Loss of histone acetyltransferase cofactor transformation/transcription domain-associated protein...

Loss of Histone Acetyltransferase CofactorTransformation/Transcription Domain-Associated

Protein Impairs Liver Regeneration After Toxic InjuryVivek Shukla,1,2 Cyrille Cuenin,1 Nileshkumar Dubey,2 and Zdenko Herceg1

Organ regeneration after toxin challenge or physical injury requires a prompt and balancedcell-proliferative response; a well-orchestrated cascade of gene expression is needed to regu-late transcription factors and proteins involved in cell cycle progression and cell proliferation.After liver injury, cell cycle entry and progression of hepatocytes are believed to require con-certed efforts of transcription factors and histone-modifying activities; however, the actualunderlying mechanisms remain largely unknown. The purpose of our study was to investi-gate the role of the histone acetyltransferase (HAT) cofactor transformation/transcription do-main-associated protein (TRRAP) and histone acetylation in the regulation of cell cycle andliver regeneration. To accomplish our purpose, we used a TRRAP conditional knockoutmouse model combined with toxin-induced hepatic injury. After we treated the mice with acarbon tetrachloride toxin, conditional ablation of the TRRAP gene in those mice severelyimpaired liver regeneration and compromised cell cycle entry and progression of hepatocytes.Furthermore, loss of TRRAP impaired the induction of early and late cyclins in regeneratinglivers by compromising histone acetylation and transcription factor binding at the promotersof the cyclin genes. Our results demonstrate that TRRAP and TRRAP/HAT-mediated acetyla-tion play an important role in liver regeneration after toxic injury and provide insight intothe mechanism by which TRRAP/HATs orchestrate the expression of the cyclin genes duringcell cycle entry and progression. (HEPATOLOGY 2011;53:954-963)

After toxin challenge or physical damage, tissue andorgan regeneration requires a well-orchestrated cas-cade of gene expression regulating transcription fac-

tors and proteins involved in cell cycle progression and cellproliferation.1 A vast majority of hepatocytes in the adultliver are highly differentiated cells that are in a quiescent(nonproliferative) state and rarely divide.2 On the otherhand, hepatocytes have a capacity to rapidly reenter the cellcycle and proliferate in a highly synchronized manner afteracute liver injury, such as damage induced by chemical ex-posure or partial hepatectomy.3 Thus, the lost hepatocytescan be replaced rapidly, and a damaged liver can regeneratewithin a few days.3 However, the mechanisms underlyingliver regeneration and the molecular participants that governthis process remain poorly understood.One of the most widely used approaches to studying

the mechanism of liver regeneration after injury inrodents is treatment with carbon tetrachloride(CCl4).

4,5 CCl4 is metabolized in the centrilobularzone of the liver, where the production of trichloro-methyl radicals leads to necrotic death of pericentralhepatocytes.6 These events stimulate liver cells, princi-pally hepatocytes, within the periportal and intermedi-ate zones to synchronously exit the quiescent state (G0

phase), reenter the cell cycle, and undergo replicationbefore returning to the G0/G1 phase.

1

Abbreviations: BrdU, 5-bromo-2-deoxyuridine; CCl4, carbon tetrachloride;CDK, cyclin dependent kinase; ChIP, chromatin immunoprecipitation; Cre,Type I topoisomerase from P1 bacteriophage; ES cells, embryonic stem cells;HAT, histone acetyltransferase; PCNA, proliferating cell nuclear antigen; pIpC,polyinosinic-polycytidylic acid; TRRAP, transformation/transcription domain-associated protein; TRRAP-CO, TRRAP-containing (control) mice; TRRAP-CKO, TRRAP conditional knockout mice.From the 1International Agency for Research on Cancer (IARC), Lyon, France;

and 2Department of Gastroenterology, Hepatology & Nutrition, Unit 1466,University of Texas MD Anderson Cancer Center, Houston, TX.Received September 10, 2010; accepted December 2, 2010.Supported by the Association pour la Recherche sur le Cancer (ARC, France),

the Association for International Cancer Research (AICR UK), la LigueNationale (Francaise) contre le Cancer (France), and l’Agence Nationale deRecherhe Contre le Sida et Hepatites Virales (ANRS, France).Address reprint requests to: Z. Herceg, IARC, 150 Cours Albert Thomas,

69008, Lyon, France. E-mail: [email protected]; or V. Shukla, Department ofGastroenterology, Hepatology & Nutrition, Unit 1466, University of Texas MDAnderson Cancer Center, Houston, TX. E-mail: [email protected] .CopyrightVC 2010 by the American Association for the Study of Liver Diseases.View this article online at wileyonlinelibrary.com.DOI 10.1002/hep.24120Potential conflict of interest: Nothing to report.Additional Supporting Information may be found in the online version of

this article.

