Aldehyde Dehydrogenase 1B1 in Ethanol Metabolism, Glucose

ALDEHYDE DEHYDROGENASE 1B1 IN ETHANOL METABOLISM, GLUCOSE

HOMEOSTASIS AND COLON CANCER

SURENDRA SINGH

B.V.Sc. & A.H., Rajasthan Agricultural University, 2001

M.V.Sc., Rajasthan Agricultural University, 2004

A thesis submitted to the

Faculty of the Graduate School of the

University of Colorado in partial fulfillment

of the requirements for the degree of

Doctor of Philosophy

Toxicology Program

2014

This thesis for the Doctor of Philosophy degree by

Surendra Singh

has been approved for the

Toxicology Program

David Thompson, Chair

Vasilis Vasiliou, Advisor

Dennis Petersen

David Orlicky

Ying Chen

Date 4/28/14

Singh, Surendra. (Ph.D., Toxicology)

Aldehyde Dehydrogenase 1B1 in Ethanol Metabolism, Glucose Homeostasis and

Colon Cancer

Thesis directed by Professor Vasilis Vasiliou.

ABSTRACT

Aldehyde dehydrogenases (ALDHs) are a group of NAD(P)dependent enzymes involved in the metabolism of a wide spectrum of aliphatic and aromatic aldehydes. ALDH1B1 is a mitochondrial homotetrameric enzyme which is 65% and 72% identical to ALDH1A1 and ALDH2 proteins, respectively. Our in vitro studies have shown that human ALDH1B1 metabolizes acetaldehyde with an apparent Km of 55 M, indicating an important role of this protein in alcohol metabolism. ALDH1B1 is expressed only at the crypt base, along with stem cells, in human colon. It is highly expressed in all cancerous cells of human colonic adenocarcinomas. This pattern of expression corresponds closely to that observed for Wnt/βcatenin signaling activity in normal and cancerous colon.

These findings suggest a potential role of ALDH1B1 in colon carcinogenesis.

Recently we also found that ALDH1B1 is a marker for mice pancreatic progenitor cells and is required for maintenance and expansion of progenitor pools. To assess the in vivo role of ALDH1B1, we have generated transgenic Aldh1b1(/) mice; these mice are fertile and have a normal growth pattern. ALDH1B1 messenger and protein were undetectable in examined tissues. We also

iii examined expression of ALDH2 and ALDH1A1 in these organs and did not find compensatory upregulation of these isozymes. Ethanol pharmacokinetics following a single intra peritoneal injection of ethanol 5g/kg revealed higher acetaldehyde levels in 3 and 24 hours in Aldh1b1(/) mice. At the 8weeks of age, Aldh1b1(/) mice showed higher fasting blood glucose levels and decreased glucose tolerance on intraperitoneal glucose tolerance test. The shRNAmediated knockdown of ALDH1B1 reduced the number and size of spheroids formed by SW480 colon cancer cells in 3dimensional matrigel culture.

ALDH1B1 knockdown depleted the highly carcinogenic ALDH bright cells and significantly decreased xenograft tumor formation in athymic mice. Protein and mRNA expression evaluation revealed downregulation of Wnt/βcatenin, Notch and PI3K/Aktsignaling pathways in ALDH1B1depleted colon cancer cells. In summary, our data demonstrate that ALDH1B1 is crucial for ethanol metabolism and glucose homeostasis and plays a functional role in colon cancer tumorigenesis by modulating the Wnt/βcatenin, Notch and PI3K/Akt signaling pathways and could be a possible target for more effective treatments for this devastating condition.

The form and content of this abstract are approved. I recommend its publication.

Approved: Vasilis Vasiliou

I dedicate this thesis to

My father, late Shanker Singh Rajpurohit, even though he did not live to share my achievement,

I am sure he is delighted from his heavenly home

ACKNOWLEDGEMENTS

It is not possible to express my sincere gratitude and appreciation in words to my advisor Vasilis Vasiliou for sharing his experience, scholastic guidance and scientific vision. He always been very patient and showed great confidence in me, I am especially thankful to him for constant encouragement, and personal interest in the research work right from its planning to its successful execution. The experience and knowledge I acquired here will continue to benefit me beyond my time in this lab.

I am grateful to have people with great scientific aptitude and helping nature in my advisory committee, their dedication to graduate education helped me immensely during my PhD research work. My sincere thanks to Dr. David

Thompson, for his enthusiastic and creative support, Dr. Dennis Petersen for showing continuous interest in my work, Dr. David Orlicky for guiding me throughout and for his great help with histopathological work. I can never thank enough to Dr. Ying Chen, who is not only a committee member but also been a great mentor, teacher, colleague, friend and a critic for me.

I also want to thank all the Vasiliou lab members for their cheerful company, constant encouragement and support during course of my research work, especially Chad Brocker, for his constructive critique. I am also grateful to members of Dr. Wells Messersmith lab, especially John Arcaroli for helping in planning and conducting experiments related with colon cancer cells.

I extend my cordial thanks to my amazing and supportive friends Gaurav,

Swetha and Sangeeta and their families for providing lot of memorable experiences during this period.

Last but not least I express my deep sense of gratitude and love for my family especially my mother, Sharda Rajpurohit, who is there for me with her continuous encouragement, inspiration and support throughout all the ups and downs. It gives me immense pleasure to express my gratitude to my wife, Priti, and children, Srishti and Aditya, for their patience, understanding and unconditional love during the years when this research work occupied a large chunk of my life, most of which originally belonged to them.

vii

TABLE OF CONTENTS

CHAPTER

I. BACKGROUND .. 1

Introduction ...... 1

Acetaldehyde and retinaldehydemetabolizing ALDHs in

colorectal cancer ........ 2

Acetaldehyde: a carcinogen . 7

Opposing effects of retinoic acid on cancer cell proliferation .. 9

ALDH and cancer stem cells ...... ........ 11

Cancer stem cells in colorectal cancer 12

ALDH isozymes in pancreatic functions and progenitor cells. 16

Summary .. 17

Objectives .... 18

II. CHARACTERIZATION OF ALDEHYDE DEHYDROGENASE 1B1 KNOCKOUT MICE: PHYSIOLOGICAL IMPLICATION OF ALDH1B1 IN ETHANOL METABOLISM AND GLUCOSE HOMEOSTASIS ...... 20

Introduction.. 20

Materials and methods .. 22

Results . 31

Discussion 40

viii

III. ALDH1B1 IS CRUCIAL FOR COLON TUMORIGENESIS BY MODULATING WNT/βCATENIN, NOTCH AND PI3K/AKT SIGNALING PATHWAYS .... 46

Introduction.. 46

Materials and methods ..... 50

Results .. 58

Discussion 71

IV. CONCLUSION .... 77

Summary... 77

Significance . 80

Future directions . 80

REFERENCES . 83

LIST OF TABLES

TABLE

1.1. ALDH expression in various progenitor, stem and cancer cell types . 6

1.2. Affinity of ALDHs for acetaldehyde and retinaldehyde . 8

2.1. Primers for QPCR analysis 25

LIST OF FIGURES

FIGURE

1.1. ALDHs modulate carcinogenesis by metabolizing acetaldehyde and retinaldehyde .... 5

1. 2. ALDHexpressing cells are responsible for chemoresistance and relapse of many tumors after chemotherapy ...... 12

1.3. ALDH1B1 expression pattern in normal colon and colon adenocarcinoma... 15

2.1. Generation of Aldh1b1(/) null mouse line... 28

2.2. Expression analysis of ALDH1B1 mRNA and protein in Aldh1b1(/) mice 33

2.3. Growth curve of Aldh1b1(+/+) and Aldh1b1(/) mice 34

2.4. Aldh1b1(/) mice have normal cytoarchitecture 35

2.5. Lack of ALDH1B1 does not affect number of proliferating cells in colon. 37

2.6. Lack of ALDH1B1 does not affect number of goblet cells in colon ..... 38

2.7. Pharmacokinetics of ethanol and acetaldehyde in Aldh1b1(/) mice . 39

2.8. Intraperitoneal glucose tolerance test (GTT) in Aldh1b1(/) mice ... 41

2.9. Photomicrographs of mouse pancreas after immunostaining for insulin and glucagon..... 42

3.1. Illustration of Wnt/βcatenin signaling and distribution of ALDH1B1 in wild type and Apc Min mice 60

3.2. Evaluation of human ALDH1B1 promoter activity in colon cancer cell lines. 62

3.3. ALDH1B1 promoter activity in colon cancer cell lines 63

3.4. ALDH1B1 knockdown in the SW480 cell line inhibits spheroid formation and tumor growth ... 65

3.5. ALDH1B1 knockdown in SW480 cells depletes ALDH bright cells .. 68

3.6. ALDH1B1 positively regulate Wnt/βcatenin, Notch and PI3K/Akt signaling pathways .. 70

3.7. Possible mechanism for the role of ALDH1B1 in colon tumorigenesis ... 76

xii

ABBREVIATIONS

ADH Alcohol dehydrogenase

ALDH Aldehyde dehydrogenase

AODS Antioxidative defense systems

APC Adenomatous polyposis coli

ATRA Alltrans retinoic acid

AUC Area under the curve

B2M β2 microglobulin

BAA Bodipyaminoacetate

BAAA Boron dipyrromethene aminoacetaldehyde

CRBPII Cellular retinoic acid binding protein

CRC Colorectal cancer

CSC Cancer stem cell

CSL CBF1/RBPJk, Su (H), Lag1

CYP2E1 Cytochrome P4502E1

DEAB Diethylaminobenzaldehyde

Dsh Dishevelled

ECM Extracellular matrix

ESA Epithelialspecific antigen

FABP5 Fatty acid binding protein 5

GCMS Gas chromatography Mass spectrometry

GI Gastrointestinal

GSK3β Glycogen synthase kinase 3β

xiii

H&E Hematoxilin and eosin

H2O2 Hydrogen peroxide

HES Hairy and enhancer of split

HSVTK Thymidine kinase minicassete i.p. Intraperitoneally

IACUC Institutional animal care and use committee

IARC International agency for research on cancer

ICN Intracellular Notch iLBP Intracellular lipid binding protein

IPGTT Intraperitoneal glucose tolerance test

LEF Lymphoid enhancer factor

LRP Lipoprotein receptor related protein

MAPK Mitogen activated protein kinase

Min Multiple intestinal neoplasia

MS Mass spectrometry

MW Molecular weight marker

NAD(P) Nicotinamide adenine dinucleotide (phosphate)

NEO Neomycinresistance minicassette

NTP National toxicology program

PCA Perchloric acid

PDTX Patientderived tumor xenograft

PI3K Phosphoinositide3kinase

PPAR Peroxisome proliferatoractivated receptor

PTEN Phosphatase and tensin homolog deleted on chromosome 10 xiv

RA Retinoic acid

RALDH Retinaldehyde dehydrogenase

RAR RA receptor

ROS Reactive oxygen species

RXR Retinoid X receptor

Sc Scramble

SIM Selected ion monitor

SNP Single nucleotide polymorphism

SPSS SigmaStat statistical analysis software

TBE TCF/LEF binding element

TBST Trisbuffered saline with 0.1% tween 20

TCF T cell factor

TESS Transcription element search system

TGFBR2 TGFbeta type II receptor

TGTC Transgenic and gene targeting core

TIC Tumor initiatingcell

TSS Transcription start site

WT Wildtype

CHAPTER I

BACKGROUND

Introduction

The aldehyde dehydrogenase (ALDH) superfamily contains NAD(P) + dependent enzymes that oxidize a wide range of endogenous and exogenous aldehydes to their corresponding carboxylic acids [1]. The ability of ALDHs to act as ‘aldehyde scavengers’ is grounded in the observation that many have broad substrate specificities and can metabolize a wide range of chemically and structurallydiverse aldehydes. Many of the ALDH isozymes overlap in relation to substrate specificities, tissue distribution and subcellular localization but vary in their efficiency in metabolizing specific aldehydes [25]. The human genome contains 19 proteincoding ALDH genes. ALDH proteins are found in one or more subcellular compartments including the cytosol, mitochondria, endoplasmic reticulum and nucleus [2]. Mutations and polymorphisms in ALDH genes are associated with various pathophysiological conditions in humans and rodents [6,

7] including alcoholrelated diseases [8] and cancers [9]. ALDH1B1 is a mitochondrial enzyme which was previously known as ALDH X or ALDH5 [10]. It shares 72 and 65 percent peptide sequence identity with ALDH2 and ALDH1A1 respectively [11]. It shares some catalytic functions with ALDH2 by metabolizing acetaldehyde, lipid peroxidationderived aldehydes and other short chain aldehydes [11, 12]. Ethanol is metabolized mainly by class I alcohol

1 dehydrogenase (ADH1) isozymes into acetaldehyde which is then oxidized to acetic acid, primarily by mitochondrial ALDH2 [7]. ALDH1B1 has also been found to metabolize retinaldehyde [13] to retinoic acid (RA) and thereby could play crucial role in cell proliferation and differentiation. Recently ALDH1B1 has also been implicated in pancreatic development and regeneration [14].

Acetaldehyde and retinaldehydemetabolizing ALDHs in colorectal cancer

Colorectal cancer (CRC) represents a serious health concern because of its very high morbidity and mortality. Each year, more than one million new CRC cases are diagnosed and over 500,000 deaths are associated with this condition worldwide [15]. The American Cancer Society estimated diagnosis of 136,830 new cases of CRC in the USA during 2014; of these, approximately 50,310 people are expected to die [16]. Although the exact mechanisms that promote

CRC remain obscure, there is increasing evidence suggesting the involvement of lifestylerelated factors in addition to genetic predisposition. These factors include waist circumference, folate and multivitamins in the diet, high fat and high energy diet, physical exercise, tobacco smoking and alcohol consumption [17, 18].

According to doseresponse metaanalysis and pooled results from cohort studies, chronic daily consumption of approximately 50 g alcohol increases the relative risk for colon cancer by 40 per cent [19, 20]. Various theories have been advanced regarding the mechanism by which alcohol induces cancer. For example, ethanol may enhance mucosal penetration of a carcinogen by serving as a solvent. In addition, ethanol induces cytochrome P4502E1 (CYP2E1), an

2 enzyme capable of generating reactive oxygen species (Figure 1.1). However, the most well accepted theory regarding ethanolinduced cancer involves acetaldehyde acting as a carcinogen [21, 22]. Ethanol is metabolized to acetaldehyde by ADH, CYP2E1 and catalase [23, 24] (Figure 1.1). Aldehydes covalently adduct proteins, nucleic acids and cellular biomolecules leading to

DNA damage, altered cellular homeostasis and hyperregeneration of colon mucosa, all of which can result in increased cancer risk [2529]. In 2009, the

International Agency for Research on Cancer (IARC) designated acetaldehyde

(as associated with alcohol consumption) to be a group I human carcinogen [30].

Reactive aldehyde generation is elevated in all clinical stages of colorectal cancer and the levels increase with, and are tightly correlated to, disease progression [3133]. ALDH enzymes detoxify reactive aldehydes produced during oxidative stress, especially in metabolicallyactive tissues, such as cancers [3436]. Acetaldehyde is metabolized to acetate, a process catalyzed by

ALDH2, ALDH1B1 and ALDH1A1 (Figure 1.1) [11]. The ability of these ALDHs to repress cellular acetaldehyde levels is consistent with a role for ALDHs in colon cancer and is supported strongly by the association of ALDH2 deficiency with high incidence of CRC in heavy ethanol drinkers [37]. In addition to metabolizing acetaldehyde, ALDH1 isozymes are the primary enzymes involved in the metabolism of retinaldehyde to RA, a signaling molecule that plays a crucial role in cellular proliferation and differentiation [24]. Given their ability to affect cellular

RA levels, it is likely that RAgenerating ALDHs have a role in modulating carcinogenesis. Several other observations lend support to the notion that

ALDHs are implicated in cancer. First, ALDH activity has been used to identify and isolate normal and cancer stem cells of various lineages [3840] (Table 1.1).

Second, high ALDH expression has been found to be associated with poor clinical outcome in leukemia [41], ovarian [4244], prostate [45, 46], breast [47

49], colorectal [50] and pancreatic cancer [51, 52]. Third, ALDH+ cells (cells with very high ALDH expression) exhibit a greater tumorigenic capacity, as reflected in colonyforming capability in vitro and in xenograftinduced tumor formation in vivo [53]. We have found very strong upregulation of ALDH1B1 expression in an animal model of colon polyps, specifically adenomatous polyposis coli

(Apc) multiple intestinal neoplasia (Apc Min ) mice (our unpublished data). These mice have a point mutation in Apc , a tumor suppressor gene which when mutated leads to dysregulation of the Wntsignaling pathway and results in up regulation of oncogenes like cMyc [54]. Overexpression of ALDH1B1 in polyps from these mice is suggestive of a possible relationship between Wntsignaling and ALDH1B1 expression, a consideration that warrants further study.

A causal relationship exists between alcohol consumption and CRC and this may be mediated, at least in part, by acetaldehyde [29, 52]. The significance of retinaldehyde and acetaldehyde in tumor formation, and very high expression of the ALDHs in colorectal cancer are suggestive of a crucial role for acetaldehyde and retinaldehyde metabolizing ALDHs in these cancers. Lack of

ALDH2 activity and resultant high acetaldehyde levels are linked with colon cancer initiation. By contrast, high ALDH1 activity (primarily ALDH1A1 and

ALDH1B1) is required for the stemness and tumorigenic potential of cancer stem cells.

Figure 1.1. ALDHs modulate carcinogenesis by metabolizing acetaldehyde and retinaldehyde. Ethanol is metabolized by alcohol dehydrogenase (ADH), catalase and CYP2E1 to acetaldehyde. Acetaldehyde can interfere with anti oxidative defense systems (AODS) and generate reactive oxygen species (ROS); inhibits DNA repair and methylation; and forms DNA and protein adducts to promote tumor growth. Acetaldehyde is metabolized to acetate primarily by ALDH2, ALDH1B1 and ALDH1A1. Retinaldehyde, formed from retinol by ADH, is converted to retinoic acid (RA) by retinaldehydemetabolizing ALDHs. RA exerts anticarcinogenic activity by binding to cellular retinoic acid binding proteins (CRBPII) and activating the RA receptor (RAR). When RA binds to fatty acid binding protein 5 (FABP5), it activates orphan nuclear receptor peroxisome proliferatoractivated receptor (PPAR)β/δ and acts as procarcinogenic agent. ALDH, aldehyde dehydrogenase; NAD +, NAD(P), nicotinamide adenine dinucleotide (phosphate); H 2O2, hydrogen peroxide.

Table 1.1: ALDH expression in various progenitor, stem and cancer cell types

S.No. Cell or tumor type ALDH isozyme(s) a Reference 1 Hematopoietic progenitor ALDH, ALDH1A3 [3840, 55, 56] 2 Mesenchymal progenitors ALDH [55] 3 Endothelial progenitors ALDH [55] 4 Neural stem cells ALDH, ALDH1L1 [57, 58] 5 Normal mammary stem ALDH1A1 [47] cells 6 Breast cancer stem cells ALDH1A1, [47, 49, 56, 59, ALDH1A3, ALDH2, 60] ALDH6A1, 7 Prostate cancer ALDH, ALDH7A1 [45, 46, 61] 8 Ovarian cancer stem cells ALDH, ALDH1A1 [4244, 62] 9 Ovarian cancer cells ALDH1A1, [63] ALDH1A3, ALDH3A2, ALDH7A1 10 Colon stem cells ALDH1A1, [12, 64] ALDH1B1 11 Colon cancer stem cells ALDH1A1, [12, 42, 50, 53, ALDH1B1 64, 65] 12 Leukemia stem cells ALDH [41] 13 Human lung cancer cells ALDH1A1 [42, 66, 67] 14 Head and neck cancer ALDH1A1 [68] stem cells 15 Pancreatic cancer ALDH, ALDH1A1, [51, 56, 69] ALDH1A3 16 Liver cancer stem cells ALDH, ALDH1A1 [70, 71] aALDH is designated for studies in which ALDH+ cells were identified and isolated using the Aldefluor ® assay.

