Scholarly article on topic 'NLRX1 Acts as an Epithelial-Intrinsic Tumor Suppressor through the Modulation of TNF-Mediated Proliferation'

NLRX1 Acts as an Epithelial-Intrinsic Tumor Suppressor through the Modulation of TNF-Mediated Proliferation Academic research paper on "Biological sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Cell Reports
OECD Field of science
Keywords
{}

Abstract of research paper on Biological sciences, author of scientific article — Ivan Tattoli, Samuel A. Killackey, Elisabeth G. Foerster, Raphael Molinaro, Charles Maisonneuve, et al.

Summary The mitochondrial Nod-like receptor protein NLRX1 protects against colorectal tumorigenesis through mechanisms that remain unclear. Using mice with an intestinal epithelial cells (IEC)-specific deletion of Nlrx1, we find that NLRX1 provides an IEC-intrinsic protection against colitis-associated carcinogenesis in the colon. These Nlrx1 mutant mice have increased expression of Tnf, Egf, and Tgfb1, three factors essential for wound healing, as well as increased epithelial proliferation during the epithelial regeneration phase following injury triggered by dextran sodium sulfate. In primary intestinal organoids lacking Nlrx1, stimulation with TNF resulted in exacerbated proliferation and expression of the intestinal stem cell markers Olfm4 and Myb. This hyper-proliferation response was associated with increased activation of Akt and NF-κB pathways in response to TNF stimulation. Together, these results identify NLRX1 as a suppressor of colonic tumorigenesis that acts by controlling epithelial proliferation in the intestine during the regeneration phase following mucosal injury.

Academic research paper on topic "NLRX1 Acts as an Epithelial-Intrinsic Tumor Suppressor through the Modulation of TNF-Mediated Proliferation"

Cell Reports

Article

NLRX1 Acts as an Epithelial-Intrinsic Tumor Suppressor through the Modulation of TNF-Mediated Proliferation

Graphical Abstract

Authors

Ivan Tattoli, Samuel A. Killackey,

Elisabeth G. Foerster.....

Catherine J. Streutker, Dana J. Philpott, Stephen E. Girardin

Correspondence

stephen.girardin@utoronto.ca

In Brief

NLRX1 protects against colorectal tumorigenesis. Tattoli et al. now examine the mechanism underlying this effect and find that epithelial NLRX1 protects against colitis-associated colorectal cancer. In the epithelium, NLRX1 controls the expression of intestinal wound healing factors following injury and inhibits the proliferative action of TNF.

Highlights

• Epithelial NLRX1 protects against colorectal cancer

• NLRX1 dampens expression of intestinal wound healing factors

• NLRX1 inhibits proliferation during epithelial regeneration post-colitis

• In organoids, NLRX1 inhibits TNF-induced proliferation and signaling

Tattoli et al., 2016, Cell Reports 14, 2576-2586 ciossMark March 22, 2016 ©2016 The Authors

http://dx.d0i.0rg/l 0.1016/j.celrep.2016.02.065

CelPress

Cell Reports

Article

NLRX1 Acts as an Epithelial-Intrinsic Tumor Suppressor through the Modulation of TNF-Mediated Proliferation

Ivan Tattoli,12 Samuel A. Killackey,1 Elisabeth G. Foerster,2 Raphael Molinaro,1 Charles Maisonneuve,2 Muhammed A. Rahman,12 Shawn Winer,4 Daniel A. Winer,4 Catherine J. Streutker,13 Dana J. Philpott,2 and Stephen E. Girardin1*

department of Laboratory Medicine and Pathobiology 2Department of Immunology 3Saint Michael's Hospital

4Department of Pathology, Toronto General Hospital University of Toronto, M5S 1A8 Toronto, Canada 'Correspondence: stephen.girardin@utoronto.ca http://dx.doi.org/10.1016/j.celrep.2016.02.065

This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

SUMMARY

The mitochondrial Nod-like receptor protein NLRX1 protects against colorectal tumorigenesis through mechanisms that remain unclear. Using mice with an intestinal epithelial cells (IEC)-specific deletion of Nlrxl, we find that NLRX1 provides an lEC-in-trinsic protection against colitis-associated carcino-genesis in the colon. These Nlrxl mutant mice have increased expression of Tnf, Egf, and Tgfbl, three factors essential for wound healing, as well as increased epithelial proliferation during the epithelial regeneration phase following injury triggered by dextran sodium sulfate. In primary intestinal organoids lacking Nlrxl, stimulation with TNF resulted in exacerbated proliferation and expression of the intestinal stem cell markers Olfm4 and Myb. This hyper-proliferation response was associated with increased activation of Akt and NF-kB pathways in response to TNF stimulation. Together, these results identify NLRX1 as a suppressor of colonic tumorigenesis that acts by controlling epithelial proliferation in the intestine during the regeneration phase following mucosal injury.

INTRODUCTION

NLRX1 is a member of the Nod-like receptor (NLR) family of intracellular sensors of danger-associated molecular patterns (DAMPs) and microbial-associated molecular patterns (MAMPs) that function in innate immunity (Philpott et al., 2014). In contrast to other members of the NLR family, NLRX1 possesses an N-terminal addressing sequence that targets the protein to the mitochondrial matrix (Arnoult et al.,

2009). The exact function of NLRX1 has remained poorly understood. It was proposed to act as a negative regulator of antiviral immunity through interaction with the protein MAVS (Moore et al., 2008), but these findings were not reproduced by other groups (Jaworska et al., 2014; Rebsamen et al., 2011; Soares et al., 2013). Overexpression of NLRX1 was also shown to regulate the formation of reactive oxygen species (ROS) by mitochondria (Tattoli et al., 2008). Further support for this observation was obtained in mouse embryonic fibroblasts (MEFs) from Nlrx1~'~ mice, which displayed blunted tonic levels of ROS in resting condition as compared to wildtype (WT) MEFs (Soares et al., 2014).

NLRX1 was also shown to inhibit inflammatory pathways triggered by lipopolysaccharide (Allen et al., 2011; Xia et al., 2011). Moreover, Nlrxl mice display increased cerebral inflammation in a model of experimental autoimmune encephalopathy (Ei-tas et al., 2014), higher lung inflammation in a model of cigarette smoke-induced chronic obstructive pulmonary disease (Kang et al., 2015), and enhanced intestinal inflammation in a colitis model induced by dextran sodium sulfate (DSS) (Soares et al., 2014). However, the underlying mechanism through which NLRX1 controls inflammation in these various models remains unclear.

