lipocalin-2在肝再生中的作用

O R I G I N A L A R T I C L EThe role of lipocalin-2in liver regenerationKatrin Kienzl-Wagner 1*,Alexander R.Moschen 2,3*,Sabine Geiger 2,3,Alexandra Bichler 3,Felix Aigner 1,Gerald Brandacher 4,Johann Pratschke 1and Herbert Tilg 2,31Department of Visceral,Transplant and Thoracic Surgery,Center of Operative Medicine,Innsbruck Medical University,Innsbruck,Austria 2Christian Doppler Research Laboratory for Gut Inflammation,Innsbruck Medical University,Innsbruck,Austria3Department of Internal Medicine I (Gastroenterology,Endocrinology &Metabolism),Innsbruck Medical University,Innsbruck,Austria 4Department of Plastic and Reconstructive Surgery,Johns Hopkins University School of Medicine,Baltimore,MD,USAKeywordsbiomarker –hepatic proliferation –LCN2–lipocalin-2–liver regeneration –partial hepatectomyCorrespondenceHerbert Tilg,MD,Department of Internal Medicine I,Innsbruck Medical University,Anichstrasse 35,Innsbruck 6020,Austria Tel:+4351250423540Fax:+4351250423538e-mail:herbert.tilg@i-med.ac.at Received 13November 2013Accepted 1July 2014DOI:10.1111/liv.12634AbstractBackground &Aims:Various immune mediators such as interleukin-6(IL-6)have been implicated in the process of liver regeneration.Lipocalin-2(LCN2)has been recently characterized as a prototypic immune mediator produced by various cell types being involved mainly in host defence.In addition,numerous studies have demonstrated its clinical value as a biomar-ker.This study aimed at defining the role of LCN2in liver regenera-tion.Methods:We studied LCN2expression in wild-type mice in a model of partial hepatectomy (PH).Furthermore,we evaluated liver regeneration after PH in LCN-deficient mice compared to littermate controls.Serum levels of LCN2were assessed in a small group of patients undergoing hepatic resec-tion.Results:LCN2is dramatically induced in livers and sera of wild-type mice after PH,whereas liver LCN2-receptor expression was decreased.Sham operations did not affect hepatic and serum LCN2expression.Although LCN2-deficient mice exhibited increased baseline liver expression indices,LCN2-deficient mice did not differ from wild-type mice with respect to hepatic proliferation suggesting that this molecule is not involved in hepatic repair.Only serum IL-1b levels were slightly lower in LCN À/Àmice,whereas IL-6serum levels did not differ between various tested animal groups.In humans undergoing hepatic resection,LCN2levels increased significantly within 24h following surgery.Conclusions:LCN2,although massively induced in mice after PH,is not relevant in murine hepatic regeneration.Further,human studies have to define whether LCN2could evolve as biomarker after liver surgery.The liver has the unique ability to fully regenerate after major tissue loss induced by viruses,toxins,autoim-mune disease,resection of primary or metastatic tumours as well as in the context of partial liver trans-plantation (1).Enhancing the regenerative capacity of the liver is crucial for the development of specific thera-pies for chronic liver disease and expanding indications for liver resection.Even though the last years have seen remarkable progress in characterizing the molecular and cellular mechanisms of liver regeneration,the multiple components and complex networks that orchestrate the regeneration process remain to be further elucidated.The current understanding of liver regeneration encompasses various cytokines,growth factors and met-abolic networks that link liver function with cell growth and proliferation.In normal liver regeneration as seenafter partial hepatectomy (PH),the liver mass is replen-ished by replication of existing hepatocytes without involvement of stem or progenitor cells (2).Upon liver resection,expression of TNF-a and IL-6is induced in Kupffer cells by various mechanisms including lipopoly-saccharide resulting in activation of STAT3in hepato-cytes and subsequent entry into the cell cycle.Key mitogenic factors that drive cell cycle progression are hepatocyte growth factor,epidermal growth factor and vascular endothelial growth factor.Termination of hepatocyte proliferation is controlled by the action of transforming growth factor b and activins (1,3–5).Mounting evidence