Allogeneic bone marrow transplantation restores liver function in Fah-knockout mice

Allogeneic bone marrow transplantation restores liver function in Fah-knockout mice

Experimental Hematology 2008;36:1507–1513 Allogeneic bone marrow transplantation restores liver function in Fah-knockout mice Elke Eggenhofera,*, Fel...

490KB Sizes 3 Downloads 59 Views

Experimental Hematology 2008;36:1507–1513

Allogeneic bone marrow transplantation restores liver function in Fah-knockout mice Elke Eggenhofera,*, Felix C. Poppa,*, Philipp Rennera, Pczemyslaw Slowika, Annette Neuwingera, Pompiliu Pisoa, Edward K. Geisslera, Hans J. Schlitta, and Marc H. Dahlkea,b a Department of Surgery, University of Regensburg, Regensburg, Germany; bConcord Repatriation General Hospital, Sydney, Australia

(Received 4 April 2008; revised 27 May 2008; accepted 28 May 2008)

Objective. In murine models, transplantation of wild-type bone marrow cells (BMC) can counterbalance genetic liver defects by fusion between transplanted marrow cells and resident hepatocytes. This phenomenon, however, is of no immediate clinical use because all syngeneic BMC harbor the same underlying genetic defect. Materials and Methods. Describing the fusion between transplanted allogeneic BMC and resident hepatocytes in a murine model of hereditary tyrosinemia type I (fumarylacetoacetate hydrolase [Fah] knockout mouse), we transplanted BMC from fully allogeneic BALB/c donors into Fah–/– recipients after lethal total body irradiation. Results. Following hematopoietic reconstitution, recipients remained healthy without pharmacological support (withdrawal of 2-2-nitro-4-fluoromethylbenzoyl-1,3-cyclohexanedione [NTBC]). Metabolic serum parameters improved nearly to wild-type levels. Livers of recipient animals contained up to 10% functional hepatocytes that stained positive for wild-type Fah, as well as both donor and recipient major histocompatibility complex. Flow cytometry confirmed this coexpression on a single cell level. Application of T-cell–depleted bone marrow reduced onset of early graft-vs-host disease. Conclusions. We introduce the observation that allogeneic bone marrow transplantation can lead to stable cell fusion of BMC with recipient hepatocytes and restored liver function in a model of otherwise lethal genetic liver disease. Thus, in principle, allogeneic cell fusion can be a possible management of hereditary liver diseases. Long-term immunological properties of fusion cells have to be further investigated. Ó 2008 ISEH - Society for Hematology and Stem Cells. Published by Elsevier Inc.

Allogeneic liver transplantation is the only definitive therapy for end-stage hepatic disorders when conservative management fails. Unfortunately, solid organ transplantation is still associated with high morbidity, mortality, and cost, and the supply of donor livers is limited. Therefore, novel therapies are warranted, as many patients still die on waiting lists. In selected metabolic disorders of the liver with inherited protein deficiencies, e.g., Wilson’s disease, hemochromatosis, or tyrosinemia, successful transplantation of

*Drs. Eggenhofer and Popp contributed equally to the study. Offprint requests to: Marc H. Dahlke, M.D., Ph.D., Department of Surgery, University of Regensburg, Franz Josef Strauss Allee 12, 93042 Regensburg, Germany; E-mail: [email protected]

a limited number of fully functional hepatocytes is sufficient to restore liver function. In these cases, cell transplantation, rather than solid organ transplantation, might be sufficient for successful treatment. To define a novel cell-based approach for metabolic liver diseases, the fumarylacetoacetate hydrolase (Fah)–deficient mouse, an animal model of fatal hereditary tyrosinemia type 1, has been studied extensively [1]. Fah knockout mice suffer from progressive liver failure and renal tubular damage unless treated with 2-(2-nitro-4-tifluoro-methylbenzyol)1,3-cyclohexanedione (NTBC, nitisone, Orphadin; Swedish Orphan, Stockholm, Sweden) [2]. Cellular fusion between bone marrow–derived cells and host hepatocytes, leading to genetic complementation, is the principal mechanism of liver regeneration in the Fah mouse after wild-type

