HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: 2005-0405v1 24/4/1104    most recent Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints/Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kashofer, K. Articles by Bonnet, D. Search for Related Content PubMed PubMed Citation Articles by Kashofer, K. Articles by Bonnet, D.

TISSUE-SPECIFIC STEM CELLS

In Vivo Formation of Unstable Heterokaryons after Liver Damage and Hematopoietic Stem Cell/Progenitor Transplantation

Hematopoietic Stem Cell Laboratory, Cancer Research UK, London Research Institute, London, United Kingdom

Key Words. Fusion • Hepatocytes • Hematopoietic stem cells • Plasticity

Correspondence: Dominique Bonnet, Ph.D., Hematopoietic Stem Cell Laboratory, Cancer Research UK, London Research Institute, 44 Lincoln’s Inn Fields, London, WC2A 3PX, United Kingdom. Telephone: 020 72693281; Fax: 020 72693581; e-mail: d.bonnet{at}cancer.org.uk

Received August 22, 2005; accepted for publication November 2, 2005.

	ABSTRACT

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
Following reports of lineage plasticity in human hematopoietic stem cells (HSCs), we investigated the potential of human cord blood HSC-enriched cells to create hepatocytes in hosts after inducing liver damage. Carbon tetrachloride induces severe liver damage and subsequent repair via mitosis of resident hepatocytes. It additionally leads to a threefold increase in homing of human mononuclear cells to bone marrow and liver and subsequently to a substantial enhancement of bone marrow engraftment. Eight weeks after liver damage and infusion of an enhanced green fluorescent protein (eGFP) lentivirus-transduced human HSC-enriched cell population, we observed eGFP-positive cells with clear hepatocyte morphology in the livers of animals. These eGFP-positive cells co-expressed human albumin, and reverse-transcription polymerase chain reaction (PCR) analysis demonstrated the presence of human albumin and -anti-trypsin mRNA. However, two antibodies against human mitochondria and human nuclei failed to mark eGFP-positive hepatocyte-like cells but did give clear staining of donor-derived hematopoietic cells. Subsequent fluorescent in situ hybridization (FISH) analysis revealed the presence of mouse Y chromosome in eGFP-positive hepatocyte-like cells. To resolve this discrepancy, we performed single-cell PCR analysis of microdissected eGFP-positive hepatocyte-like cells and found that they contained mostly mouse and little human genomic material. FISH analysis highlighting the centromeres of all human chromosomes revealed only few human chromosomes in these cells. From these results, we conclude that similar to their murine counterparts, human hematopoietic cells have the potential to fuse with resident host hepatocytes. Because no selective pressure is applied to retain the human genomic material, it is gradually lost over time, leading to a variable phenotype of the chimeric cells and making their detection difficult.

	INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
It has now been several years since the surprising discovery that adult stem cells can give rise to a multitude of tissues when transplanted into a suitable recipient [1, 2]. Stem cells of different origin seemed to be much more potent than previously thought, giving rise to tissue normally derived from all three different germ layers. Hematopoietic stem cells (HSCs) have been the most thoroughly investigated and were shown to rescue mice from a fatal genetic liver disease by integrating into the liver parenchyme [3]. The conversion of hematopoietic cells into hepatocytes has received much attention and has been demonstrated with human HSCs in sheep [4] and mouse [5–7] models.

The issue of stem cell plasticity has since remained highly controversial. Some results could not be reproduced by other laboratories [8] and artifacts have made analysis difficult [9], and also the issue of cell fusion has complicated the debate. In vitro experiments have revealed that fusion of cells can produce new cell types that share properties of both parent cell lines. Cells derived from co-culture of adult stem cells with embryonic stem cells regain full pluripotentiality and contribute to all tissues in blastocyst-chimeras, but at the price of a tetraploid genome [10]. Cell fusion was also shown in a model of tissue repair in which mesenchymal stem cells fused with small airway epithelial cells and then participated in the repair of the damaged lung epithelium in vitro [11]. A very interesting cell fusion outcome has been reported in purkinje neurons [12]. These specialized neurons span from the cerebellum to other brain tissues and are generated only during gestation. After transplantation of HSCs, markers of the donor cells could be shown in these neurons, and careful analysis of the cell nuclei revealed that fusion was the cause of this phenomenon. The generation of hepatocytes that express donor markers and were able to rescue an otherwise lethal genetic liver disease in the FAH^–/^– mouse model could also be attributed to fusion [13, 14]. A macrophage from the myeloid lineage of hematopoietic development has been shown to be the fusion partner of resident hepatocytes [15, 16]. Although fusion is clearly the cause of phenotypic rescue in this disease model, the severity of the damage and other circumstances might favor fusion here, but real transdifferentiation is still possible in other models. In agreement with this, no evidence of fusion has been found in other models of hematopoietic to hepatic conversion [4–6], and the mechanisms that lead to one pathway or the other are unknown.