Recently, findings from several groups have identified a role for NLRX1 in the control of cell death in various cellular systems (Imbeault et al., 2014; Jaworska et al., 2014; Singh et al., 2015; Soares et al., 2014). However, it remains unclear if the effect of NLRX1 on ROS or cell death signaling is direct or whether it is a consequence of an overarching function of the protein in the mitochondria that remains to be identified. In support of the latter hypothesis, it must be noted that NLRX1 expression is significantly repressed by glucose starvation (Soares et al., 2014); in addition, NLRX1 silencing is sufficient to upregulate the basal activities of complex I and III of the mitochondrial respiratory chain (Singh et al., 2015), suggesting that NLRX1 may act to tie mitochondrial oxidative

phosphorylation (Oxphos) activity to the cellular availability of glucose. As such, NLRX1 could play a role as a tumor suppressor by inhibiting mitochondrial Oxphos-dependent generation of ATP in conditions of nutrient sufficiency. This hypothesis is supported by several lines of evidence. First, NLRX1 expression is strongly reduced by SV40- or Ras-dependent transformation in a genome-wide study (Danielsson et al., 2013), a result that we confirmed by analyzing Nlrxl expression by qPCRand western blotting in primary versus in SV40-transformed MEFs (Soares et al., 2014). Second, we recently demonstrated that Nlrxlmice display increased susceptibility to tumorigenesis in a model of colitis-associated colorectal cancer triggered by injection of azoxymethane (AOM) followed by three cycles of oral administration of DSS in drinking water (Soares et al., 2014). Finally, Singh et al. (2015) demonstrated that NLRX1 expression decreases tumor-igenic and migration properties of MCF-7 cells in vitro and the tumorigenic potential of colon cancer RKO cells in vivo using a xenograft model in nude mice.

While the critical role of NLRX1 in tumorigenesis is increasingly appreciated, the underlying mechanism remains unclear. In the AOM/DSS model of colorectal cancer, it is well established that inflammation is a major driver of tumor progression, as DSS administration strongly accelerates pathogenesis as compared to injection of AOM alone (Rosenberg et al., 2009). Because DSS-induced inflammation is increased in animals lacking Nlrxl (Soares et al., 2014), the enhanced tumor burden observed in Nlrxlin the AOM/DSS model as compared to WT mice could be caused by uncontrolled inflammation. Alternatively, NLRX1's proposed role in the control of apoptosis and mitochondrial respiration, as well as its transcriptional regulation by glucose levels and during cellular transformation, argue for a cell intrinsic effect of NLRX1 on tumorigenesis. In order to delineate the exact mechanism through which NLRX1 protects against colorectal cancer, we generated mice lacking Nlrxl only within the intestinal epithelium. Our results demonstrated an epithelial-intrinsic role of NLRX1 in the control of colitis-associated tumorigenesis. We present evidence that NLRX1 played its protective role by inhibiting TNF-induced intestinal epithelial cell (IEC) proliferation, which occurs during the healing phase post-colitis.

RESULTS

Generation and Analysis of Mice Lacking Nlrxl in the Intestinal Epithelium

C57Bl/6 mice with insertion of LoxP sites flanking Exon 3 of Nlrxl (Nlrx1fl/fl hereafter) have been described previously (Soares et al., 2013). We crossed these mice with transgenic C57Bl/6 mice expressing the Cre recombinase under the Villin gene promoter in order to excise Nlrxl specifically in lECs (Figure S1A). Nlrx1~'~, WT, Nlrx1fl/fl mice as well as mice with homozygous deletion of Nlrx1 in lECs (Nlrx1DIEC thereafter) were analyzed for NLRX1 expression in the spleen and the intestinal mucosa by western blotting (Figure 1A). As expected, only Nlrx1~'~ mice lost immunoreactivity for NLRX1 in the spleen. In intestinal mucosal samples of Nlrx1DIEC mice, NLRX1 expression was absent from the IEC fraction, but not

the remaining lamina propria fraction, indicating that Nlrx1 deletion was specific to the IEC population. We also noted that NLRX1 expression in the spleen and the lamina propria was similar between WT and Nlrx1fl/fl mice (Figure 1A), thus showing that LoxP sites insertion in the Nlrx1 locus had no overall effect on Nlrx1 expression.

Next, we aimed to determine if Nlrx1DIEC mice had intestinal defects in resting conditions. Importantly, for these experiments and the ones described in the rest of this study, Nlrx1fl/fl and Nlrx1DIEC mice were littermates from Nlrx1fl/flxNlrx1AIEC mice crosses. Histology of the colon (Figure 1B) revealed normal crypt architecture in Nlrx1DIEC mice and no signs of inflammation. IEC proliferation, as assessed by Ki-67 staining (Figures 1C and 1D) and analysis of Cyclin D1 and Cyclin B1 expression by quantitative (q)PCR, was also unaffected by Nlrx1 deletion (Figure 1D). Immunofluorescence using an antibody against chromogranin A revealed a normal number of enteroendocrine cells in the colonic epithelium of Nlrx1DIEC mice (Figures 1E and S1C) and expression of Chga, Gcg, and Cck, three markers of enteroendocrine cells, was also normal in the colon of Nlrx1DIEC mice (Figure 1F). Alcian Blue-Periodic acid-Schiff (AB-PAS) staining and immunofluorescence staining using an antibody against Muc2 also showed that goblet cell numbers were unaffected in Nlrx1DIEC mice (Figures 1G, S1B, and S1C), although analysis of Muc2 expression by qPCR revealed increased expression in Nlrx1DIEC mice (Figure 1H). Together, these results show that deletion of Nlrx1 in IECs does not significantly alter intestinal architecture.

Expression of Nlrxl in the Intestinal Epithelium Protects against Colitis-Associated Tumorigenesis

We previously reported that whole-body deletion of Nlrx1 exacerbated colitis-associated tumorigenesis in the AOM/DSS model (Soares et al., 2014), but whether this effect was epithelium-intrinsic or due to increased inflammation was not addressed. Nlrx1fl/fl and Nlrx1DIEC mice received an intraperitoneal injection of AOM (10 mg/kg) followed by three cycles of DSS (2.5% in drinking water). In both male and female cohorts of mice receiving AOM/DSS treatment, animals lost weight during DSS cycles, which they recovered during the intercycles (Figure 2A). A trend toward increased weight loss in Nlrx1fl/fl mice as compared to Nlrx1DIEC mice was observed. Colon shortening, an indication of intestinal inflammation, was observed in both Nlrx1fl/fl and Nlrx1DIEC mice, although it appeared to be slightly more pronounced in Nlrx1fl/fl mice (Figure 2B). Moreover, histological analysis of colon sections showed no significant difference overall between Nlrx1fl/fl and Nlrx1DIEC mice with regards to inflammation-associated parameters (Figure 2C). In particular, sub-mucosal edema, tissue damage, polymorphonuclear (PMN) infiltration and hyperplasia were similar between both groups of mice. However, analysis of tumor burden showed that AOM-DSS treated Nlrx1DIEC mice developed approximately twice as many tumors as Nlrx1fl/fl mice (Figure 2D) and histopathological scoring of these tumors revealed an increased number of adenomas in Nlrx1DIEC mice (Figures 2E and 2F).

The above results suggest that the impact of Nlrx1 deficiency on tumorigenesis is epithelial-intrinsic and not driven by changes

Intestinal mucosa

_ Isolated IEC without I EC

AAA KO WT fl/fl IEC KO WT fl/fl IEC KO WT fl/fl IEC

Spleen

Nlrxl m

Tubulin

Nlrxl '

Nlrxl1

fl/fl IEC-KO

Nlrxl™

fl/fl IEC-KO

Nlrxl1

fl/fl IEC-KO

O.O8-1 O.Ofr 0.040.020.00-

0.15-1

fl/fl IEC-KO

Nlrxl fl/fl

fl/fl IEC-KO

Nlrxl ^

0.00»!

