Inhibition of De Novo NAD+ Synthesis by Oncogenic URI Causes Liver Tumorigenesis through DNA Damage

Please cite this article in press as: Tummala et al., Inhibition of De Novo NAD+ Synthesis by Oncogenic URI Causes Liver Tumorigenesis through DNADamage, Cancer Cell (2014), http://dx.doi.org/10.1016/j.ccell.2014.10.002

Cancer Cell

Article

Inhibition of De Novo NAD+ Synthesisby Oncogenic URI Causes LiverTumorigenesis through DNA DamageKrishna S. Tummala,1 Ana L. Gomes,1 Mahmut Yilmaz,1 Osvaldo Grana,2 Latifa Bakiri,3 Isabel Ruppen,4

Pilar Ximenez-Embun,4 Vinayata Sheshappanavar,5 Manuel Rodriguez-Justo,6 David G. Pisano,2 Erwin F. Wagner,3

and Nabil Djouder1,*1Growth Factors, Nutrients and Cancer Group, BBVA Foundation-Cancer Cell Biology Programme, Spanish National Cancer ResearchCentre, CNIO, 28029 Madrid, Spain2Bioinformatics Unit, Structural Biology and Biocomputing Programme, Spanish National Cancer Research Centre, CNIO, 28029 Madrid,

Spain3Genes, Development, and Disease Group, BBVA Foundation-Cancer Cell Biology Programme, Spanish National Cancer Research Centre,CNIO, 28029 Madrid, Spain4Proteomics Core Unit, ProteoRed ISCIII, Biotechnology Programme, Spanish National Cancer Research Centre, CNIO, 28029Madrid, Spain5Department of Pathology, Royal London Hospital, London E1 1BB, UK6Department of Cellular Pathology, University College London NHS Trust, London NW1 2BU, UK*Correspondence: [email protected]

http://dx.doi.org/10.1016/j.ccell.2014.10.002

SUMMARY

Molecular mechanisms responsible for hepatocellular carcinoma (HCC) remain largely unknown. Usinggenetically engineered mouse models, we show that hepatocyte-specific expression of unconventional pre-foldin RPB5 interactor (URI) leads to amultistep process of HCC development, whereas its genetic reductionin hepatocytes protects against diethylnitrosamine (DEN)-induced HCC. URI inhibits aryl hydrocarbon (AhR)-and estrogen receptor (ER)-mediated transcription of enzymes implicated in L-tryptophan/kynurenine/nicotinamide adenine dinucleotide (NAD+) metabolism, thereby causing DNA damage at early stages oftumorigenesis. Restoring NAD+ pools with nicotinamide riboside (NR) prevents DNA damage and tumorformation. Consistently, URI expression in human HCC is associated with poor survival and correlates nega-tively with L-tryptophan catabolism pathway. Our results suggest that boosting NAD+ can be prophylactic ortherapeutic in HCC.

INTRODUCTION

Hepatocellular carcinoma (HCC) is the commonest, usually lethal,

humanprimary liver neoplasm (GLOBOCANv2.0, 2008). Theearly

stage is characterized by low- to high-grade dysplastic nodules,

‘‘preneoplastic lesions’’ (Kudo, 2009). These frequently develop

in chronic inflammatory liver disease or hepatitis, which can pro-

mote fibrosis, cirrhosis, and progression to HCC. Thus, precan-

cerous lesions have clinical value for HCC prediction (Libbrecht

et al., 2001), but therapeutic options are limited (El-Serag, 2011).

Significance

HCC is the third leading cause of cancer death worldwide withdeficit-induced genotoxic stress is critical to initiate liver tumobolism and genome integrity. Because our findings are relevantof NR, a vitamin B3 derivative, or other NAD+ boosters can binduced NAD+ depletion-mediated DNA damage and carcinoTherapeutic intervention on metabolic alterations prior to getumorigenesis.

In early stages of many cancers, including HCC, oncogene

activation induces replicative stress, resulting in DNA damage

leading to chromosomal instability (CIN), which accelerates

tumor development (Teoh et al., 2008). DNA damage elicits a

key repair mechanism, the DNA damage response (DDR), initi-

ated by phosphorylation of checkpoint proteins Chk1, Chk2,

and p53 (Reinhardt and Schumacher, 2012). p53-dependent

responses, including cell cycle arrest and/or senescence, are

induced, limiting preneoplastic lesions’ growth. When DNA dam-

age is too pronounced, p53 engages an apoptotic program by

limited therapeutic options. Here we demonstrate that NAD+

rigenesis and unravel a critical link between nutrient meta-in humanHCC, we propose that nutritional supplementatione used as preventive and curative therapies in oncogene-genesis, especially in patients with precancerous lesions.nomic instability should be further considered to prevent

Cancer Cell 26, 1–14, December 8, 2014 ª2014 Elsevier Inc. 1

mailto:[email protected]


Cancer Cell

URI-Induced NAD+ Depletion Causes HCC Development


upregulation of Bcl-2 family proteins (Noxa, Puma, Bid, and/or

Bax). p53 dysfunctions allow tumor cells to escape apoptosis,

and thus, mutations inactivating p53 are themost common alter-

ations observed in HCC (Reinhardt and Schumacher, 2012).

In pathophysiological situations, the balance between cell pro-

liferation and apoptosis can be altered, perturbing tissue homeo-

stasis. Apoptotic dysregulations are important in liver disease.

Insufficient apoptosis, eliminating mutated cells, combined

with inflammation-mediated proliferation can promote liver can-

cer development. Excessive or sustained apoptosis causes liver

injuries, increased hepatocyte regeneration, which enhances ge-

netic errors and predisposes to HCC (Malhi and Gores, 2008).

Still, the initiating hepatocarcinogenesis events remain unclear.

Developing experimental models mimicking distinct stages of

HCC development would help to explore molecular mechanisms

linking histopathological changes to hepatocarcinogenesis.

Unconventional prefoldin RBP5 interactor (URI), a member

of the R2TP/URI-prefoldin (PFD)-like complex containing the

heat shock protein 90 (HSP90) (Boulon et al., 2010), is an onco-

gene amplified in human ovarian carcinomas and downstream

effector of the growth factor and nutrient-regulated mTOR/

S6K1 signaling cascade (Theurillat et al., 2011). URI inhibits

phosphatase PP1g, thereby increasing S6K1 activity-dependent

survival signaling. Thus, URI/PP1g complexes maintain the

mitochondrial threshold for apoptosis in accordance to nutrient

availability. URI overexpression promotes survival, while its dele-

tion enhances cancer cell death (Djouder et al., 2007; Theurillat

et al., 2011). Prompted by these observations, and the fact

that HCC occurs on the basis of mitochondrial dysfunction-

mediated hepatocyte death and liver injury (Luedde et al.,

2014; Malhi and Gores, 2008), we investigate the role of URI in

hepatocarcinogenesis.

RESULTS

URI Expression in Mouse Hepatocytes InducesSpontaneous Liver TumorsWe generated a Col1a1 knockin mouse (Figures S1A and S1B

available online), expressing human URI (hURI) via a tetracy-

cline-dependent transactivator controlled by the hepatocyte-

specific liver activated protein promoter (Figures S1C–S1E).

These mice, designated hURI-tetOFFhep, and littermates lacking

hURI expression are referred to hereafter as ‘‘mutants’’ and

‘‘controls,’’ respectively. Without doxycycline, hURI was ex-

pressed specifically in hepatocytes from one allele from E10.5

(Carpenter et al., 2005), roughly twice as much as mouse

URI (Figures 1A, S1F, and S1G), similar to the increase of URI

expression in human HCC (see below).

We observed no pathological signs in 3-week-old mutants. In

8-week-old mutants, hematoxylin and eosin (H&E) staining re-

vealed anisokaryotic clusters (Figure 1B) resembling low-grade

dysplastic nodules observed in human hepatitis (Libbrecht

et al., 2001). At 12 weeks the clusters developed into high-grade

dysplastic nodules (Figures 1B, S1H, and S1I), similar to human

large liver cell dysplasia (LLCD) (Libbrecht et al., 2001). Fibrosis

was detected at 8 weeks and increased over time until 24 weeks,

as assessed by Sirius Red (SR), Masson Trichrome, alpha

smooth muscle actin, type I collagen (COL1A1), and reticulin

staining (Figure S1J). Quantification showed that about 1% to

2 Cancer Cell 26, 1–14, December 8, 2014 ª2014 Elsevier Inc.

3% of livers were SR positive in mutants, representing 100%

to 300% increase over littermates (Figure S1K). Increases in

fibroticmarkers weremeasured by quantitative RT-PCR (Figures

S1L–S1N), but serum alanine aminotransferase (ALT) values

remained unchanged (Figure S1O).

Between 24–54 weeks macroscopic lesions including ade-

noma and early HCC emerged. Recent reports described malig-

nant transformation of human adenomas, but tumors in our

model developed simultaneously (Pilati et al., 2014). Between

54–65 weeks low-grade and differentiated HCC were fully

apparent, and between 65–75 weeks, 40% of mutants devel-

oped macroscopic high-grade tumors occupying 20%–60%

of the liver (Figures 1B, 1C, and S1P). There were 25%–50% of

hepatocytes that were Ki67-positive, suggesting aggressive

tumors (data not shown). According to World Health Organi-

zation criteria (WHO, 2008), all tumors were well/moderately

differentiated: 20% glandular/acinar, indicative of telangiectatic

variants, and 80% trabecular. No cholangiocarcinoma were

detected (Figure 1C). Serum glucose, ALT, and total bile

acids were affected (Figure S1Q). Surprisingly, serum albumin

was increased, suggesting that liver function might not be fully

compromised (Figure S1Q).

All mutants died at �85 weeks, with a median survival

of 76 weeks before complete liver failure (Figure 1D). Immuno-

histochemistry (IHC) and pathological analyses revealed hURI-

positive hepatocyte-like cells in 30% of mutant lungs with

HCC, indicating aggressive metastases (Figures S1R and S1S).

Histopathological characterization confirmed the presence of

heterogeneous tumorswith collapsed reticulin fibers, suggesting

increased hepatocyte death and proliferation, as indicated by

Ki67 (Figure 1E). Increases in alpha fetoprotein (AFP) levels, a

clinical marker for human HCC varied, but all tumors displayed

dramatic increases in p53 abundance and phosphorylation (Fig-

ures 1E and 1F), suggesting that p53 may either carry mutations

or may be improperly folded (Trinidad et al., 2013), thus possibly

inactive.

Fully developed HCC appeared at 30 weeks in hepatocarcino-

gen diethylnitrosamine (DEN)-treated hURI-tetOFFhepmice (Ves-

selinovitch and Mihailovich, 1983) (Figure S1T). When hURI was

expressed from two alleles, increasing its expression to 6-fold

compared to heterozygous hURI-tetOFFhep mice, HCC were

detected at 10 weeks (Figure S1U), highlighting the importance

of URI dosage. Embryonic development was not involved

because liver tumors were also detected in mice kept on doxy-

cycline until 8 weeks (expressing hURI from 8 weeks) then trans-

ferred to normal (chow) diet (Figure S1V). Thus, hURI expression

in mouse hepatocytes induces spontaneous HCC.

Continuous URI Expression Is Essential forHepatocarcinogenesisCeasing hURI expression in 8-week-old mutants for 24 weeks

reduced fibrosis and abolished dysplastic foci and prevented

early tumors, without affecting liver-to-body weight ratios (Fig-

ures 2A–2D and S2A). S6K1 activity was increased in 24-week-

old mice, but remained constant when hURI expression ceased,

indicating that mTOR/S6K1 activation was hURI-independent

(Figure 2B). Switching hURI expression off until 60 weeks pre-

vented tumor development and normalized ALT levels (Figures

S2B–S2E). Similarly, when hURI was expressed for 24 weeks,

Figure 1. URI Expression in Mouse Hepatocytes Induces Spontaneous Liver Tumors

(A) Representative images of IHC stained liver sections from 3-week-old hURI-tetOFFhep mice using hURI and FLAG antibodies. Insets represent the periportal

area, showing hepatocyte specific hURI expression. (n > 10).

(B) Representative images of H&E stained liver sections from 3- (n > 6), 8- (n > 19), 12- (n > 11), and 32-week-old (n > 7) hURI-tetOFFhepmice. Bottom two rows are

representative images of whole livers from hURI-tetOFFhep mice at 32 and 75 weeks of age. Black dotted circles mark LLCD-like lesions and black arrows point

anisokaryotic clusters in mutant hURI-tetOFFhep mice. Yellow dotted circles depict adenoma and HCC at 32 and 75 weeks of age, respectively.

(C) Percentage of control and mutant hURI-tetOFFhep mice bearing liver abnormalities in 60- to 75-week-old-mice.

(D) Kaplan Meier curve of control (n = 17) and mutant (n = 17) hURI-tetOFFhep mice. Log rank test p = 0.0036; Hazard ratio = 0.1603.

(E) Representative images of H&E, IHC, and reticulin stained liver sections from control and four tumors derived from one mutant hURI-tetOFFhep. NT, PT, and

T denote nontumoral, peritumoral, and tumoral tissues, respectively.

(F) WB analysis of control and mutant hURI-tetOFFhep livers. ‘‘T’’ denotes tumor.

See also Figure S1.

Cancer Cell



until high-grade dysplastic nodules/early HCC and adenomas

were apparent, then, switched-off for 28 weeks, only residual

anisokaryotic clusters were detected, but no adenomas or

HCCs (Figure 2E). However, ultrasound analysis demonstrated

that well/moderately differentiated HCC (above 60 weeks) did

not regress when hURI expression was ceased for 5 weeks (Fig-

ures S2F–S2H). Thus, continuous hURI expression is required for

the maintenance of preneoplastic lesions and early tumors.

Aggressive HCCs with sufficient genetic mutations become

URI independent, even though ceasing URI expression for a

longer time remains to be tested.

We genetically inactivated URI specifically in hepatocytes by

crossing URI(lox/lox) and serum albumin (SA)-CreERT2 mice

(Schuler et al., 2004). URI deletion in hepatocytes after tamoxifen

treatment to obtain URI(+/D)hep or URI(D/D)hepmice, was confirmed

by IHC and western blotting (WB) (Figures 2F and 2G). Homozy-

gous deletion of URI led to death of URI (D/D)hep mice around

10 days (Figure S2I). Disruption of tissue architecture, presence

of atypia, dilated veins with intrahepatic bleeding, signs of

necrosis, and inflammatory cell infiltration were observed by

H&E staining. Additionally, SR staining, collapsed reticulin

fibers, and increased ALT indicated that hepatocytes underwent


Figure 2. Continuous URI Expression Is Essential for Hepatocarcinogenesis

(A) Representative images of H&E stained liver sections from 32-week-old hURI-tetOFFhep mice fed with (+) or without (�) doxycycline (Dox) after dysplatic lesion

formation at 8 weeks. Dotted black circle represents premalignant lesions. (n R 5).

(B) WB analysis of hURI-tetOFFhep livers as described in (A).

(C) Representative images of Sirius Red stained liver sections from mice described in (A). (n R 5).

(D) Quantification of Sirius Red stained liver sections from mice described in (C). (n R 5).

(E) Representative images of full livers and H&E stained liver sections from hURI-tetOFFhep mice treated with Dox for 28 weeks. Treatment started at 24 weeks of

age, after the appearance of dysplastic lesions, adenomas, and early HCC. Dotted black circles denote reminiscent anisokaryotic areas. (n R 5).

(F) Representative images of IHC stained liver sections for endogenous URI in URI(+/+)hep, URI(+/D)hep, and URI(D/D)hep mice. (n R 3).