Because we have previously reported the occurrence of hepatocyte-like cells expressing human markers in the nonobese diabetic/severe combined immunodeficient (NOD/SCID) xeno-transplantation model [7], we used a similar approach to determine whether these cells were generated by fusion or transdifferentiation. To potentially enhance the number of cells of human origin, we performed bone marrow transplantation into mice that received carbon tetrachloride (CCl₄) to induce liver damage. Identification of cells of donor origin was facilitated by transduction of the original stem cell population with a lentiviral construct expressing enhanced green fluorescent protein (eGFP), and additional analysis of the genetic content of the hepatocyte-like cells was performed.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
Primary Cells
We obtained cord blood from mothers attending University College Hospital or Royal London Hospital (London) after informed consent. The protocol was approved by the hospital research ethics committees. Mononuclear cells (MNCs) were obtained by Ficoll-Paque density centrifugation (Pharmacia, Buckinghamshire, U.K., http://www.apbiotech.com) according to the manufacturer’s instructions.

Animals
All animal experiments were performed in compliance with Home Office and institutional guidelines. NOD/SCID mice were originally obtained from Dr. Leonard Schultz (Jackson Laboratory, Bar Harbor, ME, http://www.jax.org) and bred at Charles Rivers Laboratories (Wilmington, MA, http://www.criver.com). They were kept in micro-isolators and fed sterile food and acidified water. Mice aged 8–12 weeks were irradiated at 375 rads (cesium 137 source) up to 24 hours before i.v. injection of cells. Conditioning with CCl₄ was performed by intraperitoneal (i.p.) injection of CCl₄ (Sigma) dissolved in sterile corn oil on the day of irradiation. Mice were transplanted with 1–3 x 10⁵ human eGFP transduced (lineage-depleted or CD34⁺) cells purified from cord blood via i.v. injection.

Stem Cell Purification, Transduction, and Labeling
Human cord blood MNCs were obtained by centrifugation onto a ficoll layer and subjected to red cell lysis using 0.8% ammonium chloride (StemCells Technologies, Meylan, France, http://www.stemcell.com) solution. HSCs were enriched by lineage depletion (StemCell Technologies) or by CD34 selection (Mini-Max; Miltenyi Biotec, Bergisch Gladbach, Germany, http://www.miltenyibiotec.com) according to the manufacturers’ instructions. Lentiviral marking was performed with an HIV-1–based self-inactivating (SIN) lentiviral vector (pHRSINcPPT-SEW), which carries the eGFP reporter gene under the control of the spleen focus-forming virus long terminal repeat (a kind gift from Prof. A. Thrasher, Institute of Child Health, London). High-titer viral stocks were obtained as previously described [17, 18].

For transduction, lineage-depleted or CD34-enriched human stem cells from cord blood were seeded into plastic dishes. High-titer virus particles were added at a multiplicity of infection of 35, and transduction was performed for 20 hours. In other experiments, MNCs were stained with PKH26 (Sigma-Genosys, Dorset, U.K., http://www.sigma-genosys.com) according to the manufacturer’s protocol. To label mitotic cells in vivo, a dose of 50 mg/kg bromo-deoxy-uridine (BRDU) (Sigma-Aldrich, Dorset, U.K., http://www.sigmaaldrich.com) dissolved in water was administered by i.p. injection 24 hours before sacrifice.

FACS Analysis of Bone Marrow and Liver Engraftment
Bone marrow cells were collected by flushing the femurs, tibias, and iliac crests from transplanted animals and were subsequently stained with antibodies against human hematopoietic markers CD45-CY5, CD19-PE, CD33-PE, and CD34-PE (Becton, Dickinson and Company, Oxford, U.K., http://www.bd.com). DAPI (4,6 diamidino-2-phenylindole) was used for exclusion of dead cells, and analysis was performed using an LSR (Becton, Dickinson and Company). Single-cell suspension from liver tissue was obtained by liver perfusion as described previously [19]. For the assessment of the homing of MNCs to the liver, the whole livers were dissolved in 5 ml of buffer and 300 µl aliquot of cells was mixed with a fixed amount of fluorescent beads (Perfect Count microspheres; Quest Biomedical, Solihull, West Midlands, U.K., http://questbiomedical.com) for fluorescence-activated cell sorting (FACS) analysis.

