cells Article The Effect of Fibroblast Growth Factor 15 Signaling in Non-Steatotic and Steatotic Liver Transplantation from Cardiocirculatory Death Cindy G. Avalos-de León 1 , Mónica B. Jiménez-Castro 1 , María Eugenia Cornide-Petronio 1 , José Gulfo 1 , Floriana Rotondo 1 , Jordi Gracia-Sancho 2,3 , Araní Casillas-Ramírez 4,5, *,† and Carmen Peralta 1,2, *,† 1 Institut d’Investigacions Biomèdiques August Pi I Sunyer (IDIBAPS), 08036 Barcelona, Spain; avalosdl.cin@gmail.com (C.G.A.-d.L.); monicabjimenez@hotmail.com (M.B.J.-C.); cornide@clinic.cat (M.E.C.-P.); gulfo@clinic.cat (J.G.); floriana.rotondo@gmail.com (F.R.) 2 Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBERehd), 08036 Barcelona, Spain; jordi.gracia@idibaps.org 3 Liver Vascular Biology Research Group, IDIBAPS, 08036 Barcelona, Spain 4 Hospital Regional de Alta Especialidad de Ciudad Victoria “Bicentenario 2010”, Ciudad Victoria 87087, Mexico 5 Facultad de Medicina e Ingeniería en Sistemas Computacionales de Matamoros, Universidad Autónoma de Tamaulipas, Matamoros 87300, Mexico * Correspondence: aranyc@yahoo.com (A.C.-R.); cperalta@clinic.cat (C.P.); Tel.: +52-834-153-6100 (A.C.-R.); +34-932-275-400 (ext. 4177) (C.P.) † These authors contributed equally to this work.  Received: 20 November 2019; Accepted: 13 December 2019; Published: 14 December 2019  Abstract: We elucidate the relevance of fibroblast growth factor 15 (FGF15) in liver transplantation (LT) using rats with both steatotic and non-steatotic organs from donors after cardiocirculatory death (DCD). Compared to LT from non-DCDs, the induction of cardiocirculatory death (CD) increases hepatic damage, proliferation, and intestinal and circulatory FGF15. This is associated with high levels of FGF15, bilirubin and bile acids (BAs), and overexpression of the enzyme involved in the alternative BA synthesis pathway, CYP27A1, in non-steatotic livers. Furthermore, CD activates the proliferative pathway, Hippo/YAP, in these types of liver. Blocking FGF15 action in LT from DCDs does not affect CYP27A1 but causes an overexpression of CYP7A, an enzyme from the classic BA synthesis pathway, and this is related to further accumulation of BAs and exacerbated damage. FGF15 inhibition also impairs proliferation without changing Hippo/YAP. In spite of worse damage, steatosis prevents a proliferative response in livers from DCDs. In steatotic grafts, CD does not modify CYP7A1, CYP27A1, BA, or the Hippo/YAP pathway, and FGF15 is not involved in damage or proliferation. Thus, endogenous FGF15 protects against BA accumulation and damage and promotes regeneration independently of the Hippo/YAP pathway, in non-steatotic LT from DCDs. Herein we show a minor role of FGF15 in steatotic LT from DCDs. Keywords: cardiocirculatory death; FGF15; liver transplantation; steatotic liver grafts; ischemia-reperfusion damage 1. Introduction Every year, many thousands of deaths result from liver failure that could not be treated due to the lack of suitable donor organs [1]. To better meet clinical demand, the use of livers from donors after cardiocirculatory death (DCDs) is being explored. However, the high incidence of hepatic dysfunction Cells 2019, 8, 1640; doi:10.3390/cells8121640 www.mdpi.com/journal/cells Cells 2019, 8, 1640 2 of 21 and biliary complications caused by disruption of the blood supply during the donor warm ischemia time after cardiocirculatory death (CD) hinders the utilization of these grafts to liver transplantation (LT) [2–4]. In addition, the use of organs from extended criteria donors has increased, including those with hepatic steatosis, due to its high worldwide prevalence [5]. Unfortunately, hepatic steatosis is one of the most important risk factors that negatively affect post-operative outcomes [6]. Fibroblast growth factor 15 (FGF15), along with its human homolog (FGF19), collectively referred to as FGF15/19, regulates bile acids (BAs) [7]. In brief, the ileal enterocytes induce the expression of FGF15/19, which is secreted into the enterohepatic circulation [8]. Through the hepatic fibroblast growth factor receptor 4 (FGFR4) signaling pathway, BA levels are tightly controlled by the enzyme cytochrome P450 7A1 (CYP7A1), the first and rate-limiting step in the classic BA synthesis pathway [9]. In addition to maintaining BA homeostasis, FGF15 regulates liver regeneration through Hippo/Yes-associated protein (YAP) signaling [10,11]. Indeed, the mammalian sterile-20 kinase 1/2 (MST1/2) activates the large tumor suppressor 1/2 (LATS1/2), that phosphorylates YAP [12]. Phospho-YAP is either degraded or sequestered in the cytoplasm [13]. When the Hippo pathway is off, YAP translocates to the nucleus and turns on pro-proliferative and pro-survival genes, thereby enabling cell proliferation [14]. Changes in circulating FGF19 have been reported in patients with liver diseases, such as cirrhosis [15], and in pathologies associated with intestinal inflammation, such as ileal dysfunction due to Crohn’s disease [16]. Several lines of evidence demonstrate that various liver diseases, such as cirrhosis and bile duct ligation, alter BA synthesis by up-regulating the expression of CYP7A1 in the liver [17]. In addition, YAP mediates cell proliferation and regulated cell death and participates in the restoration of the heart after ischemia-reperfusion (I/R) [18]. Along the same lines, ablating YAP in the mouse liver enhances hepatocyte necrosis and suppresses hepatocyte proliferation after bile duct ligation [14]. This is the first study to analyze the effect of FGF15 derived from intestinal signaling in both steatotic and non-steatotic LT from DCDs. Understanding the essential components of this response may facilitate the development of hepatoprotective and pro-regenerative therapeutic strategies in steatotic and non-steatotic LT from DCDs. 2. Materials and Methods 2.1. Experimental Animals This study was performed using homozygous (obese, Ob) and heterozygous (lean, Ln) male Zucker rats (Iffa Credo, France) aged between 10 and 11 weeks. Ob rats show moderate macrovesicular and microvesicular fatty infiltration in hepatocytes (40% to 60% steatosis), whereas Ln rats show no evidence of steatosis [19]. All procedures were approved by the Laboratory Animal Care and Use Committee of Barcelona University and by the government of Catalonia (DAAM 9353). European Union regulations (Directive 86/609 EEC) for animal experiments were respected. 2.2. Cardiocirculatory Death Induction Animals were anesthetized by isoflurane and heparin sodium solution was administered to donors. A well-established rodent CD model was selected: induction of CD by hypoxia through an incision of the diaphragm and blocking the descending aorta [20,21]. The in situ warm ischemia period was 45 min. 2.3. Surgical Procedure of Liver Transplantation Steatotic and non-steatotic liver grafts from DCDs and non-DCDs were flushed and stored in University of Wisconsin (UW) solution for 6 h. Standard orthotopic LT in Ln Zucker rats was performed according to the double cuff technique [19]. Cells 2019, 8, 1640 3 of 21 2.4. Experimental Design Group 1. Sham (n = 12). Six Ob and six Ln Zucker rats were anesthetized and maintained for 45 min. Group 2. LT (n = 24, 12 transplantations). Six Ob and six Ln Zucker rats were anesthetized and maintained for 45 min. Then, steatotic and non-steatotic liver grafts were flushed, isolated, and stored in ice-cold UW solution for 6 h. Finally, they were implanted in 12 Ln Zucker rats [19]. Group 3. CD + LT (n = 24, 12 transplantations). Six Ob and six Ln Zucker rats were anesthetized. After the induction of CD, the animals were maintained in situ in warm ischemia for 45 min. Then, steatotic and non-steatotic liver grafts were flushed, isolated, and stored in ice-cold UW solution for 6 h. Finally, they were implanted in 12 Ln Zucker rats [19–21]. Group 4. CD + FGF15inh + LT (n = 24, 12 transplantations). Procedures as per Group 3 except donors were treated with intravenous FGF401 at a dosage of 200 µg/kg, 10 min before CD induction. FGF401 is a highly selective and potent FGFR4 inhibitor aimed at blocking the actions of FGF15 [20–23]. Group 5. FGF15inh + LT (n = 24, 12 transplantations). Procedures as per Group 2 except donors were treated with intravenous FGF401 at a dosage of 200 µg/kg, 10 min before surgical procedure [19,23]. Group 6. CD (n = 12). Six Ob and six Ln Zucker rats were anesthetized. After CD induction, rats were maintained in situ in warm ischemia for 45 min [20,21]. Group 7. CD + FGF15inh (n = 12). Procedures as per Group 5 except donors were treated with intravenous FGF401 at a dosage of 200 µg/kg, 10 min before CD induction [20–23]. Plasma, liver, and intestine samples were collected 4 h after LT (Groups 2–5) as well as before the procurement of liver grafts from donors (Groups 6 and 7). For the current study, conditions, drug dosage, and pre-treatment times were selected based on previous studies [19–23] as well as on preliminary studies by our group. 