§ •

•1 0.006-

& 0.004-

■3 u 0.002- • •

0.000-

fl/fl IEC-KO

in the inflammatory environment. In agreement with a cancer cell-intrinsic role of NLRX1, publically available data show that expression of NLRX1 is lower in tumors than the normal corresponding tissue in human cancers, including bladder, cervical, esophageal, colon, and skin cancers (Figure S2A). Moreover, analysis of two large data sets of colorectal cancer demonstrated that NLRX1 expression was significantly reduced in colon adenocarcinoma or colon mucinous adenocarcinoma as compared to normal colon (Figure S2B). Finally, analysis of 1,010 human cancer cell lines (from the Cancer Cell Line Encyclopedia) showed that the cell line with the lowest expression

Figure 1. Generation and Analysis of Mice Lacking Nlrxl in the Intestinal Epithelium

(A) Western blot analysis of NLRX1 and Tubulin (loading control) in spleen, isolated IEC fraction, and intestinal mucosa depleted of lECs from Nlrx1~/~

(B-H) Colonic tissue was collected from Nlrx1 and Nlrx1 DIEC (IEC-KO) mice and processed for H&E staining (B), IHC using anti-Ki-67 antibody (C), qPCR for Ki-67, Cyclin D1, and Cyclin B1 (D), Chga, Gcg, and Cck (F), or Muc2 (H), immunofluorescence using anti-chromogranin A (Chga) antibody (E), or AB-PAS staining (G). For qPCR data, each point represents the average value for one mouse obtained as a technical triplicate. The images presented in (B), (C), and (E) are representative of data analyzed on n R 4 mice. *p < 0.05 and other comparisons showed no statistical difference.

of NLRX1 was NCI-H716, which is derived from a colon adenocarcinoma (Figure S2C). Thus, available data from human cancer and cancer cell lines indicate that NLRX1 expression is reduced in multiple cancers and in particular in colon cancer.

Mutations in the Adenomatous polyposis coli (APC) gene are found in sporadic colon cancers in humans (Raskov et al., 2014) and result in uncontrolled Wnt signaling and proliferation of IECs. Mice harboring the Apcmin mutation develop spontaneous tumors in the small intestine and in the colon (Su et al., 1992). Nlrx1 -deficient mice were crossed with mice carrying the Apcmin mutation. Interestingly, Apcmin/+ Nlrx1~'~ mice had a significantly reduced lifespan and had to be euthanized at age 101.1 ± 8.38 days as compared to littermate Apcmin/+ Nlrx1+/+ mice that survived until day 132.5 ± 6.39 (Figures 3Aand 3B). Moreover, Apcmin/+ Nlrx1~'~ mice had more colonic tumors and those were larger than the ones observed in Apcmin/+ Nlrx1+/+ mice (Figures 3C and 3D). Thus, NLRX1 protects against tumorigenesis induced by dys-regulation of the Apc/Wnt signaling axis. Together, these results suggest that NLRX1 acts in an epithelial-intrinsic manner to suppress colorectal tumorigenesis.

Epithelial-Specific Deletion of Nlrxl Does Not Affect Peak Tissue Damage Induced by DSS, but Potentiates Expression of Late Tissue Healing Factors TNFa and TGFb

In order to better understand how IEC-specific deletion of Nlrx1 contributes to tumorigenesis susceptibility in the colitis-associated AOM/DSS model, DSS-induced colitis was analyzed in Nlrx1fl/fl and Nlrx1DIEC mice. Following treatment with 2.5%

fl/fl IEC-KO

(KO), WT mice, Nlrx1fl/fl (fl/fl), and Nlrx1 DIEC (DIEC)

Cohort #1 (Females)

Nlrxl ЯП

-•- Nlrxl № AOM/DSS -O- Nlrxl

-•- Nlrxl ж AOM/DSS

Cohort #2 (Males)

Nlrxl № 130п-в- Nlrxl Ш AOM/DSS -e- Nlrxl ^

Nlrxl ж AOM/DSS

CTR АОМ/ CTR АОМ/ DSS DSS

Nbxlm Nlrxlmc

CTR АОМ/ CTR АОМ/ DSS DSS

Nlrxlш Nlrxl"*1

S H s «2

л ^ 1

■ DSS tissue damage score

□ Hyperplasia score

■ Global inflammation score

□ Goblet cell score

□ PMN inflammatory score

□ PMN infiltration in LM per HPF

■ Submucosal edema score

Nlrxl0/11 Nlrxlшс NlrxlM Nlrxlшс AOM/DSS AOM/DSS

Cohort #1

s «H i

Cohort #2

О 1 mm О 2 mm О 3 mm О 4 mm □ 5 mm О 6 mm ■ 7 mm ^ emm

Adenomas

NlrxlIEC Nlrxlш Nlrxl,EC Nlrxlш Nlrxl,EC Nlrxlш

Nlrxlm Nlrxlwc

Count per mouse

Adenoma

b* ** .

Nlrxl Nlrxl

Figure 2. Expression of Nlrxl in the Intestinal Epithelium Protects against Colitis-Associated Tumorigenesis

Two independent studies were performed in Nlrxlл/л and Nlrxl DIEC mice to analyze the effect of AOM/DSS on tumorigenesis. Cohorts #1 and #2 were littermate females and males, respectively, either left untreated (n R 3) or receiving AOM/DSS treatment (n R 7).

(A) Weight curves.

(B) Measurement of colon length.

(C) Histopathological evaluation of inflammation-associated parameters.

(D and E) Breakdown of the tumor number per size (diameter) (D) and by grade (E).

(F) Representative micrographs showing polyps in colons from AOM/DSS-treated mice (aberrant crypt foci, ACF and high grade dysplasia, HGD) (non-significant, NS) (*p < 0.05 and **p<0.01).

• AptP** / Ntxl-'-

• Apd^+ / Nirxl+I+

ApcM'"l+Nlrxl+l+ ■

Ap^+Nlrxl4

Figure 3. Nlrxl Protects against Tumorigen-esis in Apcmin/+ Mice

Analysis of Apcmin/+ Nlrx1+'+ (n = 6) versus Apcmin/+ Nlrx1~^ mice (n = 12).

(A) Survival curve.

(B) Age in days at sacrifice. Tumors were analyzed for animals sacrificed between days 100-150.

(C) Total number of colonic tumors per mouse.

(D) Cumulative tumor size per mouse (*p < 0.05 and **p<0.01).

