Original Basic ScienceçLiver
Extracellular Vesicles from Human Liver Stem Cells Reduce Injury in an Ex Vivo Normothermic Hypoxic Rat Liver Perfusion Model
Federica Rigo, MD,1 Nicola De Stefano, BMSc,1 Victor Navarro-Tableros, MD, PhD,2 Ezio David, MD,3
Giorgia Rizza, MD,1 Giorgia Catalano, MD,1 Nicholas Gilbo, MD,1 Francesca Maione, MD,1 Federica Gonella, MD,1 Dorotea Roggio, MSc,1 Silvia Martini, MD,4 Damiano Patrono, MD, PhD,1 Mauro Salizzoni, MD,1
Giovanni Camussi, MD,5 and Renato Romagnoli, MD1
Background. The gold standard for organ preservation before transplantation is static cold storage, which is unable to fully pro- tect suboptimal livers from ischemia/reperfusion injury. An emerging alternative is normothermic machine perfusion (NMP), which permits organ reconditioning. Here, we aimed to explore the feasibility of a pharmacological intervention on isolated rat livers by using a combination of NMP and human liver stem cells-derived extracellular vesicles (HLSC-EV). Methods. We established an ex vivo murine model of NMP capable to maintain liver function despite an ongoing hypoxic injury induced by hemodilution. Livers were perfused for 4 hours without (control group, n = 10) or with HLSC-EV (treated group, n = 9). Bile production was quantified; perfusate samples were collected hourly to measure metabolic (pH, pO2, pCO2) and cytolysis parameters (AST, alanine aminotransferase, lactate dehydrogenase). At the end of perfusion, we assessed HLSC-EV engraftment by immunofluorescence, tissue injury by histology, apoptosis by terminal deoxynucleotidyl transferase dUTP nick-end labeling assay, tissue hypoxia-inducible factor 1-α, and transforming growth factor-beta 1 RNA expression by quantitative reverse transcription-polymerase chain reaction. Results. During hypoxic NMP, livers were able to maintain homeostasis and produce bile. In the treated group, AST (P = 0.018) and lactate dehydrogenase (P = 0.032) levels were significantly lower than those of the control group at 3 hours of perfusion, and AST levels persisted lower at 4 hours (P = 0.003). By the end of NMP, HLSC-EV had been uptaken by hepatocytes, and EV treatment significantly reduced histological damage (P = 0.030), apoptosis (P = 0.049), and RNA overexpression of hypoxia-inducible factor 1-α (P < 0.0001) and transforming growth factor-beta 1 (P = 0.014). Conclusions. HLSC-EV treatment, even in a short-duration model, was feasible and effectively reduced liver injury during hypoxic NMP.
(Transplantation 2018;102: e205–e210)
Liver transplantation (LT) is currently the only successful therapy for end-stage liver disease, but it is limited by the discrepancy between transplant candidates and available
organs. Several new strategies are under active investigation
Received 5 August 2017. Revision received 27 December 2017. Accepted 27 December 2017.
1 General Surgery 2U, Liver Transplantation Center, AOU Città della Salute e della Scienza di Torino, University of Turin, Turin, Italy.
2 2i3T, Società per La Gestione Dell’incubatore Di Imprese e Per Il Trasferimento Tecnologico Dell’Università degli Studi di Torino, Scarl., Molecular Biotechnology Center (MBC), Turin, Italy.
3 Pathology Unit, Molinette Hospital, AOU Città della Salute e della Scienza di Torino, Turin, Italy.
4 Gastrohepatology Unit, Molinette Hospital, AOU Città della Salute e della Scienza di Torino, Turin, Italy.
5 Department of Medical Sciences, University of Turin, Turin, Italy.
F.R., N.D.S., and V.N.-T. contributed equally to this work.
G.C. and R.R. are senior authors and contributed equally to this work.
Ricerca Locale Ex 60%, University of Turin—year 2015 and year 2016.
Conflictofinterest: G.C.isnamedasinventorinpatentsrelatedtotheregenerativeeffects of HL SC-derived extracellular vesicles. All other authors declare no conflict of interest.