Acetaldehyde: a carcinogen

Acetaldehyde is categorized as ‘carcinogenic to humans’ and ‘reasonably anticipated to be a human carcinogen according to IARC regulations and United

States National Toxicology Program (NTP), respectively [30, 72]. Acetaldehyde has been shown to be a highly toxic, mutagenic and carcinogenic compound in a variety of in vitro and in vivo studies. Its effects range from damaging antioxidant defenses [24] to interfering with DNA methylation and repair mechanisms through formation of adducts with DNA and proteins (Figure 1.1) [23, 29]. In the colon, acetaldehyde is primarily produced from ethanol by resident bacteria and, to a lesser extent, by mucosal ADHs. As a result of metabolism by intracolonic microbes, large quantities (ninefold higher than normal) of acetaldehyde accumulate in the rat colon 2 hours after intraperitoneal injection of ethanol [73].

Human colon mucosal cells harbor ADH1, ADH3 and ADH5, with the ADH1 and

ADH3 isozymes being most active [74]. In an in vitro experiment, human colon contents were able to generate 60 to 250 M acetaldehyde when incubated with a concentration of ethanol (10100 mg%), which is known to be attained during normal ethanol drinking [75]. The high levels of acetaldehyde attained in the colon after drinking ethanol likely underlies the correlation between chronic, heavy ethanol consumption and CRC in humans. In ethanoltreated rats, a high concentration of acetaldehyde (50 to 350 M) in the colon mucosa has been shown to correlate positively with hyperproliferation of the colon crypt cells. Such a phenomenon would be anticipated to favor the development of CRC [27, 76].

Acetaldehyde is metabolized primarily by mitochondrial ALDH2 and

ALDH1B1 and, to lesser extent, by cytosolic ALDH1A1 (Table 1.2) [11]. The most convincing evidence for a role of acetaldehyde in CRC initiation emanates from studies involving Asians who possess a polymorphism in their ALDH2 enzyme known as ALDH2*2. These subjects possess a single nucleotide polymorphism

(SNP) that leads to a lysine to glutamate substitution at residue 487 that renders the enzyme functionallyinactive [77, 78]. Approximately 40 per cent of the Asian population carry an ALDH2*2 allele; this compromises their ability to metabolize acetaldehyde and increases their colon cancer risk 3.4 times [37].

Table 1.2: Affinity of ALDHs for acetaldehyde and retinaldehyde

S.No. ALDH Substrate Km Reference isozyme(s) 1 ALDH1A1 Acetaldehyde 180 M [11] Alltrans 11.626.8 [79],[13] Retinaldehyde M 9cis Retinaldehyde 3.59 M 2 ALDH1A2 Alltrans 0.66 M [80] Retinaldehyde 0.62 M 9cis Retinaldehyde 3 ALDH1A3 Alltrans 0.2 M [81] Retinaldehyde 4 ALDH1B1 Acetaldehyde 55 M [11] Retinaldehyde 24.9 M [13] 5 ALDH2 Acetaldehyde 3.2 M [11] 6 ALDH8A1 9cis Retinaldehyde 3.15 M [82]

Opposing effects of retinoic acid on cancer cell proliferation

Retinoids exert many physiologicallyimportant and diverse functions in relation to cellular proliferation and differentiation of normal and cancer cells. For example, retinoids are crucial for embryonic development and adult tissue remodeling. The retinoids comprise all of the derivatives of retinol, including all trans , 9cis and 13cis retinoic acid. Retinol is oxidized to retinaldehyde by retinol dehydrogenases. The resultant retinaldehydes are further metabolized to their corresponding RA by retinaldehyde dehydrogenases which include

RALDH1 (ALDH1A1), RALDH2 (ALDH1A2), RALDH3 (ALDH1A3) and RALDH4

(ALDH8A1) (Table 1.2) [8287]. Among the RAs, alltrans RA (ATRA) is the most biologically potent retinoid. Abnormally low levels of ALDH1A2 have been observed in breast and prostate cancers [88, 89]. Impaired RA formation and high levels of CYP26A1 (a RAmetabolizing enzyme) in human breast cancer are consistent with a protective role for RA in this cancer [8890]. The physiological actions of the retinoids are mediated through binding of the RA receptor (RAR) and retinoid X receptor (RXR) heterodimer to the regulatory region of retinoid responsive genes, known as RA response elements [91]. RARs and RXRs are liganddependent transcription factors and exist as α, β or γ isoforms. RAR isoforms interact with both ATRA and 9cis RA, whereas RXR isoforms interacts only with 9cis RA [92, 93]. The binding of RA with the RAR/RXR dimer recruits coactivator proteins and initiates transcriptional activation of the retinoid responsive genes [91]. Retinoids have been found to be effective for the treatment of acute promyelocytic leukemia and prevention of liver, lung, breast,

9 prostate, skin and colon cancers [9496]. In vivo studies involving rats have revealed that retinoids added to the diet reduced colon cancer cell proliferation and prevented azoxymethaneinduced aberrant crypt foci (putative precancerous lesions in colon) and colon tumor formation [96, 97]. A RXRselective retinoid,

AGN194204, has been found to inhibit the proliferation of human pancreatic cancer cells, an effect that can be reversed by a RXRselective antagonist [98].

In addition to inhibiting the growth of pancreatic cancer cells, RA increases the sensitivity of pancreatic adenocarcinoma cells to the antineoplastic drugs gemcitabine and cisplatin [99].

In contrast to the antiproliferative and antisurvival role of RA in cancer cells, dietary ATRA has been shown to enhance initiation and growth of intestinal tumors in the Apc Min mouse model in vivo [100]. RA can promote cell survival and hyperplasia in cells expressing high levels of fatty acidbinding protein 5 (FABP5) by activating an orphan nuclear receptor, peroxisome proliferatoractivated receptor (PPAR)β/δ [101]. PPARβ/δ mediates antiapoptotic properties partly by inducing the PDK1/Akt survival pathway [102]. RA binds to intracellular lipid binding proteins (iLBPs), including cellular retinoic acidbinding proteins

(CARBPII) and FABP5. CARBPII and FABP5 are selective for nuclear receptors

RARα and PPARβ/δ, respectively [101]. Hence, RA induces CARBPII or

FABP5 mediated activation of RAR or PPARβ/δ (respectively), depending on the ratio of FABP5/ CARBPII in the cells [101]. Human colorectal cancer cell lines

(specifically, T84, COLO205, SW620, SW480, HCT116 and DLD1) express ~30 fold higher levels of FABP5 relative to normal colorectal cells (CCD18Co),

10 suggesting the possibility of proproliferative and antiapoptotic roles for RA in these cells [101, 103]. However, the expression levels of PPARβ/δ in colorectal cancer cells and its role in tumorigenesis are unresolved in various cancers, including CRC [104].

ALDH and cancer stem cells

In the gastrointestinal (GI) tract, tissuespecific stem cells are at the top of the cellular hierarchy and play a critical role in regulating tissue homeostasis.

These specialized epithelial cells are characterized by their ability to selfrenew and differentiate into a variety of cellular populations that perform specific functions within the GI tract. Currently, it is believed that these tissuespecific stem cells (or progenitor cells), when oncogenically transformed, become cancer stem cells (CSCs) or tumor initiatingcells (TICs) since they functionally possess the capacity to form tumors and maintain tumor growth. Accumulating evidence also suggests that CSCs are responsible for chemotherapeutic/radiation resistance and tumor recurrence (Figure 1.2).

ALDH catalytic activity has been identified in many human cancers [49] and, as such, is used as a marker of CSCs, including colorectal cancer. The pathophysiological function of ALDH in CSCs remains unresolved. Intense research of ALDH enzymes is underway in order to elucidate the role of these proteins in the development and progression of cancer as well as drug resistance.

Figure 1.2. ALDHexpressing cells are responsible for chemoresistance and relapse of many tumors after chemotherapy. Most current chemotherapy drugs are effective against the bulk of the tumor cells. However, the high ALDH expressing (ALDH+) cancer stem cells are resistant to these treatments. As a result, during chemotherapy, the ALDH+ cells proliferate and promote tumor growth. The resultant tumors contain an increased proportion of ALDH+ cells, making them more resistant to chemotherapy than the original tumor.

Cancer stem cells in colorectal cancer

Although earlier stages of CRC are highly curable, therapeutic interventions in advanced disease have proven to be poorly effective at increasing the 5year survival rate. Recent drug development has focused on targeting the CSC population as a potential therapy. In normal colon, CSC’s reside at the bottom of the crypt and generate upward, migrating and differentiating into transit amplifying cells (in the middle of the crypt) which become terminally differentiated cells as they move upward and eventually shed into the lumen (Figure 1.3A) [105].

In CRC, several different molecules, including the cell surface markers

CD133 and CD44 as well as ALDH activity, have been proposed as biomarkers for identification and isolation of the CSC population [53, 64, 106108]. CD133+ colon cancer cells were initially shown to be tumorigenic [107, 108]. However,

12 subsequent studies identified that both CD133+ and CD133 cells possess tumorigenic potential [109]. CD44+ (either with or without epithelialspecific antigen (ESA+)) was demonstrated to be a marker in colon CSCs [109].

However, additional studies showed that CD44+ cells reside throughout the entire crypt, including the proliferative compartment, suggesting that the CD44+ colon cells are not necessarily stemlike [64]. We have examined CD44 and

ALDH together in one of our CRC patientderived tumor xenograft (PDTX) models to determine if CD44+ cells had tumorigenic properties [110]. Despite

ALDH+/CD44+ cells showing some tumorigenic growth, ALDH+/CD44 cells exhibited a higher incidence and faster growing tumors. In this same PDTX model, isolation and injection of ALDH+ and ALDH cells in mice showed a significant difference with respect to tumor growth [110]. ALDH+ cells produced fast growing and large tumors when compared to ALDH cells that either produced very small tumors or no tumors in five separate PDTX models.

Importantly, all ALDH+ tumors looked morphologically the same as the original tumor. Several other studies have shown that injection of ALDH+ cells from colitis and colon cancer patients facilitated spheroid formation ( in vitro 3dimensional spheroid cell culture that more closely resembles the in vivo environment) and tumor growth in a xenograft model, while ALDH cells were incapable of tumor growth [53, 64]. These studies demonstrate that ALDH catalytic activity appears to be a robust marker of CSCs in CRC.

Given the apparent promise of ALDH activity as a potential biomarker of

CSCs, many investigations are currently exploring the role of ALDH in CSC

13 function. In particular, a great deal of focus is being placed on which ALDH isoform(s) mediate the catalytic activity in the CSCs. In normal colon stem cells,

ALDH1 has been demonstrated to be primarily expressed at the bottom of the crypt compartment in the colon (where colonspecific stem cells are located) and

ALDH1 levels are significantly elevated in the development and progression of

CRC [64]. Interestingly, ALDH1 protein levels are elevated in the colon of patients with ulcerative colitis (a risk factor for colon cancer) compared to normal colon cells; such expression may be important in the transformation from colitis to colon cancer [53]. We have shown that ALDH1B1 protein is 5.6fold higher when compared to ALDH1A1 in CRC patients and may be a potential biomarker in CRC (Figure 1.3BC) [12]. Similarly, very high expression of ALDH1B1 was found in the colon polyps of Apc Min mice (our unpublished data). While these studies indicate elevations in individual ALDH isoforms in CRC, the contribution of these enzymes to the progression of CSCs and CRC remain to be clarified.

A common problem associated with standard chemotherapeutic regimens in CRC is treatment resistance. Although chemotherapy is effective at reducing tumor burden, many CRC patients will experience disease recurrence and ultimately succumb to their disease. CSCs are thought to be responsible for chemotherapyresistance and disease recurrence [111]. Therefore, therapeutic elimination of this population would be predicted to reduce tumor recurrence and ultimately improve survival. In our CRC PDTX model, the effects of an inhibitor of the Notch pathway (considered to be important for selfrenewal of colon stem cells) in combination with irinotecan was investigated on the ALDH+ cell

14 population [110]. The combination therapy was effective at reducing the number of ALDH+ cells as well as tumor recurrence, even after treatment was discontinued when compared to single agent Notch pathway inhibition and irinotecan. Administration of the combination therapy for 28 days prevented tumor growth in the ALDH+ cell xenograft model; this protection continued for 3 months after combination treatment was completed [110]. These data indicate that the ALDH+ population has the ability to selfrenew, and significantly reducing this population of cells delays tumor recurrence (Figure 1.2). Whether specific

ALDH isozymes contribute to chemotherapy resistance remains to be determined.

Figure 1.3. ALDH1B1 expression pattern in normal colon and colon adenocarcinoma. Location of various cell types in normal colon ( A). ALDH1B1 expression (red arrows) is strictly localized to stemlike cells at the base of crypts in the normal human colon (B). ALDH1B1 is expressed at extremely high levels throughout all cells of human colon adenocarcinomas (C). In figures B and C (reproduced from Chen et al., 2011 [12]), lower panels are higher magnification of areas identified by squares in the upper panel. 15

ALDH isozymes in pancreatic functions and progenitor cells

Heavy alcohol consumption [112] and the inactivating polymorphism of

ALDH2, resulting into impaired metabolism of ethanol have been found to be associated with increased risk of type 2 diabetes mellitus (T2D) [113]. ALDH2 has also been found to protect from diabetes induced cardiomyopathy and retinopathy possible by preserving cell survival and metabolizing reactive aldehydes respectively [114, 115]. High ALDH expression has also been shown to be linked with diabetes associated large vessel disease [116]. These studies are indicative of the possible association of acetaldehyde metabolizing ALDH isozymes and T2D and associated complications. In the mouse, putative adult pancreas stem/progenitor cells have been isolated as cells with high ALDH activity from the centroacinar compartment and they demonstrate the capacity to generate endocrine and exocrine cells in culture [117]. Strikingly, experiments in a genetically engineered mouse model of pancreatic cancer indicated that cells in the same centroacinar location might be the cells of origin of pancreatic cancer

[118]. We have recently found that in mice, ALDH1B1 is strongly expressed in the pancreas progenitor cells of the pancreatic primordia. As pancreas differentiation proceeds by the growth and branching of the pancreatic epithelium, strong expression persists in the tips and the trunks where tripotent and endocrine pancreas progenitor cells reside [14]. Aldh1b1 expression is subsequently restricted to exocrine progenitor pools and this expression is lost before birth. Inhibition of ALDH activity in explant cultures of embryonic pancreata accelerates differentiation suggesting that Aldh1b1 may be acting

16 together with Aldh1a1 , the only other known ALDH gene expressed in the developing epithelium, in maintaining the progenitor populations [14, 119].

Aldh1b1 expression persists in very rare centroacinarlike cells in the adult and the number of Aldh1b1 + cells increases dramatically following two different pharmacological treatments that induce pancreas regeneration [14]. These findings suggest a role of ALDH1B1 in maintenance and expansion of pancreatic progenitor pool as well as in pancreatic regeneration.

Summary

There is accumulating evidence that supports a role for ALDHs in cancer development and progression. The exact mechanisms by which ALDHs influence tumorigenesis remain to be defined. Certainly, metabolism of acetaldehyde and/or the generation of RA represent modalities by which ALDHs could influence CRC. ALDH catalytic activity appears to be an excellent biomarker that can be utilized for the isolation and characterization of the CSC population in tumors obtained from patients with CRC. It is becoming apparent that the various

ALDH isozymes may have different roles in tumorigenesis (from metabolism of the carcinogen to modulation of the proliferationregulating retinoids) and that the timing and cellular localization of isozyme expression may be critical factors that influence how ALDHs modulate cancer development and progression. Further studies are needed that identify (i) the importance of ALDH catalytic activity in modulation of tumorigenesis, (ii) the specific ALDH isozymes involved (and that regulate CSCs), and (iii) ALDHassociated signaling pathways in cancer cells

17 and CSCs. The results obtained from such studies should lead to the development of novel therapies that may more effectively treat these devastating diseases.

Objectives

The goal of this project is to understand the role of mitochondrial

ALDH1B1 in alcohol metabolism, glucose homeostasis and colon carcinogenesis. Chronic alcohol abuse elicits a plethora of pathological outcomes, including damage to the liver and colon. Most of the adverse effects of ethanol are attributed to its metabolite acetaldehyde. Ethanol is primarily metabolized via oxidation to acetaldehyde, which has recently been classified as a “Group 1” human carcinogen. Acetaldehyde is metabolized to acetate by aldehyde dehydrogenases (ALDHs), primarily by mitochondrial ALDH2 and, to a lesser extent, by cytosolic ALDH1A1. We have recently shown that ALDH1B1 has a high affinity for acetaldehyde and appears to be the second major enzyme in acetaldehyde elimination. Several ALDH1B1 polymorphisms also exist in humans, and recent studies suggest an association between these polymorphisms and drinking aversion and more frequent alcohol hypersensitivity reactions in Caucasians. ALDH1B1 has also been found to be associated with maintenance and proliferation of pancreatic progenitor cells and pancreatic regeneration following injury. We have recently reported that ALDH1B1 protein is expressed specifically in the stem cell population of the normal human colon. In addition to detoxifying acetaldehyde, we found that ALDH1B1 metabolizes

18 retinaldehyde to RA, a reaction that could explain its role in stem cells.

Interestingly, we also found that ALDH1B1 protein is expressed at high levels in colorectal cancer, making it a potential biomarker.

Objectives of this thesis are to: 1) determine the physiological implication of ALDH1B1 in ethanol metabolism and glucose homeostasis using Aldh1b1(/) knockout mice model 2) determine role of ALDH1B1 in colon tumorigenesis and the involved mechanism.

CHAPTER II

CHARACTERIZATION OF ALDEHYDE DEHYDROGENASE 1B1 KNOCKOUT

MICE: PHYSIOLOGICAL IMPLICATION OF ALDH1B1 IN ETHANOL

METABOLISM AND GLUCOSE HOMEOSTASIS

Introduction

The aldehyde dehydrogenases (ALDHs) are involved in metabolizing a wide range of endogenous aldehydes and xenobiotics [24]. ALDH1B1 is a mitochondrial enzyme which was previously known as ALDH X or ALDH5 [10].

ALDH1B1 is the second most efficient enzyme (K m= 55 M) in metabolizing acetaldehyde after ALDH2 (K m= 3.4 M) [11]. ALDH2 is a polymorphic enzyme with ALDH2*2 as an inactive variant, which is found in up to 50% of the Asian population [24, 120, 121]. These carriers accumulate high concentrations of acetaldehyde and show ethanolinduced hypersensitivity, hypertension and flushing syndrome [122]. The ALDH2*2 variant is nearly absent in Caucasians

[21] but they carry inactive variants of ALDH1B1 [122]. In Caucasian populations the ALDH1B1 polymorphism is associated with the symptoms of acetaldehyde toxicity including ethanol hypersensitivity, hypertension and ethanol aversion

[122, 123]. Together, these findings are suggestive of a crucial role for ALDH1B1 in ethanol metabolism.