954

Cell cycle entry and progression of hepatocytes afterliver injury are believed to require concerted efforts oftranscription factors and histone-modifying activ-ities.4,7 Histone modifications including acetylationplay an important role in these processes, and histoneacetyltransferases (HATs) are known to be importantin the control of gene transcription and cell cycle pro-gression.8,9 The HAT cofactor transformation/tran-scription domain-associated protein (TRRAP) has beenidentified as an essential gene in development10 and acomponent of several HAT complexes (includingGCN5/PCAF and Tip60/NuA4).9 Transcription fac-tors important for cell cycle entry and progression,including c-Myc and E2F, utilize the TRRAP/HATcomplexes to activate transcription of their down-stream target genes.11-13 This transcription activationis the result of these complexes catalyzing acetylationof the histone proteins in the nucleosomes spanningthe target genes. Studies have indicated that TRRAPmay function by recruiting HATs to c-Myc and E2Fand possibly to other transcription factors, resulting inan open chromatin configuration and increased tran-scription.9,14,15 However, the mechanism and func-tional significance of the TRRAP/HAT recruitmentduring organ regeneration remain largely unknown.The purpose of our study was to define the biological

function of TRRAP and histone acetylation in activatingtranscription and cell cycle reentry in liver cells under-going regeneration in vivo. To accomplish our purpose,we used an inducible TRRAP conditional knockoutmouse (TRRAP-CKO) model combined with toxin-induced hepatic injury. Our results revealed that TRRAPplays an essential role in cell cycle reentry and regenera-tion of the adult liver following acute liver damage andthat the mechanism by which TRRAP participates inliver regeneration involves histone modifications and therecruitment of transcription factors to chromatin.

Materials and MethodsGeneration of TRRAP-CKO Mice. TRRAP-CKO

mice were generated by crossing TRRAPþ/f mice har-boring a TRRAP conditional (floxed; f ) allele withboth TRRAPf/D mice (containing one TRRAP floxed[f ] and one deleted [D] allele)10 and Cre transgenicmice that express Cre recombinase under the controlof Mx promoter16 to obtain TRRAPf/DCreþ mice. Ge-notypes of all mice were assessed by southern blottingand polymerase chain reaction (PCR) analysis of tailDNA, as described.10

TRRAP Disruption in Liver. TRRAPf/DCreþ micewere injected intraperitoneally with 250 lg polyinosinic-

polycytidylic acid (pIpC; Sigma) in phosphate-bufferedsaline (PBS), 3 times at 2-day intervals. Of note, Mx-Cre-mediated recombination also occurs in other inter-feron-responsive tissues,16 therefore activation of Cre bypIpC injection may induce TRRAP deletion in severaltissues,14 as well as the liver. In the control group of thesame genotype (TRRAPf/DCreþ), only PBS was injectedintraperitoneally. As another control, TRRAPf/D micelacking Cre (TRRAPf/DCre�) were injected with pIpC.Animals, CCl4 Injection, and Preparation of Nu-

clear Extracts. All the mice were maintained as approvedby the Animal Care and Use Committee of the Interna-tional Agency for Research on Cancer (Lyon, France)(ACUC 03/4). Mice were divided into three groups:Group 1 ¼ TRRAPf/DCreþ without pIpC injection;Group 2 ¼ TRRAPf/DCreþ with pIpC (to delete TRRAP);and Group 3 ¼ TRRAPf/DCre� with pIpC (to comparethe effect of pIpC alone). At least three mice per groupwere examined per timepoint. Mice were fed a commer-cial diet and were given water ad libitum. To induce he-patic injury, 10 lL/g body weight of 10% solution ofCCl4 in olive oil or olive oil alone was injected intraperi-toneally 48 hours after the last pIpC injection, and themice were sacrificed at the times indicated. Forty-eighthours after the third injection of pIpC was considered 0hour for the collection of samples after CCl4 treatment.Mice were sacrificed and part of the liver was removedand fixed in 4% paraformaldehyde for paraffin-embeddedsectioning. Other parts of the liver were removed and fro-zen in liquid nitrogen and kept at �80�C for preparationof protein lysates and total RNA.Histology and Immunohistochemistry. Histologic