(G) WB liver analysis for endogenous URI in URI(+/+)hep, URI(+/D)hep, and URI(D/D)hep livers.

(H) Representative images of whole livers from URI(+/+)hep and URI(+/D)hep mice treated with diethylnitrosamine (DEN) and sacrificed at 24 weeks of age. Dotted

yellow circles depict liver tumors. (n R 5). Bottom pictures represent reticulin stained livers, black circle depicts the dysplastic area.

(I) Tumor burden of mice described in (H).

(J) Serum ALT levels from mice described in (H).

Data represented as mean ± SEM (p % 0.05 = *). See also Figure S2.

Cancer Cell



massive apoptotic program leading to liver injury, suggesting

that these mice die from fulminant liver failure (Figures S2J and

S2K). However, URI(+/D)hep mice, in which URI expression was

approximately halved (Figures 2F and 2G), supplied a liver dam-

age-inducing 3,5-diethoxycarbonyl-1,4-dihydrocollidine (DDC)-

supplemented diet, presented significantly less liver damage

and fibrosis as shown by ALT, SR, and reticulin stainings (Figures

S2L–S2N). Thus URI reduction protects from hepatocyte injury.

DEN-treatment increased URI levels in 3-week-old URI(+/+)hep

mouse livers and in wild-type murine HCC (Figures S2O

and S2P). Moreover, DEN induced tumor development at

24 weeks in 60% of URI(+/+)hep mice and enhanced ALT levels,

but URI(+/D)hep mice did not show any tumors at this age and dis-

played normal ALT values (Figures 2H–2J). Levels of cytochrome

P4502E1and1A1catalyzingDENwere unchanged in 3-week-old


livers ofDEN-treatedURI(+/+)hepandURI(+/D)hepmice (FigureS2O),

suggesting that halving URI levels does not affect DEN bio-

activation (Kang et al., 2007). Thus, while complete deletion of

URI induces liver injury, halving URI is beneficial to maintain liver

homeostasis and prevents liver injury and HCC development.

URI-Induced DNADamage Initiates Liver TumorigenesisTo further elucidate the mechanisms of hepatocyte death and

their contribution to HCC, we checked for phosphorylation of

histone H2AX (gH2AX), a DNA damage marker that triggers

apoptosis via the p53-DDR pathway. gH2AX and p53 phosphor-

ylation and abundance did not differ in 1-week-old livers (Fig-

ure S3A). At 3 weeks, a nonpathological stage with no dysplastic

lesions, gH2AX, phosphorylation of the 32 kDa subunit of repli-

cation protein A (RPA32) at Ser-4 and Ser-8, and p53 abundance

Figure 3. URI-Induced DNA Damage Initiates Liver Tumorigenesis

(A) Representative images of gH2AX IHC stained liver sections from 3-week-old hURI-tetOFFhep mice. Insets denote gH2AX positive nuclei. (n = 5).

(B) Quantification of (A).

(C) WB analysis of 3-week-old hURI-tetOFFhep livers.

(D) Representative images of gH2AX IHC stained liver sections from 8-week- and 12-week-old hURI-tetOFFhep mice. Dotted black shapes depict anisokaryotic

clusters positive for gH2AX. (n = 6).

(E) Quantification of (D).

(F) WB analysis of 8-week-old hURI-tetOFFhep livers

(G) WB analysis of 8-week-old hURI-tetOFFhep livers, with or without p53 inactivation.

(H) Reticulin and Sirius Red stained livers described in (G).

(I) Serum ALT levels of mutant mice with or without p53 inactivation. (n > 4).

(J) Kaplan-Meier survival curve of control and mutant hURI-tetOFFhep mice with and without p53 inactivation. (Log rank test p % 0.001.)

(K) Percentage of tumor incidence in hURI-tetOFFhep mice with or without p53 inactivation.

Data represented as mean ± SEM (*p % 0.05 and **p % 0.01). See also Figure S3.

Cancer Cell



and phosphorylation were higher in mutants (Figures 3A–3C),

suggesting that DDR precedes precancerous lesions. While

p53-dependent apoptosis occurred in cells that unsuccessfully

repair DNA (cleaved caspase 3; Figure 3C), hepatocyte prolifer-

ation rate was reduced (Figures S3B and S3C), suggesting that

high proliferation is not the initial hepatocarcinogenic event.

gH2AX-positive nuclei were more abundant in older mutant

livers (8- to 12-week-old) with obvious dysplastic lesions.

Furthermore, Ser-345 phosphorylation (and hence activation)

of Chk1 was enhanced in mutants, but not Thr-68 phosphoryla-

tion of Chk2, suggesting single strand break (Reinhardt and

Schumacher, 2012) (Figures 3D–3F). Enhanced p53 phosphory-

lation at Ser-18 (Chk1 target), acetylation at Ac-lys379 (Ito et al.,

2001) (Figure 3F), and expression of p19ARF (Reinhardt and

Schumacher, 2012) (Figure S3D) indicate p53 stabilization.

Senescence-associated b-galactosidase activity, expression

of several p53 target genes, BAX, and p21 protein abundance

were increased in 8- and 12-week-old mutants, while expression

of Xiapwas decreased (Figures 3F, S3E, and S3F) (Reinhardt and

Schumacher, 2012), suggesting increased hepatocyte death as

also shown by collapsed reticulin fibers (Figure S3G). Increased

abundance of proliferating cell nuclear antigen, cyclin D1, and

Ki67-positive nuclei suggested compensatory proliferation (Fig-

ures 3F and S3H). Finally, MAD2, a CIN marker and downstream

effector of cyclin D1, was increased (Figure 3F). Thus, hURI

expression in hepatocytes induces genotoxic stress, apoptosis,

compensatory proliferation, and CIN.

To assess genotoxic stress-induced apoptosis in HCC devel-

opment, p53 was inactivated in hURI-tetOFFhep mice. In p53-in-

activated 8-week-old mutants (hURI (+/Ki); p53ERTAM (+/Ki)),


Cancer Cell



cleaved caspase 3, Bax expression, and collapsed fibers were

decreased (Figures 3G and 3H). Furthermore, SR staining and

ALT levels were reduced (Figures 3H and 3I), indicating that

DNA damage-activated p53 is required for hepatocyte death

and liver injury. While apoptosis was drastically suppressed,

inactivation of p53 significantly reduced survival and accelerated

liver tumorigenesis (Figure 3J): 80% of mice displayed aggres-

sive HCC (Figure 3K). Deletion of Cdkn2a did not modify mouse

survival or tumor burden (data not shown). Thus, genotoxic

stress, rather than excessive apoptosis, is the critical initiating

event in liver carcinogenesis.

URI Causes DNA Damage and Liver Tumorigenesis byInhibiting De Novo NAD+ SynthesisTo identify URI-mediated hepatocarcinogenetic events, we first

examined mTOR activation, which had been implicated in HCC

development via DNAdamage (Menon et al., 2012). No increases

in S6K1 activity were detected at 1 week (Figure S4A). In sequen-

tial immunoprecipitation experiments, using 1-week-old liver ex-

tracts, free hURI molecules were revealed by WB after complete

depletion of PP1g (Figures S4B and S4C), and vice versa (data

not shown). When 3-week-old mice were supplied a rapamy-

cin-containing diet, progression to preneoplastic abnormalities

continued, if not further pronounced (data not shown). Thus,

although a fraction of hURI binds PP1g, hURI apparently has a

PP1g-independent role in DNA damage and liver tumorigenesis.

Additionally, no differences in reactive oxygen species (ROS)

were observed in 1- and 8-week-old livers (Figures S4D and

S4E), suggesting that DNA damage is ROS-independent.

Global transcriptomic and proteomic profiling were performed

in a very early nonpathological stage and early premalignant

state (1- and 8-week-old livers). Transcripts’ sequencing re-

vealed small fractions of genes differentially expressed upon

hURI expression: 303 out of 12,295 genes at 1 week, and 740

out of 11,133 (false discovery rate [FDR] < 0.05) at 8 weeks (Fig-

ures 4A and S4F). Similarly, isobaric tags for relative and abso-

lute quantification (iTRAQ) identified 2,394 proteins: 122 and

597 of which were differentially expressed in 1- and 8-week

livers, respectively (Figures 4B and S4G; Table S1).

Heatmapping revealed that most differentially expressed pro-

teins were downregulated (Figure S4H). Significant overlaps in

the differentially expressed transcripts and proteins at 1 and

8 weeks (Figures S4I and S4J), indicated hURI-dependent tran-

scriptional repression mechanisms. Ingenuity pathway analysis

(IPA) revealed that among canonical metabolic pathways, the

L-tryptophan/kynurenine catabolism leading to de novo nicotin-

amide adenine dinucleotide (NAD+) synthesis was one of the

most significant downregulated pathways (Figures 4C and S4K).

Enzymes implicated in the L-tryptophan/kynurenine degrada-

tion, including tryptophan 2,3-dioxygenase (TDO2) and arylfor-

mamidase (AFMID) catalyzing the initial rate-limiting step and

kynurenine 3-monooxygenase (KMO), kynureninase (KNYU), and

3-hydroxyanthranilate 3,4-dioxygenase (HAAO) were all downre-

gulated (Figure 4D). Gene set enrichment analysis (GSEA) (Subra-

manian et al., 2005), using the RNA sequencing data and Kyoto

Encyclopedia of Genes and Genomes database, corroborated

these defects (data not shown). WB confirmed that TDO2 and

AFMID expression was reduced >50% in these livers (Figures 4E

and S4L) and in adult livers expressing hURI (Figure S4M).


NAD+ concentrations were reduced in 3- and 6-week mutant

livers (Figures 4F and S4N), while increases in TDO2, AFMID,

and NAD+ levels were detected in URI(+/D)hep livers (Figures 4G

and 4H). Consistent with previous observations (Konishi et al.,

1986), liver NAD+ levels were depleted in DEN-treated mice,

and URI reduction enhanced NAD+ levels (Figure S4O). Thus,

URI reduction enhances NAD+ de novo synthesis, potentially ex-

plaining the protective effect of URI deletion in HCC. Further-

more, NAD+ concentrations inversely correlated with URI levels

in four human HCC cell lines (Huh-7, HepG2, SNU-398, and

SNU-449). While URI depletion significantly increased NAD+

levels, URI overexpression reduced NAD+ values (Figure S4P).

URI overexpression in SNU-449 cells, which had high NAD+

values and low endogenous URI levels, increased their growth,

whereas URI depletion in Huh-7 and HepG2 cells displaying

high endogenous URI, significantly reduced their growth (Fig-

ure S4Q). URI-regulating NAD+ levels may therefore be relevant

for human liver tumorigenesis.

Depleting TDO2 and AFMID in HCC cell line SNU-449

significantly reduced NAD+ levels (Figure S4R). Importantly, 14C-

labeled NAD+ levels in four human HCC cell lines incubated with14C-tryptophan reduced significantly following AFMID depletion

(Figure S4S), indicating that L-tryptophan degradation accounts

for de novo NAD+ synthesis. Furthermore, expression of key en-

zymes of three other pathways implicated in oncogenesis was

unaffected by hURI expression: SHMT1, G6PD, and GOT1 of

the glycine/serine/threonine, pentose phosphate and glutamine/

aspartate pathways, respectively (FigureS4T). Finally, expression

of nicotinamide phosphoribosyltransferase (NAMPT, implicated

in NAD+ biosynthesis through salvage reactions, Figure S4U)

and activity of poly (ADP-ribose) polymerase (PARP), the main

NAD+-consuming enzyme (Figure S4U) were not affected at early

stages (1week), and levels of NADH and several dehydrogenases

that reduce NAD+ to NADH were decreased (Figure S4U;

Table S2). Reduction of NAD+ is thus mainly due to downregula-

tion of L-tryptophan/kynurenine catabolism.

We next induced liver injury and hepatocyte proliferation in

C57BL/6 mice with DDC-supplemented diet for 4 days, treated

them with DMSO or Ro-61-8048, a KMO inhibitor, for the next

3 days, and sacrificed mice on day 8 (Figures S4V and S4W).

NAD+ concentrations were reduced and DNA damage foci

significantly elevated in Ro-61-8048-treated livers (Figures 4I–

4K). Thus, L-tryptophan/kynurenine pathway inhibition in vivo

leads to reduced NAD+ concentrations and DNA damage, reca-

pitulating effects of hURI expression. Finally, nontumorigenic

mouse liver cells AML-12, stably depleted of TDO2 and AFMID

and transplanted into immunodeficient mice, formed aggressive

tumors (Figure S4X), suggesting that inhibition of L-tryptophan

pathway leads to transformation and tumorigenesis.

We assessed whether DNA damage was a consequence

of inactivation of NAD+-consuming enzymes, such as SIRT1 or

PARP (Durkacz et al., 1980; Herranz et al., 2010). In SNU-449

cells, SIRT1 inhibition by EX-527, which may enhance NAD+,

reduced RPA32 phosphorylation, whereas activating SIRT1

with resveratrol, whichmay lower NAD+, increased phosphoryla-

tion of RPA32 (Figure S4Y). Because in our model NAD+ deficits

increased replicative stress, DNA damage is unlikely due

to SIRT1 inhibition alone. Additionally, in URI-overexpressing

SNU-449 cells, in which NAD+ levels were lowered (Figure S4P),

Figure 4. URI Causes DNA Damage and Liver Tumorigenesis by Inhibiting De Novo NAD+ Synthesis

(A) Volcano plots from RNA sequencing representing differentially expressed significant (blue) and unchanged (red) mRNA species in livers from 1- and 8-week-

old hURI-tetOFFhep mice. (n > 3).

(B) Histogram of differentially expressed proteins analyzed by iTRAQ in livers from 1- and 8-week-old hURI-tetOFFhep mice. Numbers of proteins significantly

downregulated (green) and upregulated (red) are shown. (n = 5).

(C) Top downregulated canonical metabolic pathways based on iTRAQ data from 8-week-old mice, analyzed by using IPA software.

(D) Scheme of de novo NAD+ synthesis. Fold change of protein expression detected in iTRAQ are represented within the brackets. Ro-61-8048 is an inhibitor

for KMO.

(E) WB analysis (left) and quantification of reduction (mutant over control, right) of TDO2 and AFMID of 8-week-old hURI-tetOFFhep livers.

(F) Liver NAD+ concentrations in 3-week-old hURI-tetOFFhep mice. (n R 10).

(G) WB analysis of URI(+/+)hep and URI(+/D)hep livers.

(H) NAD+ levels in livers from URI(+/+)hep and URI(+/D)hep mice. (n = 5).

(I) Liver NAD+ levels in C57BL/6 mice previously fed with DDC and treated with either DMSO (1%) or Ro-61-8048 (25 mg/Kg) compound. (n R 5).

(J) Representative images of gH2AX IHC stained liver sections from C57BL/6 mice described in (I). (n = 5).

(K) Quantification of (J).

Data represented as mean ± SEM (*p % 0.05 and ***p % 0.001). See also Figure S4 and Tables S1 and S2.

Cancer Cell



SIRT1 activation further increased RPA32 phosphorylation.

Thus, modulating SIRT1 activity may affect PARP activity either

via modulation of NAD+ levels or through regulation of acet-

ylation-dependent PARP1 activity (Rajamohan et al., 2009).

Notably, URI overexpression increased RPA32 phosphorylation,

which was not further enhanced when PARP was inhibited (Fig-

ure S4Y). Finally, PARP activity was reduced in 3-week-old

mutants, while NAMPT expression remained unchanged (Fig-

ure S4Z). Thus, hURI-mediated NAD+ depletion may induce

DNA damage via PARP inhibition.

