<%BANNER%>

Characterization of the Transcription Factor Network Involved in the Amino Acid Response and Regulation of the Asparagin...

Permanent Link: http://ufdc.ufl.edu/UFE0023995/00001

Material Information

Title: Characterization of the Transcription Factor Network Involved in the Amino Acid Response and Regulation of the Asparagine Synthetase (ASNS) Gene by Nutrient Availability
Physical Description: 1 online resource (194 p.)
Language: english
Creator: Su, Nan
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: aar, aare, all, asns, asparaginase, atf4, chop, foxa, leukemia, transcription, upr
Biochemistry and Molecular Biology (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: It has been recognized for more than a decade that the availability of nutrients can alter the expression of a number of genes, including asparagine synthetase (ASNS), through distinct signaling pathways. Previous studies showed that ASNS is induced by amino acid or glucose limitation through the amino acid response (AAR) pathway or the unfolded protein response (UPR) pathway, respectively. Genomic analysis identified two cis-elements that are responsible for the induction of ASNS by both AAR and UPR pathways, the nutrient sensing response element (NSRE)-1 and ?2. Evidence from both in vitro and in vivo studies suggested that the activating transcription factor 4 (ATF4) functions as the major activator of the ASNS gene, whereas some members from the ATF and C/EBP families function as repressors. The purpose of the current study was to gain more insight into the mechanism through which the ASNS gene is regulated by nutrient limitation. Yeast two-hybrid screening, in conjunction with co-immunoprecipitation assay, identified the C/EBP homology protein (CHOP) as an ATF4 interacting partner. Reporter analysis, over-expression, and siRNA assay demonstrated that CHOP inhibits ATF4 dependent transcriptional activation of the ASNS gene by the AAR or UPR pathways both in vitro and in vivo. Chromatin immunoprecipitation (ChIP) assay suggested that CHOP is recruited to the ASNS promoter, as opposed to the prevailing hypothesis that CHOP functions by sequestering ATF4 from the DNA. Previous studies raised a debate about the correlation between ASNS expression and the resistance to asparaginase (ASNase) therapy in patients with acute lymphoblastic leukemia (ALL). The current study utilized multiple ALL cell lines from different origins, and confirmed the lack of correlation between ASNS mRNA level and ASNase sensitivity, as suggested by clinical microarray studies. However, expression profiling revealed a delayed ASNS protein expression in an ASNase sensitive cell line even though the ASNS mRNA is highly induced by ASNase treatment. The ASNS protein level was found to correlate with ASNase sensitivity in each of the ALL cell lines studied. Furthermore, although the ASNS mRNA level varies over a wide range in ALL patients, expression profiling of the ASNS gene using samples from ALL patients showed extremely low ASNS protein expression in most patients, which is consistent with the overall high efficacy of ASNase therapy. Taken together, these results demonstrate that ASNS is regulated by complex mechanisms involving fast response and feed back inhibition, at both the transcriptional and translational level. The current studies also demonstrate that the forkhead box A (FOXA) family of genes are subjects to the regulation by nutrient stresses, such as amino acid deprivation and ER stress. The induction of FOXA2 and FOXA3 by amino acid deprivation is not absolutely dependent on the ATF4-driven AAR pathway. Conversely, the FOXA gene products are not required for the induction of most of the other amino acid responsive genes. However, FOXA2 and FOXA3 seem to be involved in the regulation of the sodium-coupled neutral amino acid transporter 2 (SNAT2) gene, indicating the existence of a distinct pathway during amino acid deprivation other than the ATF4-dependent AAR pathway.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Nan Su.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Kilberg, Michael S.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0023995:00001

Permanent Link: http://ufdc.ufl.edu/UFE0023995/00001

Material Information

Title: Characterization of the Transcription Factor Network Involved in the Amino Acid Response and Regulation of the Asparagine Synthetase (ASNS) Gene by Nutrient Availability
Physical Description: 1 online resource (194 p.)
Language: english
Creator: Su, Nan
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: aar, aare, all, asns, asparaginase, atf4, chop, foxa, leukemia, transcription, upr
Biochemistry and Molecular Biology (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: It has been recognized for more than a decade that the availability of nutrients can alter the expression of a number of genes, including asparagine synthetase (ASNS), through distinct signaling pathways. Previous studies showed that ASNS is induced by amino acid or glucose limitation through the amino acid response (AAR) pathway or the unfolded protein response (UPR) pathway, respectively. Genomic analysis identified two cis-elements that are responsible for the induction of ASNS by both AAR and UPR pathways, the nutrient sensing response element (NSRE)-1 and ?2. Evidence from both in vitro and in vivo studies suggested that the activating transcription factor 4 (ATF4) functions as the major activator of the ASNS gene, whereas some members from the ATF and C/EBP families function as repressors. The purpose of the current study was to gain more insight into the mechanism through which the ASNS gene is regulated by nutrient limitation. Yeast two-hybrid screening, in conjunction with co-immunoprecipitation assay, identified the C/EBP homology protein (CHOP) as an ATF4 interacting partner. Reporter analysis, over-expression, and siRNA assay demonstrated that CHOP inhibits ATF4 dependent transcriptional activation of the ASNS gene by the AAR or UPR pathways both in vitro and in vivo. Chromatin immunoprecipitation (ChIP) assay suggested that CHOP is recruited to the ASNS promoter, as opposed to the prevailing hypothesis that CHOP functions by sequestering ATF4 from the DNA. Previous studies raised a debate about the correlation between ASNS expression and the resistance to asparaginase (ASNase) therapy in patients with acute lymphoblastic leukemia (ALL). The current study utilized multiple ALL cell lines from different origins, and confirmed the lack of correlation between ASNS mRNA level and ASNase sensitivity, as suggested by clinical microarray studies. However, expression profiling revealed a delayed ASNS protein expression in an ASNase sensitive cell line even though the ASNS mRNA is highly induced by ASNase treatment. The ASNS protein level was found to correlate with ASNase sensitivity in each of the ALL cell lines studied. Furthermore, although the ASNS mRNA level varies over a wide range in ALL patients, expression profiling of the ASNS gene using samples from ALL patients showed extremely low ASNS protein expression in most patients, which is consistent with the overall high efficacy of ASNase therapy. Taken together, these results demonstrate that ASNS is regulated by complex mechanisms involving fast response and feed back inhibition, at both the transcriptional and translational level. The current studies also demonstrate that the forkhead box A (FOXA) family of genes are subjects to the regulation by nutrient stresses, such as amino acid deprivation and ER stress. The induction of FOXA2 and FOXA3 by amino acid deprivation is not absolutely dependent on the ATF4-driven AAR pathway. Conversely, the FOXA gene products are not required for the induction of most of the other amino acid responsive genes. However, FOXA2 and FOXA3 seem to be involved in the regulation of the sodium-coupled neutral amino acid transporter 2 (SNAT2) gene, indicating the existence of a distinct pathway during amino acid deprivation other than the ATF4-dependent AAR pathway.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Nan Su.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Kilberg, Michael S.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0023995:00001


This item has the following downloads:


Full Text

PAGE 1

1 CHARACTERIZATION OF THE TRANSCRIPTION FACTOR NETWORK INVOLVED IN THE AMINO ACID RESPONSE AND REGULATION OF THE ASPARAGINE SYNTHETASE (ASNS) GENE BY NUTRIENT AVAILABILITY By NAN SU A DISSERTATION PRESENTED TO THE GRADUATE S CHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008 Nan Su

PAGE 3

3 To my family and friends

PAGE 4

4 ACKNOWLEDGMENTS I extend my sincere thanks to my mentor, Dr. Kilberg, for teaching me how to become a good person and an honest scientist. He guided me through ups and downs in my career, and constantly supported me in every aspect. I would also like to thank the members of my committee, Dr. Goodenow, Dr. Nic k, Dr. Robertson, and Dr. Yang for their helpful advice throughout my research. I also thank all the members the Kilberg laboratory, both past and present, for their help and friendship. This is the best team I have ever worked with. Finally, I would lik e to thank my family and my friends for all their love and support.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .................... 4 LIST OF TABLES ................................ ................................ ................................ ................................ 8 LIST OF FIGURES ................................ ................................ ................................ .............................. 9 LIST OF ABBREVIATIONS ................................ ................................ ................................ ............ 12 ABSTRACT ................................ ................................ ................................ ................................ ........ 14 CHAPTE R 1 INTRODUCTION ................................ ................................ ................................ ......................... 1 Regulation of Gene Expression by Nutrient Limitation ................................ ............................. 1 Regulation of Gene Ex pression by Amino Acid Limitation ................................ ...................... 2 Regulation of Gene Expression by Glucose Limitation and the Unfolded Protein Response ................................ ................................ ................................ ................................ .... 5 Nutr ient Limitation and Cancer Therapy ................................ ................................ ..................... 8 ASNase Therapy of Childhood Acute Lymphoblastic Leukemia (ALL) ................................ 10 Regulation of the Asparag ine Synthetase (ASNS) Gene by Nutrient Limitation ................... 12 Regulation of Genes Other Than ASNS by Nutrient Limitation ................................ ............. 14 The Transcrip tional Regulator Network in Amino Acid Response ................................ ......... 16 2 MATERIALS AND METHODS ................................ ................................ ............................... 22 Cell culture ................................ ................................ ................................ ................................ ... 22 Treatments ................................ ................................ ................................ ................................ ... 22 Yeast two hybrid Screening ................................ ................................ ................................ ....... 23 RNA Isolation and Real time Quantitative RT PCR ................................ ................................ 25 Protein Isolation and Immunoblotting ................................ ................................ ....................... 26 Transient Transfection and Expression ................................ ................................ ...................... 27 Chromatin Immunoprecipitation (ChIP) Analysis ................................ ................................ .... 28 Double Chromatin Immunoprecipitation (Double ChIP) Analysis ................................ ......... 3 0 Sh ort Interfering RNA (siRNA) Transfection ................................ ................................ ........... 31 Calcium Phosphate (CaP) Transfection ................................ ................................ ..................... 31 Co Immunoprecipitation ................................ ................................ ................................ ............. 32 Trypan Blue Exclusion and WST 1 Cell Proliferation Assays ................................ ................ 33 mRNA Stability Assay ................................ ................................ ................................ ................ 34

PAGE 6

6 3 IDENTIFICATION AND CHARACTERIZATION OF ATF4 INTERACTING PROTEINS AND THEIR FUNCTION IN THE REGULATION OF THE ASPARAGINE SYNTHETASE (ASNS) GENE DURING NUTRIENT DEPRIVATION ................................ ................................ ................................ .......................... 41 Introduction ................................ ................................ ................................ ................................ 41 Involvement of ATF4 in the Nutrient Stress Responses ................................ ................... 41 Characteristics of CHOP as a Transcription Factor ................................ ........................... 43 Function of CHOP in Apoptosis ................................ ................................ ......................... 44 Involvement of CHOP in the Nutrient Stress Responses ................................ .................. 45 Results ................................ ................................ ................................ ................................ .......... 46 Identification of ATF4 Interacting Proteins by Yeast Two Hybrid Screening ................ 46 CHOP Interacts with ATF4 in vi vo ................................ ................................ .................... 48 CHOP Inhibits Transcriptional Activation from the ASNS Promoter ............................. 49 Effect of CHOP Over Expression on Endogenous Gene Expre ssion .............................. 50 Regulation of CHOP Expression by the AAR and the UPR Pathways ............................ 51 ER Stress, but Not Amino Acid Deprivation, Triggers Ce ll Apoptosis ........................... 52 The Induction of ASNS Expression is Enhanced by CHOP Knock down ...................... 53 The Repressive Function of CHOP Requires Both N and C Terminal Portions of the Protein ................................ ................................ ................................ ......................... 54 CHOP Binding to AARE Containing Genes ................................ ................................ ..... 55 Conclusions and Discussions ................................ ................................ ................................ ..... 57 4 REGULATION OF THE ASPARAGINE SYNTHETASE (ASNS) GENE BY ASPARAGINASE (ASNase) TREATMENT IN MOLT 4 LEUKEMIA CELLS AND ITS CORRELATION WITH ASNase RESISTANCE ................................ ............................ 79 Introduction ................................ ................................ ................................ ................................ 79 ASNase Therapy for ALL Patients ................................ ................................ ..................... 79 ASNase Resistance and ASNS Up Regulation ................................ ................................ .. 81 Cell Lines Used in the Laboratory Studies of ALL ................................ ........................... 83 Results ................................ ................................ ................................ ................................ .......... 83 Characteriz ation of ASNS and Transcription Factor Expression in the MOLT 4 Cells ................................ ................................ ................................ ................................ .. 83 Assembly of the Transcription Machinery on the ASNS Promoter in the MOLT 4 Cells ................................ ................................ ................................ ................................ .. 85 Irreversibility of the ASNase Resistance of the MOLT 4R Cells ................................ .... 86 Sensitivity of Different ALL Cell Lines to ASNase Treatment. ................................ ...... 88 Analysis of the ASNS mRNA and Protein Expression in ALL Cell Lines ..................... 90 Correlation of ASNase resistance and ASNS protein expression ................................ .... 93 ASNS mRNA and protein expression in patient samples ................................ ................. 94 Conclusions and Discussions ................................ ................................ ................................ ..... 95

PAGE 7

7 5 REG ULATION OF THE FOXA FAMILY OF GENES BY NUTRIENT STRESS AND THEIR INVOLVEMENT IN THE CONTROL OF AMINO ACID RESPONSIVE GENES ................................ ................................ ................................ ................................ ....... 112 Introduction ................................ ................................ ................................ ............................... 112 Function of FOXA Proteins in Activating Transcription ................................ ................ 112 Function of FOXA Proteins in Development and Organogenesis ................................ .. 113 Function of FOXA Proteins in Hormone Control of Gene Expression and Glucose Metabolism ................................ ................................ ................................ ..................... 114 Regulation of the FOXA Genes ................................ ................................ ........................ 116 Results ................................ ................................ ................................ ................................ ........ 117 Regulation of the FOXA Genes by Amino Acid Limitation and ER Stress .................. 117 Genes by Amino Acid Deprivation ................................ ................................ ............... 121 Genomic analysis of the FOXA3 gene ................................ ................................ ............. 123 Function of FOXA2 and FOXA3 on Amino Acid Responsive Genes ........................... 126 Conclusions and Discussions ................................ ................................ ................................ ... 128 6 CONCLUSIONS AND FUTURE DIRECTIONS ................................ ................................ .. 148 Conclusions ................................ ................................ ................................ ............................... 148 Future Direction ................................ ................................ ................................ ........................ 152 APPENDIX A EFFECT OF AMINO ACID DEPRIVATION AND ER STRESS DOUBLE STRESS ON THE EXPRESSION OF THE FOXA3 GENE ................................ ................................ 156 B INVOLVEMENT OF SEVERAL HISTONE ACETYLTRANSFERASES IN THE AMINO ACID DEPRIVATI ON MEDIATED TRANSCRIPTIONAL ACTIVATION ..... 157 C INVOLVEMENT OF HUMAN UBC9 IN THE ATF4 MEDIATED TRANSCRIPTIONAL ACTIVATION FROM THE ASNS PROMOTER .......................... 159 D INTERACTION BETWEEN ATF4 AND TBP IN VIVO ................................ ...................... 160 E IDENTIFICATION OF ATF4 INTERACTING PROTEINS BY CO IMMUNOPRECIPITATION ................................ ................................ ................................ ... 161 LIST OF REFERENCES ................................ ................................ ................................ ................. 162 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ........... 179

PAGE 8

8 LIST OF TABLES Table page 2 1. Antibodies used for ChIP and/or Immunoblotting ................................ ................................ ... 35 2 2. Primer sets and anneal ing temperatures for RT qPCR ................................ ............................ 36 2 3. Primer sets and annealing temperatures for CHIP qPCR ................................ ........................ 38 2 4. Plasmids used for protein ex pression, luciferase reporter assay, and yeast two hybrid screening. ................................ ................................ ................................ ................................ ..... 39 3 1. Genes identified from the yeast two hybrid screening. ................................ ............................ 68