0301-472X/08 $–see front matter. Copyright Ó 2008 ISEH - Society for Hematology and Stem Cells. Published by Elsevier Inc. doi: 10.1016/j.exphem.2008.05.010

1508

E. Eggenhofer et al./ Experimental Hematology 2008;36:1507–1513

bone marrow transplantation (BMT) [3–5]. Myelomonocytic cells, in particular, were identified as the most important cell source for hepatocyte fusion [6]. All previous work has focused on wild-type cells being transplanted into animals of the same genetic background. In a practical therapeutic setting, however, syngeneic bone marrow cells (BMC) without the genetic defect are not available, and autologous BMC would be of no benefit because these cells harbor the same underlying genetic defect. Therefore, allogeneic transplantation strategies remain the only possibility to implement a clinically applicable strategy for liverdirected cellular therapy. In the present work, we demonstrate that fully major histocompatibility complex (MHC)–mismatched BMC can serve as hepatocyte fusion partners in the Fah mouse model. Following allogeneic BMT, recipient animals remained healthy without NTBC. Metabolic parameters of these animals significantly improved due to the emergence of functional hepatocytes expressing donor and recipient MHC molecules, as well as Fah. Our findings raise the possibility that allogeneic cell fusion can be a new strategy for clinical cellular therapy aimed at correcting genetic liver enzyme defects.

Material and methods Mouse strains and animal husbandry Eight to 16-week-old (approximately 20–25 g) FAH mice (haplotype H2 Kb, a gift from M. Grompe) served as recipient animals and fully allogeneic BALB/c mice (haplotype H2 Kd, Charles River Laboratories, Wilmington, MA, USA) were used as donors. All FAH mutant animals were continuously treated with NTBC (nitisone, Orphadin, Swedish Orphan) in the drinking water at a concentration of 7.5 mg/L, providing an approximate dose of 1 mg/kg body weight per day. All experiments were carried out under general anesthesia and in accordance with regional authorities. Metabolic serum parameters were quantified on an Advia 1650 clinical analyzer (Bayer, Leverkusen, Germany). BMC harvest and transplantation BMC were harvested by flushing long bones of BALB/c mice with Dulbecco’s modified Eagle’s medium (Gibco, Invitrogen, Carlsbad, CA, USA) using a 22-gauge needle. T-cell depletion was achieved by anti-CD90 magnetic beads using magnetic cell separation (Miltenyi Biotec, Bergisch Gladbach, Germany). CD90negative cells were washed with phosphate-buffered saline and counted. Cell concentrations were adjusted for transplantation into the tail vein, with an injection volume of 200 mL per mouse (for injected cell numbers see Table 1). FAH mutant recipient mice were lethally irradiated with a total dose of 1000 cGy total body irradiation. Cell injection into the tail vein was carried out 4 to 6 hours later using a 27-gauge needle. FAH mutant animals were kept on NTBC for 3 weeks after BMT. To induce hepatocyte selection, NTBC was then withheld until mice dropped 20% of their initial body weight. At this point, NTBC application was restarted until the body weight recovered. NTBC cycles were repeated until animals were able to survive without NTBC.

Table 1. Summary of animal groups

Group 1 2 3

Transplanted cells Allogeneicc BMC Allogeneicc BMC CD90Syngeneicd BMC

Average no. of applied cellsa 40  106 20  106 15  106

No. of surviving animalsb/total animals 6/11 7/11 3/3

a

Applied cell number ranges 625% in the allogeneic experiments depending on the donor cell preparation yield. b Surviving means survival of 2-(2-nitro-4-tifluoromethylbenzyol)-1,3cyclohexanedione withdrawal for at least 2 months. c Donor animal BALB/c (H2 Kd). d Donor animal SvJ129 wild-type (H2 Kb).