suggests that lipocalin-2(LCN2)is an important modulator of proliferative and regenera-tive processes.LCN2is a 25kDa glycoprotein of a b -barrel shaped structure with the ability to bind small hydrophobic ligands and to form complexes with soluble macromolecules (6).LCN2so far has been ascribed great functional diversity.It has been shown to*Both authors contributed equally to this study.Liver International (2014)©2014John Wiley &Sons A/S.Published by John Wiley &Sons Ltd1Liver International ISSN1478-3223be a component of the innate immune system(7),it is highly induced during the acute phase response(8),and it acts as an immunomodulator in inflammatory colo-rectal diseases(9).Furthermore,LCN2regulates the inflammatory response during ischaemia and reperfu-sion of transplanted hearts(10).It was further demon-strated that hypoxia and anaemia induce LCN2 expression(11),whereas it inhibits differentiation of erythroid progenitor cells(12)and induces apoptosis (13).Numerous studies have proposed LCN2as a potent biomarker for acute renal injury(14–16).Recent evidence further implicates a role for LCN2in obesity, insulin resistance and hyperglycaemia(17).Previous studies have established LCN2as a growth regulator.LCN2has been associated with tissue involu-tion(18),it induces the epithelial to mesenchymal tran-sition,thereby actively influencing cancer progression by affecting tumour invasion and metastasis(19),and LCN2facilitates mucosal regeneration by promoting cell migration(20).In the developing kidney,LCN2pro-motes the organization of epithelial cells into tubular structures(21),whereas in chronic kidney disease it seems to be a key effector in aberrant growth(22).Dur-ing the regenerative process after renal ischaemia-reper-fusion injury,LCN2acts both as an inducer and inhibitor of renal cell proliferation(23,24).Data on the potential role of LCN2in liver homoeo-stasis and liver disease are limited.LCN2expression is strongly induced in models of acute and chronic liver injury(25–28)and it is significantly increased in liver biopsies from patients with chronic HCV(29)or HCC (30).To define the role of LCN2in liver regeneration, we characterized the regulation of LCN2in wild-type mice after2/3PH.Furthermore,we investigated the impact of heterozygote or homozygote LCN2deficiency on liver regeneration.MethodsMiceAll animal experiments described were performed in accordance with Austrian legal requirements after approval by the Austrian authorities(licensed to AR Moschen No BMWF-66.011/0040-II/10b/2009).All mice were maintained at the central animal facility of the Medical University Innsbruck under specific patho-gen-free conditions with ad libitum access to standard chow and water.The generation of LCN2À⁄Àmice has been described elsewhere(7).For all experiments,male littermates were used.At the time of surgery,mice were 12–14-week old with a body weight of20–30g.Murine2/3PHAll surgical procedures were performed under isoflurane inhalation anaesthesia using a vaporizer for laboratory animals(Provet,Lyssach,Switzerland)and a micro-scope with25-to50-fold magnification(Leica MZ12.5, Wetzlar,Germany).The2/3PH was performed as described previously(31,32).Briefly,the liver is exposed through midline laparotomy and the gallbladder is removed.For70%hepatectomy,the right and left upper lobe(median lobe)and the left lateral liver lobe are ligated at the pedicle of the respective liver lobe(5/0silk) and resected.Resected material was preserved as index liver tissue for further work-up.After irrigating the abdomen with sterile saline,the abdomen is closed in two layers(running5/0PDS suture for peritoneum and fascia,running5/0seralon suture for skin).To replace intra-operativefluid loss,the mouse is given2–3ml sterile saline subcutaneously.Sham procedures consisted of anaesthetizing mice followed by midline laparotomy and manipulating but not resecting the liver.Sample preparationMice were injected intraperitoneally with 2.5mg/ml BrdU(BD Biosciences,San Diego,CA,USA)2h before organ retrieval.After the indicated time points,mice were anaesthetized with isoflurane.Following relaparot-omy,blood was collected from the vena cava inferior. The remnant regenerative liver was inspected