PAGE 9

9 LIST OF FIGURES Figure page 1 1. Model for the translational control of GCN4 and ATF4. ................................ .................... 18 1 2. Signaling pathways during the unfolded pro tein response. ................................ ................. 19 1 3. Characterization of the ASNS proximal promoter. ................................ .............................. 20 1 4. A model for transcriptional regulation of the ASNS gene by amino acid limitation. ........ 21 3 1. Schematics of different ATF4 fragments tested as bait in the yeast two hybrid screen. ... 63 3 2. Self activation test of the full length ATF4 protein as bait. ................................ ................ 64 3 3. Self activation test of different fragments from the ATF4 activation domain as bait. ...... 65 3 4. Self activation test of 45 aa fragments from the ATF4 activation domain as bait. ........... 66 3 5. Yeast two hybrid screen with ATF4 bZIP domain a s bait. ................................ ................. 67 3 6. Identification of CHOP as an ATF4 interacting partner by yeast two hybrid screen. ....... 70 3 7. Interaction of CHO P and ATF4 in vivo ................................ ................................ ............... 71 3 8. Effect of CHOP on ASNS promoter driven transcription activated by HisOH or ATF4. ................................ ................................ ................................ ................................ ...... 72 3 11. Effect of siRNA mediated CHOP knock down on endogenous ASNS expression. ......... 75 3 12. E ffect of truncated CHOP proteins on transcription driven by the ASNS promoter. ........ 76 3 13. A ssociation of CHOP with genes containing a C/EBP ATF composite site. ................... 77 3 14. C o occupancy of CHOP and ATF4 at C/EBP ATF composite sites. ................................ 78 4 1. E xpression of ASNS, ATF4, and ATF3 in the parental and resistant MOLT 4 cells. ...... 99 4 2. A ssociation of Pol II, ATF4, ATF3, and parental and resistant MOLT 4 cells. ................................ ................................ .................. 100 4 3. A ssociation of GTFs and acetylated histones with the ASNS promoter in the parental and resistant MOLT 4 cel ls. ................................ ................................ ................................ 101 4 4. Characterization of the ASNS expression in the ASNase resistant MOLT 4 cells after a prolonged drug withdraw. ................................ ................................ ................................ 102

PAGE 10

10 4 5. A MOLT 4P and MOLT 4R0 cells. ................................ ................................ ........................ 103 4 6. The ALL cell lines show differential growth patterns and sensitivities to ASNa se treatment. ................................ ................................ ................................ .............................. 104 4 7. Growth inhibition effect of different concentration of ASNase on ALL cell lines. ........ 105 4 8. Effect of A SNase concentration on cell death of ALL cell lines. ................................ ..... 106 4 9. mRNA expression of ASNS in ALL cell lines during ASNase treatment. ...................... 107 4 10. Protein expression of ASNS in ALL cell lines during ASNase treatment. ...................... 108 4 11. Relationship between ASNase IC 50 and ASNS protein or mRNA expression. ............... 109 4 12. Effect of siRNA mediated ASNS knock down on the sensitivity to ASNase induced apoptosis in HepG2 cells. ................................ ................................ ................................ .... 110 4 13. Protein expression of ASN S in childhood ALL primary patient samples. ....................... 111 5 1. Functional domains in human FOXA1, 2 and 3 proteins. ................................ ................. 133 5 2. Regulatio n of the FOXA family of genes by amino acid deprivation and ER stress. ..... 134 5 3. The expression of ASNS, SNAT2, FOXA1, FOXA2, and FOXA3 in the pregnant mice fed with normal or low protein die t. ................................ ................................ .......... 135 5 4. Time couse expression of FOXA genes during amino acid deprivation and ER stress. 136 5 5. Change of FOXA2, FOXA3 and p21 mRNA stability following amino acid deprivation. ................................ ................................ ................................ ........................... 137 5 6. Effect of blockage of de novo transcription on the induction of FOXA2 and FOXA3 by amino acid deprivation. ................................ ................................ ................................ .. 138 5 7. Time couse expression of FOXA protein during amino acid deprivation and ER stress. ................................ ................................ ................................ ................................ ..... 139 5 8. Effect of ATF4 and/or ATF5 knock down on the expression of FOXA2 and FOXA3. 140 5 9. down on the expression of FOXA2 and FOXA3. .................... 141 5 10. Genomic analysis of the FOXA3 gene. ................................ ................................ .............. 142 5 11. Pr omoter and enhancer analysis of the FOXA3 gene. ................................ ....................... 143 5 12. Effect of FOXA3 on the transcription from the ASNS or SNAT2 promoter.. ................. 144

PAGE 11

11 5 13. Effect of FOXA2 and FOXA3 knock down on the expression of GLUT2.. ................... 1 45 5 14. Effect of FOXA2 and FOXA3 double knock down on the expression of AARE containing genes. ................................ ................................ ................................ .................. 146 5 15. Effect of FOXA2 or FOXA3 single knock down on the expression of SNAT2 and CHOP. ................................ ................................ ................................ ................................ ... 147 A 1. Regulation of the FOXA3 gene by amin o acid deprivation and ER stress double stress. ................................ ................................ ................................ ................................ ..... 156 B 1. Effect of over expression of several histone acetyltransferases on the ATF4 dependent transcription from the ASNS promoter. ................................ ............................ 157 B 2. Effect of E1A over expression on the transcription from the ASNS promoter. .............. 158 C 1. Effect of hUBC9 over expression on the t ranscription from the ASNS promoter. ......... 159 D 1. Test for Interaction of TATA binding protein (TBP) and TFIIB with ATF4 in vivo ..... 160 E 1. Identification of ATF4 interacting proteins by co IP analysis. ................................ ......... 161

PAGE 12

12 LIST OF ABBREVIATION S AAR Amino acid response AARE Amino acid responsive element ALL Acute lymphoblastic leukemia ASNase Asparaginase ASNS Aspar agine synthetase ATF Activating transcription factor bZIP Basic leucine zipper C/EBP CCAAT/enhancer binding protein CAT 1 Cationic amino acid transporter 1 CBP CREB binding protein ChIP Chromatin immunoprecipitation CHOP CCAAT/enhancer binding protein homo logous protein CRE cAMP response element ER Endoplasmic reticulum ERK Extracellular signal regulated kinase FBS Fetal bovine serum FOXA Forkhead box A GAPDH Glyceraldehyde 3 phosphate dehydrogenase GCN G eneral control nonderepressiable GFP Green fluorescent protein GLUT Glucose transporter GTF General transcription factor HAT Histone acetyltransferase

PAGE 13

13 HEK Human embryonic kidney HNF Hepatocyte nuclear factor mRNA Messenger RNA NSRE Nutrient sensing respon sive element ORF Open reading frame PCR Poly chain reaction Pol II RNA Polymerase II qPCR Quantitative polymerase chain reaction RT qPCR Reverse transcriptase quantitative polymerase chain reaction SEM Standard error of the means SNAT2 Sodium coupled neutr al amino acid transporter gene member 2 SV40 Simian virus 40 TAF TBP associated factor TAF TBP associated factor TBP TATA box binding protein uORF Upstream open reading frame UPR Unfolded protein response VEGF Vascular endothelial growth factor WST 1 2 (4 iodophenyl) 2 (4 nitrophenyl) 5 (2,4 disulfophenyl) 2H tetrazolium

PAGE 14

14 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy CHARACTERIZATION OF THE TRANSCRIPTION FA CTOR NETWORK INVOLVE D IN THE AMINO ACID RESPONSE AND REGULATION OF TH E ASPARAGINE SYNTHET ASE (ASNS) GENE BY NUTRIENT AVA ILABILITY By Nan Su December 2008 Chair: Michael S. Kilberg Major: Medical Sciences -Biochemistry and Molecular Biology It has b een recognized for more than a decade that the availability of nutrients can alter the expression of a number of genes, including asparagine synthetase (ASNS), through distinct signaling pathways. Previous studies showed that ASNS is induced by amino acid or glucose limitation through the amino acid response (AAR) pathway or the unfolded protein response (UPR) pathway, respectively. Genomic analysis identified two cis elements that are responsible for the induction of ASNS by both AAR and UPR pathways, th e nutrient sensing response element (NSRE) 1 and 2. Evidence from both in vitro and in vivo studies suggested that the activating transcription factor 4 (ATF4) functions as the major activator of the ASNS gene, whereas some members from the ATF and C/EBP families function as repressors. The purpose of the current study was to gain more insight into the mechanism through which the ASNS gene is regulated by nutrient limitation. Yeast two hybrid screening, in conjunction with co immunoprecipitation assay, identified the C/EBP homology protein (CHOP) as an ATF4 interacting partner. Reporter analysis, over expression, and siRNA assay demonstrated that CHOP inhibits ATF4 dependent transcriptional activation of the ASNS gene by the AAR or UPR pathways both in vitro and in vivo Chromatin immunoprecipitation (ChIP) assay suggested that

PAGE 15

15 CHOP is recruited to the ASNS promoter, as opposed to the prevailing hypothesis that CHOP functions by sequestering ATF4 from the DNA. Previous studies raised a debate about th e correlation between ASNS expression and the resistance to asparaginase (ASNase) therapy in patients with acute lymphoblastic leukemia (ALL). The current study utilized multiple ALL cell lines from different origins, and confirmed the lack of correlation between ASNS mRNA level and ASNase sensitivity, as suggested by clinical microarray studies. However, expression profiling revealed a delayed ASNS protein expression in an ASNase sensitive cell line even though the ASNS mRNA is highly induced by ASNase t reatment. The ASNS protein level was found to correlate with ASNase sensitivity in each of the ALL cell lines studied. Furthermore, although the ASNS mRNA level varies over a wide range in ALL patients, expression profiling of the ASNS gene using samples from ALL patients showed extremely low ASNS protein expression in most patients, which is consistent with the overall high efficacy of ASNase therapy. Taken together, these results demonstrate that ASNS is regulated by complex mechanisms involving fast r esponse and feed back inhibition, at both the transcriptional and translational level. The current studies also demonstrate that the forkhead box A ( FOXA ) family of genes are subjects to the regulation by nutrient stresses, such as amino acid deprivation and ER stress. The induction of FOXA2 and FOXA3 by amino acid deprivation is not absolutely dependent on the ATF4 driven AAR pathway. Conversely, the FOXA gene products are not required for the induction of most of the other amino acid responsive genes. However, FOXA2 and FOXA3 seem to be involved in the regulation of the sodium coupled neutral amino acid transporter 2 (SNAT2) gene, indicating the existence of a distinct pathway during amino acid deprivation other than the ATF4 dependent AAR pathway.

PAGE 16

1 CHA PTER 1 INTRODUCTION Regulation of Gene Expression by Nutrient Limitation Nutrients are the fundamental elements for life. All living organisms have evolved complex mechanisms to adapt to the change of nutrient availability in the environment. With the ad vancement of molecular biology, researchers have found direct links between gene expression and nutrient availability. These links are especially prominent in single cell organisms such as bacteria or yeast, as demonstrated by several classic paradigms in cluding the lactose operon and the tryptophan operon (Cenatiempo, 1986) The responses of mammalian cells to the change of environmental nutrients are much more complicated, and lead to more intriguing questions, such as: do all individual cells within a tissue respond to the change of nutrient a vailability? What cells respond the most to nutrient limitation or over enrichment? Does nutrient availability influence hormone or neuronal regulation in mammals? Is our behavior subjected to regulation by the types of food we eat? During the effort of answering these questions, it turns out that the availability of nutrients has a far more profound influence than researchers originally thought. However, it yet again surprises us that, in many cases, numerous downstream processes converge onto a single signaling pathway, such as that involved in cellular responses to amino acid limitation. As essential nutritional molecules, amino acids modulate a number of fundamental processes in mammalian cells and are involved in many diseases, including diabetes (Hoffer, 1993) Kwashiorkor (Roediger, 1995) and hepatic encephalopathy (Mizock, 1999) It has also been recorded that limiting individual essential amino acids in the diet leads to extension of life span (Zimmerman et al. 2003) Level/availability of amino acids within cells reflects the nutritional protein/amino acid status of the entire organism, and therefore, is constantly

PAGE 17

2 monitored by a sensing system. Fluctuation of amino acid levels, which results from dietary o r disease conditions, leads to the alteration of a number of cellular processes. Under these circumstances, the amino acids serve as signal molecules rather than simply as precursors for protein synthesis. Regulation of Gene Expression by Amino Acid Limit ation The regulation of gene expression in response to amino acid limitation has been most extensively investigated in the yeast Saccharomyces cerevisiae Limitation of environmental amino acids or even a single amino acid leads to lowered level of aminoa cyl tRNA. The corresponding uncharged tRNA binds with high affinity to the protein kinase GCN2 (general control non derepressible 2), because the structure of its C terminal portion resembles the histidyl tRNA synthetase tRNA binding site and other aminoa cyl tRNA synthetases (Hinnebusch, 1997) With tRNA bound, GCN2 will dimerize with another GCN2 monomer and autophosphorylates itself at specific threonine residues, which leads to the activation of the inh erent kinase activity (Hinnebusch, 2005) GCN2 phosphorylates the alpha subunit of 2a inhibits the exchange of GDP for GTP on the eIF2 complex, which is catalyz ed by eIF2B, and Met tRNA 40s ternary complex and reduces global translation (Hershey, 1991) However, the GCN2 induced decline in the rate of protein synthesis actually favors increased translation of a small subset of mRNAs, o ne of which is the transcription factor GCN4 (Hinnebusch, 1997) that have uORFs (upstream open reading frames) contain ternary complex is abundant, the ribosomes scanning downstream from uORF1 will reinitiate

PAGE 18

3 translation at inhibitory uORF 2, 3, or 4, so a very low level of GCN4 protein is made. By ternary complex is reduced, there is a delay in reinitiation that allows the scanning ribosomes to bypass the inhibitory uORFs 2 4 and instead translate the GCN4 coding region, thus high levels of the GCN2 protein are produced (Figure 1 1A). GCN4 is a member of the basic leucine zipper (bZIP) transcription factor family, and can bind as homodimer to the palindromic sequence TGACTCA (Hope and Struhl, 1985) GCN4 is a transcriptional activator of at least 40 genes encoding amino acid biosynthetic enzymes, and is responsible for the activation of hundreds of genes in response to amino acid limitation (Hinnebusch, 2005) Approximately 1000 genes were shown to be induced by histidine starvation in yeast, with a microarray analysi s (Natarajan et al., 2001) Amon g these genes, 539 genes were demonstrated to depend on GCN4 for their full induction by amino acid starvation. There are also many genes whose inhibition by amino acid master (Natarajan et al., 2001) There is no GCN4 protein homolog in mammalian cells. However, a bZIP transcription factor, ATF4, is believed to be the functional counterpart to GCN4. Although ATF4 shares ve ry little sequence similarity with GCN4, its regulation and function are amazingly similar to those of GCN4. ATF4 can bind as homodimer to the palindromic cAMP responsive element (CRE) TGATGTCA, and it can also bind to non palindromic sequences, as a hete rodimer with other bZIP transcription factors, such as members from the CCAAT element binding protein (C/EBP) family and the AP 1 family (Hsu et al., 1994; Hai and Curran, 1991; Shimizu et al., 1998; Vallejo et al., 1993) The expression of A TF4 is regulated by amino acid starvation through a

PAGE 19

4 similar mechanism to that of GCN4 (Lu et al., 2004; Vattem and Wek, 2004) frame and overlaps with the ATF4 ORF. In the fed state, when translation reinitiation is eff icient, the ribosome translates uORF2, so a very low level of ATF4 protein is made. However, in the starved state, when translation reinitiation becomes less efficient, the ribosome scans past uORF2 and initiates at the ATF4 initiation codon, so high leve ls of the ATF4 protein are produced (Figure 1 1B). The similarity between ATF4 and GCN4 in their regulation and transcription activity prompts the possibility that ATF4 is a critical component in the mammalian AAR pathway. Indeed, the study of many amino acid responsive genes has shown definite involvement of ATF4 in their regulation, and a microarray study using wild type or ATF4 knockout fibroblasts demonstrated that ATF4 controls many genes involved in amino acid metabolism and transport, and in redox chemistry (Harding et al., 2003) During early characterization of several amino acid responsive genes including asparagine synthetase (ASNS) and C/EBP amino acid response element (AARE) was identified, with a consensus sequence of TGATGXAAX (Barbosa Tessmann e t al., 1999b; Guerrini et al., 1993; Bruhat et al., 2000) Interestingly, all the AAREs identified so far are considered C/EBP ATF composite sites, containing half the consensus sequence for activating transcription factor (ATF) proteins [ TGACGT (C/A) (G /A)], and another half site for CCAAT/enhancer binding proteins (C/EBPs) (ATTGCGCAAT) (Fawcett et al., 1999) Electrophoresis mobility shifting assay (EMSA) data and chromatin immunoprecipitation (ChIP) has revealed increased binding of ATF an d C/EBP transcription factors, including ATF4 and (Siu et al., 2002; Siu et al., 2001; Chen et al., 2004)