Transplantation ■ May 2018 ■ Volume 102 ■ Number 5
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.
for their potential to ameliorate the transplant process and
1 expand the donor pool.
With the advent of modern preservation solutions, static cold storage (SCS) has become the standard for organ preser- vation. However, along with donor pool expansion, it has emerged that SCS is imperfect in preserving suboptimal or- gans from the so-called extended criteria donors.2,3
F.R., N.D.S., and V.N. performed the experiments, analyzed the data, and wrote the article. E.D. and D.R. contributed in histopathological and biomolecular analyses. G.R., G.C., N.G., F.M., F.G., S.M., and D.P. participated in surgical procedures and data analysis. M.S., G.C., and R.R. gave intellectual input for study design, analyzed the data, and revised the article.
Correspondence: Renato Romagnoli, MD, General Surgery, 2U-Liver Transplantation Center, AOU Città della Salute e della Scienza di Torino, Molinette Hospital, Department of Surgical Sciences, University of Turin, Corso Dogliotti 14, 10126 Torino, Italy. (renato.romagnoli@unito.it).
Supplemental digital content (SDC) is available for this article. Direct URL citations appear in the printed text, and links to the digital files are provided in the HTML text of this article on the journal’s Web site (www.transplantjournal.com).
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved. ISSN: 0041-1337/18/10205-e205
DOI: 10.1097/TP.0000000000002123
www.transplantjournal.com e205
Downloaded from http://journals.lww.com/transplantjournal by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC4/OAVpDDa8K2+Ya6H515kE= on 02/14/2022

e206 Transplantation ■ May 2018 ■ Volume 102 ■ Number 5
www.transplantjournal.com
Normothermic machine perfusion (NMP) is an innovative alternative to SCS. Through an extracorporeal perfusion cir- cuit, it keeps the organ at physiological temperature while continuously providing oxygen and nutrients.4,5 At variance with SCS, NMP permits real-time monitoring of biomarkers and hemodynamic perfusion parameters, potentially allowing pretransplant organ viability assessment.6,7 Animal studies demonstrated the superiority of NMP compared with SCS,8-12 and a phase I study established safety and feasibility of this technique in humans.13 Furthermore, NMP gives the unique opportunity to treat the liver with pharmacological interven- tions during preservation.14
Human liver stem-like cells (HLSC) are a population of stem-like cells resident in adult liver, which may be useful in regenerative medicine because they are easily expandable and have multiple differentiating capabilities.15 They express several mesenchymal (CD29, CD73, CD44, CD90) and em- bryonic markers (Nanog, Sox2, Musashi1, Oct 3/4, Pax2), but not hematopoietic ones. Moreover, HLSC express al- bumin, α-fetoprotein, and cytokeratin 18, supporting their partial hepatic commitment.15 We previously showed that HLSC could restore hepatic function and improve survival in a model of fulminant liver failure in immunodeficient mice16 and could generate hepatic-like tissue when seeded in liver acellular scaffolds.17
Stem cell-derived extracellular vesicles (EV) are a heteroge- neous population of cell-secreted vesicles which play a pivotal role in cell-to-cell communication; they carry specific subsets of mRNA and miRNA that regulate the behavior of target cells.18 Many in vitro and in vivo studies demonstrated the therapeutic potential of EV, showing that their benefi- cial effect is comparable to that of the stem cells they derive from.19-22
In this study, we set up a short-duration model of ex vivo isolated rat liver NMP in which oxygen delivery was kept suboptimal through low hematocrit, to investigate whether adding HLSC-derived EV (HLSC-EV) to the perfusate would result in (i) their rapid uptake by the liver, and (ii) an appre- ciable reduction of hypoxic tissue injury.
MATERIALS AND METHODS
Isolation, Characterization, and Culture of HLSC
HLSC were isolated from human cryopreserved hepatocytes obtained from Lonza Bioscience (Basel, Switzerland) and char- acterized.15,16 HLSC were cultured in a medium containing a 3:1 proportion of α-minimum essential medium and endothelial cell basal medium-1, supplemented with L-glutamine, 2 mM; penicillin, 100 U/mL; streptomycin, 100 μg/mL; and 10% fetal calf serum (α-minimum essential medium/endothelial-cell basal medium/fetal calf serum), and maintained in a humidified 5% CO2 incubator at 37°C. HLSC at passages 5 to 8 and 80% confluence were used in all experiments.