and immunohistochemical analyses were performed af-ter staining either with hematoxylin and eosin stain(H&E) or with Feulgen-solution (Schiff ’s base) orwere left unstained for marker analysis. 5-Bromo-2-deoxyuridine (BrdU) and proliferating cell nuclearantigen (PCNA) staining was performed asdescribed,17 using specific antibodies (Supporting Ta-ble 1) and Vectastatin ABC alkaline phosphatase kit orABC peroxidase kit (Vector Laboratories). At least5,000 cells were scored for BrdU and PCNA index.Western Blot Analysis. Western blot analysis of

liver nuclear proteins was carried out as described,13

using specific antibodies (Supporting Table 1).Chromatin Immunoprecipitation (ChIP) Assay. The

ChIP assay was performed as described18 and accord-ing to the manufacturer’s recommendations (UpstateBiotechnology, Lake Placid, NY), using polyclonalantibodies specific for histone modifications and tran-scription factors (Supporting Table 1). The recoveredDNA was then analyzed by PCR using primers

HEPATOLOGY, Vol. 53, No. 3, 2011 SHUKLA ET AL. 955

recognizing different regions of the cyclin A1 promoter(Supporting Table 2).

Statistical Analysis. For statistical analysis we usedStudent’s t test for comparison between groups. P-val-ues <0.05 were considered statistically significant.

ResultsLoss of TRRAP Severely Impairs Liver Regenera-

tion Following CCl4 Treatment. To study the func-tion of TRRAP and TRRAP-mediated histone acetyla-tion in transcription and cell proliferation during

Fig. 1. Loss of TRRAP severely impairs liver regeneration after CCl4treatment. (A) TRRAP-Co and TRRAP-CKO mice were injected with a singledose of liver toxin CCl4 (0 hours), and at indicated timepoints thereafter survival of the mice was monitored. (B) Cre-mediated deletion of TRRAPwas monitored by southern blotting on DNA extracted from brain (B), heart (H), and liver (L) samples taken from TRRAPf/f Mx-Creþ mice treatedeither with (þpIpC) or without pIpC (�pIpC). DNA was extracted from these tissues and subjected to Southern blotting to detect TRRAP delta(D) allele (upper panel). Depletion of TRRAP messenger RNA (mRNA) in bone marrow (BM), liver (L), and spleen (S) was monitored by perform-ing RT-PCR from TRRAP-CKO mice (lower panel). The hprt levels were used as an internal loading control. (C) Reduced survival in TRRAP-CKOmice. Asterisk (*) indicates that the differences are statistically significant. (D) Feulgen solution-stained sections of TRRAP-proficient (left andright panels) and of TRRAP-CKO livers (middle panel) at 36, 48, 72, and 96 hours after CCl4 treatment.

956 SHUKLA ET AL. HEPATOLOGY, March 2011

tissue regeneration in response to acute liver injury, weused TRRAP-CKO mice that allow inducible deletionin vivo of the TRRAP gene in a spatiotemporal man-ner (Fig. 1A).14 To avoid severe bone marrow defectand possible lethal phenotype caused by multiple pIpC

injections, we used older mice with TRRAP condi-tional allele (8 weeks) that remained viable and pheno-typically normal (see below). We found that 2 daysafter three pIpC injections the deletion of TRRAP washighly efficient in the liver (nearly 100%) and

Fig. 2. After liver damage, TRRAP-deficient hepatocytes exhibited defects in cell cycle reentry and proliferation. (A) Immunohistochemicaldetection of hepatocytes incorporating BrdU is shown with regenerating TRRAP-Co (left and right panels) and TRRAP-CKO livers (middle panel).Note that regenerating TRRAP-Co livers initiated DNA replication in hepatocytes 12 hours before and 36 hours after CCl4 treatment. (B) BrdUincorporation into hepatocyte DNA during liver regeneration in TRRAP-Co and TRRAP-CKO mice after CCl4 treatment. (C) BrdU incorporation innonparenchymal cells during TRRAP-Co, and TRRAP-CKO liver regeneration after CCl4 treatment. (D) Immunohistochemical detection of PCNA inregenerating TRRAP-Co (left and right panels) and TRRAP-CKO (middle panel) mouse livers. (E) PCNA-positive hepatocytes, i.e., PCNA index(PCNA positive cells/total number of cells �100) in TRRAP-Co, and TRRAP-CKO livers during regeneration induced by CCl4 treatment.