Restoring NAD+ Pools Protects from DNA Damage andPrevents Tumor FormationTo investigate whether restoring NAD+ pools would prevent

dysplastic nodules and tumor formation, 3-week-old hURI-

tetOFFhep mice were supplied with a nicotinamide riboside

(NR) diet. NR significantly increased hepatic NAD+ concentra-

tions (Figure S5A) without affecting liver-to-body weight ratio

(Figure S5B). We detected dysplastic lesions and DNA damage

in all mutants on chow, but not in those on NR, which also had

reduced fibrosis, p53 abundance, and Ser-18 phosphorylation

(Figures 5A–5D, S5C, and S5D). Prolonged NR treatment pre-

vented tumor development and reduced ALT levels (Figures

5E–5G). Similarly liver tumors were prevented in 30-week-old

homozygous mutants with higher URI levels (Figure S5E).

Thus, restoring NAD+ pools protects from hURI-induced DNA

damage, preneoplastic lesions, and tumor development. Sur-

prisingly, 12-week-old homozygous mutants with full blown tu-

mors then on 48 weeks of NR regimen showed significant tumor

regression (Figures S5F and S5G), and their livers had high levels


Figure 5. Restoring NAD+ Pools Protects from DNA Damage and Prevents Tumor Formation

(A) Representative images of H&E and gH2AX IHC stained liver sections from 12-week-old hURI-tetOFFhep mice fed with either chow (nR 15) or NR diets started

at 3 weeks of age (n R 15). Dotted black lines indicate anisokaryotic clusters present in mutant hURI-tetOFFhep mice under chow diet.

(B) Quantification of dysplastic lesions in the hURI-tetOFFhep mice described in (A).

(C) Quantification of gH2AX positive nuclei in the hURI-tetOFFhep mice described in (A).

(D) WB analysis of mutant hURI-tetOFFhep livers as described in (A).

(E) Representative images of whole livers and H&E stained liver sections from 30- or 60-week-old hURI-tetOFFhep mice supplemented with NR diet from 3 weeks

of age until mice were sacrificed. (n R 10 for chow fed or NR fed.) Yellow dotted circles depict early tumors and black arrows point mitotic bodies.

(F) Tumor burden of 60-week-old mice described in (E).

(G) Serum ALT levels of 60-week-old mice described in (E).

(H) WB analysis of hURI-tetOFFhep mice expressing hURI for 8 weeks and switched OFF for 24 weeks.

(I) gH2AX IHC stained liver sections from 32-week-oldmutant hURI-tetOFFhepmice fedwith either chow or Dox diets. (n = 5). Red arrows point to DNAdamage foci.

Data represented as mean ± SEM (*p % 0.05 and ***p % 0.001). See also Figure S5.

Cancer Cell



of cleaved caspase 3 (Figure S5H), suggesting that boosting

NAD+ levels may be cytotoxic for tumor cells.

Furthermore, ceasing hURI expression in 8-week-old mice for

24 weeks restored AFMID levels, suppressed DNA damage,

abolished the DDR, and reduced acetylation of p53 at Lys-379,

possibly due to activated NAD+-dependent SIRT1 (Luo et al.,

2001) (Figures 5H and 5I). Thus, continuous hURI expression

and consequent inhibition of de novo NAD+ synthesis is essential

for abolishing DNA repair and accelerating tumor formation.

Next, we explored whether other oncogenes had similar ef-

fects. Ela-1-myc mice, unlike K-RasG12V mice, develop pancre-

atic adenocarcinomas with high levels of DNA damage, while

pancreatic tumors initiated by K-RasG12V show no signs of repli-

cative stress (Murga et al., 2011). In early stages of tumori-

genesis, c-Myc, but not K-RASG12V, expression induced DNA

damage (Figures S5I and S5J). TOD2 and AFMID were clearly

downregulated in Ela-1-myc, but not in K-RasG12V pancreas (Fig-


ure S5K). In 3-week-old Ela-1-myc mice, 4 weeks of NR diet did

not affect acinar-to-ductal metaplasia (ADM), but 12 weeks of

NR diet decreased ADM and carcinomas formation compared

to chow fed mice (Figures S5L–S5O). Importantly pancreatic

NAD+ levels were significantly reduced in Ela-1-myc mutants

on chow diet, but restored to almost control levels on NR diet

(Figure S5P). Thus, oncogene-induced DNA damage has a

common bearing on NAD+ levels.

URI Regulates Kynurenine Metabolism by ModulatingAhR and ER ActivityWe found significant overlaps in differentially expressed tran-

scripts between our RNA sequencing and published microarray

data sets for livers from aryl hydrocarbon receptor (AhR) and es-

trogen receptor (ER) knockout mice (Figures 6A, S6A, and S6B),

suggesting that AhR and ER mediate hURI-induced transcrip-

tional repression of L-tryptophan/kynurenine catabolism. TDO2

Figure 6. URI Regulates Kynurenine Metabolism by Modulating AhR and ER Activity

(A) GSEA using microarray data from Ahr�/� and Esr�/� livers and RNA sequencing data from 1 week hURI-tetOFFhep mice.

(B) WB analysis of human HepG2 cells transfected with scramble (siCtr) or siRNA against URI (siURI), AhR (siAhR), or ER (siER).

(C) Immunoprecipitation of cytosolic liver fractions from 1 week hURI-tetOFFhep mice and WB analysis.

(D) WB analysis of cytosolic fractions in livers from 1 week hURI-tetOFFhep mice.

(E) AhR and ER immunofluorescence of 1 week hURI-tetOFFhep liver sections. DAPI was used for nuclear staining. Lower panels depict nuclear colocalization

performed by using Image J. (n = 5).

(F) Quantification of nuclear colocalization of AhR and ER shown in (E).

Data represented as mean ± SD (**p % 0.01). See also Figure S6.

Cancer Cell



was also found deregulated in the core enriched Esr�/� data sets

in GSEA analysis. Depletion of AhR and ER in HepG2 cells

reduced expression of TDO2 (and AFMID), while URI downregu-

lation increased their abundance (Figure 6B). ALGGEN-PROMO

v3.0 software predicted several AhR and ER binding sites in

genomic sequences 5 kilobases upstream of the transcriptional

start sites of TDO2 and AFMID. Chromatin immunoprecipitation

assays revealed that both AhR and ER bound to these promoters

in SNU-449 cells (Figures S6C and S6D). We also verified hURI-

induced downregulation of other AhR and ER targets detected

in the RNA sequencing and iTRAQ analyses, including car-

bomyl-phosphate synthase 1, glutayl-CoA dehydrogenase,

and glycine N-methyltransferase 1 (GNMT1) (Figures S6E and

S6F). Notably, Gnmt1�/� mice develop chronic hepatitis and

spontaneous HCC (Liao et al., 2009). Thus, hURI can repress

AhR and ER transcriptional activity, implicated in transcription

of several metabolic enzymes in particular from the L-tryptophan

catabolism pathway.

AhR and ER are in an inactive cytoplasmic complex with

HSP90, a member of the URI prefoldin complex (Boulon

et al., 2010; Knoblauch and Garabedian, 1999; Perdew, 1988).

Reciprocal coimmunoprecipitation experiments confirmed that

hURI and HSP90 interact with AhR or ER in cytosolic extracts

of 1-week-old mutant livers (Figure 6C). Cytoplasmic fractions

of mutant livers were also enriched with both nuclear receptors

(Figure 6D). Immunofluorescence analysis detected significant

reductions in nuclear AhR and ER in hepatocytes of 1-week-

old mutants (Figures 6E, 6F, and S6G). Finally, nuclear and

cytoplasmic fractionation of livers from 3-week-old DEN-

treated URI(+/+)hep mice showed a positive correlation between

URI expression and cytoplasmic AhR/ER localization, but

an inverse correlation between nuclear URI and AhR/ER. In

DEN-treated URI(+/D)hep livers, AhR and ER were enriched in

the nucleus (Figures S6H and S6I). Thus, hURI/HSP90 inhibitory

cytoplasmic complex prevents AhR and ER transcriptional

activity.


Figure 7. URI Expression Is Enhanced in HCC, Is Associated with Poor Survival, and Correlates with NAD+ Synthesis Inhibition

(A) Representative images of IHC stained for URI in human liver sections.

(B) Stratification of human samples according to URI expression as scored in (A). Values represent number of cases, and values within brackets represent

percentage of total.

(C) Kaplan Meier analysis of overall patient survival based on IHC for URI expression in HCC. URI positive includes weak, moderate, and strong signal in IHC

as described in (A). (Log rank test p = 0.0035; hazard ratio = 0.3163; and 95% confidence interval of ratio 0.1460 to 0.68).

(D) Stratification and correlation of URI expression in HCC samples with human HCC etiological factors.

(E) IHC for URI in normal and hepatitis human liver samples.

(F) Stratification of human hepatitis samples according to URI expression (n = 15). Values represent number of cases, and values within brackets represent

percentage of total.

(G) GSEA between differentially expressed genes in human HBV-associated HCC and RNA sequencing data sets from 1- and 8-week-old hURI-tetOFFhep mice.

(H) Linear regression analysis of URI1 and TDO2, KMO, and HAAO expressions in a human HCCmicroarray data set, showing inverse correlation between URI1

and TDO2, KMO, and HAAO expression.

(I) Multivariate Cox regression analysis for TDO2, KMO, HAAO, and QPRT, in a cohort of 221 HCC patients. (p = 0.025). ‘‘df’’ represents degrees of freedom

and ‘‘Sig.’’ represents significance. See also Figure S7.

Cancer Cell



URI Expression Is Enhanced in Human HCC, IsAssociated with Poor Survival, and Correlates with DeNovo NAD+ Synthesis InhibitionWe examined URI in a tissue-microarray (TMA) of human liver

samples (82 HCC, 4 peritumoral, and 9 normal livers), using a

specific URI antibody (Figure S7A). No URI was detected in

normal livers, whereas various levels of URI were detected in

HCC (Figures 7A and 7B). Increased URI levels in 20 human

HCC, relative to paired peritumoral samples, were also detected

by real-time PCR (data not shown) and WB. URI was approxi-

mately 2-fold higher in 70% of the tumoral tissues as in the peri-

tumoral counterparts (Figures S7B and S7C), similar to hURI

expression in the hURI-tetOFFhep mouse. URI expression and

Ki67 staining were positively correlated (Figure S7D). Importantly,

URI expression was associated with poor prognosis (Figure 7C).


Data stratification indicated a significant correlation between

URI and hepatitis B virus (HBV)- or hepatitis C virus (HCV)-asso-

ciated HCC (Figure 7D). Increased URI expression was also

observed in human hepatitis samples, which predisposes to

hepatocarcinogenesis (Figures 7E and 7F). URI expression was

thus analyzed in a concanavalin A (ConA)-induced mouse hepa-

titis model. C57BL/6 mice administered with ConA had a

dramatic increase in hepatic URI after 4 to 8 hr. IHC analysis

confirmed that URI was confined to hepatocytes (Figures S7E

and S7F). Furthermore, HCC cell lines transiently transfected

with HBV viral protein HBx enhanced URI expression (Figures

S7G and S7H). When introduced into Huh-7 cells, expression

of a mouse Uri1 promoter luciferase reporter was increased

about 1.5-fold by HBx (Figures S7I and S7J). Finally, GSEA de-

tected significant overlaps between transcriptomic signatures

Figure 8. Scheme of URI-Induced HCC

Scheme representing molecular and cellular

events of hepatocarcinogenesis induced by URI

specifically expressed in hepatocytes. Hexagons

represent hepatocytes.

Cancer Cell



of hURI genetically engineered mouse model (GEMM) and HBV-

associated human HCC (Figure 7G). Thus, in human HCC, URI

expression can be regulated by HBV infection or by infection-

induced inflammatory cues.

Next, we examined correlations between URI and L-trypto-

phan/kynurenine pathway. WB analysis of the paired peritu-

moral/HCC samples revealed that AFMID and NAD+ levels

were positively correlated, and both negatively correlated with

URI expression (Figures S7K–S7O). Analysis of a human HCC

gene expression data set (Wurmbach et al., 2007) showed in-

verse correlations between URI and TDO2, KMOandHAAO (Fig-

ure 7H). Finally, in a 221 patient data set (Roessler et al., 2010)

(Gene Expression Omnibus [GEO] ID: GSE14520), Cox regres-

sion analysis of TDO2, KMO, HAOO, and quinolinate phosphor-

ibosyltransferase (QPRT) expression, indicated that overall,

patients’ survival was significantly associated with L-tryptophan

catabolism signature (Figures 7I and S7P). Downregulation of

TDO2 or HAAO also correlated with poor prognosis for HCC

patients (Figures S7Q and S7R).

DISCUSSION

The presence of fibrosis/cirrhosis associated with chromosomal

abnormalities is the most convincing clinical aspect of HCC, but

the molecular mechanisms and pathogenesis of HCC are still

poorly understood (Teoh et al., 2008). Using GEMMs, URI-medi-

ated NAD+ depletion is shown to induce DNA damage, liver

injury, and multistep HCC (Figure 8).

Liver injury predisposes to HCC (Malhi andGores, 2008), but in

our model ALT values at precancerous stages are normal.

Normal ALT levels are observed in chronic HBV or HCV infected

patients withmild tomoderate histological liver damage (Nunnari

et al., 2013), indicating that ALT levels are not always determi-

nants for hepatic injuries. However, persistently elevated ALT

increase HCC risks and are clear indicators of HCC in patients

with viral hepatitis (Chen et al., 2011; Lee et al., 2010), suggesting

that chronic hepatocyte death is a key trigger of liver disease

Cancer Cell 26, 1–14

progression (Luedde et al., 2014). Geno-

toxic stress-induced p53 is required for

apoptosis, but when p53 is inactivated

and hepatocyte death is reduced, carci-

nogenesis is accelerated. Thus, geno-

toxic stress via NAD+ deficits, rather

than high-grade apoptosis, initiates liver

tumorigenesis.

Therapies increasing NAD+ levels

(e.g., NR) can be used to prevent HCC

and cancers resulting from oncogene-

induced DNA damage, but it remains to

be determined whether boosting NAD+

is also therapeutic in nongenotoxic

cancers. NR has also surprising therapeutic effects on fully

developed liver tumors. Given that DNA damage-mediated

chromosomal rearrangements are irreversible, boosting NAD+

levels may activate the mitochondria SIRT3, a proapoptotic

tumor suppressor (Verma et al., 2013) enhancing cancer cell

apoptosis and tumor regression. Elevated SIRT3 levels, specif-

ically in tumor cells responding to NAD+ boost, remain to be

elucidated.

Multiple epidemiologic studies report on the association be-

tween tryptophan-poor diets and increased specific cancer

types incidences (Surjana et al., 2010). Daily supplementation

of niacin, a NAD+ precursor, reduced esophageal cancer inci-

dence and mortality in a population with chronic nutritional defi-

ciency (Surjana et al., 2010), and, in which gut microbiota might

be altered. Gut bacteria cleave the side chain of tryptophan

(Burns and Demoss, 1962), limiting its concentration and use

in de novo NAD+ synthesis (Michael et al., 1964). The implication

of the intestinal microbiota was suggested in HCC development,

although NAD+ levels remain to be determined (Dapito et al.,

2012). Dysbiosis can also have a tumor promoting effect

(Schwabe and Jobin, 2013), and malnutrition altering the intes-

tinal flora induces HCC (Yoshimoto et al., 2013). Future

cartographic determination of patients’ microbiota and identifi-

cation of specific harmful bacteria in diseases would be impor-

tant to better understand the symbiosis between bacteria and

mammals.

Activating AhR or ER can also be beneficial in HCC. Inhibiting

AhR generates xenobiotic stress, which accelerates liver tumor-

igenesis (Walisser et al., 2005), and DEN-treated AhR�/� mice

increased tumor incidence compared to their littermates, sug-

gesting that AhR may be tumor suppressive (Fan et al., 2010).