PAGE 20

5 Besides amino acid limitation, glucose deprivation also causes induct ion of ATF4 expression (details will be described in later paragraphs). Given that two major nutrient stress response pathways induce ATF4 expression, ATF4 mediated transcription represents a critical component in the cellular response to these pathways. Regulation of Gene Expression by Glucose Limitation and the Unfolded Protein Response Carbohydrates are used by mammals as the main source of energy. They are broken down to monosaccharides, such as glucose, by the digestion system, and these monosaccha rides are utilized by individual cells to generate energy to maintain normal cellular processes. Decreases in cellular carbohydrate levels, especially the glucose level, alter the expression of a number of genes. Glucose regulated proteins (GRPs) GRP78 a nd GRP98 were first identifed as being highly induced when cultured mammalian cells were depleted of glucose (Pouyssgur et al., 1977) Subsequently, more genes were found to be regulated by glucose limitation such as the transcription factor CHOP (Carlson et al., 1993) and the glucose transporter GLUT1 (Kitzman, Jr. et al., 1993; Wertheimer et al., 1991) and GLUT3 (Nagamatsu et al., 1994) Glucose deprivation causes abnormal accumulation of mi sfolded glycoproteins in the endoplasmic reticulum (ER) that causes an imbalance between the demand for protein folding and the capacity of the ER for protein folding, thereby causing ER stress. Subsequently, it was discovered that a variety of other agen ts, such as the protein glycosylation inhibitor tunicamycin (Chan and Egan, 2005) and the endoplasmic reticulum (ER ) Ca 2+ ATPase inhibitor thapsigargin (Rogers et al., 1995) also induce UPR by disruption of ER functions. Eukaryotic cells have developed a mechanism to sense and respond to these ER stresses, termed ER stress response (ERSR) or unfolded protein response (UPR) (Bernales et al., 2006) During UPR, signal transduction can be initiated from three ER localized transmembrane signal transducers, inositol requiring kinase 1 (IRE1) (Sidrauski et al., 1998; Yoshida et al.,

PAGE 21

6 2001) the double stranded RNA activated protein kinase like ER kinase (PERK ) (Harding et al., 2 000b) and the transcription factor activating transcription factor 6 (ATF6) (Yoshida et al., 2000; Yoshida et al., 2001) Among these three proteins, IRE1 was the first identified as a component during UPR signaling, in the yeast Saccharomyces cerevisiae Under normal conditions, IRE1 protein is maintained in an inactive form by intereacting with the protein chaperone BiP. Upon activation of unfolded protein response, IRE1 protein is released from BiP and undergoes homodimerization. This dimerization leads to a trans autophosphorylation of IRE1 and activates its RNase activity. IRE1 then al ternatively splices the mRNA of a transcription factor HAC1, by cleaving a 252bp intron fragment. The spliced mRNA will generate a transcriptionally active HAC1 protein, which binds to and activates many genes containing a yeast UPR element (TGACGTG(C/A)) (Figure 1 (Tirasophon et al., 1998) (Wang et al., 1998a) These two IRE1 proteins do not seem to differ in function, as their cleavage specificities are very similar. However, they do exhibit pancreas and placenta (Tirasophon et al., 1998) epithelial cells (Wang et al., 1998a) Upon activation of the UPR, the activated IRE1 splices the mRNA of the transcription factor X box binding protein 1 (XBP1), generating a frameshifted mRNA, wh ich produces a larger form of XBP1 that contains a novel transcriptional activation domain it its C terminus (Figure 1 2). XBP1 subsequently activates a number of UPR genes containing either a mammalian ER stress response element (ERSE, CCAAT (N9) CCACG) or a mammalian UPRE (TGACGTGG/A) that is necessary and sufficient for UPR gene activation (Yoshida et al., 1998; Yoshida et al., 2001)

PAGE 22

7 One group of genes induced by IRE1/XBP1 pathway are those involved in en doplasmic reticulum associated protein degradation (ERAD), and cells that are deficient in either IRE1 or XBP1 are defective in ERAD. Several studies provided evidence that IRE1 may also promote cell death after ER stress by activating caspases (Yoneda et al., 2001; Nakagawa et al., 2008) The second branch of UPR signaling is initiated by activating transcription factor 6 (ATF6). Under normal conditions, ATF6 is strictly located in the ER memb rane. Upon UPR activation, ATF6 is released from BIP and translocated to the Golgi complex, where it is cleaved by the proteases S1P and S2P. The cytosolic fragment from the proteolysis migrates to the nucleus and functions as a transcriptional activator targeting many ER stress responsive genes (Figure 1 well as to the ER stress responsive eleme nt (ERSE) (Yoshida et al., 2001) However, no UPR directly regulate a number of ER stress responsible genes, such as CHOP (Yoshida et al., 2000) CHOP plays an essential role in the response to a wide variety of cell stresses and induces cell cycle arrest and apoptosis in response to ER stress (Zinszner et al., 1998) The PERK branch of the UPR initiates adaptive mechanisms to help clear the translation machinery and to protect cells from program med cell death. Upon activation of the UPR, PERK undergoes oligomerization, which activates its kinase activity that resides in its cytosolic domain (Harding et al., 1999) (Figure 1 very same t by PERK inhibits its function, and results in a global slowdown of protein translation. This confers to the cell an immediate protection by preventing further accumulation o f unfolded

PAGE 23

8 proteins in the ER. In the meanwhile, the translation of specific mRNA species, such as ATF4, is enhanced through the mechanism described above for amino acid limitation. As a consequence, amino acid limitation and ER stress overlap with regar d to the regulation of many genes, such as ASNS and CHOP. However, there are also many genes that are only regulated by one of the stress reponses but not by the other. Recent studies have demonstrated that the amino acid transporter SNAT2 is only induce d by the AAR pathway, but not by the UPR pathway. Indeed, the UPR pathway seems to generate repressive signals that override the activation signal from the AAR pathway, suggesting crosstalk between these two pathways to allow greater specificity in transcr iptional activation (Gjymishka et al., 2008) Nutrient Limitation and Cancer Therapy Tumor cells in general are more tolerant to different cellular stress conditions, such as hypoxia and nutrient deprivation, partly due to the development of adaptive mechanisms. During the early stages of carcinogenesis, especially prior to extensive vascularization, tumor cells are exposed to insufficient supply of oxygen and metabolites. Even after the tumor increases in size, the microenvironment within the tumor is still heterogeneous and usually displays a gradient of oxygen, growth fa ctors, and essential nutrients such as glucose and amino acid (Dang and Semenza, 1999; Helmlinger et al., 1997) As a consequence, one of the key steps of tumo r progression is angiogenesis, which has become one of the most popular targets for cancer therapy. It is known that during the progression of many types of tumors, the vascular endothelial growth factor (VEGF) gene is induced and contributes to angiogene sis. The expression of VEGF is controlled by many cellular stress conditions, including oxygen tension (Minchenko et al., 1994; Shweiki et al., 1992) ER stress (Parker et al., 2001; Abcouwer et al. 2002) and amino acid deprivation (Abcouwer et al., 2002) The development of several VEGF

PAGE 24

9 targeted agents has been shown to benefit patients with advanced stage malignancies (Ellis and Hicklin, 2008) Depletion of oxygen and nutrients is an effective approach to eliminate the population of fast growing cancer cells. However, certain types of c ancer cells are able to survive prolonged therapy due to their exceptional tolerance to nutrient deprivation. Izuishi et al. demonstrated that several pancreatic cell lines display extremely long survival, even under complete nutrient deficiency condition s. Moreover, some poorly differentiated colon and gastric cancer cell lines were also found to be tolerant to nutrient deprivation (Izuishi et al., 2000) These characteristics can be obstructions to traditional cancer therapy, but may also be targeted for novel cancer therapy (Izuishi et al., 2000) Another feature that allows cancer cells to survive cancer therapies is their ability to escape programmed cell death (PCD), including apoptosis (type I PCD) and autophagy (type II PCD). Apoptosis is an outcome of the caspase dependent cascade, whereas autophagy is a normal caspase independent degradation process that is evolutionarily conserved amo ng all eukaryotes (Broker et al., 2005; Saeki et al., 2000; Xu et al., 2006) When cells are exposed to nutrient limitation, they simultaneously decrease overall protein synthesis and incre ase rates of protein degradation by an autophagic pathway (Mortimore and Poso, 1988) Most cancer therapeutics target rapidly proliferating cells, and induce programmed cell death to eliminate their population. However, some cancer cells are able to enter quiescent states instead of undergoing apoptosis or autophagy, and thereby, survive t he treatment. Regrowth of these quiescent cells can lead to tumor reoccurrence and metastasis, which are the leading causes of cancer mortality.

PAGE 25

10 ASNase Therapy of Childhood Acute Lymphoblastic Leukemia (ALL) Leukemia is a type of cancer originating from blood or bone marrow cells. It is characterized by an abnormal proliferation of blood cells, usually white blood cells (leukocytes). There are two classification systems of leukemia. When classified by the progression rate, leukemia can be subdivided t o acute leukemia and chronic leukemia. Acute leukemia is characterized by the rapid increase of immature blood cells. It can occur in children and young adults, and is a more common cause of death for children in the US than any other type of malignant d isease. Immediate treatment is required in acute leukemias due to the rapid progression and accumulation of the malignant cells. Chronic leukemia is characterized by the excessive build up of relatively mature, but still abnormal, blood cells. These leu kemic cells are produced at a much higher rate than normal cells, resulting in many abnormal white blood cells in the blood, but the chronic form of the disease typically taking months or years to progress. Chronic leukemia mostly occurs in people over the age of 60, but can theoretically occur in any age group. Unlike acute leukemia, chronic leukemias are sometimes monitored for some time before treatment to ensure maximum efficacy of therapy. When classified according to what type of blood cell is affec ted, leukemias can be subdivided to lymphoblastic leukemia and myeloid leukemia. Acute lymphoblastic leukemia (ALL) is the most common form of childhood cancer accounting for 32% of pediatric patients (Ries et al., 1999) The rate of this disea se is approximately four per 100,000 per year, and it represents approximately 40% of all leukemias diagnosed in the United States (Pierce et al., 1969; Fraumeni, Jr. and Miller, 1967) A LL primarily occurs in young children, with a peak at 3 to 4 years of age. However, this disease also affects adults, especially those of age 65 and older.

PAGE 26

1 1 The treatment of acute lymphoblastic leukemia is a model of modern cancer therapy. Standard trea tments of ALL involve chemotherapy and radiation. In general, the survival rates are about 85% in children and 50% in adults. Part of the success is attributed to the discovery of the chemotherapeutic agent asparaginase (ASNase). ASNase is an enzyme tha t was serendipitously discovered in the guinea pig serum. In an effort to use rabbit anti lymphoma antibodies to treat mice with lymphoma, Kidd et al. (KIDD, 1953) discovered that the control guinea pig serum along, was sufficient to cause tumor regression. Broome (Broome, 1961) then demonstrated that the asparaginase activity in the serum was responsible for this anti lymphoma effect. One year later, ASNase isolated from E. coli was shown to be as effective as the guinea pig serum in promoting tumor regression (Mashburn and Wriston, Jr., 1964) Since then, ASNase has quickly become one of the most popular agents for the treatment of ALL. When used alone, ASNase induces remission in 63% of patients (Nesbit et al., 1981) and it is now universally used during the induction chemotherapy to bring about bone marrow remission. For children with low risk ALL, standard therapy usually consists of three drugs, prednisone, L asparaginase, and vincristine, for the first month of treatment. Despite many side e ffects induced by ASNase treatment, such as gastrointestinal system distress, impairment of CNS function, hypersensitivity, and renal failure, ASNase is so effective against ALL cells that it is still used as a key component of all induction protocols for treatment of childhood ALL. However, the main drawback of ASNase therapy is the emergence of drug resistance in relapsed patients, such resistance was shown to be related to elevated ASNS expression in ASNase resistant leukemia cells (Oettgen and Schulten, 1969) However, recent studies using m icroarray technologies with patient samples have brought controversy because the ASNS mRNA content in the ALL patients does not necessarily reflect their sensitivity to

PAGE 27

12 ASNase therapy (Fine et al., 2005) Thus, it remained unclear whether the up regulation of the ASNS gene causes ASNase resistance. The study described in Chapter 4 specifically addresses this question, and provides a hypothesis that integrate s the controversial observations made in different studies. Regulation of the Asparagine Synthetase (ASNS) Gene by Nutrient Limitation Asparagine synthetase (ASNS) is one of the earliest amino acid responsive genes to be well characterized. As described in the previous paragraph, an unusually low level of ASNS expression is responsible, though it may not be the only cause, for the selective sensitivity of leukemia cells to ASNase. ASNase resistant leukemia cells have an elevated ASNS expression (Haskell and Canellos, 1969; Hutson et al., 1997; Aslanian et al., 2001; Aslanian and Kilberg, 2001) and inhibiting ASNS activity or blocking it s synthesis in leukemia cells has become a potential strategy to enhance the efficacy of ASNase. Thus, the question of how the ASNS gene is regulated by amino acid limitation is also clinically important. Gong et al. (Gong et al., 1991) first dete rmined that the ASNS mRNA content was increased in cells deprived of amino acid. Hutson et al. (Hutson et al., 1996) then demonstrated that the ASNS mRNA content was increased following total amino acid deprivation or depletion of any single essential amino acid, and that the levels to which ASNS mRN A are elevated are similar in these two conditions. Guerrini et al. (Guerrini et al., 1993) identified a region fr CATGATG element (AARE). Barbosa Tessmann et al. (Barbosa Tessmann et al., 1999b; Barbosa Tessmann et al., 1999a) demonstrated that transcription from the huma n ASNS gene is also induced by glucose deprivation, and that this activation is mediated by the UPR. In vivo footprinting revealed five protein binding sites within the ASNS proximal promoter region, which contributes to the nutrient control this gene. T he five sites include three GC boxes (GC I,

PAGE 28

13 GC II, and GC III), and two novel sequences which showed different footprinting patterns in response to either amino acid or glucose deprivation (Figure 1 3) (Barbosa Tessmann et al., 2000) The two novel sites w ere originally labeled sites V and VI, and site V contains the AARE CATGATG (Guerrini et al., 1993) Barbo sa Tessmann et al. (Barbosa Tessmann et al., 2000) demonstrated that sites V and VI are also responsible for the induction of the ASNS transcription following activation of the UPR pathway. The UPR activation demonstrates that these ASNS promoter elements serve in a broader capacity than simply as an amino acid response element and to reflect this broader substrate detecting capability, the site V is now referred to as nutrient sensing response element 1 (NSRE 1). Site VI, which is 11 nucleotides downstrea m from NSRE 1 and also required for activation by both the AAR and the UPR pathways, is referred to as NSRE 2 (Siu et al., 2001) (Figure 1 3). Zhong et al. (Zhong et al., 2003) demonstrated that both NSRE 1 (nt 68 to 60) and NSRE 2 (nt 48 to 43) are absolutely required for induction of the ASNS gene following activation of either the AAR or the UPR pathway. Moreover, the 11 bp distance between these two elements, which corresponds to one turn of DNA helix, is critical for their activity. It was also demonstrated that the combination of NSRE 1 and NSRE 2 has enhancer activity in that they function in an orientation and position independent manner. As described in previous paragraphs, every AARE identified site for members of the ATF/CREB family and a half site for C/EBP family members (Wolfgang et al., 1997) Chrom atin immunoprecipitation (ChIP) has revealed increased binding of many bZIP transcription factors, including ATF4, (Chen et al., 2004) Chen et al. (Chen et al., 2004) proposed a model in which there is a specific temporal relationship for the de no vo synthesis of

