CELL THERAPY FOR LIVER DISEASE
Molecular and functional characterization of CD133
+stem/progenitor
cells infused in patients with end-stage liver disease reveals their
interplay with stromal liver cells
LUCIA CATANI
1, DARIA SOLLAZZO
1, ELISA BIANCHI
2, MARILENA CICIARELLO
1,
CHIARA ANTONIANI
1, LICIA FOSCOLI
1, PAOLO CARACENI
3,4,
FERDINANDO ANTONINO GIANNONE
4, MAURIZIO BALDASSARRE
4,
ROSARIA GIORDANO
5, TIZIANA MONTEMURRO
5, ELISA MONTELATICI
5,
ANTONIA D’ERRICO
3, PIETRO ANDREONE
3, VALERIA GIUDICE
6, ANTONIO CURTI
1,
ROSSELLA MANFREDINI
2& ROBERTO MASSIMO LEMOLI
71Department of Experimental, Diagnostic and Specialty Medicine, Institute of Hematology “L. e A. Seràgnoli,”
University of Bologna, Bologna, Italy,2Centre for Regenerative Medicine “Stefano Ferrari,” Department of Life
Sciences, University of Modena and Reggio Emilia, Modena, Italy,3Department of Medical and Surgical Sciences,
University of Bologna, Bologna, Italy,4Center for Applied Biomedical Research (C.R.B.A.), Azienda Ospedaliero/
Universitaria di Bologna, Bologna, Italy,5Cell Factory, Unit of Cellular Therapy and Cryobiology, Fondazione IRCCS
Ca’ Granda Ospedale Maggiore Policlinico, Milano, Italy,6Immunohematology Service and Blood Bank—Azienda
Ospedaliero/Universitaria di Bologna, Bologna, Italy, and7Clinic of Hematology, Department of Internal Medicine
(DiMI), University of Genoa, Genoa, Italy
Abstract
Background aims. Growing evidence supports the therapeutic potential of bone marrow (BM)-derived stem/progenitor cells
for end-stage liver disease (ESLD). We recently demonstrated that CD133+stem/progenitor cell (SPC) reinfusion in pa-tients with ESLD is feasible and safe and improve, albeit transiently, liver function. However, the mechanism(s) through which BM-derived SPCs may improve liver function are not fully elucidated. Methods. Here, we characterized the circu-lating SPCs compartment of patients with ESLD undergoing CD133+cell therapy. Next, we set up an in vitro model mimicking SPCs/liver microenvironment interaction by culturing granulocyte colony-stimulating factor (G-CSF)-mobilized CD133+and LX-2 hepatic stellate cells. Results. We found that patients with ESLD show normal basal levels of circulating hematopoi-etic and endothelial progenitors with impaired clonogenic ability. After G-CSF treatment, patients with ESLD were capable to mobilize significant numbers of functional multipotent SPCs, and interestingly, this was associated with increased levels of selected cytokines potentially facilitating SPC function. Co-culture experiments showed, at the molecular and function-al levels, the bi-directionfunction-al cross-tfunction-alk between CD133+SPCs and human hepatic stellate cells LX-2. Human hepatic stellate cells LX-2 showed reduced activation and fibrotic potential. In turn, hepatic stellate cells enhanced the proliferation and survival of CD133+SPCs as well as their endothelial and hematopoietic function while promoting an anti-inflammatory profile. Discussion. We demonstrated that the interaction between CD133+SPCs from patients with ESLD and hepatic stel-late cells induces significant functional changes in both cellular types that may be instrumental for the improvement of liver function in cirrhotic patients undergoing cell therapy.
Key Words: CD133+stem/progenitor cells, cell therapy, end-stage liver disease, liver microenvironment
Introduction
Cirrhosis is defined as the widespread disruption of normal liver structure by fibrosis and the formation
of regenerative nodules leading to end-stage liver disease (ESLD) [1,2]. ESLD is irreversibly associ-ated with liver failure, and currently, orthotopic liver transplantation (OLT) is the only curative treatment
Correspondence: Lucia Catani, PhD, Department of Experimental, Diagnostic and Specialty Medicine, Institute of Hematology “L. e A. Seràgnoli,” University of Bologna, Via Massarenti 9, 40138 Bologna, Italy. E-mail:lucia.catani@unibo.it
(Received 14 December 2016; accepted 3 August 2017)
ISSN 1465-3249 Copyright © 2017 International Society for Cellular Therapy. Published by Elsevier Inc. All rights reserved.
[3]. Thus, the development of therapeutic approaches to prevent or reverse the progression of chronic liver disease to ESLD is essential.
Chronic liver injury is invariably accompanied by progressive fibrosis. Indeed, hepatic stellate cells and myofibroblasts are major fibrogenic cell types that contribute to extracellular matrix accumulation during the progression from chronic liver injury/inflammation. Specifically, hepatic stellate cells, which reside in the space of Disse outside the liver sinusoids, maintain a quiescent phenotype and store vitamin A under phys-iological conditions. However, when liver injury occurs, they become activated and trans-differentiate into pro-liferating myofibroblast-like cells, which loose their vitamin A droplets, expressα-smooth muscle actin (α-SMA) and secrete pro-fibrogenic mediators and extra cellular matrix proteins[4–6]. Therefore, controlling the activation of the hepatic stellate cells population is considered a potential therapeutic target.
Growing evidence supports the role of bone marrow (BM)-derived stem/progenitor cells (SPCs) as an at-tractive therapeutic approach to ESLD. Animal models demonstrated that BM-derived SPCs home to damaged liver tissue and participate to tissue regeneration by promoting epithelial cell proliferation or decreasing liver fibrosis[7,8](reviewed in Behbahan et al.)[9]. In ad-dition, in recent years, small clinical studies suggested that the infusion of highly selected BM-derived CD34+ hemopoietic SPCs or whole mononuclear cell prepa-rations may transiently improve liver function in human cirrhosis (reviewed inVainshtein et al.,Wang et al., and Bailey et al.[10–12]. However, the underlying mecha-nisms, as well as the cellular subset responsible for this therapeutic effect, remain unclear.
CD133 is a highly conserved antigen that identi-fies a population with both hematopoietic and endothelial lineage potential[13–15]. It has been re-ported that CD133 antigen may be a marker for early progenitors and long-term repopulating cells with high engraftment potential.Therefore, purified CD133+cells are currently used in clinical transplantation proto-cols [16–19]. Moreover, the significant increase of circulating CD133+/ CD14+cells observed after partial hepatectomy, but not after other major abdominal surgery [20], suggests the mobilization of defined subsets of CD133+SPCs with potential regenerative activity after loss of liver tissue. In this view, the com-bination of portal vein embolization with infusion of CD133+ BM-derived stem cells substantially in-creased hepatic regeneration compared with portal vein embolization alone[21]in patients with malignant liver lesions. Furthermore, administration of CD133+SPCs has already been used therapeutically to promote the regeneration of post-infarction myocardium[22,23]. On the basis of this background, we previously demonstrated that CD133+SPCs spontaneously
mo-bilize in cirrhotic patients undergoing OLT [24]. Subsequently, we investigated whether infusion of pu-rified, mobilized, CD133+SPCs may be therapeutically relevant in ESLD. To this end, we recently com-pleted a phase I clinical study where increasing numbers of highly purified CD133+ SPCs, mobi-lized with granulocyte colony-stimulating factor (G-CSF), were reinfused through the hepatic artery in 12 patients with ESLD. Our results showed that CD133+ SPCs reinfusion in patients with ESLD is feasible, safe and may induce the transient improvement of liver function “to bridge” cirrhotic patients to OLT[25]. Here, for the first time, we characterized the cir-culating SPCs compartment of cirrhotic patients undergoing the cell therapy trial. We also investi-gated the interaction between pluripotent CD133+ SPCs and hepatic stellate cells, which are the major fibrogenic cell type in the liver microenvironment. Methods
Patient characteristics
CD133+SPCs were isolated from leukapheresis (AF) of 12 G-CSF–treated ESLD patients previously en-rolled in a phase I clinical trial for CD133+ SPCs reinfusion through the hepatic artery[25]. Detailed description of patients characteristics are shown in Sup-plementary Table SI. As control, CD133+SPCs were also isolated from the AF of G-CSF–treated hema-topoietic stem cell donors (five cases). Peripheral blood (PB) was obtained from 17 normal age-matched vol-unteers.The study conformed to the ethical guidelines of the 1975 Declaration of Helsinki, and written in-formed consent was obtained by patients and controls. Ex vivo characterization of the mobilized cells
Phenotypic and functional characterization of the mobilized cells
Ethylenediamine tetraacetic acid–treated PB samples were taken from 12 patients with ESLD during the mobilization phase (1 day before, and then 5 and 14 days after G-CSF treatment). Blood samples from 17 healthy subjects served as controls.