0 40 80 120 160 200 Age at sacrifice (days)

Apd^Ntocl*»* Apc^'Nlrxl-l-

I € h 3

1 ■S

Apcm'"NlrxlwAp<r~"'Mrxl+

DSS in drinking water for 5 days, both groups of mice lost weight similarly, up to day 7 (i.e., 2 days after removal of DSS from drinking water) (Figure 4A). From that time point, both animal groups recovered weight progressively, indicative of a healing process, but Nlrx1DIEC mice recovered faster during the d10-d15 period (Figure 4A). Despite this difference in the recovery from DSS-induced colitis, inflammation appeared to be comparable between the two groups of mice, as evidenced by similar colon length at sacrifice (Figure 4B), histology (Figure 4C), blinded inflammatory scores on colonic sections (Figure 4D), levels of keratinocyte chemoattractant or CXCL1 in the serum (Figure S3A), and of secreted IL-1b from intestinal explants in culture (Figure S3B). Of note, this analysis further showed that the peak of intestinal erosion and inflammation was observed at day 7 in both Nfrx1fl/fl and Nlrx1DIEC mice, which correlates well with the weight loss data (see Figure 4A), but suggests that the faster recovery of Nlrx1DIEC mice is independent from inflammation. Next, we performed qPCR analysis on colon mucosa samples from DSS-treated Nfrx1fl/fl and Nlrx1DIEC mice at day 7 and d12. At the peak of tissue damage (day 7), inflammation-associated genes such as A20, Il10, and Il6 were transcriptionally upregulated in both groups of mice and this activation persisted, at least partially up to d12 (Figures S3C-S3E). Although we noted a trend for increased upre-gulation of Il6 at day 7 and d12 in Nlrx1DIEC mice as compared to Nlrx1fl/fl mice, this effect was not significant. Moreover, Il22 and Il22ra1 were transiently upregulated at day 7, but this induction was similar between Nlrx1fl/fl and Nlrx1DIEC mice (Figures S3F and S3G). Interestingly, while Tnf, Egf, and Tgfb1 were not transcriptionally upregulated at day 7 in either group of mice, we observed a delayed upregulation of these genes

at d12, and this upregulation was significantly more robust in Nlrx1DIEC mice as compared to Nlrx1fl/fl mice (Figures 4E-4G). Previous work demonstrated that mi-crobial products, in synergy with Notch signaling, could potentiate Tnf expression in the intestine (Hu et al., 2008). Interestingly, Notch signaling was activated early (day 7) during colitis as evidenced by the upregulation of the Notch target Hey1 , and this induction was significantly more pronounced in Nlrx1DIEC mice than Nlrx1fl/fl mice (Figure S3H). At d12, Notch signaling was still active since expression of Atoh1, which is inhibited by Notch, was more blunted in Nlrx1DIEC mice than Nlrx1fl/fl mice (Figure S3I). Thus, early upregulation of Notch signaling could account, at least in part, for the increased upregulation of Tnf observed in Nlrx1DIEC mice during colitis.

TNF, EGF, and TGFp have well-characterized roles in promoting the healing of the intestinal epithelium following injury (Lep-pkes et al., 2014; Sturm and Dignass, 2008). In particular, TNF stimulates the proliferation of IEC, TGFp potentiates their migration, and EGF promotes both. Thus, these results suggest that Nlrx1 deficiency in IEC does not impact the initial wave of induction of inflammatory mediators, which likely originate from immune cells of the lamina propria, but upregulates second-wave mediators involved in tissue healing such as TNF, EGF, and TGFp.

Increased Epithelial Proliferation in Nlrx1DIEC Mice during Epithelial Regeneration

Treatment with DSS results in a massive epithelial erosion and healing requires upregulation of IEC proliferation and migration. Because we observed increased expression of TNF in Nlrx1DIEC mice during the regeneration phase following DSS treatment (see above), we aimed to analyze epithelial proliferation.

Immunohistochemistry (IHC) staining using anti-Ki-67, a marker of cellular proliferation, revealed that an increased number of epithelial cells were proliferative in the crypts of Nlrx1DIEC mice as compared to Nlrx1fl/fl mice at d12 and d20 post-DSS treatment (Figures 5A-5C). It must be noted that at the peak of DSS-induced tissue damage (day 7), the epithelium was too eroded to perform analysis of Ki-67. In the AOM/DSS-treated animals analyzed above (see Figure 2), which were sacrificed at d20 of the third DSS cycle, a similar increase in the number of

- Mnlm

- 2.5% dss

- fcl1

- mini 2.5% c

10№ NS

•i* T

t 8I> S •it

a 7» * • •

1EC-KO fl/fl 1EC-KO fl/ft ŒC-KO fl/fl 1EO-KO

Nlrxl1

DSS d7

DSS dl2 .

Wffl).

DSS d20

CTR DSS d7 DSS dl2 DSSd20

Figure 4. Epithelial-Specific Deletion of Nlrxl Does Not Affect Peak Tissue Damage Induced by DSS, but Potentiates Expression of Late Tissue Healing Factors TNFa and TGFp

Nlrx1ml and Nlrx1 DIEC mice were challenged with 2.5% DSS for 5 days in drinking water.

(A) Weight curves.

(B) Measurement of colon length of animals sacrificed at day 7 (5 days DSS/2 days recovery in normal water), d12 (5 days DSS/7 days recovery), and d20 (5 days DSS/15 days recovery).

(C) Representative H&E micrographs of colons of animals unchallenged (CTR) or challenged with DSS at day 7, d12, and d20.

(D) Histopathological evaluation of inflammation-associated parameters. The numbers on the right side of the image: 1, total area affected; 2, destruction of mucosal architecture; 3, % severe destruction; 4, degree of cellular infiltration; 5, % severely affected; 6, muscle thickening; 7, crypt abscess; and 8, goblet cell depletion.

(E-G) qPCR analysis of Tnf (E), Egf (F), and Tgfb1 (G) in colonic tissues of animals unchallenged (CTR) or challenged with DSS at day 7, d12, and d20. Each point represents the average value for one mouse obtained as a technical triplicate. (*p < 0.05 and **p < 0.01). (non-significant, NS).

_••«'

U 0.02-

Ki-67+ cells was observed in the crypts of Nlrx1DIEC mice as compared to Nlrx1fl/fl mice (Figure 5D).

In resting conditions, epithelial cell death must approximately be equal to proliferation in order to keep colonic crypt length constant. Because we previously showed that NLRX1 controls apoptotic cell death, we sought to determine if the more robust proliferation observed in DSS-treated Nlrx1DIEC IECs was the result of increased apoptosis. IHC analysis of apoptosis using an antibody against the cleaved form of caspase-3 showed an increase in apoptosis in the colonic crypts of d12 DSS-treated mice as compared to unchallenged animals, but this increase was similar between Nlrx1fl/fl and Nlrx1DIEC mice (Figures 5E and 5F), thus ruling out the possibility that increased proliferation in DSS-treated Nlrx1DIEC mice was a compensation mechanism for enhanced apoptotic cell death. Together, these results show that NLRX1 negatively regulates IEC proliferation during the epithelial regeneration phase following intestinal injury.