in situ, carefully resected,and processed according to protocol. Both index and follow-up liver tissue were systematically worked up to befixed in10%buffered formalin,stored in RNA-later,or snap-frozen for protein work-up.ImmunohistochemistryFour micrometre sections were prepared from formalin fixed,paraffin-embedded specimens of liver tissue sam-ples.Sections were deparaffinized in xylene and rehy-drated in graded alcohols.Antigen retrieval was achieved by microwaving slides for10min in antigen unmasking solution(Vector Laboratories,Burlingame,CA,USA). Endogenous peroxidase activity was quenched with3% H2O2in PBS for10min.Non-specific protein binding sites were blocked using serum-free protein block(Dako, Glostrup,Denmark).The followingfirst antibodies were applied:specific rabbit polyclonal anti-LCN2antibody (provided by Professor Marit Nilsen-Hamilton,Iowa State University,Ames IA,USA)and anti-PCNA(Prolif-erating Cell Nuclear Antigen)Ab-1,clone:PC10 (Thermo Lab Vision,Rockford,IL,USA).Specificity was demonstrated by using either isotype-specific con-trol antibodies or recombinant LCN2for blocking experiments.HRP-conjugated anti-mouse or anti-rabbit pAb(Dako)were incubated for60min at room temper-ature.LCN2or PCNA were visualized with3,30-diam-inobenzidine(DAB)as a chromogen.Slides were counterstained with haematoxylin QS(both Vector La-borytories),dehydrated and mounted permanently in Eukitt(O.Kindler GmbH,Freiburg,Germany). Nuclear incorporation of BrdU was visualized on paraffin-embedded tissue sections using a BrdU In-SituLiver International(2014)©2014John Wiley&Sons A/S.Published by John Wiley&Sons Ltd2Lipocalin-2in liver regeneration Kienzl-Wagner et al.Detection Kit(BD Pharmingen,San Diego,CA,USA) according to the manufacturer’s instructions.Pictures were acquired on a Zeiss AX10(Imager A1,Zeiss,Ober-kochen,Germany)with high-resolution camera and Ax-ioVision software.The number of BrdU-or PCNA-positive nuclei was determined using an Optimas image analysis package(Weiss imaging and solutions GmbH, Bergkirchen/Guending,Germany).mRNA and real-time qPCRTotal RNA was extracted from liver tissue using1ml/ 100mg TRIZOL reagent according to the manufac-turer’s instructions(Life Technologies,Carlsbad,CA, USA).One microgram of total RNA was reverse tran-scribed into cDNA using200units of moloney murine leukaemia virus reverse transcriptase(Life Technolo-gies)with random hexanucleotide primers(Roche, Mannheim,Germany)as previously described(33). Primers were designed using Primer Express software (Applied Biosystems,Foster City,CA,USA)and Fast-PCR professional(University of Helsinki). Quantitative real-time PCR(qPCR)was performed in a total volume of25l l in40cycles(95°C for30s fol-lowed by60°C for60s)using a Mx3000P bioanalyzer (Stratagene,now Agilent Technologies,Santa Clara,CA, USA).Master mixes were from Eurogentec(Seraign,Bel-gium).Expression levels were calculated using the stan-dard-curve method using serial dilutions of positive control cDNA(from stimulated splenocytes)as a stan-dard.The abundance of each mRNA was expressed as ratio to beta glucuronidase(Gusb)mRNA expression. The following primers have been used:Lcn2forward:50-GCC TCA AGG ACG ACA ACA TCA-30,Lcn2reverse: 50-CAC CAC CCA TTC AGT TGT CAA-30,Lcn2probe: 50-FAM-TTC TCT GTC CCC ACC GAC CAA TGC BHQ1-30,Lcn2R forward:50-GGC GAT TTC TAC AGC GAA TGA-30,Lcn2R reverse:50-CTA TCA GCC ACC GTG CAG ACT-30,Lcn2R probe:50-FAM-CCT CTT CCT GTT TTA TGG CTG GCC TGG T-BHQ1-30.Analysis of circulating LCN2and cytokines(IL-1b,IL-6, TNF-a)Venous blood was collected under microscopic guid-ance from the vena cava inferior and centrifuged at 2500g.Serum was stored atÀ80°C until assayed.Mouse LCN2(NGAL)was assayed using commercially avail-able antibody pairs from R&D Systems(DuoSet Devel-opment Kit,Minneapolis,MN,USA).IL-1b,IL-6and TNF-a were determined using BD OptEIA TM assays con-sisting of capture and detection antibodies along with streptavidin-horseradish peroxidase(HRP)conjugate. The respective recombinant cytokines served as a stan-dard(BD Biosciences,PharMingen).The absorption was read with a PowerWave HT Microplate Spectropho-tometer(Biotek,Winooski,VT,USA)at450nm with a reference wavelength of570nm.Western