ALDH1B1 also shares the physiological role of ALDH1A1 by metabolizing retinaldehyde[13] and could be associated with stem cells like properties in

20 normal and cancer stem cells [124]. ALDH activity is used to identify and isolate normal and cancer stem cells [48, 53, 66]. Initially, human hematopoietic cells were found to be rich in ALDH activity but more recently, stem cells from other cancer types including bone marrow, ovary, breast, lung, pancreas, prostate and colon cancer have also been found to possess high ALDH activity [24, 38, 51,

66]. The Aldefluor ® assay (method to determine ALDH activity) is used to identify cancer stem cells and tumor initiating cells in various cancer types [40, 42]. So far, ALDH1A1 is considered a marker for these cells [47, 64, 66]. However, the

Aldefluor ® assay is not specific for ALDH1A1 activity and other ALDHs also contribute to this phenotype [12, 63, 125]. We compared the expression of

ALDH1A1 and ALDH1B1 in human colon cancer cells and found significantly high expression of the later suggesting ALDH1B1 as a potential biomarker for colon cancer[12]. High ALDH1B1 expression has also been found to correlate with poor prognosis in colorectal cancer patients in more recent studies [50, 125].

It has been shown that ALDH1B1 is strongly expressed in the early pancreatic buds in developing mice when compared to other retinaldehyde dehydrogenases (RALDHs) including ALDH1A1, ALDH1A2, ALDH1A3 and

ALDH8A1 [14]. With further development and differentiation, strong ALDH1B1 expression remains confined exclusively to tips and the trunk of the pancreatic epithelium and persist only in centroacinarlike cells by the time of birth [14].

ALDH1B1 + cells expand dramatically in adult mice pancreas following acute caeruleininduced pancreatitis [14]. Together, these findings indicate the role of

ALDH1B1 in pancreatic development and regeneration. The understanding of the

21 molecular mechanism about the possible role of ALDH1B1 in restoring pancreatic function would be important for the effective treatment of the diabetes.

To delineate the in vivo role of ALDH1B1 in ethanol metabolism and glucose homeostasis, we have generated a mouse line with the global disruption of

Aldh1b1 gene.

Materials and methods

Preparation of targeting construct and Generation of Aldh1b1 (-/-) mice

A genomic 9.1 kb XbaIKpnI fragment of mouse Aldh1b1 locus was isolated from C57BL/6J mouse genomic DNA using highfidelity PCR and subcloned into a pBluescript(II)KS vector. Using this clone, we constructed a targeting vector in which a LoxP flanked NEO cassette carrying an extra NsiI site disrupts Aldh1b1 exon 2 and removes the complete coding region of Aldh1b1 gene; for counterselection, a HSVTK gene was placed at the 5’end of the XbaI

KpnI fragment (Figure 2.1A). The Aldh1b1 gene was targeted in the EC7.1 hybrid

(Sv129/C57Bl/6Bl6) ES cell line by the Transgenic and Gene Targeting

Core (TGTC) at the University of Colorado. Three homologous recombinant ES clones out of 280 clones resistant to both G418 and ganciclovir were identified by

PCR and Southern blotting analysis (see below) (Figure 2.1B). These clones were subjected to karyotyping for detection of genetic alterations and the healthiest clone was selected for microinjection into nonagouti C57BL/6J

22 blastocysts to generate chimeric mice. The chimeric mice were then mated with

C57BL/6J female mice and resultant agouti Aldh1b1(+/) heterozygous animals were intercrossed to generate Aldh1b1(/) homozygous knockout animals. The resulting offspring were of Sv129 and C57BL/6J mixed background. The

Aldh1b1() allele was then backcrossed into C57BL/6J background for 10 generations. All studies were carried out in accordance with the University of

Colorado Anschutz Medical Campus Institutional Animal Care and Use

Committee (IACUC).

Southern blot and PCR analysis

Successful targeting in ES clones were identified first by PCR analysis and further confirmed by Southern blotting analysis (Fig. 2.1B). PCR screening in

ES cells were performed for both short arm and long arm homologous recombination. For Southern blotting analysis, genomic DNA was digested with

NsiI , blotted, and hybridized with a 32 Plabeled probe (600 bp) 5’ outside the region encompassed by the targeting construct. The band was visualized using a

Storm 860 Phosphorimager (Molecular Dynamics; Sunnyvale, CA). Aldh1b1(+) and Aldh1b1() alleles gave rise to a 7.9kb and 3.8kb band, respectively.

Genotyping in offspring was performed by PCR analysis using tail genomic DNA

(Fig. 2.1C). The Aldh1b1(+) allele was detected using forward primer 5’

ACACTGCAACAGGAGGACCAAGAA3’) and reverse primer 5’

ACATGCCCAATGACCTCACCT3’, generating a 429 bp product. The Aldh1b1() allele was detected using same forward primer as (+) allele and reverse primer

5’TTAAACGCGGCCGCCAATTGT3’, generating a 200 bp product.

Quantitative real time PCR (qRTPCR)

Total cellular RNA from selected tissues was extracted using TRI reagent

(Sigma, St.Louis) and further purified with an RNeasy Mini Kit (Qiagen) according to the manufacturer’s protocol. For qPCR analysis, cDNA was produced from equal amount of total RNA using Maxima FirstStrand cDNA Synthesis kit

(Fermentas, K1641), following manufacturer’s instructions. The mRNA levels of

Aldh1a1 , Aldh1b1 and Aldh2 were quantified using Power SYBR Green Master

Mix (Applied Biosystems) with β2 microglobulin ( B2M) as an endogenous control. QPCR analysis was performed on a 7500 Real Time PCR System

(Applied Biosystems) using primers (given in table 2.1) using thermal cycling conditions of 95 °C for 4 min, followed by 40 cycles of 95 °C for 30 s, 61 °C for

30 s for Aldh1b1 (55 °C for Aldh1a1 and 66 °C for Aldh2) , and 72 °C for 30 s.

The relative mRNA expression levels were calculated by 2 − ∆∆CT method [126], where (– ∆ ∆ C T )= – ( ∆ C T sample – ∆CT reference ); (∆ C T ) = CT (gene) – CT

(housekeeping gene).

Table 2.1: Primers for QPCR analysis

Gene Primer sequence (5'3') Tm/ ℃℃℃ name F: AGGCCCTCAGATTGACAAGGA 60.3 ALDH1A1 R: GTTGCACTGGTCCAAATATCT 57.6 F: AGCGCGATTCGGAGCCTCA 61.9 ALDH1B1 R: TGACCGCATCATGCCACTCGT 61.5 F: AGGTCTTCTGCAACCAGATCT 58.0 ALDH2 R: AGATGCATCCATGCGGCG 58.8 F: CATGGCTCGCTCGGTGACC B2M 61.2 R: AATGTGAGGCGGGTGGAACTG 60.0

F, forward primer; R, reverse primer; Tm, melting temperature of the primer

Western blot analysis

For Western blot analysis, mice were sacrificed using CO 2 asphyxiation and tissues were snap frozen immediately using liquid nitrogen. Frozen tissues samples (100 mg) were homogenized in lysis buffer and incubated on ice for 10 min followed by centrifugation at 10,000g for 10 min at 4°C. Supernatant was collected and total protein was estimated using Pierce ® BCA protein assay

(Thermo Scientific, Rockford, IL; 23223 and 23224) using the manufacturer’s

s buffer for 5 min and were subjected׳protocol. Proteins were boiled in Lammeli to 5% SDSPAGE gels, and then transferred to nitrocellulose membrane.

Membranes were blocked in TrisBuffered Saline with 0.1% Tween 20 (TBST) containing 5% (w/v) nonfat dry milk and probed overnight by incubation with primary antibodies according to the manufacturer's instructions and followed by

25 washing in TBST. ALDH1B1 was detected using rabbit polyclonal antiALDH1B1 antibody (1:5,000 in 5% nonfat dry milk in TBST) [12] and βactin was detected using mouse monoclonal antiβactin antibody (1:10,000; Sigma, St. Louis, MO).

Membranes were then incubated for 1 h at room temperature with horseradish peroxidaseconjugated goat antirabbit or antimouse secondary antibody

(1:5,000; Sigma, St. Louis, MO). After washing in TBST, membranes were incubated in Pierce ® ECL Western blotting substrate (Thermo Scientific Rockford,

IL; 32106), and exposed to Xray film to visualize protein bands. Protein expression was normalized to βactin expression.

Histopathological and Immunohisotochemical analysis

Various tissue samples were dehydrated in ascending concentrations of ethanol and cleared in xylene. Tissues were embedded in parafﬁn, 5 M sections were cut and sections were mounted on polyLlysine coated glass slides. These tissue sections were used for hematoxilin and eosin (H&E), periodic acidSchiff

(PAS) and immunohistochemical staining after deparaffinization using xylene, followed by rehydration in a graded series of ethanol. H&E stained slides were used to examine the histology of liver, lungs, kidneys, gastrointestinal tract, pancreas, ovaries, testes and uterus. For immunohistochemistry, antigen retrieval was performed in 10 mM/l citrate buffer (pH 6.0) and 3% hydrogen peroxide (v/v) was used to block endogenous peroxidase activity. The sections were incubated with TNB blocking buffer (PerkinElmer, Waltham, MA). Sections

26 were incubated with rabbit polyclonal antiALDH1B1 antibody [11], rabbit monoclonal antiKi67, antiinsulin and antiglucagon antibody (Abcam,

Cambridge, MA) at a 1:750 dilution in TNB blocking buffer overnight in humidified chamber. TNB buffer was used as a negative staining control. HRPconjugated antirabbit secondary antibody was used at a 1:500 dilution in TNB buffer for 60 min and TSA biotin signal amplification system (PerkinElmer, Waltham, MA) was used following manufacturer’s protocol. Sections were visualized by incubation in

DAB working solution (Vector laboratories, Burlingame, CA) or AEC substrate kit

(BD Pharmingen, San Jose, CA) for 10 min at room temperature and counterstained with diluted hematoxylin (1:10 with distilled water) for 2 min.

Finally sections were dehydrated and mounted with Permount (Sigma, St. Louis) for microscopic examination. Slides with AEC were mounted without dehydration using aqueous mounting medium (Sigma, St. Louis).

Ethanol administration and blood collection

Eight w old Aldh1b1 (/) and Aldh1b1 (+/+) littermate (Sv129 and

C57BL/6J mixed background) mice (n=6) were matched for body weight and administered with a single i.p. dose of 20 percent ethanol (5g/kg). Blood was then collected at 0, 1, 3 and 24 h from right atrium using a heparincoated ice cold syringe, the blood was mixed immediately with pre cooled 0.6 N perchloric acid (PCA) solution. Tubes were weighed before and after blood collection to measure the exact amount of blood samples and samples were processed for

GCMS analysis soon after collection.

Figure 2.1. Generation of Aldh1b1(-/-) mouse line.

Figure 2.1. Generation of Aldh1b1(-/-) mouse line. (A) Structures of Aldh1b1 locus, targeting construct and of the mutant allele resulting from the homologous recombination. Although Aldh1b1 has two exons, the entire coding region is located on exon2, making this an intronless gene. Neomycinresistance minicassette ( NEO ) and herpes simplex virus thymidine kinase minicassete (HSVTK ) were used as selection markers. Arrow direction of the NEO and HSVTK minicassetes represents the 5’ to 3’ orientation of those genes. ( B) Screening of three ES clones for homologous recombination by PCR and Southern blotting; NC, negative control, (C) PCR analysis of DNA extracted from tails of offspring from Aldh1b1 (+/) heterozygous crossing. Molecular weight marker (MW), 1 Kb plus DNA ladder (Invitrogen, NY).

Ethanol and acetaldehyde concentration measurement

The PCA solution containing blood was centrifuged at 15000g for 10 min at 0°C and 300 l of the supernatant was transferred into a 10 ml glass vial in the cold room and sealed with a gastight cap. Ethanol and acetaldehyde concentrations were measured according to the previously described procedure by gas chromatography mass spectrometry (GCMS) [127, 128]. The vial with supernatant was placed in a heating block at 65°C for 10 min; the headspace gas was then transferred using a Combi PAL auto sampler (CTC Analytics, Zwingen,

Switzerland) to the GCMS (QP2010, Shimadzu, Kyoto, Japan). Ethanol and acetaldehyde were separated on a 60 m x 0.25 mm inner diameter AQUATIC capillary column (GL Sciences, Tokyo) with 1 M film thickness. The injection port temperature was kept at 200°C and column oven temperature was programmed from 40°C to 50°C at the rate of 2°C per minute and from 50°C to

170°C at the rate of 40°C per minute. Helium was used as carrier gas at a flow rate of 1.0 ml per minute. For acetaldehyde, the split ratio of the carrier gas was

20:1 and mass spectrometry (MS) was carried out by electron impact ionization at 300 eV. For ethanol, split ratio was 100:1 and MS was carried out by electron impact ionization at 70 eV. The interface and ion chamber temperatures were maintained at 230°C and 210°C, respectively. A selected ion monitor (SIM) mode was set at 43 and 29 m/z for acetaldehyde and 45 and 46 m/z for ethanol.

Glucose tolerance test

Six male mice (C57BL/6J background) from each group ( Aldh1b1 (/) and

Aldh1b1 (+/+)) were used for i.p. glucose tolerance test (GTT). Animals were fasted overnight with water available ad libitum . Mice were anesthetized with isoflurane anesthesia before collecting blood from the tail vein. Mice were injected with 15% sterile Dglucose (1.5 mg/g, i.p.) and blood samples were collected from a nick on the tail immediately prior to and 15, 30, 60 and 120 min after glucose administration. Blood glucose levels were determined using a glucometer (OneTouch Ultra, One Touch). Area under the curve (AUC) for glucose levels was calculated using trapezoidal estimation method using the results obtained in the GTT.

Statistical analysis

All quantitative experiments were performed at least in triplicate, and the data shown are the means ± S.E. Statistical significance was determined using

Student’s ttest using SigmaStat Statistical Analysis software (SPSS Inc.,

Chicago, IL). P < 0.05 was considered to be significant.

Results

Generation of Aldh1b1 (-/-) mice

The targeting construct was designed to disrupt the exon2 (2 Kb) of

Aldh1b1 gene, resulting in the removal of the complete coding region of Aldh1b1 gene (Figure 2.1A). Independent ES clones harboring successful homologous

31 recombination were confirmed by Southern blotting and PCR analysis (Figure

2.1B). Male chimeras generated from these ES clones were crossed with

C57BL/6 females to generate heterozygotes for the Aldh1b1() mutant allele.

Intercrossing of heterozygotes produced Aldh1b1 (+/+) wildtype , Aldh1b1 (+/) heterozygous and Aldh1b1 (/) knockout littermates (Figure 2.1C). Offspring of the three genotypes were born with expected Mendelian frequencies, suggesting no embryonic lethality due to the global disruption of Aldh1b1 gene.

Phenotype of Aldh1b1 (-/-) mice

Mice homozygous for the disrupted Aldh1b1 () allele showed no expression of Aldh1b1 mRNA in liver and colon tissues by qRTPCR analysis

(Figure 2.2A). To see if there are any compensatory changes in other closely related ALDHs, we examined the mRNA expression of Aldh1a1 and Aldh2 in these organs. No difference in mRNA expressions of these ALDH enzymes was observed between Aldh1b1 (/) and Aldh1b1 (+/+) mice (Figure 2.2A). Loss of

ALDH1B1 protein in Aldh1b1 (/) mice was confirmed by Western immunobloting

(Figure 2.2B) and immunohistochemistry (Figure 2.2C). ALDH1B1 immunostaining was present in the bottom of the crypt of small intestine and colon from the wildtype (WT) animals but was lacking in those from the

Aldh1b1(/) mice (Figure 2.2C). This pattern of ALDH1B1 expression in intestinal tissues is consistent with what we reported in human tissues [12].

Figure 2.2. Expression analysis of ALDH1B1 mRNA and protein in Aldh1b1(-/-) mice. (A) Comparison of liver and colon Aldh1b1 , Aldh1a1 and Aldh2 mRNA expression levels by real time PCR analysis between Aldh1b1(+/+) and (/) mice. The level of mRNA expression was normalized to β2 microglobulin (B2M) mRNA expression and expressed as ratio of that found in the Aldh1b1(+/+) mice (=1). Data are presented as mean ± SE; ND, not detectable; n=4 mice. ( B) Western blot analysis of liver, colon and small intestine lysates using antiALDH1B1 antibody revealed complete loss of ALDH1B1 protein. ( C) Photomicrographs of mouse colon and small intestine after immunostaining with antiALDH1B1 antibody. Cells at the crypt base showed ALDH1B1 reactivity in Aldh1b1(+/+) but not in Aldh1b1(/) mice. Representative images are presented at two magnifications, with the squared field in the top panel (100x) enlarged in the bottom panel (400x). 33

We observed these mice until 6 mo of age and found them to be overtly healthy. We also documented the growth curve of Aldh1b1 (/) and Aldh1b1

(+/+) WT littermates. Male Aldh1b1 (/) and Aldh1b1 (+/+) mice did not show any difference in body weight. However female Aldh1b1 (/) mice showed lower body weights between weeks 3 and 5: after that, they caught up the growth rate with their WT littermates (Figure 2.3). Both Aldh1b1 (/) male and female mice were fertile (data not shown). Histological analysis revealed no evidence for morphological abnormalities among the organs examined including liver, lung, pancreas, ovaries, uterus, small intestine and large intestine. Photomicrographs of representative tissues are presented in Figure 2.4.

Figure 2.3. Growth curve of Aldh1b1(+/+) and Aldh1b1(-/-) mice. Body weight of Aldh1b1(+/+) and Aldh1b1(/) mice as a function of age. The data presented are mean ± SE. *, P < 0.05. Students ttest, compared to levels in WT mice at same time points (B) Representative H & E staining of the sections from small intestine, colon, rectum, lung, testes and uterus from Aldh1b1(+/+) (top panel) and Aldh1b1(/) (bottom panel) mice revealed similar cytoarchitecture in both the groups.

Figure 2.4. Aldh1b1(-/-) mice have normal cytoarchitecture. Representative H & E staining of the sections from pancreas, liver, small intestine, colon, rectum, lung, testes and uterus from Aldh1b1(+/+) (left panel) and Aldh1b1(/) (right panel) mice revealed similar cytoarchitecture in both the groups. 35

Since ALDH1B1 was expressed selectively at the bottom of the crypt in the colon (i.e., putative stem cell compartment) and is a potential biomarker for colon cancer, we examined these organs for Ki67 immunostaining, a marker for cellular proliferation. We did not find significant difference in intensity or extent of the immunopositivity for Ki67 between colons from Aldh1b1(/) and WT mice

(Figure 2.5). To examine the effect of ALDH1B1 knockdown on the differentiation, we examined and quantified goblet cells following PAS staining.

We did not find any difference in the number of goblet cells in colon crypts from

Aldh1b1(+/+) and Aldh1b1(/) mice (Figure 2.6).

Loss of ALDH1B1 leads to decreased clearance of blood acetaldehyde

After intraperitoneal administration of 5 g/ Kg ethanol, the blood ethanol and acetaldehyde levels were determined by GCMS at 0, 1, 3 and 24 h.

Representative chromatogram for blood ethanol and acetaldehyde are shown in

Figure 2.7A. Ethanol and acetaldehyde were distinguished by different retention time and identical ions, ruling out their interference in each other’s measurement

(Figure 2.7A). A best ﬁt second order polynomial function was used to generate calibration curves for ethanol and acetaldehyde (Figure 2.7B). There were no differences in the blood ethanol levels in Aldh1b1(/) mice relative to their age and weightmatched Aldh1b1(+/+) mice (Figure 2.7C, left panel). However, the blood acetaldehyde levels after 3 and 24 h of ethanol administration as well as area under the curve (AUC) for circulating acetaldehyde levels were significantly

36 higher in the Aldh1b1(/) mice, indicating crucial role for this enzyme in the acetaldehyde clearance (Figure 2.7C, right panel).