The phenotype of circulating hematopoietic and endothelial SPCs was evaluated by conventional flow cytometry analysis as described in Supplementary Table SII. Endothelial progenitor cells (EPCs) were defined as CD34+KDR+ events, as previously de-scribed [26]. Cells (1 × 106) were acquired and analyzed by FACSDiva software (BD). The analysis was performed excluding cellular debris in a Side Scatter (SSC)/Forward Scatter (FSC) dot plot. The percentage of positive cells was calculated subtract-ing the value of the appropriate isotype controls. The absolute number of positive cells per milliliter was
cal-culated as follows: percentage of positive cells× white blood cell count (WBC) / 100. Assessment of circu-lating clonogenic hemopoietic progenitors (CFU-Cs) was performed with mononuclear cells (MNCs; 2× 105) in methylcellulose medium, as previously reported [25]. Colony-forming unit EPCs (CFU-EPCs) have been evaluated through a commercially available kit (CFU-Hill liquid medium kit; Stemcell Technologies, Voden Medical Instruments) [27] fol-lowing the manufacturer’s instructions. Colonies were enumerated by an inverted microscope (Axiovert 40, Zeiss). CFU-EPCs were classically defined as central clusters of round cells surrounded by spindle-shaped cells. The PB of ESLD patients (at baseline and after G-CSF mobilization—day+5) and healthy donors was centrifuged for 15 min at 1000g within 30 min of col-lection. The serum was then collected and stored at −80°C until quantification.The serum levels of study cytokines/matrix metalloproteinases (MMPs) was mea-sured by high-sensitivity enzyme-linked immunosorbent assay (ELISA). MMP-9, hepatocyte growth factor (HGF), C-X-C motif chemokine 12 (CXCL12) and vascular endothelial growth factor (VEGF) kits were from R&D Systems.
CD133+SPC isolation under Good Manufacturing Practice conditions
PB MNCs obtained from G-CSF–mobilized standard-volume AF were centrifuged 10 min at 200g and resuspended in 95 mL phosphate-buffered saline (Miltenyi Biotech)/0.01% human serum albumin (Kedrion) solution. Immunomagnetic selection of CD133+ SPCs was performed by using the human CD133+ MicroBead kit (Miltenyi Biotech) and a CliniMACS device (CliniMACS Plus, Miltenyi Biotech). After selection, cells were resuspended in normal saline solution (S.A.L.F.) with 10% of human serum albumin (Albital 200 g/L, Kedrion; vol:vol) plus 10% dimethyl sulfoxide (Bioniche Pharma; vol:vol) and cryopreserved in bags using a controlled rate cell freezer (Nicool Plus, Air Liquide). The preparations were then stored in the vapor phase of liquid nitro-gen until reinfusion. Aliquots of each cell preparation were cryopreserved in vials and employed in co-culture studies. In all the production steps, aliquots of the starting material (leukapheresis), of the inter-mediate and final product were used for quality control analysis. In the starting materials,WBCs were counted using the hemocytometer ABX Micros 60 CT-HORIBA and CD133+ SPCs were counted by flow cytometry using Stem kit reagents (Beckman Coulter) and a specific antibody against CD133/2 (Miltenyi Biotech) using ISHAGE method gating strategy, that allow to determine the enumeration of viable CD133+ SPCs (expressed as CD45+/7AAD−/CD133+ SPCs). The CD133+ selected cells were counted using an
automated validated method device based on an in-tegrated fluorescence microscope (Nucleocounter, Chemometec) that allows determining the number of the nucleated cells. The purity of the selected CD133+SPCs was assessed by flow cytometry by using CD133/2 and CD45 antibodies (Beckman Coulter). The percentage of live cells was also quantitatively determined by annexin V and propidium iodide (PI) staining (AnnexinV-FITC Apoptosis Detection kit, eBioscience). In addition, various hemopoietic/ endothelial SPCs markers (CD34, CD38, CD90, CD105, CD117, KDR, CD184, c-met,) were de-tected on isolated CD133+SPCs by flow cytometry, as described earlier. In parallel experiments, highly purified CD133+SPCs were immunomagnetically iso-lated from the AF of G-CSF–treated hematopoietic stem cell donors (five cases), following the manufac-turer’s instructions and phenotypically and functionally characterized.
In vitro co-cultures of highly purified CD133+SPCs and LX-2 cells
LX-2 cell line
LX-2 human hepatic stellate cell line[28], a gener-ous gift from Dr. Scott L. Friedman (Mount Sinai School of Medicine, New York), was maintained in Dulbecco’s Modified Minimal Essential Medium (DMEM; Gibco BRL Life Technologies) with 2% fetal bovine serum, 1% L-glutamine and 1% penicillin/ streptomycin and cultured at 37°C and 5% CO2. Co-culture assays
To determine the interaction of LX-2 and CD133+ SPCs from patients, an indirect co-culture system of LX-2 and CD133+SPCs was assembled using transwell membranes (0.45-µm pore size, Costar; Corning). Ap-proximately, 10× 103/cm2 LX-2 were placed in the lower chamber and 60× 103/cm2CD133+SPCs were placed on the membrane insert. In co-cultures with direct cell-cell contact, CD133+SPCs were seeded onto LX-2 cells. CD133+or LX-2 cells cultured alone were used as controls. To mimic the in vivo microenviron-ment, where CD133+SPCs are infused into the liver, we decided to perform monocultures and co-cultures in the presence of the medium which is optimal for LX-2 cells. Therefore, cells were plated and cultured at 37°C and 5% CO2in DMEM with 2% fetal bovine serum, 1% L-glutamine and 1% penicillin/ streptomycin. At 24, 48 and 72 h and, in selected experiments, 168 h after co-culture assay, cells/ supernatants were characterized as follows.