BlotProtein was extracted homogenizing liver tissue in M-PER Mammalian Protein Extraction Reagent con-taining Halt Protease Inhibitor Cocktail(Thermo Scien-tific,Pierce,Rockford,IL,USA)according to the manufacturer’s instructions.Tissue lysates were sepa-rated under reducing conditions on12.5%acrylamide gels using a mini-PROTEAN3electrophoresis system (Bio-Rad,Hercules,CA,USA).Gels were transferred to Hybond-P PVDF membranes(GE Healthcare,Uppsala, Sweden),blocked with 2.5%BSA containing0.1% Tween.Blots were incubated at4°C o/n applying one of the following primary antibodies:specific rabbit poly-clonal anti-LCN2(provided by Professor Marit Nilsen-Hamilton,Iowa State University)or anti-GAPDH (clone14C10;Cell Signaling Technology,Danvers,MA, USA).HRP-conjugated anti-rabbit secondary antibodies were used for secondary detection(Dako).Bands were visualized using a chemiluminescent substrate on a Hyperfilm ECL(both GE Healthcare)that was automat-ically processed on an X-rayfilm developer.Densitome-try was performed with ImageJ(34).Human serum samplesFrom March2014to May2014,five patients undergoing hepatic resection were enrolled at the Department of Visceral,Transplant and Thoracic Surgery,Innsbruck Medical University.This study protocol was reviewed and approved by the local ethics committee(AN2014-0024).All patients provided written informed consent for the participation of this study.Human serum sam-ples were collected the day prior to surgery as well as24 and48h post-operatively.Blood samples were centri-fuged at2500g.Serum was stored atÀ20°C until assayed.Human LCN2(NGAL)was assayed using com-mercially available human LCN2ELISA from R&D Sys-tems(according to manufacturer’s instructions).StatisticsThe differences among groups were analysed by Mann–Whitney U-test and,where appropriate,by Kruskal–Wallis ANOVA.Significance was assumed for P<0.05. All data analyses were carried out using the IBM SPSS Statistics19.0software package(IBM,Armonk,NY, USA)and graphical illustrations were prepared using GraphPad Prism version6(GraphPad Software, Jolla,CA,USA).ResultsLipocalin-2is strongly induced during liver regeneration We investigated the regulation of LCN2during liver regeneration.LCN2was strongly induced as early as6h after PH and peaked after12h reaching52.9-foldLiver International(2014)©2014John Wiley&Sons A/S.Published by John Wiley&Sons Ltd3 Kienzl-Wagner et al.Lipocalin-2in liver regenerationhigher levels over baseline controls.Thereafter,expres-sional levels declined until 48h,plateaued but remained 7.9-fold elevated at 96h (Fig.1A).The expression of the LCN2receptor (LCN2R)was slightly reduced (2.1-fold after 6h,2.7-fold after 12h,2.2-fold after 24h and 2.5-fold after 48h)after PH but returned to baseline levels at 72h (Fig.1B).RNA data were confirmed by Western blot analysis.Accordingly,protein concentra-tions peaked at 12h as shown by Western blot analysis (Fig.1D).LCN2protein levels gradually declined until(A)(B)(C)(D)(E)(G)(F)Fig.1.(A)mRNA was extracted from liver tissue at the indicated time points after PH and relative LCN2expression was determined by quan-titative real-time PCR (purple line,n =5per group).Steady state levels were assayed from index liver tissue that was obtained at the time of resection (blue line,n =5per group);*P <0.05,**P <0.01.(B)Index and follow-up liver tissue mRNAs as mentioned above were assayed for Lipocalin receptor (LCN2R)mRNA (n =5per group,*P <0.05,**P <0.01).(E)Serum was collected at the indicated time points after PH and circulating LCN2concentrations were determined by ELISA.(C)Hepatic LCN2and LCN2R mRNA expression were compared between mice that underwent PH (black bars)and sham-operated controls (white bars).(D)Liver tissue was collected at the indicated timepoints after PH and protein was extracted and LCN2content determined by Western Blot.Protein content of Gapdh served as a loading control.Protein concentration was quantitated by densitometry.(F)Hepatic LCN2protein expression was determined by immunohistochemistry (brown:LCN2,blue:nuclei).