significantly less efficient in other organs such as brain,heart, and bone marrow as monitored by southernblotting reverse-transcription (RT)-PCR (Fig. 1B, anddata not shown).14 All TRRAP-CKO mice injectedwith three doses of pIpC remained viable for the dura-tion of the experiments. Thereafter, TRRAPf/DCreþ

mice treated with pIpC were designated TRRAP-CKOmice, whereas TRRAPf/DCreþ injected with PBS andTRRAPf/DCre� injected with pIpC were designated thecontrol group (TRRAP-Co) (Fig. 1A).

To examine the impact of TRRAP deletion on liverregeneration, we used a mouse model of toxic liverinjury induced by a single injection of liver toxinCCl4.

8,19 After we induced CCl4 damage, mice weresacrificed at different timepoints (Fig. 1A). Weobserved that TRRAP-deficient mice (TRRAP-CKO)exhibited significantly lower survival than did TRRAPcontaining control mice (TRRAP-Co) (Fig. 1C).Before CCl4 treatment, adult TRRAP-CKO liverswere histologically normal, and liver histology wasindistinguishable from that of TRRAP-containing con-trols (Fig. 1D; timepoint ¼ 0 hours), suggesting thatloss of TRRAP compromises mouse survival after toxicliver injury.Analysis of CCl4-induced damage revealed markedly

less regeneration in livers from TRRAP-CKO com-pared to TRRAP-Co mice (Fig. 1D). These resultsshow that loss of TRRAP impairs liver regenerationwithout altering the degree of initial liver injury andindicate that TRRAP may be an important factor inliver regeneration.TRRAP-Deficient Hepatocytes Exhibit Defects in

Cell Cycle Reentry and Proliferation Following LiverDamage. We next assessed cell proliferation in theregenerating liver (by BrdU incorporation and PCNAimmunostaining). Neither BrdU nor PCNA stainingoccurred in TRRAP-Co or TRRAP-CKO livers beforeCCl4 treatment (0 hours after CCl4 treatment), con-sistent with the cells being in the quiescent (G0) phase(Fig. 2A). Importantly, a sharp increase in hepatocyteproliferation in TRRAP-Co livers after CCl4 treatment(as judged BrdU and PCNA index) was markedlyimpaired in TRRAP-CKO livers (statistically signifi-cant, *P > 0.05) (Fig. 2A,B,D,E). Of note, DNA syn-thesis in nonparenchymal liver cells was also impairedin TRRAP-CKO mice compared to control mice (sta-tistical significance P > 0.05) after CCl4 injection(Fig. 2C). These results suggest that TRRAP is impor-tant for proliferation of both hepatocytes and nonpar-enchymal liver cells during liver regeneration.Loss of TRRAP in Regenerating Hepatocytes

Results in Defective Mitotic Progression. To investi-gate the function of TRRAP in liver regeneration, wecounted mitotic figures and examined them for abnor-malities and found that the number of mitotic figureswas strikingly lower in livers of TRRAP-CKO micethan in TRRAP-Co mice (Fig. 3A), suggesting thepossible involvement of TRRAP in mitotic progres-sion. However, a decrease in mitotic figures may be atleast in part caused by a block in the G0 and G1/Sphase transition, a notion supported by the data onBrdU incorporation and PCNA staining (Fig. 2). In

Fig. 3. Loss of TRRAP in regenerating hepatocytes resulted in defec-tive mitotic progression. (A) Kinetics of hepatocyte mitosis duringTRRAP-Co and TRRAP-CKO liver regeneration between 0 and 96 hoursafter CCl4 treatment. Hepatocyte mitotic figures were counted in 25high-power fields at the indicated times after CCl4 treatment. (B) Mi-totic figures in TRRAP-Co and TRRAP-CKO livers. Arrows point to nor-mal anaphase and chromatin ‘‘bridges’’ in TRRAP-Co (upper panels)and TRRAP-CKO cells (lower panels), respectively, 48 to 96 hours afterCCl4 treatment. (C) Quantitative summary of liver cells in mitosisexhibiting chromatin bridges as in (B).


addition, mitotic aberrations such as anaphase bridgeswere more frequently observed in the livers ofTRRAP-CKO mice than in the controls (Fig. 3B,C),suggesting that cell cycle defects might be responsiblefor impaired hepatocyte proliferation and liver regener-ation in the absence of TRRAP.Loss of TRRAP Impairs the Induction of Early

