SEARCH

SEARCH BY CITATION

Keywords:

  • Orthotopic liver transplantation;
  • Spontaneous mobilization;
  • Hematopoietic and endothelial stem cells

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

In animals, the bone marrow (BM) is a source of liver-repopulating cells with therapeutic potential in case of tissue damage. However, the early response of human BM-derived stem cells (SC) to liver injury is still unknown. Here, we studied 24 patients undergoing orthotopic liver transplantation (OLT) for end-stage liver disease or hepatocellularcarcinoma, and 13 patients submitted to liver resection. The concentration of circulating BM-derived SC was determined by phenotypic analysis and clonogenic assays. Moreover, we assessed the serum level of inflammatory and tissue-specific cytokines. Reverse transcriptase-polymerase chain reaction and fluorescence-in situ hybridization were also used to characterize mobilized SC. At baseline, patients showed a significant lower concentration of circulating CD133+, CD34+ SC and clonogenic progenitors (colony-forming unit cells) than healthy controls. However, the time-course evaluation of peripheral blood cells after OLT demonstrated the significant early mobilization of multiple subsets of hematopoietic and endothelial stem/progenitor cells. Cytogenetic and molecular analyses of CD34+ cells showed the host origin of mobilized SC and the expression of transcripts for GATA-4, cytokeratin 19, and α-fetoprotein hepatocyte markers. In contrast with OLT, only total circulating CD34+ cells significantly increased after liver resection. Mobilization of BM cells after OLT or liver surgery was associated with increased serum levels of granulocyte-colony stimulating factor, interleukin-6, stem cell factor, hepatocyte growth factor, and vascular endothelial growth factor. In summary, we demonstrate that tissue damage after OLT and liver resection induces increased serum levels of multiple cytokines but only ischemia/reperfusion injury associated with OLT results in the remarkable mobilization of BM stem/progenitor cells.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

At least two resident cell populations have been shown to act as putative stem cells (SC) in the liver [1]. Whereas moderate cell loss is replaced by the proliferation of mature hepatocytes, more severe liver injury induces the activation of a “facultative” SC compartment located within the smallest branches of the intrahepatic biliary tree, the so-called “oval cells” which give origin to hepatocytes and the bile duct epithelium. Interestingly, liver oval SC share the CD34 and c-kit antigens with hematopoietic stem/progenitor cells [2].

Recently, a third source of liver-repopulating cells has proven to be the bone marrow (BM) as animal studies have shown that BM cells contribute, at low levels, to liver regeneration after tissue injury [3, [4], [5], [6]7]. In humans, results from gender mismatch liver transplantation studies demonstrated that BM-derived cells of host origin can differentiate into hepatocytes and cholangiocytes in transplanted livers [8, 9], although the mechanism by which hematopoietic stem cells (HSCs) acquire the function of mature hepatocytes (i.e., cell fusion or true trans-differentiation) is still debated [10, [11], [12]13]. Of note, different cellular types from the BM or the peripheral blood (PB) [14] of hematopoietic [3, 4, 7, 15, [16]17] and non-hematopoietic origin [18, 19] share the capacity to generate liver cells in vitro and in vivo. Therefore, the BM is generally regarded as a novel source of liver-repopulating cells that can be exploited for therapeutic purposes [20, 21].

The potential of human CD34+ HSCs to migrate to the liver in response to stress signals to repair non-hematopoietic tissue has been recently established in nonobese diabetic/severe combined immunodeficiency (NOD/SCID) mice [22]. In this model, stress-induced signals, such as increased expression of the chemokine stromal cell-derived factor-1 (SDF-1), stem cell factor (SCF), and hepatocyte growth factor (HGF) were able to recruit to the liver human CD34+ HSCs with hepatic-like potential. Similarly, acute and chronic liver injury models were generated by injecting carbon tetrachloride (CCl4) in C57Bl6 mice followed by HSC mobilizing doses of the hematopoietic cytokine granulocyte colony-stimulating factor (G-CSF) [23]. In both experimental models, G-CSF administration improved the histological damage and accelerated the regeneration process. These findings translated into a strong survival benefit in G-CSF-treated group versus CCl4 group. Thus, G-CSF treatment significantly improved the liver histology of chemically injured mice by promoting endogenous repair mechanisms and by mobilizing HSCs.