Cell cycle and apoptosis
Cell viability was assessed with Trypan blue dye ex-clusion. Cell cycle analysis was performed by flow
cytometry (FACSCanto BD) after PI (BD Biosci-ence) staining.The results were analyzed with a ModFit 3.0 software. Apoptosis was evaluated by using a commercially available kit (Annexin-V-FLUOS Staining Kit, Roche), following the manufacturer’s instructions. Flow cytometry analysis ofα-SMA expression
Intracellularα-smooth muscle actin (α-SMA) stain-ing was assessed in LX-2 cells usstain-ing a fluorescein isothiocyanate (FITC)-conjugated MoAb anti-human α-SMA (clone EPR5368; Abcam). Briefly, after fix-ation with paraformaldehyde (2%) in phosphate-buffered saline for 30 min at 4°C, cells were permeabilized with saponin (0.1%) and subsequently labeled with the FITC-conjugated MoAb anti- α-SMA for 30 min at 4°C. After washing, a minimum of 1× 104cells were acquired by flow cytometer BD Accuri C6. Analysis was performed excluding cellu-lar debris in a SSC/FSC dot plot. Results are expressed as mean fluorescence intensity values.
Progenitor cell assays
CFU-Cs, together with granulocyte macrophage-CFU (macrophage-CFU-GM) and erythroid-burst forming unit (BFU-E) growth of CD133+SPCs (5× 103) was evalu-ated after culture in methylcellulose medium (MethoCult H4434; Stem Cell Technologies), as pre-viously reported[29]. For long-term culture-initiating cells (LTC-IC) assays, 10× 103/mL CD133+ SPCs were plated onto irradiated murine stromal cells (M2-10B4) with bi-weekly half-medium change. After 5 weeks at 37°C, the cells were then evaluated for their secondary CFU-C activity in a methylcellulose-based assay, and the number of LTC-IC was calculated as previously reported[29].
In vitro angiogenesis
The supernatants of CD133+SPCs and LX-2 cells, either alone or cocultured with/without contact, were harvested and stored (−80°C) until further characterization. Metalloproteinase (MMP)-9, MMP-2, tissue inhibitor of metalloproteinase (TIMP)-1, hepatocyte growth factor (HGF), vascular endothe-lial growth factor (VEGF), G-CSF, CXC motif ligand (CXCL)12 and stem cell factor (SCF) con-centrations were measured in the culture supernatants by ELISA, according to the manufacturer’s instruc-tions (R&D Systems).
The angiogenic potential of CD133+ SPCs and LX-2 cells, either alone or co-cultured with/without contact, was investigated in the cell culture superna-tants with the AngioKit assay (TCS Cellworks), according to the manufacturer’s protocol. AngioKit assay is supplied as a growing co-culture of human matrix and endothelial cells (HUVEC) at the earli-est stages of tubule formation in multi-well plates.
Briefly, equal volumes of cell culture supernatants were added to the supplied growth medium. Condi-tioning medium was changed on days 1, 2, 4, 7 and 9. After 12 days of incubation, the cultures were fixed in cold formalin 2% and subsequently stained by a tubule staining kit containing anti-CD31 primary antibody (PECAM-1) conjugated with phosphatase, relevant secondary antibody and substrate (p-nitrophenol phosphate) to allow tubule visualization. The reliability of the Angio-Kit was assessed using positive (Ctrl-pos) and negative (Ctrl-neg) controls performed by adding VEGF (2µg/mL) and suramin (2 mmol/L), respectively, to the culture media.Tubular density was monitored using a Zeiss inverted micro-scope (Axiovert 40) at 2.5 magnification. Micrographs were collected, and images were analyzed with ImageJ. CD133+and LX-2 cells RNA extraction and
microarray data analysis
CD133+SPCs, isolated from two patients with ESLD enrolled in the clinical trial, and LX-2 cells were cul-tured either alone or in contact/no contact co-cultures for 24 h in the presence of SCF (5 ng/mL). After in-cubation, cells were harvested and counted.To increase the purity of CD133+SPCs after contact co-cultures, CD133+SPCs were immunomagnetically isolated, as described earlier. In a similar way, the fraction of monocultured CD133+SPCs was immunomagnetically selected to exclude differences in gene expression due to the sorting step/procedure. CD133+ SPCs purity assessed by flow cytometry was >95% for both co-cultured and monoco-cultured cells. RNA extraction and gene expression profiling (GEP) were performed as previously described[30]. Briefly, total RNA was iso-lated from 1× 105 cells for each sample using the miRNeasy MicroRNA isolation kit (Qiagen) accord-ing to the manufacturer’s protocol. RNA samples concentration and purity (assessed as 260/280 nm and 260/230 nm ratios) were evaluated by NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies). RNA integrity was assessed by using the Agilent 2100 Bioanalyzer (Agilent Technologies). GEP was per-formed on RNA samples extracted from ESLD patients-derived CD133+ SPCs (n = 2) and kept/ maintained in mono-culture or in co-culture with or without contact with LX-2 cells for 24 h. Similarly, LX-2 cells were profiled at 24 h after mono-culture or in co-culture with or without contact with ESLD patients-derived CD133+SPCs. cDNA synthesis and biotin-labeled target synthesis were performed using the GeneAtlas IVT Express Kit according to the pro-tocol supplied by Affymetrix, starting from 100 ng of total RNA for each sample.The HG-U219 Array Strips (Affymetrix) hybridization, staining and scanning were performed by using the GeneAtlas Platform. Differ-entially expressed genes (DEGs) were selected on
robust multiarray average–normalized data through a supervised analysis by using the Partek GS. 6.6 Soft-ware Package (http://www.partek.com). DEGs (fold change [FC]≥ 1.5 or FC ≤ −1.5) were identified for the following pair-wise comparisons: CD133+ cells grown in co-culture with or without contact with LX-2 versus CD133+cells grown alone, and CD133+cells grown in co-culture with contact with LX-2 versus CD133+cells grown in co-culture without contact with LX2. In a similar way, GEP was performed on RNA samples isolated from LX-2 cells at 24 h after mono-culture or in co-mono-culture with or without contact with ESLD patients-derived CD133+ SPCs. DEGs (FC ≥1.5 or FC ≤ −1.5) were identified for the following pair-wise comparisons: LX-2 cells grown in co-culture with or without contact with CD133+ cells versus LX-2 cells grown alone, and LX-2 cells grown in co-culture with contact with CD133+cells versus LX-2 cells grown in co-culture without contact with CD133+cells.
Statistical analysis
The results were expressed as mean± SEM or median (range). When appropriate, the data were analyzed by
the Wilcoxon test for paired data or Student’s t-test. Data handling and analysis were performed with SPSS software version 17.0. A P value <0.05 was consid-ered statistically significant.
Results
G-CSF mobilized progenitor cells with hemopoietic and endothelial potential in patients with ESLD enrolled in the cell therapy trial
To characterize the progenitor cell compartment in the PB of cirrhotic patients undergoing cell therapy, upon G-CSF–induced mobilization, we first enumer-ated and performed functional and phenotypic analysis of various cellular subsets. The median value of total WBC count before G-CSF treatment was compara-ble between ESLD patients and healthy donors (data not shown) as well as the numbers of circulating CD133+and CD34+cells (Figure 1A,B).
After 5 days of G-CSF treatment, we detected the mobilization into the PB of CD133+cells (P< 0.001), CD34+ cells (P < 0.001; Figure 1A,B) and of se-lected subsets of the CD34+population, such as cells co-expressing the CXCL12 receptor CXCR4 (CD184; P < 0.001; Figure 1C) or HGF receptor (c-met;
Figure 1. G-CSF mobilizes progenitor cells with hemopoietic potential in patients with ESLD. Phenotypic and functional analysis of PB cells with ESLD with hemopoietic potential from 13/14 patients before (baseline) and after G-CSF treatment (day 5) and 17 healthy sub-jects is shown. After G-CSF treatment (day 5), CD133+(A), CD34+(B), CD34+184+(C), CD34+c-met+(D) cells and CFU-C (E) were significantly higher than baseline values. Progenitors with epithelial potential (CD34+c-met+; D) were significantly increased at baseline and further increased after G-CSF treatment. The bold line indicates the median value (*P< 0.05; **P < 0.001).