Figure 2.5. Lack of ALDH1B1 does not affect number of proliferating cells in colon. (A) Photomicrographs of mouse colon after immunostaining with anti Ki67 antibody revealed no difference in number of proliferating cells between Aldh1b1(+/+) and Aldh1b1(/) mice. Representative images are presented at two magnifications, top panel (100x) and bottom panel (400x). (B) quantification of Ki67 positive cells in proximal and distal colon crypt from Aldh1b1(+/+) and Aldh1b1(/) mice.

Figure 2.6 Lack of ALDH1B1 does not affect number of goblet cells in colon. (A) Proximal and distal colonic crypts were stained for goblet cells using periodic acidSchiff (PAS) stain that detect mucus secreting cells and these were quantified for Aldh1b1(+/+) and Aldh1b1(/) mice (B). Bars represent mean ± SEM (Student’s ttest), n = 3 mice per genotype.

Figure 2.7. Pharmacokinetics of ethanol and acetaldehyde in Aldh1b1(-/-) mice. (A) Representative total ion chromatogram and selected ion monitoring profile of a blood sample for ethanol and acetaldehyde using GCMS. Ion structure was set at 45 and 46 m/z for ethanol and 29, and 43 m/z for acetaldehyde. (B) Standard curve for ethanol (left panel) and acetaldehyde (right panel) measurement. ( C) Blood ethanol and acetaldehyde levels in Aldh1b1(+/+) (closed symbol) and Aldh1b1(/) (open symbol) mice after ethanol administration (5 g/kg, i.p). Area under the curve (AUC) is presented at the upper right corner. *, P < 0.05; data represent Mean±SEM (n=6); **, P < 0.001, Student’s ttest, compared to levels in Aldh1b1(+/+) mice at same time points.

Aldh1b1(-/-) mice have compromised glucose homeostasis

Since ALDH1B1 is crucial for embryonic development of pancreas, we performed an i.p GTT to evaluate the effect of ALDH1B1 deletion on systemic glucose homeostasis. Aldh1b1(/) mice showed elevated fasting blood glucose levels accompanied by elevated blood glucose levels following i.p. glucose administration (Figure 2.8AB). Higher AUC for glucose in the Aldh1b1(/) mice was recorded following i.p. GTT ( Figure 2.8 A and C). However, we did not observe any difference in the pancreatic cytoarchitecture or immunostaining for insulin and glucagon in pancreatic islets of Aldh1b1(/) and Aldh1b1(+/+) mice

(Figure 2.9).

Discussion

Aldh1b1 gene has two exons and an intronless coding region completely confined to exon2 [11, 129]. We have generated homozygous Aldh1b1(/) mice by deleting the entire coding region of the Aldh1b1 gene and backcrossed them into a C57BL/6 background. These mice did not show any reproductive, developmental or anatomical abnormalities, indicating that ALDH1B1 is dispensable for development and survival. This may be explained by the overlapping physiological functions of ALDH1B1 with other ALDHs like ALDH2

(e.g., acetaldehyde metabolism) and other ALDH1 isozymes (e.g., retinaldehyde metabolism) [1, 24].

Figure 2.8. Intraperitoneal glucose tolerance test (GTT) in Aldh1b1(-/-) mice. (A) Overnight fasted mice were given an i.p. injection of 15% glucose (1.5 mg/g of body weight). Blood samples were collected from tail vein before (0) and 15, 30, 60 and 120 min after glucose administration and analyzed for glucose concentration. Histograms showing fasting blood glucose ( B) and area under the curve (AUC) for GTT ( C) in Aldh1b1(/) and Aldh1b1(+/+) mice. Each data point represents mean ± SE (n=6 mice); *, P < 0.05; **, P < 0.001. Student’s ttest, compared to levels in Aldh1b1(+/+) mice at same time point.

Figure 2.9. Photomicrographs of mouse pancreas after immunostaining for insulin and glucagon. Pancreatic islets of Aldh1b1(/) and Aldh1b1(+/+) mice showed similar staining for insulin and glucagon. Representative images are presented at 100x magnification.

The lower body weight of female (but not male) Aldh1b1(/) mice without clinical or histological correlates is an intriguing finding, especially considering there was no weight difference in these female mice after 6 weeks of age.

Further research is required to confirm whether the ALDH1B1 knockdown has any possible effects during early development in female mice or to rule this out as a spurious result. ALDH1B1 has been suggested to be a biomarker for colon cancer and Ki67 is a nuclear protein found in proliferating cells [130], however we did not find any difference in the expression of this protein in colon of

Aldh1b1(/) mice suggesting that there is no difference in the number and distribution of actively proliferating cells in the colon of these mice.

The mouse Aldh1b1 gene encodes a protein with 519 amino acids that has highest expression level in the liver and parts of small intestine (ileum and jejunum) and moderate expression in the colon [11]. The expression pattern of

ALDH1B1 is similar to that of ALDH2, the most efficient acetaldehyde metabolizing enzyme [11, 131]. Since ALDH1B1 is second only to ALDH2 in metabolizing acetaldehyde [11] and has been found to be associated with ethanolinduced hypersensitivity [122] and ethanolrelated hypertension [123] in humans, we hypothesized that Aldh1b1(/) mice may have reduced acetaldehyde clearance. We did not find any difference in blood ethanol levels in

Aldh1b1(/) mice relative to their Aldh1b1(+/+) littermates after i.p. ethanol. This was expected because this enzyme is not involved in conversion of ethanol to acetaldehyde. However, despite normal levels of ALDH2 (the principle

43 acetaldehyde metabolizing enzyme), Aldh1b1(/) mice exhibited higher blood acetaldehyde levels after ethanol administration. These findings corroborate the important contribution of ALDH1B1 to acetaldehyde clearance and also helps explain the hypersensitivity observed in the human population with SNP encoding ALDH1B1 (rs2228093) [122], which is a slow acetaldehyde metabolizer relative to WT ALDH1B1[13]. This raises an important question: why is it that

ALDH1B1 does not offer compensation for the acetaldehyde metabolism in people carrying ALDH2*2 variant, i.e, those having the inactivating ALDH2 mutation [77, 132135]? This could be explained by the recent finding based on computational molecular modeling that suggests that ALDH1B1 may form a heterotetramer with inactive ALDH2 mutant subunits, the result of which would be decreased ALDH1B1 catalytic activity [78].

High ALDH activity is used to isolate putative progenitor cells from the centroacinar compartment of adult mice pancreas; these cells are capable of generating endocrine and exocrine cells in culture [117]. Recently, we have found that ALDH1B1 is expressed in pancreatic progenitor cells of the developing embryo and in adult centroacinar cells in mice [14]. On the basis of these findings, we anticipated Aldh1b1 (/) mice would have compromised glucose homeostasis. Our results revealed Aldh1b1 (/) mice to have higher fasting glucose and poor glucose tolerance, suggesting an important role for this enzyme in maintaining glucose homeostasis. However, pancreatic islets architecture in Aldh1b1 (/) mice was ostensibly similar to Aldh1b1(+/+) mice.

The glucose intolerance could be caused by a change in the insulin levels. Also,

44 overtly normallooking islet with normal islet volume and βcell area can have altered insulin secretion and/or islet microvascular morphology resulting into hyperglycemia and glucose intolerance [136, 137]. Further studies are required to ascertain the cause of compromised pancreatic function in Aldh1b1 (/) mice.

In summary, we showed that Aldh1b1(/) mice have slow acetaldehyde clearance and glucose intolerance. Slow acetaldehyde clearance support the contribution of ALDH1B1 in metabolizing acetaldehyde along with ALDH2.

Glucose intolerance in Aldh1b1(/) corroborates our earlier finding suggesting role of ALDH1B1 in pancreatic development and confirms its role in the maintaining glucose homeostasis.

CHAPTER III

ALDH1B1 IS CRUCIAL FOR COLON TUMORIGENESIS BY MODULATING

WNT/βCATENIN, NOTCH AND PI3K/AKT SIGNALING PATHWAYS

Introduction

Colorectal cancer is the fourth most commonly diagnosed cancer and second leading cause of cancer related deaths in the United States with 136,830 new cases and 50,310 deaths estimated for year 2014 [16]. Effective prevention and treatment of colorectal cancer are dependent upon a comprehensive understanding of risk factors and the molecular mechanisms underlying disease progression. Disruption of several oncogenic signaling pathways has been found to be involved in colon cancers, with Wnt/βcatenin being the most significant pathway. Mutations that result in constitutive Wnt/βcatenin signaling pathway activation are believed to play a crucial role in colorectal carcinogenesis [138,

139]. For example, eightyfive percent of human colorectal cancer cases are associated with deleterious mutations within the APC gene [140, 141]. In cases where the APC gene is not mutated, mutations in βcatenin or other Wnt/β catenin signalingrelated genes are reported and these mutations promote stabilization, nuclear translocation and constitutive activation of Wnt/βcatenin pathway [142]. There is a large body of evidence suggesting the significance of

Wnt/βcatenin signaling activation for the maintenance of proliferative state of colon crypt cells. The inactivation of Wnt signaling by deleting βcatenin or by using Dkk1 (Wnt inhibitor) results into substantial loss of proliferative epithelial

46 cells [143145]. In addition, overactivation of Wnt/βcatenin signaling is associated with an increase in the number of proliferating epithelial cells and prevention of differentiation in small intestine and colon [146148]. Other important pathways include the Notch, phosphoinositide3kinase (PI3K)/Akt, mitogen activated protein kinase (MAPK) and TGF β signaling pathways [149].

Notch signaling is essential for maintaining normal intestinal homeostasis by being crucial for cell fate and playing an important role in regulating cell proliferation, differentiation and apoptosis [150, 151]. Dysregulation of Notch signaling has a synergistic effect with Wnt signaling activation and positively regulates colon cancer development [151, 152]. The role of PI3K/Akt pathway in colon cancer is evident by the loss of phosphatase and tensin homolog deleted on chromosome 10 (PTEN), a negative regulator of this pathway in approximately 30 % of the CRC cases [153]. Inactivating mutations in the TGF beta type II receptor (TGFBR2) which block TGFβ signaling are associated with approximately 30 % of colorectal cancers [154].

ALDH1B1 as possible stem cell marker

ALDH activity has been identified as a biomarker for many cancers and cancer stem cells [24, 155]. ALDH1B1 is a relatively unexplored member of the

ALDH superfamily. It shares 62% protein identity with ALDH1A1, an ALDH that has garnered much attention recently as a biomarker of cancer stem cells [64,

156, 157]. ALDH1B1 metabolizes retinaldehyde to generate RA [13], a vitamin A derivative necessary for cell growth and development [7, 24]. High Wnt activity

47 is a distinguishing feature of colon cancer stem cells and is confined to the stem cell compartment of the normal colon [158]. We have recently shown that

ALDH1B1 expression is strictly localized to stemlike cells at the base of crypts in the normal human colon. In contrast, extremely high ALDH1B1 expression was observed throughout the cells of human colon adenocarcinomas [12]. These results indicate a close association between activation of Wnt/βcatenin signaling and overexpression of ALDH1B1.

Wnt/βcatenin signaling pathway activation

The Wnt/βcatenin signaling pathway plays an important pathophysiological role in colorectal carcinogenesis by driving the transformation and tumorigenic progression of colonic epithelial cells [142, 159, 160]. In the absence of Wnt signaling, βcatenin forms a complex with adenomatous polyposis coli (APC), axin and glycogen synthase kinase (GSK3β) and is phosphorylated by GSK3β. The resultant phosphoβcatenin undergoes proteosomal degradation after ubiquitination (Figure 3.1A) [161, 162]. As a result of such continuous elimination from the cytoplasm, nuclear translocation of β catenin is prevented and Wnt target genes are repressed. The Wnt/βcatenin pathway is activated by the binding of Wnt ligand to its transmembrane frizzled receptor and coreceptor lipoprotein receptor related protein 5 or 6 (LRP5/6).

This complex recruits dishevelled (Dsh), resulting in the phosphorylation of LRP6 and recruitment of axin to the receptor complex. The interaction of axin with the receptor complex prevents axindependent phosphorylation and degradation of

βcatenin [142, 162]. Free βcatenin translocates into the nucleus where it binds with T cell factor (TCF)/ lymphoid enhancer factor (LEF) transcription factors

[163]. The TCF/LEF complex are DNA binding factors associated with TCF/LEF binding elements (TBEs) within the promoter [164]. Activation of the Wnt/β catenin signaling pathway results in an increase expression of genes involved in cell proliferation and differentiation (e.g., CMyc , CyclinD1 ) [142]. In the healthy colon, such activation is normally confined to stem or progenitor cells.

Notch pathway in colon cancer

The Notch signaling has also been shown to play important role in regulating proliferation, differentiation and determination of cell fate in various organs including intestinal mucosa [150, 165]. There are four notch receptors

(Notch14) and five ligands (Jagged1, Jagged2, Deltalike (DLL)1, DLL3, and

DLL4) [150, 166]. Activation of this pathway occurs by the interaction between a

Notch ligand and receptor on adjacent cells. This binding results in the cleavage of the intracellular Notch (ICN) receptor by the γsecretase complex resulting in the release of ICN from the cell membrane and subsequent translocation into the nucleus [166, 167]. Once in the nucleus, ICN binds with CSL [CBF1/RBPJk, Su

(H), Lag1], which activates the transcription of target genes, such as Hairy and enhancer of split (HES)1, and HES5 [166]. Inhibition of Notch signaling results into increased differentiated cells and loss of proliferation in intestinal crypt epithelial cells [168]. Conversely, Notch signal activation leads to inhibition of differentiation and amplification of proliferating cells in the intestinal crypt [169].

Jagged1, a Notch ligand, has been shown to be directly controlled by Wnt signaling and so the Notch pathway is indirectly regulated by Wnt/βcatenin signaling [152].

Retinaldehyde and PI3K pathway

The PI3K/Akt signaling pathway also plays a crucial role in colon cancer development and maintenance by regulating cell survival, cell cycle progression and cellular growth [170, 171]. These findings suggest that, like the Wnt pathway,

Notch and PI3K/Akt signaling pathways are also crucial for maintaining undifferentiated and proliferative status of colon crypt epithelial cells [172174].

Given the differential expression pattern of ALDH1B1 in normal and cancerous colon tissues, which correlates with that of the Wnt/βcatenin pathway

[158], and the metabolic capacity of ALDH1B1 towards retinaldehyde [13], we hypothesize that ALDH1B1 plays an important role in colon tumorigenesis. In this study, we tested this hypothesis using human colon cancer cell lines. We demonstrate that ALDH1B1 is not only a potential colon cancer biomarker, but may also significantly contribute to the development of tumors by modulating canonical Wnt/βcatenin, Notch and PI3K/Akt signaling pathways.

Materials and methods

Cell culture

SW480, COLO320, HCT116 and HT29 human colon cancer cells (ATCC,

Manasss, VA) were cultured in RPMI 1640 medium (Life Technologies, Carlsbad,

CA) supplemented with 10% fetal bovine serum (Sigma, Saint Louis, MO), 50 penicillin (100 units/ml), and streptomycin (100 g/ml) (Life Technologies,

Carlsbad, CA) at 37°C in a humidified 5% CO 2 atmosphere.

Computational analysis of the ALDH1B1 promoter region

The 3kb promoter region (1 to 3000) of human ALDH1B1 gene

(GenBank accession No. NM_000692.4) was analyzed for candidate TCF/LEF binding elements using the Transcription Element Search System (TESS)

(www.cbil.upenn.edu/cgibin/tess/) program.

Luciferase reporter assay

DNA fragments corresponding to ALDH1B1 3 Kb promoter region from nucleotides 3064 to +115 (numbering relative to the transcription start site) were amplified using human genomic DNA (Promega, Madison, WI; G3041). The primers for PCR of ALDH1B1 promoter were F (5’

TTAAGGTACCTTCAATCCACACAGGCTCCA3’) (introduced Kpn I as underlined) and R (5’GTTAAGCTAGCGTTAGTTACCTGCAGCAGG3’). PCR conditions for this reaction consisted of a denaturing step at 98°C for 30 s and 30 cycles at 98°C for 5 s, 72°C for 50 s and 72°C for 50 s. Resultant PCR products were purified using QIAEX II gel extraction kit (Qiagen, #20021). The promoter fragments and pGL3Enhancer vector were digested with KpnI and NheI and ligated into pGL3Enhancer vector to generate ALDH1B1pGL3 construct. Using sequential deletion of the putative TBEs from the ALDH1B1pGL3 construct, four deletion constructs were generated containing: TBE1 to TBE4 (2381/+115 bp);

TBE1 and TBE2 (1678/+115 bp); TBE1 (1206/+115 bp) and no TBE (492/+115 bp). The reporter constructs were confirmed by restriction digestion as well as sequencing using GL2 and RV3 primers (Promega, Madison, WI). Promoter activity for cMyc was examined using pBVluciferase vector with cMyc promoter; pBV empty vector was used as a control (Addgene, Cambridge, MA) [175].

SW480 cells were transiently transfected with the reporter constructs or control plasmid (empty pGL3enhancer or pBVvectors) using FuGENE HD transfection reagent (Promega, E2311). Briefly, single cell suspensions of

SW480 cells were seeded in a 12well plate (2.5x10 4 cells/ well) and 2 g construct was added to each well with 0.2 g pRLTK vector (Promega) as an internal control for transfection efficiency. Cells were lysed after 24 h using passive lysis buffer (Promega) and luciferase activity was measured in lysates using DualLuciferase ® Reporter Assay System following the manufacturer’s protocol (Promega, E1910). Luminescence was detected using the Synergy 2 plate reader (BioTek, Winooski, VT). This experiment was repeated three times and averaged.

ALDH1B1 shRNA knockdown

ALDH1B1 shRNA in pRFPCRS plasmid vector and pRS vectornegative control (scramble) were purchased from OriGene (Rockville, MD, USA, #

TF314844). Stable clones were generated by transfecting the SW480 cell line in a 6well plate with 1 g of each of the shRNA plasmids using FuGENE HD transfection reagent (Promega, Madison, WI, USA) according to the

52 manufacturer’s recommendations. Seventytwo h after transfection, the cells were placed under selection by inclusion of 2 g /ml puromycin into the medium.

Multiple clones from the same transfection were pooled and grown under puromycin selection. Successful knockdown of specific genes and gene products were confirmed by semiquantitative reverse transcription PCR and Western blot.

Luciferase reporter assay for Wnt/βcatenin activity

Scramble or ALDH1B1 shRNA transfected and nontransfected SW480 cells were plated in 12 well plates to achieve about 50% confluency in 24 h. RKO colon cancer cells were used as negative control. Cells were then transiently transfected using FuGENE HD transfection reagent (Promega) according to manufacturer’s protocol with either 2.5 g TOPflash, or FOPflash reporter plasmid (Millipore, Temecula, CA) along with 0.2 g pRLTK vector (Promega) as an internal control for transfection efficiency. The pRLTK vector is a wildtype renilla luciferase control reporter vectors for dual luciferase reporter assay. The

TOPflash plasmid contains TCF4 binding sites upstream of the luciferase gene, which is responsive to the presence of active Wnt/βcatenin signaling, whereas the FOPflash plasmid contains mutated TCF4 binding sites. Cells were lysed with

Reporter Lysis Buffer (Promega) and luciferase and renilla luminescence were measured in a 96well plate using the DualLuciferase Reporter Assay System

(Promega) on a Synergy 2 plate reader (BioTek microplate reader and Gen5 software, Winooski, VT). The ratio of TOPflash/FOPflash signal was calculated and normalized to control conditions.

