Eastern Illinois University The Keep

Masters Theses Student Theses & Publications

2019 CREG1: Its Role as a Master Regulator of Liver Function Iffat Jahan Eastern Illinois University

Recommended Citation Jahan, Iffat, "CREG1: Its Role as a Master Regulator of Liver Function" (2019). Masters Theses. 4477. https://thekeep.eiu.edu/theses/4477

This Dissertation/Thesis is brought to you for free and open access by the Student Theses & Publications at The Keep. It has been accepted for inclusion in Masters Theses by an authorized administrator of The Keep. For more information, please contact [email protected]. TheGraduate School� E>srf.RN111 1�11\USl\ 'F.R.S1n·

Thesis Maintenance and Reproduction Certificate

FOR: Graduate Candidates Completing Theses in Partial Fulfillment of the Degree Graduate Faculty Advisors Directing the Theses

RE: Preservation, Reproduction, and Distribution of Thesis Research

Preserving, reproducing, and distributing thesis research is an important part of Booth Library's responsibility to provide access to scholarship. In order to further this goal, Booth Library makes all graduate theses completed as part of a degree program at Eastern Illinois University available for personal study, research, and other not-for­ profit educational purposes. Under 17 U.S.C. § 108, the library may reproduce and distribute a copy without infringing on copyright; however, professional courtesy dictates that permission be requested from the author before doing so.

Your signatures affirm the following:

• The graduate candidate is the author of this thesis.

•The graduate candidate retains the copyright and intellectual property rights associated with the original research, creative activity, and intellectual or artistic content of the thesis.

•The graduate candidate certifies her/his compliance with federal copyright law (Title 17 of the U.S. Code) and her/his right to authorize reproduction and distribution of all copyrighted materials included in this thesis.

• The graduate candidate in consultation with the faculty advisor grants Booth Library the nonexclusive, perpetual right to make copies of the thesis freely and publicly available without restriction, by means of any current or successive technology, including but not limited to photocopying, microfilm, digitization, or internet.

•The graduate candidate acknowledges that by depositing her/his thesis with Booth Library, her/his work is available for viewing by the public and may be borrowed through the library's circulation and interlibrary loan departments, or accessed electronically. The graduate candidate acknowledges this policy by indicating in the foll7g manner: ___Yes, I wish to make accessible this thesis for viewing by the public

___No, I wish to quarantine the thesis temporarily and have included the Thesis Withholding Request Form

• The graduate candidate waives the confidentiality provisions of the Family Educational Rights and Privacy Act (FERPA) (20 U. S. C. § 1232g; 34 CFR Part 99) with respect to the contents of the thesis and with respect to information concerning authorship of the thesis, including name and status as a student at Eastern Illinois University. I have conferred with my graduate faculty advisor. My signature below indicates that I have read and agree with the above statements, and hereby give my permission to allow Booth Library to reproduce and distribute my thesis. My adviser's signature indicates concurrence to reproduce and distribute the tl),e�s.

uraouate 1..anu1udlt:.>1�11dLu1e

IFFAT JAl·MN Printed Name

�-lw, M Sc.Wn<.g. Graduate Degree Program Date

Please submit in duplicate. CREG1: ITS ROLE AS MASTER REGULATOR

OF LIVER FUNCTION

(TITLE)

BY

IFFAT JAHAN

THESIS

SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF Masters of Science

IN THE GRADUATE SCHOOL, EASTERN ILLINOIS UNIVERSITY CHARLESTON, ILLINOIS

2019

YEAR

I HEREBY RECOMMEND THAT THIS THESIS BE ACCEPTED AS FULFILLING THIS PART OF THE GRADUATE DEGREE CITED ABOVE

f4L/? DATE (lµfDATE !f!/11 !f DATE DA�f

THESIS COMMITIEE MEMBER DATE THESIS COMMITIEE MEMBER DATE CREG1: Its Role as a Master

Regulator of Liver Function

Iffat Jahan

Department of Biological Science

Eastern Illinois University

Charleston, IL-61920

2019 © 2019 by I ffat J ahan

ii COMMITTEE IN CHARGE OF CANDIDACY

Gary A. Bulla, Ph.D. Advisor

Anthony Olouch, Ph.D.

Britto Nathan, Ph.D.

iii Abstract

Mammalian development involves a complex system of regulatory signals and reactions resulting in highly differentiated cell types with specific structure and function. The liver is a major organ that has been studied extensively to understand underlying genetic processes responsible for specification, establishment and maintenance of tissue identity. Hepatoma and hepatoma variant cell lines have been used as models to understand genetic networks responsible for liver function. Whole genome microarray analysis of hepatocyte cell lines has revealed candidate genes that may serve as regulators or master regulators of liver specificity. In two previous studies in our lab, the role of candidate gene Cellular Repressor of EIA Stimulated

Gene (CREG1) in regulation of liver-specific gene expression was determined using transfection studies combined with utilization of quantitative real-time polymerase chain reaction (qRTPCR).

Both studies found strong activation (10-100-fold) of transcription factor Hn/6 and the downstream gene Serpinal (a marker gene used to identify liver function). These findings suggested that CREG 1 might act through Hnf6 to regulate Serpinal via a Locus Control Region

(LCR). However, while one study showed that CREGI overexpression in a hepatoma variant cell line resulted in modest activation of liver-specific transcription factors Hnfl, Hn/4 and Hnf3, the second study suggested robust activation of these genes as measured by qRT-PCR. In the current study, a CREG1 expression vector was reintroduced into one of the variant cell lines (H 11) and gene activation profiles monitored to establish validity of the previous studies. However, we postulated that CREG-1 overexpression, validated by qRT-PCR, can activate expression of hepatocyte transcription factor Hn/6 and the downstream gene Serpinal along with other liver­ specific downstream genes. Results showed that CREGl can fully rescue expression of liver-

iv specific genes including transactivation genes, suggesting its role as a master regulator of liver function. CREG l action appears to act at least partially through HNF6 gene activation, as well as through activation of HNF4/HNF1 pathway, both of which act to increase expression of the serpin locus through LCR activation through HNF6.

v Dedicated to My Parents

vi Acknowledgement

I would like to express my sincere gratitude to all individuals who helped me in completing this project while working on my degree. It would not have been possible without their constant support, encouragement and help in facilitating this work. It is a pleasure to acknowledge them for their motivation, guidance and numerous suggestions.

I express my deep sense of thanks and gratitude to my advisor and guide Dr. Gary A.

Bulla, whose keen interest and, above all, his overwhelming attitude in helping his students. He has been largely responsible for my completing of this work. His timely advice, meticulous scrutiny, scholarly advice and scientific approach have helped me to accomplish this work in an interesting and expanding field.

I am thankful to Dr. Britto P Nathan for providing indispensable advice, information and support on several aspects of my project while working in laboratory, and who make me look at research and my work in different ways.

I owe a deep sense of gratitude to Dr. Anthony Oluoch for his keen interest at every stage of my research. His timely suggestions along with with kindness, enthusiasm and dynamism have enabled me to complete my thesis. Moreover, for providing an opportunity to serve as a teaching assistant in his class and helping me to learn the art of teaching and learning at the same time.

The assistance, co-operation and experience of my fellow lab members was essential for completion of this project. Of course, family and friends who were there to give me the extra push necessary to succeed. Last, but not least, I am thankful to the Department of Biological

Sciences and EasternIllinois University forthis opportunity to be trained and to be an alumnus.

vii Table of Contents

Abstract ...... iv Acknowledgement ...... vi Table of Contents ...... vii List of Figures ...... ix List of Tables ...... x

Chapter 1 : Introduction ...... 1

1.1. Liver Development ...... !

1.2. Cell Culture Model Systems ...... 5

1.3. Serpinal Gene Regulation As A Model For Liver Gene Regulation ...... 7

1.4. DNA Microarray Analysis ...... 9

1.5. Project Overview And Goals ...... 10

Chapter 2: Materials and Methods ...... 16

2.1 . Cell Culture ...... 16

2.2. Cell Transfection ...... 16

2.3. RNA Isolation ...... 18

2.4. Reverse Transcription Polymerase Chain Reaction (RT-PCR) ...... 18

2.5. Quantitative-Polymerase Chain Reaction (qPCR) ...... 19

2.6. Statistical Analysis ...... 21

Chapter 3: Results...... 22

3.1. Part A. CREGlAs A Candidate forLiver Gene Activation ...... 22

3.1.1. Stable Transfection and Expression ofCREGI Clones ...... 22

3.1.2. Overexpression Effects of CREG1 On Transcription Factors Expression .. 24

3.1.3. CREGI Overexpression - Effect on Down Stream Genes ...... 28

3.1.4. CREGI Overexpression - Effect on SerpinA Gene Cluster ...... 28

Chapter 4: Discussion ...... 40

4. 1. CREG1 As A Candidate forLiver Gene Activation ...... 40

4. 2. Future Direction ...... 45

Chapter 5: References ...... 46

viii List of Figures

1. Derivation of the hepatoma variant cell lines ...... 7

2. Genes repressed and activated in hepatoma variant cells using Illumina whole genome rat

microarray ...... 12

3. Candidate genes predicted to be involved in regulation of liver phenotype ...... 13

4. Flowchart showing schematic representation of the strategy ...... 15

5. True Mouse Clone Vector ...... 31

6. Expression of CREG 1 in transfected variants H11 cells ...... 32

7. CREG1 activates Hnfl and Hnf4 in the H11 cells ...... 30

8. CREG 1 overexpression effects in transfected H11 cells ...... 31

9. CREG1 activates Hnf6 in the HI I cells ...... 32

10. CREG1 activates liver downstream hepatic genes in the HI I cells ...... 33

11. Downstream genes required for cellular maintenance, proliferation and differentiation

were tested in the H11 -CREG 1 cells...... 34

12. CREGl activates SerpinAl expression in the HI 1 cells Downstream gene Serpinal (a

liver phenotypic biomarker) was strongly activated...... 35

13. CREG 1 over-expression activates serpin cluster genes in transfected Hl1 cells ...... 36

ix List of Tables

1. List of candidate genes identifiedfrom repressed genes ...... 12

2. List of candidate genes identified from activated genes ...... 13

3. Primers used in the qPCR, primer sequences, melting and annealing temperatures of

primers ...... 20

4. Expression levels of candidate gene by microarray analysis ...... 26

5. Expression levels of transcription factors in hepatoma and hepatoma variant cell lines as

determined by micro array data ...... 26

6. H1 I - CREG I transfectants...... 28

x Chapter 1

Introduction

The mammalian liver plays a critical role in the human digestive system. As with all mammalian organ development, a complex array of regulatory signals and responses are involved, which are dependent on positive and negative regulation. These signals drive establishment of cell specialization through transcriptional regulation, and regulation of tissue specific gene expression is controlled by trans-acting factors interacting with cis-acting elements within DNA sequences (2). Still the mechanisms of tissue differentiation are not clear. To understand genetic processes which are required for specification, establishment and maintenance of tissue identity, extensive studies have been done. One model with which to study cellular differentiation is the mammalian hepatocyte and/or hepatoma cells. Large numbers of liver enriched genes have been characterized- some of which are controlled by common regulatory pathways, and others are controlled by characterized master regulators.