Several epidemiologic and animal studies also suggest a protec-

tive effect of estrogen (Naugler et al., 2007). Additionally, estro-

gen was effective in the treatment of traumatic liver injury (Hsieh

et al., 2007). The nongender disparity in HCC development in our

hURI mouse model is consistent with the role of ER inhibition in

HCC progression.

, December 8, 2014 ª2014 Elsevier Inc. 11

Cancer Cell



URI inhibition may thus represent a therapeutic option at early

stages of liver tumorigenesis, and combined therapies that syn-

ergistically activate AhR and ER should be tested in preclinical

models for HCC treatment, in particular in patients with high

URI expression. Finally, the development of more efficient and

stable NAD+ boosters could provide therapies to prevent or

cure cancers and associated metabolic dysfunctions.

EXPERIMENTAL PROCEDURES

Generation and Handling of Mice

All mice have been backcrossed to C57BL/6 for at least seven generations and

housed in pathogen-free conditions. All experiments were approved by the

Centro Nacional de Investigaciones Oncologicas (CNIO)-Instituto de Salud

Carlos III Ethics Committee and performed in accordance with the guidelines

for ethical conduct in the care and use of animals as stated in the international

guiding principles for biomedical research involving animals, developed by the

Council for International Organizations of Medical Sciences. Littermates were

always used as controls.

Liver Carcinogenesis, Injury Models, and Mouse Treatments

Fourteen-day-old mice were injected intraperitoneally with 25 mg/kg of DEN

(Sigma) (Vesselinovitch and Mihailovich, 1983).

For ConA treatment, 8-week-old C57/BL6 male mice were intravenously in-

jected with 15 mg/kg of ConA (Sigma Aldrich) and sacrificed at 2 hr intervals.

DDC was mixed with chow diet to a final concentration of 0.5% w/w (Harlan)

and supplied as indicated in the experiment.

Ro-61-8048 (Sigma-Aldrich, SML0233) was dissolved to 593 mM in DMSO.

The 9-week-old C57BL/6 mice were given DDC for 4 days and then switched

to chow and injected with either Ro-61-8048/Sunflower Seed Oil (25mg/Kg) or

DMSO/Sunflower Seed Oil (1:100) intraperitoneally for three consecutive days.

Nicotinamide riboside (97% purity, Waterstonetech Pharma) was dissolved

in ice cold water and immediately mixed thoroughly with cold amorphous

chow diet (Harlan) at 500 mg/kg/day and supplied ad-libitum.

Tumor Quantification

Macroscopically visible tumor nodules were counted and sizes measured with

Vernier calipers. Total liver size was measured in the same position, and rela-

tive tumor burden was expressed as percentage of tumor volume relative to

the whole liver.

Human Samples

Human samples were obtained from the histopathology files of University Col-

lege Hospital, University College London (UCL), from patients after approval

by the Institutional Research Ethics Committee (Central London REC 3, Refer-

ence 06/Q0512/106) and from the CNIO-Biobank. The construction and anal-

ysis of a tissue microarray of HCC was approved by the appropriate ethics

committee, and informed consent was obtained from all subjects.

[14C]-Tryptophan Metabolic Tracing

HCC cells were transfected with siCtr or siAFMID grown in a 12 well plate

until they reached 60% confluence and starved for 5 hr for tryptophan

in tryptophan-free media. 2.5 mM of [benzene-ring-U-14C]-tryptophan was

provided to cells and incubated at 37�C for 5 hr. Cells were then washed

three times with cold PBS, and metabolites extracted in a methanol and

water (80:20) mixture and incubated for 10 min at 4�C. Metabolic lysates

were centrifuged at maximum speed for 20 min at 4�C. Radiolabeled sam-

ples were separated by thin layer chromatography (TLC) using ammonium

acetate (1M, pH5) and ethanol (30:70) and cellulose F plates. Labeled

NAD+-[carbonyl-14C] was used as positive control to calibrate the relative

migration of labeled metabolites. TLC plates were dried and exposed to a

PhosphorImager.

Proteomic Analysis

Extracted liver proteins were digested using a modified filter aided sample

prep protocol. Peptides were labeled with iTRAQ reagents and samples

were pooled. The complex mixture was subjected to isoelectric focusing frac-


tionation. The resulting fractions were separated by on-line nano-liquid chro-

matography and analyzed by electrospray mass spectrometry (MS)/MS using

a linear trap quadrupole Orbitrap Velos mass spectrometer (Thermo Scienti-

fic). Raw files were searched against UniProtKB/Swiss-Prot mouse database

(release date, October 19, 2011; 16,407 entries) using MASCOT (Matrix Sci-

ence, 2013). Peptides were filtered at 1% FDR using a concatenated database

(see also Supplemental Experimental Procedures).

Reporter Assays

Evolutionarily conserved, 440 base pairs regulatory sequence of URI ORF

were cloned in pGL4.10-Luc vector to generate the URI reporter plasmid.

Huh-7 cells were transfected with 200 ng URI reporter and 5 ng Renilla encod-

ing plasmids using Lipofectamine 2000. After 2 days of transfection, cells

were analyzed using the Dual-Luciferase Reporter Assay System (#E1960,

Promega). Values after pCDNA3-GFP or pCDNA3-HA-HBx transfection were

normalized to a Renilla control.

Statistical Analyses

Statistical analyses were performed using GraphPad Prism V5.0 software

(GraphPad Software, 2007). Statistical significance (p) (p % 0.05 = *, p %

0.01 = **, p% 0.001 = ***, and p% 0.0001 = ****) between the means of a min-

imum of three groups was determined using unpaired two-tailed Student’s

t test. Results are expressed as the mean value ± SD or ± SEM as indicated.

All results including WB analysis are representative of at least three indepen-

dent experiments. The Kaplan-Meier method was used to estimate survival

curves for mouse and human, and log rank was used to evaluate statistical dif-

ferences. Statistical parameter (FDR) estimates the probability of a gene set

with false positive finding. Normalized enrichment score allows comparison

of enrichment analysis results across gene sets. Cox regression and survival

analysis were performed for target genes in human HCC data sets, using

SPSS software (v20) (IBM, 2011).

ACCESSION NUMBERS

The proteomic data are deposited to the ProteomeXchange Consortium

(http://proteomecentral.proteomexchange.org) via the PRIDE partner reposi-

tory with the data set identifier PXD000296 (Vizcaino et al., 2013). RNA

sequencing data are available from GEO (http://www.ncbi.nlm.nih.gov/geo)

with the accession number GSE48654.

SUPPLEMENTAL INFORMATION

Supplemental Information includes Supplemental Experimental Procedures,

seven figures, and two tables and can be found with this article online at

http://dx.doi.org/10.1016/j.ccell.2014.10.002.

AUTHOR CONTRIBUTIONS

K.S.T. designed and performed most of the experiments. K.S.T. and N.D. en-

gineered the ColhURI allele and generated the hURI-tetOFFhepmouse. A.L.G.

analyzed fibrosis data. M.Y. and N.D. engineered the URI(lox/lox) allele and

analyzed the URI promoter. O.G. and D.G.P. performed the bioinformatic

analysis. L.B. and E.W. helped in generating the ColhURI allele. I.R. and

P.X.E. performed the iTRAQ experiment. V.S. and M.R.J. analyzed the human

liver samples. K.S.T. and N.D. analyzed all the data. N.D. designed the exper-

iments and conceived, developed, and wrote the project and the manuscript.

Funding was secured by N.D.

ACKNOWLEDGMENTS

We are thankful to F. Real and M. Barbacid for providing the Ela-1-myc and

K-RasG12V pancreatic models, respectively. We thank R. Ricci, M. Serrano,

R. Hamacher, G. Gomes, S. Wurm, F. Diaz, and S. Anderson for support

and advice. K.S.T. is a recipient of La Caixa predoctoral fellowship. A.L.G. is

a recipient of the Caja Navarra postdoctoral fellowship. P.X.E. is a recipient

of the Fondo de Investigaciones Sanitarias grant (CA10/01231). M.R.J. is sup-

ported by UCL Hospitals Biomedical Research Centre. The E.F.W. lab is

http://proteomecentral.proteomexchange.org

http://www.ncbi.nlm.nih.gov/geo


Cancer Cell



supported by F-BBVA, the Spanish Ministry of Economy and Competitiveness

(BFU201240230) and the European Research Council (ERC)-Advanced grant

(ERC-FCK/2008/37). N.D. is a recipient of the Spanish Ramon y Cajal

fellowship. This work was supported by the Spanish Ministry of Economy

and Competitiveness (SAF2010 - 18518), the Association for International

Cancer Research AICR-UK (11-0242), CNIO (BC1104-08), and the European

Foundation for the Study of Diabetes.

Received: March 22, 2014

Revised: July 23, 2014

Accepted: October 2, 2014

Published: November 20, 2014

REFERENCES

Boulon, S., Pradet-Balade, B., Verheggen, C., Molle, D., Boireau, S.,

Georgieva, M., Azzag, K., Robert, M.C., Ahmad, Y., Neel, H., et al. (2010).

HSP90 and its R2TP/Prefoldin-like cochaperone are involved in the cyto-

plasmic assembly of RNA polymerase II. Mol. Cell 39, 912–924.

Burns, R.O., and Demoss, R.D. (1962). Properties of tryptophanase from

Escherichia coli. Biochim. Biophys. Acta. 65, 233–244.

Carpenter, B., Lin, Y., Stoll, S., Raffai, R.L., McCuskey, R., and Wang, R.

(2005). VEGF is crucial for the hepatic vascular development required for lipo-

protein uptake. Development 132, 3293–3303.

Chen, C.F., Lee, W.C., Yang, H.I., Chang, H.C., Jen, C.L., Iloeje, U.H., Su, J.,

Hsiao, C.K., Wang, L.Y., You, S.L., et al. (2011). Changes in serum levels of

HBV DNA and alanine aminotransferase determine risk for hepatocellular car-

cinoma. Gastroenterology 141, 1240–1248, 1248 e1241–1242.

Dapito, D.H., Mencin, A., Gwak, G.Y., Pradere, J.P., Jang,M.K., Mederacke, I.,

Caviglia, J.M., Khiabanian, H., Adeyemi, A., Bataller, R., et al. (2012).

Promotion of hepatocellular carcinoma by the intestinal microbiota and

TLR4. Cancer Cell 21, 504–516.

Djouder, N., Metzler, S.C., Schmidt, A., Wirbelauer, C., Gstaiger, M.,

Aebersold, R., Hess, D., and Krek, W. (2007). S6K1-mediated disassembly

of mitochondrial URI/PP1gamma complexes activates a negative feedback

program that counters S6K1 survival signaling. Mol. Cell 28, 28–40.

Durkacz, B.W., Omidiji, O., Gray, D.A., and Shall, S. (1980). (ADP-ribose)n par-

ticipates in DNA excision repair. Nature 283, 593–596.

El-Serag, H.B. (2011). Hepatocellular carcinoma. N. Engl. J. Med. 365, 1118–

1127.

Fan, Y., Boivin, G.P., Knudsen, E.S., Nebert, D.W., Xia, Y., and Puga, A. (2010).

The aryl hydrocarbon receptor functions as a tumor suppressor of liver carci-

nogenesis. Cancer Res. 70, 212–220.

Herranz, D., Munoz-Martin, M., Canamero,M., Mulero, F., Martinez-Pastor, B.,

Fernandez-Capetillo, O., and Serrano, M. (2010). Sirt1 improves healthy

ageing and protects from metabolic syndrome-associated cancer. Nat.

Commun. 1, 3.

Hsieh, Y.C., Yu, H.P., Frink, M., Suzuki, T., Choudhry, M.A., Schwacha, M.G.,

and Chaudry, I.H. (2007). G protein-coupled receptor 30-dependent protein ki-

nase A pathway is critical in nongenomic effects of estrogen in attenuating liver

injury after trauma-hemorrhage. Am. J. Pathol. 170, 1210–1218.

Ito, A., Lai, C.H., Zhao, X., Saito, S., Hamilton, M.H., Appella, E., and Yao, T.P.

(2001). p300/CBP-mediated p53 acetylation is commonly induced by p53-

activating agents and inhibited by MDM2. EMBO J. 20, 1331–1340.

Kang, J.S., Wanibuchi, H., Morimura, K., Gonzalez, F.J., and Fukushima, S.

(2007). Role of CYP2E1 in diethylnitrosamine-induced hepatocarcinogenesis

in vivo. Cancer Res. 67, 11141–11146.

Knoblauch, R., and Garabedian, M.J. (1999). Role for Hsp90-associated

cochaperone p23 in estrogen receptor signal transduction. Mol. Cell. Biol.

19, 3748–3759.

Konishi, Y., Takahashi, S., Nakae, D., Uchida, K., Tsutsumi, M., Shiraiwa, K.,

and Denda, A. (1986). Possible model of liver carcinogenesis using inhibitors

of NAD+ ADP ribosyl transferase in rats. Toxicol. Pathol. 14, 483–488.

Kudo, M. (2009). Multistep human hepatocarcinogenesis: correlation of imag-

ing with pathology. J. Gastroenterol. 44 (Suppl 19 ), 112–118.

Lee, M.H., Yang, H.I., Lu, S.N., Jen, C.L., Yeh, S.H., Liu, C.J., Chen, P.J., You,

S.L., Wang, L.Y., Chen, W.J., and Chen, C.J. (2010). Hepatitis C virus sero-

markers and subsequent risk of hepatocellular carcinoma: long-term predic-

tors from a community-based cohort study. J. Clin. Oncol. 28, 4587–4593.

Liao, Y.J., Liu, S.P., Lee, C.M., Yen, C.H., Chuang, P.C., Chen, C.Y., Tsai, T.F.,

Huang, S.F., Lee, Y.H., and Chen, Y.M. (2009). Characterization of a glycine

N-methyltransferase gene knockout mouse model for hepatocellular carci-

noma: Implications of the gender disparity in liver cancer susceptibility. Int.

J. Cancer 124, 816–826.

Libbrecht, L., Craninx, M., Nevens, F., Desmet, V., and Roskams, T. (2001).

Predictive value of liver cell dysplasia for development of hepatocellular carci-

noma in patients with non-cirrhotic and cirrhotic chronic viral hepatitis.

Histopathology 39, 66–73.

Luedde, T., Kaplowitz, N., and Schwabe, R.F. (2014). Cell death and cell

death responses in liver disease: Mechanisms and clinical relevance.

Gastroenterology 147, 765, e4.

Luo, J., Nikolaev, A.Y., Imai, S., Chen, D., Su, F., Shiloh, A., Guarente, L., and

Gu, W. (2001). Negative control of p53 by Sir2alpha promotes cell survival

under stress. Cell 107, 137–148.

Malhi, H., and Gores, G.J. (2008). Cellular and molecular mechanisms of liver

injury. Gastroenterology 134, 1641–1654.

Menon, S., Yecies, J.L., Zhang, H.H., Howell, J.J., Nicholatos, J., Harputlugil,

E., Bronson, R.T., Kwiatkowski, D.J., and Manning, B.D. (2012). Chronic acti-

vation of mTOR complex 1 is sufficient to cause hepatocellular carcinoma in

mice. Sci. Signal. 5, ra24.

Michael, A.F., Drummond, K.N., Doeden, D., Anderson, J.A., and Good, R.A.

(1964). Tryptophan metabolism in man. J. Clin. Invest. 43, 1730–1746.

Murga, M., Campaner, S., Lopez-Contreras, A.J., Toledo, L.I., Soria, R.,

Montana, M.F., D’Artista, L., Schleker, T., Guerra, C., Garcia, E., et al.