Immunohisotchemical analysis

Colon tissue sections from APC Min mice were kindly provided by Dr.

Jeffrey Peters (Pennsylvania State University). These tissue sections were used for immunohistochemical staining after deparaffinization using xylene, followed by rehydration in a graded series of ethanol. Antigen retrieval was performed in

10 mmol/l citrate buffer (pH 6.0) and 3% hydrogen peroxide (v/v) was used to block endogenous peroxidase activity. The sections were incubated with TNB blocking buffer (PerkinElmer, Waltham, MA). Sections were incubated with rabbit polyclonal antiALDH1B1 antibody [11] at a dilution of 1:750 in TNB blocking buffer overnight in humidified chamber and TNB buffer was used as a negative staining control. HRPconjugated antirabbit secondary antibody was used at a dilution of 1:500 in TNB buffer for 60 min and TSA biotin signal amplification system (PerkinElmer, Waltham, MA) was used following manufacturer’s protocol.

Sections were visualized by incubating with DAB working solution (Vector laboratories, Burlingame, CA) for 10 min at room temperature and counterstained with diluted hematoxylin (1:10 with distilled water) for 2 min. Finally, sections were dehydrated in ascending concentrations of ethanol and mounted with

Permount (Sigma, St. Louis) for microscopic examination.

Western blotting

Cells were harvested at 80 to 90% confluency using RIPA lysis buffer supplemented with protease inhibitors, PMSF, and sodium orthovanadate (Santa

Cruz Biotechnology, Santa Cruz, CA; sc24948). Protein contents of cell lysates

54 were measured using the Pierce ® BCA protein assay (Thermo Scientific,

Rockford, IL; 23223 and 23224) using the manufacturer’s protocol. Fifty g of protein was loaded and separated on a 10% SDS polyacrylamide gel. Proteins were then transferred to nitrocellulose membrane and blocked in TrisBuffered

Saline with 0.1% Tween 20 (TBST) containing 5% (w/v) nonfat dry milk.

Membranes were probed overnight by incubation with primary antibodies according to the manufacturer's instructions and followed by washing in TBST.

ALDH1B1 was detected using rabbit polyclonal antiALDH1B1 antibody (1:5,000 in 5% nonfat dry milk in TBST) [12] and βactin was detected using mouse monoclonal antiβactin antibody (1:10,000; Sigma, St. Louis, MO). Membranes were then incubated for 1 h at room temperature with horseradish peroxidase conjugated goat antirabbit or antimouse secondary antibody (1:5,000; Sigma,

St. Louis, MO). After washing in TBST, membranes were incubated in Pierce ®

ECL Western blotting substrate (Thermo Scientific Rockford, IL; 32106), and exposed to Xray film to visualize protein bands. Protein expression was normalized to βactin expression and quantitation of band density was conducted using NIH ImageJ software (http://rsbweb.nih.gov/ij/) [176].

Gene expression

RTPCR was used to confirm knockdown of ALDH1B1 in SW480 cells using human ALDH1B1 shRNA and to evaluate canonical Wnt/βcatenin pathway genes in these cells. Total RNA was extracted using RNeasy Mini kit (Qiagen,

Valencia, CA) followed by cDNA synthesis using Maxima Universal FirstStrand

55 cDNA Synthesis kit (Thermo, K1661) following manufacturer’s instructions.

Validated and predesigned primer/probes for ALDH1B1, Axin2 , cMyc , LGR5 and

GAPDH were purchased from Applied Biosystems (Carlsbad, CA). AB7500 system (Applied Biosystems) was used to run RTPCR using taqman master mix.

Relative quantification of the steadystate target mRNA levels was calculated after normalization of total amount of cDNA to GAPDH (endogenous reference).

Relative mRNA expression was analyzed by using the ∆ ∆ C T method, where

(∆ C T ) = CT (gene) – CT (housekeeping gene). Absolute Quantification Real time PCR Assay [177] was used to screen colon cancer cells for ALDH1B1 mRNA levels. QPCR analysis was performed on a 7500 Real Time PCR System

(Applied Biosystems) using Power SYBR Green Master Mix (Applied

Biosystems) and ALDH1B1 primers (Forward: AGCGCGATTCGGAGCCTCA,

Reverse: TGACCGCATCATGCCACTCGT).

Threedimensional cell cultures

SW480 cells were cultured by the overlay method of 3D culture on solidified Matrigel ® (BD Biosciences, San Jose, CA, #354234) [176]. Briefly, cells were grown to ~ 80% confluency in a monolayer. A single cell suspension was prepared in RPMI 1640 medium supplemented with 10% fetal bovine serum, penicillin (100 units/ml), streptomycin (100 g/ml) (Life Technologies) and 2% matrigel and seeded onto 80 l Matrigel ® in the 8well chamber slide (BD

Biosciences, #354118) at 2000 cells/well. All cultures were then maintained at

37°C in a 5% CO 2 humidified incubator for up to 10 d. Cell morphology was

56 examined every alternate day using phase contrast microscopy. The size of resulting spheroids was measured on day 10 and volume was calculated by formula V= (4/3)πr 3.

Xenograft in athymic mice

Scramble control and ALDH1B1 shRNA knockdown SW480 cells were injected into the left and right flanks of 4 to 6 wkold female athymic (nu +/nu +) mice (Harlan Laboratories, Indianapolis, IN). Mice were monitored daily for signs of toxicity, and the tumor size was evaluated twice per w by caliper measurements using the following formula: tumor volume= (length × width 2) ×

0.52. Mice were euthanized at the end of the experiment and all of the protocols used were approved by the Institutional Animal Care and Use Committee of the

University of Colorado Denver.

Flow cytometric analysis of cells

Cell count and viability of scramble and ALDH1B1 shRNAtransfected

SW480 cells were determined on a Countess Automated Cell Counter (Life

Technologies). Cells in both groups exhibited more than 80% viability. The

Aldefluor ® reagent system (Stem Cell Technologies, Vancouver, Canada) was used to detect intracellular catalytic activity of ALDH. This assay uses boron dipyrromethene aminoacetaldehyde (BAAA), a fluorescent substrate of ALDH, which is converted into bodipyaminoacetate (BAA) by intracellular ALDH. BAA is retained intracellularly because of its net negative charge, so cells expressing

57 high ALDH accumulate more BAA and thus show higher fluorescence. In brief, cells were centrifuged (250 G for 5 min) and pellets were resuspended in

Aldefluor ® assay buffer containing 1 mol/l BAAA at a concentration of 10 6 cells/ml. Activated Aldefluor ® (5 l) was added to 1 ml of resuspended cells and mixed well. For a negative control, 500 l from this cell suspension was immediately placed in another tube with the ALDH inhibitor

[diethylaminobenzaldehyde (DEAB)]. All samples were incubated at 37°C for

45 min and then centrifuged and washed with 500 l Aldefluor ® buffer. ALDH + cells were evaluated by the University of Colorado Cancer Center flow cytometry core facility.

Statistical analysis

All quantitative experiments were performed at least in triplicate, and the data shown are the means ± S.E. of one representative experiment. Statistical significance was determined by Student’s ttest or oneway ANOVA using

SigmaStat Statistical Analysis software (SPSS Inc., Chicago, IL). P < 0.05 was considered to be significant.

Results

ALDH1B1 expression in colon polyps from Apc Min mice

The colon from wild type mice expressed immunopositivity for ALDH1B1 in a small subset of the columnar cells at the bottom of crypt (putative stem cell compartment) and in some cells along the side of the lower middle part of the

58 crypt. However, cells from stroma and differentiated mucosal cells remained completely negative. In contrast, colon polyps from Apc Min mice showed very strong and universal ALDH1B1 expression in the epithelial cells. Some of the stomal cells from Apc Min mice were also stained positively with ALDH1B1 (Figure

3.1B).

Identification of novel TBE sites in the promoter of the human ALDH1B1 gene

Since ALDH1B1 is overexpressed in human colon cancer and is associated with activation of Wnt/βcatenin pathway, we analyzed the human

ALDH1B1 3kb promoter region using the TESS program. Six binding sites for

TCF/LEF transcription factors were identified by scanning the 3000 bp upstream sequence from TSS of the ALDH1B1 gene, which comprise a minimum element of 5′(A/T)(A/T)CAA(A/T)G3′ (Figure 3.2A) [178]. TBE6 (2782 bp), TBE5 (2701 bp), TBE4 (1724 bp), TBE3 (1684 bp) and TBE2 (1209 bp) were in sense orientation whereas TBE1 (499 bp) was in antisense orientation. Therefore, we cloned the 3064 bp segment from the human ALDH1B1 promoter including all six

TBE sites, into the pGL3enhancer vector. This vector was used for the subsequent reporter assay (Figure 3.2A).

Figure 3.1 (A) Illustration of Wnt/βcatenin signaling pathway. ( B) Distribution of ALDH1B1 in wild type and Apc Min mice. Expression of ALDH1B1 is confined to stem cell region of the normal colon (left panel) from wildtype mice; in contrast, colon polyp from Apc Min mice showed universal expression in the epithelial cells with some staining in the stromal cells (right panel). ALDH1B1 expression exhibited cytosolic punctate pattern. (100x)

ALDH1B1 promoter is active in colon cancer cells

Investigation of ALDH1B1 gene expression among colon cancer cells showed that the COLO320 cell line had the greatest expression of ALDH1B1 followed by the SW480 and HT29 (Figure 3.2B). Since these cells could exhibit high levels of ALDH1B1, we transfected the SW480, COLO320 and HT29 cell lines with ALDH1B1pGL3 or cMycpBV luciferase vector and used corresponding empty plasmids as control. Luciferase activity was measured to determine the basal level of activation of the promoters for ALDH1B1 and cMyc in these colon cancer cells. ALDH1B1 promoter activity was 2, 17 and 233fold that of the control plasmid in the HT29, SW480 and COLO320 cells, respectively.

Similar results were observed for cMyc promoter activity in these cells (Figure

3.2C). These results suggest that the ALDH1B1 promoter activation pattern in colon cancer cells closely resemble to that of ALDH1B1 mRNA expression and cMyc promoter activity in these cells. To dissect the contribution of Wnt/β catenin responsive TBE sites on the transcriptional activity of the ALDH1B1 promoter, four ALDH1B1 promoter deletions constructs were assayed for their activity through transient transfections in COLO320, SW480 and HT29 cells.

Surprisingly, sequential deletion of all the putative TBEs from the ALDH1B1 pGL3 construct did not make any difference in the promoter activity in COLO320 and HT29 cells whereas deletion of TBE2 to TBE6 in SW480 cell led to increased luciferase activity (Figure 3.3AC).

Figure 3.2. Evaluation of human ALDH1B1 promoter activity in colon cancer cell lines. (A) Putative TCF/LEF binding elements (TBEs) in the human ALDH1B1 3kb promoter region. TBE6, TBE5, TBE4, TBE3 and TBE2 are in sense orientation whereas TBE1 is in antisense orientation. Numbers given below TBEs represent the distance from transcription start site (TSS) in bp. (B) mRNA levels were measured by QPCR analysis and reported as copy numbers using a corresponding standard curve. (C) Luciferase promoter activity (measured 24h after transfection) of the ALDH1B1pGL3 and cMycpBV promoter reporter constructs compared to the empty pGL3enhancer and pBV vectors, respectively in COLO320, SW480 and HT29 cells. The relative luciferase activity is normalized to the internal control renilla luciferase. Data is expressed as relative luciferase activity (fold of control plasmid) and presented as mean ± SE (representative from three individual experiments done in triplicate).

Figure 3.3. ALDH1B1 promoter activity in colon cancer cell lines. COLO320, SW480 and HT29 cells were transfected with pGL3promoter reporter construct containing the ALDH1B1 3 kb promoter fragment or with constructs having sequential deletions of putative TCF/LEF binding elements (TBEs). Luciferase promoter activity was measured 24h after transfection and the cells transfected with empty plasmid were used as control. The relative luciferase activity is showed after normalization to the internal control Renilla luciferase. Luciferase activity was calculated as fold of empty pGL3 enhancer vector and data expressed as per cent luciferase activity of TBE16 construct and presented as mean ± SE (representative from three individual experiments done in triplicates). *: P < 0.05, oneway ANOVA.

ALDH1B1 plays a critical role in colon carcinogenesis

We used the SW480 cell line for shRNAmediated ALDH1B1 knockdown

(data not shown). For evaluating the effect of ALDH1B1 knockdown on tumor growth, we used an in vitro threedimensional culture system and in vivo xenograft model. Prior to these experiments, ALDH1B1 shRNA knockdown was evaluated in the SW480 cells by RT–PCR and Western blotting. Protein and mRNA expression of ALDH1B1 was decreased in the ALDH1B1 shRNA cells by

80% (Figure 3.4A).

Cells from the scramble control and ALDH1B1 knockdown groups were cultured in threedimensional Matrigel ® to examine spheroid formation. After 10 d of incubation, the volume and number of spheroids were lower in the ALDH1B1 knockdown cells than in control cells (Figure 3.4B). Furthermore, evaluation of tumor growth in a xenograft model showed that tumors with ALDH1B1 knockdown exhibited a significant reduction in tumor volume when compared to scramble control (Figure 3.4C). In summary, knockdown of ALDH1B1 in the

SW480 cell line greatly reduced spheroid formation in vitro and tumor growth in vivo . Using this shRNA approach, we have shown that ALDH1B1 has a pro tumorigenic role in human colon cancer cells, which may explain its widespread expression in human colon cancers.

Figure 3.4. ALDH1B1 knockdown in the SW480 cell line inhibits spheroid formation and tumor growth.

Figure 3.4. ALDH1B1 knockdown in the SW480 cell line inhibits spheroid formation and tumor growth. SW480 cells were transfected with ALDH1B1 shRNA or scramble shRNA to develop stable cell colonies. (A) ALDH1B1 expression was examined by Western blotting and quantified by densitometric analysis. Numbers at the bottom of Western blot bands are the average densitometry of three replicates expressed relative to scramble shRNA transfected control cells. Rec, recombinant ALDH1B1; CT, Control SW480 cells without transfection; Sc, scramble. ( B) Spheroid formation was measured 10 d after plating of cells. Representative phase contrast micrographs (left) and histograms (right) showing decreased number and smaller size of spheroids from SW480 cells transfected with ALDH1B1 shRNA (open bars) or with ScshRNA control vector (closed bars). Data are presented as mean ± S.E. (n = 3). #: P < 0.01, Student’s ttest, compared with scramble control. ( C) Growth of tumors induced by injection of SW480 cells transfected with ALDH1B1 shRNA or scramble shRNA into the left and right flanks of female athymic (nu+/nu+) mice. Tumor volume was calculated as (length × width 2 × 0.52). Data were reported as mean ± S.E. (n = 10 tumors per group). *: P < 0.05, Student’s ttest, compared with ScshRNA control at same time point.

ALDH1B1 knockdown depletes ALDH bright cells

Control SW480 cells contain a small population of cells (~ 3 %) with very high ALDH activity, socalled ALDH bright cells (Figure 3.5A). ALDH bright cells are considered to possess high tumorigenic potential [179] and ALDH catalytic activity can be inhibited by DEAB. ALDH1B1 knockdown in SW480 cells using shRNA decreased the number of ALDH bright cells by about 90% (Figure 3.5BC).

Our findings suggest that ALDH1B1 may play an important role in tumorigenesis of colon cancer cells.

ALDH1B1 regulates Wnt/βcatenin, Notch and PI3K/Akt signaling pathways in colon cancer

Because Wnt signaling activation and ALDH1B1 expression are closely linked in human colon cancer and APC Min mice, we examined the effect of

ALDH1B1 knockdown on Wnt/βcatenin signaling. We employed the

TOPFlash/FOPFlash Wnt activity reporter assay using ALDH1B1shRNA and scrambleshRNA transfected in the SW480 cell line. Since RKO human malignant cells express wildtype APC and βcatenin proteins, we used these cells as negative control. A decrease in βcateninmediated transcriptional activity in the SW480 ALDH1B1 knockdown cells was observed 24 h post transfection (Figure 3.6A).

Figure 3.5. ALDH1B1 knockdown in SW480 cells depletes ALDH bright cells. Flow cytometry was conducted in SW480 cells transfected with either ALDH1B1 shRNA or scramble shRNA. Cells were incubated with Aldefluor ® substrate in the presence or absence of the ALDH inhibitor diethylaminobenzaldehyde (DEAB). Cells fluorescing as a result of high ALDH activity (ALDH bright cells) were determined. (A) Flow cytometric analysis of scramble shRNA control cells in the absence (DEAB) or presence (+DEAB) of DEAB. (B) Flow cytometric analysis of ALDH1B1 shRNA cells in the absence (DEAB) or presence (+DEAB) of DEAB; (C) Proportion of ALDH bright cells in ALDH1B1 shRNA (black bar) and scramble shRNA (grey bar) cell populations. Data represent mean ± S.E. from 3 experiments. * P < 0.05, Student’s ttest, compared to scramble shRNA cells.

We also examined the expression levels of the Wnt/βcatenin signaling related genes Axin2 , CTNNB1 (βcatenin), cMyc and LGR5 . RTPCR analysis showed a strong induction of Axin2 and suppression of CTNNB1, cMyc and

LGR5 expression in the ALDH1B1 knockdown SW480 cells (Figure 3.6B). The

68 effect of ALDH1B1 knockdown on expression of active βcatenin, LEF1 (crucial proteins for Wnt/ βcatenin signaling) and cMyc (Wnt target) was evaluated by

Western blotting. Figure 3.6C shows that the expression of all three proteins examined were decreased in the SW480 ALDH1B1 knockdown cells. Since the

Notch ligand JAG1 has been shown to be regulated by Wnt signaling, we investigated the effects of ALDH1B1 knockdown on JAG1 protein levels as well as cleaved Notch1 activity. As shown in Figure 3.6C, both JAG1 and cleaved

Notch1 were decreased in the SW480 ALDH1B1 knockdown cells. Taken together, these findings suggest that ALDH1B1 positively modulates Wnt/β catenin signaling in SW480 colon cancer cells and these results are consistent with a role of ALDH1B1 in modulating the Wnt/βcatenin and Notch signaling pathways.

ALDH1B1 is involved in the generation of RA [13] and in cells expressing high FABP5, RA plays a proproliferative role by modulating PPARβ/δ activity.

PPARβ/δ has been shown to enhance the activity ofPI3K/Akt signaling [101].

Therefore, we examined the expression of FABP5, PPARβ/δ and PI3K/Akt pathway related proteins (PI3Kp85, Akt and MMP2) using Western blotting.

Knockdown of ALDH1B1 in the SW480 cells resulted in the downregulated in

PI3Kp85, Akt and MMP2 (Figure 3.6 D). Since the PI3K pathway is one essential pathway in promoting cell proliferation and survival, these results indicate a possible mechanistic basis for the decreased tumor growth observed in SW480 cells in which ALDH1B1 expression has been repressed by shRNA.