1.1. Liver Development

As an essential organ in the human digestive system, the mammalian liver has important role in endocrine and exocrine functions. These functions include detoxification, metabolism

(36), glycogen storage, secretion of digestive fluids, maintenance of homeostasis (39), regulation of cholesterol synthesis and transport, metabolism of urea and secretion of plasma proteins such as apo-lipoproteins and albumin. Liver diseases such as hepatic fibrosis, cirrhosis, hepatitis and hepatocellular carcinoma (HCC) are characterized by loss of liver function and are associated with high mortality rates ( 40). The liver is composed of various differentiated cell types: hepatocytes, cholangiocytes (bile duct epithelial cells), Kupffer cells (liver resident macrophages), hepatic stellate cells, and sinusoidal endothelium and pit cells (natural killer cells) (31 ). Each cell type has specific functionsduring proliferation and regeneration of the liver after hepatectomy and injury.

Multiple stages are involved in liver development. An array of genes is expressed by the liver in order to control metabolism and catabolism, with processes ranging from enzymatic to hematopoetic components, driven by liver enriched transcription factors (TFs) (31 ). The most common markers for liver identity include albumin, transthyretin and a-fetoprotein (37), each of which is controlled by liver-enriched TFs. For example, the promotor region of the albumin gene binds transcription factors Forkhead Box Protein A (FoxA) and GATA4. This binding results in the opening of chromatin to allow binding of Hepatocyte Nuclear Factor la (HNFla) and CCAAT-enhancer binding protein p(C/EBPP) to initiate transcription of the albumin gene

(37). However, FoxA2 gene expression must first be activated by expression of Hepatocyte

Nuclear Factor- IP (HNF IP) in early embryonic development, and impairment of this activation step results in absence of hepatic specification due to lack of competence by endodermal cells

(37). Hormonal factors are also required for interaction of intercellular and matrix-cellular proteins, and are essential for liver ontology which starts at embryonic day 9 in the mouse (22).

The earliest events in the initiation of liver generation depend on interaction of cardiogenic mesoderm and foregut endoderrn, which initiates differentiation, as in tissue recombination experiments (12, 22). Also required is the interaction of cardiogenic growth factor (FGF) and bone-morphogenic proteins (BMP2 and BMP4), produced by cardiogenic mesoderm and septum transversum respectively (37).

Competent endoderm cells then respond to fibroblast growth factor (FGF) produced by cardiogenic mesoderm cooperating with bone-morphogenic proteins (BMP2 and BMP4) produced by the septum transversum, resulting in expression of hepatic genes (37).

2 An inductive signal is introduced generating liver bud formation, which is then transduced to liver primodium (12, 22). Interkinetic nuclear migration, proliferation and differentiation of hepatoblasts and expression of homeobox factors Hex and GATA-6 are induced by that signal (3 7).

To generate a functional liver, a process of maturation occurs in cells, which include physiologic functions and morphologic changes, exclusively dependent on precise control and co-ordination of genetic networks. Furthermore, during embryogenesis a dynamic transcriptional network is involved. This network helps establish regulatory loops (cross regulatory and auto regulatory), with the help of liver enriched transcription factors (37). This phenomenon is observed by in vivo analysis of DNA -protein interactions that shows the complex and myriad of interactions of cis-acting hepatocyte-specific DNA regulatory regions (21). Tissue specific expression of hepatic and pancreatic regulation is dependent on liver-specific TFs (26).

Critical proteins involved in liver identity include HNF-la (a homeodomain protein), HNF-1 p,

HNF3P!FoxA2 (a winged helix factor), HNF-4a.l (an orphan nuclear receptor), HNF-

6/0NECUTl (a member of one cut homeodomain protein) and LRH-1 ( 42, 26, 37) and CCAAT enhancer binding protein (C/EBP) (37), are essential for phenotypic character for liver and tightly regulated by TFs.

Gene-regulatory regions cooperatively work with cofactors for stimulating transcription of common targets. However, liver-specific TFs are interdependent and important for basic functioning of liver tissue (26). Biochemical fu nctions such as gluconeogenesis, regulation of carbohydrate synthesis and storage, detoxification and synthesis of serum proteins, complement factors and coagulation factors (26). HNFI a has a partial role in hepatic phenotype maintenance, acts through promoters of more than 222 different liver specific genes (26). Moreover,

3 involvement of HNFI � in transcriptional regulation of genes responsible for bile acid sensing and fatty acid oxidation (37).

Along with HNFl, HNF3WFoxA2 controls the regulation of hepatic function as it has overlapping DNA binding properties (37). Similarly, liver receptor homolog-1 (LRH-1) is associated with a nuclear hormone receptor family, that has control over bile acid and cholesterol metabolism (37).

Along with these, Hepatocyte Nuclear Factor 4a (HNF4a) has a significant impact on the liver phenotype. In hepatocytes, HNF4a has been shown to bind to 1262 genes (26). HNF4a is essential for transcriptional regulation as it binds to RNA polymerase II, a process which has been confirmed by transcriptome analysis (26). Dysfunction of HNF4al or HNFla leads to maturity onset diabetes of the young (MODY), which indicates problems in glucose metabolism.

HNF6binds to 227 hepatocyte genes and 189 genes in pancreatic islet cell(26). It also mediates the effects of growth hormone, inhibits glucocorticoid activity, and stimulates expression of genes in the gluconeogenic, glycolytic and bile acid synthesis pathways (37). Also, HNF6 is required for hepatocyte proliferation and feed forward loop switch (26). Hepatic transcriptional regulation is dependent on the synergistic association among HNFla, HNF4al and HNF6 (26).

Developmental and metabolic functions in hepatocytes are regulated by these HNF's genes that encode transcription factors and cofactors (17). From transcriptional studies it is known that

HNFla and HNF4al create a multicomponent loop which forms a bi-stable system to switch between two alternate states and establish a feedback control mechanism, as they occupy each other's promoters in hepatocytes (17).

Therefore, maintenance of terminal phenotype of hepatocytes is dependent on several

HNF's - HNF6, HNFla and HNF4al, which form multi-input motifs. Binding of these HNF's

4 plays a vital role in regulation and the tissue specific expression of the downstream genes (26).

Finally, development, differentiation and homeostasis are controlled by transcription regulators through tissue specific gene expression, ensuring the simultaneous repression of unnecessary and conflicting genes (32).

1.2. Cell Culture Model Systems

Different techniques are used to study live gene hierarchies. One of the techniques is cell fusion. In cell fusion techniques, two distinct and highly differentiated cell types such as hepatoma cells and fibroblast cells are fused to generate somatic cell hybrids. These cannot express liver specific genes in a process known as extinction (4). This extinction process can sometimes be reversed if chromosomes from one of the parents are lost during cell division. In most cases fibroblast chromosomes are lost (4). Though tissue specific genes do not express, housekeeping genes of both the cell types are expressed.

Another method is used to develop variants, which are particularly used to understand the driving force of liver identity, particularly mutant hepatocytes identification (9). Variants have multiple genetic or epigenetic lesions which turns in de-differentiation. To reveal the effects of

HNFs in hepatocytes such as - activation and transcriptional regulation of downstream genes, it is essential to observe screening of these variants and expression profile (24). These studies have impact over understanding the mechanism of HNFs - how they react in regulatory pathways when there is a defect and how they act while maintaining the liver phenotype (13). The rescue of the liver phenotype in hepatoma variant cells has revealed the understanding of liver genetics.

Both positive and negative selection strategies were considered to get hepatoma variants from rat hepatoma cells, which are different in morphology and liver specific gene expression (9). Studies

s show that variants do not express HNF4 and HNFlcx genes (13). Ectopic expressions of HNF4 and/or HNFlcxrescue each other as well as downstream liver genes. This study suggests that its important in transcriptional regulation of hierarchical pathways (9).

In the current project, dedifferentiated cells are generated from the original rat tumor cell line H4IIEC3. This hepatoma cell line is the parent cell line of Fg14 cells, which is derived by selection against integrated trangsgenes expressing adenosine phosphoribosyl transferase

(APRT) and hypoxanthine phosphoribosyl transferase (HPRT/GPT) under control of the liver specific human SERPINAJ (aka alpha-I antitrypsin) promoter.

Transfected cells were selected by growth in media containing drugs inhibiting the de novo nucleotide biosynthesis and require nucleotide precursors that can be utilized by transgene expression: adenine-aminopetrin-thymidine (AAT) for SERPINAJ-aprt expression and hypoxanthine-aminopetrin-thymidine (HAT) for SERPINAJ-gpt expression (Fig. 1).

Hepatoma variant cell lines (Fig. I) were generated by negative selection, using toxic analogs-

2.6-diaminopurine (DAP) and 6-thioxanthine (6-TX). The surviving cells lack APRT and HPRT enzyme production as well as lack expression of the endogenous Serpinal gene (9, 13, 17).