2.5. Biochemical Assays Plasma transaminases (aspartate aminotransferase (AST) and alanine aminotransferase (ALT)) were measured photometrically using standard procedures. Total bilirubin in plasma was determined by an enzyme-linked immune absorbent assay (MyBiosourse Inc., San Diego, CA, USA) according to the manufacturer’s instructions. Plasma and liver total BAs were determined by a colorimetric assay kit (Biovision, Milpitas, CA, USA). Lipid peroxidation in the intestine was determined by measuring the formation of malondialdehyde (MDA) as an indirect index of the oxidative injury induced by the reactive oxygen species [24]. Briefly, 0.5 mL of 0.5% butylated hydroxytoluene was added to 2 mL of the intestine homogenate to prevent lipid autoxidation. For protein precipitation, 2 mL of 20% trichloroacetic acid was added to 2 mL of homogenate. After mixing and centrifuging, 1 mL of 0.67% thiobarbiturate solution was added to the supernatant and boiled for 60 min. After cooling, optical density at 530 nm was assayed [24]. Myeloperoxidase (MPO), was measured photometrically as an index of neutrophil accumulation, using tetramethyl-benzidine as a substrate. Intestine samples were macerated with 0.5% hexadecyltrimethylammonium bromide in 50 mM phosphate buffer pH 6.0. Homogenates were then disrupted for 30 s using a sonicator at 20% power and subsequently snap-frozen in dry ice and thawed on three consecutive occasions before a final 30-s sonication. Samples were incubated at 60 ◦ C for 2 h and then spun down at 4000× g for 12 min. Supernatants were collected for the MPO assay. Enzyme activity was assessed photometrically at 630 nm. The assay mixture consisted of 20 µL supernatant, 10 µL tetramethylbenzidine (final concentration 1.6 mM) dissolved in dimethyl sulfoxide, and 70 µL H2 O2 (final concentration, 3.0 mM) diluted in 80 Mm phosphate buffer, pH 5.4. An enzyme unit is defined as the amount of enzyme that produces an increase of 1 absorbance unit per minute [25]. Cells 2019, 8, 1640 4 of 21 2.6. Western Blotting Liver and intestine tissues were homogenized in a lysis buffer, 150 mM NaCl, 50 mM Tris pH 7.5, 1 mM EDTA, 0.5% Triton X-100, 0.5% Nonidet P-40, with protease and phosphatase inhibitors (Sigma Aldrich, St Louis, MO, USA). Samples were sonicated at 60 w for 15 s and then centrifuged at 20,000× g for 10 min. Plasma, liver, and intestine homogenates containing an equal amount of protein were mixed in Laemmli loading buffer and were separated on a sodium dodecyl sulfate (SDS-PAGE) 8–12% poly-acrylamide gel electrophoresis and transferred to polyvinylidene fluoride membranes. After assessing transfer, the membranes were saturated in 4 mM Tris–HCl, pH 7.6, 30 mM NaCl (TBST) containing 5% non-fat milk and 0.1% Tween-20 and incubated over-night at 4 ◦ C, using antibodies against the following proteins: FGF15 (LS-B15011) (Life-Span BioSciences Inc., Seattle, Washington, USA); CYP7A1 (ab65596) (Abcam, Cambridge, UK); cytochrome P450 27A1 (CYP27A1) (SAB1400066) (Sigma-Aldrich, St Louis, MO, USA); YAP (4912S), p-YAP (4911S), LATS1 (9135S), p-LATS1 (9157) (Cell Signaling Technology, Danvers, MA, USA); cyclin A1 (ab65596) (Abcam, Cambridge, UK) and β-Actin (A5316) (Sigma-Aldrich, St. Louis, MO, USA) as a loading control. Signals were detected by enhanced chemiluminescence, using peroxidase-conjugated secondary antibodies and Clarity Western ECL substrate (Bio-Rad Laboratories, Hercules, CA, USA) and quantified with standard densitometric scanning software (Quantity One; BioRad Laboratories, Hercules, CA, USA). 2.7. Reverse Transcription and Quantitative Polymerase Chain Reaction Total RNA was isolated from frozen rat liver and intestine sections using TRIzol reagent (Invitrogen, Madrid, Spain) and was quantified with a NanoDrop 1000 spectrophotometer. Two µg of RNA was reverse transcribed using the high capacity cDNA reverse transcription kit. Real-time PCR was performed in an ABI PRISM 7900 HT detection system by using 5 ul of PowerUp SYBR green master mix (ThermoFisher Scientific, Life Technologies, Carlsbad, CA, USA) in a total of 10 µL amplification mixture, containing 10 ng of reverse-transcribed RNA and 300 nM of rat primers. 