(G)Circulating LCN2levels in patients undergoing liver resection.Serum samples were obtained prior to hepatic surgery,24and 48h after liver resection.Concentration of circulating LCN2was determined by ELISA.Liver International (2014)©2014John Wiley &Sons A/S.Published by John Wiley &Sons Ltd4Lipocalin-2in liver regeneration Kienzl-Wagner et al.96h(Fig.1D).As determined by immunohistochemis-try,hepatic LCN2positivity was strongest after24h (Fig.1F).In addition to strong LCN2regulation,we observed significant microvesicular steatosis following PH occurring12–24h after hepatic resection(Fig.1F). Circulating LCN2levels were determined by ELISA.As outlined in Fig.1E,the highest serum levels were found 24h after PH.To rule out a significant contribution of the surgical manipulation on LCN2and LCN2R levels, we performed sham operations.As depicted in Fig.1C, laparotomy on its own did not significantly impact hepatic LCN2or LCN2R expression.To elucidate whether LCN2is also regulated following liver resection in humans,we assessed LCN2serum levels infive patients undergoing hepatic resection of different extent for variable aetiologies.Patient characteristics are pro-vided in Table1.LCN2was readily detectable by ELISA in all samples.There was a significant increase in circu-lating LCN2expression24h post-liver resection com-pared to pre-operative LCN2levels(P<0.05;Fig.1G).Impact of LCN2deficiency on liver regenerationWe next investigated the impact of LCN2deficiency on liver regeneration.Therefore,wild-type mice(LCN2+⁄+), mice heterozygote for LCN2(LCN2+⁄À)and LCN2 knockout mice(LCN2À⁄À)were subjected to PH.As shown in Fig.S1,the number of BrdU-positive cells peaked48h after PH indicating the time point of maxi-mum hepatic regeneration and was consequently used for further analysis.At the same time point,we also defined the maximum positivity for PCNA(data not shown).Interestingly,LCN2À/Àmice demonstrated sig-nificantly elevated baseline liver regeneration compared to LCN2+/+and LCN2+/Àlittermates as demonstrated by increased numbers of PCNA-positive hepatic nuclei (Fig.2A).However,no significant differences in hepatic proliferation were observed at48h after PH between wild-type(LCN2+⁄+),LCN2+⁄Àand LCN2À⁄Àlittermates with regard to number of PCNA-positive cells(Fig.2B) and BrdU-positive nuclei(Fig.2C).Impact of LCN2genotype on circulating cytokine levels Pro-inflammatory cytokines have been shown to be potent inducers of the initiation phase of liver regeneration.As LCN2has been implicated in the regu-lation of immune functions,we hypothesized that the LCN2genotype might affect the cytokine profile after PH.As shown in Fig.2D–F,TNF-a serum concentra-tions peaked6h after PH,whereas IL-1b and IL-6 showed their peaks after12h.Interestingly,the overall IL-1b response was higher in wild-type mice compared to LCN2+⁄Àand LCN2À⁄Àmice(Fig.2D).While IL-1b was detectable until48h in wild-type mice,it was unde-tectable in knockout mice after12h.No significant dif-ferences were seen for IL-6concentrations(Fig.2E). Regarding circulating TNF-a concentrations,lowest lev-els were observed in mice heterozygote for LCN2 (Fig.2F).Liver cytokine expression data after12,24 and48are depicted in Fig.S2and did not show any differences between study groups.DiscussionRecent evidence demonstrates that LCN2expression shows clear correlations with liver damage.LCN2 expression was significantly increased in acute and chronic liver injury models(26,27)as well as in chronic HCV infection(29).A recent study observed a substan-tial increase in LCN2liver expression after PH in rats (25),although the role of LCN2remains unclear.We now also observe that PH is followed by an impressive LCN2induction.As extensive LCN2overexpression did not influence hepatocyte proliferation at48h post-PH in LCN2À/Àmice,the reasons for this prominent LCN2 upregulation need to be explored:whether it is just a bystander effect or whether it is necessary for liver regeneration.One possible explanation is that LCN2is in fact a mediator of liver regeneration and constitutes another redundant pathway(4).The idea that LCN2influences liver regeneration is supported by the fact that LCN2À/Àmice exhibit significantly higher baseline regeneration compared to their wild-type littermates.Consequently, LCN2was supposed to act as an inhibitor of hepatocyte proliferation in a way that under