Isolation of DIL-Labeled HLSC-EV
HLSC were previously labeled with the DIL Stain (1,1′-Dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine per- chlorate “DiI”; DiIC18(3)) (Molecular Probes Life Tech- nology, New York, NY)23 and cultured. Once at 80% confluence, cells were starved overnight in Roswell Park Memo- rial Institute/penicillin-streptomycin medium deprived of FCS at 37°C in a humidified incubator with 5% CO2. Supernatants
were collected, centrifuged at 3500 rpm for 15 minutes to remove cell debris, and submitted to ultracentrifugation at 100000g for 2 hours at 4°C (Beckman Coulter Optima L-90K, Fullerton, CA). EV were collected and used fresh or stored at −80°C after resuspension in RPMI plus 1% di- methyl sulfoxide. No difference in biological activity was ob- served between fresh and stored EV (data not shown). Quantification and size distribution of EV were performed using NanoSight LM10 (NanoSight Ltd, Minton Park, UK).21
Animals
Animal studies were approved by the Ethic Committee of the Italian Institute of Health (N.1164/2015-PR; Istituto Superiore di Sanità) and conducted in accordance with the National Institute of Health Guide for Care and Use of Lab- oratory Animals. Male Wistar rats aged 8 to 12 weeks (200-250 g weight) obtained from Charles Rivers (Italy) were maintained on standard conditions and provided with food and water ad libitum.
Isolation of Rat Livers
Animals were anesthetized through an intramuscular in- jection of Zolazepam (0.2 mg/kg) and Xilazyne (16 mg/kg). After intraperitoneal heparin (1250 U) administration, a midline laparotomy was performed. The bowel was retracted to expose the liver and the hepatic pedicle. The common bile duct was cannulated with a 22-G cannula, the hepatic artery was ligated, and the portal vein (PV) was cannulated with an 18-G cannula. After sternotomy, the heart was opened to ex- sanguinate the animal. The liver was flushed with 40 mL of cold Celsior solution (IGL, France) through PV cannula. After perfusion, the liver was removed by transecting its ligaments, PV, common bile duct, and suprahepatic and infrahepatic in- ferior vena cava. It was then weighed, placed into a Petri dish filled with ice-cold Celsior solution and transported to the perfusion room.
NMP
The NMP circuit was made up of a perfusion chamber, a peristaltic pump, an oxygenator (Hollow Fiber Oxygenator D150, Hugo Sachs Elektronik) and a bubble trap. A Trans- ducer Amplifier Module (TAM-D) and a Servo Controller (SCP Type 704) (Harvard Apparatus, Hugo Sachs Elektronik) allowed constant pressure perfusion (8-10 mm Hg) and continuous monitoring of perfusate flow (1.1- 1.3 mL/min per gram of liver). The perfusion chamber was linked to a warming circuit made of a thermocirculator (Lauda) with temperature set at 37°C.
The perfusion solution consisted of phenol red-free Williams E Medium, supplemented with 11.6 mM glucose, 50 U/mL penicillin, 50 μg/mL streptomycin, 5 mM L-glutamine (all from Sigma), 1 U/mL insulin (Lilly, Italy), 1 U/mL heparin (PharmaTex, Italy), named complete Williams Medium. An isovolemic hemodilution was performed by adding 20 mL of fresh rat blood to 50 mL of complete Williams Medium, thus obtaining a mean hematocrit of 9.67 ± 0.66%. This low hemoglobin content (roughly half of what is usually em- ployed in similar perfusion settings)11 was devised to provide suboptimal oxygen delivery and induce a limited but progres- sive hypoxic injury.24 The perfusion solution was supple- mented with 99% oxygen, and 2 mEq of bicarbonate was added as a pH buffer.
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.

© 2018 Wolters Kluwer
Rigo et al e207
Livers were flushed with complete Williams Medium, con- nected to the NMP circuit, and perfused at 37°C through PV cannula under controlled pressure and flow conditions dur- ing 4 hours. Heparin (500 U) was added hourly during perfu- sion. Bile duct cannula was connected to a reservoir to quantify bile production.