Cell Cycle Genes in Regenerating Livers. To assesswhether deregulation of critical cell cycle players maybe responsible for the observed decrease in hepatocyteproliferation in livers of TRRAP-CKO mice, we meas-ured the steady-state levels of early cyclins D and A(reliable markers of liver regeneration), cyclin-depend-ent kinases (cdk2 and cdk4), and CDC25A (a memberof the CDC25 family of phosphatases), as well as c-

Myc, a TRRAP-interacting transcription factorinvolved in cell cycle control. Although cyclin D1 andD2 levels in control livers increased and reached apeak between 36 and 48 hours after CCl4 treatment(Fig. 4A-C), consistent with a synchronous exit of qui-escent hepatocytes from G0 and entry into the cellcycle (Fig. 2), their levels were dramatically lower inTRRAP-CKO livers after CCl4 treatment, which is inagreement with impaired cell cycle reentry. Similarly,an increase in cyclins E and cyclin A levels were alsostrongly counteracted in TRRAP-CKO livers afterCCl4 treatment (Fig. 4A,D,E). Protein levels of othercell cycle regulators investigated (c-Myc, cdc25A, cdk2,and cdk4) were similar in both TRRAP-CKO andcontrol livers after CCl4 treatment (Fig. 4A;

Fig. 4. Loss of TRRAP impairs the induction of early cell cycle genes in regenerating livers. (A) Immunoblot analyses were performed on livertissue lysates prepared from TRRAP-Co and TRRAP-CKO mice at the indicated timepoints after CCl4 treatment using the indicated antibodies. Toconfirm equal loading, the blot was probed with an anti-actin antibody. (B-E) Quantification of protein levels of critical cell cycle players inTRRAP-Co and TRRAP-CKO liver cells as in (A) by densitometric analysis and normalization versus loading control.


Supporting Fig. 1), suggesting that expression of thesecell cycle genes in regenerating livers is not controlledby TRRAP. These results show that TRRAP may beimportant for expression of cyclins D, A, and E andthat it is dispensable for expression of cdk2 and cdk4,cdc25A, and c-Myc during liver regeneration.

TRRAP Is Essential for Histone Acetylation andTranscription Factor Binding at the Cyclin GenesDuring Liver Regeneration. To elucidate the mecha-nism by which TRRAP regulates the expression of cellcycle regulators in liver regeneration, we next usedChIP assay to examine the status of histone acetylation

Fig. 5. Effect of TRRAP deletion on histone acetylation at the cyclin gene promoters in regenerating livers. (A) ChIP analysis of histone H3acetylation at the cyclin A promoter. TRRAP-Co and TRRAP-CKO mice were treated with CCl4, and at indicated times thereafter hepatocytes werecross-linked and immunoprecipitated either by antiacetyl-histone H3 or irrelevant antibodies (No Ab). Immunoprecipitated DNAs were analyzed byPCR using primers recognizing the cyclin A promoter. (B) Quantification of acetylation levels at the cyclin A promoter as in (A) by densitometricanalysis and normalization versus input. (C) DNAs immunoprecipitated by anti-acetylhistone H4 antibodies as in (A) were analyzed by PCR usingprimers recognizing the cyclin A promoter. (D) DNAs immunoprecipitated by anti-c-Myc antibodies as in (A) were analyzed by PCR using primersrecognizing the cyclin A promoter. (D,E) ChIP analysis of transcription factor binding at the cyclin A promoter. Samples prepared as in (A) wereanalyzed by ChIP assay using anti-c-Myc (D) and anti-E2F1 (E) antibodies. (F) Quantification of E2F1 binding at the cyclin A promoter was car-ried out as in (B). As a control, mock immunoprecipitations without addition of antibodies (No Ab) were included.


and transcription factor binding within the cyclin Agene promoters after CCl4 treatment. Although inTRRAP-containing livers histone H3 acetylation levelswere significantly increased at different timepoints afterCCl4 treatment, TRRAP-deficient cells failed to showan increase in histone H3 acetylation at later time-points (Fig. 5A,B). Similar to histone H3, analysis ofhistone H4 acetylation status revealed that loss ofTRRAP also compromised the increase in acetylationof histone H4 at the cyclin A promoter after CCl4treatment (Fig. 5C). These results indicate thatTRRAP is needed for the hyperacetylation of histoneH3 and histone H4 associated with an increasedexpression of the cyclin gene in regenerating liver(Figs. 2, 4).We next examined the binding of c-Myc and E2F1