Hepatocyte isolation and flow cytometry To obtain single cell suspensions of hepatocytes, livers were treated by a four-step perfusion protocol as we and others have described previously [7]. The resulting cell suspension was separated by gradient centrifugation with 35% isotonic Percoll (Amersham Biosciences, Piscataway, NJ, USA) at 700g for 20 minutes at 20 C to select for viable single cells. For flow cytometric analysis, cells were washed with phosphate-buffered saline (PBS) and resuspended in PBS. Cell suspensions were incubated with monoclonal antibodies at 4 C for 20 minutes with combinations of saturating amounts of purified, fluorescein isothiocyanate (FITC)–conjugated or phycoerythrinconjugated antibodies (Becton Dickinson, Heidelberg, Germany). Cells were analyzed on a FACS Calibur flow cytometer (Becton Dickinson). Data analysis of living cells (propidium iodide–negative) was carried out with FlowJo V7.1.3 software (TreeStar, Inc., Ashland, OR, USA). Histology and immunohistology Liver tissues were washed in PBS to remove blood cells and immediately frozen in liquid nitrogen. Five-micrometer sections were stained with FITC-conjugated anti-H2 Kb (clone AF6-88.5; Becton Dickinson), anti-H2 Kd (clone SF1-1.1; Becton Dickinson), and polyclonal anti-FAH antibody (kindly provided by M. Grompe). Antibodies were diluted in PBS and applied at concentrations of 1:50 for anti-H2 Kb-FITC and 1:600 for anti-H2 KdFITC at room temperature for 60 minutes. Anti-FAH was used at a 1:3000 dilution, with an overnight incubation at 4 C. To colorimetrically visualize the FITC-conjugated antibodies, horseradish peroxidase–conjugated anti-FITC antibody (Dianova, Hamburg, Germany) was applied at a 1:100 dilution. Secondary antibody for anti-FAH was horseradish-conjugated goat anti-rabbit IgG (Dianova) at a dilution of 1:200; color development was achieved by adding AEC solution. This final incubation was at room temperature for 60 minutes.

Results Fah-negative 129 SvJ mice survive with stable weight after fully allogeneic BMT from BALB/c donors and NTBC withdrawal To determine whether allogeneic BMT can augment liver function of Fah knockout mice, we transplanted allogeneic

E. Eggenhofer et al./ Experimental Hematology 2008;36:1507–1513

1509

Figure 1. Body weight measurements from one representative group 2 animal. 2-(2-nitro-4-tifluoro-methylbenzyol)-1,3-cyclohexanedione (NTBC) was deprived at day 21 and day 47 and was added whenever body weight was !80% of the initial body weight (dotted line). Gray areas indicate when NTBC was given, white areas show NTBC withdrawal periods.

BALB/c (haplotype H2 Kd, Fahþ/þ) bone marrow into Fah–/–/129SvJ (haplotype H2 Kb) recipients. Pressure for selection was controlled by repeated cycles of NTBC withdrawal. In the first group of experiments, Fah–/– recipients were transplanted with 40  106 unfractionated bone marrow cells (group 1, Table 1). In the second group of experiments, bone marrow grafts were depleted of T cells and 20  106 cells were injected intravenously (group 2). All transplantations were performed 4 hours after a lethal dose of total body irradiation (1000 cGy). Animals in group 3 were transplanted with wild-type BMC from SvJ129 Fahþ/þ mice and served as a control group. At day 21 after transplantation, NTBC was discontinued. Because Fah–/–-negative hepatocytes accumulate toxic metabolites, this allowed for the selection of liver repopulating cells as described previously [8]. Six of 11 mice (55%) survived NTBC withdrawal in group 1, 7 of 11 (64%) survived discontinuation of the drug in group 2, and all animals in the control group survived. Survivors showed healthy

behavior, including a stable body weight at a level O80% of the initial weight for more than 2 months (Fig. 1). Animals were monitored daily for clinical signs of graft-vshost disease (GVHD) and were sacrificed between days 100 and 150 after BMT. Surviving recipients show stable multilineage hematopoietic chimerism after allogeneic BMC transplantation Donor-type hematopoietic chimerism was determined 4 weeks after transplantation and at the end of the observation period. The fraction of donor-derived hematopoietic cells was determined by flow cytometry (anti-H2 Kd) in the blood, spleen, lymph node, thymus, and bone marrow samples of recipient mice. Untreated Fah–/– animals served as controls to exclude the possibility of unspecific antibody binding. In the early phase after transplantation (days 20– 30, Table 2), O80% of donor-derived cells could be demonstrated in the majority of mice from groups 1 and 2.