Despite extensive investigations in animal models of liver injury [21], no information is presently available on the mechanisms and kinetics of mobilization of BM SC after liver damage in humans. Recently, it has been shown that partial hepatectomy in living liver donors induces the mobilization of BM-derived myelomonocytic progenitor cells with hepatic differentiation potential in vitro that may play a role in liver regeneration [24]. In this investigation, we studied the early response of BM-derived SC of patients undergoing orthotopic liver transplantation (OLT) or liver resection.

Our phenotypic, functional and molecular studies demonstrated: (a) the significant, physiologic mobilization of host-derived hematopoietic and endothelial stem/progenitor cells after tissue injury; (b) this finding was associated with increased serum levels of cytokines involved in SC mobilization and/or liver repair; (c) ischemia/reperfusion organ damage associated with OLT is a more efficient stimulus to SC mobilization than massive liver resection.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

Patients and Control Group

We assessed the peripheral blood (PB) stem/progenitor cells compartment of 24 and 13 adult patients undergoing OLT and liver resection, respectively. The clinical characteristics of study patients are reported in Table 1. Blood samples were taken one day before (day −1) and then one, three, seven, and 14 days after surgery when possible. Single samples from 12 healthy subjects served as controls. Study patients underwent liver surgery according to standard procedures either at the Liver and Multi Organ Transplant Center of the University of Bologna, Bologna, Italy or at the Department of Surgery and Organ Transplantation, University of Udine, Udine, Italy. The study protocol was approved by the local Ethic Committees and conformed to the ethical guidelines of the 1975 Declaration of Helsinki. Written informed consent was obtained by all patients and healthy controls.

Table Table 1.. Characteristics of patients
Thumbnail image of

Flow Cytometry Analysis

The phenotype of circulating cells was evaluated by conventional dual color immunofluorescence using a panel of fluorescein isothiocyanate (FITC)-conjugated and phycoerythrin (PE)-conjugated monoclonal antibodies (MoAbs): FITC-conjugated anti-CD34, PE-conjugated anti-CD38, PE-conjugated anti-CXCR4, PE-conjugated anti-CD90 (Becton, Dickinson and Company, Franklin Lakes, NJ, http://www.bd.com). PE-conjugated anti-CD133 was purchased from Miltenyi Biotec (Bergisch Gladbach, Germany, http://www.miltenyibiotec.com). MoAbs against vascular endothelial growth factor receptor (VEGFR)-2 (KDR) (clone KDR-1), VEGFR-1 (Flt-1 receptor) (clone FLT-19) (Sigma, Saint Louis, MI, http://www/sigmaaldrich.com) antigens were tested by indirect immunofluorescence. Negative controls were isotype-matched irrelevant MoAbs (Becton Dickinson). Briefly, 100 μl of PB in heparin-containing tubes (Becton Dickinson) were incubated for 15 minutes at room temperature with 10 μl of MoAbs. In case of indirect immunofluorescence, cells were then incubated for 15 minutes with 5 μl of PE/FITC-conjugated goat anti-mouse immunoglobulins (Becton Dickinson). After red-cell lysis (FACS Lysing Solution, Becton Dickinson), the samples were centrifuged, washed twice with phosphate buffer and fixed with 1% paraformaldehyde (Sigma). Five × 104 cells were acquired by flow cytometer (FACSCalibur; Becton Dickinson) and analyzed by CellQuest software (Becton Dickinson) [25, 26]. The analysis was performed excluding cellular debris in a side scatter/forward scatter dot plot. The percentage of positive cells was calculated subtracting the value of the appropriate isotype controls. The absolute number of positive cells per microliter was calculated as follows: percentage of positive cells × white blood cells count/100.

Hematopoietic Colony-Forming Unit Cells Assays

Assessment of clonogenic hematopoietic progenitors was performed in methylcellulose as previously reported [25, 26]. In brief, mononuclear light density (<1.077 g/ml) cells (MNC) were isolated by density gradient centrifugation from 30 ml of PB. Subsequently 2 × 105 cells were plated in duplicate in culture medium MethoCult H4433 (Stem Cell Technologies Inc., Vancouver, BC, Canada, http://www.stemcell.com), following the manufacturer's instructions. Burst-forming units-erythroid (BFU-E) and colony-forming units-granulocyte-macrophage (CFU-GM) (together referred to as colony-forming unit cells [CFU-C]), were scored after 14 days of incubation at 37°C in a fully humidified 5% CO2 atmosphere, as described previously [25, 26].