6 1. SERPINA 1-Aprt 2. SERPINA 1-Gpt

Fado-2 L. Fg-14 --+ H11, M38 HS2, M29 AAT, OAP , Variants HAT + 6TX APRT­ APRT APRT­ + GPT­ GPT GPT­ + + SERPINA1 SERPINA1 SERPINA1 -

Fig.1. Derivation of the hepatoma variant cell lines. Fgl4 cells contain two selectable trans­ genes - adenine-aminopterin-thymidine (AAT) and hypoxanthine-aminopetrin-thymidine (HAT) under the control of the liver-specific human SERPINAJ gene promoter. Selection against - 2,4- diaminopurine(DAP) and 6-thioxanthine (6TX), expression of these genes resulted in the isolation of hepatoma variant cell lines (e.g. HS2, M38, Hl 1 and M29) which lack expression of several liver genes. (APRT- Adenine phosphoribosyJ transferase; OPT- Guanosine phosphoribosyl transferase).

1.3. Serpinal Gene Regulation as A Model for Liver Gene Regulation

The SERPINAJ gene is highly conserved in the liver has been studiedto understand the molecular mechanism of transcriptional regulation in this organ (9) For example, gene silencing in hepatoma variants cell lines has been shown to be due to loss of trans-acting factors HNF 1a andHNF4a.

In this current study, the hepatoma variant cell lines Hl l, M29, M38 and HS2 are deficient in HNF4a /HNF a transactivation pathway, which results in the loss of expression of many liver specific downstream genes including SERPINAJ.

7 SERPINAJ is a member of serine-proteinase inhibitor SERPIN super family (45). It is an abundantly produced circulating protease inhibitor, encoded by a single gene on the long arm of human chromosome 14 (23). In 1955, Schultze isolated these highly polymorphic 55 kda serum glycoproteins. They occur in the alpha-I globulin fraction (45) and have ability to inhibit the destructive neutrophil-secreted protease, elastase, and cathepsin G ( 19). SERP !NA J is principally synthesized (70-80 %) in liver cells (45) but also produced by pulmonary alveolar cells, macrophages, monocytes and intestinal epithelial cells (19, 45). The SERPINAJ half-life is 3-5 days with normal plasma concentration ranging from 0.9-1.75 g/L. Auto- regulation of

SERPINAJ was observed when exposed to a SERPINAJ-elastase complex or elastase alone.

Also, SERPINAJ was shown to be synthesized de novo in human cancer cells (45).

Through in-situ hybridization studies, it has been shown that hepatocytes and hepatoma cell lines synthesize SERPINAJ in ample amounts in cell culture (19). SERPINAJ is an ideal model to study tissue specific gene expression and extinction because of the following characteristics: (i) liver enriched trans-acting factors HNFl a and HNF4 binding site at the

proximal promoter of SERPINAI has been well characterized; (ii) in hepatoma x fibroblast hybrids, SERPINAJ gene expression is reduced by> 1000 fold; (iii) the genes encoding HNFla and HNF4 are highly characterized and (iv) tissue specific expression of a gene can be obtained by utilization of only the 5' proximal promoter of the SERPINA 1 gene (13). The expression of the SERP!NA 1 gene is regulated synergisticalJy by hepatocyte nuclear factor-I a (HNFl a) and

HNF4 (19). Mutations in the SERPINAJ promoter region prevent binding of HNFla and reduce cellular expression (24). Previous studies have shown that ectopic expression of trans-acting elements results in chromatin remodeling around the Serpinal locus, suggesting that a single

8 transcription factor can change the chromatin structure of DNA and rescue of hepatic expression in hepatoma variants (14).

1.4. DNA Microarray Analysis

The DNA microarray, a tool initially developed in -1987, provides a high throughput data platform with which to analyze gene expression of most or all genes within a cell type or tissue. This technique provides whole gene expression profiling (i.e. the quantitative data of an expression level of thousands of genes simultaneously under different conditions). For example, the expression pattern of the same gene under different times and conditions can be monitored to elucidate the role of a given gene in genetic pathways ( 15, 18). The process of gene clustering is based upon regulatory pathways or disease association. Various methods that have been used for analysis, including hierarchical clustering, k-means, partitions around memoids (PAM a.ka K­ memoids), self-organizing maps (SOM), mixed model-based clustering and tight clustering (33).

DNA microarrays have become highly utilized to determine differential gene expression patterns or to compare the differences in mRNA expression between similar cells subjected to different stimuli or between developmental stages or different cellular phenotypes.

Analysis of expression arrays is based upon successful implementation, development and normalization of fluorescent signals which is broadly divided into three different stages: (I) array fabrication (II) probe preparation and hybridization and (Ill) data collection, normalization and analysis (15). The methods of target-sample preparation generally include generation of the first-strand of cDNAs from RNA that are labelled or tagged with a capture sequence, or by generating antisense RNA (aRNA) via an in-vitro transcription reaction (18).

A common type of microarray chip is constructed on a derivatized glass microscope slides by arraying PCR amplified cDNA clones or genes at high density. Expression sequence tags (ESTs)

9 are extensively used for the gene identification, gene discovery and mapping in human and other organism, which are single pass partial sequences of cDNA clones. Representative unique transcripts of cDNA clones are then selected for analyses. The cDNA clone inserts from cultured cells are typically amplified by PCR using universal amplification primers. Products are then purified forthe efficient binding of the amplified clone inserts onto the microscopic slide. A high speed robotic system is then used to print or coat the amplified PCR product suspended in a high-salt denaturing buffer and applied to poly-L-lysine or aminosilane-coated glass microscopic slides. RNA from different sources is typically labelled with fluorescent dyes such as Cy3- or

Cy5- dUTP for co-hybridization experiments. The microscopic slide is pre-hybridized with washing buffer for removal of unbound DNA impurities and then hybridized with the labelled probe. Differential gene expression level is measured by using a confocal laser scanner, capable of distinguishing signals generated between Cy3 and Cy5 labeled probes. An image is generated and processed based upon the identifying and distinguishing arrayed genes against spurious signal, estimation of background signal and the background -subtracted hybridized intensities.

The generated data is normalized by adjusting the differences in labeling and detection efficiencies of fluorescent labels and quantity of the starting RNA between the assayed samples

(15). Finally, data is analyzed and difference in expression level is determined based upon the cut-off value of specific threshold levels.

1.5. Project Overview and Goals

In order to elucidate the pathways for normal regular liver function in hepatoma variant cells, whole genome expression profiling approach was carried out in this laboratory in order to identify the liver specific genes that are responsible for activation or repression of gene

10 expression. Four variant cell lines (M29, HS2, Hl 1, and M38), were tested against the parental

hepatoma cell line Fg 14. Whole genome microarray experiments were conducted using the RNA

isolated from parental and variant cell lines. The experiment was performed in triplicate for

reproducibility and viability and tested by comparing against the biological replicates for each

cell line and gene variability within the microarray.

A total of -22,500 target genes were analyzed. Based upon the generated data two criteria

were implemented when comparing the gene expression between the parental hepatoma cell and

the hepatoma variants: (i) a > 5-fold signal variability difference within replicates of parental

against variant cell lines and (ii) at least two of the four variant cell lines tested must show these

> 5-fold differences in expression. This led to the identification of 355 repressed genes, 132 of

which were common in all of them. Also, 213 genes were found to be activated among the

variant cell lines, although only two genes were activated in all four variant cell lines, as shown

in the Fig. 2. Further analysis was done to identify those repressed/activated genes that are either

transcription factors or encode products that play a known role in signaling pathways. It was

hypothesized that at least some of these genes are regulated by liver specific master regulators

such as HNFla, HNF3p (FOXA2), HNF4 and HNF6 (ONECUTI). From the pool of these

genes, candidate genes were identified using two criteria: (i) a > 5 fold increase or decrease in

expression in 2 of 4 variants compare to parental Fg14 cells and (ii) a known role in

transcriptional activation or interaction with transcriptional activation pathways. Based upon the

criteria, 14 repressed genes (Table 1) and 7 activated genes (Table 2) were identified as

potentially having a capacity to contribute in maintaining the liver (hepatic) phenotype (Fig. 3).

These preliminary data suggest that a high level of reprogramming has occurred in variant cells,

11 leading to activation of few sub-sets of genes along with loss of liver-specific gene expression

(47).

> S·folciRepr=ed

M38 Hll M38 Hll

HS2 M29

Fig. 2. Gene repressed and activated in hepatoma variants cells using Illumina whole genome rat rnicroarray. A total of 355 genes were identified the cut-off value of > 5 folds repression compared to Fgl4 levels, 132 genes were shared in repressed in all four variant cell lines whereas only 2 genes are activated by> 5 fold in all 4 variant cell lines.

Table.I. List of candidate genes identified from list of repressed genes.

Fold Repression Gene name Description M38 HS2 Hll M29 6.2 I.7 3.6 5.5 BHLBH2 Basic helix-loop-helix domain containing Class 2B

12.9 15.0 14.6 13.7 CREB3L3 cAMP responsive element binding protein 3-like 3

3.6 6.4 2.7 5.9 CREG EI A-stimulated gene (predicted) 17.4 18.7 18.7 16.7 DPPA4 Developmental pluripotency associated 4 (LOC680293) (predicted) 12.2 11.9 11.9 IO.I DPPA5 Similar to DPPA5developmental pluripotency associated 5 (RGD1564306( (predicted) 9.3 8.5 7.9 8.8 GAS2 Growth arrest-specific 2 (RGDl 562167) (predicted)

5.1 6.0 2.8 5.8 HHEX Hemtopoietically-expressed homeobox protein 118.6 128.0 131.9 107.4 IGFBPl Insulin like growth factor binding protein I

8.3 9.6 0.1 7.4 MDK Midkine (neurite growth-promoting factor 2) 10.7 5.3 14.9 16.9 ONECUTI One cut homeobox 1 7.0 14.9 2.1 43.3 RNF125 Ring Finger protein 125 (predicted) 9.4 10.3 10.6 5.3 SECI6b SECl6 homolog B (S. cerevisiae homolog) 19.1 19.7 19.7 19.0 STRA8 Stimulated by Reinoic Acid 8 (RGD1562852) (predicted)

12 3.5 9.5 1.0 21.0 TCFAP2B Transcription factor Ap-2 beta

Table. 2. List of candidate genes identified from list of activated genes.