(the transcription factors known to bind TRRAP) tochromatin at the promoter of the cyclin A, a down-stream target of these transcription factors.20,21 Wefound that in TRRAP-containing livers c-Myc andE2F1 association with the cyclin A promoter wasincreased at specific timepoints after CCl4 treatmentand correlated with hepatocyte proliferation after liverinjury (Figs. 2, 4). Of note, loss of TRRAP wasaccompanied by a dramatic loss of both c-Myc andE2F1 binding at the cycin A promoter (Fig. 5D-F),suggesting that TRRAP and/or HAT-mediated acetyla-tion play an important role in the recruitment of thesetranscription factors to the cyclin promoters in regen-erating livers.

Discussion

Our results demonstrate that TRRAP and HATsplay a critical role in liver regeneration through theirfunction in cell cycle reentry and proliferation ofquiescent hepatocytes following liver damage. AfterCCl4treatment, TRRAP-deficient mice revealedimpaired liver regeneration without alteration of thedegree of initial histologic injury, indicating thatTRRAP is needed for regeneration of damaged livertissue. TRRAP-deficient hepatocytes exhibited a dra-matic proliferation defect as seen by the dramaticreduction in both BrdU- and PCNA-positive cells inTRRAP-deficient livers after CCl4 treatment. Thesefindings show that TRRAP is essential for cell cycleentry of hepatocytes after liver injury and that thereduced viability of TRRAP-deficient mice treatedwith CCl4 may be caused by the hepatocytes’ failureto regenerate the injured liver. However, the possibilitythat loss of TRRAP may affect other liver functions,

which may contribute to the phenotype of TRRAP-de-ficient mice, could not be formally ruled out.The observation that TRRAP-CKO livers after

CCl4 treatment did show an increase in hepatocyteproliferation, although more severely attenuated thanthat seen in TRRAP-containing livers, may beexplained by the presence of the remaining TRRAPprotein in a fraction of the hepatocytes (‘‘escapers ofdeletion’’), which may be sufficient to support cellcycle entry. However, TRRAP protein levels may befurther diluted during cell cycle progression. In sup-port of this is the frequent appearance of mitotic errorsin TRRAP-deficient livers, a finding that is consistentwith the role of TRRAP in mitotic progression.10,13

Together with previous findings on TRRAP-deficientcells,13 our results show that TRRAP plays a key roleduring entry into and progression through the cellcycle (including mitosis) owing to its association withdifferent transcription factors and expression of distinctsets of cell cycle-specific genes (Supporting Fig. 2A).However, the mechanism by which TRRAP mediateshepatocyte regeneration may involve other layers ofregulation (including posttranslational modifications ofkey cell cycle players).Of note, we found that loss of TRRAP also com-

promised the proliferation of nonparenchymal cells,suggesting that TRRAP is important for cell cycle pro-gression and proliferation of not only parenchymalcells (hepatocytes) but also of nonparenchymal cellsduring liver regeneration. In response to hepatic dam-age, differentiated hepatocytes proliferate in an attemptto restore the functional mass of liver as well as itsphysiological function.1 Liver regeneration is main-tained through the orchestration of regenerative signal-ing pathways involving an interplay between nonpar-enchymal and parenchymal cells.22 Therefore, ourresults suggest that the failure of TRRAP-deficient liv-ers to regenerate may be caused by proliferation defectsin both hepatocytes and nonparenchymal cells. How-ever, further studies are needed to determine to whatextent the proliferation defect in nonparenchymal cellsmay contribute to the TRRAP-deficient phenotypeinduced by liver injury.Along with the defect in hepatocyte proliferation in