physiological condi-tions LCN2prohibits excess hepatocyte proliferation to maintain liver homoeostasis.In line with these data,Mi-harada et al.demonstrated in a mouse model of acute haemolytic anaemia that under normal physiological conditions,LCN2secreted by erythroid progenitor cellsTable1.Characteristics of patients undergoing liver surgeryPatient Age Sex Diagnosis PVE/PVL Liver resection158Male HCC No Segment II/III extended to IVa 272Male Recurrent HCC Yes Right hemihepatectomy334Female Donation for LDLT No Segment II/III471Male Hepatic metastases from CRC No Atypical resection II/III+VI 562Female CCC Yes TrisegmentectomyPVE,portal vein embolization;PVL,portal vein ligature;HCC,hepatocellular carcinoma;CCC,cholangiocellular carcinoma;CRC,colorectal cancer; LDLT,living donor liver transplantation.Liver International(2014)©2014John Wiley&Sons A/S.Published by John Wiley&Sons Ltd5 Kienzl-Wagner et al.Lipocalin-2in liver regenerationin bone marrow suppressed the survival and differentia-tion of erythroid progenitors.Once an urgent haemato-poiesis is required,the erythroid progenitors appear to escape from LCN2-mediated suppression thereby per-mitting rapid production of mature cells (12).However,considering LCN2an inhibitor of liver regeneration under physiological conditions neither explains the dramatic increase in LCN2expression following 70%hepatectomy nor the absent impact on hepatocyte proliferation at 48h post-PH in LCN2À/Àmice.Hypothetically,LCN2could exert a dual role in the process of liver regeneration as has been proposed for renal regeneration.Depending on the phase of reper-fusion (early te)LCN2acts as an inducer and inhibitor of renal cell proliferation.LCN2blockade with anti-LCN2antibody in the late regenerative phase(A)(B)(C)(D)(E)(F)Fig.2.(A)Liver tissue was resected from wild-type,LCN2+⁄Àand LCN2À⁄Àmice at the time of PH.Tissue sections were stained for prolifera-tive cell nuclear antigen (PCNA).Pictures were acquired at 1009magnification and the number of PCNA-positive nuclei was determined using an Optimas software package.(B)Livers were harvested from wild-type,LCN2+⁄Àand LCN2À⁄Àmice 48h after PH.Paraffin-embedded tissue sections were stained for PCNA.The number of PCNA-positive cells was determined using an Optimas software package.(C)Mice were injected with BrdU 2h before livers were collected and processed.Liver sections were stained for BrdU.Absolute numbers of positive nuclei per 1009high-power field was determined with Optimas.(D –F)Serum samples were obtained 6,12,24and 48h after PH.Concen-trations of circulating cytokines were determined by ELISA.(D)IL-1b ,(E)IL-6,(F)TNF-a concentrations.*P >0.05.Liver International (2014)©2014John Wiley &Sons A/S.Published by John Wiley &Sons Ltd6Lipocalin-2in liver regeneration Kienzl-Wagner et al.reduced renal cell proliferation,whereas anti-LCN2 antibody administration in the early inflammatory phase had the opposite effect with increased expression of proliferation markers(23).Adopting thesefindings from renal regeneration to our data would propose LCN2as an inhibitor of hepatocyte proliferation in physiological liver homoeostasis and an inducer–though redundant–of liver regeneration upon PH. Alternatively,the fact that LCN2induction does not alter hepatocyte regeneration at48h post-PH which represents the time of maximum proliferative activity suggests that LCN2upregulation in the liver and serum upon major hepatic resection simply is a bystander effect.Already in1995,Liu et al.identified SIP24/24p3 as an acute phase protein(8).Major hepatic surgery with resection of70%of the liver parenchyma is of course an appropriate stimulus that elicits an acute phase response.However,as shown in our sham opera-tions,midline laparotomy and manipulation of the vis-ceral organs on its own did not significantly induce hepatic LCN2expression.Furthermore,LCN2has been shown to be strongly upregulated during anaemia.In a mouse model of acute hypovolaemic anaemia Jiang et al.found LCN2mRNA dramatically increased in the liver of phlebotomized mice and LCN2protein levels were increased in the cor-responding sera(11).Major surgery such as70%hepa-tectomy is inevitably accompanied by a certain blood loss,solely