P< 0.001;Figure 1D). From the functional point of view, ESLD patients showed decreased hematopoi-etic function at baseline as demonstrated by the significantly lower concentration of PB CFU-C com-pared with healthy donors (P< 0.001;Figure 1E) although clonogenic progenitors mobilized at day 5 of G-CSF treatment (P< 0.001;Figure 1E).
Interestingly, the frequency of endothelial CD34+/ VEGFR-2 (KDR)+progenitors (Figure 2A), as well as more mature CD34+CD105+and CD34+CD117+ endothelial cells (Figure 2B,C), significantly in-creased 5 days after G-CSF treatment (P< 0.05; P< 0.05 and P < 0.001, respectively). Consistently, a trend toward increased G-CSF mobilized clonogenic endothelial CFU-EPC progenitors was observed despite their low baseline concentration as com-pared to healthy individuals (P< 0.0001;Figure 2D). All the SPC subsets detected into the PB upon G-CSF treatment returned to baseline values by day 14 of G-CSF administration (data not shown).
Taken together, these results corroborated our pre-vious results[25,31]showing that ESLD patients are capable of mobilizing hematopoietic/endothelial SPCs upon G-CSF treatment or tissue damage.
Next, we investigated the concentration of se-lected cytokines in the serum of ESLD patients before and after G-CSF–induced mobilization. At base-line, HGF/CXCL12 and VEGF/MMP-9 were significantly increased (P< 0.01) and decreased (P< 0.05), respectively, compared with normal donors. However, we found that SPC mobilization was asso-ciated with increased serum level of molecules such as MMP-9, HGF and VEGF that facilitate migra-tion, homing and engraftment of BM-derived SPCs to the damaged liver[32–35], whereas, as expected, CXCL12 concentration decreased (Figure 3). Simi-larly to SPC mobilization, serum level of cytokines returned to baseline values by day 14 of G-CSF ad-ministration (data not shown).
CD133+SPCs inhibit activation of liver stromal cells which, in turn, promote the hematopoietic and endothelial phenotype and function of CD133+SPCs To shed light on the mechanism of CD133-dependent transient liver improvement, we performed in vitro experiments mimicking the interplay between CD133+SPCs and a selected cell population of liver
Figure 2. G-CSF mobilizes progenitor cells with endothelial potential in patients with ESLD. Phenotypic and functional analysis of PB cells with endothelial potential from 14 of 17 patients before and after G-CSF treatment and 10 of 17 healthy subjects is shown. After G-CSF treatment (day 5), the median numbers of CD34+KDR+(A), CD34+CD105+(B) and CD34+117+(C) progenitors were signifi-cantly higher than baseline values. At variance, the number of CFU-EPC remained almost unchanged, with only three patients showing increased number after G-CSF treatment. The bold line indicates the median value (*P< 0.05; **P < 0.001; ***P < 0.0001).
microenviroment. To this aim, the human hepatic stellate cell line LX-2 and human primary CD133+ SPCs from ESLD patients enrolled in the cell therapy trial were co-cultured (contact or no-contact cultures; see Methods). Then, we performed a functional and molecular characterization of both cell populations.
Flow cytometry analysis showed that the immuno-magnetically isolated CD133+ SPCs from ESLD patients were 90%± 5 pure and expressed, to a various degree, the following hemopoietic (CD34+, CD90+, CD38+, CD184+) and endothelial markers (CD105+, CD117+, KDR+), as previously described in[25] (Sup-plementary Table SIII).
Next, we analyzed the phenotype of CD133+SPCs from ESLD patients (CD34, CD133, CD90, CD38, CD117, CD105, KDR, CD184, CD31, CD44 ex-pression) and of LX-2 cells (CD44, c-met, CD105, CD90, KDR expression) upon co-culture. We did not find any significant modification up to 72 h of co-culture (data not shown).
CD133+/LX-2 cells mutually modify their molecular phenotype
Next, we investigated the changes in gene expres-sion induced in ESLD patients-derived CD133+SPCs by the co-culture with LX-2 cells by means of the HG-U219 Array (Affymetrix). Microarray data anal-ysis allowed us to identify a list of 291 DEGs with FC ≥1.5 or FC ≤ −1.5 in at least one of the pair-wise com-parisons among CD133+cells grown in co-culture with contact with LX-2 cells and CD133+cells, grown in co-culture without contact with LX-2 cells and CD133+ cells grown alone (Supplementary Figure S1and Sup-plementary Table SIV). The heat map in Figure 4
displays a selection of the DEGs obtained by using a FC≥2 as cutoff in at least one of the pairwise com-parisons listed in Methods. GEP data showed the downregulation of negative regulators of cell cycle progression (CDKN1C and CDKN2C) and the upregulation of FHL2, a positive regulator of cell proliferation in co-cultured (either in contact or
Figure 3. G-CSF treatment induces increased serum levels of multiple cytokines facilitating migration, homing and engraftment. The base-line levels of HGF (A) and CXCL12 (C) were significantly higher, and VEGF (B) and MMP-9 (D) levels significantly lower, in patients with ESLD (n= 12) than in healthy controls (n = 13–16). After G-CSF (day 5), we observed the significant increase of HGF, VEGF and MMP-9. By contrast, the serum level of CXCL12 significantly decreased after G-CSF treatment. The bold line indicates the median value (*P< 0.05; **P < 0.001).
no-contact) CD133+SPCs versus CD133+SPCs cul-tured alone. In addition, only the co-culture with LX-2 cells induced the upregulation of apoptosis inhibi-tors (PIM1, PIM2, BCL6) in CD133+ SPCs
(Figure 4). Taken together, these results indicate that the interaction with hepatic stellate cells LX2 induces a proliferative molecular phenotype of CD133+SPCs. Noteworthy, the co-culture in contact with LX-2 cells
Figure 4. The co-culture with LX-2 cells affects the molecular phenotype of CD133+SPCs from ESLD patients. Hierarchical clustering and heat map analysis for the transcripts differentially expressed (i.e., up-regulated, FC≥2 or down-regulated, FC ≤ −2) in CD133+SPCs from ESLD patients (n= 2) co-cultured (with or without contact) with LX-2 cells as compared with CD133+SPCs cultured alone. The signal-based coloring legend is shown at the bottom of the panel. For each transcript, red bars indicate relatively high signal intensity, while green bars represent lower intensity and black intermediate. The clustering of the samples is indicated by the dendrogram on top. Microarray analysis was performed in duplicate (indicated as ESLD patient 1 and ESLD patient 2).
induced the up-regulation of several genes promot-ing angiogenesis (CXCL3, ID1,VEGFC, SERPINE1, SELP, SCL2A14, VEGFA, CYP1B1) and the downregulation of genes encoding for inhibitors of an-giogenesis (SPON2, vWF) and promoters of inflammation (LTC4S, SPON2) in CD133+ SPCs (Figure 4). Thus, the interaction with LX-2 liver stromal cells induces a pro-angiogenic and anti-inflammatory phenotype in CD133+SPCs, the former requiring cell–cell contact.