18S was used as a housekeeping gene. The ∆∆Ct method was used for relative quantifications. Data were calculated with respect to the sham group and expressed as percentages. The primer sequences were as follows: Fgfr4 (forward 50 -CAGTGCTACCCGCAGAGGAAG-30 and reverse 50 -TGGATTGCTTGTCGGTGA TACA-30 ); Iqgap1 (forward 50 -ACAATCTGGAGACGCAAGCA-30 and reverse 50 -AGCTGCTCTCGG TTATACGC-30 ); Lats1 (forward 50 -TCTGCATCGCAAGAAGCAAC-30 and reverse 50 -TCTCATTTG ATCCTGGGCATCT-30 ); Mst1 (forward 50 -GAGACCGTGCAACTGAGGAA-30 and reverse: 50 -TGA ATCGCCTTGTACACGCT-30 ); Yap (forward 50 -CCTGATGGATGGGAGCAAGC-30 and reverse 50 -ACTCTGAGTGATCCTCTGGTTC-30 ); Ctgf (forward 50 AGACACATTTGGCCCTGACC-30 and reverse 50 -TCTTAGAACAGGCGCTCCAC-30 ); 18S (18S ribosomal RNA, forward 50 -GGGAGCCTG AGAAACGGC-30 and reverse 50 -GGGTCGGGAGTGGGTAATTT-30 ). 2.8. Liver and Intestine Histology To assess the severity of hepatic injury, paraffin-embedded liver sections were stained with hematoxylin and eosin, and blind histological scoring was performed by a board-certified pathologist, using a point-counting method on an ordinal scale, as follows: grade 0, minimal or no evidence of injury; grade 1, mild injury consisting of cytoplasmic vacuolation and focal nuclear pyknosis; grade 2, moderate to severe injury with extensive nuclear pyknosis, cytoplasmic hypereosinophilia, and loss of intercellular borders; grade 3, severe necrosis with disintegration of hepatic cords, hemorrhage, and neutrophil infiltration; grade 4, very severe necrosis with disintegration of hepatic cords, hemorrhaging, and neutrophil infiltration [19]. The severity of intestinal damage was scored on a scale from 0 to 5, as described by Chiu et al. [26]. Liver steatosis was visualized by Oil Red staining of liver cryosections. Liver tissues were frozen in optimal cutting temperature (OCT) compounds. The sections were fixed with 10% formalin and the slides were placed in 100% propylene glycol and stained in 0.5% Oil Red solution in propylene glycol. The slides were transferred to an 85% propylene glycol Cells 2019, 8, 1640 5 of 21 solution and processed for hematoxylin counterstaining. The percentage of steatosis was calculated by image analysis [24]. At least 30 high-power fields were counted per slide. The score described by Brunt et al. [27] is suitable for the assessment of the full spectrum of nonalcoholic fatty liver disease, including simple steatosis. This score is calculated as the weighted sum of the scores for steatosis (the % of liver cells containing fat, range 0–3; 0, <5%; 1, 5–33%; 2, >33–66%; and 3, >66%), ballooning (0–2), lobular inflammation (0–3), and fibrosis state (1, perisinuoidal or periportal; 2, perisunoidal and periportal; 3, briding fibrosis, and 4: cirrhosis). Thus, this score ranges from 0 to 8. The score described by Matteoni et al. [28] is defined as the following types: type 1, fatty liver alone; type 2, fat accumulation and lobular inflammation; type 3, fat accumulation and ballooning degeneration; type 4, fat accumulation, ballooning degeneration, and fibrosis. As previously reported [19,24,29,30], steatosis in Zucker rats used in the current study is not associated with inflammation. Ob rats exhibited moderate macrovesicular and microvesicular fatty infiltration in hepatocytes (40–60% steatosis) but no apparent inflammation or fibrosis. These observations in Ob rats correspond to grade 1 according to Brunt’s classification, and type 1 according to Metteoni’s classification [27,28]. 2.9. Immunohistochemistry After fixation with 4% formalin/phosphate-buffered saline, paraffin-embedded livers were sliced and immunostained using mouse monoclonal antibody anti-proliferating cell nuclear antigen (PCNA) (Agilent, Santa Clara, CA, USA) and anti-YAP (Cell Signaling Technology, Danvers, MA, USA). Staining was developed with diaminobenzidine (DAB), using DAKO Envision System (Agilent, Santa Clara, CA, USA) according to the manufacturer’s protocol. Sections were counterstained with hematoxylin. At least 30 high-power fields were counted per slide. 2.10. Statistics Data are expressed as means ± standard error and were statistically analyzed via a one-way analysis of variance, followed by post hoc Student–Newman–Keuls test. p < 0.05 was considered significant. 