Rat livers undergoing NMP were divided into 2 groups: NMP alone (NMP group, n = 10) and NMP enriched with HLSC-EV (NMP + HLSC-EV group, n = 9). In the NMP + HLSC-EV group, a single dose of 5
108 HLSC-EV/g of liver was added to the circuit 15 minutes after perfusion start.22
Perfusate Analysis
Blood gas analysis (ABL 705L Radiometer, Copenhagen) was performed hourly during perfusion on inflow and out- flow perfusate samples, testing pO2, pCO2, and pH. In addi- tion, outflow samples (1 mL) were collected every 60 minutes and centrifuged at 3500 rpm for 10 minutes at 4°C; the cell-free supernatants were stored at −80°C until aspartate aminotransferase (AST), alanine aminotransferase (ALT), and lactate dehydrogenase (LDH) levels were assessed by the Biochemistry Laboratory.
Immunofluorescence and Histological Analysis
To analyze HLSC-EV uptake, 2 liver lobes were included in cryostat embedding medium (Killik, Bio-Optica) and frozen at −80°C; then serial slides were cut (3-5 μm) by a cryostat and fixed in acetone. After rinsing in phosphate-buffered saline (PBS), slides were incubated with a blocking solution of 3% bovine serum albumin and 0.1% Tween-20 for 1 hour at room temperature, and then incubated overnight at 4°C with a primary antibody directed against rat cytochrome P450-4A (Invitrogen) at 1:50 dilution. After washing in PBS, the slides were incubated with the secondary antibody Alexa Fluor 488 (Invitrogen) for 1 hour at room temperature. Finally, nuclei were stained with 4′,6-diamidino-2-phenylindole and, after washing in PBS, slides were mounted with Fluoromount (Sigma). Microscopy analysis was performed using a Cell Ob- server SD-ApoTome laser-scanning system (Carl Zeiss).
Two other liver lobes were formalin-fixed and paraffin- embedded, then sections were obtained from most macro- scopically altered areas. After hematoxylin-eosin staining, severity of histological damage was blindly scored by an experienced liver pathologist (E.D.) according to modified Suzuki criteria, by which sinusoidal congestion, hepatocyte necrosis, and ballooning degeneration are graded from 0 to 4 points and the final score is the sum of the grades for each item.25 Apoptosis was quantified by terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL assay).26 Positive and negative deoxynucleotidyl transferase dUTP nick-end labeling cells were blindly counted on 20 microscopic fields at 200
magnification, then the apoptosis index was calcu- lated as the ratio between the number of positive cells and the number of total cells.
Quantitative Reverse Transcription Polymerase Chain Reaction
RNA was extracted from all paraffin embedded liver sam- ples using RecoverAll Total Nucleic Acid Isolation Kit for FFPE (Invitrogen), then quantified spectrophotometrically (NanoDrop 2000; Thermo Scientific). The subsequent analy- ses could be performed on more than half of the samples of both NMP and NMP + HLSC-EV groups. High-capacity
cDNA reverse transcription (RT) Kit (Applied Biosystems) was used to synthesize cDNA from 400 ng/μL of RNA. Then, a Real-Time Polymerase Chain Reaction (PCR) (Applied Biosystems) was performed on triplicate cDNA samples ac- cording to the chemistry of Power SYBR Green PCR Master Mix (Applied Biosystems), using the following primers of hypoxia-induced markers27: (a) hypoxia-inducible factor 1-α (HIF-1α), forward 5′-TGTGTGTGAATTATGTTGTAA GTGGTATT-3′, reverse 5′-GTGAACAGCTGG GTCATT TTCAT-3′; (b) transforming growth factor-beta 1 (TGF- β1), forward 5′-TTGCCCTCTACACCAACACAA-3’, reverse 5′-GGCTTGCGACCCACGTAGTA-3′. The primer of the house-keeping gene actin β was forward 5′-ACCGTGAAA AGATGACCCAGAT-3′, reverse 5′-CACAGCCTGGATGGC TACGT-3′. Comparative ΔΔCt method was used to calculate the expression levels of the genes of interest normalized to the housekeeping gene expression. One liver explant from a healthy rat (sham) was used as reference sample.