We then analyzed the gene expression changes induced in LX-2 cells by the co-culture with ESLD patients-derived CD133+ SPCs by Affymetrix HG-U219 Array. Microarray data analysis highlighted a list of 664 DEGs with FC ≥1.5 or FC ≤ −1.5 in at least one of the pair-wise comparisons (see Methods) among LX-2 cells grown in co-culture with or without contact with CD133+ cells and LX-2 cells grown alone
(Supplementary Figure S2 and Supplementary
Table SV). Of note, microarrays analysis highlighted the down-regulation of genes promoting hepatic stellate cells activation and fibrosis (CEBPB, FOXF1, TLR4, SMAD1, SMAD5, COL1A1, COL3A1, COL6A3, COL12A1, MMP1, MMP2, TIMP1) in co-cultured LX-2 cells in comparison with LX-2 cells grown alone, whereas SMAD6 (a negative regulator of the pro-fibrotic TGF-β signaling pathway) and NFKBIB (an inhibitor of NFkB-mediated proinflammatory signal-ing) were up-regulated upon co-culture. Interestingly, microarray data also unveiled the down-regulation of genes coding for proinflammatory cytokines (interleukin-1B and -8) and profibrotic mediators (fibroblast growth factor-2 and -5) in co-cultured LX-2 cells compared with LX-2 cells alone (Figure 5).
CD133+SPCs inhibit activation of LX-2 cells
During liver fibrosis hepatic stellate cells express fibrosis-related genes such as α-SMA, which pro-motes contractile force and extra-cellular matrix stiffness[36]. Thus, to support functionally GEP data relative to the down-regulation of genes promoting hepatic stellate cells activation and fibrosis in co-cultured LX-2 cells, we tested by flow cytometry the expression of α-SMA. Consistently with GEP data, at the functional level, the percentage of the cells ex-pressingα-SMA was significantly decreased in LX-2 cells after 48 h of co-culture with CD133+ SPCs (Figure 6; contact co-cultured cells versus LX-2 cells alone P< 0.05; no contact co-cultured cells vs LX-2 cells P< 0.05).
LX-2 cells promote viability and proliferation of CD133+SPCs
To investigate whether LX-2 cells may influence the function of CD133+SPCs from ESLD patients, we
studied their viability, apoptosis and hematopoietic function upon co-culture. As shown in Figure 7A
and 7B, CD133+ SPCs from patients with ESLD,
co-cultured in contact or without contact, with LX-2 cells showed, at different time points, the significant increase of survival coupled with the reduction of apoptosis as compared with CD133+ SPCs cultured alone. These data are consistent with microarray data showing the modulation of proliferation/survival genes in CD133+ SPCs co-cultured with LX-2 cells. Of note, the same pattern (viability/apoptosis) was observed when CD133+SPCs from healthy donors were co-cultured with LX-2 cells up to 72 hours (Supplementary Figure S3A,B). In contrast, the viability of LX-2 cells was unaffected
Figure 5. The gene expression changes unveil that patient-derived CD133+ SPCs could restrain LX-2 cell activation. Hierarchical clustering and heat map analysis for the transcripts coding for proteins involved in hepatic stellate cells activation and differentially expressed (i.e., up-regulated, FC≥1.5 or down-regulated, FC≤ −1.5) in LX-2 cells co-cultured (with or without contact) with ESLD patient-derived CD133+SPCs compared with LX-2 cells cultured alone.The signal-based coloring legend is shown at the bottom of the panel. For each transcript, red bars indicate relatively high signal intensity, whereas green bars represent lower intensity and black intermediate. The clustering of the samples is indicated by the dendrogram on top.
by the co-culture with CD133+ SPCs from ESLD patients (Supplementary Figure S4).
Cell cycle analysis showed that CD133+SPCs from ESLD patients and co-cultured in direct contact with LX-2 cells for 48 h increased significantly (P< 0.05) the percentage of cells in the G2/M phase (12± 6.2%) compared with CD133+SPCs alone (6.1± 4.4%) and decreased the percentage of cells in the G0/G1 phase (Figure 8). Similarly, normal CD133+SPCs showed the same kinetic pattern when cocultured with LX-2 cells (data not shown). On the contrary, cell cycle dis-tribution in LX-2 cells was unaffected by the co-culture with CD133+ SPCs (Supplementary Figure S5). Thus, functional assays demonstrated that the interaction with LX-2 liver stromal cells led to en-hanced cycling and reduced apoptosis in CD133+ SPCs.
Moreover, we found that the number of CFU-C from contact co-cultured CD133+SPCs was signifi-cantly increased (P< 0.05), compared with CD133+
SPCs alone, after 72 and 168 h of co-culture (Figure 7C). Interestingly, as shown inSupplementary Figure S6, also the size of either CFU-GM or BFU-E colonies increased, suggesting the higher cycling ca-pacity of lineage-specific cells. In addition, we found that CD133+SPCs co-cultured with LX-2 cells, either with or without contact, showed the significant in-crease of primitive hematopoietic progenitors as determined by LTC-IC assay (P< 0.05 and P < 0.01, respectively;Figure 7D). Parallel experiments dem-onstrated that the clonogenic ability of co-cultured CD133+SPCs from healthy donors was superimposable to that of ESLD patients (Supplementary Figure S3C).
LX-2 cells enhance angiogenetic potential of CD133+SPCs
We next evaluated whether the bi-directional cross-talk between CD133+SPCs and LX-2 cells is regulated by cytokines/growth factors release in the
superna-Figure 6. CD133+SPCs from ESLD patients inhibit activation of LX-2 cells. Flow cytometry analysis ofα-SMA expression in LX-2 cells after in vitro co-cultures, either in contact or without contact, with CD133+SPCs from ESLD patients (n= 13 is shown. Representative dot plots showingα-SMA expression of isotype control (A), LX-2 cells alone (B) and LX-2 cells after in vitro co-cultures with CD133+ SPCs in contact (C) and without contact (D). The gates show the percentage of positive cells using the isotype control as reference. (E) Bar graph: the percentage ofα-SMA positive cells was significantly decreased in LX-2 cells after 48 h of in vitro co-culture with CD133+ SPCs (light gray and white columns) compared with LX-2 cells cultured alone (black column). Results are expressed as mean± SEM (*P< 0.05).
tants of co-cultured cells. As shown in Figure 9A, LX-2 cells produced a time-dependent increasing amount of VEGF independently from the presence of CD133+ SPCs. Interestingly, HGF, SCF, G-CSF and CXCL12 proteins were absent in the supernatants at any condition/time tested as well as the MMP9, MMP2 and TIMP1 (data not shown).
On the basis of the VEGF production results, we functionally tested the angiogenic potential of the su-pernatants of mono/co-cultured cells. We found that supernatants of co-cultured cells stimulated tube for-mation when added to HUVEC cell line. Quantitative analysis of the total area (mm2) of tubes showed that both the supernatant of contact and no contact co-cultures as well as that of LX-2 alone increased the total tubules area indicating that LX-2-produced soluble factors are able to induce angiogenic activity (Figure 9B,C). Interestingly, a slight, but significant, increase in the total tube area was observed when CD133+SPCs were cultured in contact with LX-2 liver cells (Figure 9C).
Discussion
We have recently completed a phase I clinical trial in which we demonstrated the safety and the feasibility of the intrahepatic reinfusion of G-CSF–mobilized, highly purified CD133+SPCs in cirrhotic patients[25]. Similarly to other studies, we were able to show the significant, although transient, improvement of liver function after CD133+SPC injection (as reviewed in Houlihan and Newsome, Nikeghbalian et al., and Margini et al.) [37–40]. Thus, we suggest that stem cell therapy may be an attractive “bridge to trans-plant” approach for ESLD patients.