(2011). Exploiting oncogene-induced replicative stress for the selective killing

of Myc-driven tumors. Nat. Struct. Mol. Biol. 18, 1331–1335.

Naugler, W.E., Sakurai, T., Kim, S., Maeda, S., Kim, K., Elsharkawy, A.M., and

Karin, M. (2007). Gender disparity in liver cancer due to sex differences in

MyD88-dependent IL-6 production. Science 317, 121–124.

Nunnari, G., Pinzone, M.R., and Cacopardo, B. (2013). Lack of clinical

and histological progression of chronic hepatitis C in individuals with true

persistently normal ALT: the result of a 17-year follow-up. J. Viral Hepat.

20, e131–e137.

Perdew, G.H. (1988). Association of the Ah receptor with the 90-kDa heat

shock protein. J. Biol. Chem. 263, 13802–13805.

Pilati, C., Letouze, E., Nault, J.C., Imbeaud, S., Boulai, A., Calderaro, J.,

Poussin, K., Franconi, A., Couchy, G., Morcrette, G., et al. (2014). Genomic

profiling of hepatocellular adenomas reveals recurrent FRK-activating

mutations and the mechanisms of malignant transformation. Cancer Cell 25,

428–441.

Rajamohan, S.B., Pillai, V.B., Gupta, M., Sundaresan, N.R., Birukov, K.G.,

Samant, S., Hottiger, M.O., and Gupta, M.P. (2009). SIRT1 promotes cell

survival under stress by deacetylation-dependent deactivation of poly(ADP-

ribose) polymerase 1. Mol. Cell. Biol. 29, 4116–4129.

Reinhardt, H.C., and Schumacher, B. (2012). The p53 network: cellular and

systemic DNA damage responses in aging and cancer. Trends Genet. 28,

128–136.

Roessler, S., Jia, H.L., Budhu, A., Forgues, M., Ye, Q.H., Lee, J.S.,

Thorgeirsson, S.S., Sun, Z., Tang, Z.Y., Qin, L.X., and Wang, X.W. (2010).

A unique metastasis gene signature enables prediction of tumor relapse

in early-stage hepatocellular carcinoma patients. Cancer Res. 70, 10202–

10212.

Schuler, M., Dierich, A., Chambon, P., andMetzger, D. (2004). Efficient tempo-

rally controlled targeted somatic mutagenesis in hepatocytes of the mouse.

Genesis 39, 167–172.

Schwabe, R.F., and Jobin, C. (2013). The microbiome and cancer. Nature

Reviews 13, 800–812.

Subramanian, A., Tamayo, P., Mootha, V.K., Mukherjee, S., Ebert, B.L.,

Gillette, M.A., Paulovich, A., Pomeroy, S.L., Golub, T.R., Lander, E.S., and


Cancer Cell



Mesirov, J.P. (2005). Gene set enrichment analysis: a knowledge-based

approach for interpreting genome-wide expression profiles. Proc. Natl.

Acad. Sci. USA 102, 15545–15550.

Surjana, D., Halliday, G.M., and Damian, D.L. (2010). Role of nicotinamide

in DNA damage, mutagenesis, and DNA repair. J. Nucleic Acids 2010,

2010.

Teoh, N.C., Dan, Y.Y., Swisshelm, K., Lehman, S., Wright, J.H., Haque, J., Gu,

Y., and Fausto, N. (2008). Defective DNA strand break repair causes chromo-

somal instability and accelerates liver carcinogenesis in mice. Hepatology 47,

2078–2088.

Theurillat, J.P., Metzler, S.C., Henzi, N., Djouder, N., Helbling, M.,

Zimmermann, A.K., Jacob, F., Soltermann, A., Caduff, R., Heinzelmann-

Schwarz, V., et al. (2011). URI is an oncogene amplified in ovarian cancer cells

and is required for their survival. Cancer Cell 19, 317–332.

Trinidad, A.G., Muller, P.A., Cuellar, J., Klejnot, M., Nobis, M., Valpuesta, J.M.,

and Vousden, K.H. (2013). Interaction of p53 with the CCT complex promotes

protein folding and wild-type p53 activity. Mol. Cell 50, 805–817.

Verma, M., Shulga, N., and Pastorino, J.G. (2013). Sirtuin-3 modulates Bak-

and Bax-dependent apoptosis. J. Cell Sci. 126, 274–288.

Vesselinovitch, S.D., and Mihailovich, N. (1983). Kinetics of diethylnitrosamine

hepatocarcinogenesis in the infant mouse. Cancer Res. 43, 4253–4259.


Vizcaino, J.A., Cote, R.G., Csordas, A., Dianes, J.A., Fabregat, A., Foster, J.M.,

Griss, J., Alpi, E., Birim, M., Contell, J., et al. (2013). The PRoteomics

IDEntifications (PRIDE) database and associated tools: status in 2013.

Nucleic Acids Res. 41, D1063–D1069.

Walisser, J.A., Glover, E., Pande, K., Liss, A.L., and Bradfield, C.A. (2005). Aryl

hydrocarbon receptor-dependent liver development and hepatotoxicity are

mediated by different cell types. Proc. Natl. Acad. Sci. USA 102, 17858–17863.

WHO. (2008). Tumours of the liver and intrahepatic bile ducts. In Pathology

and Genetics of Tumors of the Digestive System, S.R. Hamilton and L.A.

Aaltonen, eds. (IARC Press). http://www.iarc.fr/en/publications/pdfs-online/

pat-gen/bb2/BB2.pdf.

Wurmbach, E., Chen, Y.B., Khitrov, G., Zhang, W., Roayaie, S., Schwartz, M.,

Fiel, I., Thung, S., Mazzaferro, V., Bruix, J., et al. (2007). Genome-wide

molecular profiles of HCV-induced dysplasia and hepatocellular carcinoma.

Hepatology 45, 938–947.

Yoshimoto, S., Loo, T.M., Atarashi, K., Kanda, H., Sato, S., Oyadomari, S.,

Iwakura, Y., Oshima, K., Morita, H., Hattori, M., et al. (2013). Obesity-induced

gut microbial metabolite promotes liver cancer through senescence secre-

tome. Nature 499, 97–101.

http://www.iarc.fr/en/publications/pdfs-online/pat-gen/bb2/BB2.pdf

http://www.iarc.fr/en/publications/pdfs-online/pat-gen/bb2/BB2.pdf

Cancer Cell, Volume 26

Supplemental Information

Inhibition of De Novo NAD+ Synthesis

by Oncogenic URI Causes Liver

Tumorigenesis through DNA Damage

Krishna S. Tummala, Ana L. Gomes, Mahmut Yilmaz, Osvaldo Graña, Latifa Bakiri, Isabel Ruppen, Pilar Ximénez-Embún, Vinayata Sheshappanavar, Manuel Rodriguez-Justo, David G. Pisano, Erwin F. Wagner, and Nabil Djouder

1

SUPPLEMENTAL DATA

2

3

4

Figure S1, related to Figure 1. URI Expression in Mouse Hepatocytes Induces

Spontaneous Liver Tumors

5

(A) Scheme of knock-in strategy of hURI in the Col1a1 locus. Red line depicts 3’ probe

used.

(B) Southern blot analysis using 3’probe showing correct targeting of hURI in the Col1a1

locus.

(C) Schematic representation of the model hURI-tetOFFhep mouse in which URI expression

is under the control of the hepatocyte-specific LAP promoter.

(D) Quantification of hURI mRNA expression in different organs derived from hURI-

tetOFFhep mice. Denote hepatic specificity of the model. (n ≥ 5).

(E) WB analysis of hURI expression in different organs derived from hURI-tetOFFhep mice.

(F) WB analysis of hURI and endogenous URI (mURI) expression in livers from control and

mutant hURI-tetOFFhep mice.

(G) Quantification of (F).

(H) Hematoxylin stained liver sections in 12-week-old mutant hURI-tetOFFhep mice. Image

displays characteristics of human Large Liver Cell Dysplasia (LLCD) like lesions. Insets

show nuclear pseudo inclusions, large pleomorphic nuclei and hepatocytes with binucleus

and prominent nucleoli. (n ≥ 5).

(I) Table showing disease progression with age in mutant hURI-tetOFFhep mice, and

resembling the human pathological state.

(J) Representative images of Sirius Red (with/without green color), Masson Trichrome, alpha

smooth muscle actinACTA2), collagen, type 1 (COL1A1) and Reticulin stained livers

showing the fibrotic area in 8-, 12- and 24-week-old hURI-tetOFFhep mice. (n ≥ 5).

(K) Quantification of Sirius Red positive area in 8-, 12-, 24-, 32- and 54-week-old hURI-

tetOFFhep mice. (n ≥ 5).

6

(L-N) qRT-PCR analysis of (Acta2) (L), collagen, type1, alpha 1 (Col1a1) (M) and tissue

inhibitor of metalloproteinase 1 (Mmp1) (N) transcripts expressed in livers of mice at 8, 12

and 24 weeks of age. -Actin was used as housekeeping gene. (n ≥ 5).

(O) Serum alanine aminotransferase (ALT) from 8-, 12- and 24-week-old hURI-tetOFFhep

mice.

(P) Tumor burden defined as tumor size relative to the total liver size and total number of

tumors per liver in hURI-tetOFFhep mutant mice. (n = 12).

(Q) Quantification of ALT, albumin, bile acids and glucose in serum from 65-week-old

hURI-tetOFFhep mice. (n ≥ 5).

(R) Representative images of lungs from hURI-tetOFFhep mice at 75 weeks of age. Yellow

arrows depict lung metastasis. Insets show H&E images with macro metastatic nodules in

mutant lungs (dotted yellow circles). Lower images represent immunohostochemistry (IHC)

for hURI in lung tissues. Hepatocyte-like cells are positively stained for hURI.

(S) Percentage of mice presenting lung metastasis in control and mutant hURI-tetOFFhep

mice. (n ≥ 5).

(T) Scheme of DEN treatment in control and mutant hURI-tetOFFhep mice. Right panel

depicts WB analysis in hURI-tetOFFhep mice treated with the carcinogen DEN. Bottom left

panel represents pictures of full livers and H&E stained sections from hURI-tetOFFhep mice

treated with DEN. Dotted yellow circle denotes liver tumor. Dotted black line in the mutant

H&E picture denotes border between HCC and peritumoral (PT) tissue. Bottom right panel

represents histogram of tumor incidence observed in hURI-tetOFFhep mice treated with DEN.

(n ≥ 5).

(U) WB analysis for hURI in livers derived from 12-week-old hURI-tetOFFhep mice. Right

histogram represents hURI quantification. Bottom left panel represents pictures of full livers

7

and H&E stained sections from hURI-tetOFFhep mice at 10 weeks of age. Dotted black line in

the mutant H&E picture denotes border between HCC and peritumoral (PT) tissue. Bottom

right histogram depicts tumor incidence in hURI-tetOFFhep mice of 10 weeks of age. (n ≥ 5).

(V) Schematic representation of the doxycycline treatment used in hURI-tetOFFhep mice as to

take advantage of the switch-ability system. Mice were kept in doxycycline (Dox) diet until 8

weeks of age (+ doxycycline (hURI-)). From 8 weeks on, mice were fed chow diet for hURI

expression in adult stage (- doxycycline (hURI+)) and sacrificed for analysis from 70 weeks

of age. Top right panel represents pictures of hURI IHC from 9-week-old hURI-tetOFFhep

liver sections. (n ≥ 5). Bottom left panel represents pictures of full livers and H&E stained

liver sections from hURI-tetOFFhep mice. Red arrows in mutant H&E picture denote tumoral

and necrotic areas. Bottom right panel represents tumor incidence of mice. (n ≥ 5).

Data represented as mean ± SD. (p ≤ 0.05 = *, p ≤ 0.01 = ** and p ≤ 0.001 = ***).

8

9

Figure S2, related to Figure 2. Continuous URI Expression Is Essential for

Hepatocarcinogenesis

10

(A) Liver to body weight ratio of hURI-tetOFFhep mice treated with doxycycline for 24

weeks. Treatment started at 8 weeks of age, after the appearance of anisokaryotic

hepatocytes. (n ≥ 5). Error bars represent ± SD.

(B) Schematic representation of the long term doxycycline treatment of hURI-tetOFFhep

mice. Treatment started at 8 weeks of age, after the appearance of anisokaryotic hepatocytes

and mice were sacrificed at 60 weeks of age.

(C and D) Representative images of full livers (C) and H&E stained liver sections (D) from

hURI-tetOFFhep mice described in (B).

(E) Serum ALT and Albumin levels of the mice described in (B). Error bars represent ± SD.

(F) Representative images of ultrasound analysis of 60-week-old hURI-tetOFFhep mutants fed

with normal or doxycycline diet for the following 5 weeks (65 weeks). Yellow dotted circles

represent the tumoral area.

(G) Tumor area in 60 week-old hURI-tetOFFhep mutants described in (F). Each dots represent

individual tumors. (n ≥ 5).

(H) Fold increase in total tumor area of mice described in (F). (n ≥ 5).

(I) Kaplan Meier curve of URI(+/+)hep (n = 20) and URI(hep (n = 19) mice .

(J) Representative images of IHC for endogenous URI, H&E, Sirius Red and Reticulin

stained liver sections from URI(+/+)hep and URI(+hep mice. (n ≥ 5).

(K) Quantification of ALT in serum from URI(+/+)hep, URI(+hep and URI(hep mice. (n ≥

5). Error bars represent ± SEM.

(L) Quantification of ALT in serum from URI(+/+)hep and URI(+hep mice treated with DDC

for 1 month. (n ≥ 5). Error bars represent ± SEM.

11

(M) Representative images of Sirius Red and Reticulin stained liver sections from URI(+/+)hep

and URI(+hep mice treated with DDC for 1 month. (n ≥ 5).

(N) Quantification of fibrotic area. Denote decreased fibrotic area with reduction of URI

protein levels. (n ≥ 5). Error bars represent ± SEM.

(O) WB analysis of livers from 3 week-old URI(+/+)hep, URI(+hep mice treated with/without

DEN.

(P) WB analysis for URI in liver tumors from C57Bl/6 mice treated with the carcinogen

DEN. (n = 3 without DEN and n = 4 with DEN).

Data represented as mean ± SEM or ± SD, as indicated. (p ≤ 0.01 = **, p ≤ 0.001 = *** and p

≤ 0.0001 = ****).

12

13

Figure S3, related to Figure 3. URI-Induced DNA Damage Initiates Liver

Tumorigenesis

(A) WB liver analysis of 1 week hURI-tetOFFhep mice.

(B) Representative images of Ki67 stained liver sections of 3-week-old hURI-tetOFFhep mice.

(n ≥ 5).

(C) Quantification of (B). Error bars represent ± SEM.

(D) WB liver analysis of 8-week-old hURI-tetOFFhep mice.

(E) Representative images of -gal stained liver sections in 12-week-old hURI-tetOFFhep

mice. (n ≥ 5).

(F) qRT-PCR analysis of pro-apoptotic p53 upregulated modulator of apoptosis (Bbc3) and

Bcl-2-associated X protein (Bax) and anti-apoptotic X-linked inhibitor of apoptosis protein

(Xiap) transcripts expressed in livers from 8-week-old hURI-tetOFFhep mice. -Actin was

used as housekeeping gene. (n ≥ 5). Error bars represent ± SD.

(G) Representative images of reticulin stained liver sections in 8- and 12-week-old hURI-

tetOFFhep mice. (n ≥ 5).

(H) Quantification of Ki67 positive nuclei in the liver, detected by IHC in 8- and 12-week-old

hURI-tetOFFhep mice. (n ≥ 5). Error bars represent ± SEM.