TRRAP-CKO livers, we found a striking defect in theinduction of cyclin D1/2, cyclin E, and cyclin A.These findings suggest that in regenerating livers,TRRAP is essential for the induction of both early andlate cyclins, although TRRAP appears to be dispensa-ble for expression of CDK genes. Because expressionof the cyclin genes, in contrast to the CDK genes, isregulated in a cell cycle stage-specific manner, TRRAP


may be involved in the coordinated expression of thecyclin genes needed for cell cycle entry and progressionduring liver regeneration (Supporting Fig. 2B).D-type cyclins are important for early cell cycle reen-

try and G1-S progression23,24; therefore, the defect incell proliferation of hepatocytes in TRRAP-deficient liv-ers may be due to cellular failure to induce cyclin D1/2in the absence of TRRAP. Our findings agree with pub-lished data showing that reduced levels of D cyclins areassociated with impaired liver regeneration.25-27 BecauseTRRAP has been shown to bind, together with HATs,to cyclin D promoters to mediate histone acetylationand gene expression,12,28 our results suggest that duringliver regeneration TRRAP regulates cell cycle reentry ofhepatocytes through expression of D type cyclins,whereas TRRAP promotes cell cycle progression throughexpression of cyclin E and cyclin A.The precise mechanism by which TRRAP mediates

transcription of the cyclin genes remains elusive. ChIPanalysis of the cyclin A promoter revealed significantlylower levels of both histone H3 and H4 acetylation,consistent with previous studies showing that TRRAPregulates expression of genes by acetylation of histoneH3 or histone H413,18 and that TRRAP is importantfor the integrity and function of HAT complexes.9 Ofnote, despite a clear trend showing an impaired acety-lation in TRRAP-deficient livers, we observed amarked variability in acetylation levels using differentamplicons, suggesting that TRRAP-dependent HATsmay exhibit differential activity across different regionsof the cyclin A promoter.Our study revealed that loss of TRRAP compro-

mised the binding of transcription factors c-Myc andE2F1 at the cyclin gene promoter during liver regener-ation. Based on these findings, we propose thatTRRAP/HAT-mediated histone acetylation may be adeterminant of transcription factor binding (Support-ing Fig. 2B). In other words, histone acetylation is keyfor recognition and/or binding to chromatin by actingas binding sites that are recognized by transcriptionmachinery, rather than by simply facilitating access tochromatin.8,9,29 Alternatively, TRRAP may serve notonly as a scaffold for HAT complexes but also as aplatform for recruitment of transcription machineryincluding transcription factors to chromatin (Support-ing Fig. 2B). Future studies are needed to define theprecise mechanism by which TRRAP and histone acet-ylation mediate transcription activation and orchestratetimely expression of different cyclins throughout cellcycle during liver regeneration.In summary, our study demonstrates that TRRAP

and TRRAP/HAT-mediated acetylation play an impor-

tant role in liver regeneration after toxic injury andprovides insight into the mechanism by whichTRRAP/HATs orchestrate expression of the cyclingenes during cell cycle entry and progression.

Acknowledgment: We thank Marie-Pierre Cros forexcellent assistance in the maintenance of mouse colo-nies and with experiments on mice. We thank KristiM. Speights for editing the article.

References1. Michalopoulos GK, DeFrances MC. Liver regeneration. Science 1997;

276:60-66.2. Fausto N. Liver regeneration and repair: hepatocytes, progenitor cells,

and stem cells. HEPATOLOGY 2004;39:1477-1487.3. Taub R. Liver regeneration: from myth to mechanism. Nat Rev Mol

Cell Biol 2004;5:836-847.4. Yu C, Wang F, Jin C, Wu X, Chan WK, McKeehan WL. Increased

carbon tetrachloride-induced liver injury and fibrosis in FGFR4-defi-cient mice. Am J Pathol 2002;161:2003-2010.

5. Reed CA, Mayhew CN, McClendon AK, Yang X, Witkiewicz A,Knudsen ES. RB has a critical role in mediating the in vivo checkpointresponse, mitigating secondary DNA damage and suppressing liver tu-morigenesis initiated by aflatoxin B1. Oncogene 2009;28:4434-4443.

6. Taniguchi M, Magata S, Suzuki T, Shimamura T, Jin MB, Iida J, et al.Dipyridamole protects the liver against warm ischemia and reperfusioninjury. J Am Coll Surg 2004;198:758-769.

7. Shepard BD, Tuma PL. Alcohol-induced protein hyperacetylation:mechanisms and consequences. World J Gastroenterol 2009;15:1219-1230.

8. Agalioti T, Chen G, Thanos D. Deciphering the transcriptional histoneacetylation code for a human gene. Cell 2002;111:381-392.

9. Murr R, Vaissiere T, Sawan C, Shukla V, Herceg Z. Orchestration ofchromatin-based processes: mind the TRRAP. Oncogene 2007;26:5358-5372.

10. Herceg Z, Hulla W, Gell D, Cuenin C, Lleonart M, Jackson S, et al.Disruption of Trrap causes early embryonic lethality and defects in cellcycle progression. Nat Genet 2001;29:206-211.