Cell Preparation and HSC Purification

Enriched MNC were resuspended in 1% bovine serum albumin (BSA) (Sigma) and then processed by MiniMacs high-gradient magnetic separation column (Miltenyi Biotec) to obtain highly purified CD34+ cells as already reported [25, 26]. To assess the percentage of CD34+ elements, aliquots of the CD34+ target cells were restained with a MoAb (HPCA-2-FITC; Becton Dickinson) directed to a different epitope of CD34 antigen than that (Qbend 10) used with the magnetic system. Propidium iodide (2 μg/ml) was added for the detection of nonviable cells, which were excluded from the analyses. Cytometric analysis was performed on a gated population set on scatter properties by using FACScan equipment (Becton Dickinson). A minimum of 10,000 events was collected in list mode on FACScan software. The percentage of CD34+ cells after magnetic separation was 93 ± 6%.

Fluorescence In Situ Hybridization and Reverse Transcriptase-Polymerase Chain Reaction

Both fluorescence in situ hybridization (FISH) and reverse transcriptase-polymerase chain reaction (RT-PCR) assays were performed on immunomagnetically highly purified CD34+ cells. The slides were prepared immediately before hybridization. We used two probes: CEP X that hybridizes to centromere region Xp11.1-q11.1 (Spectrum Orange, Vysis Inc., Downers Grove, IL, http://www.vysis.com) and CEP Y to chromosome band Yq13 (satellite III, Spectrum Green, Vysis Inc.). The normal female pattern was two red signals, and the normal male pattern was one red and one green signal. Hybridization was performed overnight at 37°C on a Hybrite semiautomated FISH hybridization chamber (Vysis), according to the manufacturer's guide with some modifications. Slides were mounted and counterstained with 4′,6-diamidino-2-phenylindole (DAPI) in antifade. The samples were examined with a Nikon E-1000 fluorescent microscope equipped with FITC/tetramethylrhodamine isothiocyanate/DAPI filter and a Genikon FISH system image capture (Nikon Instruments, Rockville, MD, http://www.nikonusa.com). Two-hundred and ten CD34+ HSC were analyzed.

The expression of GATA-4, cytokeratin (CK)-19 and α-fetoprotein (AFP) transcripts was assessed by RT-PCR amplification using gene-specific primers (GATA-4, F: 5′-aggcattacatacaggctcacc-3′, R: 5′-ctgtggcctctatcacaagatg-3′; CK-19, F: 5′-atggccgagcagaaccggaa-3′, R: 5′-ccatgagccgtcggtactcc-3′; AFP, F: 5′-tgcagccaaagtgaagagggaaga-3′, R: 5′-catagcgagcagcccaaagaagaa-3′) [19]. In brief, total cellular RNA was extracted from PB mononuclear cells by TRIzol reagent (Invitrogen, Milan, Italy, http://www.invitrogen.com). Reverse transcription was performed using the SuperScript II First-Strand Synthesis System (Invitrogen) following manufacturer's instructions. The expression of different transcripts was assessed by PCR amplification following standard protocols. PCR products were analyzed by agarose gel electrophoresis and visualized by ethidium bromide staining. RNA integrity was evaluated by the expression of β2-microglobulin.

Cytokine Measurements

The serum levels of study cytokines was measured before OLT or liver resection and on days 1, 7, and 14 after surgery by high-sensitivity enzyme-linked immunosorbent assays. SCF, G-CSF, HGF, SDF-1, vascular endothelial growth factor (VEGF), and tumor necrosis factor-α (TNF-α) kits were from R&D Systems (Wiesbaden, Germany, http://www.rndsystems.com). Interleukin (IL)-6 kit was from Biosource International (Camarillo, CA, http://www.biosource.com).