Data represented as mean ± SEM or ± SD, as indicated. (p ≤ 0.01 = ** and p ≤ 0.001 = ***).

14

15

16

17

18

Figure S4, related to Figure 4. URI Causes DNA Damage and Liver Tumorigenesis by

Inhibiting De Novo NAD+ Synthesis

(A) WB analysis of livers derived from 1-week-old control and mutant hURI-tetOFFhep mice.

(B) Sequential immunoprecipitation (I.P) of PP1 in livers from 1 week hURI-tetOFFhep

mice.

(C) WB analysis of inputs before the first I.P and output after the last I.P performed in (B).

Denote the presence of hURI after complete depletion of PP1 (samples 11, 12).

(D) Quantification of reactive oxygen species (ROS) in 1-week-old hURI-tetOFFhep livers. (n

≥ 4). Error bars represent ± SEM.

(E) Quantification of reactive oxygen species (ROS) in 8-week-old hURI-tetOFFhep livers. (n


(F) Scatter plots representing liver transcripts from 1 and 8 weeks hURI-tetOFFhep mice.

(G) Dot plots of differentially expressed liver proteins analyzed in iTRAQ of 1- and 8-week-

old hURI-tetOFFhep mice.

(H) Heatmaps of downregulated (green) and upregulated (red) liver proteins analyzed in

iTRAQ of 1- and 8-week-old hURI-tetOFFhep mice.

(I) Scatter plots of differentially expressed transcripts and proteins from 1 and 8 weeks RNA

sequencing and iTRAQ data. iTRAQ data set was plotted on x-axis as log (fold change) and

RNA sequencing data set was plotted on y-axis.

(J) GSEA between upregulated and downregulated proteins analyzed in iTRAQ and

upregulated transcripts analyzed in RNA sequencing in livers from 1- and 8-week-old hURI-

tetOFFhep mice.

19

(K) Top downregulated canonical metabolic pathways in 1 week iTRAQ data. Data were

analyzed using Ingenuity Pathway Analysis (IPA) software.

(L) WB analysis of 1 week hURI-tetOFFhep mice. Right histogram represents the

quantification depicted as percentage enzyme reduction in mutant hURI-tetOFFhep mice.

Error bars represent ± SEM.

(M) WB analysis of hURI-tetOFFhep mice with 12 weeks of hURI expression, starting from 8

weeks of age (adult stage).

(N) Liver NAD+ concentrations in 6-week-old hURI-tetOFFhep mice. (n ≥ 5). Error bars

represent ± SEM.

(O) Quantification of NAD+ levels in 3-week-old URI(+/+)hep and URI(+hep mice treated or

not with DEN. Error bars represent ± SEM.

(P) WB analysis of URI levels in different human HCC cells lines: Huh-7, HepG2, SNU-398

and SNU-449. Panel below represents quantification of URI levels with respective to loading

control. Right panel represents quantification of NAD+ levels in human HCC cell lines in

which URI was either depleted with siRNA or transiently overexpressed using pCDNA3-HA-

URI. WB analysis depicts the URI levels after URI depletion with siRNA or after HA-URI

overexpression. Error bars represent ± SEM.

(Q) Cell number of transiently URI-overexpressed (using pcDNA3-HA-URI) in SNU-449

presenting low endogenous URI levels. Cell number of URI depletion (using URI siRNA) in

Huh-7 and HepG2 cells presenting high endogenous URI levels. Error bars represent ± SEM

(Anova test).

(R) Quantification of NAD+ levels in SNU-449 cells previously treated with siRNA Ctr

(siCtr), TDO2 (siTDO2) or AFMID (siAFMID). WB analysis for TDO2 and AFMID in

20

SNU-449 cells previously treated with siCtr, siTDO2 or siAFMID. Error bars represent ±

SEM.

(S) Thin Layer Chromatography (TLC) of [14C]-Tryptophan incorporation into [14C]-NAD+.

Right panel represents quantification of [14C]-NAD+ signal over several TLC experiments. (n


(T) WB of enzymes from other pathways predicted to be downregulated in IPA in 1-week-old

hURI-tetOFFhep livers.

(U) WB of NAMPT (NAD+ salvage pathway) enzyme in 1-week-old hURI-tetOFFhep liver

extracts. Right panels represent quantification of PARP activity and liver NADH levels in 1

week hURI-tetOFFhep mice. Error bars represent ± SEM.

(V) DDC followed by DMSO (1 %) or Ro-61-8048 (25 mg/kg) treatment in C57BL/6 mice.

(W) Representative images of Ki67-stained liver sections from C57Bl/6 mice treated

with/without DDC-containing diet and its quantification (n ≥ 5). Error bars represent ± SEM.

(X) WB analysis of TDO2 and AFMID downregulation in AML-12 cells using three different

shRNA. Separate pools of puromycin-selected AML-12 cells depleted for TDO2 and AFMID

were generated and finally mixed in order to have one pool of three AML-12 cell pools for

TDO2 downregulation (AML-12-shTDO2) and one pool of three AML-12 cells for AFMID

downregulation (AML-12-shAFMID). AML-12-shTDO2 and AML-12-shAFMID pooled

cells were injected subcutaneously into the left and right flanks of 5-week-old female athymic

nude mice. Histograms represent percentage of tumors implanted and tumor volume (mm3)

after 5 months. H&E Pictures are representative of anaplastic lesions and lung metastases.

(shCtr n = 5; shTDO2 n = 14; shAFMID; n = 13). Error bars represent ± SEM.

21

(Y) WB analysis of whole SNU-449 cell extracts treated for 48 hr with Sirt1 inhibitor (EX-

527; 10 M), Sirt1 activator (Resveratrol; 10 M) or PARP inhibitor (10 M; Olaparib) with

or without transient URI overexpression using pCDNA3-HA-URI.

(Z) Quantification of PARP activity and WB liver analysis of the NAMPT levels in 3-week-

old hURI-tetOFFhep mice. Error bars represent ± SEM.

Data represented as mean ± SEM or ± SD, as indicated. (p ≤ 0.05 = *, p ≤ 0.01 = ** and p ≤

0.001 = ***).

22

Table S1, related to Figure 4. iTRAQ Data, provided as an Excel File.

Table S2, related to Figure 4. Dehydrogenases Significantly Downregulated in iTRAQ

Datasets

Protein symbol Fold change* Protein symbol Fold change*Ldhb 0.5 Hsd3b5 0.05

Hasd17b6 0.53 Hsd17b6 0.35Hsd17b11 0.64 Gldc 0.39

G6pd 0.7 Dhtkd1 0.42Aldh16a1 0.53Hsd3b2 0.54Aspdh 0.55

Hsd17b2 0.57Gcdh 0.57Hpgd 0.57Rdh7 0.58Idh2 0.59Ldhd 0.6

Ndufv3 0.61Aldhli1 0.61Hibadh 0.62

Hsdll1b1 0.62Sardh 0.62

Ivd 0.66Prodh 0.66Pdk2 0.66

Acadsb 0.66Aldh6a1 0.66

Dhdh 0.67Ndufa12 0.68L2hgdh 0.68Sdhc 0.7Idh1 0.71

*Fold change represents the protein expression in mutant samples relative to the controls in iTRAQ analysis

23

24

25

Figure S5, related to Figure 5. Restoring NAD+ Pools Protects from DNA Damage and

Prevents Tumor Formation

(A) Liver NAD+ levels in mutant hURI-tetOFFhep mice fed either chow or NR diet. (n ≥ 5).

(B) Liver to body weight ratio in mutant hURI-tetOFFhep mice fed either chow or NR diet. (n

≥ 5).

(C) Representative images of Sirius Red stained liver sections from hURI-tetOFFhep mice fed

with chow or NR diets. (n = 6).

(D) Quantification of the Sirius red positive area from (C).

(E) Representative images of whole livers and H&E stained liver sections from 30-week-old

mutant homozygous hURI-tetOFFhep mice fed either chow or NR diet. Dotted black circles

depict tumor area and red arrows depict tumor present in the homozygous mutant hURI-

tetOFFhep mice under chow diet. (n = 9).

(F) Representative images of whole livers from 60-week-old homozygous mutant hURI-

tetOFFhep mice fed either chow or NR diet from 12 weeks of age.

(G) Relative tumor size to mouse livers described in (F).

(H) H&E and cleaved caspase 3 stained liver sections from 60-week-old homozygous mutant

hURI-tetOFFhep mice described in (F). Red arrows depict the apoptotic bodies and the yellow

arrow depicts the inflammatory clusters.

(I) Representative images of H&E and H2AX stained pancreas sections from 4-week-old

Ela-1-myc and K-RasG12V mice. (n ≥ 5).

(J) Quantification of H2AX positive nuclei in Ela-1-myc and K-RasG12V mice.

(K) WB analysis of Ela-1-myc and K-RasG12V mice pancreas.

26

(L) H&E representative images of normal, acinar-to-ductal metaplasia (ADM), acinar cell

carcinomas and ductal adenocarcinomas from 15-week-old Ela-1-myc mice.

(M) Percentage of normal pancreas or pancreas with pancreatitis and ADM in 7-week-old

mice under chow or NR diet for 4 weeks, starting from 3 weeks of age. (n = 3)

(N) Percentage of normal pancreas or pancreas with pancreatitis, ADM and carcinomas in 15-

week-old mice under chow or NR diets for 12 weeks, starting from 3 weeks of age (n ≥ 3).

(O) Representative images of pancreas from 15-week-old mutants Ela-1-myc supplemented

with either chow or NR diets for 12 weeks, starting from 3 weeks of age.

(P) Pancreatic NAD+ levels in 15-week-old Ela-1-myc mice fed either chow or NR diet for 12

weeks. (n ≥ 7).

Data represented as mean ± SEM. (p ≤ 0.05 = * and p ≤ 0.01 = **)

27

28

Figure S6, related to Figure 6. URI Regulates Kynurenine Metabolism by Modulating

AhR and ER Activity

(A and B) GSEA analysis between liver gene arrays of Ahr-/- (A) and Esr-/- (B) mice and liver

RNA sequencing data from 8-week-old hURI-tetOFFhep mice.

(C and D) ChIP assays showing association of AhR (C) and ER (D) with promoters of TDO2

and AFMID in SNU-449 cells. Semi-quantitative PCR of the intragenic region of

chromosome 3 was used for normalization.

(E and F) WB analysis of 1- (E) and 8- (F) week-old hURI-tetOFFhep mice.

(G) Co-immunofluorescence of AhR or ER and hURI in 1-week-old hURI-tetOFFhep mice.

DAPI was used for nuclear staining. Dotted red circles represent colocalization between AhR

or ER and hURI while dotted yellow circles represent the decrease in colocalization between

AhR or ER and hURI.

(H and I) WB analysis of cytosolic (H) and nuclear (I) fractions from 3-week-old URI(+/+)hep

and URI(+hep mice treated or not with DEN. Immunoblots was performed with the indicated

antibodies.

29

30

31

Figure S7, related to Figure 7. URI Expression Is Enhanced in Human HCC, Is

Associated with Poor Survival and Correlates with De Novo NAD+ Synthesis Inhibition

(A) Characterization of URI mouse monoclonal antibody by WB and IHC. Endogenous URI

was knock-downed in HEK-293T and in HeLa cells using siRNA against URI mRNA. IHC

using URI antibody in HEK-293T cells previously depleted for URI.

(B) WB analysis for URI expression in paired peritumoral (PT) and tumoral (T) liver samples

derived from human patients.

(C) Quantification of (B). Denote increased URI levels in tumoral (T) compared to the paired

peritumoral (PT) samples. (p = 0.0086)

(D) Stratification of human HCC liver samples according to Ki67 levels (less aggressive

(Ki67 < 10%) and more aggressive (Ki67 > 10%)) in tumors with weak, moderate or high

URI expression.

(E) WB analysis of URI expression in Concanavalin A (ConA) treated C57BL/6 mouse livers

at different time points (0, 2, 4 and 8 hr).

(F) IHC analysis of URI expression in ConA treated C57BL/6 mouse liver tissues at different

time points (0, 2, 4 and 8 hr).

(G and H) WB analysis for URI expression in Huh-7 (G) and HepG2 (H) cells transiently

transfected with either pCDNA3-GFP or pCDNA3-HA-HBx.

(I) Scheme representing URI promoter cloned in frame with the luciferase reporter.

(J) Luminescence ratio of luciferase reporter and background renilla in Huh-7 cells that were

co-transfected with either the reporter plasmid and with pCDNA3 or with the reporter

plasmid and with different concentrations of pCDNA3-HA-HBx. Right panel represents the

relative URI luciferase reporter activity after transfecting 16 ng of pCDNA3-HA-HBx.

32

(K) WB analysis of paired peritumoral (PT) and tumoral (T) liver samples from human

patients.

(L) Pearson correlation analysis between URI and AFMID protein levels in peritumoral (PT)

and tumoral (T) human liver sample. Denote a significant inverse correlation between URI

and AFMID levels. (p = 0.0047)

(M) Quantification of liver NAD+ levels in the peritumoral (PT) and tumoral (T) paired

samples used in (B). (n ≥ 5).

(N) Pearson correlation analysis between URI protein and NAD+ levels in peritumoral (PT)

and tumoral (T) human liver samples. (p = 0.0136)

(O) Pearson correlation analysis of AFMID protein and NAD+ levels in peritumoral (PT) and

tumoral (T) human liver samples. (p = 0.0173)

(P) Linear regression analysis of URI1 and HAAO, KMO, TDO2 and QPRT expressions in a

human HCC microarray dataset, showing inverse correlation between URI1 and HAAO,

KMO, TDO2 and QPRT expression.

(Q and R) Kaplan Meier curve analysis of human HCC patients based on gene expression of

TDO2 (Q) and HAAO (R). “df” represents degrees of freedom and “Sig.” represents

significance.

Data represented as mean ± SEM.

33

SUPPLEMENTAL EXPERIMENTAL PROCEDURES

Generation and Handling of Mice

hURI knock-in mouse model (ColhURI) has been generated by flippase-mediated targeting

and recombination of inducible human URI (hURI) cDNA tagged with the Flag peptide in the

3’ untranslated region of the homing locus of the collagen type I, alpha 1 locus (Col1a1)

Cola1 gene in KH2 embryonic stem cells (ESCs), having a tetracycline operator (Tet-op). In

order to express hURI specifically in hepatocytes, ColhURI mouse has been crossed with a

line containing the tetracycline-dependent transactivator (tTA) under the control of the liver

activated protein (LAP) promoter to generate LAP-tTA/hURItetOFF mouse, named hURI-

tetOFFhep mouse (Kistner et al., 1996). Hepatocyte-specific ectopic hURI expression can be

switched off by administration of doxycycline. hURI is expressed since conception and mice

were off doxycycline, unless otherwise stated. The URI conditional knockout mouse with uri

floxed allele was generated by homologous recombination in embryonic stem (ES) cells (URI

lox). In this allele, the exon 4 of uri was flanked by two LoxP sites. In addition, a neomycin

resistance gene (Neo) flanked by two FRT sites was inserted before the second LoxP site for

drug selection and was removed by expressing Flp recombinase. Expression of Cre

recombinase deletes the targeted exon 4 and generates a delta allele (URI, leading of

inactivation of the URI gene. The germ line transmission of the founder mouse generated

through an intercross between the chimeras and the Flp-deleter strain in order to remove the

neomycin cassette was checked by Southern blot (5’ and 3’ arms were checked) and PCR

analysis. Deletion of one allele by Cre recombinase in heterozygous ES cells was verified by

reduced URI expression via WB analysis. Homozygous (lox/lox) and heterozygous (+/lox)

mice carrying the URI lox allele were viable, fertile, and exhibited no overt abnormalities.