Figure 3.6. ALDH1B1 positively regulates Wnt/βcatenin, Notch and PI3K/Akt signaling pathways. (A) For the Wnt/βcatenin reporter assay, scramble or ALDH1B1 shRNA transfected SW480 cells were transiently transfected with either 2.5 g TOPflash, or FOPflash reporter plasmid along with 0.2 g pRLTK vector as an internal control. Data are expressed as TOP/FOPflash ratio and presented as mean ± SE (representative from three individual experiments done in triplicate). *: P < 0.05, Student’s ttest. (B) RT PCR analysis of RNA isolated from SW480 cells transfected with scramble shRNA or ALDH1B1 shRNA. mRNA levels were expressed as fold change of scramble shRNA, GAPDH gene was used to normalize the data. Results are presented as the mean ± S.E. from three experiments. *P<0.05, Student’s ttest, compared to results in scramble shRNA cells. ( CD) Western blot analysis revealed decreased expression of members of Wnt/βcatenin and Notch pathways ( C); that of FABP5, PPARβ/δ and PI3K/Akt pathway members (D) in ALDH1B1 knock down SW480 cells relative to scramble control cells. Proteins from western blot representative of three independent experiments were quantified by densitometry.

Discussion

Despite a better understanding and substantial advances in our knowledge of the molecular mechanisms involved in CRC development and progression, this disease still remains one of the leading causes of cancer related deaths in the developed world. CRCs show dysregulation of the Wnt/β catenin signaling pathway, including Wnt dependent transcription of cMyc and

CyclinD1 . These genes have TBEs in their promoter regions which are required for all genes under direct control of the Wnt/βcatenin signaling pathway [180

182].

Given that overexpression of ALDH1B1 and Wnt/βcatenin signaling activity have been observed in human colon cancer cells [12, 158], we hypothesized that ALDH1B1 may also be regulated by Wnt/βcatenin signaling.

In this study, we found six TBE sites present in the ALDH1B1 3kb promoter region, making it a potential target for control by βcatenin and, by extension, the

Wnt/βcatenin signaling pathway [142]. In addition, we demonstrated that the

ALDH1B1 promoter is transcriptionally active in colon cancer cells. However, even after sequential deletion of all the six TBE sites from the ALDH1B1pGL3 promoter construct, promoter activity remained the same. This indicates that six potential TBE sites found within the 3 kb promoter region of ALDH1B1 are not contributing to transcriptional activation of this promoter. In SW480 cells, the promoter region containing TBE 2 to TBE6 exhibited increased promoter activity, suggesting that transcription repression factors are present in this region of the promoter and are functional in these cells. All of the examined cell lines showed

ALDH1B1 promoter activity even after deletion of all the six putative TBEs, so we speculate that the remaining promoter region (492/+115 bp) has some transcription enhancing factor that needs to be evaluated further. We have examined this remaining promoter region and found binding sites for various transcription factors including those known to play important role in colon cancer e.g., AP2, C2H2 zinc finger, E2F4, EGR1,ETS1, serum response factors [183

188]. Further research is required to ascertain that which of these sites are involved in colon cancer initiation and/ or propagation.

To explore the possibility of ALDH1B1 being upstream of the Wnt cascade, we used an shRNA approach to knock down expression of ALDH1B1 in the SW480 cell line. We determined that the high ALDH1B1 expressing SW480 cell line had an 80% reduction in ALDH1B1 levels following knockdown. Of note, this cell line has very low ALDH1A1 expression. As such, we selected this cell line for tumor formation assays to rule out any possible obfuscating effects of

ALDH1A1 on tumorigenicity. When examining the possible correlation between

Wnt/βcatenin signaling and ALDH1B1, we predicted that ALDH1B1 knockdown should modulate the tumorigenic potential of the colon cancer cells. We utilized threedimensional cancer cell culture since the response of cancer cells are significantly affected by their microenvironment, e.g., tissue architecture and extracellular matrix (ECM). Further, it more closely resembles in vivo tumors than conventional twodimensional cultures [189]. Using a shRNA approach, we showed that ALDH1B1 appears to have an important role for in vitro spheroid formation and in vivo tumor formation in the SW480 cell line. These results, for

72 the first time, demonstrate that ALDH1B1 may have a protumorigenic role in human colon cancer cells. This is contrary to the majority of previous studies which suggest ALDH1A1 is the isozyme primarily responsible for the effect of

ALDH activity on the phenotype of cancer cells [64, 157, 190194].

ALDH1A1 has also been suggested as a marker of ALDH activity based on the Aldefluor ® assay. However, we and others have shown that the Aldefluor ® assay is not specific for the ALDH1A1 isoform, meaning that other ALDHs may contribute to this phenotype [12, 125]. ALDH bright cells are a small subpopulation of tumor cells with stem celllike properties (stemness), including high tumorigenicity, selfrenewal and chemoresistance [42, 64]. Our results show that

ALDH1B1 is responsible for the high ALDH activity (and possibly stemness) of

ALDH bright SW480 colon cancer cells. It is not surprising to find that ALDH1B1 also has ALDH catalytic activity because it is structurally and functionally similar to ALDH1A1 [11]. As noted in introduction (chapter I), RA has been reported to increase expression of prosurvival genes and tumor growth by activating

PPARβ/δ in colon cancer in cells with high FABP5 relative to CARBPII (Figure

3.7) [101, 104]. This may explain the decrease in tumorigenicity in the ALDH1B1 knockdown cells in our in vitro and in vivo tumor formation assays.

Several inherited and sporadic colorectal cancer studies have established that Wnt/βcatenin signaling plays a critical role in colon tumorigenesis [142].

With our discovery that ALDH1B1 may be central for colon cancer formation and is a major contributor of ALDH activity in highly tumorigenic ALDH bright SW480 colon cancer cells, we hypothesized that ALDH1B1 affects colon cancer

73 formation by modulating Wnt/βcatenin signaling. This was indeed found to be the case, revealing for the first time that ALDH1B1 modulates Wnt/βcatenin signaling in colon cancer cells. In this study, we showed that repression of

ALDH1B1 in SW480 cells leads to increased expression of Axin2 mRNA and suppression of cMyc and LGR5 mRNAs. Axin2 is a negative regulator of Wnt/β catenin signaling by enhancing the phosphorylation, ubiquitination and degradation of βcatenin. Mutations in Axin2 have been described to be oncogenic in CRC [195, 196]. cMyc is an oncogene and downstream target of

Wnt/βcatenin signaling [142, 197] and overexpression of cMyc has been shown to enhance cell proliferation and survival in cancer cells [198]. Downregulation of cMyc suppresses growth in colon cancer cells and induces apoptosis [199].

LGR5 is a target gene of Wnt/βcatenin signaling and is expressed in stemlike cells in colon cancers. In addition, overexpression of LGR5 is associated with a poor prognosis in colorectal cancer patients [197, 200]. Collectively, these findings suggest that ALDH1B1 may mediate colon cancer formation through a mechanism involving the Wnt/βcatenin signaling pathway.

Overexpression of Notchrelated proteins in cells with active Wnt signaling at the bottom of intestinal crypt is suggestive of cross talk between Wnt and

Notch signaling [169, 174]. Also, APC Min mice, which have mutated APC gene and resultant hyperactive Wnt signaling show enhanced Notch signaling activity

[146]. Overexpression of the Notch signaling pathway has been shown to dramatically increase proliferating cells in wildtype mice. However, a similar effect is not seen in TCF4 null mice, which have blocked Wnt activity. This

74 indicates that expansion of proliferative compartments in response to hyperactivation of Notch signaling requires an intact Wnt signaling cascade [201].

The synergistic effect of these two pathways on tumor growth was also shown in

Notch/APC Min double mutant mice. These mice exhibited an increase in the number of intestinal tumors and dysplastic lesions in colon relative to the APC Min mice [201]. In addition, the Notch ligand Jagged1has been found to be transcriptionally activated by βcatenin, suggesting Notch signaling is downstream of the Wnt pathway in colorectal cancer [152]. Our results showing downregulation of the Notch pathway along with Wnt signaling downregulation in ALDH1B1 knockdown cells corroborate these results. These findings, when taken together with the enhanced Notch activity identified in human colon adenoma [201] suggest that Notch signaling, may enhance tumor development by hyperactivation of Wnt signaling.

Activation of the PI3K/Akt pathway has been reported in colorectal cancers [171] and our results corroborate these data. We also found a possible link between RAgenerating ALDH1B1 and PI3K/Akt signaling activation as summarized in Figure 3.7. This pathway crosstalk could also be due to Hes1, the downstream target of Notch signaling, negatively regulating PTEN expression, which has inhibitory effects on PI3K/Akt signaling [202, 203]. Further research is required to clarify the exact mechanism of this crosstalk.

These findings demonstrate that ALDH1B1 may not only be a potential marker for colon cancer but also play a functional role in the formation of tumors in colorectal cancers. In addition, the present study provides mechanistic insights

75 into the means by which ALDH1B1 may promote tumor formation, i.e. by modulating the Wnt/βcatenin, Notch, and PI3K/Akt signaling pathways. Finally, it is apparent that ALDH1B1 can be a significant contributor to ALDH activity measured by the Aldefluor ® assay. These results substantiate the accumulating evidence that many ALDH isozymes other than ALDH1A1 (especially ALDH1B1) may contribute to cancer stem cell phenotype [12, 49, 50, 125]. Hence,

ALDH1B1 may be one of the factors imparting high tumorigenicity to these tumorinitiating cells in colon cancer. It is anticipated that the findings from this study will serve as a foundation that may allow earlier detection of colorectal cancer and the identification of novel therapeutic approaches for the prevention and treatment of this devastating disease, such as specific inhibitors or substrates of ALDH1B1.

Figure 3.7. Possible mechanism for the role of ALDH1B1 in colon tumorigenesis. ALDH1B1 play important role in cell survival and proliferation possibly by modulating Wnt, Notch and PI3K/Akt signaling pathways and thereby contributes in colon tumorigenesis. 76

CHAPTER IV

CONCLUSION

Summary

Most of the pathological effects of ethanol are attributed to its reactive metabolites. Ethanolderived acetaldehyde, responsible for some of the major toxic effects of alcohol consumption, is metabolized to acetate by aldehyde dehydrogenases (ALDHs), mostly by mitochondrial ALDH2 and, to a lesser extent, by cytosolic ALDH1A1. We have recently shown that ALDH1B1, which shares a considerable identical amino acid sequence with ALDH2 and ALDH1A1,

(a) has a high affinity for acetaldehyde and appears to be the second major enzyme responsible for acetaldehyde elimination, (b) generates RA, a molecule crucial for cell proliferation and differentiation, (c) is crucial for maintaining the pancreatic progenitor pool and (d) is a potential biomarker and participant in human colorectal cancer.

Therefore, to delineate the role of ALDH1B1 in ethanol metabolism and glucose homeostasis, we have generated and characterized Aldh1b1(/) mice, as described in chapter II of this thesis. Our data show that ALDH1B1 is dispensable for the survival and reproduction in these mice. Prior work has indicated a possible role of ALDH1B1 in ethanol metabolism so we used our

Aldh1b1(/) mice for an ethanol pharmacokinetics study. This study revealed that

ALDH1B1 deletion makes these mice less efficient in clearing blood

77 acetaldehyde; however there was no difference in blood ethanols level after i.p. ethanol administration. We also performed a glucose tolerance test in Aldh1b1(/

) mice, which revealed high fasting blood glucose and poor glucose tolerance in knockout mice suggesting a possible role for ALDH1B1 in glucose homeostasis.

However, we did not observe a difference in the pancreatic islet architecture in

Aldh1b1(/) mice. Studies presented in the chapter II of this thesis demonstrate that, although ALDH1B1 is not essential for survival and fertility, it plays a significant role in ethanol metabolism and pancreatic function.

Our previous work showed that ALDH1B1 is a potential biomarker for colon cancer and the expression of this protein is closely associated with the activation of Wnt/βcatenin signaling, a crucial pathway for colon cancer generation. We examined colon polyps from the Apc Min mice for ALDH1B1 expression and, similar to our previous study on human colon cancers, it was significantly upregulated in mice colon polyps. To delineate the association between ALDH1B1 and Wnt/βcatenin signaling, we examined TBE within the 3 kb promoter region of ALDH1B1 gene and found six TBE sites. The promoter reporter assay revealed ALDH1B1 promoter activity in the examined CRC cell lines. However, sequential deletion of these sites showed that the transcriptional activation of ALDH1B1 promoter is not mediated trough examined six TBE sites in these cells. To decipher the mechanism for ALDH1B1 upregulation and role of

ALDH1B1 in colon tumorigenesis, we used SW480 cells with stable ALDH1B1 knock down (using shRNA), for Wnt activity reporter assay, three dimensional spheroid formation assay and nude mice xenograft study. We found decreased

Wnt activity in ALDH1B1depleted colon cancer cells. Our data also show that

ALDH1B1 knock down inhibits spherogenicity and tumorigenicity of SW480 cells.

The Aldefluor ® assay, as presented in the chapter III of this thesis, shows that

ALDH1B1 significantly contributes to ALDH catalytic activity of SW480 colon cancer cells. Cancer cells with high ALDH activity have been shown to possess properties like stemness, high tumorigenicity and chemoresistance. Given that

Wnt/βcatenin, Notch and PI3K/Akt signaling pathways play critical roles in colon cancer, we examined the mRNA and protein expression of molecules involved in these pathways in the ALDH1B1 knock down SW480 cell line. We found increased expression of Axin2 (a negative regulator of Wnt signaling) and decreased expression of cMyc (oncogene and downstream target of Wnt signaling) and LGR5 (Wnttarget gene). Expression of Jagged1 (the target gene of Wnt/βcatenin signaling and lingand for the Notch pathway), and other Notch targets (HES1, cNotch) were repressed upon ALDH1B1 knock down in these cells. We also found a possible association between ALDH1B1 knockdown and suppression of FABP5 and PPARβ/δ and subsequent inhibition of the PI3K/Akt signaling pathway. The data presented in this thesis shows that ALDH1B1 is important for colon tumorigenesis and these findings greatly enhance our understanding about the mechanistic basis for the inhibition of tumor growth in

SW480 cells with ALDH1B1 knockdown.

Significance

Our research work has revealed that ALDH1B1 is a vital player in ethanol metabolism and it will help in identifying new targets for the treatment of alcohol toxicity, and provide a basis for understanding interindividual vulnerabilities to the deleterious effects of alcohol consumption. A novel result suggests an important role for ALDH1B1 in pancreatic function. Dysregulation of pancreatic homeostasis can lead to diabetes or pancreatic cancer, both diseases with very high prevalence and high associated human cost in terms of both quality of life and mortality. Our results provide a basis for the further elucidation of the role of

ALDH1B1 in the maintenance of pancreatic progenitors and stem cells, which may provide venues for both restoring pancreatic function in diabetes and efficiently targeting pancreatic cancer stem cells. Our data also suggest that

ALDH1B1 is an early marker of colon cancer and contributes to the oncogenic potential of colon cancer cells. The studies presented in this thesis also begin to explain the mechanism by which ALDH1B1 may function in colon cancer, and make it a potential target for anticancer therapy.

Future directions

Research presented in this thesis showed that ALDH1B1 is involved in ethanol metabolism and is also crucial for colon cancer generation. We now wish to extend our studies and determine the role of ALDH1B1 in ethanolinduced colon cancer using transgenic knockout animals (Aldh1b1(/) mice reported in chapter II) and human tissue samples and elucidate the role of this protein as a

80 colon cancer biomarker. A causal link has been established between alcohol consumption and various forms of cancers. Worldwide, approximately 389,000 cases of cancer (representing 3.6% of all cancers) derive from chronic alcohol consumption. In 2007, two of the most common cancers, colon and breast cancer, were added by the International Agency for Research on Cancer (IARC) to the list of cancers causally related to alcohol [204]. Ethanol has been shown to potentiate chemically and geneticallyinduced colon cancer in experimental animals [27, 205, 206]. Specifically, ethanol administration in drinking water promotes intestinal tumorigenesis in the Apc Min mouse model [206]. To further evaluate role of ALDH1B1 in colon cancer, we plan to intercross our Aldh1b1(/) mice with mice having colonspecific disruption of Apc [207]. We will also examine the expression of ALDH1B1 during the various stages of colon disease

(including polyps and inflammation) and in tumors obtained from Caucasian and

Japanese alcoholics. We have access to the tissue microarray blocks from the

Aspirin/Folate Polyp Prevention Study (AFPPS) [208210]. A valuable advantage of using AFPPS tissue is that specifics regarding the alcohol consumption of each patient will be available to us.

We also want to identify the role of ALDH1B1 in pancreas development and regeneration by transgenically modulating Aldh1b1 expression (i.e., loss and gain) in mouse models. We will also like to study the mechanism by which

ALDH1B1 is involved in glucose homeostasis. Additionally, we will seek to establish whether ALDH1B1 constitutes the elusive adult pancreas stem cell marker. The proposed studies will greatly advance our understanding of the

81 pancreas stem cell in development, regeneration and cancer. Such knowledge may be exploited in the development of novel approaches to enhance pancreas regeneration and/or halt pancreatic cancer progression.

REFERENCES

1. Vasiliou V, Pappa A, Petersen DR. Role of aldehyde dehydrogenases in endogenous and xenobiotic metabolism. Chem Biol Interact 2000; 129 : 1 19.

2. Marchitti SA, Brocker C, Stagos D, Vasiliou V. NonP450 aldehyde oxidizing enzymes: the aldehyde dehydrogenase superfamily. Expert Opin Drug Metab Toxicol 2008; 4: 697720.

3. Sladek NE. Human aldehyde dehydrogenases: potential pathological, pharmacological, and toxicological impact. J Biochem Mol Toxicol 2003; 17 : 723.

4. Black WJ, Stagos D, Marchitti SA et al. Human aldehyde dehydrogenase genes: alternatively spliced transcriptional variants and their suggested nomenclature. Pharmacogenet Genomics 2009; 19 : 893902.

5. Song BJ, Abdelmegeed MA, Yoo SH et al. Posttranslational modifications of mitochondrial aldehyde dehydrogenase and biomedical implications. J Proteomics 2011; 74 : 2691702.

6. Vasiliou V, Pappa A. Polymorphisms of human aldehyde dehydrogenases. Consequences for drug metabolism and disease. Pharmacology 2000; 61 : 1928.

7. Vasiliou V, Pappa A, Estey T. Role of human aldehyde dehydrogenases in endobiotic and xenobiotic metabolism. Drug Metab Rev 2004; 36 : 27999.

8. Enomoto N, Takase S, Takada N, Takada A. Alcoholic liver disease in heterozygotes of mutant and normal aldehyde dehydrogenase2 genes. Hepatology 1991; 13 : 10715.

9. Yokoyama A, Muramatsu T, Omori T et al. Alcohol and aldehyde dehydrogenase gene polymorphisms and oropharyngolaryngeal, esophageal and stomach cancers in Japanese alcoholics. Carcinogenesis 2001; 22 : 4339.

10. Stewart MJ, Malek K, Xiao Q et al. The novel aldehyde dehydrogenase gene, ALDH5, encodes an active aldehyde dehydrogenase enzyme. Biochem Biophys Res Commun 1995; 211 : 14451.

11. Stagos D, Chen Y, Brocker C et al. Aldehyde dehydrogenase 1B1: molecular cloning and characterization of a novel mitochondrial acetaldehydemetabolizing enzyme. Drug Metab Dispos 2010; 38 : 1679 87.

12. Chen Y, Orlicky DJ, Matsumoto A et al. Aldehyde dehydrogenase 1B1 (ALDH1B1) is a potential biomarker for human colon cancer. Biochem Biophys Res Commun 2011; 405 : 1739.

13. Jackson BC, Reigan P, Miller B et al. Human ALDH1B1 polymorphisms may affect the metabolism of acetaldehyde and alltrans retinaldehyde – in vitro studies and computational modeling. Pharmaceutical Research 2014; Under revision .