Table 2. Detection of donor derived cells after allogeneic bone marrow transplantation Group 1

Group 2

Day 20–30a Animal 1 2 3 4 5 6

Day 20–30a

Day 100–150a

Blood (%)

Animal

Blood (%)

Blood (%)

Spleen (%)

Lymph node (%)

99.9 99.3 99.2 89.0 81.0 84.8

1 2 3 4 5 6 7

94.9 96.0 87.6 42.6 63.4 55.1 68.9

ND ND ND 82.0 94.3 91.2 79.3

ND ND ND 85.0 93.2 88.1 81.2

ND ND ND 83.0 87.8 89.3 80.4

BM 5 bone marrow, ND 5 not determined. a Time after transplantation.

Thymus (%) ND ND ND 95.2 ND ND ND

BM (%) ND ND ND 82.2 66.4 60.2 55.9

1510

E. Eggenhofer et al./ Experimental Hematology 2008;36:1507–1513

All animals of group 2 that were tested (n 5 4) and presented with early mixed chimerism also showed a stable chimerism at a later point in time (days 100–150). Thus, most Fah–/– recipients of allogeneic BALB/c bone marrow developed stable hematopoietic chimerism after BMT. Some of the long-term chimeric hosts suffered from GVHD throughout the observation period [9,10]. The degree of systemic GVHD was assessed using a scoring system assessing four clinical parameters: posture (hunching), activity, fur texture, and skin integrity [11]. Weight loss was not assessed because it is affected by NTBC withdrawal in this present model. GVHD monitoring revealed higher scores among group 1 animals during the first 3 weeks after transplantation. Thereafter, no differences were observed between the GVHD scores in group 1 and group 2 animals (Fig. 2). Functional Fah-positive cells express donor and recipient MHC Long-term survivors were sacrificed and analyzed between days 100 and 150. At harvest, livers were individually processed, sectioned, and stained for Fah immunoreactivity. Liver cryosections of all NTBC-free survivors displayed functional Fah-positive nodules, indicating expansive repopulation. Each section was systematically scanned and Fah-positive nodules, as well as the cell numbers in each nodule, were counted with ImageJ 1.37v software (National Institutes of Health, Bethesda, MD, USA). Fah–/– recipients transplanted with wild-type BMC served for comparison. Both animals from group 1 and group 2 presented with the same number of nodules per liver section (group 1 5 6.1 6 4.1, group 2 5 7.8 6 2.7) as the syngeneic control group 3 (7.9 6 2.9, Fig. 3A). While there were no Figure 3. Quantification of fumarylacetoacetate hydrolase (Fah)–positive nodules. (A) Numbers of nodules from group 1 and 2 animals were not significantly different (p O 0.05) to the syngeneic control (group 3). (B) Nodules of group 1 and 2 animals are significantly smaller than in group 3 (Student’s two-tailed t-test: *p 5 0.0281, **p 5 0.0210, n 5 4).

Figure 2. Graft-vs-host disease (GVHD) monitoring of group 1 and group 2 animals 2 months after transplantation. Degree of systemic GVHD was assessed by scoring four clinical parameters: posture (hunching), activity, fur texture and skin integrity compared to the method of Min et al. [11] with modifications as described in the text. Standard deviation (SD) is shown.

distinct differences in the number of nodules, the number of cells per nodule was significantly higher in the control group than in group 1 and 2 (Fig. 3B), reflecting a limited extent of immunological selectivity in the allogeneic transplant groups. To further characterize the Fah-positive cells, serial cryosections were taken from recipient livers. Staining with anti-Fah antibody, as well as H2 Kd and anti-H2 Kb antibodies demonstrated that hepatocytes within these nodules were positive for all three markers and appeared as microscopically normal (Fig. 4). Furthermore, single hepatocytes obtained through perfusion digest were analyzed by flow cytometry. Using antibodies specifically recognizing the BALB/c and SvJ129 MHC class I molecules (H2 Kd and H2 Kb), this confirmed that single hybrid cells were positive for both donor and recipient MHC antigen. On average,