Immunohistochemistry
Liver tissue was collected from animals and either immediately frozen in liquid nitrogen or fixed overnight in 4% neutral buffered formalin (NBF). Cryosections were thawed, fixed for 5 minutes in NBF at room temperature, washed, incubated in acetone at –20°C for 10 minutes, and blocked with 1:25 swine serum with 0.1% Triton-X for 30 minutes. After washing, antibodies against human nuclei or human mitochondria (cat. nos. MAB1281 and MAB1273, respectively; Chemicon, Temecula, CA, http://www.chemicon.com) were applied at 1:20 dilution in phosphate-buffered saline (PBS) for 1 hour. Secondary anti-mouse antibodies bearing fluorochromes Alexa 488 or Alexa 594 were used at a 1:100 dilution for visualization. Paraffin sections were cut, de-waxed in Histoclear (RaLamb, East Sussex, U.K., http://www.ralamb.co.uk), and blocked in 3% hydrogen peroxide for 10 minutes. Antigen unmasking was performed by microwaving at 700 W for 10 minutes in citrate buffer (pH 6). After blocking, anti-eGFP polyclonal rabbit serum (Invitrogen, Paisley, U.K., http://www.invitrogen.com) was applied 1:500 in PBS while anti-human-albumin antibody (Cedar Lane, Hornby, ON, Canada, http://www.cedarlanelabs.com) was used 1:25. Secondary antibodies conjugated to a fluorochrome (anti-mouse or anti-rabbit Alexa Fluor 594 or Alexa Fluor 488; Invitrogen) were used at a 1:100 dilution, while horseradish peroxidase–conjugated secondary antibody (DakoCytomation, Cambridgeshire, U.K., http://www.dakocytomation.com) was used at 1:400 for visualization with diaminobenzidine (Sigma-Aldrich). Detection of incorporated BrdU into mitotic cells was performed on paraffin sections. After dewaxing and blocking, rat anti-BrdU antibody (Seralab, Leicestershire, U.K., http://www.seralab.co.uk) was applied in a 1:500 dilution for 1 hour. Anti-rat HRP-conjugated secondary antibody (Sigma) was used at 1:100.

RT-PCR
Approximately 25 mg of liver tissue was cut from frozen tissue samples. Tissue was homogenized in Trizol reagent (Invitrogen) with a pistil, and RNA was extracted according to the manufacturer’s protocol. Three micrograms of RNA was subjected to DNase digestion (Quiagen, Sussex, U.K., http://www1.qiagen.com), and subsequent generation of cDNA was performed with the Sensiscript kit (Quiagen) in 20 µl volume. Two microliters of cDNA was added to polymerase chain reaction (PCR) reactions with primers for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (5'-CATCAAGAAGGTGGTGAAGCAG, 3'-TGTGGGCCATGAGGTCCACCAC), human ß-Actin (5'-cAGGCTGCTTCCAGCTCC, 3'-GGGTATAACGCAACTAAGTCATAG), eGFP (5'-ACCCCGACCACATGAAGCAGC, 3'-CGTTGGGGTCTTTGCTCAGGG), human -anti-trypsin (AAT) (5'-GCTGAAGACCTTAGTGATGC, 3'-CTTTGAAGTCAAGGACACCG), and human albumin (5'-CATTAGCTGCTGATTTTGTTGAAAG, 3'-TGTGCAGCATTTTGTGACTCTG). PCR was performed with the High Fidelity PCR kit (Roche, Basel, Switzerland, http://www.roche.com) using buffer 3. PCR conditions were 94°C for 30 seconds, annealing temperature for 30 seconds, and extension at 72°C for 1 minute. Annealing temperature was 60°C for ß-actin, albumin, and GAPDH, 62°C for AAT, and 67°C for eGFP. PCR products were separated by agarose gel electrophoresis and stained with ethidium bromide.

Microdissection and Single-Cell PCR
Frozen tissue sections were stained for eGFP, but after dehydration no coverslip was applied. Sections were placed on the stage of a PALM micro-dissection microscope (PALM Micro-laser Technologies AG, Bernried, Germany, http://www.palm-microlaser.com). A standard PCR tube cap containing 15 µl of PCR buffer supplemented with 0.5% Triton was placed above the section, and eGFP-stained cells were positioned in the center of the field of view. The laser was used to excise the single eGFP-positive cell from the surrounding tissue and then to catapult it into the tube cap. The tube was closed, supplemented with 2 µl proteinase K (20 µg/ml), and incubated at 48°C overnight. This was followed by 60 rounds of improved primer extension pre-amplification (i-PEP) PCR as described [20]. Three microliters of the reaction was used as template for nested PCR specific to the human and mouse tumor necrosis factor- (TNF-) locus and for eGFP. The following primers were used: human TNF-, outer: 5'-AGGAACAGCACAGGCCTTAGTG, 3'-AGGAACAGCACAGGCCTTAGTG, inner: 5'-GGATACTCAGAACGTCATGGCC, 3'-CTCATACCAGGGCTTGGCCT; and mouse TNF-, outer: 5'-CCACCATCAAGGACTCAAATG, 3'-CACTGGGTCCTCCAGGACA, inner: 5'-GGCTTTCCGAATTCACTGGAG, 3'-CCCCGGCCTTCCAAATAAA. eGFP outer was the same as mentioned above; inner: 5'-GCATCGACT TCAAGGAGGAC, 3'-TGCTCAGGTAGTGGTTGTCG. PCR was performed as mentioned above, with an annealing time of 65°C for the first round and 55°C for the second round. Efficiency of PCR was determined by cutting hepatocytes from normal human and mouse liver tissue.