3. Results 3.1. Role of FGF15 in Recipients of Non-Steatotic and Steatotic Liver Grafts from DCDs First, we determined the effect of CD on damage and regeneration in both non-steatotic and steatotic liver grafts after transplantation. CD was seen to exacerbate damage in both types of grafts since transaminase levels and damage score values were higher in the CD+LT group than in the LT group (LT without CD) (Figure 1A). The histological study of liver grafts from DCDs (CD + LT) showed moderate and multifocal areas of coagulative necrosis and neutrophil infiltration, randomly distributed through the parenchyma, in non-steatotic livers whereas in steatotic grafts, severe, extensive, and confluent areas of coagulative necrosis were observed (Figure 1A). In LT using non-steatotic grafts, CD triggered an increment in regeneration markers, PCNA and cyclin A (CD + LT group), compared to those without CD (LT group). On the other hand, in steatotic livers undergoing transplantation, regeneration parameters were unchanged by CD (Figure 1B). Second, we evaluated the role of FGF15 in damage and regeneration in grafts from DCDs after transplantation. In non-steatotic liver grafts from DCDs, our findings indicated a rise in FGF15 protein levels, in comparison to the results of the LT group. On the other hand, in steatotic livers, FGF15 protein levels were similar in both the CD + LT and LT groups (Figure 2A). Circulatory FGF15/19 is assumed to come from the intestine, but, under conditions of obstructed bile flow, FGF15/19 is also synthesized by the liver [31,32]. However, in our experiment, Fgf15 was not expressed by the liver in any of the study groups, which was confirmed by a positive control (intestine from the Sham group) (Figure 2A). Cells 2019, 8, 1640 6 of 21 Cells 2019, 8, x FOR PEER REVIEW 6 of 21 Figure Figure 1. 1.Liver Liver injury injury and andregeneration regenerationin in non-steatotic and and non-steatotic steatotic livers livers steatotic from DCDs from after DCDsLT.after (A) LT. Hepatic damage parameters: plasma transaminases, liver damage score (A) Hepatic damage parameters: plasma transaminases, liver damage score and representative and representative photographs photographs ofof histologicalchanges histological changes in in LT LT from from DCDs. DCDs.InInnon-steatotic non-steatoticgrafts, LT LT grafts, exhibited moderate exhibited moderate and multifocal areas of coagulative necrosis, the extent and the number of necrotic areas and multifocal areas of coagulative necrosis, the extent and the number of necrotic areas increasedincreased in in CD + LT, and extensive and confluent areas of coagulative necrosis were observed in CD + FGF15inh CD + LT, and extensive and confluent areas of coagulative necrosis were observed in CD + FGF15inh + LT. In steatotic grafts, LT showed extensive and confluent areas of coagulative necrosis, whereas such Cells 2019, 8, 1640 7 of 21 coagulative necrosis was even more severe in CD + LT and CD + FGF15inh + LT (4×). (B) Liver regeneration parameters: percentage of positive hepatocytes of PCNA, protein expression of cyclin A1 in liver tissue, and representative photographs of immunohistochemical staining of PCNA-positive cells. In non-steatotic livers, CD + LT increased the number of PCNA-positive hepatocytes compared with the LT group. CD + FGF15inh + LT showed fewer positive cells than the CD + LT group. In steatotic livers, PCNA-positive cells were similar in LT, CD + LT and CD + FGF15inh + LT groups Cells 2019, (20×). 8, x FOR p < 0.05 vs. Sham; + p < 0.05 vs. LT; # p < 0.05 vs. CD + LT. PEER *REVIEW 8 of 21 Figure Figure 2. 2. FGF15levels FGF15 levelsand andintestinal intestinal damage damage inin LT LT with withDCDs. DCDs.(A) (A)FGF15 FGF15protein expression protein in liver, expression in liver, intestine, intestine, andand plasma.(B) plasma. (B)Intestinal Intestinalinjury injury parameters: parameters: MDA, MDA,MPO MPOandandhistological histologicalinjury according injury according to to Chiu-score.* p* p<<0.05 Chiu-score. 0.05vs. vs. Sham; Sham; ++ PP<<0.05 0.05vs. vs.LT. LT. 3.3.When The Effects of FGF15the we inhibited on the BA Synthesis action of FGF15,Pathway in LT from DCDs the non-steatotic grafts from DCDs showed exacerbated damageAs and impaired regenerative response. Indeed, (a) increased previously mentioned, FGF15/19 regulates BA synthesis through levels ofFGFR4 transaminases signalingand damage [7,9]. We score values, and (b) reduced PCNA and cyclin A, were observed in the CD + FGF15inh registered an increased expression of Fgfr4 in non-steatotic livers when comparing the CD + LT and + LT group when compared LT groups. to the CD In steatotic + LT grafts, group. these groupsHowever, showed the inhibitioninofgene no difference FGF15 action in expression steatotic of Fgfr4 livers (Figure from 3A). Interestingly, bilirubin and both