Statistical Analysis

The results were expressed as median (range). The data were analyzed by the Wilcoxon test for paired data, Mann-Whitney test and by the Fisher's exact test. Spearman rank correlation test was used for correlation analysis. A p value <.05 was considered statistically significant. Data processing was carried out with the SPSS statistical package for Windows (11.0.1, Chicago, http://www.spss.com).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

OLT Patients Mobilize Hematopoietic and Endothelial Stem/Progenitor Cells

At baseline, patients with liver disease showed decreased hematopoietic function as demonstrated by the significant lower concentration of PB CD133+ (median 0.35 cells per microliter; range 0–1.8), CD34+ (0 cells per microliter; 0–0.7) SC, and CFU-C (19 cells per microliter; 0–525) than in healthy donors, respectively (1.44 cells per microliter; 0–10.8), (0 cells per microliter; 0–1.6), (554.5 cells per microliter; 362.5–1100) (Fig. 1).

thumbnail image

Figure Figure 1.. Impaired bone marrow (BM) function in patients with liver disease. Circulating BM-derived cells were evaluated by phenotypic analysis and clonogenic assay. Baseline concentrations of CD34+, CD133+cells, and CFU-C were significantly lower in patients undergoing liver resection and OLT, respectively, than in healthy controls. The bold line indicates the median value. Abbreviations: CFU-C, colony-forming unit cells; OLT, orthotopic liver transplantation.

Download figure to PowerPoint

Nonetheless, the longitudinal phenotypic evaluation of circulating cells after OLT demonstrated the small, but significant, early mobilization of CD34+ HSCs (day +7 0.13 cells per microliter; 0–9.3; p = .014) (Fig. 2A), which remained higher than the baseline value until day +14 (0 cells per microliter; 0–12.1; p = .041) and returned to pretreatment levels on day +30 after OLT (data not shown). In keeping with the spontaneous mobilization of CD34+ cells, we observed a significantly higher number of circulating CFU-C (85.6 CFU-C per milliliter; 0–1830; p = .028) on day +14 after transplantation (Fig. 2A). The number of CFU-C was higher than the baseline value when the results were expressed both as number of colonies/106 MNC and colonies/PB milliliter. Thus, we ruled out any confounding effect due to the possible increase of the white blood cells count after OLT.

thumbnail image

Figure Figure 2.. Phenotypic and functional analysis of peripheral blood (PB) cells before and after orthotopic liver transplantation (OLT) demonstrates the mobilization of bone marrow-derived stem/progenitor cells. (A): Following OLT, we observed the significant early increase of hematopoietic and endothelial stem/progenitor cells and the mobilization of CFU-C on day +14. The bold line indicates the median value. (B): OLT induces the mobilization of host-derived hematopoietic stem cells coexpressing mRNA for liver-specific epithelial markers. Fluorescence-in situ-hybridization analysis was performed on purified circulating CD34+ cells from a male recipient of a female-derived liver and no female cells (XX, orange/orange pattern) were observed. (C). CK-19 (i),AFP (ii), GATA-4 (iii) transcripts were assessed in circulating CD34+ cells by reverse transcription-polymerase chain reaction. Lanes 1 and 2, PB mononuclear cells from mobilized healthy donors; lanes 3 and 5, baseline samples from two nonmobilizer OLT patients; lanes 4 and 6, day +7 post-OLT samples from the same nonmobilizers; lanes 7 and 9, baseline samples from two mobilizers OLT patients); lanes 8 and 10, day +7 post-OLT samples from the same mobilizers; lane 11, positive control (human liver); lane 12, negative control (no cDNA); lane 13, molecular weight markers. Abbreviations: AFP, α-fetoprotein; β2-M, β2-microglobulin; CFU-C, colony-forming unit cells; CK, cytokeratin.

Download figure to PowerPoint

When we analyzed selected subsets of the CD34+ HSC population, we found a significant increase of very immature CD34+/CD90+ cells (day +3 0 cells per microliter; 0–7.4 vs. day −1 0 cells per microliter; 0–0.4; p = .043) and CD34+ cells coexpressing the chemokine SDF-1 receptor CXCR4 (day +3 0.65 cells per microliter; 0–3.9, day +7 0.5 cells per microliter; 0–4.6, day +14 0.83 cells per microliter; 0–3 as compared to day −1 0 cells per microliter; 0–0.7; p < .05) (Fig. 2A).

Also the concentration of multipotent PB CD133+ stem/progenitor cells and endothelial CD34+/VEGFR-2 (KDR)+ progenitors (Fig. 2A) augmented following OLT reaching the statistical significance (p < .05) at days +3 (day +3 1.17 cells per microliter; 0–7.36 vs. day −1 0.35 cells per microliter; 0–1.8) and +7 after surgery (day +7 0 cells per microliter; 0–6.3 vs. day −1 0 cells per microliter; 0–0), respectively, and returning to baseline levels on day +14.