because of the fact that a significant amount of portal,arterial and venous blood is trapped within the three liver lobes that are being resected.This state of post-operative hypovolaemia may therefore very well explain the striking induction of LCN2that we observed upon PH.In addition to strong LCN2regulation,we observed significant microvesicular steatosis following PH occur-ring12–24h after hepatic resection.Liver regeneration following PH is physiologically associated with transient microvesicular steatosis(35–37).which seems to be required for the regenerative process.The injured liver produces cytokine signals that trigger the release of fatty acids from adipose tissue into circulation that are then taken up by hepatocytes and stored as triacylglycerols. Their hydrolysis releases fatty acids for mitochondrial b-oxidation and ATP production and provides substrate for the biosynthesis of membrane phospholipids to sup-port rapid cell division(38).Strikingly,LCN2expres-sion has been linked to obesity and non-alcoholic fatty liver disease(17,39).As the peak of steatosis within the regenerating liver correlates with the peak of hepatic LCN2expression upon PH,the striking LCN2induction that we observed could be interpreted as an epiphenom-enon of the transient microvesicular steatosis that is physiologically associated with liver regeneration. Previous studies have established the role of IL-1b to induce LCN2expression via the NF-j B pathway (40–42).In line with thesefindings,serum IL-1b concentrations peaked at12h after PH.As IL-1b is known to exert a mitoinhibitory effect on hepatocyte proliferation,the fact that the overall IL-1b response was markedly higher in wild-type mice compared to LCN2-deficient littermates might explain why LCN2À/Àmice demonstrated significantly elevated baseline liver regeneration.In summary,our data convincingly demonstrate that LCN2is potently expressed after PH.We further observed a significant increase in circulating LCN2levels 24h post-liver resection in humans suggesting that lar-ger prospective studies using LCN2as a potential bio-marker for liver regeneration should be initiated. Studies in LCN2-deficient mice,however,clearly dem-onstrate that this acute phase protein is not involved in hepatic regeneration.AcknowledgementsConflict of interest:The authors do not have any disclo-sures to report.References1.Bohm F,Kohler UA,Speicher T,Werner S.Regulation ofliver regeneration by growth factors and cytokines.EMBO Mol Med2010;2:294–305.2.Zaret KS,Grompe M.Generation and regeneration of cellsof the liver and pancreas.Science2008;322:1490–4.3.Fausto N,Campbell JS.The role of hepatocytes and ovalcells in liver regeneration and repopulation.Mech Dev 2003;120:117–30.4.Fausto N,Campbell JS,Riehle KJ.Liver regeneration.Hepatology2006;43(2Suppl.1):S45–53.5.Clavien PA,Petrowsky H,Deoliveira ML,Graf R.Strate-gies for safer liver surgery and partial liver transplantation.N Engl J Med2007;356:1545–59.6.Flower DR.The lipocalin protein family:structure andfunction.Biochem J1996;318(Pt1):1–14.7.Flo TH,Smith KD,Sato S,et al.Lipocalin2mediates aninnate immune response to bacterial infection by seques-trating iron.Nature2004;432:917–21.8.Liu Q,Nilsen-Hamilton M.Identification of a new acutephase protein.J Biol Chem1995;270:22565–70.9.Nielsen BS,Borregaard N,Bundgaard JR,et al.Inductionof NGAL synthesis in epithelial cells of human colorectal neoplasia and inflammatory bowel diseases.Gut1996;38: 414–20.10.Aigner F,Maier HT,Schwelberger HG,et al.Lipocalin-2regulates the inflammatory response during ischemia and reperfusion of the transplanted heart.Am J Transplant 2007;7:779–88.11.Jiang W,Constante M,Santos MM.Anemia upregulateslipocalin2in the liver and serum.Blood Cells Mol Dis 2008;41:169–74.12.Miharada K,Hiroyama T,Sudo K,Nagasawa T,Nakam-ura Y.Lipocalin2functions as a negative regulator of red blood cell production in an autocrine fashion.FASEB J 2005;19:1881–3.13.Devireddy LR,Gazin C,Zhu X,Green MR.A cell-surfacereceptor for lipocalin24p3selectively mediates apoptosis and iron uptake.Cell2005;123:1293–305.Liver International(2014)©2014John Wiley&Sons A/S.Published by John Wiley&Sons Ltd7 Kienzl-Wagner et al.Lipocalin-2in liver regeneration。