Expression of NLRX1 Inhibits TNF-Mediated Proliferation and Signaling in Primary Intestinal Organoids

In order to gain insights into the mechanism through which epithelial-intrinsic expression of NLRX1 inhibits epithelial proliferation during the regeneration phase post-DSS, we isolated intestinal -1-.-.— crypts and cultured primary intestinal

mc-ko m mc-ko fl/fl diec

dss«--5ssdi2— organoids from Nlrx1 and Nlrx1

mice in Matrigel, as previously described (Sato et al., 2009). Since we observed increased expression of TNF in the colons of Nlrx1DIEC mice as compared to Nlrx1fl/fl mice during the epithelial regeneration phase (see Figure 4), we hypothesized that TNF could contribute to the hyper-proliferative response observed in Nlrx1DIEC mice recovering from DSS. TNF stimulation for 24 hr of freshly passaged organoids from Nlrx1fl/fl mice resulted in a modest non-significant increase in the diameter of the organoids (Figures 6A and 6B). In contrast, TNF stimulation more than doubled the surface of organoids from Nlrx1DIEC mice (Figures 6A and 6B), and those also had more numerous and longer digitations, showing that these structures displayed increased crypt growth. Next, TNF-stim-ulated organoids were pulsed with EdU, which incorporates into proliferative cells, and analyzed by flow cytometry. Interestingly, organoids from Nlrx1DIEC mice displayed more EdU+ cells than those from Nlrx1fl/fl mice (Figures 6C and 6D), thus showing that the increase in organoid size following

Nlrxl M

Nlrxl ŒC

CTR -

dl2 T . . • r-, . v A ■ ■ " : ■:> '•"' , r ï J

Ki-67 DSS d20

1.00-1

fl/fl IEC-KO fl/D IEC-KO

Nlrxl '

DSS d20

Nlrxl1

DSS dl2

1 " «.-. v' i\ > ' L V rJ/M

> v SV , ■Y- V"- . s* '' ,-i v »>'••

Cleaved caspase-3

DSS dl2

•-E-I

1 £ s &

fl/fl IEC-KO fl/fl IEC-KO

AOM/DSS

DSS dl2

•!•§ ® S

l.OO-i

Figure 5. Increased Epithelial Proliferation in Nlrx1DIEC Mice during Epithelial Regeneration

(A) IHC using anti-Ki-67 antibody of colonic sections of Nlrxlfl/" and Nlrxl DIEC mice unchallenged (CTR) or challenged with 2.5% DSS at d12 (5 days DSS/7 days recovery).

(B-D) Quantification of the average proportion of the crypt length that is Ki-67+ at post-DSS d12 (B), d20 (C), or animals challenged with AOM/DSS (D). Each point represents the quantification of n R 100 crypts of one animal.

(E and F) Representative IHC micrographs (E) and quantification (F) or colonic sections at d12 post-DSS using an antibody against the cleaved caspase-3.

expressed in the TA zone (Itzkovitz et al., 2012). Interestingly, we observed stronger upregulation of Olfm4 in organoids

fl/fl IEC-KO fl/fl IEC-KO

AOM/DSS

1-1 «1

fl/fl IEC-KO fl/fl IEC-KO

DSS dl2

TNF stimulation was caused, at least in part, by hyper-proliferation.

Intestinal organoids grow from rapidly dividing pluripotent intestinal stem cells (ISCs) that are found at the bottom of the crypts and express the marker Lgr5 (Barker et al., 2007). Another ISC population, likely acting as a reserve ISC, expresses Bmi1 (Sangiorgi and Capecchi, 2008; Tian et al., 2011). Cells leaving the ISC niche lose expression of the Lgr5 and Bmi1 markers and form the transit-amplifying (TA) zone, in which cells remain highly proliferative and still express numerous ISC markers, but progressively differentiate into specific IEC populations. qPCR analysis revealed that stimulation with TNF resulted in a modest increase in the expression of Lgr5 and Bmi1 in intestinal organoids from both Nlrx1DIEC and Nlrx1fl/fl mice; however, this increase was not significantly different between the two groups of organoids (Figures S4A and S4B). Next, we analyzed the expression of Olfm4, an ISC marker that, in contrast to Bmi1 and Lgr5, remains

mice following TNF stimulation (Figure 6E), suggesting that NLRX1 specifically represses proliferation of cells in the TA zone, rather than cells in the ISC niche.

Uncontrolled proliferation of cells in the ISC niche or the TA zone can drive tumor-igenesis. It was shown that the transcription factor Myb, which is highly expressed in ISCs (Munoz et al., 2012), was required for proliferation in colonic crypts (Mala-terre et al., 2007) and Myb activation results in colon tumorigenesis in mice (Malaterre et al., 2015). Strikingly, TNF stimulation resulted in only a transient upregulation of Myb expression in organoids from Nlrx1fl/fl mice, while this upregulation persisted in those from Nlrx1DIEC mice (Figure 6F). Thus, dysregulated expression of Myb in the colonic crypts of Nlrx1DIEC mice during epithelial restitution following injury could contribute to the exacerbated tumorigenesis observed in these mice in the AOM/DSS model.

We have demonstrated above that epithelial regeneration following DSS-induced injury resulted in exacerbated expression of Tnf, Egf, and Tgfb1 in Nlrx1DIEC mice as compared to Nlrx1fl/fl mice, suggesting that these factors could be expressed by the intestinal epithelium itself, since non-IECs in both groups of mice are WT. In support for this, we observed that stimulation of organoids from Nlrx1DIEC and Nlrx1fl/fl mice with TNF resulted in potent transcriptional upregulation of both Tnf and Tgfb1 (Figures 6G and 6H), as well as a more modest induction of Egf (Figure S4C). In contrast, Il6 transcript was undetectable in these conditions (data not shown). Therefore, these data support the notion that Tnf, Tgfb1, and Egf can be expressed through paracrine regulation by the intestinal epithelium.

from Nlrx1 C than those from NlrxVm

IEC-KO

TNF 24h

Jij'j w^Êël. 10000-1 8000-I 6000- ■o o 4000-

Js 2000-o

Nlrxl™ Nlrxlmc D

TNF (h) 0 4g 04g TNF (h) 0 4 8 0 4 8

fl/fl IEC-KO

0.025-

g 0.020*3>

5 0.015-

£ 0.010-^ 0.005-

fl/fl IEC-KO Tnf

0.0100

0.0075

0.0050

1 H 0.0025

TNF (h) 0 4 8 0 4 8 TNF (h) 0 4 8 0 4 8

fl/fl IEC-KO

fl/fl IEC-KO

fl/fl IEC-KO 11/11 IEC-KO

TNF 24h

fl/fl IEC-KO TNF4h

Nlrxl1

Nlrxl <

0 15 30 45 60 0

15 30 45 60 TNF (min.) P-Akt

In order to better understand the mechanism through which TNF triggered a hyper-proliferative response in organoids from

NlrxlD

mice, we analyzed by western blotting the activation

of Akt, a protein that plays a central role in the regulation of epithelial cell proliferation (Lee et al., 2010) and whose dysregu-lation is strongly associated with colorectal cancer (Agarwal et al., 2013; Leystraet al., 2012). TNF stimulation resulted in rapid upregulation of Akt phosphorylation in both groups of organoids; however, the effect was stronger and more sustained in organo-ids from Nlrx1MEC mice than those from Nlrx1fl/fl mice (Figure 6I). Finally, we analyzed NF-kB signaling, since this pathway is also critical for the control of IEC proliferation and intestinal tumori-genesis (Myant et al., 2013; Steinbrecher et al., 2008). Similar to Akt activation, we observed that TNF stimulation triggered degradation of the inhibitory molecule Ikba, which indicates NF-kB activation, in both groups of organoids, but the effect was again more sustained in organoids from Nlrx1DIEC mice than those from Nlrx1fl/fl mice (Figure 6I). Together, these results indicate that TNF-mediated hyper-proliferative response in orga-

Figure 6. Expression of NLRX1 Inhibits TNF-Mediated Proliferation and Signalization in Primary Intestinal Organoids