E. Eggenhofer et al./ Experimental Hematology 2008;36:1507–1513

1511

Figure 4. Liver histology of serial sections of a fumarylacetoacetate hydrolase (Fah)–/– mouse liver at 120 days after allogeneic bone marrow transplantation. (A) Repopulating FAH-positive cells (brown staining) are organized in a cell cluster. (B) Overall staining of liver cells with antibody against recipient major histocompatibility class (MHC) class I antigen H2 Kb. (C) Antibody against donor MHC class I antigen shows colocalization of staining, indicating that the FAH expression is derived from fusion cells. Dashed line shows corresponding area of FAH-positive cells in the serial sections (original magnification 100). Inserts show single cells of serial sections in higher magnification (original magnification 400).

20% 6 11% of hepatocytes of group 1 and 2 animals stained positive for both antibodies (Fig. 5). Metabolic parameters of transplanted animals Serum biochemical markers were assessed for both allogeneic transplant groups (urea, bilirubin, and albumin) and were compared with levels in wild-type and Fah–/– animals. Recipients of nondepleted marrow (group 1) showed a substantial improvement in all tested parameters to levels near that found in wild-type mice (Fig. 6). Group 2 animals, receiving T-cell–depleted marrow demonstrated a significant improvement in albumin and urea (Fig. 6A and C), but serum bilirubin levels resembled those of Fah–/– animals (Fig. 6B).

Discussion It has previously been established that wild-type BMC can migrate to the livers of Fah–/– mice, secure their survival, and restore liver function by cellular fusion with resident hepatocytes. However, an introduction of this concept into clinical cell transplantation requires application of allogeneic BMC. In the present work, we demonstrate that allogeneic BMC from BALB/c mice can repopulate Fah–/– recipients, which survive without pharmacological support by NTBC. Recipient metabolic serum parameters improve to near wild-type levels and recipient livers contain fully functional Fahþ/þ hepatocytes, which stain positive for both donor and recipient MHC. BMT is long established for treatment of hematopoietic malignancies [12]. Most nonhematopoietic tissues, however, cannot normally be ‘‘repopulated’’ by transplantation of bone marrow [13]. Recently, Petersen, Theise, and others have described hepatocytic cells derived from rodent bone marrow after BMT [14,15]. The nature of these cells was most convincingly established experimentally in the

Fah–/– mouse, a widely studied model for hepatic repopulation therapies, initially introduced in 1993 by Markus Grompe and coworkers [1]. Later, Lagasse et al. [8] demonstrated that bone marrow cells in the Fah mouse model not only restore liver function, but also achieve complete reconstitution of the liver. At this stage, the finding was interpreted as the transdifferentiation of highly potent bone marrow–derived stem cells. In later reports, several laboratories have provided evidence that cellular fusion, rather than transdifferentiation, is responsible for the therapeutic effect of liver repopulation from bone marrow in Fah knockout mouse [4,5], as well as in other murine models [3]. Additionally, it was found that predominantly myelomonocytic donor cells fuse with resident hepatocytes with or without liver damage [6,16–18]. Despite the success story for cellular fusion in Fah knockout mouse, it must be noted that fusion events are rare overall and expansion of functional hepatocytes is strongly dependent on a special and constant selective pressure in the surrounding hepatic microenvironment [19,20]. In all previous BMC transplantation studies aimed at liver repopulation, syngeneic donor cells (without the genetic defect) were used for grafting. A critical aspect of our study was that fully MHC mismatched BMC were used to mimic a potential clinical setting where cell therapy could be applied. Four weeks after transplantation we detected O85% donor-derived blood cells (H2 Kd-positive cells) in recipient blood samples. This finding was confirmed on day 150 after donor BMC transplantation for blood, as well as in other peripheral lymphoid tissues. Because stable mixed hematopoietic chimerism is defined as the presence of O5% donor cells during an observation period of at least 100 days [21], we can state that stable mixed bone marrow chimeras were created in our study. In accordance with previous syngeneic experiments, chimeric mice survived NTBC withdrawal and remained healthy