Statistical Analysis
Data are expressed as mean ± standard error of the mean (SEM). Student t test or analysis of variance with Sidak’s multicomparison test was used where appropriate (GraphPad Prism, version 6.00, USA). A P value less than 0.05 was con- sidered as statistically significant.
RESULTS
After standardization, surgical procedures showed low variability between groups. Before perfusion, liver weight, tissue temperature, warm ischemia, and total ischemia time were comparable between NMP group and NMP + HLSC-EV group. During perfusion, all organs were able to self-regulate pH, pO2, and pCO2, maintaining stable flow and resistance (Table S1, SDC, http://links.lww.com/TP/B532).
Immunofluorescence analysis (Figure 1) revealed the pres- ence of DIL-stained HLSC-EV in treated livers by the end of the experiment (Figure 1C). The colocalization of HLSC-EV with the hepatocyte marker cytochrome P450-4A demon- strated their internalization within hepatocytes (Figure 1D).
Histological analysis evidenced overt damage in the NMP livers, characterized by areas of necrosis and apopto- sis, which were reduced in the NMP + HLSC-EV livers (Figures 1E and G). Tissue injury quantification by the Suzuki score (Figure 1F) showed a significant decrease in the NMP + HLSC-EV livers (3.9 ± 0.4) compared with the NMP ones (5.7 ± 0.6) (P = 0.030). In particular, livers treated with HLSC-EV displayed a reduced extension of necrotic areas, which never exceeded the mild degree (data not shown). Also apoptosis was significantly reduced in NMP + HLSC-EV group (apoptosis index, 0.06 ± 0.01 vs 0.14 ± 0.03; P = 0.049) (Figure 1H).
Biochemical perfusate analyses showed a gradual increase of AST, ALT, and LDH levels in both NMP and NMP + HLSC-EV group throughout perfusion. The AST, ALT, and LDH levels were much higher at 4 hours than at 1 hour in each group (P < 0.0001) (Figure 2). When com- pared with controls, livers treated with HLSC-EV released significantly less AST and LDH at 3 hours (AST, 92 ± 14 vs 47±7U/Lpergram,P=0.018;LDH:619±104vs. 340 ± 47 U/L per gram, P = 0.032) (Figures 2A and B) and lessASTat4hours(134±20U/Lpergramvs80±14U/L per gram of liver, P = 0.003) (Figure 2A), whereas no
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.

e208 Transplantation ■ May 2018 ■ Volume 102 ■ Number 5
www.transplantjournal.com
differences were observed for ALT (Figure 2C). Small amounts of bile were consistently produced during the ex vivo perfusion without differences between groups (Figure 2D).
Quantitative RT-PCR analysis showed that (a) RNA expres- sion of HIF-1α and TGF-β1 was increased in the NMP group, if compared with the sham reference sample (Figures 3A and B); (b) HLSC-EV treatment significantly reduced both HIF-1α and TGF-β1 RNA expression levels when compared with the NMP group (P < 0.0001 and P = 0.014, respectively) (Figures 3A and B).
DISCUSSION
NMP is quickly emerging as a preservation technique po- tentially able to improve LT outcomes using extended criteria grafts.8-12 Thanks to physiological temperature and active
hepatic metabolism, NMP also allows to pharmacologically treat the livers,14 to reduce preservation injury or even ame- liorate their quality before implantation.
HLSC are liver-resident stem-like cells, partially com- mitted to the hepatic lineage, which carry regenerative and hepatoprotective properties.15,16 We already demon- strated that EV derived from HLSC and mesenchymal stem cells are able to mimic most of the cell effects (includ- ing apoptosis inhibition and mitogenic activity) by trans- ferring proteins, mRNAs, and micro-RNAs.20,21 Possible mechanisms involved in EV effects on injured tissue in- clude upregulation of antiapoptotic genes (Bcl-xL, Bcl2, and BIRC8) and downregulation of proapoptotic genes (Casp1, Casp8, and LTA).18,21 Therefore, HLSC-EV may represent a treatment option for liver diseases, avoiding stem cells transplantation.