(A and B) Representative bright field micrographs (A) and quantification of the surface (B) of intestinal organoids derived from Nlrxlfl/fl (fl/fl) and Nlrxl DIEC (IEC-KO) mice in normal growth media supplemented or not with TNF at 10 ng/ml for 24 hr. For quantifications, each point represents the average surface area of all the organoids found in one individual culture well. The results are representative of two independent experiments. (C and D) Analysis by flow cytometry (C) and quantification of the average mean fluorescence intensity (MFI) (D) of EdU incorporation in fl/fl and IEC-KO organoids stimulated or not with TNF at 10 ng/ml for 4 hr. The results are representative of three independent experiments. (E-H) qPCR analysis of the expression of Olfm4 (E), Myb (F), Tnf (G), or Tgfbl (H) in fl/fl and IEC-KO organoids stimulated or not with TNF at 10 ng/ml for 4 or 8 hr. Each point represents the average value for technical triplicate of individual organoid cultures derived from different mice. (I) Western blot analysis of fl/fl and IEC-KO organoids stimulated or not with TNF at 10 ng/ml for 15 to 60 min. The blots are representative of two independent experiments (*p < 0.05; **p < 0.01; ***p < 0.001; and ****p < 0.0001) (non-significant, NS).

noids from Nlrx1DIEC mice was associated with exacerbated activation of the Akt and NF-kB signaling pathways and upregula-tion of the transcription factor Myb.

DISCUSSION

Chronic injury of the intestinal mucosa can favor the development of colorectal cancer. This is highlighted by the fact that inflammatory bowel disease confers increased risk of developing gastrointestinal neoplasia in humans. Similarly, repeated exposure to the epithelial irritant DSS greatly accelerates tumor formation in the AOM/DSS model in mice (Rosenberg et al., 2009). It is clear that inflammation plays a critical role in this effect, and inflammation-associated cyto-kines such as IL-6 are known to contribute to the pathogenesis of colorectal cancer (Waldner and Neurath, 2014). However, it is less appreciated that epithelial regeneration and wound repair pathways, which are triggered as a direct consequence of mucosal injury and are amplified by inflammatory mediators, likely play central roles in colitis-associated intestinal tumorigen-esis. Indeed, these important processes rely on cellular proliferation and migration and on signaling pathways dependent on NF-kB, Akt/PI3K, and Wnt/p-catenin, all of which being critical actors of colorectal cancer pathogenesis. Because inflammation and epithelial repair pathways are tightly interconnected and are spatially and temporally coordinated, it is difficult to determine what aspects of inflammation (i.e., recruitment of immune cells, upregulation of phagocytosis, secretion of inflammatory or

Tubulin

immune defense mediators, microbiota dysbiosis, and upregula-tion of epithelial repair) have the most important impact on tumorigenesis. One way to approach this question and to identify key molecular pathways linking specific arms of inflammation to tumorigenesis is to perform analysis of mice with tissue-specific gene deletion. Here, by using mice harboring a deletion of Nlrx1 only in IECs, we identified an epithelial-intrinsic role for this NLR molecule in the control of IEC proliferation during the epithelial regeneration phase following DSS treatment. Our data suggest that Nlrx1 deletion exacerbates cellular responses to TNF, which could generate a paracrine or autocrine amplification loop in IECs by triggering Tnf expression. This effect might also amplify the expression of other wound healing genes, such as Tgfb1 and Egf, as suggested by our results in intestinal organoids. Thus, our data suggest that, by restricting the activation of Akt and NF-kB in IECs in response to TNF stimulation, NLRX1 tones down the effect of TNF on the proliferation of TA cells in the intestinal crypt, providing a protection against colitis-associated tumorigenesis. It is possible that a similar mechanism underlies the protective role of NLRX1 against tumorigenesis in mice carrying the Apcmin mutation, since increased mucosal expression of Tnf was observed in these mice and correlated with tumor formation (McClellanetal., 2012).

Our previous results demonstrated that full-body deletion of Nlrx1 resulted in increased inflammation following DSS treatment (Soares et al., 2014), which is in line with other studies showing an inhibitory effect of NLRX1 on inflammation (Allen et al., 2011; Xia et al., 2011; Eitas et al., 2014; Kang et al., 2015). Although this effect is likely immune cell-intrinsic, the underlying mechanism is not clearly elucidated, nor are the immune cell subsets identified that are undergoing uncontrolled activation in Nlrx1 -deficient mice. Nevertheless, these observations, together with the results of the present study, suggest that NLRX1 might mitigate colitis-associated tumorigenesis at two levels: (1) by downregulating inflammation and (2) by inhibiting pro-proliferative responses in the intestinal epithelium in response to inflammatory and pro-regeneration cues.

Our data show that, in intestinal organoids, NLRX1 diminishes TNF-induced activation of both Akt- and NF-kB-dependent pathways. Since Akt was shown to be essential for TNF-medi-ated activation of NF-kB (Ozes et al., 1999), it is likely that the primary effect of NLRX1 is at the level of Akt signaling. Alternatively, NLRX1 could regulate a step or an event that controls both pathways, such as mitochondrial ROS levels or expression of TNF receptors. Future studies will aim at identifying the molecular mechanism by which NLRX1 mitigates Akt and NF-kB signaling in response to TNF stimulation in IECs.

Overall, our study identifies an epithelial-intrinsic role for NLRX1 in the control of colon tumorigenesis through the modulation of epithelial proliferation following injury. Experiments in primary intestinal organoids further revealed that NLRX1 directly impacted Akt and NF-kB pathways, expression of the proliferation-associated transcription factor Myb, and proliferation of crypt TA cells in response to TNF. Identifying the mechanism through which NLRX1 controls these pathways will allow the designing of novel therapeutic strategies aimed at regulating hy-per-proliferative responses in colon cancer without having to suppress inflammatory cytokine networks.

EXPERIMENTAL PROCEDURES Animals

All mice were bred and housed at the Division of Clinical Medicine (DCM) of the University of Toronto and experiments were performed according to guidelines of the DCM and following protocols approved by the University of Toronto Committee on Use and Care of Animals. Mice were kept in specific pathogen-free conditions with periodic testing for murine norovirus and used for experiments at age 6-10 weeks except otherwise indicated. All experiments were performed using littermate animals that were separated at weaning. Nlrx1~'~ mice have been described previously (Soares et al., 2013).

Generation of IEC-Specific NLRX1-Deficient Mice

NLRX1-deficient mice (Nlrx1DEC) were generated by crossing Nlrx1fl/fl mice (Soares et al., 2013) with transgenic mice B6.SJL-Tg(Vil-cre)997Gum/J (Jackson Laboratories), which express Cre recombinase under the control of the mouse villin promoter. The following PCR primers (5-3') were used to genotype WT, floxed, and Nlrx1DIEC mice: primer 1- (CCA TTT GCC AAT CCC ACT CAC), primer 2 (ACC AAG AAC CTA ACC CAC GGT C), primer 3 (TTG CCA GCC ATC TGT TGT TTG), oIMR1878 (GTG TGG GAC AGA GAA CAA ACC), and oIMR1879 (ACA TCT TCA GGT TCT GCG GG).