Fold Activation Gene name Description M38 HS2 HI I M29 21.8 15.9 19.5 6.2 TIMPI TIMP metallopeptidase inhibitor I 4.9 11.0 35.6 6.3 BMP7 Bone morphogenetic protein 7 11.5 6.2 0.9 8.8 GPRl77 G protein-coupled receptor 177 1.0 15.6 0.9 27.1 PITXI Paired-like homeodomain I 4.6 11.0 0.8 9.0 IGFBP7 Insulin-like growth factor binding protein 7 10.4 0.6 I.I 11.5 WIFI Wnt inhibitory factor I 23.2 4.1 0.9 12.8 lGFBP2 Insulin-like growth factor binding protein 2

Hepatic Rhenotype

HNF network Wnt pathway Growth Hormone Pathway CREG ???

Non-Hepatic phenotype

Fig. 3. Candidate gene predicted to involve in regulation of liver phenotype.

13 Cellular repressor of EIA stimulated gene-1 (CREGI) is one such candidate gene identified through microarray analysis that falls under the repressed category. Our hypothesis was that CREG I plays a role in maintaining the hepatic phenotype through activation of liver specific transcription factors and may therefore partially revert a non-hepatic cell type to hepatic phenotype. In this study, we are presuming that CREG I plays a vital role in regulating transcription factors and signaling proteins that play a significant role in the tissue-specific gene activation in hepatoma variant cell. This will shed light upon the role of CREG I on liver specific gene regulation. To this end, CREG 1 will be introduced into Hl 1 bepatoma variant cells and transfectants screened for reactivation of liver gene expression. The specific goals of this work include I) Validation of tissue specific liver gene expression in hepatoma variant cell line HI I by reverse transcriptase quantitative Polymerase Chain Reaction (RT-qPCR), 2) Identification of liver-specific transcription factors whose expression is affected by CREG 1 3) Determination of the level of up-regulation or down-regulation of the downstream liver genes by CREG 1 and 4)

Comparison and validation of RT-qPCR data with microarray data. The overall goal of this project is to use whole genome functional data to understand the putative molecular role of

CREG I in liver function.

14 Hepatoma Variant Cells l Whal• G•n• • Miuoarray

Identification ofI Candidate Genes

Repressed I(e.c. CREG1)

TransfectI H11 cells

Generation and selectionI of G418 resistance clones

Mon1tor . tver Icene repression ,. qRT-PCR .

Fig.4. Flowchart showing the schematic representation of the strategy to be implemented to study the effects of CREG1 in hepatoma variants (Hl 1 ).

15 Chapter 2

Materials and Methods

2.1. Cell Culture

All cells were maintained in a medium containing 1:1 Ham's F12/Du1becco's modified

Eagle's medium (FDV) containing 10% fetal bovine serum (FBS) (GIBCO BRL) and 5 µg/100 ml penicillin-streptomycin (GIBCO) at 37° C in a humid 5% C02 water-jacketed incubator. All cells were derived from the parental cell line H4IIEC3, a rat tumor. The rat hepatoma variant cell line HI 1 was derived from Fg-14 cells by negative selection against Aprt (adenine phosphoribosyltransferase) and Gpt (xanthine - guanine phosphoribosyltransferase) transgene.

Hll cells were generated from Fg-14 cells by using 20 µg/ml of2, 6-diaminopurine (DAP) and

30 µg/ml 6-thioxanthine in the culture medium. Surviving clones were picked using cloning rings and each clone was expanded.

2.2. Cell Transfection

Semi-confluent cell plates were used for the transfection of candidate genes. FDV (0.5 ml) media without penicillin and streptomycin was added to a microfuge tube. DNA (1 µg/l µl) along with 5 µl Lipofectamine Plus reagent (Invitrogen, Inc) was added and mixed gently for five minutes at room temperature. Then, 5 µl of Lipofectamine L TX reagent was added and mixed gently by pipetting and incubated for 30 min at room temperature.

Media on cell plates was removed by pipetting. The transfection mixture prepared in the microfuge tubes were added in the wells of the plates and incubated at 37° C in 5% C02 incubator for 6 - 8 hrs, rocking periodically. After 6 - 8 hrs, medium was replaced with FDV

16 containing 10% fe tal bovine serum plus penicillin and streptomycin and incubated for 2 days at

37° C in 5% C02 incubator changing media after 18-24 hrs. For stable transfection, cells were split 1:10 into complete medium containing 500 µg/ml G418 and incubated for 2-3 weeks. G418 resistant clones were either picked individually or pooled and expanded so that RNA could be extracted.

2.3. RNA Isolation

RNA was extracted from confluent cell lines using a RNeasy Mini Kit (Qiagen, Cat

#74104) following the RNeasy Mini kit Protocol (Qiagen, Cat#79254), with a DNase-I step modification. Nutrient medium was removed from the culture dish, and cells were lysed by treating with highly denaturing guanidine isothiocyanate (GITC) containing RL T buffer and P­ mercaptoethanol. Cells were then scraped from the dishes and homogenized by applying the cell mixture to Qia-shredder columns placed in a collection tube followed by centrifugation. The flow-through was treated with 70% ethanol which was then applied to resin filter within the

RNeasy column in a collection tube. The resin filter was washed with R WI buffer followed by digestion with DNase I for 15 min at room temperature and centrifuged. The filter was again washed with RWI, and then washed with RPE buffer twice followed by quick centrifugation after each wash. After the purification process, RNA was eluted from resin filter by applying 40

µL RNase-free water into the column followed by centrifugation at 10,000 x g for 60 sec. All

RNA samples were collected in microfuge tubes and stored at -70° C. Concentration and purity of the RNA was determined by using a nano-drop spectrometry at 260 and 280 nm.

2.4. Reverse Transcription Polymerase Chain Reaction (RT-PCR)

17 Complementary DNA (cDNA) was synthesized from purified RNA by using a four-step

RT-PCR protocol and the MasterAmp High Fi delity RT-PCR kit (Applied Biosystem, part#

4368814, Lot#09 l l I 00). A final volume of 50 µl of reaction mixture contained 3 µl of RNA template (5 ng-10 ng), 19 µl of sterile nuclease free water, 25 µJ of Master Amp 2X RT-PCR

PreMix, 2 µl of 25 pmoles of oligo (dT), and I µl (50 U) of Moloney Murine Leukemia Virus

Reverse Transcriptase (MMLV-RT) Plus. The reaction mixture was incubated at 37° C for 30 min. The Bio-rad Thermal Cycler was used to synthesize cDNA. The firststep was carried at 25°

C for 10 min, followed by second step at 37° C for 2 hrs, then third step at 85° C for 5 min and finally, fourth step at 4° C for infinity. Newly synthesized cDNA was stored at -20° C, to be used for the quantitative polymerase chain reaction (qPCR).

2.5. Quantitative-Polymerase Chain Reaction (qPCR)

All the cDNA samples were diluted and concentration (5 ng/µl) was standardized among the samples. A final volume of 20 µl of reaction mixture contained 2 µI of cDNA template (5 ng/µl), 6.75 µl of sterile nuclease free water, 10 µl of Fast SYBR® Green Master Mix (Applied

Biosystem) and 1.25 µI of gene specific primer (0.5 µM) (IDTDNA). The applied Biosystem thermocycler was used to measure the expression of mRNA level. The first step was carried at

95° C for 3 sec, followed by extension/annealing step at 5° C above melting temperature (Tm) for 30 sec. A final melt curve step was performed at 95° C for 15 sec down to 60° C for 60 sec and a final step at 95° C for 15 sec. The control for RT-qPCR contains 8.75 µl of sterile nuclease free water, 10 µl of Fast SYBR® Green Master Mix (A pplied Biosystem) and 1.25 µl of gene specific primer (0.5 µM) (IDTDNA). The list of primers used for genes and their Tm is given in the Table 3. The cDNA quality was tested against the reference gene Gapdh. Duplicate assay was performedfor each cell line and reaction was repeated three times.

18 Raw threshold (Ct) value was averaged after the amplification for each of the cell line.

The amplified target genes in each hepatoma and hepatoma variant were normalized to the respective Gapdh Ct-value, generating a delta-Ct value (6Ct). Fold differences in gene expression were determined with a delta delta-Ct (66Ct) calculation. The 6Ct of the control cell line (Fg l4) was subtracted from the hepatoma variant 6Ct value, generating a 66Ct value. Using a log base 2 scale, the calculated difference was placed into the 66Ct equation: 2(6Ct (sample - experimental) - (6Ct Control).

2.6. Statistical Analysis

Statistical significance was determined by student t-test assuming unequal vanance.

Significance value was accepted at p < 0.05 level.

19 Table 3. Primers used in the quantitative Polymerase Chain Reaction (qPCR), primer sequences, melting temperature and annealing temperature of primers:.