Crossing URI (+/lox) heterozygous mice with a Mox-Cre line, to deplete URI ubiquitously,

generated viable URI (+/) mice (homozygotes obtained from intercrosses were non-

34

viable). p53 inactivation in the hURI-tetOFFhep mouse was performed by crossing hURI-

tetOFFhep mice with p53ERTAM mice, in which p53 activation requires ectopic 4-

hydroxytamoxifen provision (Christophorou et al., 2005).

Food (Harlan Laboratories) and water were provided ad libitum. Same experiments have

been performed by sacrifying mice at the same time and same lobes were used for the same

experiment.

Mouse Treatments

Food containing either doxycycline for LAP-tTA inhibition or tamoxifen for CreERT2-

mediated recombination was prepared by Harlan Laboratories and mice were dietary fed.

Mice were supplied with tamoxifen-containing diet for 2 weeks and then transferred to chow

diet. For DEN-treated mice, tamoxifen was previously administered at 1 week of age by

intraperitoneally injection (1 mg per dose; 1 dose/day for 2 days) and DEN injection was

applied at 2 weeks of age.

Rapamycin (Sirolimus) was purchased from TOKU-E biosciences (R001) and stored at 4 oC.

Rapamycin diet was prepared by Harlan laboratories to a final concentration of 14 mg/kg of

food. Mice were fed 2.24 mg/kg of body weight per day (Menon et al., 2012).

Ultrasound Imaging

Animals were anesthetized with Isofluorane (Isovet, Braun Vetcare) (4 % during anesthetic

induction and 2 % as maintenance level) and livers were scanned (Vevo 770, 40 Mhz

frequency) for structural alterations, including echogenicity variations, by using the probe

RMV707b (Visualsonics, Canada). The frame rate used is in the range of 60 Hz and 11X11

mm FOV (field of view). The biggest and lowest diameters were chosen for all liver

structural anomalies.

35

Antibodies

Antibodies recognizing hURI and mURI have been described previously (Djouder et al.,

2007). Ki67 (1:1) and Cyclin D1 (1:1000) were purchased from Dako. MAD2 (1:250),

PCNA (1:1000), ACTA2 (1:1000) and p65 (1:1000) were purchased from Santa Cruz.

Phospho-S6K1 (Thr389) (1:1000), BAX (1:1000), phospho-p53 (Ser18) (1:500), p53 (1:500),

acetyl-p53 (Lys379) (1:500), cleaved caspase3 (1:500), phospho-Chk1 (Ser345) (1:500),

Lamin (1:1000) and HSP90 (1:1000) were from Cell Signaling Technology. Phospho-H2AX

(Ser139) (1:1000) was purchased from Merck Millipore. Phospho-Chk2 (Thr68) (1:500),

Chk1 (1:500), Chk2 (1:500), Cyp2E1 (1:2000) and Cyp1A1 (1:2000) were purchased from

Upstate (Millipore). S6K1 (1:1000), c-Myc (1:200) and GAPDH (1:2000) were from Abcam.

Flag antibody (1:1000) was from Agilent. p21 (1:1000) was from BD-Pharmingen. Vinculin

(1:1000) was purchased from Sigma. TDO2 (1:1000), AFMID (1:500), GCDH (1:500), CPS1

(1:1000), GNMT1 (1:1000) and NAMPT (1:1000) were purchased from Proteintech. AhR

(1:1000) was from Enzo life sciences and ER (1:1000) was from Bethyl laboratories. AFP

(1:500) was from R&D systems. HA (1:1000) was obtained from Covance.

Cell Culture, siRNA, shRNA, Plasmids and Xenografts Studies

HEK-293T, HeLa, HepG2, Huh-7, SNU-398 and SNU-449 were grown in complete DMEM

medium (Lifetechnologies) supplemented with 10 % fetal calf serum and 100 units/ml

penicillin and 0.1 mg/ml streptomycin purchased from Gibco. Non tumorigenic alpha mouse

liver (AML)-12 cells were grown in DMEM/F12 supplemented with 10 % fetal calf serum,

100 units/ml penicillin and 0.1 mg/ml streptomycin, 0.005 mg/ml insulin, 0.005 mg/ml

transferrin, 5 ng/ml selenium, and 40 ng/ml dexamethasone purchased from Gibco (Wu et

al., 1994). Plasmid transfections were performed using FUGENE 6 following the

manufacturer instructions. siRNA transfections were performed using Lipofectamin

36

RNAiMAX following the manufacturer’s instructions. pCDNA3-HA-HBx purchased from

Addgene was used (http://www.addgene.org/24930/). Cell number, size and diameter were

measured in triplicates at the indicated times using CASY® cell counter.

Knockdown experiments were performed using ON-TARGET plus SMART pool siRNA

targeting human AhR, ER, TDO2, AFMID and URI as well as control siRNA purchased

from Dharmacon.. Three different murine TDO2 and AFMID MISSION shRNA Lentiviral

Transduction Particles (pLKO-puro-shRNA system) were purchased from Sigma Aldrich.

AML-12 cells were infected according to the manufacturer instructions. Separate pools of

puromycin-selected AML-12 cells depleted for TDO2 and AFMID were generated (levels of

depletion were checked by WB) and finally mixed in order to have one pool of three AML-12

cell pools for TDO2 downregulation (AML-12-shTdo2) and one pool of three AML-12 cells

for AFMID downregulation (AML-12-shAfmid). AML-12 cells harvested from culture were

resuspended in PBS at a concentration of 5.106 cells/0.1 ml and were injected subcutaneously

into the left and right flanks of 5-week-old female athymic nude mice purchased from Harlan

laboratories. Tumor volume was monitored over 5 months. The following hairpins shRNA

sequences were used: sh1Tdo2

(CCGGCCAAAGATGAATCCGATCATTCTCGAGAATGATCGGATTCATCTTTGGTTT

TTG); sh2Tdo2

(CCGGGCTGGAAAGAACACCTGGTTTCTCGAGAAACCAGGTGTTCTTTCCAGCTT

TTTG); sh3Tdo2

(CCGGGCGCAAGAACTTCAGAGTGAACTCGAGTTCACTCTGAAGTTCTTGCGCTT

TTTG). Sh1Afmid

(CCGGGTTGGGAACTTCGTGCAGATACTCGAGTATCTGCACGAAGTTCCCAACTT

TTTG); sh2Afmid

(CCGGCCCTGGAAGATGCTCAGAGAACTCGAGTTCTCTGAGCATCTTCCAGGGTT

37

TTTG); sh3Afmid

(CCGGCCGGCAGTCCAAGGAGTTCTACTCGAGTAGAACTCCTTGGACTGCCGGTT

TTTG).

Cytoplasmic Fractionation

For cytoplasmic extracts, livers were homogenized with the cytoplasmic buffer (10 mM

Hepes pH 7.4, 3 mM MgCl2, 40 mM KCl, 5 % Glycerol, 0.5% NP-40, 2 mM DTT)

supplemented with protease inhibitors. Pellets were incubated on ice 5 min and centrifuged

for 5min at 1500 rpm at 4°C. The supernatant fractions correspond to the cytoplasmic

fractions. Protein concentrations of the fractions were determined using Bradford assay

(BioRad) and equal amounts of each fraction were subjected for SDS-PAGE and blotted to

nitrocellulose membranes for Western blotting with the indicated proteins.

Immunoblotting and Immunoprecipitation

For immunoblotting, liquid N2 snap frozen liver tissues were used to prepare lysates. 50 to

100 mg of tissues were lysed using RIPA lysis buffer containing 10 mM Tris pH 7.5, 100

mM NaCl, 1 mM EDTA, 1 mM EGTA, 1 mM NaF, 20 mM Na4P2O7, 2 mM Na3VO4, 0.1%

SDS, 0.5 % sodium deoxycholate, 1 % Triton-X 100, 10 % glycerol and supplemented with

10 mg/ml proteases inhibitor aprotinin and 1 mM PMSF followed by homogenization using

Precellys 24 Bead Mill homogenizer (Bertin Technologies) (15 x 2 s, power set to 5500 w)

and then clarified by centrifugation at 4 ºC and 10.000 g for 10 min. Protein concentration

was measured by using Bio-Rad Bradford reagent (Bio-Rad) and bovin serum albumin (BSA)

as standard protein. 1 mg/ml concentrated lysates were made by boiling the appropriate

amount of protein lysates with 2X laemmli buffer (4 % SDS, 20 % glycerol, 10 % -

mercaptoethanol, 0.004 % bromophenol blue in 0.2 M Tris-HCL of pH 7) at 70 ºC, for 10

38

min. 10-30 g of protein lysates were subjected into SDS-PAGE gels, and transferred to

nitrocellulose membranes. The membranes were blocked with 5 % nonfat milk in Tris-

buffered saline containing 1 % Tween 20 for 1h at room temperature. Blots were immune-

stained with indicated antibodies. Immunoblots were processed by ECL (Amersham)

according to the manufacturer's instructions.

For immunoprecipitation, samples and assays were performed as described previously

(Djouder et al., 2007).

Chromatin Immunoprecipitation (ChIP)

For Chromatin Immunoprecipitation (ChIP) assay, 6.107 SNU-449 cells were cross-linked for

10 min at RT in 1 % formaldehyde. Cross-linking reactions were stopped by adding 1.25 M

glycine to a final concentration of 125 mM. Cells were centrifuged for 10 min at 4 0C and

then washed in cold PBS. Cells were lysed with 1 ml of lysis buffer-1 (50 mM HEPES-KOH,

pH 7.5, 140 mM NaCl, 1 mM EDTA, 10 % glycerol, 0.5 % NP-40, 0.25 % Triton X-100, 1 X

protease inhibitor) followed by centrifugation at 4 0C and resuspension of the pellet in lysis

buffer-2 (10 mM Tris-HCl, pH 8.0, 200 mM NaCl, 1 mM EDTA, 0.5 mM EGTA, 1 X

protease inhibitors. Finally the pellets were resuspended in 1 mL of lysis buffer-3 (10 mM

Tris-HCl, pH8.0, 100 mM NaCl, 1 mM EDTA, 0.5 mM N-lauroylsarcosine, 1 X protease

inhibitors), and 100 l of 10 % Triton X-100 were added and sonicated for 20 min in a

Covaris sonicator. The soluble fraction was quantified with Bradford, and 400 g was used to

immunoprecipitate the transcription factors and IgG’s used as a control. Chromatin and

antibody mixtures were incubated overnight at 4 0C in total volume of 500 l.

Immunoprecipitated mixture was washed with a low salt buffer (20 mM Tris-HCl, pH 8.0,

150 mM Nacl, 2 mM EDTA, 0.1 % SDS and 1 % Triton X-100), followed by high salt buffer

(20 mM Tris-HCl, pH 8.0, 500 mM NaCl, 2 mM EDTA, 0.1% SDS and 1 % Triton X-100)

39

and finally with LiCl wash buffer (10 mM Tris-HCl, pH 8.0, 250 mM LiCl, 1 mM EDTA, 1

% Na-Deoxycholate and 1 % NP-40). Samples were decross-linked by resuspending the

beads in 210 l of elution buffer (50 mM Tris-HCl, pH 8.0, 10 mM EDTA and 1 % SDS),

incubating at 65 0C for 15 min and treating with 2.5 ul (from 32 mg/mL) of RNAse A for 2

hr. Next, 4 l of (20 mg/ml of stock) of Proteinase K was added and incubated at 55 0C for 2

hr. DNA was extracted by phenol/chloroform/isoamylalcohol mixtures, washed with 80 %

EtOH and resuspended in 200 l of TE buffer. This was followed by quantitative real-time

PCR. The specific primers used are listed below.

P1-TDO2-F (AGGAGACTACACTACTGGGAGAAG); P1-TDO2-R-

(CTTGTTTGCTAGCAGGGCGT); P4-TDO2-F (TTGTCTATGGGCAGGGTGAT); P4-

TDO2-R (GCCTGGTGCGAAAAACGAG); P5-TDO2-F

(CTCCTGTAAGGACCTACCTAGC); P5-TDO2-R (ACCAAACTCCTCTGGCGTATC);

P6-AFMID-F (CTAGAGGCTAGCAGCAGTGTG); P6-AFMID-

R (CTTCCAGATCGCCCAGTACA); P7-AFMID-F (CAGGCACAGTTGGATCTCT); P7-

AFMID-R (AGAGGACAGCCTCGGGTTAT); P8-AFMID-F

(GCGGGTGTATGGTTACCTGT); P8-AFMID-R (ATGCCCGAGTGTTGTGTTGT); P9-

AFMID-F (GCCTCAGTAAATGGGTTGAGAGA); P9-AFMID-R

(CTGCACGCTACTTGCTTGTC); P10-AFMID-F (CACTGAGGCCAAGGCTAGAG);

P10-AFMID-R (AGGAACACCATGGACGCATT); P11-AFMID-F

(TAAAAGCAGCCCAATGCGGG); P11-AFMID-R (GTCGACCTCTGACCTCCACAT)

Chr3-intron-F (CTTGGCCCTTCCTCTCCTAA); Chr3-intron-R

(TGCAGTGGTCAGAAGATGTGT)

Immunohistochemistry and Immunofluorescence

40

Freshly harvested murine livers were fixed immediately in 10 % buffered formalin solution

overnight and embedded in paraffin. Sections of 3 m were deparafinized, rehydrated and

antigen retrieved by using 1 M sodium citrate buffer (pH 6.5). After blocking endogenous

peroxidase using 3 % H2O2 (in methanol) for 5 min, sections were then blocked with 1:200

goat serum 5 % BSA/PBST for 1h at RT. Furthermore, sections were incubated with primary

antibodies overnight at 4 ºC, after which Vectastain ABC kit (Vector Laboratories, Inc) was

used, following manufacturer’s instructions. Sections were then incubated with 33-

diaminobenzidinetetracloride (DBA) and counterstained with hematoxylin. In the case of

immunofluorescence, sections were incubated with PE-conjugated secondary antibodies

(1:500) for 1 hr at RT, followed by DAPI staining. Quantification was performed either by

counting the number of positive cells in a minimum of six 10X or 20X microscopic fields, or

the percentage of positive area using Color Deconvolution plug-in in Image J.

Reactive Oxygen Species (ROS), NAD+ and PARP Activity Determination

Reactive oxygen species (ROS) using the commercial kit OxiSelect™ Intracellular ROS

(Green Fluorescence) purchased from Cell Biolabs.

NAD+ levels were determined using a commercial kit (Enzychrom, BioAssays Systems, CA),

as previously described (Canto et al., 2012). NAD+/NADH assay kit is based on a lactate

dehydrogenase cycling reaction, in which the formed NADH reduces a formazan (MTT)

reagent.

PARP activity was measured from the liver lysates using TRIVIGEN colorimetric assay

(4677-096-K). In brief, fresh livers were lysed in PARP buffer, and 10 g of the protein from

the supernatant was used. 96 well plate coated with histones were rehydrated and the sample

mixtures were prepared using 10 l of protein lysate, 15 l of water and 25 l of PARP

cocktail (biotinylated NAD+). Additionally, protein standards were also added to the wells by

41

diluting 25 l standards and 25 l PARP cocktail. After washing with PBS and 0.1 % Triton

X-100 followed by 2 washes of PBS, diluted strep-HRP was added and incubated for 60 min.

Finally, color was developed by adding TACS-Sapphire colorimetric substrate and reading

the absorbance at 450 nm.