PAGE 29

14 these factors in response to nutrient stress and subsequently, a specific order for their binding to the AARE site. As shown in Figure 1 4, during Phase I (0 4 h after amino acid limitation), ATF4 binding to the ASNS promoter NSRE 1 site i ncreases as a result of an increased abundance of nuclear ATF4 protein. This event takes place within 30 45 min of amino acid removal and leads to a localized histone acetylation, presumably through the recruitment of an unidentified histone acetyltransfe unidentified co activator complexes, it is assumed that ATF4 functions to recruit the general transcription machinery and the transcription initiation complex, including RNA polymerase II (Chen et al., 2004) As the period of amino acid deprivation continues during hours 4 12 (Phase II), transcription from the ATF3 gene is enhanced and the ATF3 mRNA is stabilized (Pan et al., 2007) resulting in increased nuclear ATF3 protein levels. Within the same time frame, there is (Chen et al., 2004) As a result, the possibly in cooperation with co repressor complexes, their binding results in a relative decline in the rate of ASNS transcription. The self limiting program has been show n to occur for several other AARE containing genes (Thiaville et al., 2008) Regulation of Genes Other Than ASNS by Nutrient Limitation Besides ASNS, the current study also involv es a number of other genes that are regulated by the AAR or UPR pathways, including CHOP, SNAT2, VEGF, cationic amino acid transporter 1 (CAT 1), and TRB3. All these genes are up regulated upon AAR activation, and each of them has at least one AARE that i s responsive to the induction by amino acid limitation. CCAAT/enhancer binding protein (C/EBP) homology protein (CHOP), also known as growth arrest and DNA damage protein 153 (GADD153), is a bZIP transcription factor. The expression of CHOP is rapidly in creased after either amino acid starvation or ER stress. The CHOP gene

PAGE 30

15 contains at least two ERSE (ERSE 1 and ERSE 2) elements and one AARE element (Yoshida et al., 2000; Ubeda and Habener, 2000) Although over expression o expression of CHOP, its induction by ER stress is nearly completely blocked in PERK null cells S51A cells. These results suggest that the PERK/eIF2a signaling pathway plays an essential role in the induction of CHOP in ER st ress, and is dominant over that through the ATF6 or Ire1/XBP 1 signaling pathway (Scheuner et al., 2001; Harding et al., 2000a) The CHOP AARE element was identified to be essential for transcription activation from the CHOP promoter by amino acid limitation. The bZIP transcription f actor ATF2 and ATF4 were shown to be involved in the amino acid regulation of CHOP (Bruhat et al., 2000) Whereas the induction of CHOP by stress conditions and its involvement in cell cycle arrest and apoptosis has been well characterized, less is known about its function as a transcriptional activator/repressor. The expression of the Na + coupled neutral amino acid transporter 2 (SNAT2) is induced by amino acid limitation through the Amino Acid Response (AAR) pathway (Bain et al., 2002; Gaccioli et al., 2006) Previous analysis of the SNAT2 gene structure identi fied a 9 bp AARE in the first intron that acts as a transcriptional enhancer upon amino acid starvation (Palii et al., 2004) Evidence from in vitro studies established that members of the ATF and C/EBP families regulate SNAT2 gene transcription by interacting with this intronic AARE (Palii et al., 2006) The expression of SNAT2 is not affected by ER stress, consistent with the lack of an E R stress response element (ERSE) in the SNAT2 gene. Interestingly, whereas the NSRE 1/NSRE 2 element in the ASNS gene can mediate the induction of ASNS expression by both AAR and UPR pathways, the AARE element of SNAT2, which is only 2 nucleotides differe nt from NSRE 1, does not carry this dual function, and is specific for the induction by the AAR pathway (Gjymishka et al., 2008)

PAGE 31

16 Vascular endothelial growth factor (VEGF) has potent angiogenic properties and contributes to angiogenesis in both normal and tumor tissues. Abcouwer et al. (Abcouwer et al., 2002) showed that VEGF mRNA expression was increased by amino acid deprivation or glucose deprivation and other ER stress inducers, in a retinal pigmented epithelial cell line. Fo ur AARE like elements were identified in the VEGF gene, but no ERSE element was found. Consistent with the property of AARE sequences, the induction of VEGF expression by amino acid deprivation is dependent on ATF4 (Roybal et al., 2005) The cationic amino acid transporter 1 (CAT 1) is a transporter for the ess ential amino acids arginine and lysine. CAT 1 expression was first shown to be induced by various of stress conditions, including amino acid deprivation and ER stress, through coordinated stabilization of the mRNA and increased mRNA translation (Yaman et al., 2003; Fernandez et al., 2001) Later, it was shown that the CAT 1 gene transcription is also increased by cellular stress, and an AARE element was identified in the first exon of the CAT 1 gene (Fernandez et al., 2003) TRB3 is a human ortholog of drosophila tribble 3 protein and associates with CHOP to suppress the CHOP dependent trans activation (Ohoka et al., 2005) T RB3 is induced by amino acid deprivation (Jousse et al., 2007) as well as ER stress (Ohoka et al., 2005) There are three tand em AARE elements in the TRB3 promoter, and they were shown to be responsible for the induction of TRB3 by either the AAR or the UPR (Jousse et al., 2007; Ohoka et al., 2005) CHOP was demonstrated to be an activator of the TRB3 gene, an effect that was mediated by there AARE sequences (Ohoka et al., 2005) The Transcriptional Regulator Network in Amino Acid Response Many of the gen es induced by amino acid deprivation are transcription factors or transcriptional regulators. These transcriptional regulators form a complex network in the sense that not only do they cooperatively regulate many genes directly involved in metabolic contr ol

PAGE 32

17 such as ASNS and SNAT2, but the expression of the transcription factors themselves is also subject to regulation.. For example, ATF4 activates the CHOP gene, both ATF4 and CHOP bind to and activate the TRB3 gene (Oho ka et al., 2005) and the TRB3 protein inhibits ATF4 function on the CHOP promoter, and therefore represses CHOP expression in a negative feedback manner (Jousse et al., 2007) In the current study, I also demonstrated that CHOP, for which the expression is directly induced by ATF4 during amino acid limitation or ER stress (Averous et al., 2004) inhibits ATF4 function on the ASNS gene. More details are described in Chapter 3. Interestingly, many of the transcription factors induc ed by amino acid limitation belong to the bZIP transcription factor family, indicating the critical role of this super family of proteins in the regulation of stress responsive genes. Transcription factors of other families, such as the zinc finger transc ription factors Sp1 and Sp3, are also involved in the regulation of amino acid responsive genes (Leung Pineda and Kilberg, 2002) Continuin g to identify and characterize transcriptional regulators that are up regulated by amino acid limitation will lead to a more comprehensive understanding of the cellular response to nutrient deprivation. In the current study, I identified two genes, FOXA2 and FOXA3, which belong to the FOXA winged helix transcription factor family, to be induced by amino acid limitation. I characterized their regulation by nutrient availability, and also tested their functions in the regulation of other amino acid responsi ve genes. More details are described in Chapter 5.

PAGE 33

18 Figure 1 1. Model for the translational control of GCN4 and ATF4. The model shows the mechanism by which yeast GCN4 (Panel A) and mammlian ATF4 (Panel B) are translationally up regulated under amino acid starvation conditions. Details are described in Chapter 1. The figure in Panel A was taken from Hinnebusch JBC 1997, and the figure in Panel B was taken from Vattem and Wek, PNAS, 2004.

PAGE 34

19 Figure 1 2. Signaling pathways during the unfolded protei n response. The figure shows the UPR signaling cascades initiated by three proximal sensors IRE1, PERK and ATF6. Under non stressed conditions, BiP binds to the lumenal domains of IRE1, ATF6, and PERK. Details are described in Chapter 1. Upon accumulat ion of unfolded proteins in the ER lumen, IRE1, released from BIP, dimerizes to activate its kinase and RNase activities to initiate XBP1 mRNA splicing thereby creating a potent transcriptional activator. Primary targets that require IRE1/XBP1 pathway for induction include genes involved in ERAD. Similarly, ATF6 released from BiP transits to the Golgi compartment where cleavage by S1P and S2P proteases yields a cytosolic fragment that migrates to the nucleus to further activate transcription of UPR respon sive genes. Finally, PERK, released from BiP, dimerizes and then autophosphorylates and phosphorylates responsive genes and also s ome pro apoptotic genes, such as CHOP. The figure was taken from Malhotra and Kaufman, Cell and Dev. Biol., 2007.

PAGE 35

20 Figure 1 3. Characterization of the ASNS proximal promoter. There are five protein binding sites within the ASNS proximal promoter regio n that contribute to nutrient control of the human ASNS gene, three GC boxes (GC I, GC II, and GC III), NSRE 1, and NSRE 2 sites. NSRE 1 and NSRE 2 form a functional unit (NSRU) that is responsible for the induction of ASNS following either amino acid sta rvation or ER stress. The figure was taken from Zhong et al, Biochem J., 2003

PAGE 36

21 Figure 1 4. A model for transcriptional regulation of the ASNS gene by amino acid limitation. The figure demonstrates a two phased model of amino acid regulation of the AS NS gene through the NSRE 1 element. During Phase I (0 4 h after amino acid limitation), ATF4 binding to the ASNS promoter NSRE 1 site increases, therefore, results in a transcriptional activation of the ASNS gene. During Phase II (8 24 h), the repression 1 while ATF4 binding is diminished thereby results in a relatively low rate of transcription compared to Phase I. This figure was modified from Chen et al. JBC, 2004.

PAGE 37

22 CHAPTER 2 MATERIALS AND METHODS Cell culture The cell lines used in the studies described in this thesis were: human hepatoma HepG2 cells; human embryonic kidney HEK293T cells; three sublines from the acute lymphoblastic leukemia cell line MOLT 4: parental MOLT 4 cell line (MOLT 4P), ASNase r esistant cell line (MOLT 4R), parental MOLT 4 cell line with constitutive ASNS over expression (MOLT 4P/ASNS). A B cell derived ALL cell line with TEL AML1 fusion [TEL AML(+)] REH. A B cell derived ALL cell line without TEL AML1 fusion [TEL AML( )] NALM6 All cell types were maintained at 37C in a 5% CO 2 /95% air incubator. medium; pH 7.4) (Mediatech, Herndon, VA), supplemented with 1x non essential amino acids, 2 mM glut amine, 100 g/mL streptomycin sulfate, 100 units/mL penicillin G, 0.25 g/mL amphotericin B and 10% (v/v) FBS (fetal bovine serum). Human embryonic kidney (HEK) 293T cells were cultured in Dulbecco's modified Eagle's medium (DMEM) with glutamine (Mediatec h), supplemented with 1X non essential amino acids, 10% fetal bovine serum, 100 g/mL streptomycin sulfate, 100 units/mL penicillin G, and 0.25 g/mL amphotericin B. All of the ALL cell lines were cultured in RPMI 1640 (Cellgro) medium supplemented with 1 0% (v/v) fetal bovine serum (FBS) and 10 ml/L ABAM (100 U/mL penicillin, 100 g/mL streptomycin, 0.25 g/mL amphotericin B) (GIBCO, Gaithersburg, MD). The MOLT 4R cells were maintained in medium containing 1 U/mL of ELSPAR ASNase (MERCK). Treatments Cell cultures were replenished with fresh medium and serum for 12 h prior to initiating all amino acid

PAGE 38

23 deprivation, cells were treated with 2 mM HisOH (histidinol). HisOH blocks ch arging of histidine on to the corresponding tRNA and thus mimics histidine deprivation, thereby triggering activation of the AAR cascade (Thiaville et al., 2008) To induce e ndo plasmic reticulum (ER) stress, cells were treated with 300 nM of Thapsigargin (Tg). Tg activates the UPR pathway due to its inhibition of the ER resident Ca ++ ATPase (Rogers et al., 1995) For ASNase treatment of the ALL cells, ASNase was diluted to 0.5U/ l in sterile 50%(v/v) glycerol, stored in 1 ml aliquots at 20 C, and thawed only once. Twelve hours before all experiments, cells were collected by centrifugation for 5 min at 230 X g, rinsed once with phosphate buffered saline (PBS) (0.15 M sodium chloride, 10 mM sodium pho sphate, pH 7.4), and resuspended at a density of approximately 5 X 10 5 cells/mL in fresh medium. 0.5U/ l ASNase was added to the media to desired concentrations. Yeast two hybrid Screening BD Matchmaker Library Construction & Screening Kits were purcha sed from Clontech protocols, using total RNA from HepG2 cells treated with HisOH for 8h. cDNAs corresponding to full length or different domains of human ATF4 (Figure 3 1) were generated by PCR, using hATF4/pcDNA3.1 zeo+ plasmid as template, and subcloned into the pGBKT7 bait vector (Clontech). The procedures for culture media and plates for yeast growth and selection were as Briefly, AH109 yeast strain was propagated in YPDA medium (20 g/L Difco peptone, 10 g/L Yeast extract, 30 mg/L adenine hemisulfate) or on YPDA plates (YPDA medium containing 20 g/L Agar). Selections were carried out on SD/dropout (DO) plates (6.7 g/L Yeas t nitrogen base without amino acids, 20 g/L Agar, appropriate DO supplement lacking specific amino acids). All media were supplemented with 2% glucose. The

PAGE 39

24 selection conditions are as follow: for selection of successful transformation of pGBKT7 bait plas mid, SD Trp; for selection of successful transformation and recombination of pGADT7 Rec prey plasmid, SD Leu; for selection of the activation of the HIS3 reporter gene, SD His; for selection of the activation of the ADE2 reporter gene, SD Ade. Media combi ning several selection conditions will select for specific phenotypes of the transformed cells, as indicated in each figure legend. For the transformation protocol, the competent cells were made on the same day of experiments. A single colony (< 4 weeks old, 2 3 mm in diameter) was inoculated into 3 ml of YPDA medium in a sterile, 15 ml centrifuge tube, and incubated at 30C with shaking for ml flask containing 50 ml of YPDA, and incubate at 30C with sh aking at 230 250 rpm for 16 20 hr. When the OD600 reached 0.15 0.3, the cells were collected by centrifugation at 700 X g for 5 min at room temperature. Supernatant was discarded and the cell pellet was resuspended in 100 ml of YPDA, and then incubated a t 30C for another 3 5 hr until OD600 reached 0.4 0.5. The cells were washed with deionized H 2 O by centrifugation and resuspension, and then suspended in 3 ml of 1.1 X TE/LiAc Solution (0.1 M Tris HCl, 10 mM EDTA, pH 7.5, and 0.1M LiAc). The cells were a gain collected by centrifugation at high speed for 15 sec, and resuspended in 1.2 ml of 1.1 X TE/LiAc Solution. For the screening, the following components were added to a sterile, pre chilled, 15 ml 2.5 ml PEG/LiAc solution [0.1 M Tris HCl, 10 mM EDTA, pH 7.5, 0.1M LiAc, and 40% PEG (polyethylene glycol 3350)] was added and mi xed by gentle vortexing. The mixure was mixed by inverting the tube. The tube was placed in a 42C water bath for 20 min (cells were

PAGE 40

25 mixed every 10 min), then collec ted by centrifugation at 700 X g for 5 min and resuspended in 3 ml of YPDA medium. The cells were then incubated at 30C with shaking for 90 min. After the incubation, cells were collected by centrifugation at 700 X g for 5 min and resuspended in 30 ml o f NaCl solution (0.9%), and then plated on to appropriate selection plates, as indicated in each figure legend. Positive clones were sequenced and Blast searches were performed against the NRdb and ESTdb in the National Center for Biotechnology Informatio n. RNA Isolation and Real time Quantitative RT PCR Total RNA was isolated using the Qiagen RNeasy kit (Qiagen), including a DNase I treatment before the final elution to eliminate DNA contamination, according to the ll RNA sample s were diluted in RNase free H 2 O to a final To measure the relative amount of specific mRNA, quantitative real time RT PCR (qRT PCR) analysis was performed using a DNA Engine Opticon 3 system (MJ Research, Reno, NV) and detection with SYBR Green I. For quant ification of real time PCR data, a relative standard curve method was used as follows. In order to generate a standard curve, RNA samples from HisOH treated cells were pooled and diluted to a concentration of 100 half conce in duplicat e wells to ensure accuracy. The master mix used for all RT qPCR experiments primer (5 sense primer (5 2 The PCR reactions were incubated at 48C for 30