Primer Primer Sequence Tm Annealing (°C) temperature (o C) HNF6 F-5' CAGTGGCTCTAAGCACAGT AA 3' 54.5 60 R-5' CAGTGTGGTGGAACAGATAAGA 3' 54.3 HNFl F-5' CCA TIC TGA AAG AGC TGG AGA AC3' 57.l 60 R-5' AGG TAC GAC TTG ACC A TC TTT GC3' 55.6 HNF4 F-5' TGG CAO AAC TCC ATC CGT CAT TC3' 56.6 60 R-5' AAC TGG ATC TGC TCG ATC ATC TG3' 56.5 SERPINAl F-5' CCTATACCGGGAGCTGGTCCAT 3' 60.3 64 R-5' TTGCGAGTGTCACCCTTGCT 3' 59.5 SERPINA3n F-5' ACT GGG AGA GGA CAC TCT ATT3' 54.5 60 R-5' GTA GAG GCT GAA GGC AAA GT3' 54.7 SERPINA3m F-5' CCC ACA TGG ACC AAG AAG AA3' 54.7 64 R-5' GTG CCC AGA GGA AGG AAA TAG3' 55.l SERPINA4 F-5' GAC CTG AGC CCA GAT GTA AAG3' 55.1 64 R-5' ACT GGT AGA GAC TCT GGA AGA A3' 54.4 SERPINA6 F-5' AAC CAG CAT GTC AAG GAT AAG A3' 54. l 60 R-5' GCC TCT GAG GAA GAT GTA GTT G3' 54.8 SERPINAlO F-5' GGA ACT TTG CCT CTA CCT TTG A3' 54.8 64 R-5' GGT CAC CCG ATT TCT CCA TAA G3' 55.1 SERPINAl l F-5' ACA AAG TCA CAC CCA CCA TTA3' 54.2 64 R-5' GAA GAG GAT GTT CCC TGG TAT TT3' 54.4 ALB F- 5' CATCCTGAACCGTCTGTGTG 3' 55.7 60 R-5' TTTCCACCAAAGACCCACTA 3' 55.8 rKNGl(l) F- 5' AACACAA TTGCCGCCTTCTCACAG 3' 60.2 64 R-5' GTGCAATGGAATGACCAAGTGCCT 3' 60.2 rFGBl (4) F-5' AAGGAGACAAGGTGAAGGCACACT 3' 60.3 64 R-5' AAGAACATGCCGTTGTGGATGCTC 3' 59.9 rPCKl (2) F- 5' CGCTATGCGGCCCTTCTTT 3' 58. 3 60 R-5' CGTGAAAGATCTTGGGCAACT 3' 57.5 rHHEXl (4) F-5' AGGTGCCTCTTTGGACAGTTCTCA 3' 60.l 64 R-5' AGCCTTTATCACCCTCGATGTCCA 3' 59.8 rGAPDH F-5' TGA TTCTACCCACGGCAAGTT 3' 56.5 60 R-5' TGA TGGGTTTCCCA TTGATGA 3' 54.2 rCREGl (3) F-5' CATCAGACACCCTGAGATGAAA 3' 54.4 60 R-5' GGTCCACCAAAGT AGTCCAAA 3' 54.8

20 Chapter 3

Results

To understand the loss of liver cell function, it is essential to observe the hierarchy of signaling pathways. Hepatoma variant cell lines were derived from parent hepatoma cell lines and selected according to their inability to activate an introduced SERPIN AJ-Aprt transgene in parental cell line Fg14. Hepatoma variants were generated at a frequency < io-6 from the Fg l4 parent (47). The hepatic phenotype of variant cells was strongly repressed in the hepatoma variant cell lines including loss of expression of the chromosomal Serpinal gene; also found was the repression of trans-acting factors and loss of transgene expression controlled by the

SERPINAJ promoter. To identify the defects in regulatory pathway, variant cells were found to lack expression of liver enriched transcription factors, hepatocyte nuclear factors Hnfl a and

Hnf4 (9) as well as many downstream liver-specific genes.

3.1. Part A. CREGl as a candidate for liver gene activation

3.1.1. Stable transfection of the CREGl gene in the Hll variant cell line

Based upon RatRef-12vl Expression Bead Chips (Illumina, Inc., San Diego, CA) whole genome microarrays, it was predicted that Cellular repressor of EIA-stimulated gene- I

(CREG I), a putative candidate gene, may have role in activating liver-specific genes and transcription factors. Microarray data showed that in each of four hepatoma variant cell lines,

CREG1 expression was repressed from 3 to 7 folds compared to hepatoma cells. Unexpectedly

RATl fibroblast cells also abundantly expressed this protein (Table 4). A similar comparison

21 was done for other known liver-enriched genes, including C/ebpa, HnD, Hnfla, Hnf4 and Hnf6.

Microarray results showed that each of the transcription factors, with the exception of HNF3, was substantially reduced in the hepatoma variant cell lines ((Table 5).

Based on the above results, we asked whether reintroduction of CREG 1 would partially restore the liver phenotype in the hepatoma variant cell line H 11. Previous data showed reactivation of approximately 25% of repressed hepatic upon forced over-expression of the

HNFla gene in the Hl 1 cells (47) .

To this end, Hl 1 cells were stably transfected with an exogenous CREGl gene, which was incorporated in cell genome, for studying CREG 1 's role in hepatoma cells. Plasmid vector pCMV6-Kan/Neo (Fig. 5), is introduced in Hl 1 cells. This vector expresses CREG 1 containing the selectable neo marker gene providing geneticin ( G418) resistance. In this plasmid, a strong enhancer/promoter from the cytomegalovirus (CMV)

(http://wwv. .origene.com/cdna/111ouseclones.111spx) controls CREGI gene expression. Variant cells were transfected, followed by splitting after 2 days under G4 l 8 selective pressure resuting in several G4 I 8-resistant clones . After 2-3 weeks, these clones were picked individually and or pooled (Table 6) and expanded for RNA extraction. Transfection efficiencies were observed by parallel transfections with a plasmid expressing the green fluorescence protein (GFP) reporter

(results not shown). After a few rounds of transfections, a total of 21 Hl 1-CREGl clones were picked individually and 4 pools of clones were generated (70 to 150 clones/pool). For a negative transfection control, FDV media without plasmid was used in the transfection procedure and resulted on no G418 resistant clones. A total of 12 of the individual clones were analyzed and 2 of 4 pooled samples were analyzed for over-expression of CREGI (Table 6). The introduced

22 CREGI transgene expression was monitored by qRT-PCR for reactivaion of hepatocyte nuclear factors and downstream liver genes (see below).

3.1.2. Overexpression effects of CREGl on transcription factor expression

In the genome, transcription factors act as an activators or repressors of genes. In human hepatocytes, four master transcription factors HNFla, HNF3P (FOXA2), HNF4a and HNF6

(ONECUTl) have been identified (26). These HNFs operate synergistically to regulate liver gene expression (26), which has been confirmed through previous studies of microarray analysis data (47). Also, they bind to 800-4000 gene promoters in chromatin immunopreciptation (ChIP) assays.

Expression levels of several of these HNFs was measured by qRT-PCR. Results from the pooled clones showed that forced expression of CREG 1 in the H 11 variant cells resulted in robust (100-fold) activation of both Hnfl a and Hn/4 to levels similar to those observed in the parental Fg14 cells (Fig. 7). Hnf3 levels were also monitored. These levels were found to be similar between the Fgl4 and HI 1 variant cells original. As expected, these levels were similar for the Hl 1-CREG pool as well (Fig. 8). Likewise, Hn/6 was also dramatically activated (30-

35 folds) in the pooled Hl 1-CREG cells and showed variable activation in the four clones tested

(Clones 7, 12, 17 and 20) ranging from 6-fold to 18-fold compared to the HI I parental cells.

(Fig. 9). Levels in some of the clones (clones 7 and 12) were similar to those of the Fg14 hepatoma cells.

3.1.3. CREGl overexpression effects on downstream genes

Regulation of downstream genes in distinct cell types depends largely upon transcriptional control. Previous studies have also shown that regulation and/or activation of liver

23 specific downstream genes requires binding of hepatocyte nuclear factors and co-activators to promoter regions. Because CREG 1 is a putative transcription factor (predicted based upon published data, 47), we asked if CREGI could activate liver specific downstream genes.

We first monitored expression of �hree key liver markers, Alb, Pckl and Hhexl. Results show that each of these marker genes is strongly activated (up to 20-fold) in the CREGl­ overexpressing H 11 cells (Fig. 10) to levels just below those of the parental Fg 14 hepatoma cell line (Fig. 9). We next tested two additional liver gene markers, Kngl and Fgbl, both of which were also strongly activated (Fig. 11).

3.1.4. CREGloverexpression effects on SerpinA genes Cluster

It was first asked that whether CREG I could activate expression of liver genes. The cellular repressor of EIA- stimulated gene-I (CREGl) plays an important role in mammalian cells. It is a secreted protein that could be a regulator of many aspects of early development ( 16).

Quantitation of RNA levels using qRT-PCR amplification of cDNA generated from RNA.

Expression of the SerpinAl gene was tested in the HI 1- CREGl cells. Serpinal was strongly activated (150 folds) compared to the Hl 1 cells (Fig. 12). These results were compared to the differences in expression found between FG 14 and H 11 cells using whole genome microarrays.

Microarray analysis showed SerpinAl expression to be expressed 55- fold higher in the FG14 cells compared to the Hl 1 cells (Table 5).

Next we asked whether other members of the rat Serpin locus were rescued by the introduction of CREG 1 into the H 11 cells. We first examined microarray data which showed that expression of several serpin locus genes repressed in the HI 1 cells compared to the FG14 parental cells. Serpina3n, 4 and 6 showed the most obvious expression differences, with the other

24 serpin genes showing near background levels (-100 units) of expression in the FG 14 cells.

However, it was reported previously that the rnicroarray data lacks sensitivity to monitor gene expression compared to qPCR (47).

We monitored gene expression for several serpin locus genes using qPCR. Nearly all (6 of 7) of the serpin locus genes tested showed a dramatic reduction of expression in the hl 1 cells compared to the FG 14 parental cells (Fig.13). These results suggest that the microarray was unable to detect diffe rential expression for many of the lower- expressed serpin genes.

Importantly, the Hl 1- CREGl cells showed strong rescue of nearly all serpin genes tested. In most ofthe cases, H 1 1-CREG 1 cells expressed levels or serpin genes higher than the FG 14 cells.

25 Table 4. Comparison of expression levels of CREG1 hepatoma variant cells lines compared to parental Fgl4 hepatom cells as determined by by microarray analysis. Raw fluorescent values are shown for FG 14 and RATI cells. Ratios of gene expression of the variant cells compared to the parental Fg 14 cells are shown for the four variant cells lines M38, HS2, HI 1 and M29.