14. Ioannou M, Serafimidis I, Arnes L et al. ALDH1B1 is a potential stem/progenitor marker for multiple pancreas progenitor pools. Dev Biol 2013; 374 : 15363.

15. Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics, 2002. CA Cancer J Clin 2005; 55 : 74108.

16. Siegel R, Ma J, Zou Z, Jemal A. Cancer statistics, 2014. CA Cancer J Clin 2014; 64 : 929.

17. Wei EK, Wolin KY, Colditz GA. Time course of risk factors in cancer etiology and progression. J Clin Oncol 2010; 28 : 40527.

18. Kaneko R, Sato Y, An Y et al. Clinicoepidemiologic study of the metabolic syndrome and lifestyle factors associated with the risk of colon adenoma and adenocarcinoma. Asian Pac J Cancer Prev 2010; 11 : 97583.

19. Cho E, SmithWarner SA, Ritz J et al. Alcohol intake and colorectal cancer: a pooled analysis of 8 cohort studies. Ann Intern Med 2004; 140 : 60313.

20. Fedirko V, Tramacere I, Bagnardi V et al. Alcohol drinking and colorectal cancer risk: an overall and doseresponse metaanalysis of published studies. Ann Oncol 2011; 22 : 195872.

21. Brennan P, Boffetta P. Mechanistic considerations in the molecular epidemiology of head and neck cancer. IARC Sci Publ 2004: 393414.

22. Seitz HK, Stickel F. Molecular mechanisms of alcoholmediated carcinogenesis. Nat Rev Cancer 2007; 7: 599612.

23. Seitz HK, Stickel F. Acetaldehyde as an underestimated risk factor for cancer development: role of genetics in ethanol metabolism. Genes Nutr 2010; 5: 1218.

24. Singh S, Brocker C, Koppaka V et al. Aldehyde dehydrogenases in cellular responses to oxidative/electrophilic stress. Free Radic Biol Med 2013; 56 : 89101.

25. Comporti M. Lipid peroxidation and biogenic aldehydes: from the identification of 4hydroxynonenal to further achievements in biopathology. Free Radic Res 1998; 28 : 62335.

26. Poschl G, Seitz HK. Alcohol and cancer. Alcohol Alcohol 2004; 39 : 155 65.

27. Simanowski UA, Suter P, Russell RM et al. Enhancement of ethanol induced rectal mucosal hyper regeneration with age in F344 rats. Gut 1994; 35 : 11026.

28. Seitz HK, Matsuzaki S, Yokoyama A et al. Alcohol and cancer. Alcohol Clin Exp Res 2001; 25 : 137S43S.

29. Yu HS, Oyama T, Isse T et al. Formation of acetaldehydederived DNA adducts due to alcohol exposure. Chem Biol Interact 2010; 188 : 36775.

30. Secretan B, Straif K, Baan R et al. A review of human carcinogensPart E: tobacco, areca nut, alcohol, coal smoke, and salted fish. Lancet Oncol 2009; 10 : 10334.

31. Skrzydlewska E, Sulkowski S, Koda M et al. Lipid peroxidation and antioxidant status in colorectal cancer. World J Gastroenterol 2005; 11 : 4036.

32. Skrzydlewska E, Stankiewicz A, Sulkowska M et al. Antioxidant status and lipid peroxidation in colorectal cancer. J Toxicol Environ Health A 2001; 64 : 21322.

33. Skrzydlewska E, Kozuszko B, Sulkowska M et al. Antioxidant potential in esophageal, stomach and colorectal cancers. Hepatogastroenterology 2003; 50 : 12631.

34. Choudhary S, Xiao T, Vergara LA et al. Role of aldehyde dehydrogenase isozymes in the defense of rat lens and human lens epithelial cells against oxidative stress. Invest Ophthalmol Vis Sci 2005; 46 : 25967.

35. Makia NL, Bojang P, Falkner KC et al. Murine hepatic aldehyde dehydrogenase 1a1 is a major contributor to oxidation of aldehydes formed by lipid peroxidation. Chem Biol Interact 2011; 191 : 27887.

36. Lindahl R. Aldehyde dehydrogenases and their role in carcinogenesis. Crit Rev Biochem Mol Biol 1992; 27 : 283335.

37. Yokoyama A, Muramatsu T, Ohmori T et al. Alcoholrelated cancers and aldehyde dehydrogenase2 in Japanese alcoholics. Carcinogenesis 1998; 19 : 13837.

38. Storms RW, Trujillo AP, Springer JB et al. Isolation of primitive human hematopoietic progenitors on the basis of aldehyde dehydrogenase activity. Proc Natl Acad Sci U S A 1999; 96 : 911823.

39. Hess DA, Meyerrose TE, Wirthlin L et al. Functional characterization of highly purified human hematopoietic repopulating cells isolated according to aldehyde dehydrogenase activity. Blood 2004; 104 : 164855.

40. Armstrong L, Stojkovic M, Dimmick I et al. Phenotypic characterization of murine primitive hematopoietic progenitor cells isolated on basis of aldehyde dehydrogenase activity. Stem Cells 2004; 22 : 114251.

41. Ran D, Schubert M, Pietsch L et al. Aldehyde dehydrogenase activity among primary leukemia cells is associated with stem cell features and correlates with adverse clinical outcomes. Exp Hematol 2009; 37 : 1423 34.

42. Deng S, Yang X, Lassus H et al. Distinct expression levels and patterns of stem cell marker, aldehyde dehydrogenase isoform 1 (ALDH1), in human epithelial cancers. PLoS One 2010; 5: e10277.

43. Silva IA, Bai S, McLean K et al. Aldehyde dehydrogenase in combination with CD133 defines angiogenic ovarian cancer stem cells that portend poor patient survival. Cancer Res 2011; 71 : 39914001.

44. Wang YC, Yo YT, Lee HY et al. ALDH1bright epithelial ovarian cancer cells are associated with CD44 expression, drug resistance, and poor clinical outcome. Am J Pathol 2012; 180 : 115969.

45. van den Hoogen C, van der Horst G, Cheung H et al. High aldehyde dehydrogenase activity identifies tumorinitiating and metastasisinitiating cells in human prostate cancer. Cancer Res 2010; 70 : 516373.

46. van den Hoogen C, van der Horst G, Cheung H et al. The aldehyde dehydrogenase enzyme 7A1 is functionally involved in prostate cancer bone metastasis. Clin Exp Metastasis 2011; 28 : 61525.

47. Ginestier C, Hur MH, CharafeJauffret E et al. ALDH1 is a marker of normal and malignant human mammary stem cells and a predictor of poor clinical outcome. Cell Stem Cell 2007; 1: 55567.

48. CharafeJauffret E, Ginestier C, Iovino F et al. Aldehyde dehydrogenase 1positive cancer stem cells mediate metastasis and poor clinical outcome in inflammatory breast cancer. Clin Cancer Res 2010; 16 : 4555.

49. Marcato P, Dean CA, Pan D et al. Aldehyde dehydrogenase activity of breast cancer stem cells is primarily due to isoform ALDH1A3 and its expression is predictive of metastasis. Stem Cells 2011; 29 : 3245.

50. Langan RC, Mullinax JE, Ray S et al. A Pilot Study Assessing the Potential Role of nonCD133 Colorectal Cancer Stem Cells as Biomarkers. J Cancer 2012; 3: 23140.

51. Rasheed ZA, Yang J, Wang Q et al. Prognostic significance of tumorigenic cells with mesenchymal features in pancreatic adenocarcinoma. J Natl Cancer Inst 2010; 102 : 34051.

52. Kanda J, Matsuo K, Suzuki T et al. Impact of alcohol consumption with polymorphisms in alcoholmetabolizing enzymes on pancreatic cancer risk in Japanese. Cancer Sci 2009; 100 : 296302.

53. Carpentino JE, Hynes MJ, Appelman HD et al. Aldehyde dehydrogenase expressing colon stem cells contribute to tumorigenesis in the transition from colitis to cancer. Cancer Res 2009; 69 : 820815.

54. McCart AE, Vickaryous NK, Silver A. Apc mice: models, modifiers and mutants. Pathol Res Pract 2008; 204 : 47990.

55. Gentry T, Foster S, Winstead L et al. Simultaneous isolation of human BM hematopoietic, endothelial and mesenchymal progenitor cells by flow sorting based on aldehyde dehydrogenase activity: implications for cell therapy. Cytotherapy 2007; 9: 25974.

56. Marcato P, Dean CA, Giacomantonio CA, Lee PW. Aldehyde dehydrogenase: its role as a cancer stem cell marker comes down to the specific isoform. Cell Cycle 2011; 10 : 137884.

57. Corti S, Locatelli F, Papadimitriou D et al. Identification of a primitive brain derived neural stem cell population based on aldehyde dehydrogenase activity. Stem Cells 2006; 24 : 97585.

58. Foo LC, Dougherty JD. Aldh1L1 is expressed by postnatal neural stem cells in vivo. Glia 2013; 61 : 153341.

59. Tanei T, Morimoto K, Shimazu K et al. Association of breast cancer stem cells identified by aldehyde dehydrogenase 1 expression with resistance to sequential Paclitaxel and epirubicinbased chemotherapy for breast cancers. Clin Cancer Res 2009; 15 : 423441.

60. CharafeJauffret E, Ginestier C, Iovino F et al. Breast cancer cell lines contain functional cancer stem cells with metastatic capacity and a distinct molecular signature. Cancer Res 2009; 69 : 130213.

61. Le Magnen C, Bubendorf L, Rentsch CA et al. Characterization and Clinical Relevance of ALDHbright Populations in Prostate Cancer. Clin Cancer Res 2013; 19 : 536171.

62. Landen CN, Jr., Goodman B, Katre AA et al. Targeting aldehyde dehydrogenase cancer stem cells in ovarian cancer. Mol Cancer Ther 2010; 9: 318699.

63. Saw YT, Yang J, Ng SK et al. Characterization of aldehyde dehydrogenase isozymes in ovarian cancer tissues and sphere cultures. BMC Cancer 2012; 12 : 329.

64. Huang EH, Hynes MJ, Zhang T et al. Aldehyde dehydrogenase 1 is a marker for normal and malignant human colonic stem cells (SC) and tracks SC overpopulation during colon tumorigenesis. Cancer Res 2009; 69 : 33829.

65. Dylla SJ, Beviglia L, Park IK et al. Colorectal cancer stem cells are enriched in xenogeneic tumors following chemotherapy. PLoS One 2008; 3: e2428.

66. Jiang F, Qiu Q, Khanna A et al. Aldehyde dehydrogenase 1 is a tumor stem cellassociated marker in lung cancer. Mol Cancer Res 2009; 7: 330 8.

67. Zhang Q, Taguchi A, Schliekelman M et al. Comprehensive proteomic profiling of aldehyde dehydrogenases in lung adenocarcinoma cell lines. Int J Proteomics 2011; 2011 : 145010.

68. Chen YC, Chen YW, Hsu HS et al. Aldehyde dehydrogenase 1 is a putative marker for cancer stem cells in head and neck squamous cancer. Biochem Biophys Res Commun 2009; 385 : 30713.

69. Kim SK, Kim H, Lee DH et al. Reversing the Intractable Nature of Pancreatic Cancer by Selectively Targeting ALDHHigh, Therapy Resistant Cancer Cells. PLoS One 2013; 8: e78130.

70. Ma S, Chan KW, Lee TK et al. Aldehyde dehydrogenase discriminates the CD133 liver cancer stem cell populations. Mol Cancer Res 2008; 6: 1146 53.

71. Colombo F, Baldan F, Mazzucchelli S et al. Evidence of distinct tumour propagating cell populations with different properties in primary human hepatocellular carcinoma. PLoS One 2011; 6: e21369.

72. Zakhari S, Vasiliou, V, Guo, QM, ed. Alcohol and Cancer. 1 ed: Springer 2011.

73. Visapaa JP, Jokelainen K, Nosova T, Salaspuro M. Inhibition of intracolonic acetaldehyde production and alcoholic fermentation in rats by ciprofloxacin. Alcohol Clin Exp Res 1998; 22 : 11614.

74. Jelski W, Zalewski B, Chrostek L, Szmitkowski M. The activity of class I, II, III, and IV alcohol dehydrogenase isoenzymes and aldehyde dehydrogenase in colorectal cancer. Dig Dis Sci 2004; 49 : 97781.

75. Salaspuro M. Microbial metabolism of ethanol and acetaldehyde and clinical consequences. Addiction Biology 1997; 2: 3546.

76. Seitz HK, Simanowski UA, Garzon FT et al. Possible role of acetaldehyde in ethanolrelated rectal cocarcinogenesis in the rat. Gastroenterology 1990; 98 : 40613.

77. Chen YC, Peng GS, Tsao TP et al. Pharmacokinetic and pharmacodynamic basis for overcoming acetaldehydeinduced adverse reaction in Asian alcoholics, heterozygous for the variant ALDH2*2 gene allele. Pharmacogenet Genomics 2009; 19 : 58899.

78. Jackson BC, Holmes RS, Backos DS et al. Comparative genomics, molecular evolution and computational modeling of ALDH1B1 and ALDH2. Chem Biol Interact 2013; 202 : 1121.

79. Gagnon I, Duester G, Bhat PV. Enzymatic characterization of recombinant mouse retinal dehydrogenase type 1. Biochem Pharmacol 2003; 65 : 1685 90.

80. Gagnon I, Duester G, Bhat PV. Kinetic analysis of mouse retinal dehydrogenase type2 (RALDH2) for retinal substrates. Biochim Biophys Acta 2002; 1596 : 15662.

81. Graham CE, Brocklehurst K, Pickersgill RW, Warren MJ. Characterization of retinaldehyde dehydrogenase 3. Biochem J 2006; 394 : 6775.

82. Lin M, Napoli JL. cDNA cloning and expression of a human aldehyde dehydrogenase (ALDH) active with 9cisretinal and identification of a rat ortholog, ALDH12. J Biol Chem 2000; 275 : 4010612.

83. Duester G. Families of retinoid dehydrogenases regulating vitamin A function: production of visual pigment and retinoic acid. Eur J Biochem 2000; 267 : 431524.

84. Yoshida A, Hsu LC, Dave V. Retinal oxidation activity and biological role of human cytosolic aldehyde dehydrogenase. Enzyme 1992; 46 : 23944.

85. Suzuki R, Shintani T, Sakuta H et al. Identification of RALDH3, a novel retinaldehyde dehydrogenase, expressed in the ventral region of the retina. Mech Dev 2000; 98 : 3750.

86. Niederreither K, Fraulob V, Garnier JM et al. Differential expression of retinoic acidsynthesizing (RALDH) enzymes during fetal development and organ differentiation in the mouse. Mech Dev 2002; 110 : 16571.

87. Sima A, Parisotto M, Mader S, Bhat PV. Kinetic characterization of recombinant mouse retinal dehydrogenase types 3 and 4 for retinal substrates. Biochim Biophys Acta 2009; 1790 : 16604.

88. Kim H, Lapointe J, Kaygusuz G et al. The retinoic acid synthesis gene ALDH1a2 is a candidate tumor suppressor in prostate cancer. Cancer Res 2005; 65: 811824.

89. Mira YLR, Zheng WL, Kuppumbatti YS et al. Retinol conversion to retinoic acid is impaired in breast cancer cell lines relative to normal cells. J Cell Physiol 2000; 185 : 3029.

90. Touma SE, Perner S, Rubin MA et al. Retinoid metabolism and ALDH1A2 (RALDH2) expression are altered in the transgenic adenocarcinoma mouse prostate model. Biochem Pharmacol 2009; 78 : 112738.

91. Tang XH, Gudas LJ. Retinoids, retinoic acid receptors, and cancer. Annu Rev Pathol 2011; 6: 34564.

92. Chambon P. A decade of molecular biology of retinoic acid receptors. FASEB J 1996; 10 : 94054.

93. Shimizu M, Takai K, Moriwaki H. Strategy and mechanism for the prevention of hepatocellular carcinoma: phosphorylated retinoid X receptor alpha is a critical target for hepatocellular carcinoma chemoprevention. Cancer Sci 2009; 100 : 36974.

94. Boone CW, Kelloff GJ, Malone WE. Identification of candidate cancer chemopreventive agents and their evaluation in animal models and human clinical trials: a review. Cancer Res 1990; 50 : 29.

95. Warrell RP, Jr., Frankel SR, Miller WH, Jr. et al. Differentiation therapy of acute promyelocytic leukemia with tretinoin (alltransretinoic acid). N Engl J Med 1991; 324 : 138593.

96. Zheng Y, Kramer PM, Lubet RA et al. Effect of retinoids on AOMinduced colon cancer in rats: modulation of cell proliferation, apoptosis and aberrant crypt foci. Carcinogenesis 1999; 20 : 25560.

97. Wargovich MJ, Jimenez A, McKee K et al. Efficacy of potential chemopreventive agents on rat colon aberrant crypt formation and progression. Carcinogenesis 2000; 21 : 114955.

98. Balasubramanian S, Chandraratna RA, Eckert RL. Suppression of human pancreatic cancer cell proliferation by AGN194204, an RXRselective retinoid. Carcinogenesis 2004; 25 : 137785.

99. Pettersson F, Colston KW, Dalgleish AG. Retinoic acid enhances the cytotoxic effects of gemcitabine and cisplatin in pancreatic adenocarcinoma cells. Pancreas 2001; 23 : 2739.

100. Mollersen L, Paulsen JE, Olstorn HB et al. Dietary retinoic acid supplementation stimulates intestinal tumour formation and growth in multiple intestinal neoplasia (Min)/+ mice. Carcinogenesis 2004; 25 : 149 53.

101. Schug TT, Berry DC, Shaw NS et al. Opposing effects of retinoic acid on cell growth result from alternate activation of two different nuclear receptors. Cell 2007; 129 : 72333.

102. DiPoi N, Tan NS, Michalik L et al. Antiapoptotic role of PPARbeta in keratinocytes via transcriptional control of the Akt1 signaling pathway. Mol Cell 2002; 10 : 72133.

103. Koshiyama A, Ichibangase T, Imai K. Comprehensive fluorogenic derivatizationliquid chromatography/tandem mass spectrometry proteomic analysis of colorectal cancer cell to identify biomarker candidate. Biomed Chromatogr 2013; 27 : 44050.

104. Peters JM, Shah YM, Gonzalez FJ. The role of peroxisome proliferator activated receptors in carcinogenesis and chemoprevention. Nat Rev Cancer 2012; 12 : 18195.

105. McDonald SA, Preston SL, Lovell MJ et al. Mechanisms of disease: from stem cells to colorectal cancer. Nat Clin Pract Gastroenterol Hepatol 2006; 3: 26774.

106. Dalerba P, Dylla SJ, Park IK et al. Phenotypic characterization of human colorectal cancer stem cells. Proc Natl Acad Sci U S A 2007; 104 : 10158 63.

107. RicciVitiani L, Lombardi DG, Pilozzi E et al. Identification and expansion of human coloncancerinitiating cells. Nature 2007; 445 : 1115.

108. O'Brien CA, Pollett A, Gallinger S, Dick JE. A human colon cancer cell capable of initiating tumour growth in immunodeficient mice. Nature 2007; 445 : 10610.

109. Shmelkov SV, Butler JM, Hooper AT et al. CD133 expression is not restricted to stem cells, and both CD133+ and CD133 metastatic colon cancer cells initiate tumors. J Clin Invest 2008; 118 : 211120.

110. Arcaroli JJ, Powell RW, VarellaGarcia M et al. ALDH+ tumorinitiating cells exhibiting gain in NOTCH1 gene copy number have enhanced regrowth sensitivity to a gammasecretase inhibitor and irinotecan in colorectal cancer. Mol Oncol 2012; 6: 37081.