hepatic and plasma BAs were also higher only in non-steatotic of DCDs did not induce changes in either hepatic damage or regeneration. Thus, in such types liver LT graft transaminases, from DCDs damage (Figure 3B). In fact, scores, PCNA and in the presence cyclin Asuch of steatosis, values were similar parameters in both in the CD + LTthe groupCD + FGF15inh resembled+ LT theand CD levels + LTin found groups the LT(Figure group. To1). investigate the signaling pathway associated with the Control increase in experiments based oninthe the BA pool observed administration non-steatotic LT fromof FGF15inh DCDs, theinclassic LT without CD wereBA and alternative also synthesis pathways performed. were evaluated. The inhibition of FGF15 The protein action (FGF15inhof+CYP7A1, in LTexpression LT group)theresulted main enzyme of the in parameters classic BA synthesis pathway, was similar in non-steatotic grafts in the CD + LT and LT groups (Figure 3A). On the other hand, protein expression of CYP27A1, a representative enzyme of the alternative BA synthesis pathway, was higher in non-steatotic liver grafts in the CD + LT group than the LT group. In steatotic liver grafts from CD + LT group, CYP7A1 and CYP27A1 protein expression were no different from the LT group (Figure 3A). Cells 2019, 8, 1640 8 of 21 of proliferation and damage in non-steatotic livers similar to those of the LT group. For instance, the results corresponding to transaminases and PCNA were the following: ALT levels: 1345 ± 201 and 1281 ± 191 U/L for the FGF15inh + LT and LT groups, respectively; AST levels: 1547 ± 101 and 1488 ± 97 U/L for the FGF15inh + LT and LT groups, respectively; and PCNA: 52.47 ± 4.68 and 49.5 ± 4.42% PCNA-positive hepatocytes for FGF15inh + LT and LT groups, respectively; p > 0.05 for all determinations, data not shown in graphs. Similarly, the inhibition of FGF15 action in LT (FGF15inh + LT group) resulted in parameters of proliferation and damage in steatotic livers similar to those of the LT group. For instance, the results corresponding to transaminases and PCNA were the following: ALT levels: 2671 ± 441 and 2544 ± 420 U/L for the FGF15inh + LT and LT groups, respectively; AST levels: 2790 ± 244 and 2683 ± 235 U/L for the FGF15inh + LT and LT groups, respectively; and PCNA: 25.05 ± 3.97 and 23.63 ± 3.74% PCNA-positive hepatocytes for the FGF15inh + LT and LT groups, respectively; p > 0.05 for all determinations, data not shown in graphs. 3.2. FGF15 Levels in Intestine and Plasma in LT from DCDs Next, we evaluated the possible contribution of intestine in FGF15 signaling in LT from DCDs. Similar intestinal tissue histology and levels of MDA, MPO were found in the Sham, LT, and CD + LT groups, suggesting that there is no intestinal damage when LT is performed nor when CD is induced, in any type of liver (Figure 2B). Regarding FGF15, an intestinal rise in protein levels was observed in both non-steatotic and steatotic grafts in the CD + LT group when compared to levels recorded in the LT group. This pattern of levels of FGF15 in intestinal tissue was mirrored in circulatory FGF15: levels of FGF15 in plasma were higher in the CD + LT group compared to the LT group, in both non-steatotic and steatotic grafts (Figure 2A). 3.3. The Effects of FGF15 on the BA Synthesis Pathway in LT from DCDs As previously mentioned, FGF15/19 regulates BA synthesis through FGFR4 signaling [7,9]. We registered an increased expression of Fgfr4 in non-steatotic livers when comparing the CD + LT and LT groups. In steatotic grafts, these groups showed no difference in gene expression of Fgfr4 (Figure 3A). Interestingly, bilirubin and both hepatic and plasma BAs were also higher only in non-steatotic LT from DCDs (Figure 3B). In fact, in the presence of steatosis, such parameters in the CD + LT group resembled the levels found in the LT group. To investigate the signaling pathway associated with the increase in the BA pool observed in non-steatotic LT from DCDs, the classic and alternative BA synthesis pathways were evaluated. The protein expression of CYP7A1, the main enzyme of the classic BA synthesis pathway, was similar in non-steatotic grafts in the CD + LT and LT groups (Figure 3A). On the other hand, protein expression of CYP27A1, a representative enzyme of the alternative BA synthesis pathway, was higher in non-steatotic liver grafts in the CD + LT group than the LT group. In steatotic liver grafts from CD + LT group, CYP7A1 and CYP27A1 