FIGURE1. ImmunofluorescenceandhistologyinHLSC-EV–treated(NMP+HLSC-EV)(panelsA,B,C,D,E,F,G,andH)andincontrol(NMP) (panels E, F, G, and H) rat livers after 4 hours of ex vivo isolated perfusion. Representative micrographs of fluorescence microscopy showing the cell nuclei (A), rat P450-4A immunofluorescence (B), DIL-stained HLSC-EV (C) and the merge (D) (original magnification, 630
; magnification of the insert representing the center of the image, 2520
; scale bar, 20 μm). (E) Representative micrographs of hematoxylin-eosin staining show- ing the grade of tissue injury (original magnification, 200
; scale bar, 50 μm). (F) Quantitative analysis for tissue damage (Suzuki score; *P = 0.030). (G) Representative micrographs of TUNEL assay showing apoptotic cells (brown; original magnification, 200
; scale bar, 50 μm). (H) Quantitative analysis for apoptosis grade (apoptosis index; *P = 0.049). Data are represented as mean ± SEM.
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.

© 2018 Wolters Kluwer Rigo et al e209
FIGURE 2. Biochemical profile of hepatic cytolysis and function markers in HLSC-EV treated (NMP + HLSC-EV) and control (NMP) rat livers assessed at different time points during ex vivo isolated perfusion. (A) AST (P = 0.018; P = 0.003), (B) LDH (P = 0.032), (C) ALT, and (D) Bile production. All values are normalized to the animal liver weight in grams. Data are represented as mean ± SEM.
Combining these 2 innovative approaches, we demon- strated that HLSC-EV can be rapidly uptaken by the he- patocytes and can reduce hepatic injury in an ex vivo model of hypoxia-induced liver damage, thus providing the rationale for a pharmacological intervention with HLSC-EV during NMP.
For our purposes, we set up a simplified circuit for small organs. Because of the large body of literature on isolated rat liver perfusion,11,28,29 an isolated-perfused rat liver model was chosen for the experimental protocol. The morphologi- cal aspect of livers after 4 hours of NMP (alone or with HLSC-EV treatment), together with the maintained bile pro- duction, evidenced that the experimental conditions allowed both survival and function of organs. Nonetheless, the preperfusion ischemic period (about 30 minutes) followed by an NMP with low hematocrit induced a limited but pro- gressive hypoxic injury, which was proven by both the in- creasing levels of cytolysis enzymes in the perfusate during perfusion and the areas of necrosis and apoptosis observed in liver samples at the end of the experiments.
Fluorescent microscopy confirmed the ability of the rat liver to uptake HLSC-EV20 and revealed the presence of
HLSC-EV within hepatocytes at 4 hours. This uptake is more rapid than that observed with EV derived from mesenchymal stem cells, which were found within the injured tissue after at least 5 hours from their intravenous in vivo inoculation.23 We hypothesize that the ex vivo isolated liver perfusion allows a faster and organ-specific distribution. To avoid modifications of hemodynamic parameters, we did not collect biopsy sam- ples during perfusion; therefore, additional experiments are necessary to explore the precise timing of start of EV uptake within the liver during NMP.
In both groups, cytolysis enzyme levels increased with a steep slope to reach a peak at the fourth hour. This was more likely due to the insufficient oxygen delivery during perfusion rather than to an ex vivo ischemia/reperfusion phenomenon. During this injury, in the NMP + HLSC-EV group, AST and LDH levels were found to be lower at different time points, suggesting a protective effect of HLSC-EV against hypoxia.
This finding was confirmed by better histological integrity of the hepatic parenchyma and by halving of apoptosis which were observed in livers treated with HLSC-EV.
Finally, the RNA overexpression of hypoxia-induced markers was significantly reduced by HLSC-EV treatment.
FIGURE 3. Real-time RT-PCR mean relative quantification (Relative Quantification Mean) of RNA expression of hypoxia-induced markers in a sample of rat livers after 4 hours of ex vivo isolated perfusion. Gene level expression in sham (n = 1), control (NMP, n = 6), and HLSC-EV–treated (NMP + HLSC-EV, n = 5) rat livers of (A) HIF-1α (*P < 0.0001) and (B) TGF-β1 (P = 0.014). All values are normalized to actin β. Data are rep- resented as mean ± SEM. RT, reverse transcription.
Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.

e210 Transplantation ■ May 2018 ■ Volume 102 ■ Number 5
www.transplantjournal.com
In a study on livers perfused at 37°C with an oxygen carrier deficiency, HIF-1α gene and protein expression levels were increased.29 In our setting, EV treatment significantly limited the HIF-1α–dependent response to hypoxia, replicating the evidence reported in a rat model of acute kidney injury.30 This finding is consistent with the cross-talk between HIF-1 and TGF-β1 (a multifunctional cytokine involved in many cellular pathways including inflammation, apoptosis, and fibro- sis), which was recently described in hypoxic hepatocytes.27
As a whole, our data strongly suggest a role of HLSC-EV against hypoxic injury. Studies focusing on the characteriza- tion of HLSC-EV content are needed to clarify their mecha- nisms of protection.
We acknowledge that our study did not include trans- plantation of the livers after hypoxic NMP, yet they were overtly damaged and unsuitable for LT. We aimed indeed to investigate whether HLSC-EV could reduce hypoxic in- jury, a necessary proof of concept before proceeding to other experiments on organ preservation/reconditioning in normoxic conditions.
In conclusion, this study demonstrates that the NMP system can deliver stem cell–derived products to an ex vivo perfused liver and suggests that NMP + HLSC-EV could represent an innovative approach to recondition organs before transplant. Further investigations on NMP models using HLSC-EV in other experimental conditions, including transplantation, are now warranted.
ACKNOWLEDGMENTS
The authors thank Paola Caropreso and Angela Gariboldi (Biochemistry Laboratory, AOU Città della Salute e della Scienza di Torino, Molinette Hospital) for helping with bio- chemical analyses, and Federica Antico for providing techni- cal assistance. We thank Unicyte AG for graciously providing the human liver stem cells (HLSC)-derived extracellular vesi- cles. The study was in part supported by an Unicyte grant to GC.
REFERENCES

1. Gilbo N, Catalano G, Salizzoni M, et al. Liver graft preconditioning, preser- vation and reconditioning. Dig Liver Dis. 2016;48:1265–1274.
2. O’Callaghan JM, Morgan RD, Knight SR, et al. The effect of preservation solutions for storage of liver allografts on transplant outcomes: a system- atic review and meta-analysis. Ann Surg. 2014;260:46–55.
3. Briceño J, Marchal T, Padillo J, et al. Influence of marginal donors on liver preservation injury. Transplantation. 2002;74:522–526.
4. Ravikumar R, Leuvenink H, Friend PJ. Normothermic liver preservation: a new paradigm? Transpl Int. 2015;28:690–699.
5. Laing RW, Bhogal RH, Wallace L, et al. The use of an acellular oxygen car- rier in a human liver model of normothermic machine perfusion. Transplan- tation. 2017;101:2746–2756.
6. Karangwa SA, Burlage LC, Adelmeijer J, et al. Activation of fibrinolysis, but not coagulation, during end-ischemic ex situ normothermic machine per- fusion of human donor livers. Transplantation. 2017;101:e42–e48.
7. van Smaalen TC, Beurskens DM, Hoogland ER, et al. Presence of cyto- toxic extracellular histones in machine perfusate of donation after circula- tory death kidneys. Transplantation. 2017;101:e93–e101.
8. Brockmann J, Reddy S, Coussios C, et al. Normothermic perfusion: a new paradigm for organ preservation. Ann Surg. 2009;250:1–6.
9. JamiesonRW,ZilvettiM,RoyD,etal.Hepaticsteatosisandnormothermic perfusion-preliminary experiments in a porcine model. Transplantation. 2011;92:289–295.
10. Butler AJ, Rees MA, Wight DG, et al. Successful extracorporeal porcine liver perfusion for 72 hr. Transplantation. 2002;73:1212–1218.
11. Tolboom H, Pouw RE, Izamis ML, et al. Recovery of warm ischemic rat liver grafts by normothermic extracorporeal perfusion. Transplantation. 2009;87:170–177.