111. Dean M, Fojo T, Bates S. Tumour stem cells and drug resistance. Nat Rev Cancer 2005; 5: 27584.

112. Kao WH, Puddey IB, Boland LL et al. Alcohol consumption and the risk of type 2 diabetes mellitus: atherosclerosis risk in communities study. Am J Epidemiol 2001; 154 : 74857.

113. Xu F, Chen Y, Lv R et al. ALDH2 genetic polymorphism and the risk of type II diabetes mellitus in CAD patients. Hypertens Res 2010; 33 : 4955.

114. Zhang Y, Babcock SA, Hu N et al. Mitochondrial aldehyde dehydrogenase (ALDH2) protects against streptozotocininduced diabetic cardiomyopathy: role of GSK3beta and mitochondrial function. BMC Med 2012; 10 : 40.

115. Morita K, Saruwatari J, Miyagawa H et al. Association between aldehyde dehydrogenase 2 polymorphisms and the incidence of diabetic retinopathy among Japanese subjects with type 2 diabetes mellitus. Cardiovasc Diabetol 2013; 12 : 132.

116. Jerntorp P, Ohlin H, Almer LO. Aldehyde dehydrogenase activity and large vessel disease in diabetes mellitus. A preliminary study. Diabetes 1986; 35 : 2914.

117. Rovira M, Scott SG, Liss AS et al. Isolation and characterization of centroacinar/terminal ductal progenitor cells in adult mouse pancreas. Proc Natl Acad Sci U S A 2010; 107 : 7580.

118. Guerra C, Schuhmacher AJ, Canamero M et al. Chronic pancreatitis is essential for induction of pancreatic ductal adenocarcinoma by KRas oncogenes in adult mice. Cancer Cell 2007; 11 : 291302.

119. Ostrom M, Loffler KA, Edfalk S et al. Retinoic acid promotes the generation of pancreatic endocrine progenitor cells and their further differentiation into betacells. PLoS One 2008; 3: e2841.

120. Matsuse H, Shimoda T, Fukushima C et al. Screening for acetaldehyde dehydrogenase 2 genotype in alcoholinduced asthma by using the ethanol patch test. J Allergy Clin Immunol 2001; 108 : 7159.

121. Yoshida A, Huang IY, Ikawa M. Molecular abnormality of an inactive aldehyde dehydrogenase variant commonly found in Orientals. Proc Natl Acad Sci U S A 1984; 81 : 25861.

122. Linneberg A, GonzalezQuintela A, Vidal C et al. Genetic determinants of both ethanol and acetaldehyde metabolism influence alcohol hypersensitivity and drinking behaviour among Scandinavians. Clin Exp Allergy 2010; 40 : 12330.

123. Husemoen LL, Fenger M, Friedrich N et al. The association of ADH and ALDH gene variants with alcohol drinking habits and cardiovascular disease risk factors. Alcohol Clin Exp Res 2008; 32 : 198491.

124. Ma I, Allan AL. The role of human aldehyde dehydrogenase in normal and cancer stem cells. Stem Cell Rev 2011; 7: 292306.

125. Moreb JS, Ucar D, Han S et al. The enzymatic activity of human aldehyde dehydrogenases 1A2 and 2 (ALDH1A2 and ALDH2) is detected by Aldefluor, inhibited by diethylaminobenzaldehyde and has significant effects on cell proliferation and drug resistance. Chem Biol Interact 2012; 195 : 5260.

126. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using realtime quantitative PCR and the 2(Delta Delta C(T)) Method. Methods 2001; 25 : 4028.

127. Isse T, Oyama T, Kitagawa K et al. Diminished alcohol preference in transgenic mice lacking aldehyde dehydrogenase activity. Pharmacogenetics 2002; 12 : 6216.

128. Eriksson CJ, Mizoi Y, Fukunaga T. The determination of acetaldehyde in human blood by the perchloric acid precipitation method: the characterization and elimination of artefactual acetaldehyde formation. Anal Biochem 1982; 125 : 25963.

129. Hsu LC, Chang WC. Cloning and characterization of a new functional human aldehyde dehydrogenase gene. J Biol Chem 1991; 266 : 1225765.

130. Gerdes J, Lemke H, Baisch H et al. Cell cycle analysis of a cell proliferationassociated human nuclear antigen defined by the monoclonal antibody Ki67. J Immunol 1984; 133 : 17105.

131. Lassen N, Estey T, Tanguay RL et al. Molecular cloning, baculovirus expression, and tissue distribution of the zebrafish aldehyde dehydrogenase 2. Drug Metab Dispos 2005; 33 : 64956.

132. Yoshida A, Ikawa M, Hsu LC, Tani K. Molecular abnormality and cDNA cloning of human aldehyde dehydrogenases. Alcohol 1985; 2: 1036.

133. Goedde HW, Agarwal DP. Polymorphism of aldehyde dehydrogenase and alcohol sensitivity. Enzyme 1987; 37 : 2944.

134. Higuchi S, Muramatsu T, Shigemori K et al. The relationship between low Km aldehyde dehydrogenase phenotype and drinking behavior in Japanese. J Stud Alcohol 1992; 53 : 1705.

135. Cook TA, Luczak SE, Shea SH et al. Associations of ALDH2 and ADH1B genotypes with response to alcohol in Asian Americans. J Stud Alcohol 2005; 66 : 196204.

136. Kostromina E, Gustavsson N, Wang X et al. Glucose intolerance and impaired insulin secretion in pancreasspecific signal transducer and activator of transcription3 knockout mice are associated with microvascular alterations in the pancreas. Endocrinology 2010; 151 : 2050 9.

137. Akerblom B, Barg S, Calounova G et al. Impaired glucose homeostasis in Shb/ mice. J Endocrinol 2009; 203 : 2719.

138. Korinek V, Barker N, Morin PJ et al. Constitutive transcriptional activation by a betacateninTcf complex in APC/ colon carcinoma. Science 1997; 275 : 17847.

139. Mahmoudi T, Boj SF, Hatzis P et al. The leukemiaassociated Mllt10/Af10 Dot1l are Tcf4/betacatenin coactivators essential for intestinal homeostasis. PLoS Biol 2010; 8: e1000539.

140. Fujise T, Iwakiri R, Kakimoto T et al. Longterm feeding of various fat diets modulates azoxymethaneinduced colon carcinogenesis through Wnt/betacatenin signaling in rats. Am J Physiol Gastrointest Liver Physiol 2007; 292 : G11506.

141. Groden J, Thliveris A, Samowitz W et al. Identification and characterization of the familial adenomatous polyposis coli gene. Cell 1991; 66 : 589600.

142. Reya T, Clevers H. Wnt signalling in stem cells and cancer. Nature 2005; 434 : 84350.

143. Ireland H, Kemp R, Houghton C et al. Inducible Cremediated control of gene expression in the murine gastrointestinal tract: effect of loss of beta catenin. Gastroenterology 2004; 126 : 123646.

144. Korinek V, Barker N, Moerer P et al. Depletion of epithelial stemcell compartments in the small intestine of mice lacking Tcf4. Nat Genet 1998; 19 : 37983.

145. Pinto D, Gregorieff A, Begthel H, Clevers H. Canonical Wnt signals are essential for homeostasis of the intestinal epithelium. Genes Dev 2003; 17 : 170913.

146. Su LK, Kinzler KW, Vogelstein B et al. Multiple intestinal neoplasia caused by a mutation in the murine homolog of the APC gene. Science 1992; 256 : 66870.

147. Sansom OJ, Reed KR, Hayes AJ et al. Loss of Apc in vivo immediately perturbs Wnt signaling, differentiation, and migration. Genes Dev 2004; 18 : 138590.

148. Andreu P, Colnot S, Godard C et al. Cryptrestricted proliferation and commitment to the Paneth cell lineage following Apc loss in the mouse intestine. Development 2005; 132 : 144351.

149. Comprehensive molecular characterization of human colon and rectal cancer. Nature 2012; 487 : 3307.

150. Miyamoto S, Rosenberg DW. Role of Notch signaling in colon homeostasis and carcinogenesis. Cancer Sci 2011; 102 : 193842.

151. Qiao L, Wong BC. Role of Notch signaling in colorectal cancer. Carcinogenesis 2009; 30 : 197986.

152. Rodilla V, Villanueva A, ObradorHevia A et al. Jagged1 is the pathological link between Wnt and Notch pathways in colorectal cancer. Proc Natl Acad Sci U S A 2009; 106 : 631520.

153. Yu M, Trobridge P, Wang Y et al. Inactivation of TGFbeta signaling and loss of PTEN cooperate to induce colon cancer in vivo. Oncogene 2014; 33 : 153847.

154. Bellam N, Pasche B. Tgfbeta signaling alterations and colon cancer. Cancer Treat Res 2010; 155 : 85103.

155. Gires O. Lessons from common markers of tumorinitiating cells in solid cancers. Cell Mol Life Sci 2011; 68 : 400922.

156. Eirew P, Kannan N, Knapp DJ et al. Aldehyde dehydrogenase activity is a biomarker of primitive normal human mammary luminal cells. Stem Cells 2012; 30 : 3448.

157. Sanders MA, Majumdar AP. Colon cancer stem cells: implications in carcinogenesis. Front Biosci (Landmark Ed) 2011; 16 : 165162.

158. Vermeulen L, De Sousa EMF, van der Heijden M et al. Wnt activity defines colon cancer stem cells and is regulated by the microenvironment. Nat Cell Biol 2010; 12 : 46876.

159. Morin PJ, Sparks AB, Korinek V et al. Activation of betacateninTcf signaling in colon cancer by mutations in betacatenin or APC. Science 1997; 275 : 178790.

160. Rubinfeld B, Albert I, Porfiri E et al. Loss of betacatenin regulation by the APC tumor suppressor protein correlates with loss of structure due to common somatic mutations of the gene. Cancer Res 1997; 57 : 462430.

161. Orford K, Crockett C, Jensen JP et al. Serine phosphorylationregulated ubiquitination and degradation of betacatenin. J Biol Chem 1997; 272 : 247358.

162. MacDonald BT, Tamai K, He X. Wnt/betacatenin signaling: components, mechanisms, and diseases. Dev Cell 2009; 17 : 926.

163. Daniels DL, Weis WI. Betacatenin directly displaces Groucho/TLE repressors from Tcf/Lef in Wntmediated transcription activation. Nat Struct Mol Biol 2005; 12 : 36471.

164. Mao CD, Byers SW. Cellcontext dependent TCF/LEF expression and function: alternative tales of repression, derepression and activation potentials. Crit Rev Eukaryot Gene Expr 2011; 21 : 20736.

165. van der Flier LG, Clevers H. Stem cells, selfrenewal, and differentiation in the intestinal epithelium. Annu Rev Physiol 2009; 71 : 24160.

166. Noah TK, Shroyer NF. Notch in the intestine: regulation of homeostasis and pathogenesis. Annu Rev Physiol 2013; 75 : 26388.

167. Katoh M. Notch signaling in gastrointestinal tract (review). Int J Oncol 2007; 30 : 24751.

168. van Es JH, van Gijn ME, Riccio O et al. Notch/gammasecretase inhibition turns proliferative cells in intestinal crypts and adenomas into goblet cells. Nature 2005; 435 : 95963.

169. Fre S, Huyghe M, Mourikis P et al. Notch signals control the fate of immature progenitor cells in the intestine. Nature 2005; 435 : 9648.

170. Johnson SM, Gulhati P, Rampy BA et al. Novel expression patterns of PI3K/Akt/mTOR signaling pathway components in colorectal cancer. J Am Coll Surg 2010; 210 : 76776, 768.

171. Fresno Vara JA, Casado E, de Castro J et al. PI3K/Akt signalling pathway and cancer. Cancer Treat Rev 2004; 30 : 193204.

172. Milano J, McKay J, Dagenais C et al. Modulation of notch processing by gammasecretase inhibitors causes intestinal goblet cell metaplasia and induction of genes known to specify gut secretory lineage differentiation. Toxicol Sci 2004; 82 : 34158.

173. Wong GT, Manfra D, Poulet FM et al. Chronic treatment with the gamma secretase inhibitor LY411,575 inhibits betaamyloid peptide production and alters lymphopoiesis and intestinal cell differentiation. J Biol Chem 2004; 279 : 1287682.

174. Nakamura T, Tsuchiya K, Watanabe M. Crosstalk between Wnt and Notch signaling in intestinal epithelial cell fate decision. J Gastroenterol 2007; 42 : 70510.

175. Chen CR, Kang Y, Massague J. Defective repression of cmyc in breast cancer cells: A loss at the core of the transforming growth factor beta growth arrest program. Proc Natl Acad Sci U S A 2001; 98 : 9929.

176. Girish V, Vijayalakshmi A. Affordable image analysis using NIH Image/ImageJ. Indian J Cancer 2004; 41 : 47.

177. Ball HC, Holmes RK, Londraville RL et al. Leptin in whales: validation and measurement of mRNA expression by absolute quantitative realtime PCR. PLoS One 2013; 8: e54277.

178. Gustavson MD, Crawford HC, Fingleton B, Matrisian LM. Tcf binding sequence and position determines betacatenin and Lef1 responsiveness of MMP7 promoters. Mol Carcinog 2004; 41 : 12539.

179. Visvader JE, Lindeman GJ. Cancer stem cells in solid tumours: accumulating evidence and unresolved questions. Nat Rev Cancer 2008; 8: 75568.

180. Erisman MD, Scott JK, Watt RA, Astrin SM. The cmyc protein is constitutively expressed at elevated levels in colorectal carcinoma cell lines. Oncogene 1988; 2: 36778.

181. He TC, Sparks AB, Rago C et al. Identification of cMYC as a target of the APC pathway. Science 1998; 281 : 150912.

182. Shtutman M, Zhurinsky J, Simcha I et al. The cyclin D1 gene is a target of the betacatenin/LEF1 pathway. Proc Natl Acad Sci U S A 1999; 96 : 55227.

183. Schewe DM, Leupold JH, Boyd DD et al. Tumorspecific transcription factor binding to an activator protein2/Sp1 element of the urokinasetype plasminogen activator receptor promoter in a first large series of resected gastrointestinal cancers. Clin Cancer Res 2003; 9: 226776.

184. Garneau H, Paquin MC, Carrier JC, Rivard N. E2F4 expression is required for cell cycle progression of normal intestinal crypt cells and colorectal cancer cells. J Cell Physiol 2009; 221 : 3508.

185. Qian J, Liu H, Chen W et al. Knockdown of Slug by RNAi inhibits the proliferation and invasion of HCT116 colorectal cancer cells. Mol Med Rep 2013; 8: 10559.

186. Song LJ, Liu RJ, Zeng Z et al. Gastrin inhibits a novel, pathological colon cancer signaling pathway involving EGR1, AE2, and PERK. J Mol Med (Berl) 2012; 90 : 70718.

187. Suzuki H, RomanoSpica V, Papas TS, Bhat NK. ETS1 suppresses tumorigenicity of human colon cancer cells. Proc Natl Acad Sci U S A 1995; 92 : 44426.

188. Modak C, Chai J. Serum response factor: look into the gut. World J Gastroenterol 2010; 16 : 2195201.

189. Luca AC, Mersch S, Deenen R et al. Impact of the 3D microenvironment on phenotype, gene expression, and EGFR inhibition of colorectal cancer cell lines. PLoS One 2013; 8: e59689.

190. Yang L, Ren Y, Yu X et al. ALDH1A1 defines invasive cancer stemlike cells and predicts poor prognosis in patients with esophageal squamous cell carcinoma. Mod Pathol 2013: 19.

191. Alam M, Ahmad R, Rajabi H et al. MUC1C oncoprotein activates ERK >C/EBPbeta signaling and induction of aldehyde dehydrogenase 1A1 in breast cancer cells. J Biol Chem 2013; 288 : 30892903.

192. Bartley AN, Parikh N, Hsu CH et al. Colorectal adenoma stemlike cell populations: associations with adenoma characteristics and metachronous colorectal neoplasia. Cancer Prev Res (Phila) 2013; 6: 116270.

193. Alamgeer M, Ganju V, Szczepny A et al. The prognostic significance of aldehyde dehydrogenase 1A1 (ALDH1A1) and CD133 expression in early stage nonsmall cell lung cancer. Thorax 2013; 68 : 1095104.

194. Li X, Wan L, Geng J et al. Aldehyde dehydrogenase 1A1 possesses stem like properties and predicts lung cancer patient outcome. J Thorac Oncol 2012; 7: 123545.

195. Liu W, Dong X, Mai M et al. Mutations in AXIN2 cause colorectal cancer with defective mismatch repair by activating betacatenin/TCF signalling. Nat Genet 2000; 26 : 1467.

196. Jho EH, Zhang T, Domon C et al. Wnt/betacatenin/Tcf signaling induces the transcription of Axin2, a negative regulator of the signaling pathway. Mol Cell Biol 2002; 22 : 117283.

197. Takahashi H, Ishii H, Nishida N et al. Significance of Lgr5(+ve) cancer stem cells in the colon and rectum. Ann Surg Oncol 2011; 18 : 116674.

198. Chen C, Cai S, Wang G et al. cMyc enhances colon cancer cellmediated angiogenesis through the regulation of HIF1alpha. Biochem Biophys Res Commun 2013; 430 : 50511.

199. Hongxing Z, Nancai Y, Wen S et al. Depletion of cMyc inhibits human colon cancer colo 320 cells' growth. Cancer Biother Radiopharm 2008; 23 : 22937.

200. Barker N, van Es JH, Kuipers J et al. Identification of stem cells in small intestine and colon by marker gene Lgr5. Nature 2007; 449 : 10037.

201. Fre S, Pallavi SK, Huyghe M et al. Notch and Wnt signals cooperatively control cell proliferation and tumorigenesis in the intestine. Proc Natl Acad Sci U S A 2009; 106 : 630914.

202. Gutierrez A, Look AT. NOTCH and PI3KAKT pathways intertwined. Cancer Cell 2007; 12 : 4113.

203. Palomero T, Sulis ML, Cortina M et al. Mutational loss of PTEN induces resistance to NOTCH1 inhibition in Tcell leukemia. Nat Med 2007; 13 : 120310.

204. Baan R, Straif K, Grosse Y et al. Carcinogenicity of alcoholic beverages. Lancet Oncol 2007; 8: 2923.

205. Simanowski UA, Stickel F, Maier H et al. Effect of alcohol on gastrointestinal cell regeneration as a possible mechanism in alcohol associated carcinogenesis. Alcohol 1995; 12 : 1115.

206. Roy HK, Gulizia JM, Karolski WJ et al. Ethanol promotes intestinal tumorigenesis in the MIN mouse. Multiple intestinal neoplasia. Cancer Epidemiol Biomarkers Prev 2002; 11 : 1499502.

207. Manna SK, Tanaka N, Krausz KW et al. Biomarkers of Coordinate Metabolic Reprogramming in Colorectal Tumors in Mice and Humans. Gastroenterology 2014.

208. Baron JA, Cole BF, Sandler RS et al. A randomized trial of aspirin to prevent colorectal adenomas. N Engl J Med 2003; 348 : 8919.

209. Cole BF, Baron JA, Sandler RS et al. Folic acid for the prevention of colorectal adenomas: a randomized clinical trial. JAMA 2007; 297 : 23519.

210. Grau MV, Sandler RS, McKeownEyssen G et al. Nonsteroidal anti inflammatory drug use after 3 years of aspirin use and colorectal adenoma risk: observational followup of a randomized study. J Natl Cancer Inst 2009; 101 : 26776.

100