1512

E. Eggenhofer et al./ Experimental Hematology 2008;36:1507–1513

Figure 5. Isolation and major histocompatibility class (MHC) class I expression of hepatocytes. (A) Dot plot of single hepatocytes. (B) Analysis of hepatocytes with anti-H2 Kd-phycoerythrin-conjugated antibodies (gate R1). (C) H2 Kb-Fluorescein isothiocyanate staining of the H2 Kd-positive hepatocytes (gate R2), gray area indicates isotype control.

[5,8]. Animals transplanted with CD90-depleted BMC (group 2) presented with a higher percentage of survivors (64% vs 55%). Nevertheless, significant differences in the onset of GVHD occurred during the first 3 weeks after transplantation only. We then further analyzed the metabolic serum parameters on day 150 and observed an

improvement for both mice receiving whole BMC and lymphocyte-depleted BMC. This is comparable to findings in recipients of syngeneic marrow [6,8]. Immunohistochemical assessment of the livers of transplanted animals on day 150 revealed up to 20% Fah-positive hepatocytes. Further characterization of these cells by serial sections and flow cytometry of single hepatocytes demonstrated expression of both donor- and recipient-type MHC molecules. This observation suggests that Fah-positive recipient hepatocytes result from the same type of cellular fusion as cells in the syngeneic transplantation model [4,5]. Interestingly, we observed that similar numbers of Fah-positive nodules developed for both allogeneic groups, but the size of the nodules in the allogeneic groups was significantly smaller as compared to controls. This observation suggests that clonal expansion of fusion cells from allogeneic parent cells is comparatively decreased. Alternatively, this finding may be explained if there are more fusion events in the syngeneic setting, which results in aggregation of single nodules appearing as larger clusters. Whether this difference is due to impaired proliferation or the number of initial fusion events remains to be elucidated. It is also possible that an ongoing anti-donor response against donor MHC would play a role in reducing nodule size. Nevertheless, even with smaller FAH-positive allogeneic nodules, animals remain healthy without pharmacological support. In conclusion, our present findings confirm the potential of cellular fusion between BMC and resident hepatocytes as an interesting tool to introduce new genetic content into genetically impaired hepatocytes. Use of allogeneic cells in the setting of the Fah knockout mouse opens up the general possibility of introducing a clinical therapy based on the fusion of resident hepatocytes and allogeneic cells from a healthy donor. Further work is necessary to determine if the use of low-level immunosuppression can increase the frequency of fusion events. Concerning the ongoing development of murine models, the emergence of doublepositive fusion cells also opens up the possibility to isolate and characterize fused cells. Because data describing the tumor-initiating potential of fusion cells is still unclear, proper isolation of fused cells will also provide additional information on this important issue. When finally a greater understanding of the molecular nature of the ‘‘nuclear transfer’’ behind cellular fusion (and the associated dangers) can be achieved, these cells will have a promising future.

Acknowledgments The authors have received Fah knockout mice, the Fah antibody and kind advice from Markus Grompe and coworkers. Expert technical assistance and animal care was carried out by Irina Kucuk and Friederike Thamm. This work was generously supported by Novartis Pharma, Transplantation and Immunology, Nu¨rnberg

E. Eggenhofer et al./ Experimental Hematology 2008;36:1507–1513

1513

Figure 6. Analysis of metabolic serum parameters. Wild-type SvJ129 and untreated fumarylacetoacetate hydrolase (Fah)–/– mice (2 weeks off NTBC) served as controls (n 5 4 each). (A) Albumin (g/L). (B) Bilirubin (mg/dL). (C) Urea (mg/dL); *p ! 0.05, relative to untreated Fah–/– mice.

and a program grant from the Deutsche Forschungsgemeinschaft to M.H.D.