12. Fondevila C, Hessheimer AJ, Maathuis MH, et al. Superior preservation of DCD livers with continuous normothermic perfusion. Ann Surg. 2011;254: 1000–1007.
13. Ravikumar R, Jassem W, Mergental H, et al. Liver transplantation after ex vivo normothermic machine preservation: a phase 1 (first-in-man) clinical trial. Am J Transplant. 2016;16:1779–1787.
14. Goldaracena N, Spetzler VN, Echeverri J, et al. Inducing hepatitis C virus resistance after pig liver transplantation—a proof of concept of liver graft modification using warm ex vivo perfusion. Am J Transplant. 2017;17: 970–978.
15. HerreraMB,BrunoS,ButtiglieriS,etal.Isolationandcharacterizationofa stem cell population from adult human liver. Stem Cells. 2006;24: 2840–2850.
16. HerreraMB,FonsatoV,BrunoS,etal.Humanliverstemcellsimproveliver injury in a model of fulminant liver failure. Hepatology. 2013;57:311–319. 17. Navarro-Tableros V, Herrera MB, Figliolini F, et al. Recellularization of rat
    liver scaffolds by human liver stem cells. Tissue Eng Part A. 2015;21:
    1929–1939.
17. QuesenberryPJ,AliottaJ,DeregibusMC,etal.RoleofextracellularRNA-
    carrying vesicles in cell differentiation and reprogramming. Stem Cell Res
    Ther. 2015;6:153.
18. Biancone L, Bruno S, Deregibus MC, et al. Therapeutic potential of mes-
    enchymal stem cell-derived microvesicles. Nephrol Dial Transplant. 2012;
    27:3037–3042.
19. Herrera MB, Fonsato V, Gatti S, et al. Human liver stem cell-derived
    microvesicles accelerate hepatic regeneration in hepatectomized rats. J Cell
    Mol Med. 2010;14:1605–1618.
20. Bruno S, Grange C, Collino F, et al. Microvesicles derived from mesenchy-
    mal stem cells enhance survival in a lethal model of acute kidney injury.
    PLoS One. 2012;7:e33115.
21. HerreraMB,BrunoS,GrangeC,etal.Humanliverstemcellsandderived
    extracellular vesicles improve recovery in a murine model of acute kidney
    injury. Stem Cell Res Ther. 2014;5:124.
22. Grange C, Tapparo M, Bruno S, et al. Biodistribution of mesenchymal
    stem cell-derived extracellular vesicles in a model of acute kidney injury
    monitored by optical imaging. Int J Mol Med. 2014;33:1055–1063.
23. Gravante G, Ong SL, Metcalfe MS, et al. Effects of hypoxia due to isovolemic hemodilution on an ex vivo normothermic perfused liver model.
    J Surg Res. 2010;160:73–80.
24. Suzuki S, Toledo-Pereyra LH, Rodriguez FJ, et al. Neutrophil infiltration
    as an important factor in liver ischemia and reperfusion injury. Modu- lating effects of FK506 and cyclosporine. Transplantation. 1993;55: 1265–1272.
25. Fonsato V, Collino F, Herrera MB, et al. Human liver stem cell-derived microvesicles inhibit hepatoma growth in SCID mice by delivering antitu- mor microRNAs. Stem Cells. 2012;30:1985–1998.
26. Roth KJ, Copple BL. Role of hypoxia-inducible factors in the development of liver fibrosis. Cell Mol Gastroenterol Hepatol. 2015;1:589–597.
27. Schlegel A, Kron P, Graf R, et al. Warm vs. cold perfusion techniques to rescue rodent liver grafts. J Hepatol. 2014;61:1267–1275.
28. Ferrigno A, Di Pasqua LG, Bianchi A, et al. Metabolic shift in liver: correla- tion between perfusion temperature and hypoxia inducible factor-1α. World J Gastroenterol. 2015;21:1108–1116.
29. Zou X, Gu D, Xing X, et al. Human mesenchymal stromal cell-derived ex- tracellular vesicles alleviate renal ischemic reperfusion injury and enhance angiogenesis in rats. Am J Transl Res. 2016;8:4289–4299.
    Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.