Interestingly, the cold ischemia time of the liver transplant (Table 1) directly correlated with the number of circulating CD133+ cells at day +3 after OLT (r = .5; p = .01). No difference was found in terms of mobilization taking into account the etiology of liver disease, the immunosuppression regimen or the post-transplant complications such as acute allograft rejection.

Mobilized CD34+ SC Are of Host Origin and Coexpress mRNAs for Epithelial Liver-Specific Markers

In a representative case of sex-mismatch transplant, circulating CD34+ cells from a male recipient of a female-derived liver, were purified and analyzed by FISH (Fig. 2B). Two-hundred and ten cells were analyzed, and no female cells (XX, orange/orange pattern) were observed. This result demonstrated the host origin of circulating CD34+ cells and ruled out that the observed increase of PB HSCs may be due to hematopoietic cells carried over by the transplanted liver [27].

Moreover, the RT-PCR assay showed that circulating CD34+ cells analyzed after OLT, but not baseline steady-state cells from the same patients or PB samples from G-CSF-mobilized healthy donors, expressed mRNA for GATA-4, CK-19, and AFP liver markers (Fig. 2C). Therefore, we demonstrated that OLT induces the release from the BM of HSCs expressing liver markers.

SC Mobilization in Patients Undergoing Liver Resection

Thirteen patients, mostly with primary (seven cases) or metastatic liver cancer (three cases) (Table 1), who were submitted to liver resection were also studied to answer the question of whether the surgical procedure, per se, induces SC mobilization and whether the removal of a large part of the liver (up to 55%) represents an efficient stimulus to recruit BM cells into PB.

Similarly to OLT patients, resected individuals showed impaired BM function as demonstrated by the significantly reduced number of circulating CD34+ (0 cells per microliter; 0–0), CD133+ cells (0 cells per microliter; 0–1.36) and CFU-C (70 CFU-C per milliliter; 0–684) at baseline (Fig. 1). The longitudinal study performed after surgery showed only the significant mobilization (p = .028) of total CD34+ cells (0 cells per micrliter; 0–1.2) at day +3 after surgery (Fig. 3). The extent and the length of mobilization of CD34+ cells after liver resection was markedly lower than after OLT. PB endothelial stem/progenitor cells as well as selected subsets of HSCs did not increase in response to liver resection even after removal of 55% of the organ (data not shown).

thumbnail image

Figure Figure 3.. Liver resection induces the mobilization of CD34+ HSCs. The phenotype of PB cells of resected patients was evaluated by conventional immunofluorescence. Following liver resection we observed the early mobilization of total CD34+ at day +3. The bold line indicates the median value.

Download figure to PowerPoint

SC Mobilization Is Associated with Increased Serum Levels of Selected Cytokines

When we analyzed the serum level of cytokines involved in SC mobilization and/or liver repair, we found that the baseline levels of IL-6 (60 pg/ml; 0–453) and HGF (1,866 pg/ml; 997–3,960) were significantly higher (p < .03), and VEGF levels (40 pg/ml; 0–182) lower (p = .001), in patients undergoing OLT than in healthy controls, respectively (0 pg/ml; 0–46, 793 pg/ml; 578–1,109, 225 pg/ml; 85–450) (Fig. 4). In liver resected patients, HGF (1,850 pg/ml; 454–2,670) and SCF (316 pg/ml; 0–520) serum concentrations were significantly higher and lower (p < .03), respectively, than in control samples (SCF 904 pg/ml; 453–1,720) (Fig. 4). No difference was found as for SDF-1, G-CSF, and TNF-α (data not shown). In parallel with mobilization of BM-derived SC, after OLT, we observed the significant increase (p < .03) of SCF (day +7 1,303 pg/ml; 682–1,867), G-CSF (day +1 85 pg/ml; 0–2,275), IL-6 (day +7 252 pg/ml; 0–933), and VEGF (day +7 178 pg/ml; 23–613 and day +14 684 pg/ml; 31–2,340) (Fig. 5) but not of HGF and TNF-α (data not shown). Of note, the serum level of the hematopoietic and mobilizing cytokine G-CSF peaked at day +1 after OLT and returned to baseline levels at day +7 (Fig. 5). The serum level of SDF-1 significantly decreased early after transplantation (Fig. 5), and this finding may be consistent with the proteolytic degradation of this chemokine in the BM inducing the release of HSCs into PB [28]. Conversely, SDF-1 may be highly expressed at the injury site to recruit circulating CD34+ cells [22].

thumbnail image

Figure Figure 4.. Baseline serum levels of study cytokines. The baseline levels of IL-6 and HGF were significantly higher, and VEGF levels lower, in patients undergoing OLT than in healthy controls. In liver resected patients, HGF and SCF serum concentrations were significantly higher and lower, respectively, than in control samples. The bold line indicates the median value. Abbreviations: HGF, hepatocyte growth factor; IL, interleukin; OLT, orthotopic liver transplantation; SCF, stem cell factor; VEGF, vascular endothelial growth factor.