References 1. Grompe M, al-Dhalimy M, Finegold M, et al. Loss of fumarylacetoacetate hydrolase is responsible for the neonatal hepatic dysfunction phenotype of lethal albino mice. Genes Dev. 1993;7:2298–2307. 2. Grompe M, Lindstedt S, al-Dhalimy M, et al. Pharmacological correction of neonatal lethal hepatic dysfunction in a murine model of hereditary tyrosinaemia type I. Nat Genet. 1995;10:453–460. 3. Alvarez-Dolado M, Pardal R, Garcia-Verdugo JM, et al. Fusion of bone-marrow-derived cells with Purkinje neurons, cardiomyocytes and hepatocytes. Nature. 2003;425:968–973. 4. Vassilopoulos G, Wang P-R, Russell DW. Transplanted bone marrow regenerates liver by cell fusion. Nature. 2003;422:901–904. 5. Wang X, Willenbring H, Akkari Y, et al. Cell fusion is the principal source of bone-marrow-derived hepatocytes. Nature. 2003;422:897–901. 6. Camargo FD, Finegold M, Goodell MA. Hematopoietic myelomonocytic cells are the major source of hepatocyte fusion partners. J Clin Invest. 2004;113:1266–1270. 7. Dahlke MH, Loi R, Warren A, et al. Immune-mediated hepatitis drives low-level fusion between hepatocytes and adult bone marrow cells. J Hepatol. 2006;44:334–341. 8. Lagasse E, Connors H, Al-Dhalimy M, et al. Purified hematopoietic stem cells can differentiate into hepatocytes in vivo. Nat Med. 2000; 6:1229–1234. 9. Reddy P. Pathophysiology of acute graft-versus-host disease. Hematol Oncol. 2003;21:149–161. 10. Shlomchik WD, Couzens MS, Tang CB, et al. Prevention of graft versus host disease by inactivation of host antigen-presenting cells. Science. 1999;285:412–415.

11. Min CK, Kim BG, Park G, Cho B, Oh IH. IL-10-transduced bone marrow mesenchymal stem cells can attenuate the severity of acute graft-versus-host disease after experimental allogeneic stem cell transplantation. Bone Marrow Transplant. 2007;39:637–645. 12. Mathe G, Thomas ED, Ferrebee JW. The restoration of marrow function after lethal irradiation in man: a review. Transplant Bull. 1959;6: 407–409. 13. Ferrari G, Cusella-De Angelis G, Coletta M, et al. Muscle regeneration by bone marrow-derived myogenic progenitors. Science. 1998; 279:1528–1530. 14. Petersen BE, Bowen WC, Patrene KD, et al. Bone marrow as a potential source of hepatic oval cells. Science. 1999;284:1168–1170. 15. Theise ND, Nimmakayalu M, Gardner R, et al. Liver from bone marrow in humans. Hepatology. 2000;32:11–16. 16. Camargo FD, Green R, Capetanaki Y, Jackson KA, Goodell MA. Single hematopoietic stem cells generate skeletal muscle through myeloid intermediates. Nat Med. 2003;9:1520–1527. 17. Doyonnas R, LaBarge MA, Sacco A, Charlton C, Blau HM. Hematopoietic contribution to skeletal muscle regeneration by myelomonocytic precursors. Proc Natl Acad Sci U S A. 2004;101:13507–13512. 18. Willenbring H, Bailey AS, Foster M, et al. Myelomonocytic cells are sufficient for therapeutic cell fusion in liver. Nat Med. 2004;10: 744–748. 19. Kanazawa Y, Verma IM. Little evidence of bone marrow-derived hepatocytes in the replacement of injured liver. Proc Natl Acad Sci U S A. 2003;100(Suppl 1):11850–11853. 20. Wang X, Montini E, Al-Dhalimy M, Lagasse E, Finegold M, Grompe M. Kinetics of liver repopulation after bone marrow transplantation. Am J Pathol. 2002;161:565–574. 21. Fandrich F, Lin X, Chai GX, et al. Preimplantation-stage stem cells induce long-term allogeneic graft acceptance without supplementary host conditioning. Nat Med. 2002;8:171–178.