PAGE 41

26 min followed by 95C for 15 min to activate the Taq polymerase and amplification of 35 cycles of 95C for 15 s, and XC for 60 s. Table 2 2 illustrates all the primer sets used and the specific annealing temperature (XC) for each primer set. After PCR, melting curves were acquired by stepwise increase of the temperatur e from 55C to 95C to ensure that a single product was amplified in the reaction. PCR was done in duplicates with samples from at least three independent experiments, and the means the standard error of the means (S.E.M.) between conditions were compar ed by Students t test. To measure steady state mRNA level, primers were designed to amplify a 60 200 bp region inside one exon or spanning two adjacent exons. To measure the transcriptional activity, primers were designed to span an exon intron junction a nd assay for hnRNA amplification based on a method described by Lipson and Baserga (Lipson and Baserga, 1989) Table 2 2 illustrates all the primer sets used for the qRT PCR experiments and the specific annealing temperature (XC) for each primer set. Protein Isolation and Immunoblotting For whole cell protein extraction from adherent cells, the cells were washed with 1 X PBS and collected in sample dilution buffer (62.5 mM Tris HCl, pH 6.8, 2% SDS, 25% glycerol, 0.01% Bromophenol Blue, 5% 2 mercaptoethanol). For whole cell protein extraction from suspension cells, the cells were collected and washed with 1X PBS by centr ifugation at 500 X g, and then cell pellets were dissolved in sample dilution buffer (62.5 mM Tris HCl, pH 6.8, 2% SDS, 25% glycerol, 0.01% Bromophenol Blue, 5% 2 mercaptoethanol). Protein concentrations were determined using a modified Lowry method (Markwell et al., 1978) For immunoblotting analysis, total cell extracts, nuclear extracts or eluates from Co immunoprecipitation experiments were prepared. An aliquot equaling 30 g/lane (total cell extract and nuclear extract) o r 45 IP eluate) was separated on a 10 14.5%

PAGE 42

27 Tris HCl polyacrylamide gel (Bio Rad) and then electrotransferred to a Protran nitrocellulose membrane (Schleicher & Schuell). The m embranes were stained with Fast Green FCF (Sigma) to ensur e equal loading by incubating 5 min in Fast Green stain solution (0.1% Fast Green FCF, 50% methanol, 10% acetic acid) followed by 3 X rinses in de stain solution (50% methanol, 10% acetic acid). Stained membranes were incubated with 10% blocking solution (10% (w/v) Carnation nonfat dry milk, 30 m M Tris Base, pH 7.5, 0.1% (v/v) Tween 20, and 200 m M NaCl) for 1 h at room temperature with mixing. Immunoblotting was performed by diluting primary antibodies in 10% dry milk blocking solution to a final concentr ation of 0.2 0.4 g/mL, and incubated with the membranes for overnight at 4C with rotation The blots were washed 3 X 10 min in 5% blocking solution on a shaker and then incubated with peroxidase conjugated secondary antibody for 2 h at room temperature. For immunoblots probed with rabbit polyclonal antibodies, a goat anti rabbit secondary antibody (Kirkegaard & Perry Laboratories, Gaithersburg, MD) was used at a 1:10,000 dilution. For immunoblots probed with mouse monoclonal antibodies, a goal anti mou se secondary antibody (Bio Rad) was used at a 1:5,000 dilution. The blots were then washed for 2 X 10 min in 5% dry milk blocking solution and 2 X 10 min in freshly made TBS/Tween (30 m M Tris base, 0.1% Tween 20, and 200 m M NaCl, pH 7.5). The bound secon dary antibody was detected using an enhanced chemiluminescence kit (Amersham Biosciences) and exposing the blot to the BioMax MR film (Kodak, Rochester, NY). Table 2 1 illustrates all the antibodies used for immunoblotting analysis. Transient Transfection and Expression HepG2 cells (0.2 X 10 6 cells/well) were seeded on 24 well plates 18 24 h before transfection with Superfect reagent (Qiagen) at a ratio of 6 l of Superfect to 1 g of DNA. For amounts of the transcription factor expression plasmids in each experiment. For the information

PAGE 43

28 of luciferase reporter cons tructs, see Table 2 4. The total amount of transfected DNA was kept constant among experimental groups by the addition of empty pcDNA3.1 zeo+ plasmid. At 24 hours following transfection, the cells were transferred to fresh complete MEM for 12 h before pr eparing cellular extracts for analysis of luciferase activity. To prepare the extract, cells were washed with phosphate supplied by Promega. The lysates were collected and stored at 8 0C until use. Ten microliters of cell extract was used for Firefly luciferase assays using the Luciferase Reporter Assay System (Promega), and the luminescence was measured with a luminometer. The luciferase values are normalized to microgram protein co ntent for each sample. For each experimental condition, three assays were performed for each transfection and at least two independent transfections were done. Chromatin Immunoprecipitation (ChIP) Analysis ChIP analysis was performed according to a modi fied protocol of Upstate Biotechnology, Inc. (Charlottesville, VA) and described by Chen et al. (2004). Briefly, HepG2 cells were seeded at 1.5 X 10 7 /150 mm dish in complete MEM with dishes enough for 3 IP conditions per plate. MOLT 4 cells were seeded a t 2.5 X 10 7 /50mL in complete RPMI 1640, in T 175 flask, with flasks enough for 5 IP conditions per flask. HepG2 c ells were transferred to fresh media 12 h before transfer to either complete MEM or MEM containing 2mM HisOH or MEM containing 300nM Tg for t he time period indicated in each expeiment. MOLT 4 cells were transferred to fresh media 12 h before transfer to either complete RPMI 1640 or RPMI 1640 containing 1U/mL ASNase. Protein DNA was cross linked by adding a 37% formaldehyde stock solution dire ctly to the culture medium to a final concentration of 1% and then stopped 10 min later by adding 2 M glycine to a final concentration of 0.125 M Cross linked chromatin was solubilized by sonication using a Sonic Dismembrator (Model 60, Fisher Scientific Co.) for five bursts of 40

PAGE 44

29 s at power 10 with 2 min cooling on ice between each burst. Extract from 1 X 10 7 HepG2 or MOLT 4 cells was incubated with 2 1 lists all the antibodies used in the ChIP anaysis. The antibody bound com plex was precipitated by Rec protein G Sepharose 4B beads (Invitrogen). Beads were incubated in a blocking solution (3% bovine serum albumin, 0.05% sodium azide, and protease inhibitor in TE pH 8.0) as a 50% slurry overnight at 4 o C. After incubation, 60 antibody aliquot and incubated at 4 o C with rotation for 2 h. Antibody bead complexes were pelleted by centrifugation at 500 g X 5 min, and washed by suspension and centrifugation with a series of wash buff ers, 1 mL of each washing buffer was incubated with the beads for a 5 min at 4 o C with rotation. The beads were washed with the following buffers, as used in order: Low salt buffer (0.1% SDS, 1% Triton X 100, 2 mM EDTA, 150 mM NaCl, 20 mM Tris HCl pH 8.0); High salt buffer (0.1% SDS, 1% Triton X 100, 2 mM EDTA, 500 mM NaCl, 20 mM Tris HCl pH 8.0); LiCl buffer (0.25 M LiCl, 1% NP 40, 1% sodium deoxycholate, 1 mM EDTA, 10 mM Tris HCl pH 8.0); and TE buffer (1 mM EDTA, 10 mM Tris HCl pH 8.0). After the final wash, antibody bead complexes were resuspended in 65 o C elution buffer (1% SDS, 0.1 M NaHCO 3 ), and incubated at 37 o C for 30 min with vigorous shaking. The DNA fragments in the immunoprecipitated complex were released by reversing the cross linking at 65C for 5 h and purified using a QIAquick PCR purification kit (Qiagen). Purified, immunoprecipitated DNA was analyzed by quantitative real time PCR (qPCR) with primers amplifying specific genomic regions. Table 2 3 lists all the ChIP primers with their ampl ified regions and the annealing temperature for each primer set. qPCR analysis was performed using the DNA Engine Opticon 3 system and detected with SYBR Green I. Serial dilutions of one fourth concentration were performed for determining the relative am ount of product, starting with the 1/4 input sample to give five tubes

PAGE 45

30 with final concentrations of 1/4, 1/16, 1/64, 1/256, and 1/1024 of the input sample. Duplicates for both the standards and the samples were simultaneously amplified using the same reac tion master mixture. The reactions were incubated at 95C for 15 min to activate the polymerase, followed by amplification at 95C for 15 s and X C for 60 s for 35 cycles. Table 2 3 illustrates all the primer sets used and the specific annealing tempera ture (XC) for each primer set. After PCR, melting curves were acquired by stepwise increases in the temperature from 55 to 95C to ensure that a single product was amplified in the reaction. The results are expressed as the ratio to input DNA. Samples from at least three independent immunoprecipitations were analyzed, and the means S.E.M between conditions were compared by the Students t test. Double Chromatin Immunoprecipitation (Double ChIP) Analysis HepG2 cells were seeded at 1.5 X 10 7 /150 mm dish with complete MEM and cultured for 24 h. Cells were transferred to fresh MEM 12 h before transfer to complete MEM, MEM containing 2 mM HisOH or MEM containing 300 nM Tg for 4 h. ChIP analysis was performed as described above, using CHOP antibody (sc 575, Santa Cruz). Immunoprecipitated chromatin Triton X 100, 1.2 mM EDTA, 16.7 mM Tris HCl, pH 8.0, 167 mM NaCl) containing 10mM DTT. The eluate was diluted 100 times w ith ChIP dilution buffer and divided evenly into two fractions. Each fraction was subjected again to the ChIP procedure using ATF4 antibody (Cocalico) or normal rabbit IgG antibody (sc 2027, Santa Cruz), respectively. Purified, immunoprecipitated DNA was analyzed by (qPCR) as described in the ChIP procedure. The results are expressed as the ratio to input DNA. Samples from three independent immunoprecipitations were analyzed in duplicate PCR assays, and the means S.E.M. between conditions were compare d by the Students t test. Table 2 3 lists all the ChIP primers with their amplified regions and the annealing temperature for each primer set.

PAGE 46

31 Short Interfering RNA (siRNA) Transfection The human CHOP (DDIT3) siRNA (catalog number L 004819 00 ), ATF4 siRNA ( catalog number L 005125 00), ATF5 siRNA ( catalog number L 008822 00), siControl nontargeting siRNA (catalog number D 00 1210 006423 01), FOXA2 siRNA (L 010089 00), FOXA3 siRNA (L 008863 00), and DharmaFECT 4 transfection reagent were purchased from Dharmacon, Inc. (Lafayette, CO). HepG2 cells were seeded in 12 well plates at a density of 2.5 X 10 5 cel ls/well in MEM and cultured for 16 h. Transfection was performed according to the instructions of Dharmacon using 3 l of DharmaFECT 4 and a 100 n M per well final siRNA concentration. HepG2 cells were treated with siRNA and transfection reagent for 24 h and then rinsed with PBS, given fresh MEM, and cultured for another 12 h. The medium was then removed and replaced with control MEM, MEM containing 2mM HisOH, or MEM containing 300nM Tg. Total RNA and protein extracts were isolated at specific times and analyzed by reverse transcription PCR or immunoblotting, respectively. Calcium Phosphate (CaP) Transfection HEK293T cells were plated at 30% confluency in DMEM medium 24h before transfection. Cells were replenished with 3 ml/60mm dish or 18 ml/150mm dish of fresh DMEM medium. CHOP pcDNA3.1 zeo + HA ATF4 pCGN or GFP pcDNA3.1 zeo + plasmids were transfected at 5 10 minutes prior to transfection, cells were replenished with fresh media w ith 18 ml/150 mm dish or 3 ml/60 mm dish. The following reagents were added in order to a 50 ml polypropylene tube and mixed: plasmid DNA, 2620 l/150 mm dish or 438 l/60 mm dish of ddH 2 O, 366 l/150 mm dish or 61 l/60 mm dish 2 M CaC1 2 Then 3 ml/150 mm dish or 500 l/60 mm dish 2 X HBS (50 mM HEPES, 270 mM NaCl, 1.5 mM Na 2 HPO4, pH7.1 ) was added drop wise to the DNA/CaC1 2 mixture and mixed by bubbling vigorously with an automatic pipette for 30 40 sec. The pH

PAGE 47

32 value of the HBS buffer is critical for ac hieving maximal transfection efficiency. A series of HBS solutions with pH values of 7.0, 7.1, and 7.2 were tested in the CaP transfection experiment, and the one that gave the best transfection efficiency (pH7.1) was used in the subsequent experiments. The HBS/DNA/CaC1 2 mixture was incubated at RT for 3 min, and then added to each plate at 6 ml/150 mm dish or 1 ml/60 mm dish. The cells were incubated in 37C incubator for 15 hr, washed twice with PBS, replenished with complete DMEM, and incubated for an other 12 h. Then the cells were moved to complete DMEM, DMEM containing 2 mM HisOH or DMEM containing 300 nM Tg for appropriate time. Transfection efficiency was evaluated by estimating the ratio of total cells and GFP fluorescent cells. Co Immunoprecipi tation HEK 293T cells were cultured in 150mm dishes and transfected with CHOP pcDNA3.1 zeo + HA ATF4 pCGN or GFP pcDNA3.1 zeo + plasmid either alone or in combination, as indicated in each experiment, using CaP transfection protocol described above. Four p lates were used for each condition. At 36 h after transfection, cells were washed twice with PBS, harvested by scraping and centrifuged for 5 min at 500 X g. Cell pellets were resuspended in lysis buffer (20 mM HEPES, pH7.6,10 mM NaCl, 1.5 mM MgCl 2 0.2 mM EDTA, 20% glycerol, 0.5% Triton X 100) and incubated for 30 min at 4 C with shaking. Lysates were cleared by centrifugation for 15 min at 14,000 X g. For the immunoprecitation of HA ATF4, cell lysates were incubated with HA affinity agarose beads (Sig ma, catalog number A2095) for 16 h at 4 C, with shaking. The beads were precipitated by centrifugation for 5 min at 500 X g, washed with lysis buffer for 5 X 10 min. The immunoprecipitated proteins were eluted with 100 l sample dilution buffer (2% SDS, 1 25mM Tris HCl pH 6.8, 20% glycerol and 30 g/mL bromophenol blue) at 65 C. For the immunoprecipitation of CHOP, cell lysates were incubated with anti CHOP monoclonal antibody (Santa Cruz, catalog number sc 7351) and Rec protein G Sepharose

PAGE 48

33 4B beads (Invitr ogen) for 16 h at 4 C, with shaking. The beads were precipitated by centrifugation for 5 min at 500 X g and washed with lysis buffer 5 X 10 min. The immunoprecipitated proteins were eluted with 100 l sample dilution buffer at 65 C. After 5% beta mercapt oethanol was added to the eluates, the samples were subjected to Western analysis l/lane. All antibodies used are listed in Table 2 1. Trypan Blue Exclusion and WST 1 Cell Proliferation Assays Cell number and viability was established for the leukem ia cell cultures by staining with 0.2% trypan blue solution (Sigma) and then counting replicate samples on a hemocytometer. The cell viability was calculated as the percentage of trypan blue stained cells. The proliferation of the ALL cells was determine d by a 2 (4 Iodophenyl) 3 (4 nitrophenyl) 5 (2,4 disulfophenyl) 2H tetrazolium (WST 1) assay (Boehringer Manheim, Indianapolis, IN). Cells were collected by centrifugation, washed once with PBS, and then resuspended at a concentration of 5 X 10 4 cells/mL in RPMI 1640 medium. An aliquote of 100 l/well was seeded in 96 well plates and then subjected to an ASNase challenge, as described in each figure legend. After incubation for 72 h at 37 C, 10 l/well of WST 1 reagent was added and the cells incubated f or 1 h. The optical density of the solution, which is linearly related to the viable cell number, was read on a plate reader (SLT Lab Instruments) at 450 nm with a reference wavelength of 690 nm. Proliferation curves were plotted as the absolute optical density readings versus the ASNase concentration tested. To calculate IC 50 after ASNase treatment, ALL Cells were treated with ASNase (0, 0.01, 0.02, 0.05, 0.1, 0.2, 0.5, 1, 2 and 5 U/mL) for 72 h. Percent viability was determined using a Trypan Blue ass ay and the data plotted versus ASNase concentration. A single, 3 parameter exponential decay curve was fit to the data points and used to calculate an IC 50 (SigmaPlot Systat Software, Inc.).

PAGE 49

34 mRNA Stability Assay HepG2 cells were plated at 40% confluency in 60 mm dishes. Cells were cultured in MEM containing 2mM HisOH, both containing 100 ng/mL Actinomycin D (ActD), and cultured for 12h. Total RNA were collec ted at 0h, 2h, 4h, 6h, 9h, and 12h, and subjected to quantitative RT PCR analysis to measure the steady state mRNA content of FOXA2, FOXA3, p21, or GAPDH. The values of the ratio of a target gene relative to GAPDH were plotted as the semi logarithm agains t the time.