Fluorescence In Situ Hybridization
Frozen sections were fixed in 4% paraformaldehyde for 10 minutes, and images of natural eGFP fluorescence were immediately taken on a Zeiss LSM 510 confocal microscope (Carl Zeiss, Jena, Germany, http://www.zeiss.com). The position of individual eGFP-positive hepatocytes was saved with custom-made software, and subsequent fluorescence in situ hybridization (FISH) analysis was performed on the slide. The slides were digested with 0.005% pepsin in 0.9% saline at pH 1.5 for 2–5 minutes. After dehydration, probes for mouse Chromosome Y, human Chromosome 1, or human centromeres (Cambio, Cambridge, U.K., http://www.cambio.co.uk) were applied according to the manufacturer’s recommendations. The tissue section was then covered with a coverslip and sealed with rubber glue. Sections were then incubated at 80°C for 10 minutes for co-denaturation of nuclei and probes. After overnight incubation at 37°C, the sections were washed in 0.4% SSC at 72°C for 30 seconds and mounted in fluorescence mounting medium (DakoCytomation) supplemented with DAPI. The slides were again inserted into the confocal microscope, the exact position saved earlier was reloaded, and images of the nuclei of the same cells photographed before were taken. Overlay images of eGFP and FISH signals were produced using Adobe Photoshop (Adobe Systems Incorporated, San Jose, CA, http://www.adobe.com).

Statistical Analysis
The effect of CCl₄ on the homing of MNCs to the bone marrow or liver was analyzed using the Student’s paired t test. Generalized linear models were used to assess the significance of the association between the dose of CCl₄ and the outcomes (i.e., percentage engraftment, eGFP-positive cells in the liver). Also by this technique, we tested whether engraftment or the number of eGFP liver-like cells was associated with the source of cells used (i.e., CD34⁺ or lineage-depleted cells).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
Increased Homing of HSCs after Systemic Administration of CCl₄
Preliminary experiments determined that 40 µl of CCl₄ induced severe liver damage visualized by strong centrolobular necrosis (Fig. 1A) and subsequent regeneration by mitosis (Fig. 1D, 1E). Damage was repaired by day 14 but was still visible after 7 days (Fig. 1B, 1C). CCl₄ is known to enhance homing of HSCs to the liver due to upregulation of SDF1 expression [21]. To quantify the homing potential of MNCs to both liver and bone marrow after CCl₄ treatment, we transplanted MNCs that had been labeled with PKH26 into CCl₄-treated or control animals. CCl₄-treated (40 µl, n = 5) or control animals (n = 5) were transplanted with the same number of PHK26-labeled MNCs. After 24 hours, animals were sacrificed, and a single-cell suspension was prepared from bone marrow by flushing the femur and tibia and from the liver by perfusion [19]. FACS analysis of bone marrow revealed a threefold marked increase in homing to the bone marrow in the animals that had received CCl₄ compared with untreated controls (Fig. 1F) (0.184% ± 0.09% vs. 0.382% ± 0.17%, respectively; p < .05). A 2.5-fold increase (although not statistically significant) in homing to the liver in the animals that had received CCl₄ compared with controls (Fig. 1F) was also observed (0.088% ± 0.045% vs. 0.2575% ± 0.26%; p = .11, respectively). We also tested the long-term effect of this increase in homing to the bone marrow on the level of engraftment at 8 weeks. FACS analysis demonstrated that the increased homing due to CCl₄ resulted in an enhancement in the level of human bone marrow engraftment after 8 weeks (Table 1) (p < .05). In summary, we found that not only the use of CCl₄ but the dose of CCl4 used significantly increased the level of engraftment (p = .0005 and p = .017, respectively).

View larger version (85K): [in this window] [in a new window]   Figure 1. Effect of CCl4 on liver morphology and hematopoietic stem cell homing. (A): Intraperitoneal administration of CCl4 (40 µl) results in massive centrolobular necrosis 96 hours after injection (hematoxylin eosin staining). Gross liver morphology remains abnormal 7 days later: control normal liver (B), CCl4-damaged liver (C). BrdU staining reveals increased mitotic activity in CCl4-damaged (D) compared to healthy liver (E). Homing of transplanted mononuclear cells to the bone marrow and liver in the first 24 hours is increased up to threefold by treatment with CCl4 (40 µl) (F). Schematic outline of the experimental procedures used in this study (G). Abbreviations: BM, bone marrow; BrdU, bromo-deoxy-uridine; CCl4, carbon tetrachloride; i.p., intraperitoneal, i.v., intravenously; MNC, mononuclear cells.