However, the mechanism(s) through which BM-derived SPCs improve liver function are not fully elucidated. Some insights from animal models support the view that SPCs may exert their activity through the release of cytokines acting through a paracrine mechanism [9,41,42] rather than by trans-differentiation[43,44]or cell fusion[45,46]. In this view, clinical trials did not provide any further data
Figure 7. LX-2 cells promote viability and stimulate the short- and long-term clonogenic ability of patients-derived CD133+SPCs. CD133+ SPCs from patients with ESLD (n= 10) were in vitro co-cultured with LX-2 cells, either in contact or without contact, at different time points and, along with survival/apoptosis, short- (CFU-C) and long-term (LTC-IC) clonogenic activity were monitored. (A and B) The CD133+SPCs from patients with ESLD co-cultured in contact with LX-2 cells showed a significant increase of survival with concomitant reduction of apoptosis as compared with the respective mono-cultured counterparts. Results are expressed as mean fold change compared with CD133+SPCs alone± SEM. (C) The mean number of CFU-C from contact co-cultured CD133+SPCs was significantly increased compared with the mono-cultured counterparts after 72 and 168 h of incubation. Results are expressed as mean± SEM. (D) After 72 h of in vitro co-cultures, either in contact or without contact, CD133+SPCs from ESLD patients showed a significant increase in the long-term clonogenic capacity (LTC-IC number) compared with the mono-cultured counterparts. Results are expressed as mean± SEM (*P < 0.05; **P< 0.001; ***P < 0.0001).
owing to the variability in cell source (e.g., BM or mo-bilized SPCs), route of administration, use of selected/ unselected stem/progenitor cells and lack of liver biopsies before and after cell therapy. In the present study, we characterized CD133+SPCs of ESLD pa-tients previously enrolled in our clinical study, with specific reference to the interaction between CD133+ SPCs and the human hepatic stellate cells LX-2.
Here, we demonstrated that ESLD patients show basal levels of circulating hematopoietic and endo-thelial progenitors comparable with those detected in healthy donors. Consistent with a previous study, pa-tients with chronic cirrhosis had similar numbers of circulating CD133+/CD34+cells compared with healthy people[47]. This finding indicates that severe chronic liver injury, by itself, probably does not stimulate mo-bilization of BM-derived cells into PB. However, both the hematopoietic and the endothelial potential, evalu-ated as CFU-C and CFU-EPC growth, respectively, were impaired in cirrhotic patients.
Conversely, after G-CSF treatment, patients with chronic liver diseases were able to mobilize signifi-cant numbers of functional hemopoietic/endothelial stem/progenitor cells as previously reported by our group and others [31,48–51], although to a lesser extent than healthy control subjects as demon-strated by the higher dose of G-CSF required to mobilize relatively low numbers of SPCs in compar-ison with healthy BM stem cells donors (15 versus 10µg/kg/day). This is probably due to sequestration
of circulating BM-derived stem cells in the spleen and/ or impaired BM function secondary to cirrhosis and/ or infection-related cirrhosis.
We have demonstrated that SPC mobilization was associated with increased serum levels of molecules such as VEGF and MMP-9, which are involved in the process of cell migration, matrix degradation and an-giogenesis. In addition, in line with the increased number of circulating CD34+CXCR4+progenitors, the serum levels of SDF-1 were significantly reduced after G-CSF administration, likely due to CXCR4 binding. Moreover, the increased number of circulating pro-genitors co-expressing CXCR4 suggests increased trafficking and, perhaps, homing and engraftment to the damaged liver as shown in animal models[33]. Thus, it may well be that BM-derived SPCs with hematopoietic/endothelial potential contribute to im-proving tissue function in case of acute or chronic liver injury.
As previously reported[25], performing liver bi-opsies in ESLD patients enrolled in our trial was considered not ethically justified; indeed, the histo-pathology examination of the explanted liver in five patients undergoing OLT did not reveal the pres-ence of residual CD133+SPCs, and the fibrotic and angiogenetic pattern was not significantly different, at first glance, from control samples (i.e., OLT recipi-ents who did not receive cell therapy). Whether this finding was due to the timing of this retrospective anal-ysis (performed at a median time of 109 days after cell therapy)[25], which may have missed some early biological modifications induced by the reinfused CD133+SPCs, remains a matter of discussion.
We therefore investigated, in vitro, the putative mechanisms of interaction between CD133+SPCs and the human hepatic stellate cells LX2. In this regard, molecular and functional assays unveiled the inter-play between the two cell populations.
The proliferation and activation of hepatic stel-late cells play an essential role during the process of liver fibrogenesis. Therefore, reversal or suppression of hepatic stellate cells activation is considered as a main target for therapeutic approaches addressing hepatic fibrosis. In this study, we used the immortal-ized LX-2 cell line, which has been shown to retain the features of activated hepatic stellate cells, includ-ingα-SMA expression[28]and therefore are suitable to give an indication of the responses of hepatic stel-late cells in vivo[52]. Of note, despite the fact that CD133/LX2 co-culture experiments do not dupli-cate the complexity of the liver microenvironment in vivo and other cellular elements should probably be taken into account as playing a role in the hepatic mi-croenvironment, LX-2 cells are nonetheless considered a valuable tool for analysis of human hepatic fibro-sis.We performed LX-2 co-culture experiments to gain
Figure 8. LX-2 cells promote cell proliferation in CD133+SPCs from patients with ESLD. CD133+SPCs from patients with ESLD (n= 10) were in vitro co-cultured for 48 h with LX-2 cells, either in contact or without contact, and their distribution in the differ-ent phases of cell cycle was assessed by PI staining. Cells cultured in direct contact presented a slight, but significant, increase in the percentage of cells in phase G2-M, whereas the percentage of cells in G0-G1 was decreased. Data are represented as mean± SEM (*P< 0.05).
insight into the potential effects of CD133+ on key hepatic fibrogenic cells.
Indeed, the profiling of the gene expression changes associated with the co-culture of CD133+and LX-2 cells provided important insights into the cross-talk between these populations by highlighting the up-regulation of antiapoptotic and pro-angiogenic genes and the down-regulation of inhibitors of cell cycle pro-gression and angiogenesis in CD133+SPCs, together with the down-regulation of genes promoting hepatic stellate cell activation and fibrosis in LX-2 cells. In line with these data, functional assays showed that CD133+ SPCs inhibit LX-2 cells activation by de-creasing α-SMA expression. In turn, LX-2 cells promoted CD133+SPCs proliferation/survival and
he-matopoietic function, suggesting supportive effects of human hepatic stellate cells on hemopoiesis in ESLD patients.These data are consistent with a recently pub-lished paper demonstrating that the microevironmental cues of diseased liver are able to influence the cell cycle and the hemopoietic programming fate of liver resi-dent CD34+cells[53]. In addition, LX-2 cells, through VEGF production, stimulated, in vitro, CD133+ cell-driven angiogenesis. This finding may be important in ESLD patients, in whom endothelial cell dysfunc-tion is a central mechanism in the pathogenesis of portal hypertension. In addition, although VEGF is unequivocally a pro-fibrogenic mediator and has a well established role in endothelial cells survival/proliferation, it may have also a role in fibrosis resolution by
po-Figure 9. LX-2 cells enhance the angiogenetic potential of CD133+SPCs from ESLD patients. (A) ELISA analysis of VEGF levels in the supernatants of co-cultures of CD133+SPCs from ESLD patients (n= 10) and LX-2 cells after 24, 48, 72 and 168 h of incubation. LX-2 cells synthesize a time-dependent increasing amount of VEGF; however, this production was not significantly modified by the presence of CD133+SPCs. Results are expressed as mean± SEM. (B and C) Supernatants (n = 10) of mono-cultured CD133+SPCs from ESLD patients/LX-2 cells or co-cultured cells (in contact and no contact) after 1 week of incubation have been added to HUVEC cell line. Su-pernatants of co-cultured cells stimulate tubules formation (B,d) and increased the total tubules area (C) compared with mono-cultured CD133+SPCs. However, a significant difference was observed only with the contact co-cultures supernatants (C). Results are expressed as mean± SEM (*P < 0.05).
tentially increasing the recruitment of “resolution” macrophages, which have an important role in fibrosis reversal[54]. Moreover,VEGF may accelerate fibrolysis via CXCL9/MMP13 axis[55].