PAGE 50

35 Table 2 1. Antibodies used for ChIP and/or Immunoblotting Antibody Catalogue Number Company Animal in which antibody was produced Actin A2066 Sigma Aldrich rabbit Acetylated H3 (specific for acetylated Lys 9 and Lys 14) 06 599 Millipore (formerly Upstate) rabbit Acetylated H4 (recognizes acetylated H4 at Lys 5, 8, 12, and 16) 06 866 Millipore (formerly Upstate) rabbit ASNS University of Florida Hybridoma Core mouse ATF3 sc 188 Santa Cruz Biotechnology, Inc. rabbit ATF4 sc 200 San ta Cruz Biotechnology, Inc. rabbit ATF4 (2) CoCalico Biologicals, Inc. rabbit sc 150 Santa Cruz Biotechnology, Inc. rabbit CHOP sc 575 Santa Cruz Biotechnology, Inc. rabbit CHOP sc 7351 Santa Cruz Biotechnology, Inc. mouse FOXA1 sc 22841 Santa Cruz Biotechnology, Inc. rabbit FOXA2 sc 20692 Santa Cruz Biotechnology, Inc. rabbit FOXA3 sc 25357 Santa Cruz Biotechnology, Inc. rabbit Normal rabbit IgG (non specific IgG) sc 2027 Santa Cruz Biotechnology, Inc. rabbit RNA Pol II sc 899 Santa Cruz B iotechnology, Inc. rabbit TAF1/TAFII250 (TFIID) sc 17134 Santa Cruz Biotechnology, Inc. goat TBP (TFIID) sc 204 Santa Cruz Biotechnology, Inc. rabbit TFIIB sc 274 Santa Cruz Biotechnology, Inc. rabbit v5 R96025 Invitrogen Inc. mouse

PAGE 51

36 Table 2 2. Prim er sets and annealing temperatures for RT qPCR Primer Pair Gene Region Primer Sequences Annealing Temperature ( o C) Human ASNS mRNA, human +13723 to +13784 FP 5 GCAGCTGAAAGAAGCCCAAGT 3 RP 5 TGTCTTCCATGCCAATTGCA 3 60 ASNS transcription activity, h uman +19577 to +19653 (intron 12 to exon 13) FP 5 CCTGCCATTTTAAGCCATTTTGC 3 RP 5 TGGGCTGCATTTGCCATCATT 3 58 ATF4 mRNA, human +1154 to +1221 FP 5 TGAAGGAGTTCGACTTGGATGCC 3 RP 5 CAGAAGGTCATCTGGCATGGTTTC 3 60 ATF5 mRNA, human +876 to +967 FP 5 CCTC CTCCTTCTCCACCTCAA 3 RP 5 TGGTCTCTCTTCTTTTGCTTGC 3 60 Cat 1 mRNA, human +470 to +601 TGCCATTGTCATCTCCTTCCTG AGCTCTCCAACGGTGACATAGC 60 mRNA, human +967 to +1054 FP 5 CAGGTCAAGAGCAAGGCCAAGA 3 RP 5 TGCGCACGGCGATGTTGT 3 60 CHO P mRNA, human +128 to +191 CATCACCACACCTGAAAGCA TCAGCTGCCATCTCTGCA 60 FOXA1 mRNA, human +1793 to +1996 CAGCAAACAAAACCACACAAACC ACACTTGTGGATCATTAAACTTCGC 60 FOXA2 mRNA, human +1892 to +1988 FP 5 GTTGTTGTTGTTCTCCTCCATTG C 3 RP 5 AACTACATGGTTTTACACCGAGTCAC 3 60 FOXA3 mRNA, human +1361 to +1584 FP 5 GGTCTATTACTTACTGTGATGACTGCTG 3 RP 5 CCAAAGAAGATGTCACTGAAATGCT 3 60 GAPDH mRNA, human +1382 to +1442 FP 5 TTGGTATCGTGGAAGGACTC 3 RP 5 ACAGTCTTCTGGGTGGCAGT 3 60 SNA T2 mRNA, human +8243 to +8319 FP 5 GTGTCCTGTGGAAGCTGCTTTGA 3 RP 5 CAGGTACAAGAGCTGTTGGCTGTGT 3 60 SNAT2 transcription activity, human +2205 to +2299 (exon 4 to intron 4) FP 5 GCAGTGGAATCCTTGGGCTTTC 3 RP 5 CCCTGCATGGCAGACTCACTACTTA 3 60 TRB3 mRNA, human +205 to +343 ATTAGCTCCGGTTTGCATCAC 3 60 VEGF mRNA, human +4 to +91 FP 5 AGCTCCAGAGAGAAGTCGAGGAAGA 3 RP 5 TCACTTTGCCCCTGTCGCTTT 3 60

PAGE 52

37 Table 2 2. Continued Primer Pair Gene Region Primer Sequences Annealing T emperature ( o C) mouse ASNS mRNA, mouse +10892 to +11197 FP 5 CATGCCATCTATGACAGCGTGGA 3 RP 5 CGCAGATTGTTCTTCACGGTCTCT 3 60 SNAT2 mRNA, mouse +4225 to +4430 FP 5 CTCCTCCTCAAGACTGCCAACGA 3 RP 5 TTCCAGCCAGACCATACGCCTTA 3 60 GAPDH mRNA, mouse + 738 to +879 FP 5 GCCTTCCGTGTTCCTACCC 3 RP 5 CCTCAGTGTAGCCCAAGATGC 3 60 FOXA1 mRNA, +379 to +539 FP 5 GCTCCAGGATGTTAGGGACTGT 3 RP 5 GTTCATGGTCATGTAGGTGTTCAT 3 60 FOXA2 mRNA, +117 to +202 FP 5 GGGAGCCGTGAAGATGGA 3 RP 5 TCATGTTGCTCACGGAAGAGTA 3 60 FOXA3 mRNA, +780 to +1180 FP 5 AAGGCAAAGAAAGGAAACAGCGCCA 3 RP 5 GATGCATTAAGCAGAGAGCGGGA 3 60

PAGE 53

38 Table 2 3. Primer sets and annealing temperatures for CHIP qPCR Primer Pair Gene Region (bp) Primer Sequences Annealing Temperature ( o C) Human ASNS coding region, human +13723 to +13784 FP 5 GCAGCTGAAAGAAGCCCAAGT 3 RP 5 TGTCTTCCATGCCAATTGCA 3 60 ASNS promoter/AARE, human 87 to 22 FP 5 TGGTTGGTCCTCGCAGGCAT 3 RP 5 CGCTTATACCGACCTGGCTCCT 3 61.4 CAT 1 AARE, human +470 to 601 TGCCATTGTCATCTCCTTCCTG AGCTCTCCAACGGTGACATAGC 60 FOXA3 AARE 1, human 7890 to 7780 CCTGAAAAGAGGTCCCTGACA RP 5 TGTCTTGCCTCAGGTCATTCTG 3 60 FOXA3 AARE 2, human 2440 to 2318 CAGATCACTTGAGGTCAGGAGTTC RP 5 TCGAGTAGCTGGGAT TACAAACG 3 60 FOXA3 AARE 3, human 294 to 385 GACTACAAGGACCCGTAAAAGG RP 5 AAGTTAAGCATTTCTAGCCGTC 3 60 FOXA3 AARE 4, human +3834 to +3935 CAACGGCAAAAATAGCCAACAG GAGTTCCAGTCGAGAAATCCTCTG 60 FOXA3 promoter, human +27 to +149 GTGTCCCGGCTATAAAGCGTG RP 5 GAGCGCTCTGGATCTCTCAGC 3 60 TRB3 AARE, human +205 to +343 ATTAGCTCCGGTTTGCATCAC RP 5 TCCACTTCCGCTGCGAGTCTC 3 60 SNAT2 AARE, human +672 to +745 FP 5 GGGAAGACGAGTTGGGAACATTTG 3 RP 5 CCCTCCTATGTCCGGAAAGAAA AC 3 60 VEGF AARE, human +1626 to +1705 FP 5 CTCTGCCCAGTGCTAGGAGGAATT 3 RP 5 CAGGGGCTTCTCTCCAGGCTAAA 3 60

PAGE 54

39 Table 2 4. Plasmids used for protein expression, luciferase reporter assay, and yeast two hybrid screening. Construct Name Parental Vector Insert and Function Species Specificity Over expression plasmid hATF4/pcDNA3.1 zeo+ pcDNA3.1 zeo+ Expression of full length human ATF4 Mammal hCHOP/pcDNA3.1 zeo+ pcDNA3.1 zeo+ Expression of full length human CHOP Mammal CHOP FL V5 /pEF6 V5 His, pE F6 V5 His Expression of full length (aa 1~169) of human CHOP Mammal CHOP N Term V5 /pEF6 V5 His, pEF6 V5 His Expression of N terminal portion (aa 1~100) of human CHOP Mammal CHOP C Term V5 /pEF6 V5 His: pEF6 V5 His Expression of C terminal portion (aa 10 1~169) of human CHOP Mammal HA hATF4/pCGN pCGN Expression of N teminally HA isotope tagged full length human ATF4 Mammal hFOXA3/pCMV pCMV Expression of full length human FOXA3 Mammal Luciferase reporter plasmid ASNS 173/+51 /luc pGL3 ASNS promoter re gion (nt 173/+51) upstream of the Firefly luciferase reporter Mammal FOXA3 1000/+151/luc pGL3 FOXA3 promoter region (nt 1000/+151) upstream of the Firefly luciferase reporter Mammal FOXA3 510/+151/luc pGL3 FOXA3 promoter region (nt 510/+151) upstrea m of the Firefly luciferase reporter Mammal FOXA3 304/+151/luc pGL3 FOXA3 promoter region (nt 304/+151) upstream of the Firefly luciferase reporter Mammal

PAGE 55

40 Table 2 4. Continued Yeast two hybrid plasmid hATF4 FL/pGBKT7 pGBKT7 Expression of full le ngth human ATF4 as bait in yeast Yeast hATF4 AD (1 270) /pGBKT7 pGBKT7 Expression of human ATF4 activation domain (aa 1 270) as bait in yeast Yeast hATF4 AD (1 130) /pGBKT7 pGBKT7 Expression of human ATF4 activation domain (aa 1 130) as bait in yeast Y east hATF4 AD (130 270) /pGBKT7 pGBKT7 Expression of human ATF4 activation domain (aa 130 270) as bait in yeast Yeast hATF4 AD (1 45) /pGBKT7 pGBKT7 Expression of human ATF4 activation domain (aa 1 45) as bait in yeast Yeast hATF4 AD (45 90) /pGBKT7 pGBKT7 Expression of human ATF4 activation domain (aa 45 90) as bait in yeast Yeast hATF4 AD (90 135) /pGBKT7 pGBKT7 Expression of human ATF4 activation domain (aa 90 135) as bait in yeast Yeast hATF4 bZIP (270 351) /pGBKT7 pGBKT7 Expression of human A TF4 basic leucine zipper domain (aa 270 351) as bait in yeast Yeast

PAGE 56

41 CHAPTER 3 IDENTIFICATION AND CHARACTERIZATION OF ATF4 INTERACTING PROTEINS AND THEIR FUNCTION IN THE REGULATION OF THE ASPARAGINE SYNTHETASE (ASNS) GENE DURING NUTRIENT DEPRIVATION Intr oduction Involvement of ATF4 in the Nutrient Stress Responses As described in Chapter 1, both the unfolded protein response (UPR) pathway and the amino acid response (AAR) pathway lead to phosphorylation of the eukaryotic initiation factor 2 alpha (eIF2 ) on serine 51. The corresponding kinases are GCN2 in the amino acid response (Sood et al., 2000; Zhang et al., 2002; Hinnebusch, 2005) and PERK in the unfolded protein response (Okada et al., 2002; Harding et al., 2000b; Liang et al., 2006) The phosphorylation of eIF2 provokes a suppression of global protein synthesis, but a paradoxical increase in the translation of selected mRNAs through a mechanism involvi ng short upstream opening reading frames (ORFs) that overlap with the actual coding ORF. Among the proteins for which translation is increased is activating transcription factor 4 (ATF4) (Vattem and Wek, 2004; Lu et al., 2004) ATF4 is a member of the ATF subfamily of the basic leucine zipper (bZIP) factor super family (Ameri and Harris, 2008) With regard to function, it is thought to be the mammalian counterpart to the yeast transcription factor GCN4 (Hinnebusch, 1997) although these two proteins have no sequence homology. ATF4 protein has been demonstrated to form a homo dimer with itself or hetero dimers with a variety of bZIP proteins through its leucine zipper mot if. These bZIP proteins include Fos (Hsu et al., 1994) Jun (Hai and Curran, 1991; Chevray and Nathans, 1992) JunD (Shimizu et al., 1998) and several C/EBP (CCAAT/enhancer E (Nishizawa and Nagata, 1992; Vallejo et al., 1993) Therefore, ATF4 has been demonstrated to form hetero dimers with a wide sp ectrum of members of the AP 1 and

PAGE 57

42 C/EBP subfamily of bZIP proteins, rather than within its own ATF/CREB subfamily. The C/EBP family consists (Ramji and Foka, 2002) A C/EBP ATF4 complex has been detected at cAMP response elements (Vallejo et al., 1993; Liang and Hai, 1997) as well as at C/EBP ATF composite sites (Fawcett et al., 1999; Chen et al., 2004; Lopez et al., 2007) These pathways initiate a wide array of adaptive mechanisms, and if necessary, programmed cell d eath (Harding et al., 2003) In a generalized stress response, the targets of ATF4 include many of those genes involved in amino acid metabolism and tr ansport, as well as in redox chemistry (Harding et al., 2000a) ATF4 is also demonstrated to be important for a few b asic aspects of cell proliferation and differentiation, because ATF4 knock out mice display abnormal lens formation (Tanaka et al., 1998) and defects in cell proliferation in fetal liver, e mbryonic lens and hair follicles, as well as an overall reduction in size of the animals (Masuoka and Townes, 2002) In addition, ATF4 is a critical regulator of osteoblast differentiation and function (Yang et al., 2004) and bone resorption (Elefteriou et al., 2006) and it is involved in long term memory induction (Chen et al., 2003) Among the A TF4 target genes are asparagine synthetase (ASNS) and CCAAT/enhancer binding protein (C/EBP) homology protein (CHOP), also known as growth arrest and DNA damage protein 153 (GADD153). Both ASNS and CHOP are transcriptionally activated by ATF4 (Siu et al., 2002; Averous et al., 2004) The ASNS gene has been studied in the Kilberg laboratory as a model for stress induced genes, and its regulation by amino acid limitation can be attributed to two cis acting elements in its proximal promoter, the Nutrient Sensing Response Elements 1 and 2 (NSRE 1, NSRE 2). These two elements together function as an enhancer element, nutrient sensing response unit (NSRU) and mediate the transcriptional activation of the gene by either the AAR or the UPR pathway (Barbosa Tessmann et al., 2000)

PAGE 58

43 The NSRE 1 sequence is a C/EBP ATF composite site that has been shown by both EMSA and ChIP assays to bind ATF 4 following increased de novo synthesis of ATF4 following activation of either the AAR or the UPR (Siu et al., 2002; Chen et al., 2004) Characteristics of CHOP as a Transcription Factor CHOP was originally isolated as a gene induced in response to DNA damaging agents (Cenatiempo, 1986) and subsequently demonstrated to be induced by amino acid deprivation (Bruhat et al., 1997) and ER stress (Hershey, 1991) The prototypic C/EBP protein contains a transcriptional activation/repression domain at its N terminus and a basic leucine zipper (bZIP) region at its C terminus, which consists of a basic DNA binding domain a nd a leucine zipper motif for dimerization. As a member of the C/EBP family, CHOP protein is also composed of two known functional domains, an N terminal transcriptional activation domain and a C terminal basic leucine zipper (bZIP) domain consisting of a basic amino acid rich DNA binding region followed by a leucine zipper dimerization motif. Furthermore, CHOP protein contains two adjacent serine residues that can serve as substrates of the p38 MAP kinase family (Wang and Ron, 1996) The basic region of CHOP contains proline and glycine substitutions in conserved residues believed to be essential to the interaction of these proteins with most C/EBP DNA binding sites (Ron and Habener, 1992; Wertheimer et al., 1991; Inesi et al., 1998) Consequently, it was originally proposed that CHOP lacked DNA binding activity for prototypi c C/EBP binding sites and instead, negatively regulated the activity of C/EBP proteins by sequestering them in non functional heterodimers, thereby inhibiting their ability to bind DNA (Ron and Habener, 1992) Indeed, CHOP C/EBP heterodimers fail to bind several known (A/G)TTGCG(C/T)AA(C/T) in vitro and when expressed in cells, CHOP attenuates the transcription activity of other C/EBP proteins to activate promoters sites (Ron and Habener, 1992) Howeve r, subsequent studies revealed that CHOP C/EBP heterodimers can