Download figure to PowerPoint

thumbnail image

Figure Figure 5.. Orthotopic liver transplantation induces increased serum levels of multiple cytokines. After liver transplantation, we observed the significant increase of IL-6, SCF, VEGF and G-CSF but not of HGF. The serum level of SDF-1 significantly decreased at days +1 and +7 after transplantation. The bold line indicates the median value. Abbreviations: G-CSF, granulocyte colony-stimulating factor; HGF, hepatocyte growth factor; IL, interleukin; SCF, stem cell factor; SDF, stromal cell-derived factor; VEGF, vascular endothelial growth factor.

Download figure to PowerPoint

When serum cytokine levels were analyzed in patients undergoing liver resection, we found the significant increase of SCF, IL-6, VEGF, HGF, and G-CSF (Fig. 6). Of note, the serum level of SCF was remarkably higher after OLT (day +7 1,303 pg/ml; 682–1,867) than after liver resection (day +7 499 pg/ml; 313–913). Again, G-CSF peaked at day +1 and rapidly decreased to pretreatment levels.

thumbnail image

Figure Figure 6.. Liver resection augments the serum level of multiple cytokines. When serum cytokine levels were analyzed in patients undergoing liver resection, we found a significant increase of IL-6, HGF, SCF, VEGF, and G-CSF. The bold line indicates the median value. Abbreviations: G-CSF, granulocyte colony-stimulating factor; HGF, hepatocyte growth factor; IL, interleukin; SCF, stem cell factor; SDF, stromal cell-derived factor; VEGF, vascular endothelial growth factor.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

To the best of our knowledge, this is the first study reporting the mobilization of both hematopoietic and endothelial stem/progenitor cells after OLT. We thus provide a strong rationale for the clinical observation of human liver cells deriving from circulating cells of extrahepatic origin upon tissue injury [8, 9, 29]. Consistently, hematopoietic stem cells mobilization contributing to liver repair after full/partial OLT or partial hepatectomy have been recently described in animal models [30, [31], [32]33]. In the present study, we demonstrated that despite baseline impaired hematopoietic function patients undergoing OLT and ischemia/reperfusion damage have a 3.4-fold increase of circulating CD34+ cells, which is appreciable within 24 hours of surgery, peaks at days +7 and +14, and normalizes within 30 days. Phenotypic and functional analyses further showed that both primitive CD34+/CD90+ HSCs and more mature, committed CFU-C were mobilized into PB within 14 days of OLT. Of note, the early and significant elevation of the concentration of different subsets of hematopoietic and endothelial stem/progenitor cells into PB occurred despite similar values of circulating leukocytes before and after OLT.

Mobilization of HSCs was likely enhanced by increased serum levels of IL-6, SCF, and G-CSF after OLT. In particular, G-CSF and SCF have been shown to enhance, in animals, the expression of the chemokine SDF-1 and its cognate receptor, CXCR4, on HSCs to recruit marrow cells to the injured tissue [22, 34]. Our results also suggest that the SDF-1/CXCR4 axis may be implicated in stress-induced trafficking and migration to the liver of human HSCs for tissue repair. Thus, we extend previous studies that demonstrated the significant increase of the number of circulating CXCR4+ cells and the early decrease of SDF-1 plasma level after acute myocardial infarction [35].

Similarly, mobilization of endothelial progenitors may be due to the increased serum level of VEGF following the vascular damage associated with OLT. As a matter of fact, circulating CD133+ and CD34+/VEGFR-2+ (KDR+) cells are known to contribute to neoangiogenesis after tissue ischemia and organ regeneration in animal models [36, [37]38] after the release of angiogenic factors including VEGF. In humans, increased serum levels of VEGF have been recently associated with the mobilization of endothelial SC after acute myocardial infarction [35, 39, 40].