View larger version (49K): [in this window] [in a new window]   Figure 2. Analysis of eGFP-positive hepatocyte-like cells. (A): Detection of eGFP-positive hepatocyte-like cells with an anti-GFP antibody in frozen (a, b) and paraffin sections (c) seen in the liver of CCl4-treated animals 8 weeks after CD34+ cell engraftment. These eGFP-positive hepatocyte-like cells (d, g) co-express human albumin ([e, f], merged images) but are negative for human CD45 ([h, i], merged images). eGFP-positive donor-derived hematopoietic cells (j) are also present in the liver and express the pan-leukocyte marker CD45 ([k, l], merged images). (B): Human liver–specific mRNA in the livers of CCl4-treated mice after CD34+ cell engraftment. Human mRNA for ß-actin, AAT, and albumin could be detected in human liver RNA (positive control: lane 1) but not in mouse liver RNA (negative control: lane 2) or in a mouse RNA that had been mixed with 10% of human MNC (specificity test: lane 3). eGFP was detected in the positive control containing eGFP plasmid (lane 4) and in five different mouse livers (lanes 5–9). The expression of human liver–specific markers was detectable only in mice, which had received 40 µl CCl4 (lanes 5–7), whereas in mice treated with 10 µl CCl4, the expression was absent (lanes 8, 9). Abbreviations: AAT, -anti-trypsin; CCl4, carbon tetrachloride; eGFP, enhanced green fluorescent protein; GFP, green fluorescent protein.

eGFP-Positive Hepatocyte-Like Cells Do Not Express Pan-Human Protein Markers
To further investigate the origin of the eGFP-positive hepatocytes, we performed immunohistochemistry with anti-human mitochondria and anti-human nuclei antibodies. The former antibody efficiently stains the cytoplasm of all human cells, including hepatocytes (Fig. 3A). The latter antibody gives bright staining in the nuclei of human cells (Fig. 3I). Surprisingly, in the experimental animals, donor-derived hematopoietic cells were clearly labeled with both antibodies, whereas staining was completely absent in the eGFP-positive hepatocyte-like cells (Fig. 3).

View larger version (63K): [in this window] [in a new window]   Figure 3. eGFP-positive hepatocyte-like cells do not stain with anti-human-mitochondria or anti-human-nuclei antibody. Anti-human-mitochondria antibody highlights human mitochondria in the cytoplasm of human cells (A), but no signal is obtained in mouse liver cells (E). eGFP-positive hepatocyte-like cells in the liver of experimental animals (B, F) were negative for human mitochondria (C, G), whereas adjacent hematopoietic cells (arrows) were clearly labeled ([D, H], merged images). Likewise, anti-human-nuclei antibody highlights human nuclei (I) but not mouse liver cell nuclei (M). eGFP-positive hepatocyte-like cells (J, N) were negative for anti-human nuclei (K, O), whereas adjacent hematopoietic cells (arrows) again stained positive ([L, P], merged images). All the cells analyzed are from mice injected with human cord blood CD34+ cells. Abbreviation: eGFP, enhanced green fluorescent protein.

View larger version (21K): [in this window] [in a new window]   Figure 4. eGFP-positive hepatocyte-like cells contain mouse Y chromosome and lack human chromosome 1. eGFP-positive cells in liver sections were photographed, and the location of each cell was saved ([A, D] native eGFP fluorescence). Subsequently, fluorescence in situ hybridization for mouse Y chromosome (red) or human chromosome 1 (yellow) was performed. Images corresponding to the same area of the slide as the original eGFP-positive cells were taken (B, E). Overlay images (C, F) demonstrate the presence of mouse Y chromosome in eGFP-positive hepatocyte-like cells as well as in surrounding eGFP-negative cells. Simultaneous detection of mouse Y chromosome and human chromosome 1 ([H, I] overlay image) revealed absence of human chromosome 1 in eGFP-positive hepatocyte-like cells but presence of human chromosome 1 in eGFP-positive hematopoietic donor-derived cells (left arrow image [H] and overlay image [I]). All the cells shown are from mice injected with cord blood CD34+ cells. Abbreviation: eGFP, enhanced green fluorescent protein.