Gene RATl Fg14 RAT1/Fg14 Fg14/M38 Fg14/HS2 Fg14/H ll Fg14/M29

CREG1 4885 2791 2 3 7. 6 6

Table 5. Expression levels of transcription factors in hepatoma and hepatoma

variant cell lines as determined by microarray data.

Transcription Average values Fold Exl!ression

factors Fg14 Hll M38 Hll HS2 M29

Cebpa 250.0 100.7 2.1 2.5 1.5 2.1 Hnf3 (FoxA2) 3417.9 5632.0 0.6 0.6 1.1 1.1 Hnfla 455.2 107.7 3.4 4.2 4.5 4.2 Hnf4 247.0 88.6 2.9 2.8 2.9 2.7 Hnf6 (Onecutl) 201 1.9 134.6 10.7 14.9 5.3 16.9

26 Sgfl Asel pCMV6-Kan/Neo Kpnl 4906bp Rsrll EcoRI Not l Xbal Xhol HSVTK CotE� PolyA ,,, Hind Ill

Fig. 5. True Mouse Clone Vector pCMV6-Kan/Neo. The CMV promoter and a Kozak consensus sequence drive protein expression in mammalian cells. An antibiotic selection cassette (Kanr/ Neo� confers resistance to kanamycin in E. coli and neomycin analogs (G418) in mammalian cells. The T7 promoter upstream of the ORF allows protein expression in cell-free system. Each clone contains a fully verified insert sequence (h ttp: /v.,,n\.o rigene.com/cdna/mouseclones.mspx).

27 Table 6. Hll-CREGI transfectants ge nerated Sampl e # of Clones/pools Analyz ed Over-Expr essed Control 0 NIA NIA Transfected Clones 17 4 2 Transfected Pools 3 I 1

28 mCREG 1s expressed in transfected Hll cells

500

450 r

400

350

300

200

150

1 100 I I I

T so T iI l I 0 Hll FG14 Hll-CREG Pool Hll-CREG HllCREG HllCREG HllCREG Clone 7 Clone 12 Clone 17 Clone 20 [{:ell linesI

Fig. 6. Expression of transfected mouse CREG l in transfected hepatoma variant Hl 1 cells. CREG I is overexpressed in individual clones (H 1 1-CREG 1-Clone7, H 11-CREG 1-Clone20) and pooled clones (HI 1-CREGI-Pool) as detected by qR-TPCR. Fold differences compared to the parental HI I cells is shown. All samples were measured in triplicate by qRT-PCR reactions Values are normalized for RNA quality using Gapdh expression values for each specific cell lines. Note that the mouse primers do detect expression of the rat CREG gene (data not shown)

29 CREGl effects on Hnfl & Hnf4 expression

350

300

250

Qi ::I ;;; > ti 200 s Cll u c � � 150 0 "O 0 ...

100 so 11 0 Hll FG14 H 11-CREG Pool

Cell lines

•HNFl • HNF4

Fig. 7. CREGI activates Hnfl and Hnf4 in the HI 1 cells. CREGl overexpression resulted in activation of liver- specific HNF transcription factors. Fold differences compared to the parental Hll cells is shown. All samples were measured in triplicate by qRT-PCR reactions Values are normalized for RNA quality using Gapdh expression values for each specific cell lines.

30 CREGl effects on Hnf3 expression

1.4

1.2 T I

QJ 1 :J iii > ... u

0.2

0 Hll FG14 H 11-CREG Pool Cell lines

Fig. 8. CREGI overexpression effects in transfected HI I cells. Fold differences compared to the parental Hl 1 cells is shown. All samples were measured in triplicate by qRT-PCR reactions Values are normalized for RNA quality using Gapdh expression values for each specific cell lines.

31 CREG1 effects on Hnf6 expression

35

30 T

QI 25 BIV > T t;

10 T

T 5 T • 0 J. Hll FG14 Hll-CREG Pool Hll-CREG HllCREG HllCREG HllCREG Clone 7 Clone 12 Clone 17 Clone 20 Cell line

Fig. 9. CREG1 activates Hnf6 in the Hl 1 cells. Pooled clones and individual clones were monitored for Hnf6 expression. Fold differences compared to the parental Hl I cells is shown. All samples were measured in triplicate by qRT-PCR reactions Values are normalized for RNA quality using Gapdh expression values for each specific cell lines.

32 CREGl effects on downstream gene expression

a; :I ii > lj so � 45 GI 40 u c GI 35 ... � 30 0 25 "O 0 20 T .... 15 10 5 I •• I 0 RHHEXl ALB PCKl

Downstream genes

• Hll • FG14 • Hll-CREG Pool

Fig. 10. CREG l activates liver downstream hepatic genes in the Hll cells. Downstream genes, required for cellular maintenance, proliferation and differentiation, were tested in the Hl 1- CREG l cells. Liver-specific genes Hhex l, Pckl, and Alb were activated by CREG. Fold differences compared to the parental Hl 1 cells is shown. All samples were measured in triplicate by qRT-PCR reactions Values are normalized for RNA quality using Gapdh expression values for each specificcell lines.

33 CREG1 effects on downstream gene expression 1200 1000 I

� 600 c � ;; -0 400 -0 :2 200

0 KNGl FGBl Cell Lines

• Hll • FG14 • Hll-CREG Pool

Fig. 11. Downstream genes required for cellular maintenance, proliferation and differentiation were tested in the Hl 1-CREGJ cells. Liver-specific genes, including Kngl and Fgbl, were strongly activated. All values are normalized for RNA quality using Gapdh expression values for each specific cell lines. All samples were measured in triplicate by qRT-PCR reactions.

34 CREGl rescues Serpinal expression

200

180

160

140

T• 120 Qj v c � Qj :;: 100 0 "O 0 .... T 80 I

60 TI i

40

20 !

0 .l Hll FG14 Hll-CREG Pool Hll-CREG HllCREG HllCREG HllCREG Clone 7 Clone 12 Clone 17 Clone 20

Cell lines

Fig. 12. CREGI activates SerpinAl expression in the Hl 1 cells Downstream gene Serpinal (a liver phenotypic biomarker) was strongly activated. All samples were measured in triplicate by qRT-PCR reactions Values are normalized for RNA quality using Gapdh expression values for each specific cell lines.

35 CREGl rescues serpin cluster expression

• SerpinA3m •SerpinA3n 30 8 T c: 25 T f 20 T £ 15 i5 10 T "O 5 1. I. 0� 0 2:..l. •• . .. Hll IIFG14 Hll-CREG Hll-CREG HllCREG HllCREG HllCREGii Pool Clone 7 Clone 12 Clone 17 Clone 20 Cell lines

Serpina4 and Serpina6 • SerpinA4 • SerpinA6 250 8 200 c :!! 150 £ 0 100 "O ;f so •• 0 T T •• � I Hll FG14 Hll-CREG Pool Hll-CREG HllCREG CloneHllCREG CloneHllCREG Clone �� 7 u v �

Cell lines

Fig. 13. CREG 1 over-expression activates serpin cluster genes in transfected Hl 1 cells. Fold activation for Serpin3A and Serpin3 (Top) and SerpinA4 and Serpin A6 (Bottom) are shown. All samples were measured in triplicate by qRT-PCR reactions Values are normalized for RNA quality using Gapdh expression values for each specific cell lines.

36 Chapter 4

Discussion

Regulation of gene expression in mammals involves both activation and repression of tissue-specific genes which dictate tissue function and phenotype. Hepatoma variants are derived from rat liver tumor cells, but lack liver gene expression. These variant cells lines were generated using a selection scheme. In these variants,hundreds of liver specific genes have been stably silenced, including those encoding tissue specific transcription factors (47). In this study, through whole genome microarray analysis, cell specific transcription factors and genes responsible for cell signaling, activated or repressed in the system, were identified. One

candidate gene - CREG1 (repressed) was identified as candidate genes in at-least two out of the four hepatoma variant cells (M29, HS2, Hl 1, M38) tested via microarray analysis comparing expression levels to those of the parental Fg 14 hepatoma cells ( 42).

4. CREG1 As A Candidate for Liver Gene Activation

Though the underlying pathways disrupted in the variant cell lines which result in loss of liver gene expression have yet to be elucidated, these cell lines have been used to identify the role of liver-specific transcription factors in driving phenotype. In this study, the dedifferentiated hepatoma variant HI I cells were used. These have been shown to be defective in expression of

Hnfl a and Hnf4, along with other downstream genes (Bulla et al, 2010). Stable introduction of a

CREG1 expression plasmid into the Hl1 cells was found to reactivate many liver-specific transcription factors and downstream genes.

37 Cellular repression of EIA stimulated genes-I (CREG 1 ), a secretory glycoprotein, was first discovered while screening a cDNA library of a model organism -Drosophila melanogaster

(fruit fly). The screening was for proteins that interact with transcription factors (43). It has been expressed ubiquitously in differentiated tissues; not being restricted to a single type of cell (43).

In humans, CREGI is located on chromosome lq24.2 (shorter arm of chromosome!). The

CREG 1 gene is 12.8 kb in size and composed of four exons and introns ( 44). The transcriptional product of CREG 1 is 220 amino acids ( 16), which is a conserved protein, shares a similar sequence with the EIA protein of adenovirus known to regulate the repression and activation of genes via E2F. E2F is the gene that codes fa milies of transcription factors (16). CREG I, resides at the perinuclear region i.e. cytoplasmic area that surrounds the nucleus. An extensive study revealed that it regulates transcription via RNA polymerase II promoter i.e. it is a transcription factor binding complex (16). Furthermore, this complex can antagonize transcriptional activation and adenovirus mediated cellular transformation, which is dependent on interaction of insulin like growth factor receptor 2 (IGF2R) (48). Through gene card analysis, it is known that CREG 1 has the ability to bind to Transcription factor TBP and tumor suppressor, Rb.

The ability of CREG 1 to bind the general transcriptional factor TBP and tumor suppressor, Rb is also suggested by Gene Card analysis (http: //www.genecards.org/cgi ­ bin/carddisp.pl?ge ne=CREG I &search=4 1 d1 78bt28c87c08d7ecfab386a9cf9c#pa thways source).