protein expression were no different from the LT group (Figure 3A). Then, we attempted to clarify the possible implications of FGF15 for the synthesis pathways of BAs. As previously demonstrated, when we inhibited the action of FGF15 (CD + FGF15inh + LT), hepatic damage in non-steatotic grafts from DCDs was exacerbated (Figure 1). In the same type of liver, blocking the action of FGF15 (CD + FGF15inh + LT group) reduced the expression of Fgfr4 and resulted in high levels of bilirubin and hepatic and plasma BAs compared to the CD + LT group. This was related to an overexpression of CYP7A1 in the CD + FGF15inh + LT group compared to the CD + LT group, whereas CYP27A1 did not alter its expression (Figure 3). As we saw for damage, the inhibition of FGF15 action in steatotic LT from DCDs (CD + FGF15inh + LT) did not induce changes in either Fgfr4, bilirubin, BAs, CYP7A1 or CYP27A1 levels when compared to results from the CD + LT group. resulted in high levels of bilirubin and hepatic and plasma BAs compared to the CD + LT group. This was related to an overexpression of CYP7A1 in the CD + FGF15inh + LT group compared to the CD + LT group, whereas CYP27A1 did not alter its expression (Figure 3). As we saw for damage, the inhibition of FGF15 action in steatotic LT from DCDs (CD + FGF15inh + LT) did not induce changes in either Cells 2019, 8,Fgfr4, 1640 bilirubin, BAs, CYP7A1 or CYP27A1 levels when compared to results from the CD +9 of 21 LT group. Figure 3. Figure 3. Effect Effect of of FGF15 FGF15ononFGFR4-CYP7A1/CYP27A1 FGFR4-CYP7A1/CYP27A1 pathway pathway andand BA BAaccumulation accumulationin LTinwith LT with DCDs. (A) DCDs. (A) mRNA mRNAexpression expressionofofFgfr4 Fgfr4and andprotein expression protein expression of CYP7A1 of CYP7A1andandCYP27A1 CYP27A1in liver. (B) (B) in liver. Bilirubin in Bilirubin in plasma, plasma, and and BAs BAs in inliver liverand plasma.* *p p< < andplasma. 0.05 vs.vs. 0.05 Sham; + p+< p0.05 Sham; < 0.05 vs. LT; vs. and p < 0.05 # p <#0.05 LT; and vs. CD vs. CD ++ LT. LT. 3.4. 3.4. The The Effect of FGF15 FGF15 on onthe theHippo/YAP Hippo/YAPSignaling SignalingPathway Pathwayinin Liver Liver Grafts Grafts from from DCDs DCDs after after LT LT FGF15/19 regulates FGF15/19 regulates liver liverregeneration regenerationthrough through Hippo/YAP Hippo/YAP signaling [11]. signaling OurOur [11]. results indicated results indicated thatmRNA that mRNAexpression expression of of mediators mediators of Hippo signaling such such as as Mst1 Mst1 and and Lats1; Lats1;and andprotein proteinlevels levelsof of the final effector LATS (in its phosphorylated form), were decreased in non-steatotic grafts in the CD + LT the final effector LATS (in its phosphorylated form), were decreased in non-steatotic grafts in the CD + LT group, group, whenwhen compared compared to thetoLT thegroup LT group (Figure (Figure 4A).4A). In contrast, In contrast, mRNAmRNA expression expression and and protein protein levels levels of YAP were higher, and p-YAP protein levels lower, in the CD + LT group of YAP were higher, and p-YAP protein levels lower, in the CD + LT group (Figure 4B). Furthermore, (Figure 4B). Furthermore, YAP expression YAP wasexpression markedlywas markedly higher higher innuclei in hepatocyte hepatocyte in the nuclei CD + LTin the CD +This group. LT group. Thiswith coincided coincided with over-expression of YAP target genes, such as connective tissue growth factor over-expression of YAP target genes, such as connective tissue growth factor (Ctgf ). In non-steatotic (Ctgf). In non-steatotic liver grafts, the mRNA expression of Mst1, Lats1, and Ctgf and the protein expression liver grafts, the mRNA expression of Mst1, Lats1, and Ctgf and the protein expression of p-LATS and of p-LATS and p-YAP, and the YAP immunohistochemical positivity were similar, in the CD + p-YAP, and the YAP immunohistochemical positivity were similar, in the CD + FGF15inh + LT and FGF15inh + LT and CD + LT groups. In steatotic grafts, the expression of all the parameters of the CD + LT groups. In steatotic grafts, the expression of all the parameters of the Hippo/YAP signaling pathway was unchanged across all experimental groups (Figure 4). BA levels, through the IQ motif containing GTPase activating protein 1 (IQGAP1), are able to induce YAP activation [33,34]. However, in our hands, Iqgap1 mRNA expression was similar across all groups of the study, for both liver types (Figure 4B). Cells 2019, 8, 1640 10 of 21 C