View larger version (37K): [in this window] [in a new window]   Figure 5. Presence of both human and mouse TNF- in single eGFP-positive hepatocyte-like cells after single-cell PCR analysis. Frozen tissue sections of experimental livers were stained with eGFP antibody. Single eGFP-positive hepatocyte-like cells were cut from the section with a PALM laser microdissection microscope ([A], before [left] and after [right] dissection) and captured into PCR tubes containing 1x PCR buffer with 0.5% tween. After digestion with proteinase K overnight, 60 cycles of i-PEP PCR were performed (see Materials and Methods). Three microliters of this reaction was then used as template in duplicate nested PCR for human TNF-, mouse TNF-, and eGFP. Efficiency and specificity of PCR were determined by cutting cells from mouse liver or human liver ([B, D], average of three experiments). Twenty-four single eGFP-positive hepatocyte-like cells were cut out and processed. Human TNF- could be detected in only two of 24 samples, whereas mouse TNF- was present in 10 of 24 and eGFP in 7 of 24 (C). The PCR upper and lower gel panels represent replicates of the same reaction. Cells 2 and 5 are positive for human and mouse TNF-. Samples 1–3 were additionally tested for the presence of human TCR locus, which was detected only in sample 2 (E). The cells analyzed were derived from mice injected with either CD34+ or lineage-depleted cells. Abbreviations: eGFP, enhanced green fluorescent protein; i-PEP, improved primer extension pre-amplification; PCR, polymerase chain reaction; TCR, T cell receptor; TNF, tumor necrosis factor.

View larger version (48K): [in this window] [in a new window]   Figure 6. eGFP-positive hepatocyte-like cells contain few human centromeres. FISH for human centromeres highlights multiple spots and clusters (red) in human hepatocytes (A) but not in mouse hepatocytes (D). eGFP-positive hepatocyte-like cells in experimental animals injected with human cord blood CD34+ cells (B–F) contain only few human centromeres ([C, F], high magnification). Nuclei were counter-stained with DAPI (blue). Abbreviations: DAPI, 4,6 diamidino-2-phenylindole; eGFP, enhanced green fluorescent protein; FISH, fluorescence in situ hybridization.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
In this study, we have shown that the pretreatment of experimental animals with the hepatotoxin CCl₄ has a profound effect on stem cell homing. Originally intending to provide an enhancement to liver engraftment, we found that not only was homing to the damaged liver tissue increased twofold but that simultaneously the homing of stem cells to the bone marrow was enhanced. Tissue damage leads to a drastic increase in the levels of secreted chemokines, cytokines, and proteolytic enzymes as part of the regeneration and repair process, which has a profound impact on stem cell migration [21] and repopulation [22]. Although the effect of CCl₄ is mediated via P-450 enzymes and is described as liver-specific [23], the systemic distribution of chemokines and cytokines in response to liver injury may well lead to the observed increase in bone marrow homing. Unlike the FAH^–/– model, in which donor-derived hepatocytes can be selectively expanded, the repeated administration of an indiscriminate drug like CCl₄ will damage not only host hepatocytes but also newly formed donor-derived hepatocytes. Thus, repeated rounds of damage to further increase liver engraftment were not possible, restricting any beneficial effect of CCl₄ to the initial phase of transplantation. Nevertheless, the observation that the percentage of final engraftment in the bone marrow was strongly correlated with the elevated initial seeding efficiency warrants a closer investigation of the influence exerted on the bone marrow by CCl₄ intoxication.

After transplantation of human cord blood–enriched HSCs transduced to express eGFP, we observed cells of hepatocyte morphology that were eGFP-positive and thus demonstrated a property of the originally transplanted cell population. This observation is well in line with the literature [4, 6, 24] although most other studies have largely relied on staining with the HepPar1 antibody to detect cells of human origin, highlighting mainly cells close to vessels. The eGFP-positive hepatocyte-like cells in our study were evenly distributed throughout the parenchyme of the liver and only rarely were situated close to vessels. The concurrent detection of hepatocyte-specific markers and markers to establish the origin of cells is one of the major challenges in this field of research. The eGFP-positive hepatocyte-like cells observed in our study showed a clear co-expression of human albumin and lack of the human pan-leukocyte marker CD45. RT-PCR analysis revealed the presence of human liver-specific mRNA species (albumin, AAT) as previously reported [7]. The infrequent incidence of the eGFP-positive hepatocyte-like cells and the low levels of these mRNA species did not allow for a conclusive quantitative correlation between the two (data not shown).

Further investigation into the origin of these eGFP-positive hepatocyte-like cells surprisingly revealed no expression of human mitochondrial or nuclear proteins after staining with the respective antibodies. On the other hand, eGFP-positive hematopoietic cells clearly stained positive with the same antibodies, suggesting fundamental differences in the genetic content between the former and the eGFP-positive hepatocyte-like cells. This notion was reinforced by the presence of the murine Y chromosome in the eGFP-positive hepatocyte-like cells but not in the eGFP-positive hematopoietic-derived cells, clearly demonstrating a different origin for these two populations. Collectively, the identification of very few human markers expressed in the eGFP-positive hepatocyte-like cells, as revealed by different methods, suggested that a simple fusion model between a donor-derived hematopoietic cell with a resident hepatocyte with retention of two full sets of chromosomes could not explain these results.