The gene ontology database (http: //am igo .geneontology.org/ amigo/gene pr oduct/RGD: 1306804) has suggested CREG 1 to be a member of a transcription factor binding - transcription factor complex that has the ability to regulate gene transcription.

CREG 1 has not previously been reported to play a role in liver development or maintenace of liver-specific gene expression, and it is unclear how liver function is activated by

38 CREG I in the current study. Previous studies revealed that hepatocyte nuclear factors (HNFs) function synergistically by forming a complex of inter-dependent signaling meshwork (27, 42).

Also, transcriptional regulation in liver is known to be controlled by DNA binding proteins and co-activators (26).

This study found that HNF6 was found to be dramatically activated (>25 fold) in both single clone and pooled HI I clones overexpressing CREG 1 from an introduced expression plasmid. HNF6 is considered a major transcription factor in mammalian development ( 40). In a mouse model, if HNF6 is not present during development, there occur abnormalities in intrahepatic and extrahepatic bile ducts, defects in gallbladder formation. This observation suggests that HNF6 is essential for development of biliary and gall bladder (27, 28).HNF6 has a single domain that binds to DNA and has sequence homology between ONECUT (HNF6) families and functional redundancy, can regulate the nuclear localization and transcription activation and homeodomain motif (21). Studies in zebrafish suggests that HnflP regulates Hnf6

(25). Though in current study, influence of CREG 1 overexpression on Hnfl p expression was not tested.

HNF6 regulates transcriptional activity via binding the proximal region sequence of the

RNA polymerase II core promoter, as it has DNA binding motif, which is revealed by UniProt data (http://www.unipr ot.org/ unipr ot/09UBCO) analysis. Furthermore, HNF6 binding sites can be found in the promoter or enhancer regions in many of the liver specific downstream genes

(28). Also, an HNF6-C/EBP transcription complex is involved in this independent transcription process via synergy, without involvement of any other hepatocyte transcription factors (30). To express the HNF6 targeted genes, it is required to recruit CBP co-activator protein. TBP (7),

C/EBP (11) and TFIID (I 0) are transcriptional components which work as cofactor or co

39 activator of transcriptional control regulatory machinery. From gene card analysis it is understood that CREG 1 regulates transcription through these transcription components.

Interestingly, it is unclear as there is no direct link between HNFs and CREG 1. One possibility is that CREG 1 has a cofactorbinding region- FMN (Flavin mono- nucleotide), which is responsible for interaction with other proteins (perhaps HNF6) through protruding extended loop (29, 46). Secondly, it has a secretory property, which can either easily enter the membrane or accumulate on extracellular space and cell surface. If this assumption is true, CREG l could regulate morphogenesis, cellular differentiation and transcriptional modulation (20). This could result in stabilization of cellular identity and regulate a myriad of biological functions through receptor interaction like IGF2R, protein- protein adhesion and cell communication (35).

Serpinal (45), Hex], Alb, Wnt4 (40) are liver specific downstream genes, all of which were acrtivated the H 11-CREG I cells, are expressed during early stages of liver development.

However, other liver enriched transcription factors have been identified that regulate these.

Traditionally the liver specific gene is SERPINAI has been used as as phenotypic marker for liver function. SERPINAI gene mutations result in emphysema or liver disease (45). HNFl and

HNF4 have been shown to bind to promoter region for expressing SERPINAl (2). The current studies suggest that CREG 1 strongly activates SERPINA 1 gene, though there was no activation of Hn/4 or Hnfl genes (which are strongly down- regulated in the hepatoma variant cells) (47).

This leads to assumption that there might be a common co-activator/co-factor binding site in

SERPINAI promoter. Two types of interactions are predicted: 1) An interaction between HNF6 and SERPINAl mediated via an alternative protein-RNA interaction domain in SERPINAl and

2) an indirect effect mediated via protein-protein interaction between SERPINAl and an unknown RNA binding protein. However, there is no evidence that protein - protein interaction

40 may affect transcriptional regulation of Serpinal gene. One study suggests that expression of

Serpinal gene cluster of five genes can be controlled by locus control region (LCR) (citation).

However, recruitment of liver specific transcription factor- HNF6a, HNF3p, C/EBP-a. and -P activates as well as RNA polymerase II responsible for activating gene through serpin LCR (34).

Therefore, it is possible that over-expression of SERPINA I is responsible for synergistic association of HNFs with CREGI. It might be that expression of SERPINAl increases when

CREG I binds or CREG 1 regulates the expression of Serpina I gene in an independent fa shion.

Hematopoietically expressed homeobox (HHex) is homeobox transcription factor which ts regulated by Wnt4 signaling in conjunction with SOX13 by dissociating the P-catenin/TCFl complex (38, 42). Also, lack of HNF4a. interaction with LEFI (P-catenin regulated protein) leads to repression of Wnt4 expression ( 40). In the current study, Hhex expression was affected in the

Hl 1 cells over-expressing CREGI, suggesting that the observed expression might have some impact over the pathway.

41 Future Directions

We have identified the cellular repressor of E 1 A-stimulated gene (CREG 1) through whole genome analysis of hepatoma and hepatoma variant cells as a candidate master regulator.

We asked whether the ability of CREGI for rescuing expression of both liver maker gene

SerpinAl and several liver- specific transcription factors (HNF6, HNF4 and HNFl) suggested its role as a master regulator of liver function. The fact that CREG 1 also rescued most genes of the

Serpin locus suggests that the LCR was activated. The overexpression of Serpin genes by

CREG 1 could be the result of CREG 1 binding to Serpin promoters. This could be determined through the use of software that predicts transcription factor binding sites (TFBS) paired with functional assays like ChlP to detect the binding site sequences of CREG 1 on promoters.

Likewise, genome- wide binding sites for HNF6, which has not been reported in vitro, could be predicted using prediction software. Since, it is unclear whether HNF6 is required for this activation, gene knockdown experiments using siRNA could be used to reduce HNF6 expression in the H 11- CREG 1 cells and determine if SerpinA 1 expression is affected. Also, knocking down of HNFl and HNF4 expression may clarify whether CREG 1 can activate Serpin locus expression independent or other liver transcription factors.

Finally, it may be necessary to determine how CREG 1 might function in development and maintenance of liver tissue. The use of other cell culture model systems as well monitoring

CREG 1 expression during embryogenesis in normal and CREG 1 knockout mice model could help clarify the role of the protein in liver function. It would also be useful to conduct whole­ genome microarray analysis on the Hl 1- CREGI cells compared to FG14 cell and Hll cells to identify the extent of genomic remodeling that has occurred due to over expression of CREG 1.

42 References

1. Schultze, H. E., Goilner, I., Heide, K., Schoenenberger, M., Schwick, G. (1955). Zur

kenntnis der alpha-globulin des menschichen nirmal serums. Naturforsch 10: 463

2. Li, Y., Shen, R. F., Tsai., S. Y., and Woo, S. L. C. (1988). Multiple hepatic trans-acting

factors are required for in vitro transcription of the human alpha-1 -antitrypsin gene.

Molecular And Cellular Biology 8 (I 0): 4362-4369

3. Costa, R. H., Grayson, D. R., and Darnell Jr, J. E. (1989). Multiple hepatocyte-enriched

nuclear factors function in the regulation of transthyretin and a.1-antitrypsin genes.

Mo lecular and Cellular Biology 9 (4): 1415 - 1425

4. Gourdeau, H., and Fournier, R. E. K. (1990). Genetic analysis of mammalian cell

differentiation. Annual Review of Cell Biology 6: 69 - 94

5. Defrances, M. C., Wolf, H. K., Michalopoulos, G. K. and Zarnegar, R. (1992). The

presence of hepatocyte growth factor in the developing rat. Development 116: 387-395

6. Kuo, C. J., Conley, P. B., Chen, L., Sladek, F. M., Darnell, J. E., and Crabtree Jr, G. R.

(1992). A transcriptional hierarchy involved in mammalian cell-type specification.

Nature 355 : 457 - 461

7. Geisberg, J. V., W. S. Lee, A. J. Berk, and R. P. Ricciardi. (1994). The zinc finger region

of the adenovirus EIA transactivating domain complexes with the TATA box binding

protein. Proceeding of National Academy of Science 91 : 2488 - 2492.

8. Aso, T., Serizawa, H., Conaway, R. C., and Conaway, J. W. (1994). A TATA sequence­

dependent transcriptional repressor activity associated with mammalian transcription

factorIIA. The European Molecular Biology Organization Journal 13 (2) : 435 - 445

43 9. Bulla, G. A., and Fournier, R. E. K. (1994). Genetic analysis of a transcriptional

activation pathway by using hepatoma cell variants. Molecular And Cellular Biology 14

(11): 7086-7094

IO. Song, C. Z., P. M. Loewenstein, K. Toth, and M. Green. (1995). Transcription factor

TFIID is a direct functional target of the adenovirus e 1 a transcription-repression domain.

Proceedings of the National Academy of Sciences 92 : 10330 - 10333

11. Lundblad, J. R., R. P. Kwok, M. E. Laurance, M. L. Harter, and R. H. Goodman. (1995).

Adenoviral ElA-associated protein p300 as a functional homologue of the transcriptional

co-activator CBP. Nature 374 :85 - 88

12. Kamiya, A., Kinoshita, T., Ito, Y., Matsui, T., Morikawa, Y., Senba, E., Nakashima, K.,

Taga, T., Yoshida, K., Kishimoto, T., and Miyajima, A. (1999). Fetal liver development

requires a paracrine action of oncostatin M through the gp 130 signal transducer. The

European Mo lecular Biology Organization Journal I 8 (8) : 2127 - 2136

13. Bulla, G. A. (1999). Extinction of al-antitrypsin expression in cell hybrids is independent

of HNFla and HNF4 and involves both promoter and internal DNA sequences. Nucleic

Acid Research 27 (4) : 1190-1 197

14. Rollini, Pierre., and Fournier, R. E. K. (1999). The HNF-4/HNF-lo.transactivation

cascade regulates gene activity and chromatin structure of human serine protease

inhibitor gene cluster at l 4q32. l. Proceedings of National Academy of Sciences 96:

10308 - 10313.