Induction of DSS-Induced Colitis

Colitis was induced with 2.5% (w/v) DSS (MP Biomedicals) dissolved in sterile, distilled waterfor experimental days 5, followed by normal drinking water until the end of the experiment.

Colitis-Associated Colorectal Cancer

Mice were injected intraperitoneally with 10 mg/kg AOM (Sigma-Aldrich). After 5 days, 2.5% DSS was given in drinking water over 5 days, followed by regular drinking water for 2 weeks. This cycle was repeated three times, and mice were sacrificed 2 weeks after the last DSS cycle. The experiments were performed with at least 4-8 mice in each AOM-treated condition.

Reagents and Antibodies

AOM (Sigma-Aldrich, A5486). DSS (MP Biomedicals, 160110). The following antibodies were used for protein detection: rabbit polyclonal AKT rabbit poly-clonal (Cell Signaling Technology, 9272S), rabbit polyclonal Phospho-AKT (Ser473) (Cell Signaling Technology, 9271S), rabbit polyclonal anti-cleaved caspase-3 (Asp175) (Cell Signaling Technology, 9661S), rabbit anti-chromog-ranin A (Abcam, ab15160), rabbit anti-Muc2 (H300) (Santa Cruz, sc-15334) mouse monoclonal anti-tubulin (Sigma-Aldrich, T9026), rabbit polyclonal anti-Ki-67 (D3B5) (Cell Signaling Technology, 12202), mouse monoclonal anti-NLRX1 (Millipore, 04-146) and rabbit monoclonal anti-IkB-a (E130) (Abcam, ab32518), goat anti-rabbit (ThermoScientific / Molecular Probes, A-21245), AB solution (Sigma-Aldrich, B8438), Periodic Acid- Schiff (PAS) kit (Sigma-Aldrich, 395B), Alexa Fluor 647 donkey anti-rabbit (Life Technologies, A31573), and Click-iT EdU Flow Cytometry Assay Kit (Life Technologies, C10419).

Statistical Analysis

Prism software was used to plot data and determine statistical significance using a Student's t test (for two groups) or ANOVAfor data sets with three or more sets of data to analyze. Data are presented as means ± SEM. A p value of <0.05 was considered to be statistically significant.

SUPPLEMENTAL INFORMATION

Supplemental Information includes Supplemental Experimental Procedures, four figures, and one table and can be found with this article online at http:// dx.doi.org/10.1016/j.celrep.2016.02.065.

AUTHOR CONTRIBUTIONS

I.T., D.J.P., and S.E.G. designed the study and analyzed data. I.T. performed most experiments of the study. S.E.G. wrote the manuscript. S.A.K. and

M.A.R. performed western blots, E.G.F. performed Immunofluorescence, R.M. did EdU assays, and C.M. performed some qPCRs. S.W., D.A.W., and C.J.S. performed blinded pathological scoring.

ACKNOWLEDGMENTS

This project is supported in part by the Nick Natale Innovation Grant of the Canadian Cancer Society (grant #702392) (to S.E.G.), the Natural Sciences and Engineering Research Council of Canada (to S.E.G.), and from a team grant from the Canadian Institutes of Health Research on NLRs and inflammation (to D.J.P. and S.E.G.). We thank Dr. Sheila Costford for some of the statistical analyses and Dr. A. Martin (University of Toronto) for providing the Apcmin/+ mice.

Received: January 4, 2016 Revised: February 8, 2016 Accepted: February 22, 2016 Published: March 10, 2016

REFERENCES

Agarwal, E., Brattain, M.G., and Chowdhury, S. (2013). Cell survival and metastasis regulation by Akt signaling in colorectal cancer. Cell. Signal. 25, 1711-1719.

Allen, I.C., Moore, C.B., Schneider, M., Lei, Y., Davis, B.K., Scull, M.A., Gris, D., Roney, K.E., Zimmermann, A.G., Bowzard, J.B., et al. (2011). NLRX1 protein attenuates inflammatory responses to infection by interfering with the RIG-I-MAVS and TRAF6-NF-kB signaling pathways. Immunity 34, 854-865. Arnoult, D., Soares, F., Tattoli, I., Castanier, C., Philpott, D.J., and Girardin, S.E. (2009). An N-terminal addressing sequence targets NLRX1 to the mitochondrial matrix. J. Cell Sci. 122, 3161-3168.

Barker, N., van Es, J.H., Kuipers, J., Kujala, P., van den Born, M., Cozijnsen, M., Haegebarth, A., Korving, J., Begthel, H., Peters, P.J., and Clevers, H. (2007). Identification of stem cells in small intestine and colon by marker gene Lgr5. Nature 449, 1003-1007.

Danielsson, F., Skogs, M., Huss, M., Rexhepaj, E., O'Hurley, G., Klevebring, D., Ponten, F., Gad, A.K., Uhlen, M., and Lundberg, E. (2013). Majority of differentially expressed genes are down-regulated during malignant transformation in a four-stage model. Proc. Natl. Acad. Sci. USA 110, 6853-6858. Eitas, T.K., Chou, W.C., Wen, H., Gris, D., Robbins, G.R., Brickey, J., Oyama, Y., and Ting, J.P. (2014). The nucleotide-binding leucine-rich repeat (NLR) family member NLRX1 mediates protection against experimental autoimmune encephalomyelitis and represses macrophage/microglia-induced inflammation. J. Biol. Chem. 289, 4173-4179.

Hu, X., Chung, A.Y., Wu, I., Foldi, J., Chen, J., Ji, J.D., Tateya, T., Kang, Y.J., Han, J., Gessler, M., et al. (2008). Integrated regulation of Toll-like receptor responses by Notch and interferon-gamma pathways. Immunity 29, 691-703. Imbeault, E., Mahvelati, T.M., Braun, R., Gris, P., and Gris, D. (2014). Nlrx1 regulates neuronal cell death. Mol. Brain 7, 90.

Itzkovitz, S., Lyubimova, A., Blat, I.C., Maynard, M., van Es, J., Lees, J., Jacks, T., Clevers, H., and van Oudenaarden, A. (2012). Single-molecule transcript counting of stem-cell markers in the mouse intestine. Nat. Cell Biol. 14, 106-114.

Jaworska, J., Coulombe, F., Downey, J., Tzelepis, F., Shalaby, K., Tattoli, I., Berube, J., Rousseau, S., Martin, J.G., Girardin, S.E., et al. (2014). NLRX1 prevents mitochondrial induced apoptosis and enhances macrophage antiviral immunity by interacting with influenza virus PB1-F2 protein. Proc. Natl. Acad. Sci. USA 111, E2110-E2119.

Kang, M.J., Yoon, C.M., Kim, B.H., Lee, C.M., Zhou, Y., Sauler, M., Homer, R., Dhamija, A., Boffa, D., West, A.P., et al. (2015). Suppression of NLRX1 in chronic obstructive pulmonary disease. J. Clin. Invest. 125, 2458-2462. Lee, G., Goretsky, T., Managlia, E., Dirisina, R., Singh, A.P., Brown, J.B., May, R., Yang, G.Y., Ragheb, J.W., Evers, B.M., et al. (2010). Phosphoinositide 3-ki-

nase signaling mediates beta-catenln activation In Intestinal epithelial stem and progenitor cells in colitis. Gastroenterology 139, 869-881. Leppkes, M., Roulis, M., Neurath, M.F., Kollias, G., and Becker, C. (2014). Pleiotropic functions of TNF-a in the regulation of the intestinal epithelial response to inflammation. Int. Immunol. 26, 509-515.