Genotyping and qRT-PCR

For genotyping tail DNA was extracted by overnight incubation of tails with the following

buffer (1 % SDS, 0.1 M NaCl, 0.1 M EDTA and 0.05 M Tris (pH 8)). Extracted DNA was

precipitated using ice-cold isopropanol which was further washed with 70% ethanol. The

DNA pellet was further dried and resuspended in 400 l of water. 1 l of DNA was used for

genotyping. For the ColhURI locus the following primers were used: P1-F:

GCACAGCATTGCGGACATGC; P2-R: CCCTCCATGTGTGACCAAGG; P3-R:

GCAGAAGCGCGGCCGTCTGG. For the LaptTA transactivator locus the following primers

were used: P1-F: TCTGAGCATGGCCTCTAA; P2-R: GCTGGAGTAAATTTCACAGTG;

P3-R: TCTCACTCGGAAGGACAT.

For qRT-PCR, total RNA was extracted from 20-50 mg snap frozen liver tissue, using Trizol

(Sigma). First strand cDNA synthesis was performed by using Ready-to-go first strand beads

(GE Healthcare). qRT-PCR was performed using an ABI PRISM 7700 (Life Technologies),

and GoTaq Real-Time qPCR mix (Promega). Fold changes have been determined by using 2-

CT (Livak and Schmittgen, 2001) and normalized to -actin. Finally, the fold changes were

obtained by converting the logarithmic scale to an exponential scale (2^CT). The

following primers were used Actin-F: CACAGCTGAGAGGGAAATCG. Actin-R:

AGTTTCATGGATGCCACAGG. Timp1-F: AGGTGGTCTCGTTGATTCGT. Timp1-R

GTAAGGCCTGTAGCTGTGCCActa2-F: CAATGGCTCTGGGCTCTGTA. Acta2-R:

42

TCATCCCCCACATAGCTGTC. Col1a1-F: AGGGCCAGGGGGTCCAGCATTTC.

Col1a1-R: GGTGCCCCCGGTCTTCAG. Bax-F: GTGAGCGGCTGCTTGTCT. Bax-R:

GGTCCCGAAGTAGGAGAGGA. Xiap-F: GAGGCAGGAAGCTAACGTTTT. Xiap-R:

GCAGCTCGTCTGAGATCAATG. Bbc3-F: CTGTATCCTGCAGCCTTTGC Bbc3-R:

ACGGGCGACTCTAAGTGCT.

RNA Extraction and Sequencing

For RNA extraction, 20 livers (5 per group) were individually extracted. RNA Integrity

Numbers were 8.3 on average (range 7.5 to 9.2) when assayed by Lab-chip technology on a

2100 Bioanalyzer (Agilent). Sequencing libraries were prepared as in "TruSeq RNA Sample

Preparation Guide" (Part # 15008136 Rev. A) with final PCR amplification limited to 10

cycles. The resulting purified cDNA libraries were sequenced on a Genome Analyzer IIx,

following manufacturer's protocols. Samples were analyzed in different paired end

sequencing runs. In order to balance read length and adjust base calling quality within

conditions, reads from 8-week samples were trimmed to 50 bases. Likewise, reads from 1-

week samples were adjusted to 78 bases. Reads were aligned to the mouse genome

(NCBI37/mm9) with Tophat (Trapnell et al., 2012) version 1.3.1 (including Bowtie 0.12.7

(Langmead et al., 2009) and Samtools version 0.1.16 (Li et al., 2009) ) allowing a maximum

of 2 mismatches and 5 multihits for each read. The RNA fragment length was 230bp on

average. Transcripts were assembled and their abundances estimated with Cufflinks version

1.0.3 (Trapnell et al., 2012). After that, we performed a differential expression analysis with

Cuffdiff 1.0.3.

iTRAQ

43

For protein extraction, 20 mouse livers (5 per group) were individually processed. Lysis

solution (4 % SDS, 10 mM DTT, 100 mM HEPES, pH 7.6) including 0.1 % Benzonase

Nuclease (Novagen) plus a cocktail of protease inhibitors was added to small pieces of tissue

(1 mg tissue: 20 µL buffer). Samples were homogenized in a Precellys 24 Bead Mill

Homogenizer (Bertin Technologies) (15 x 2 s, power set to 5500 w) and then clarified by

centrifugation at 4 ºC and 16000 rpm for 15 min. Recovered supernatants were cleaned-up by

methanol-chloroform extraction (Kline et al., 2008) and pellets were dissolved in 7 M urea 2

M tiourea 0.1 M TEAB by sonication and vortexing. The protein concentration of the

samples was determined according to the Bradford assay using BSA as standard (Protein

Assay Kit, Bio-Rad). Samples for each group were prepared by pooling an equal amount of

protein from each individual.

Protein Digestion and Labelling with iTRAQ Reagents

Samples were digested using the filter aided sample preparation (FASP) method (Wisniewski

et al., 2009) with some modifications. Briefly, 100 ug of each sample dissolved in 7M urea 2

M tiourea was loaded on the filter, reduced with 10 mM DTT for 1 h at 37 ºC and alkylated

using 50 mM iodoacetamide for 20 min in the dark. The excess of reduction and alkylation

reagents was washed. The proteins were digested overnight at RT using endoproteinase Lys-

C from Acromobacter lyticus M497-1 (Wako Pure Chemical Industries) with 1:50 enzyme to

protein ratio. Finally, trypsin (Promega) was added and samples were subjected to a second

digestion for 6 h. Each tryptic digest was labeled according to the manufacturer's instructions

(ABSciex) with one isobaric amine-reactive tag as follows: liver peptides from 1 week

control mice were tagged with Tag114. Tag115 was used for 8 week control mice. Tag116

was used for 8 week mutant mice and Tag117 for 1 week mutant mice. After 1 h incubation,

labeled samples were pooled, and evaporated to dryness in a vacuum centrifuge. The iTRAQ

44

sample was cleaned up using a Sep-Pak C18 cartridge for SPE (Waters Corp) (Ernoult et al.,

2008). Eluted peptides were vacuum-dried and reconstituted in OFFGEL solution (5 %

glycerol, 1 % ampholytes pH 3-10) prior to electrofocusing.

OFFGEL Fractionation

For pI-based peptide separation, we used the 3100 OFFGEL Fractionator system (Agilent

Technologies) with a 24-well set-up. The IPG gel strips of 24 cm-long (GE Healthcare) with

a 3–10 linear pH range were rehydrated for 15 min with the Peptide IPG Strip Rehydratation

Solution according to the protocol of the manufacturer. Subsequently, 150 μL of sample was

loaded in each well. Electrofocusing of the peptides was performed at 20 °C and 50 μA until

the 50 kVh level was reached. After focusing, the 24 peptide fractions were withdrawn and

the wells rinsed with 100 μL of a solution of 0.1%TFA. Rinsing solutions were pooled with

their corresponding peptide fraction. All fractions were evaporated by centrifugation under

vacuum. Solid phase extraction and salt removal was performed with home–made columns

based on Stage Tips with C8 Empore Disks (3M) (Rappsilber et al., 2003) filled with Poros

Oligo R3 resin (Life Technologies). Eluates were evaporated to dryness and maintained at 4

°C. Just prior nano-LC, the fractions were resuspended in 0.1 % formic acid (FA).

Peptide Analysis by NanoLC-MS/MS

Digested samples were separated by on-line reversed-phase nanoscale capillary LC and

analyzed by electrospray MS/MS. The experiments were performed on an Eksigent nano LC

system (Eksigent technologies) coupled to an LTQ Orbitrap Velos mass spectrometer

(Thermo Scientific) equipped with a nanoelectrospray ion source (Proxeon Biosystems).

Peptides were loaded from a cooled nanoLC AS-2 autosampler (Eksigent). In order to pre-

concentrate and desalt the samples before switching the pre-column in line with the

45

separation column, 5 μL from each sample was loaded onto a reversed-phase ReproSil Pur

C18-Aq 5 µm 0.3 x 10 mm trapping cartridge (SGE Analytical, Victoria, Australia), and

washed for 10 min at 2.5 μL/min with loading buffer (0.1% FA). The peptides were eluted

from a RP ReproSil Pur C18-AQ 3 µm 200 x 0.075 mm (Dr. Maisch GmbH) by application

of a binary gradient consisting of 2 % ACN in 0.1 % FA (buffer A) and 100 % ACN in 0.1 %

FA (buffer B), with a flow rate of 300 nL/min. Peptides were separated using the following

gradient: 0-5 min 2 % B, 5-150 min 60 % B and 150-165 min 98 % B. The column was

operated at a constant temperature of 30 ºC. The LTQ Orbitrap Velos was operated in

positive ionization mode. The MS survey scan was performed in the FT analyzer scanning a

window between 250 and 1750 m/z. The resolution was set to 60 000 FWHM at m/z 400. The

m/z values triggering MS/MS with a repeat count of 1 were put on an exclusion list for 60 s.

The minimum MS signal for triggering MS/MS was set to 1000 counts. In all cases, one

microscan was recorded. The lock mass option was enabled for both MS and MS/MS mode

and the polydimethylcyclosiloxane ions (PDMS, protonated (Si(CH3)2O))6; m/z 445.120025)

were used for internal recalibration of the mass spectra (Olsen et al., 2005). For the HCD, up

to the 15 most abundant isotope patterns with charge ≥ 2 from the survey scan were selected

with an isolation window of 2 m/z fragmented in the C-trap collision cell. Normalized

collision energy was set to 42 %, the Q value to 0.25 and an activation time to 0.10 ms.

Waveform filter was activated. The resulting fragments were detected in the Orbitrap system

with a resolution of 7500 FWHM at m/z 400. The maximum ion injection times for the

survey scan and the MS/MS scans were 500 ms and 250 ms respectively and the ion target

values were set to 1E6 and 5E4, respectively for each scan mode. The 24-fractions were run

in duplicates.

iTRAQ Data Analysis

46

The raw files were processed using the Proteome Discoverer 1.3.0.339 software suite

(Thermo Scientific). The fragmentation spectra were searched against the UniProtKB/Swiss-

Prot mouse database (released date: October 19, 2011; 16407 entries) using MASCOT

(Perkins et al., 1999) as the search engine (v 2.2) with the precursor and fragment mass

tolerances set to 10 ppm and 0.075 Da, respectively, and with up to two missed cleavages.

Lysine and peptide N-termini labeling with iTRAQ-4plex reagent as well as

carbamidomethylation of cysteine were considered as fixed modifications, while oxidation of

methionine was chosen as variable modification for database searching. Peptides

identification was filtered at 1% false discovery rate (FDR) and thus not dependent on the

peptide score. Only peptides with high confidence were considered. The results were then

exported into Excel for manual data interpretation. Although relative quantification and some

statistical data were provided by the Proteome Discoverer software, an additional 1.3-fold

change cut-off for all iTRAQ ratios (ratio < 0.77 or > 1.3) was selected to classify proteins as

up- or down-regulated (Chen et al., 2007; Gan et al., 2007; Ho et al., 2009). Proteins with

iTRAQ ratios below the low range (0.77) were considered to be under-expressed, while those

above the high range (1.3) were considered overexpressed. Only proteins identified with 2 or

more unique peptides and at least 3 iTRAQ peptide counts were considered. Gene Ontology

analyses of the identified proteins were performed.

SUPPLEMENTAL REFERENCES

Canto, C., Houtkooper, R.H., Pirinen, E., Youn, D.Y., Oosterveer, M.H., Cen, Y., Fernandez-

Marcos, P.J., Yamamoto, H., Andreux, P.A., Cettour-Rose, P., et al. (2012). The NAD(+)

precursor nicotinamide riboside enhances oxidative metabolism and protects against high-fat

diet-induced obesity. Cell Metab 15, 838-847.

47

Chen, Y., Choong, L.Y., Lin, Q., Philp, R., Wong, C.H., Ang, B.K., Tan, Y.L., Loh, M.C.,

Hew, C.L., Shah, N., et al. (2007). Differential expression of novel tyrosine kinase substrates

during breast cancer development. Mol Cell Proteomics 6, 2072-2087.

Christophorou, M.A., Martin-Zanca, D., Soucek, L., Lawlor, E.R., Brown-Swigart, L.,

Verschuren, E.W., and Evan, G.I. (2005). Temporal dissection of p53 function in vitro and in

vivo. Nat Genet 37, 718-726.

Ernoult, E., Gamelin, E., and Guette, C. (2008). Improved proteome coverage by using

iTRAQ labelling and peptide OFFGEL fractionation. Proteome Sci 6, 27.

Gan, C.S., Chong, P.K., Pham, T.K., and Wright, P.C. (2007). Technical, experimental, and

biological variations in isobaric tags for relative and absolute quantitation (iTRAQ). J

Proteome Res 6, 821-827.

Ho, J., Kong, J.W., Choong, L.Y., Loh, M.C., Toy, W., Chong, P.K., Wong, C.H., Wong,

C.Y., Shah, N., and Lim, Y.P. (2009). Novel breast cancer metastasis-associated proteins. J

Proteome Res 8, 583-594.

Kistner, A., Gossen, M., Zimmermann, F., Jerecic, J., Ullmer, C., Lubbert, H., and Bujard, H.

(1996). Doxycycline-mediated quantitative and tissue-specific control of gene expression in

transgenic mice. Proc Natl Acad Sci USA 93, 10933-10938.

Kline, K.G., Frewen, B., Bristow, M.R., Maccoss, M.J., and Wu, C.C. (2008). High quality

catalog of proteotypic peptides from human heart. J Proteome Res 7, 5055-5061.

Langmead, B., Trapnell, C., Pop, M., and Salzberg, S.L. (2009). Ultrafast and memory-

efficient alignment of short DNA sequences to the human genome. Genome Biol 10, R25.

Li, H., Handsaker, B., Wysoker, A., Fennell, T., Ruan, J., Homer, N., Marth, G., Abecasis,

G., and Durbin, R. (2009). The Sequence Alignment/Map format and SAMtools.

Bioinformatics 25, 2078-2079.

48

Livak, K.J., and Schmittgen, T.D. (2001). Analysis of relative gene expression data using

real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25, 402-408.

Olsen, J.V., de Godoy, L.M., Li, G., Macek, B., Mortensen, P., Pesch, R., Makarov, A.,

Lange, O., Horning, S., and Mann, M. (2005). Parts per million mass accuracy on an Orbitrap

mass spectrometer via lock mass injection into a C-trap. Mol Cell Proteomics 4, 2010-2021.

Perkins, D.N., Pappin, D.J., Creasy, D.M., and Cottrell, J.S. (1999). Probability-based protein

identification by searching sequence databases using mass spectrometry data. Electrophoresis

20, 3551-3567.

Rappsilber, J., Ishihama, Y., and Mann, M. (2003). Stop and go extraction tips for matrix-

assisted laser desorption/ionization, nanoelectrospray, and LC/MS sample pretreatment in

proteomics. Anal Chem 75, 663-670.

Trapnell, C., Roberts, A., Goff, L., Pertea, G., Kim, D., Kelley, D.R., Pimentel, H., Salzberg,

S.L., Rinn, J.L., and Pachter, L. (2012). Differential gene and transcript expression analysis

of RNA-seq experiments with TopHat and Cufflinks. Nat Protoc 7, 562-578.

Wisniewski, J.R., Zougman, A., Nagaraj, N., and Mann, M. (2009). Universal sample

preparation method for proteome analysis. Nat Methods 6, 359-362.

Wu, J.C., Merlino, G., and Fausto, N. (1994). Establishment and characterization of

differentiated, nontransformed hepatocyte cell lines derived from mice transgenic for

transforming growth factor alpha. Proc Natl Acad Sci USA 91, 674-678.

Inhibition of De Novo NAD+ Synthesis by Oncogenic URI Causes Liver Tumorigenesis through DNA Damage

Documents