PAGE 59

44 (A/G) (A/ G) (A/G)TGCAAT(A/C)CCC and a CHOP C/EBP heterodimer could function as a positive trans activator (Wang et al., 1998b) More recently, TRB3 was also identified as a CHOP inducible gene, during ER stress (Ohoka et al., 2005) Furthermore, CHOP can enhance the transcriptional activation of AP 1 by tethering to the AP 1 complex without direct binding of DNA (Ubeda et al., 1999). Thus, CHOP has a dual role both as an inhibitor of C/EBP function on some genes and as an activator of other genes. Function of CHOP i n Apoptosis At least three apoptosis pathways are known to be involved in the apoptosis induced by ER stress. The first is through the activation of the cJUN NH2 terminal kinase (JNK) pathway, re1) TNF receptor associated factor 2 (TRAF2) apoptosis signal (Urano et al., 2000; Nishitoh et al., 2002) The second is through the activation of ER associated caspase 12 (Yuan and Yankner, 2000) And the third is through the transcriptional activation of C HOP (Oyadomari and Mori, 2004) Over expression of CHOP and microinjection of CHOP protein have been reported to lead to cell cycle arrest and/or apoptosis (Matsumot o et al., 1996; Wang and Ron, 1996; Maytin et al., 2001; Oyadomari and Mori, 2004) Attenuation of CHOP expression by BiP over expression (Wang and Ron, 1996) and CHOP knock down in transgenic mice (O yadomari et al., 2001; Zinszner et al., 1998) both let to reduced apoptosis during ER stress. The pro apoptotic activity which is a CHOP dimerization partner, are also resistant to ER stress induced apoptosi s (Zinszner et al., 1998)

PAGE 60

45 On the other hand, CHOP may be an upstream regulator of the Bcl 2 family of proteins in an apoptosis eve nt. This is supported by the fact that over expression of CHOP leads to decrease in Bcl 2 protein and over expression of Bcl 2 blocks CHOP induced apoptosis (Matsumoto et al., 1996; McCullough et al., 2001) Furthermore, over expression of CHOP leads to translocation of Bax protein from the cytosol to the mitochondria (McCullough et al., 2001) consequently transmits the apoptotic signals to the mitochondria, which leads to cytochrome c release and subsequent apoptosis. Cells from Bax / and Bak / mice are resistant to apoptosis induced by ER s tress, therefore Bax and Bak may function as executioners in ER stress mediated apoptosis (Wei et al., 2001) However, it is not clear whether CHOP dir ectly or indirectly affects the Bcl 2 family of proteins. Involvement of CHOP in the Nutrient Stress Responses The basal expression level of CHOP is almost undetectable in most cell types, but its expression is rapidly induced through ATF4 dependent transc ription (Ron and Habener, 1992; Averous et al., 2004) Therefore, CHOP is one of the most important components in the network of stress inducible transcription factors and the investigation of CHOP actions is critical for understanding the molecular mechanisms for the cellular response to stress. Many studies have been done to investigate the CHOP interaction with the C/EBP proteins. However, the number of investigations focused on interaction of CHOP with ATF transc ription factors and the physiological role of those interactions is limited. Chen, et al. (Chen et al., 1996) demonstrated that CHOP inhibits ATF3 function by forming a non function al heterodimer, similar to the dominant negative effect on C/EBP proteins. Ohoka, et al. (Ohoka et al., 2005) showed that CHOP and ATF4 cooperatively activate TRB3 expression during ER stress, but they did not investig ate whether or not this action is due to direct action of an ATF4 and CHOP homodimer, or is a secondary effect. In the contrast to its action on the TRB3 gene, CHOP antagonizes the

PAGE 61

46 action of ATF4 and ATF5 at C/EBP ATF composite sites, but not the action o f CREB at cyclic AMP response elements (Yoon et al., 2004) Furthermore, Gachon et al. (Gachon et al., 2001) showed that CHOP binds to ATF4 in vitro based on glutathione S transferase pull down assays, and CHOP blocks the action of a fusion protein composed of the ATF4 bZIP domain and the GAL4 binding domain, in a GAL4 reporter assay. In the current study, in an effort to identify binding partners for ATF4 CHOP was identified as an ATF4 interacting protein. Their interaction was confirmed in vivo in mammalian cells. The results from this study demonstrate the involvement of CHOP in the ATF4 dependent regulation of the ASNS gene, and provide an initial an alysis of the mechanism for this CHOP inhibitory activity. The role of CHOP in nutrient stress induced apoptosis is also discussed. Results Identification of ATF4 Interacting Proteins by Yeast Two Hybrid Screening Given the pivotal role of ATF4 in nutrien t stress response pathways, I decided to utilize an unbiased approach to identify ATF4 interacting proteins. The most classical approach, yeast two hybrid screening, was performed in the current study. I started by cloning the complete ATF4 cDNA into the bait vector pGBKT7, and used the full length ATF4 protein as bait (Figure 3 1). However the full length ATF4 showed strong activation of the selection marker genes ( HIS3 and ADE2 ) without prey, as evidenced by growth on the most stringent selection medi um ( Trp/ His/ Ade) (Figure 3 2). A protein that activates the selection markers cannot be used as bait because it will give false positive results. Therefore, a deletion analysis was performed and different ATF4 domains were cloned into the bait vector, as shown in Figure 3 1, and tested for self activation activity. All the ATF4 constructs containing the transcriptional activation domain (aa 1 130), including aa 1 130, aa 1 200, and aa 1 270 showed strong self activation activities, as shown in Figure 3 3. Surprisingly, the linker domain alone (aa 130 270) also showed self

PAGE 62

47 activation activity (Figure 3 3). The ATF4 activation domain was further split into three fragments containing 45 amino acids each, aa 1 45, aa 45 90, and aa 90 135, but all three f ragments still exhibited self activation (Figure 3 4). The only ATF4 fragment that did not have self activation activity was the bZIP domain (aa 270 351) (Figure 3 5), and consequently, it was used as bait in the screening. For the prey, a HepG2 cDNA lib rary was prepared by using RNA from cells that had been incubated for 8 h in 2 mM HisOH to obtain higher mRNA abundance for those genes that are up regulated by amino acid deprivation. Screening was carried out as described in Chapter 2. A typical screen ing experiment is shown in Figure 3 5. Successful transformation of bait plasmid was shown by growth on the Trp selection plate, and successful co transformation of the bait and the prey was shown by growth on the Trp/ Leu plate. The Trp/ Leu/ His pla tes supplemented with 10mM 3 AT (3 aminotriazole) were used as a weak selection condition, and typically had 0 10 positive colonies on each plate. The Trp/ Leu/ His/ Ade plates were used as a strong selection condition, and typically had 0 3 colonies on each plate. Due to the extreme stringency of the Trp/ Leu/ His/ Ade selection condition, the majority of the screening was carried out with the Trp/ Leu/ His/10mM 3 AT selection, which has intermediate selection strength. A total of 63 positive clones were identified, and listed in Table 3 1. The positive clones included several bZIP transcription factors, such as ATF3, C/EBP and ATF4 itself. Also identified were nuclear factor erythroid2 related factor (Nrf 2) and factor inhibiting ATF4 mediated t ranscription (FIAT), two proteins known to bind ATF4 (He et al., 2001; Yu et al., 2005) Among the positive clones obtained, the two highest scored genes were the homolog of yeast ubiquitin conjugating enzyme UBC9 (hUBC9 or E2I) with seven independent clone s identified, and the C/EBP homology protein (CHOP) with five independent clones. The hUBC9

PAGE 63

48 was demonstrated later by luciferase reporter assay to have no major effect on transcription from the ASNS promoter (Figure A 1), therefore was put on hold for fut ure investigation. The five CHOP clones all corresponded to the open reading frame of CHOP; two clones were the full length CHOP sequence, one was the entire bZIP domain, and two were the leucine zipper domain (Figure 3 6A). This result indicated that CH OP interacts with ATF4 through its leucine zipper motif. A back transformation experiment was performed using the prey plasmid expressing full length CHOP to test if CHOP has self activation activity. As shown in Figure 3 6B, transformation of CHOP/pGAD Rec plasmid alone did not cause activation of the marker genes as evidenced by a lack of growth on the Trp/ Leu/ His or the Trp/ Leu/ His/ Ade plate. A confirmation experiment was also performed to confirm ATF4/CHOP interaction in yeast. In this experi ment, the ATF4/pGBKT7 bait plasmid and the CHOP/pGAD Rec prey plasmid were co transformed into AH109 cells, and the growth of the transformed cells on both the Trp/ Leu/ His and the Trp/ Leu/ His/ Ade plates demonstrated that CHOP interacts with ATF4 in yeast (Figure 3 6C). CHOP Interacts with ATF4 in vivo Gachon et al. (Gachon et al., 2001) demonstr ated that CHOP directly interacts with ATF4 in a glutathione s transferase pull down assay, in vitro However, whether or not CHOP also interacts with ATF4 in vivo was not answered in their study. To further investigate the interaction of CHOP and ATF4 i n intact cells, a co immunoprecipitation (co IP) assay was performed. An HA ATF4 fusion protein and CHOP were expressed either alone or in combination in human HEK293T cells. Whole cell extracts were collected and subjected to immunoprecipitation (IP) wi th HA antibody (mouse monoclonal) conjugated agarose beads, and the presence of CHOP was detected by immunoblotting analysis using CHOP antibody (rabbit polyclonal) (Figure 3 7A). CHOP was only detected in the immunoprecipitate when HA ATF4

PAGE 64

49 and CHOP were co expressed, but not when HA ATF4 or CHOP was expressed alone, indicating that the co IP of CHOP with ATF4 was specific. In a reciprocal co IP experiment, whole cell extracts from HEK293T cells co expressing CHOP and HA ATF4 were subjected to IP with ant i CHOP antibody (mouse monoclonal) or normal mouse IgG antibody, as a negative control (Figure 3 7B). ATF4 was detected in the immunoprecipitate after IP with the CHOP antiboby, but was absent from the immunoprecipitate after IP with normal mouse IgG (n/s IgG). Taken together, these experiments, in conjuction with the observation from Gachon et al. (2001), demonstrate that ATF4 and CHOP directly interact with each other in vivo CHOP Inhibits Transcriptional Activation from the ASNS Promoter Given that th e induction of the ASNS gene by amino acid limitation is mediated by ATF4 (Chen et al., 2004; Siu et al., 2002; Pan et al., 2003) I investigated whether CHOP, as an ATF4 interacting protein, plays a role in the regulation of the ASNS gene. To test CHOP function on the ASNS promoter, a luciferase reporter assay was performed. HepG2 cells were co transfected with the ASNS 173/+51 pro moter/Luc reporter plasmid (Figure 3 8A) and a gradient concentration of CHOP expression vector. To test the function of CHOP on amino acid limitation induced transcription from the ASNS promoter, the cells were incubated in MEM or MEM containing 2 mM His OH for 12 h. As shown in Figure 3 8B, ASNS promoter activity was induced by HisOH treatment and over expression of CHOP negatively regulated the induction in a concentration dependent manner. To determine if CHOP directly antagonizes ATF4 activity on the ASNS promoter, HepG2 cells were co transfected with the ASNS 173/+51 promoter/Luc reporter plasmid and expression vectors for ATF4 and CHOP (Figure 3 8C). In a previous study by Pan et al. (Pan et al., 2003) the Kilberg laboratory showed that 10 ng/well of ATF4 expression plasmid activated the ASNS promoter activity to a level approximately comparable to that by HisOH treatment. Therefore, in the present study, the cells were co transfected with 10 ng/well of ATF4

PAGE 65

50 expression plasmid and a concentration gradient of CHOP expression vector. As sh own in Figure 3 8C, co expression of CHOP and ATF4 revealed a concentration dependent CHOP repression of the promoter activity induced by ATF4. Taken together, these experiments demonstrate that CHOP negatively regulates the ATF4 driven transcription medi ated by the NSRU within the ASNS promoter. Effect of CHOP Over Expression on Endogenous Gene Expression To examine whether CHOP over expression affects the endogenous expression of ASNS in the context of chromatin, HEK293T cells were transfected with CHOP or, as a negative control, the green fluorescent protein (GFP). The HEK293T cells were chosen because these cells can be transfected at very high efficiency using a CaP transfection protocol, as described in Chapter 2. As shown in Figure 3 9A, an estimat ed transfection efficiency of 50 60% was achieved as shown by the ratio of GFP fluorescent cells to total cells. To monitor the expression of CHOP protein, immunoblotting was performed and it showed a concentration dependent expression from a 0 2 g plasm id/60 mm dish (Figure 3 9B). Although CHOP is also induced by amino acid deprivation, the CHOP protein level from over expression is higher than the endogenous CHOP, as incubating the transfected cells in HisOH for 8 h did not alter the CHOP protein level with or without over expression. Therefore the contribution from the endogenous CHOP can be neglected, and the over expression really represents an over production of CHOP (Figure 3 9B). To investigate the effect of CHOP over expression on the induction of ASNS, as well as several other amino acid responsive genes, by amino acid deprivation, total RNA was collected and subjected to qRT PCR to measure the mRNA abundance of specific genes. As shown in Figure 3 9C, CHOP over expression did not affect the b asal level of the endogenous ASNS mRNA, but the induction of ASNS expression by HisOH treatment was decreased by

PAGE 66

51 ~40%. Given the transfection efficiency of only 50 60%, the repressive effect of CHOP on the endogenous ASNS gene is likely to be quite strong In the contrast, other genes that are activated by amino acid limitation, SNAT2 (Palii et al., 2004) VEGF (Abcouwer et al., 2002) CAT1 (Fernandez et al., 2003) and CHOP itself (Bruhat et al., 1997) were largely unaffected by CHOP over expression. As a positive control, the T RB3 gene was also investigated for its response to CHOP over expression. Jousee et al. (Jousse et al., 2007) have shown that TRB3 expression is increased following amino acid limitation and Ohoka et al. (Ohoka et al., 2005) reported that the TRB3 gene contains AARE like genomic elements that are responsive to activation by ATF4 and CHOP. As shown in Figure 3 9C, in the HEK293T ce lls, TRB3 mRNA content was modestly increased by HisOH treatment, but the increase was substantially enhanced by over expression of CHOP. These results are consistent with TRB3 as a CHOP inducible gene and show striking contrast between ASNS and TRB3 in t erms of their response to CHOP over expression. Regulation of CHOP Expression by the AAR and the UPR Pathways CHOP is induced by both amino acid deprivation (Bruhat et al., 1997) and ER stress (Yoshida et al., 2000) Its induction pattern by either stress has been addressed in isolated studies (Bruhat et al., 1997; Okada et al., 2002; Wang et al., 1998b) However there has not been a systematic study that compared the CHOP expression patterns under these two stress conditions. To establish the induction pattern of CHOP expression by amino acid deprivation and ER stress, HepG2 cells were incubated in MEM, MEM con taining 2 mM HisOH, or MEM containing 300 nM thapsigargin (Tg) for 0 24 h, RNA and protein extracts were collected, and subjected to qRT PCR and immunoblotting, respectively. As shown in Figure 3 10A, after either HisOH or Tg treatment, CHOP mRNA expressi on was rapidly induced within 2 h and remained elevated throughout the entire 24 h time course for both treatments. However,

PAGE 67

52 compared to amino acid deprivation, the Tg triggered ER stress caused a much greater induction of CHOP mRNA by about two fold at a ny specific time point (Figure 3 10A). Although the basal level of CHOP protein was below detection under normal culture condition, HisOH or Tg treatment quickly induced CHOP protein within 2 h (Figure 3 10B). Surprisingly, HisOH treatment only caused a transient induction of CHOP protein with a peak at about 4 h and then a decrease after 8 12 h. This lack of correlation between CHOP mRNA and protein expression (compare Figure 3 10A and Figure 3 10B) is consistent with a previous report that CHOP is also regulated by nutrient stresses at the translational level (Jousse et al., 2001) In striking contrast, during Tg tr eatment the protein level of CHOP peaked at 4 h, at a level much higher than that during HisOH treatment, and remained at the high level throughout the entire 48 h time course, roughly paralleling the mRNA expression pattern (compare Figure 3 10A and Figur e 3 10B). Given that CHOP is reported to be an inducer of cell cycle arrest and apoptosis (Zinszner et al., 1998) the transient and relatively low level of CHOP protein expression following amino acid deprivation may have physiological significance as to temporarily protecting the cells from programmed cell death while trying to resolve the stress. As a positive control, the ATF4 pro tein abundance was also measured. Both pathways caused a robust and prolonged induction of ATF4 protein, and exhibited less of a difference as compared to that for CHOP protein expression (Figure 3 10B). These data suggest that during the AAR there may b e translational or post translational mechanisms that hold the CHOP protein production in check. ER Stress, but Not Amino Acid Deprivation, Triggers Cell Apoptosis As noted above, the difference in magnitude and time course of CHOP protein expression durin g amino acid deprivation or ER stress may be an indicator for differential activation of the apoptotic pathway. To test this hypothesis, experiments were designed to determine if these two pathways have different kinetics in terms of apoptosis and if thos e kinetics correlate with CHOP