15. Hedge, P., Qi, R., Abernathy, K., Gay, C., Dharap, S., Gaspard, R., Hughes, J. E.,

Snesrud, E., Lee, N., and Quackenbush, J. (2000). A concise guide to cDNA microarray

analysis. BioTechniques 29 : 548 - 562

44 16. Veal, E., Groisman, R., Eisenstein, M., and Gill, G. (2000). The secreted glycoprotein

CREG enhances differentiation of NTERA-2 human embryonal carcinoma cells.

Oncogene 19: 2120-2128

17. Bulla, G. A., Givens, E., Brown, S., Oladiran, B., and Kraus, D. (2000). A common

regulatory locus affects both HNF4/HNF1 o. pathway activation and sensitivity to LPS­

mediated apoptosis in rat hepatoma cells. Journal of Cell Science 114 : 1205 -1212

18. Dorris, D. R., Ramakrishnan, R., Trakas, D., Dudzik, F., Belva!, R., Zhao, C., Nguyen,

A., Domanus, M., and Mazumder, A. (2002). Genome Research 12 : 976 - 984

19. Hu, C., and Perlmutter, D. H. (2002). Cell-specific involvement of HNF-1� in o.1-

antitrypsin gene expression in human respiratory epithelial cells. American Journal of

Physiology Lung Cell Molecular Physiology 282 : 757-765

20. Clark, H. F., Gurney, A. L., Abaya E., Baker, K., Baldwin, D., Brush, J., Chen, J., Chow,

B., Chui, C., Crowley, C., Currell, B., Deuel, B., Dowd, P., eaton, d., Foster, J., Grimaldi,

C., Gu, Q., Hass, P. E., Heldens, S., Huang, A., Kim, H. S., Klimowski, L., Mark, M.,

Robbie, E., Sanchez, C., Schoenfeld, J., Seshagiri, S., Simmons, L., Singh, J., Smith, V.,

Stinson, J., Vagts, A., Vandlen, R., Watanabe, C., Wieand, D., Woods, K., Xie, M. H.,

Yansura, D., Yi, S., Yu, G., Yuan, J., Zhang, M., Zhang, Z., Goddard, A., Wood, W. I.,

and Godowski,P. (2003). The secreted protein discovery initiative (SPDI), a large-scale

effort to identify novel human secreted and transmembrane proteins: a bioinformatics

assessment. Genome Research 13: 2265 - 2270

21. Costa, R. H., Kalinichenko, V. V., Holterman, A-X. L., and Wang, X. (2003).

Transcription factors in liver development, differentiation and regeneration. Hepatology

38 (6): 1331 - 1347

45 22. Ober, E. A., Field, H. A., and Stainier, D. Y. R. (2003). Form endoderm formation to

liver an dpancreas development in zebrafish. Mechanism of Development 120 : 5 - 18

23. Parfrey, H., Mahadeva, R., and Lomas, D. A. (2003). al-Antitrypsin deficiency, liver

disease and emphysema. The International Journal of Biochemistry & Cell Biology 35 :

1009-1014

24. Bulla, G. A., and Kruas, D. M. (2004). Dissociation ofthe hepatic phenotype from HNF4

and HNFla expression. Bioscience Reports 24 (6): 595 - 608

25. Matthews, R. P., Lorent, K., Russo, P., and Pack, M. (2004). The zebrafish onecut gene

hnf-6 functions in an evolutionary conserved genetic pathway that regulates vertebrate

biliary development. Developmental Biology 274: 245-259

26. Odom, D. T., Zizlsperger, N., Gordon, D. B., Bell, G. W., Rinaldi, N. J., Murray, H. L.,

Volkert, T. L., Schreiber, J., Rolfe, P. A., Giffo rd, D. K., Fraenkel, E., Bell, G. I., and

Yound, R. A. (2004). Control of pancreas and liver gene expression by HNF transcription

factors. Science 303: 1378 - 1381

27. Rausa III, F. M., Hughes, D. E., and Costa, R. H. (2004). Stability of the hepatocyte

nuclear factor 6 transcription factor requires acetylation by the CREB-binding protein co­

activator. The Journal of Biological Chemistry 279 (41):43070-43076

28. Rubins, N. E., Friedman, J. R., Le, P. P., Zhang, L. Brestelli J., and Kaestner, K. H.

(2005). Transcriptional network in the liver: Hepatocyte nuclear factor 6 fu nction is

largely independent of Foxa2. Molecular And Cellular Biology 25 (16): 7069-7077

29. Sacher, M., Bacco, A. D., Linin, V. V., Ye, Z., Wagner, J., Gill, G., and Cygler, M.

(2005). The crystal strycture of CREG, a secreted glycoprotein involved in cellular

46 growth and differentiation. Proceedings of National Academy of Sciences 102 (5 1):

18326 - 18331.

30. Yoshida, Y., Hughes, D. E., Rausa III, F. M., Kim, 1-M., Tan, Y., Darlington,G. J., and

Costa, R. H. (2006). CIEBPa and HNF6 protein complex formation stimulates HNF­

dependent transcription by CBP coactivator recruitment in Hepg2 cells. Hepatology 43

(2) : 276 - 286

31. Cheng, W., Guo, L., Zhang, Z., Soo, H. M., Wen, C., Wu, W., and Peng, J. (2006). HNF

factors form a network to regulate liver-enriched genes in zebrafish. Developmental

Biology 294: 482 - 496

32. Teitell, M. A., and Mikkola, H. K. A. (2006). Transcriptional activators, repressors and

epigenetic modifiers controlling hematopoietic stem cell development. Pediatric

Research 59 (4): 33 - 39

33. Thalamuthu, A., Mukhopadhyay, I., Zheng, X., and Tseng, G. C. (2006). Evaluation and

comparison of gene clustering methods in microarray analysis. Bioinformatics 22 : 2405

-2412

34. Zhao, H., Friedman, R. D., and Fournier, R. E. K. (2007). The locus control region

activates serpin gene expression throught recruitment of liver specific transcription

factors and RNA polymerase II. Molecular and Cellular Biology 27 (15) : 5286 - 5295

35. Chen, R., Jiang, X., Sun, D., Han, G., Wang, F., Ye, M., Wang, L., and zou, H. (2008).

Glycoproteeomics analysis of human liver tissue by combination of multiple enzyme

digestion and hydrazide chemistry. Jo urnal of Proteome Research 8 : 651 - 661

36. Tao T., and Peng, J. (2009). Liver development in zebrafish (Danio rerio). Jo urnal of

Genetics and Genomics 36 : 325 - 334

47 37. Lemeigre, F. P. (2009). Mechanism of liver development: Concept for understanding

liver disorders and design of novel therapies. Gastroenterology 137: 62-79

38. Marfil, V., Moya, M., Pierreux, C. E., Castell, J. V., Lemaigre, F. P. Real, F. X., and

Borr, R. (2009). Interaction between HHEX and SOX13 modulates WNT/TCF activity.

The Journal of Biological Chemistry 285 (4) : 5726-5737

39. Li, T., Wan, B., Huang, J., and Zhang, A. (2010). Comparision of gene expression in

hepatocellular carcinoma, liver development, and liver regeneration. Molecular Genetic

and Genomics 283 : 485 - 492

40. Si-Tayeb, K., Lemaigre, F. P., and Duncan, S. A. (2010). Organogenesis and

development of the liver. Developmental Cell 18 : 175-181

41. Reiterer, V., Nyfeler, B., and Hauri, H-P. (2010). Role of the lectin VIP36 in post-ER

quality control of human al-antitrypsin. Traffic 11 : 1044 - 1055

42. Bulla, G. A., Luong, Q., Shrestha, S., Reeb, S., and Hickman, S. (2010). Genome-wide

analysis of hepatic gene silencing in mammalian cell hybrids. Genomics 96 : 323-332

43. Han, Y., Luan, B., Sun, M., Guo, L., Guo, P., Tao, J., Deng, J., Wu, G., Liu, S., Yan, C.,

and Li, S. (2011 ). Glycosylation-independent binding to extracellular domains 11-13 of

mannose-6-phosphate/insulin-like growth factor-2 receptor mediates the effects of

soluble CREG on the phenotypic modulation of vascular smooth muscle cells. Jo urnal of

Molecular and Cellular Cardiology 50: 723-730.

44. Wang, N., Han, Y., Tao, J., Huang, M., You, Y., and Zhang, H. (2011). Over-expression

of CREG attenuates atherosclerotic endothelium apoptosis via VEGF/PI3K/AKT

pathway. Atherosclerosis 218: 543-551

48 45. Janciauskiene, S. M., Bals, R., Koczulla, R., Vogelmeier, C., Kohnlein, T., and Welte, T.

(2011 ). The discovery of a.1-antitrypsin and its role in health and disease. Respiratory

Medicine 105 : 1129-1139

46. Yang, G., Han, Y., Tian, X., Tao, J., Sun, M., Kang, J., and Yan, C. (201 1). Pattern of

expression of the CREG gene and CREG protein in the mouse embryo. Molecular

Biology Replication 38: 2133 -2140

47. Bulla, G. A., Aylmer, C. M., Dust, A. L., Kurkewich, J. L., Mire, L. K., and Estanda, A.

B. (2012). Genome-wide analysis of hepatic gene silencing in hepatoma cell variants.

Genomics 100: 176-183

48. Probst, Olivia C., Karayel, Evren., Schida, Nicole., Nimmerfall, Elisabeth., Hehenberger,

Elisabeth., Puxbaum, Verena., and Mach, Lukas. (2013 ). The mannose 6-phosphate­

bindingsites of M6P/IGF2R determines its capacity to suppress matrix invasion by

squamous cell carcinoma cells. Jo urnal of Biochemistry 451: 91 -99

49