Leystra, A.A., Deming, D.A., Zahm, C.D., Farhoud, M., Olson, T.J., Hadac, J.N., Nettekoven, L.A., Albrecht, D.M., Clipson, L., Sullivan, R., et al. (2012). Mice expressing activated PI3K rapidly develop advanced colon cancer. Cancer Res. 72, 2931-2936.

Malaterre, J., Carpinelli, M., Ernst, M., Alexander, W., Cooke, M., Sutton, S., Dworkin, S., Heath, J.K., Frampton, J., McArthur, G., et al. (2007). c-Myb is required for progenitor cell homeostasis in colonic crypts. Proc. Natl. Acad. Sci. USA 104, 3829-3834.

Malaterre, J., Pereira, L., Putoczki, T., Millen, R., Paquet-Fifield, S., Germann, M., Liu, J., Cheasley, D., Sampurno, S., Stacker, S.A., et al. (2015). Intestinal-specific activatable Myb initiates colon tumorigenesis in mice. Oncogene. http://dx.doi.org/10.1038/onc.2015.305.

McClellan, J.L., Davis, J.M., Steiner, J.L., Day, S.D., Steck, S.E., Carmichael, M.D., and Murphy, E.A. (2012). Intestinal inflammatory cytokine response in relation to tumorigenesis in the Apc(Min/+) mouse. Cytokine 57, 113-119. Moore, C.B., Bergstralh, D.T., Duncan, J.A., Lei, Y., Morrison, T.E., Zimmermann, A.G., Accavitti-Loper, M.A., Madden, V.J., Sun, L., Ye, Z., et al. (2008). NLRX1 is a regulator of mitochondrial antiviral immunity. Nature 451, 573-577.

Munoz, J., Stange, D.E., Schepers, A.G., van de Wetering, M., Koo, B.K., Itzkovitz, S., Volckmann, R., Kung, K.S., Koster, J., Radulescu, S., et al. (2012). The Lgr5 intestinal stem cell signature: robust expression of proposed quiescent '+4' cell markers. EMBO J. 31, 3079-3091.

Myant, K.B., Cammareri, P., McGhee, E.J., Ridgway, R.A., Huels, D.J., Cordero, J.B., Schwitalla, S., Kalna, G., Ogg, E.L., Athineos, D., et al. (2013). ROS production and NF-kB activation triggered by RAC1 facilitate WNT-driven intestinal stem cell proliferation and colorectal cancer initiation. Cell Stem Cell 12, 761-773.

Ozes, O.N., Mayo, L.D., Gustin, J.A., Pfeffer, S.R., Pfeffer, L.M., and Donner, D.B. (1999). NF-kappaB activation by tumour necrosis factor requires the Akt serine-threonine kinase. Nature 401, 82-85.

Philpott, D.J., Sorbara, M.T., Robertson, S.J., Croitoru, K., and Girardin, S.E. (2014). NOD proteins: regulators of inflammation in health and disease. Nat. Rev. Immunol. 14, 9-23.

Raskov, H., Pommergaard, H.C., Burcharth, J., and Rosenberg, J. (2014). Colorectal carcinogenesis-update and perspectives. World J. Gastroenterol. 20, 18151-18164.

Rebsamen, M., Vazquez, J., Tardivel, A., Guarda, G., Curran, J., and Tschopp, J. (2011). NLRX1/NOD5 deficiency does not affect MAVS signalling. Cell Death Differ. 18, 1387.

Rosenberg, D.W., Giardina, C., and Tanaka, T. (2009). Mouse models for the study of colon carcinogenesis. Carcinogenesis 30, 183-196. Sangiorgi, E., and Capecchi, M.R. (2008). Bmi1 is expressed in vivo in intestinal stem cells. Nat. Genet. 40, 915-920.

Sato, T., Vries, R.G., Snippert, H.J., van de Wetering, M., Barker, N., Stange, D.E., van Es, J.H., Abo, A., Kujala, P., Peters, P.J., and Clevers, H. (2009). Single Lgr5 stem cells build crypt-villus structures in vitro without a mesenchymal niche. Nature 459, 262-265.

Singh, K., Poteryakhina, A., Zheltukhin, A., Bhatelia, K., Prajapati, P., Sripada, L., Tomar, D., Singh, R., Singh, A.K., Chumakov, P.M., and Singh, R. (2015). NLRX1 acts as tumor suppressor by regulating TNF-a induced apoptosis and metabolism in cancer cells. Biochim. Biophys. Acta 1853, 1073-1086. Soares, F., Tattoli, I., Wortzman, M.E., Arnoult, D., Philpott, D.J., and Girardin, S.E. (2013). NLRX1 does not inhibit MAVS-dependent antiviral signalling. Innate Immun. 19, 438-448.

Soares, F., Tattoli, I., Rahman, M.A., Robertson, S.J., Belcheva, A., Liu, D., Streutker, C., Winer, S., Winer, D.A., Martin, A., et al. (2014). The mitochondrial

protein NLRX1 controls the balance between extrinsic and intrinsic apoptosis. J. Biol. Chem. 289, 19317-19330.

Steinbrecher, K.A., Harmel-Laws, E., Sitcheran, R., and Baldwin, A.S. (2008). Loss of epithelial RelA results in deregulated intestinal proliferative/apoptotic homeostasis and susceptibility to inflammation. J. Immunol. 180, 2588-2599. Sturm, A., and Dignass, A.U. (2008). Epithelial restitution and wound healing in inflammatory bowel disease. World J. Gastroenterol. 14, 348-353. Su, L.K., Kinzler, K.W., Vogelstein, B., Preisinger, A.C., Moser, A.R., Luongo, C., Gould, K.A., and Dove, W.F. (1992). Multiple intestinal neoplasia caused by a mutation in the murine homolog of the APC gene. Science 256, 668-670. Tattoli, I., Carneiro, L.A., Jehanno, M., Magalhaes, J.G., Shu, Y., Philpott, D.J., Arnoult, D., and Girardin, S.E. (2008). NLRX1 is a mitochondrial NOD-like re-

ceptor that amplifies NF-kappaB and JNK pathways by inducing reactive oxygen species production. EMBO Rep. 9, 293-300.

Tian, H., Biehs, B., Warming, S., Leong, K.G., Rangell, L., Klein, O.D., and de

Sauvage, F.J. (2011). A reserve stem cell population in small intestine renders

Lgr5-positive cells dispensable. Nature 478, 255-259.

Waldner, M.J., and Neurath, M.F. (2014). Master regulator of intestinal disease:

IL-6 in chronic inflammation and cancer development. Semin. Immunol. 26,

75-79.

Xia, X., Cui, J., Wang, H.Y., Zhu, L., Matsueda, S., Wang, Q., Yang, X., Hong, J., Songyang, Z., Chen, Z.J., and Wang, R.F. (2011). NLRX1 negatively regulates TLR-induced NF-kB signaling by targeting TRAF6 and IKK. Immunity 34, 843-853.