11. Frank SR, Schroeder M, Fernandez P, Taubert S, Amati B. Binding ofc-Myc to chromatin mediates mitogen-induced acetylation of histoneH4 and gene activation. Genes Dev 2001;15:2069-2082.

12. Frank SR, Parisi T, Taubert S, Fernandez P, Fuchs M, Chan HM, et al.MYC recruits the TIP60 histone acetyltransferase complex to chroma-tin. EMBO Rep 2003;4:575-580.

13. Li H, Cuenin C, Murr R, Wang ZQ, Herceg Z. HAT cofactor Trrapregulates the mitotic checkpoint by modulation of Mad1 and Mad2expression. EMBO J 2004;23:4824-4834.

14. Loizou JI, Oser G, Shukla V, Sawan C, Murr R, Wang ZQ, et al. His-tone acetyltransferase cofactor Trrap is essential for maintaining thehematopoietic stem/progenitor cell pool. J Immunol 2009;183:6422-6431.

15. Taubert S, Gorrini C, Frank SR, Parisi T, Fuchs M, Chan HM, et al.E2F-dependent histone acetylation and recruitment of the Tip60 ace-tyltransferase complex to chromatin in late G1. Mol Cell Biol 2004;24:4546-4556.

16. Kuhn R, Schwenk F, Aguet M, Rajewsky K. Inducible gene targetingin mice. Science 1995;269:1427-1429.

17. Rudolph KL, Chang S, Millard M, Schreiber-Agus N, DePinho RA.Inhibition of experimental liver cirrhosis in mice by telomerase genedelivery. Science 2000;287:1253-1258.

18. Herceg Z, Li H, Cuenin C, Shukla V, Radolf M, Steinlein P, et al. Ge-nome-wide analysis of gene expression regulated by the HAT cofactorTrrap in conditional knockout cells. Nucleic Acids Res 2003;31:7011-7023.


19. Meng Z, Wang Y, Wang L, Jin W, Liu N, Pan H, et al. FXR regulatesliver repair after CCl4-induced toxic injury. Mol Endocrinol 2010;24:886-897.

20. Sylvester AM, Chen D, Krasinski K, Andres V. Role of c-fos and E2Fin the induction of cyclin A transcription and vascular smooth musclecell proliferation. J Clin Invest 1998;101:940-948.

21. Adachi S, Obaya AJ, Han Z, Ramos-Desimone N, Wyche JH, SedivyJM. c-Myc is necessary for DNA damage-induced apoptosis in theG(2) phase of the cell cycle. Mol Cell Biol 2001;21:4929-4937.

22. Zheng ZY, Weng SY, Yu Y. Signal molecule-mediated hepatic cell com-munication during liver regeneration. World J Gastroenterol 2009;15:5776-5783.

23. Morgan DO. SnapShot: cell-cycle regulators II. Cell 2008;135:974-974e971.

24. Morgan DO. SnapShot: cell-cycle regulators I. Cell 2008;135:764-764e761.

25. Berasain C, Garcia-Trevijano ER, Castillo J, Erroba E, Lee DC, PrietoJ, et al. Amphiregulin: an early trigger of liver regeneration in mice.Gastroenterology 2005;128:424-432.

26. Desbois-Mouthon C, Wendum D, Cadoret A, Rey C, Leneuve P, BlaiseA, et al. Hepatocyte proliferation during liver regeneration is impaired inmice with liver-specific IGF-1R knockout. FASEB J 2006;20:773-775.

27. Natarajan A, Wagner B, Sibilia M. The EGF receptor is required forefficient liver regeneration. Proc Natl Acad Sci U S A 2007;104:17081-17086.

28. Bouchard C, Dittrich O, Kiermaier A, Dohmann K, Menkel A, EilersM, et al. Regulation of cyclin D2 gene expression by the Myc/Max/Mad network: Myc-dependent TRRAP recruitment and histone acetyla-tion at the cyclin D2 promoter. Genes Dev 2001;15:2042-2047.

29. Fernandez PC, Frank SR, Wang L, Schroeder M, Liu S, Greene J,et al. Genomic targets of the human c-Myc protein. Genes Dev 2003;17:1115-1129.


Loss of histone acetyltransferase cofactor transformation/transcription domain-associated protein...

Documents

Transcript of Loss of histone acetyltransferase cofactor transformation/transcription domain-associated protein...