Therefore, we proceeded to analyze individual cells by single-cell PCR to identify the presence of mouse genetic material (TNF- locus) and eGFP gene in 25% of eGFP-positive hepatocyte-like cells. The frequency of detection of the murine TNF- locus was similar to that achieved for control tissues. Although there are numerous factors that influence detection of genomic loci in single-cell PCR, including tissue preservation state, freshness and thickness of the section, and effectiveness of the PCR, the frequency of detection of the human TNF- locus in the eGFP-positive hepatocyte-like cells was much lower than what was observed for control cells. This phenomenon could not be explained by the spatial organization of the particular genomic locus. Genomic loci that are equally distributed in the spherical nucleus could be separated into two different sections when cut, explaining the occurrence of samples that show no locus detection by PCR. In the case of eGFP, this led to the occurrence of samples that were negative by PCR for the eGFP gene although the cells were expressing eGFP protein as determined by immunohistochemistry. Further FISH analysis of eGFP-positive hepatocyte-like cells for the presence of human centromeres confirmed that only residual human chromosomes were retained in these cells. The loss of substantial amounts of donor-derived genetic material demonstrated here is consistent with earlier reports from experiments with murine stem cells, in which genetic material from the donor cell was gradually lost from the heterokaryons [13].

Our data suggest that fusion between a donor-derived human hematopoietic cell and a resident hepatocyte, similar to the results obtained in the FAH^–/– model [13–15], has taken place. Interestingly, in the FAH^–/– model, the number of nodules derived from clonal expansion of single cells suggested a starting population of approximately 300 fusion events in the whole liver. Presumably because a lack of growth advantage, our cells did not form nodules, but we speculate that the number of initial fusion events observed was greater than in the FAH^–/– model, because we saw approximately 10 eGFP-positive cells per 10⁵ mouse liver cells. On one hand, this number might be overestimated. Indeed, the nature and sensitivity of the technique of laser microdissection doesn’t preclude amplification of potential contaminating RNA from human hematopoitic cells lying below the dissected nucleus. On the other hand, this number excludes an unknown number of cells, which have lost the eGFP locus after the fusion event. The organic solvent CCl₄ could be the causative agent of this fusion because of interference in the stability of membranes. Although the amount of eGFP-positive cells and hence heterokaryons correlates with the amount of CCl₄ administered to the animals, a time-course analysis up to 2 weeks after cell transplantation did not reveal the formation of any eGFP-positive hepatocyte-like cells. This suggested that CCl₄ was not directly involved—at least at the initial phase of transplantation—in the fusion mechanism (data not shown).

In summary, we have shown that transplantation of human-enriched HSCs into NOD/SCID mice after CCl₄-induced liver damage leads to the emergence of unstable cell chimeras, which arise from fusion. After initial cell fusion, the human genome is gradually lost from the cells, and only remnants of human chromosomes are present after 8 weeks. Which chromosomes are retained under normal or selective pressure conditions is a question that remains to be elucidated. The phenomenon of cell fusion after HSC/progenitor transplantation and the unpredictability of the cellular phenotype of the resultant hybridoma cells in vivo may complicate the prospective use of these cells for general cell therapy purposes. However, the long-term effect and fate of such chimeric cells after enriched HSC transplantation will determine the potential therapeutic use of hematopoietic stem/ progenitor cells in human liver diseases.

	ACKNOWLEDGMENTS

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
This work was supported by Cancer Research UK. K.K. and E.K.S. contributed equally to this work.

	DISCLOSURES

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
The authors indicate no potential conflicts of interest.

	REFERENCES

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 

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

2. Jiang Y, Jahagirdar BN, Reinhardt RL et al. Pluripotency of mesenchymal stem cells derived from adult marrow. Nature 2002;418:41–49.[CrossRef][Medline]

3. 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.[CrossRef][Medline]

4. Almeida-Porada G, Porada CD, Chamberlain J et al. Formation of human hepatocytes by human hematopoietic stem cells in sheep. Blood 2004; 104:2582–2590.[Abstract/Free Full Text]

5. Jang YY, Collector MI, Baylin SB et al. Hematopoietic stem cells convert into liver cells within days without fusion. Nat Cell Biol 2004; 6:532–539.[CrossRef][Medline]

6. Wang X, Ge S, McNamara G et al. Albumin-expressing hepatocyte-like cells develop in the livers of immune-deficient mice that received transplants of highly purified human hematopoietic stem cells. Blood 2003; 101:4201–4208.[Abstract/Free Full Text]

7. Danet GH, Luongo JL, Butler G et al. C1qRp defines a new human stem cell population with hematopoietic and hepatic potential. Proc Natl Acad Sci U S A 2002;99:10441–10445.[Abstract/Free Full Text]