Overall, the present study demonstrates that there is bi-directional cross-talk between CD133+SPCs from patients with ESLD and the human hepatic stellate LX-2 cells. This interaction promotes positive func-tional modifications in both cellular types. We can therefore hypothesize a scenario in which, after cell therapy, the activated hepatic stellate cells of the liver microenvironment could stimulate the hemopoietic and endothelial potential of the CD133+SPCs, which, in turn, might contribute to switch off the activation of the hepatic stellate cells. These findings open new per-spectives for cell therapy in patients with advanced chronic liver diseases.
Conclusions
We demonstrated that mobilized CD133+SPCs from patients with ESLD may be a good candidate for cell therapy in cirrhotic patients given their potential to participate/trigger reparative processes. In turn, hepatic stellate cells, key players in the liver microenviron-ment, enhance the endothelial and hemopoietic potential of the mobilized cells from ESLD patients. This finding highlights the potential reparative role of CD133+SPCs for severely diseased liver.
Acknowledgments
This work was supported by a grant from Research Program Region-University 2007/2009 1b “Regen-erative Medicine” area, Regione Emilia-Romagna, Italy.
Contributions: RML, LC, and RM—conception
and design, manuscript writing, final approval of manuscript. DS, MC, CA, EB, LF, MB, FG, TM, EM—collection and assembly of data, data analysis and interpretation, final approval of manuscript. AC, PA, RG,VG, PC—data analysis and interpretation and final approval of manuscript.
Disclosure of interest: The authors have no
com-mercial, proprietary, or financial interest in the products or companies described in this article.
References
[1]Lim YS, Kim WR. The global impact of hepatic fibrosis and end-stage liver disease. Clin Liver Dis 2008;12:733–46. [2]Trautwein C, Friedman SL, Schuppan D, Pinzani M. Hepatic
fibrosis: concept to treatment. J Hepatol 2015;62(1 Suppl.): S15–24.
[3]Schuppan D, Afdhal NH. Liver cirrhosis. Lancet 2008; 371:838–51.
[4]Wake K. Perisinusoidal stellate cells (fat-storing cells, interstitial cells, lipocytes), their related structure in and around the liver
sinusoids, and vitamin A-storing cells in extrahepatic organs. Int Rev Cytol 1980;66:303–53.
[5] Lee YA, Wallace MC, Friedman SL. Pathobiology of liver fibrosis: a translational success story. Gut 2015;64:830–41. [6] Pinzani M, Macias-Barragan J. Update on the pathophysiology
of liver fibrosis. Expert Rev Gastroenterol Hepatol 2010;4:459–72.
[7] Caraceni P, Giannone F, Catani L, Talarico S, Pertosa AM, Domenicali M, et al. Effects of granulocyte colony stimulating-factor in a rat model of acute liver injury. Dig Liver Dis 2007;39:943–51.
[8] Domenicali M, Caraceni P, Giannone F, Baldassarre M, Lucchetti G, Quarta C, et al. A novel model of CCl4-induced cirrhosis with ascites in the mouse. J Hepatol 2009;51:991–9. [9] Behbahan IS, Keating A, Gale RP. Concise review: bone marrow autotransplants for liver disease? Stem Cells 2013;31:2313–29.
[10] Vainshtein JM, Kabarriti R, Mehta KJ, Roy-Chowdhury J, Guha C. Bone marrow-derived stromal cell therapy in cirrhosis: clinical evidence, cellular mechanisms, and implications for the treatment of hepatocellular carcinoma. Int J Radiat Oncol Biol Phys 2014;89:786–803.
[11] Wang K, Chen X, Ren J. Autologous bone marrow stem cell transplantation in patients with liver failure: a meta-analytic review. Stem Cells Dev 2015;24:147–59.
[12] Bailey AS, Jiang S, Afentoulis M, Baumann CI, Schroeder DA, Olson SB, et al. Transplanted adult hematopoietic stems cells differentiate into functional endothelial cells. Blood 2004; 103:13–19.
[13] Loges S, Fehse B, Brockmann MA, Lamszus K, Butzal M, Guckenbiehl M, et al. Identification of the adult human hemangioblast. Stem Cells Dev 2004;13:229–42.
[14] Handgretinger R, Kuçi S. CD133-positive hematopoietic stem cells: from biology to medicine. Adv Exp Med Biol 2013;777:99–111.
[15] Handgretinger R, Gordon PR, Leimig T, Chen X, Buhring HJ, Niethammer D, et al. Biology and plasticity of CD133+ hematopoietic stem cells. Ann NY Acad Sci 2003;996:141–51. [16] Mizrak D, Brittan M, Alison M. CD133: molecule of the
moment. J Pathol 2008;214:3–9.
[17] Boxall SA, Cook GP, Pearce D, Bonnet D, El-Sherbiny YM, Blundell MP, et al. Haematopoietic repopulating activity in human cord blood CD133+ quiescent cells. Bone Marrow Transplant 2009;43:627–35.
[18] Brunet de la Grange P, Vlaski M, Duchez P, Chevaleyre J, Lapostolle V, Boiron JM, et al. Long-term repopulating hematopoietic stem cells and “side population” in human steady state peripheral blood. Stem Cell Res 2013;11:625–33. [19] Gehling UM, Willems M, Dandri M, Petersen J, Berna M, Thill M, et al. Partial hepatectomy induces mobilization of a unique population of haematopoietic progenitor cells in human healthy liver donors. J Hepatol 2005;43:845–53. [20] am Esch JS 2nd, Knoefel WT, Klein M, Ghodsizad A, Fuerst
G, Poll LW, et al. Portal application of autologous CD133+ bone marrow cells to the liver: a novel concept to support hepatic regeneration. Stem Cells 2005;23:463–70.
[21] Fürst G, Schulte am Esch J, Poll LW, Hosch SB, Fritz LB, Klein M, et al. Portal vein embolization and autologous CD133+bone marrow stem cells for liver regeneration: initial experience. Radiology 2007;243:171–9.
[22] Stamm C, Westphal B, Kleine HD, Petzsch M, Kittner C, Klinge H, et al. Autologous bone-marrow stem-cell transplantation for myocardial regeneration. Lancet 2003; 361:45–6.
[23] Klein HM, Ghodsizad A, Borowski A, Saleh A, Draganov J, Poll L, et al. Autologous bone marrow-derived stem cell therapy in combination with TMLR. A novel therapeutic
option for endstage coronary heart disease: report on 2 cases. Heart Surg Forum 2004;7:E416–19.
[24] Lemoli RM, Catani L, Talarico S, Loggi E, Gramenzi A, Baccarani U, et al. Mobilization of bone marrow-derived hematopoietic and endothelial stem cells after orthotopic liver transplantation and liver resection. Stem Cells 2006;24:2817– 25.
[25] Andreone P, Catani L, Margini C, Brodosi L, Lorenzini S, Sollazzo D, et al. Reinfusion of highly purified CD133(+) bone marrow-derived stem/progenitor cells in patients with end-stage liver disease: a phase I clinical trial. Dig Liver Dis 2015; 47:1059–66.