Thus, it is conceivable that hematopoietic and endothelial progenitors can play different roles, if any, in tissue repair. Since regeneration of the injured organ involves the proliferation of parenchymal cells as well as neovascularization, it may well be that tissue injury also induces the activation of progenitors for endothelium. These findings support the concept that “plasticity” may not be restricted to a unique SC population but, rather, may be a general property of marrow cells that redirect their transcriptional program under appropriate stimuli. Of note, mobilization of CD133+ hematopoietic and endothelial stem/progenitor cells directly correlated with the length of the cold ischemia of the transplanted organ. Thus, it may be hypothesized that the greater the extent of the tissue damage (reflected by the cold ischemia time) is, the higher SC mobilization is.

FISH analysis formally demonstrated the host origin of mobilized BM-derived CD34+ HSCs after OLT. In addition, molecular studies showed that only HSCs circulating after OLT, but not CD34+ cells evaluated before surgery or purified from G-CSF-mobilized healthy donors, expressed mRNAs for liver-specific epithelial markers such as AFP and CK-19 and for the GATA-4 transcription factor. Therefore, our data support the hypothesis of the pool of tissue-committed BM SC readily available after tissue damage as previously proposed [35, 41]. The engraftment of mobilized BM-derived SC to the injured liver, their actual role, and the mechanism of action for tissue regeneration after OLT were topics beyond the scope of this study. However, previous papers [8, 9] have demonstrated that BM-derived cells can give rise to hepatocytes and cholangiocytes in transplanted livers, although their engraftment efficiency and long-term survival was questioned [42, 43]. Concerning the mechanism of action, both cell fusion and trans-differentiation may be considered mechanisms of SC plasticity, although SC are not necessarily involved [10, 11]. From a therapeutic point of view, as long as the function of the damaged tissue is restored it may not be crucial to determine whether true trans-differentiation occurs as demonstrated in the fumarylacetoacetate hydrolase model of liver injury [3]. A third mechanism has been suggested. Recent data reported that bone marrow may contain some nonhematopoietic tissue-committed stem cells [44, [45]46], thereby suggesting the existence of an heterogenous population of stem cells. In particular, a small population of mobilizing cells that express mRNA for early liver tissue-committed stem/progenitor cells has been described. These circulating tissue specific cells may play an important role in tissue repair following injury [41, 47, [48]49].

Whereas ischemia/reperfusion liver damage associated with OLT induced the extensive mobilization of several subsets of hematopoietic and endothelial BM-derived SC, liver resection was a weaker stimulus to recruit significant numbers of BM cells into PB despite the increase of the serum level of hemopoietic/mobilizing cytokines. Thus, the physiological stress associated with major liver surgery, per se, only induced the significant mobilization of total CD34+ HSCs. Moreover, we did not find any correlation between the extent of liver resection (and subsequent tissue reconstitution) and mobilization. Taken together, these findings suggest that activation of resident stem/progenitor cells may play the major role in tissue repair after liver resection, although cell therapy with HSCs may significantly help this process [50, 51]. Thus, it may be hypothesized that different stress signals (i.e., OLT or liver resection) may induce the activation of different BM-derived cell populations. Of note, the reduced serum levels of SCF, one of the main cytokines to stimulate HSC proliferation and mobilization, may account for the decreased concentration of circulating BM-derived SC in resected patients as compared to healthy controls. Similarly, SCF concentration was lower after liver resection than after OLT and this finding may help explaining the different extent of mobilization of patients submitted to different surgical procedures. This observation is consistent with previous findings of Baccarani et al. [52] showing that liver injury (transplantation or hepatic resection) affects circulating levels of SCF. At variance with OLT patients, the serum levels of SDF-1 did not decrease significantly after liver resection. This finding suggests a lower proteolytic degradation at bone marrow level and might contribute to explaining the different mobilization pattern between OLT and partial liver resection. The baseline serum level of HGF was higher in both study populations as compared to normal controls and increased further after liver resection, thus suggesting that HGF may be the main cytokine involved in tissue repair by acting on resident liver cells and by recruiting HSCs. In summary, our data extend to the liver the observation that SC recruitment into PB may be a relevant physiological process in case of tissue injury [35, 41] and underline the potential role of BM-derived cells for cell therapy of liver diseases as recently proposed [50, 51].

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References

The research was supported by the University of Bologna (funds for selected topics) and the Italian Association Against Leukemia, Bologna (BolognAil).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgements
  9. References