8. Wagers AJ, Sherwood RI, Christensen JL et al. Little evidence for developmental plasticity of adult hematopoietic stem cells. Science 2002; 297:2256–2259.[Abstract/Free Full Text]

9. Jackson KA, Snyder DS, Goodell MA. Skeletal muscle fiber-specific green autofluorescence: Potential for stem cell engraftment artifacts. STEM CELLS 2004;22:180–187.[Abstract/Free Full Text]

10. Ying QL, Nichols J, Evans EP et al. Changing potency by spontaneous fusion. Nature 2002;416:545–548.[CrossRef][Medline]

11. Spees JL, Olson SD, Ylostalo J et al. Differentiation, cell fusion, and nuclear fusion during ex vivo repair of epithelium by human adult stem cells from bone marrow stroma. Proc Natl Acad Sci U S A 2003;100: 2397–2402.[Abstract/Free Full Text]

12. Weimann JM, Johansson CB, Trejo A et al. Stable reprogrammed heterokaryons form spontaneously in Purkinje neurons after bone marrow transplant. Nat Cell Biol 2003;5:959–966.[CrossRef][Medline]

13. Wang X, Willenbring H, Akkari Y et al. Cell fusion is the principal source of bone-marrow-derived hepatocytes. Nature 2003;422:897–901.[CrossRef][Medline]

14. Vassilopoulos G, Wang PR, Russell DW. Transplanted bone marrow regenerates liver by cell fusion. Nature 2003;422:901–904.[CrossRef][Medline]

15. Willenbring H, Bailey AS, Foster M et al. Myelomonocytic cells are sufficient for therapeutic cell fusion in liver. Nat Med 2004;10:744–748.[CrossRef][Medline]

16. Camargo FD, Finegold M, Goodell MA. Hematopoietic myelomonocytic cells are the major source of hepatocyte fusion partners. J Clin Invest 2004;113:1266–1270.[CrossRef][Medline]

17. Demaison C, Parsley K, Brouns G et al. High-level transduction and gene expression in hematopoietic repopulating cells using a human immuno-deficiency virus type 1-based lentiviral vector containing an internal spleen focus forming virus promoter. Hum Gene Ther 2002;13:803–813.[CrossRef][Medline]

18. Siapati EK, Bigger BW, Miskin J et al. Comparison of HIV- and EIAV-based vectors on their efficiency in transducing murine and human hematopoietic repopulating cells. Mol Ther 2005;12:537–546.[Medline]

19. Seglen PO. Preparation of rat liver cells. 3. Enzymatic requirements for tissue dispersion. Exp Cell Res 1973;82:391–398.[CrossRef][Medline]

20. Dietmaier W, Hartmann A, Wallinger S et al. Multiple mutation analyses in single tumor cells with improved whole genome amplification. Am J Pathol 1999;154:83–95.[Abstract/Free Full Text]

21. Kollet O, Shivtiel S, Chen YQ et al. HGF, SDF-1, and MMP-9 are involved in stress-induced human CD34⁺ stem cell recruitment to the liver. J Clin Invest 2003;112:160–169.[CrossRef][Medline]

22. Mallet VO, Mitchell C, Mezey E et al. Bone marrow transplantation in mice leads to a minor population of hepatocytes that can be selectively amplified in vivo. Hepatology 2002;35:799–804.[CrossRef][Medline]

23. Wu J, Danielsson A, Zern MA. Toxicity of hepatotoxins: New insights into mechanisms and therapy. Expert Opin Investig Drugs 1999;8:585–607.[CrossRef][Medline]

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

This article has been cited by other articles:

				D. A. Hess, T. P. Craft, L. Wirthlin, S. Hohm, P. Zhou, W. C. Eades, M. H. Creer, M. S. Sands, and J. A. Nolta Widespread Nonhematopoietic Tissue Distribution by Transplanted Human Progenitor Cells with High Aldehyde Dehydrogenase Activity Stem Cells, March 1, 2008; 26(3): 611 - 620. [Abstract] [Full Text] [PDF]

				Y N Kallis, M R Alison, and S J Forbes Bone marrow stem cells and liver disease Gut, May 1, 2007; 56(5): 716 - 724. [Full Text] [PDF]

				R. Poulsom, E. I. Prodromidi, C. D. Pusey, and H. T. Cook Cell therapy for renal regeneration--time for some joined-up thinking? Nephrol. Dial. Transplant., December 1, 2006; 21(12): 3349 - 3353. [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 2005-0405v1 24/4/1104    most recent Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints/Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kashofer, K. Articles by Bonnet, D. Search for Related Content PubMed PubMed Citation Articles by Kashofer, K. Articles by Bonnet, D.