Figure 4. Effect of FGF15 on Hippo/YAP signaling pathway in LT with DCDs. (A) mRNA expression of Mst1, Lats1, and protein levels of LATS, p-LATS, and p-LATS/LATS ratio in liver. (B) mRNA levels of Yap, protein levels of YAP, p-YAP, p-YAP/YAP ratio; mRNA expression levels of Ctgf and Iqgap1 in liver. * p < 0.05 v.s Sham and + p < 0.05 vs. LT; # p < 0.051vs. CD + LT. (C) Representative photomicrographs of YAP immunohistochemical positivity. Inmunostaining of cytoplasmatic YAP in hepatocyte (thick arrow) of LT group and YAP nuclear staining in CD + LT and CD + FGF15Inh + LT groups (thin arrow) (40×). Figure 4. Effect of FGF15 on Hippo/YAP signaling pathway in LT with DCDs. (A) mRNA expression of Mst1, Lats1, and protein levels of LATS, p-LATS, and p-LATS/LATS ratio in liver. (B) mRNA levels of Yap, protein levels of YAP, p-YAP, p-YAP/YAP ratio; mRNA expression levels of Ctgf and Iqgap1 in liver. * p < 0.05 v.s Sham and + p < 0.05 vs. LT; # p < 0.05 vs. CD + LT. (C) Representative photomicrographs Cells 2019, 8, 1640 of YAP immunohistochemical positivity. Inmunostaining of cytoplasmatic YAP in 11 of 21 hepatocyte (thick arrow) of LT group and YAP nuclear staining in CD + LT and CD + FGF15Inh + LT groups (thin arrow) (40×). 3.5. FGF15 and Related Signaling Pathways in Liver Grafts from DCDs before Procurement 3.5. FGF15 and Related Signaling Pathways in Liver Grafts from DCDs before Procurement The changes recorded after transplantation, with respect to FGF15 and associated signaling The changes pathways, were alsorecorded observedafter transplantation, immediately withinduction after the respect toofFGF15 CD, just and associated before signaling of the procurement pathways, were also observed immediately after the induction of CD, just before the procurement of grafts. In non-steatotic livers, a rise in FGF15 levels was observed in the CD group, when compared grafts. In non-steatotic livers, a rise in FGF15 levels was observed in the CD group, when compared to the Sham group (Figure 5A). On the other hand, in steatotic livers, there was no difference in to the Sham group (Figure 5A). On the other hand, in steatotic livers, there was no difference in FGF15 FGF15 levels between those experimental groups. Induction of CD increased transaminases in both levels between those experimental groups. Induction of CD increased transaminases in both non- non-steatotic and steatotic livers from the CD group compared to the Sham group. Also, we observed an steatotic and steatotic livers from the CD group compared to the Sham group. Also, we observed an increment in cyclin A protein expression in the CD group with non-steatotic livers, but this regeneration increment in cyclin A protein expression in the CD group with non-steatotic livers, but this parameter regeneration wasparameter unchanged wasin groups unchanged of steatotic in groupsgrafts of (Figure steatotic5A). Increased grafts (Figuretransaminase 5A). Increasedlevels and reduced cyclin transaminase levelsprotein expression and reduced cyclinwere recorded protein expression CD +recorded in thewere FGF15inh in group, the CDwhen compared + FGF15inh to group, when compared to CD group, in non-steatotic livers. In steatotic livers, there were hepatic CD group, in non-steatotic livers. In steatotic livers, there were no differences in either no injury or regeneration differences parameters in either hepatic injury between or regeneration + FGF15inh the CDparameters and CD between thegroups (Figure 5A). CD + FGF15inh andACDrise in FGF15 groupsintestinal levels (Figure 5A). wasin A rise observed in the CDlevels FGF15 intestinal group when was compared observed in thetoCDthegroup Shamwhengroupcompared and this was mirrored to the Shamin circulatory FGF15 group and this was(Figure mirrored5B). Induction of in circulatory CD in FGF15 donors (Figure did 5B). not leadoftoCD Induction intestinal injury in donors prior to the did not leadprocurement of liver to intestinal injury grafts. prior to the procurement of liver grafts. Figure 5. Figure 5. FGF15 FGF15 levels levels and andintestinal intestinaldamage damagebefore before procurement procurementfrom DCDs. from DCDs.(A) (A) FGF15 levels FGF15 in in levels liver and liver injury and regeneration parameters. (B) FGF15 in plasma and intestine and liver and liver injury and regeneration parameters. (B) FGF15 in plasma and intestine and MDA and MDA and MPO in MPO in intestine. intestine. ** pp<<0.05 0.05vs. vs.Sham; Sham;+ p+