. Iron accumulation in the bone marrow niche of thalassemia patients which might increase ROS production to impair the most primitive MSC pool as well as MSC stemness.
A weakened antioxidative response and diminished expression of bone marrow niche-associated genes in the MSC were found. This caused a functional impairment in MSC hematopoietic supportive capacity in vitro and in vivo [17]. Since the beneficial effects of MSC were hampered by iron overload, we, therefore, replenished the MSC pool by IBM co-transplantation of MSC with HSC. In agreement, this approach is successful in overcoming the iron overload-mediated impairment of hematopoiesis capacity. Whether ROS involves the beneficial effects is worth of further investigation.
MSCs are important in the regulation of hematopoietic stem cell niche due to their self-renewal and differentiation capacity into tissues of mesodermal origin. These MSC derivatives support hematopoiesis through release various molecules that play a crucial role in migration, homing, selfrenewal, proliferation, and differentiation of HSC [18]. For example, MSC could differentiate to osteoblasts which are closely related to the successful engraftment of stem cells [19,20]. MSC also secrete extracellular matrices such as aminoglycan and proteoglycan and maintain the hematopoietic microenvironment by releasing soluble factors for the interaction between cells [21,22]. Different subtypes of MSC interact with hematopoietic stem cells in specific areas around blood vessels. CD271 + and CD271 + /CD146 −/low MSCs are described as bone lining cells associated with long-term hematopoietic stem cells in the hypoxic region of bone marrow. CD146 + and CD271 + /CD146 + MSCs are located in the perivascular regions of the bone marrow sinus and related to the activation and proliferation of hematopoietic progenitor stem cells [23]. Specific subsets of MSC might be considered for the co-transplantation therapy of MSC and HSC to further improve the engraftment efficacy.
Mechanistically, our study found co-transplantation with UC-MSC significantly increased the expression of VEGF-a, OPN, and SDF-1a in the bone marrow, which might be closely related to the paracrine effect of MSC. VEGF-a, a critical factor for angiogenesis, also mediates the proliferation and migration of osteoblasts [24]. Previous studies found that the beneficial effects of VEGF-a on stem cells might be via inhibition of irradiation-induced apoptosis and facilitation of cell survival by paracrine effect [25,26]. Bone marrow morphogen osteopontin (OPN), which is abundantly present in the bone marrow extracellular matrix, can maintain hematopoietic stem cell reconstitution potential and the progenitor pool in the bone marrow [27,28]. SDF-1/CXCR4 axis is indispensable to the homing of HSC [29,30]. University. Consent to participate is not applicable.

Consent for publication
Not applicable.

Availability of data and materials
All the data and material have been presented in the main paper.

Competing interests
The authors declare that they have no competing interests.

Funding
This work was supported by the grant from the Guangdong Science and Technology Department of

Authors' contributions
We declare that all the listed authors have participated actively in the study and all meet the  H&E staining was used to examine iron deposition (arrow) in bone marrow, heart, and liver.