[26] Mancuso P, Antoniotti P, Quarna J, Calleri A, Rabascio C, Tacchetti C, et al. Validation of a standardized method for enumerating circulating endothelial cells and progenitors: flow cytometry and molecular and ultrastructural analyses. Clin Cancer Res 2009;15:267–73.
[27] Hill JM, Zalos G, Halcox JP, Schenke WH, Waclawiw MA, Quyyumi AA, et al. Circulating endothelial progenitor cells, vascular function, and cardiovascular risk. N Engl J Med 2003;348:593–600.
[28] Xu L, Hui AY, Albanis E, Arthur MJ, O’Byrne SM, Blaner WS, et al. Human hepatic stellate cell lines, LX-1 and LX-2: new tools for analysis of hepatic fibrosis. Gut 2005;54:142–51. [29] Lemoli RM, Tafuri A, Fortuna A, Petrucci MT, Ricciardi MR, Catani L, et al. Cycling status of CD34+ cells mobilized into peripheral blood of healthy donors by recombinant human granulocyte colony-stimulating factor. Blood 1997;89:1189–96. [30] Bianchi E, Bulgarelli J, Ruberti S, Rontauroli S, Sacchi G, Norfo R, et al. MYB controls erythroid versus megakaryocyte lineage fate decision through the miR-486-3p-mediated downregulation of MAF. Cell Death Differ 2015;22:1906–21. [31] Lorenzini S, Isidori A, Catani L, Gramenzi A, Talarico S, Bonifazi F, et al. Stem cell mobilization and collection in patients with liver cirrhosis. Aliment Pharmacol Ther 2008;27:932–9.
[32] Dalakas E, Newsome PN, Harrison DJ, Plevris JN. Hematopoietic stem cell trafficking in liver injury. FASEB J 2005;19:1225–31.
[33] Kollet O, Shivtiel S, Chen YQ, Suriawinata J, Thung SN, Dabeva MD, 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;112:160–9.
[34] Armbrust T, Batusic D, Xia L, Ramadori G. Early gene expression of hepatocyte growth factor in mononuclear phagocytes of rat liver after administration of carbon tetrachloride. Liver 2002;22:486–94.
[35] Petit I, Szyper-Kravitz M, Nagler A, Lahav M, Peled A, Habler L, et al. G-CSF induces stem cell mobilization by decreasing bone marrow SDF-1 and up-regulating CXCR4. Nat Immunol 2002;3:687–94. Erratum in: Nat Immunol 2002;3:787.
[36] Hinz B, Phan SH, Thannickal VJ, Prunotto M, Desmoulière A, Varga J, et al. Recent developments in myofibroblast biology: paradigms for connective tissue remodeling. Am J Pathol 2012;180:1340–55.
[37] Houlihan DD, Newsome PN. Critical review of clinical trials of bone marrow stem cells in liver disease. Gastroenterology 2008;135:438–50.
[38] Nikeghbalian S, Pournasr B, Aghdami N, Rasekhi A, Geramizadeh B, Hosseini Asl SM, et al. Autologous transplantation of bone marrow-derived mononuclear and CD133(+) cells in patients with decompensated cirrhosis. Arch Iran Med 2011;14:12–27.
[39] Margini C, Vukotic R, Brodosi L, Bernardi M, Andreone P. Bone marrow derived stem cells for the treatment of end-stage liver disease. World J Gastroenterol 2014;20:9098–105.
[40] Terai S, Takami T, Yamamoto N, Fujisawa K, Ishikawa T, Urata Y, et al. Status and prospects of liver cirrhosis treatment by using bone marrow-derived cells and mesenchymal cells. Tissue Eng Part B Rev 2014;20:206–10.
[41] Sakaida I, Terai S, Yamamoto N, Aoyama K, Ishikawa T, Nishina H, et al. Transplantation of bone marrow cells reduces CCl4-induced liver fibrosis in mice. Hepatology 2004;40: 1304–11.
[42] Wang L, Wang X, Xie G, Wang L, Hill CK, DeLeve LD. Liver sinusoidal endothelial cell progenitor cells promote liver regeneration in rats. J Clin Invest 2012;122:1567–73. [43] Terai S, Sakaida I, Yamamoto N, Omori K, Watanabe T,
Ohata S, et al. An in vivo model for monitoring trans-differentiation of bone marrow cells into functional hepatocytes. J Biochem 2003;134:551–8.
[44] Jang YY, Collector MI, Baylin SB, Diehl AM, Sharkis SJ. Hematopoietic stem cells convert into liver cells within days without fusion. Nat Cell Biol 2004;6:532–9.
[45] Wang X, Willenbring H, Akkari Y, Torimaru Y, Foster M, Al-Dhalimy M, et al. Cell fusion is the principal source of bone-marrow-derived hepatocytes. Nature 2003;422:897– 901.
[46] Vassilopoulos G, Wang PR, Russell DW. Transplanted bone marrow regenerates liver by cell fusion. Nature 2003;422:901– 4.
[47] Ghanem LY, Nosseir MM, Lotfi AA, Mohamed AS, Ibrahim RA, Hassanein MH, et al. Hematopoietic stem cell mobilization into the peripheral circulation in patients with chronic liver diseases. J Dig Dis 2012;13:571–8.
[48] Gaia S, Smedile A, Omedè P, Olivero A, Sanavio F, Balzola F, et al. Feasibility and safety of G-CSF administration to induce bone marrow–derived cells mobilization in patients with end stage liver disease. J Hepatol 2006;45:13–19.
[49] Di Campli C, Zocco MA, Saulnier N, Grieco A, Rapaccini G, Addolorato G, et al. Safety and efficacy profile of G-CSF therapy in patients with acute on chronic liver failure. Dig Liver Dis 2007;39:1071–6.
[50] Spahr L, Lambert JF, Rubbia-Brandt L, Chalandon Y, Frossard JL, Giostra E, et al. Granulocyte-colony stimulating factor induces proliferation of hepatic progenitors in alcoholic steatohepatitis: a randomized trial. Hepatology 2008;48:221–9. [51] Garg V, Garg H, Khan A, Trehanpati N, Kumar A, Sharma BC, et al. Granulocyte colony-stimulating factor mobilizes CD34(+) cells and improves survival of patients with acute-on-chronic liver failure. Gastroenterology 2012;142:505–12. [52] Van de Bovenkamp M, Groothuis GMM, Meijer DKF, Olinga P. Liver fibrosis in vitro: cell culture models and precision-cut liver slices. Toxicol In Vitro 2007;21:545–57.
[53] Velazquez VM, Uebelhoer LS, Thapa M, Ibegbu CC, Courtney C, Bosinger SE, et al. Systems biological analyses reveal the hepatitis C virus (HCV)-specific regulation of hematopoietic development. Hepatology 2015;61:843–56. [54] Yang L, Kwon J, Popov Y, Gajdos GB, Ordog T, Brekken RA,
et al. Vascular endothelial growth factor promotes fibrosis resolution and repair in mice. Gastroenterology 2014; 146:1339–50, e1. doi:10.1053/j.gastro.2014.01.061. [55] Park S, Kim JW, Kim JH, Lim CW, Kim B. Differential roles
of angiogenesis in the induction of fibrogenesis and the resolution of fibrosis in liver. Biol Pharm Bull 2015; 38(7):980–5.
Appendix: Supplementary material
Supplementary data to this article can be found online atdoi:10.1016/j.jcyt.2017.08.001.