PAGE 68

53 protein expression. HepG2 cell were incubated in MEM, MEM containing 2 mM HisOH, or MEM containing 300 nM Tg for 0 48 h. Whole cell extracts were collected at specific times and subjected to immunoblotting analysis to probe for cleavage of caspase 3, which is an indicator of cell apoptosis (Mazumder et al., 2008) As shown in Figure 3 10C, cells cultured in MEM containi ng HisOH did not exhibit detectable levels of caspase 3 cleavage up to 48 h of incubation, whereas significant cleavage was observed after 24 h of Tg treatment indicating that the apoptotic pathway had been triggered by Tg, but not HisOH treatment. The Ind uction of ASNS Expression is Enhanced by CHOP Knock down Over expression assays are artificial in the sense that by producing an over whelming amount of protein, they can drive biological reactions that are less important under physiological conditions. W ith this in mind, in order to address whether the endogenouos CHOP also plays detectable inhibitory function on the ASNS expression, an siRNA approach was used to knockdown the CHOP expression and then the induction of ASNS by these stress pathways was ass essed. HepG2 cells were transfected with either control siRNA or CHOP siRNA, and at 36 h post transfection the cells were incubated in MEM, MEM containing 2 mM HisOH, or MEM containing 300 nM Tg for 4 h. The rationale of choosing the 4 h time point was b ecause firstly, the CHOP protein level peaks at 4 h during amino acid deprivation, and secondly, the transcription of ASNS is subjected to the regulation of other repressors, such as ATF3 (Chen et al., 2004) in the later phase of the activatio n. Therefore, to avoid functional overlap between CHOP and other repressors, an earlier time point (4 h) was selected. As shown in Figure 3 11, both mRNA (Figure 3 11C) and protein (Figure 3 11D) expression of CHOP were efficiently reduced by siRNA treat ment. As for the expression of the ASNS gene, I chose to measure the transcriptional activity because it is a more kinetic parameter and is especially useful during the early phase of transcriptional activation, as the steady state mRNA has yet to be accu mulated.

PAGE 69

54 As shown in Figure 3 11A, with reduced CHOP expression, the induction of ASNS transcriptional activity by either HisOH or Tg treatment was further enhanced. Consistent with the observation that more CHOP protein is produced during Tg treatment, the additional enhancement of ASNS transcription during the ER stress was larger than that during HisOH treatment. These experiments further support the proposal that CHOP functions as a repressor of ASNS transcription in response to the AAR and UPR pathw ays. The Repressive Function of CHOP Requires Both N and C Terminal Portions of the Protein There have been two distinct mechanisms proposed for CHOP action, both of which involve the formation of heterodimers between CHOP and its binding partners. The first mechanism was demonstrated by Ron and Habener (Ron and Habener, 1992) as they showed that CHOP forms non functional heterodimers with its binding partners, such as other C/EBP proteins, and sequesters them from binding to their DNA recognition sequences, whereas the second mechanism was based on the observation that CHOP heterodimers can bind to specific DNA sequences (Ohoka et al., 2005) and carry out its transcriptional activity. Given the fact that the leucine zipper domain of CHOP is sufficient to bind ATF4, as suggested by the yeast two hybrid screening (Figure 3 6A), I proposed that if CHOP inhibits ATF4 activity on the ASNS gene by sequestering ATF4 from binding to the ASNS promoter, similar to that described in the first mechanism, then the bZIP domain alone should be able to inhibit ATF4 activity. To test this pos sibility, three constructs were prepared: full length CHOP, the N terminal (aa 1 100, transcriptional activation/repression domains) or the C terminal (aa 101 169, bZIP domain) portion of CHOP were cloned into the pEF6 V5/His expression vector (Figure 3 12 A) and their expression was confirmed by transfecting these constructs into HEK293T cells followed by immunoblotting analysis (Figure 3 12B). Their functions on the amino acid deprivation or

PAGE 70

55 ATF4 mediated activation of the ASNS gene were then tested in a luciferase reporter assay. HepG2 cells were co transfected with different CHOP fragments and with the ASNS 173/+51 promoter/Luc reporter plasmid, in combination with either HisOH treatment or ATF4 over expression (Figure 3 12C). As expected, the full lengt h CHOP protein repressed the induction of ASNS promoter activity by HisOH treatment or ATF4 over expression. However, neither the N terminal nor the C terminal portion of CHOP had significant inhibitory effect on the induction of ASNS driven transcription (Figure 3 12C). CHOP Binding to AARE Containing Genes The fact that CHOP inhibition of ATF4 action requires the entire protein led to the hypothesis that the CHOP transcriptional domain is essential for its action on the ASNS gene, therefore DNA binding is likely to be required. To investigate CHOP binding to the ASNS promoter containing the AARE element, a chromatin immunoprecipitation (ChIP) analysis was performed as described in Chapter 2, in HepG2 cells incubated in MEM, MEM containing 2 mM HisOH, o r MEM containing 300 nM Tg for 4 h (Figure 3 13). As a positive control, the TRB3 promoter was also investigated because it was reported that there are three AARE like sequences in tandem in the TRB3 promoter, and these sequences direct the binding of CHO P to DNA (Ohoka et al., 2005) As expected, the binding of CHOP to the TRB3 promoter was dramatically increased following HisOH or Tg treatment (Figure 3 13). Following activation of the AAR and UPR pathways, the asso ciation of CHOP with the ASNS promoter was also significantly increased, and the magnitude was greater in Tg treated cells compared to that in the HisOH treated cells (Figure 3 13), consistent with the observation that Tg induces CHOP protein abundance to a higher level than does HisOH (Figure 3 10). As a negative control, the association of CHOP with ASNS exon 7 (coding region) was also investigated, and showed minimal binding of CHOP, indicating that the association of CHOP with the ASNS promoter is

PAGE 71

56 spec ific (Figure 3 13). These data demonstrate that CHOP is functionally associated with the ASNS promoter containing the AARE element. In conjunction with the observation that CHOP binds to the AARE like sequences of the TRB3 gene and activates its transcri ption, it is likely that DNA binding is a mechanistic step for CHOP function on AARE containing genes. To test this hypothesis and expand our study to other amino acid responsive genes that contain an AARE element, ChIP analysis was also performed on the SNAT2, VEGF, and CAT 1 genes for CHOP binding (Figure 3 13). For SNAT2 and VEGF, there was a low basal level of binding that was significantly increased in response to HisOH or Tg treatment. For CAT 1, although the increased CHOP association reached stat istical significance after Tg treatment, the absolute amount of CHOP association was quite small (Figure 3 13). These results suggest that CHOP may be associated with all functional AARE elements under amino acid deprivation or ER stress. However, given that CHOP over expression did not alter the expression pattern of SNAT2, VEGF, or CAT 1 under either amino acid deprivation or ER stress, whether or not this physical association has functional significance remains to be studied. To further investigate i f the association of CHOP with the AARE regions is always ATF4 are bound simultaneously to the same DNA fragment. HepG2 cells were incubated in MEM, MEM containing 2 mM Hi sOH, or MEM containing 300 nM Tg for 4 h. Formaldehyde cross linked chromatin was first immunoprecipitated with CHOP antibody, and then subjected to a second immunoprecipitation with ATF4 antibody or normal rabbit IgG antibody. Immunoprecipitated DNA fra gments were purified and subjected to qPCR analysis using the ATF4 and CHOP on the AARE regions of the same group of genes. As shown in Figure 3 14,

PAGE 72

57 following ac tivation of the AAR or UPR pathway, genomic fragments containing the AARE regions from the TRB3 ASNS, and VEGF genes were significantly enriched in the ATF4 immunoprecipitated chromatin compared to the control IgG. The degree of enrichment on each gene r oughly paralleled that observed in the CHOP chromatin IP experiment (Figure 3 13). Significant enrichment of CHOP/ATF4 on the SNAT2 AARE region was only observed during the UPR but not the AAR. The co occupancy of CHOP and ATF4 did not reach statistic si gnificance on the CAT 1 AARE region, which is consistent with the weak binding of CHOP on that region (Figure 3 13). As a negative control, the association of CHOP/ATF4 with ASNS exon 7 (coding region) was minimal (Figure 3 14). Conclusions and Discussi ons The pivotal role of ATF4 in the amino acid response pathway inspired me to find ATF4 interacting proteins in order to better understand the mechanism of ATF4 dependent transcription. Dr. Hong Chen and Dr. Michelle Thiaville in the Kilberg laboratory h ad utitilized ChIP analysis to screen for transcriptional co activators on both the ASNS and SNAT2 gene. However, most co activators tested in their studies, including several histone acetyltransferases (HATs), the mediator complex, several chromatin remo deling complexes, and several other transcription factors, showed no association to the ASNS or SNAT2 AARE region (Kilberg laboratory, unpublished results). The original goal of the current study was to utilize an unbiased approach to identify ATF4 intera cting proteins, especially potential co activators. However, due to the limitation of the yeast two hybrid screening, the full length ATF4 could not be used as bait because it has self activation activity on the selection marker genes in yeast. As a comp romise, the ATF4 bZIP domain was used as bait to screen for ATF4 interacting partners. Indeed, by using the bZIP domain as bait in a yeast two hybrid screen, Kawai et al. (Kawai et al., 1998) had successfully identified a novel leucine zipper containing serine/threonine kinase, the ZIP kinase,

PAGE 73

58 as an ATF4 interacting protein and mediator of apoptosis. In the current study, another bZIP transcription factor, CHOP, was identified to inte ract with ATF4 through its leucine zipper motif, by yeast two hybrid screening. Gachon et al. (Ga chon et al., 2001) documented CHOP/ATF4 interaction in vitro in a GST pull down assay. I confirmed their interaction in mammalian cells by co immunoprecipitation analysis. O ver expression analysis demonstrated that not only does CHOP inhibit the HisOH treatment or ATF4 over expression mediated transcriptional activation from the ASNS promoter in vitro but it also inhibits the endogenous ASNS expression during AAR or UPR. siRNA mediated knock down of CHOP demonstrated that the endogenous CHOP also func tions to antagonize the induction of the ASNS gene by either amino acid deprivation or ER stress. The inhibition effect of CHOP on ATF4 function requires the entire protein, arguing against a mechanism by which CHOP sequesters its binding partners from bin ding to DNA through its bZIP domain only. By ChIP/double ChIP analysis, I demonstrated that not only do CHOP and ATF4, presumably as a heterodimer, co occupy the AARE region of the ASNS gene, but they also bind to the AARE region of several other amino ac id responsive genes, such as TRB3 and SNAT2. Interestingly, as opposed to its inhibitory function of sequestering its binding partner from binding to DNA, CHOP can act as either an activator or a repressor when it physically associates with DNA. For exam ple, CHOP and ATF4 cooperatively activate the TRB3 gene by binding to its AARE sequences. For ASNS and SNAT2, while ATF4 functions to activate both genes, CHOP specifically inhibits ATF4 function on the ASNS gene, but has largely no effect on the SNAT2 ge ne. This difference in CHOP function may be partially explained by its interaction with a co activator or a co repressor depending on the gene context. However, given the fact that CHOP, and presumably the ATF4/CHOP heterodimer, acts on the

PAGE 74

59 same AARE ele ment on different genes, the AARE sequence per se does not seem to be the only determinant for CHOP function in terms of activating or repressive. The flanking sequence around the AARE element or single nucleotide differences within the AARE itself may he lp to form a functional unit, which serves as a binding site for a group of transcription factors including CHOP and ATF4, and the transcriptional output from this unit is determined by the collective function of those transcription factors. Indeed, Ohoka et al. (2005) demonstrated that the flanking sequence of the TRB3 AARE element might contain one or more binding sites for additional factors that contribute to the distinct function of the ATF4/CHOP heterodimer at a particular gene. The stress response unit (nt +201 to +312) of the TRB3 gene contains three identical tandem repeats each consisting of 33 bp. Each of the repeats contains a CCAAT like element, a CHOP binding site, and an AARE element, which partially overlaps with the CHOP binding site. Mu tagenesis analysis showed that not only were the CHOP binding site and the AARE element essential for the TRB3 induction by ER stress, but the CCAAT like element was also required (Ohoka et al. 2005). The ASNS promoter does not have the CCAAT like element and is repressed by CHOP over expression, leading to the possibility that the CCAAT like element may be involved in the discrimination between activation and repression by the ATF4/CHOP heterodimer. The responses of mammalian cells to stress conditions a re robust, yet intricately regulated. Many stress conditions, such as amino acid deprivation (Harding et al., 2000a) ER stress (Costa Mattioli et al., 2007) the presence of long double strand RNA (Barber, 2005) and heme deficiency (Yoneda et al., 2001) converge onto the same transcription factor, ATF4, which is translationally induced through the phosphorylation of eIF2 (Vattem and Wek, 2004) The importance of ATF4 is indirectly demonstrated in yeast, as its yeast functional ortholog, GCN4,

PAGE 75

60 t deprivation (Hinnebusch and Natarajan, 2002) Surprisingly, although ATF4 and GCN4 share little or no similarity in their protein sequences, when artificially recruited to DNA in the yeast two hybrid screenin g, even small fragments (45 aa) of the ATF4 trans activation domain strongly activated the yeast reporter genes, all in the absence of mammalian cofactors. The potency of ATF4 in activating its target genes and the fact that ATF4 protein level is induced dramatically by many stress conditions raises the question of whether or not ATF4 activity is counter regulated by a control mechanism to ensure a proper degree of ATF4 response, and provide gene specificity. Indeed, several negative regulators of ATF4 ha ve been identified, such as FIAT (Yu et al., 2005) ATF3 (Fawcett et al., 1999; Wolfgang et al., 1997; Pan et al., 2003) and TRB3 (Jousse et al., 2007) The Kilberg laboratory has demonstrated that during the later phase of amino acid response (8 12 the ASNS promoter and feedback repr ess the induction of ASNS by ATF4 (Chen et al., 2004; Thiaville et al., 2008) This mechanism is a self limiting process in the sense that ATF4 up discovery of CHOP as a repressor of ATF4 dependent regulation of the ASNS gene provides another example of such a regulator. Moreover, given the fact that C HOP is also translationally regulated in a similar mechanism to that for ATF4 by stress conditions (Jousse et al., 200 1) and its protein induction is as robust as ATF4 (Figure 3 10B), CHOP may serve as an early regulator at the later stage of stress responses. Besides its functio n to bind to and negatively regulate other C/EBP proteins (Ron and Habener, 1992) CHOP was also was reported to regulate the function of several bZIP

PAGE 76

61 transcription factors from the ATF family, though the information is relatively limited. Chen et al. (Chen et al., 1996) demonstrated that CHO P interacts with ATF3, and inhibits ATF3 function by forming a non functional heterodimer, a mechanism similar to that by which CHOP inhibits the activity of C/EBP proteins. And Ohoka et al. (Ohoka et al., 2005) showed that CHOP and ATF4 cooperatively activate TRB3 expression during ER stress. Al Sarray et al. (Al et al., 2005) reported that CHOP antagonizes ATF4 o r ATF5 action when they were co expressed exogenously. My results are consistent with that in vitro study and demonstrated that CHOP interacts with ATF4 in vivo and negatively regulates ATF4 functions on the ASNS gene. My ysis demonstrated that both ATF4 and CHOP are associated with the ASNS promoter region during amino acid deprivation and ER stress. These results suggest that the ASNS C/EBP ATF composite site serves as a n ATF4 CHOP binding site. Moreover, this observati on appears to hold true for several other C/EBP ATF element containing genes, including TRB3 SNAT2 and VEGF. It is interesting that the association of CHOP to the TRB3 gene is dramatically greater (at least an order of magnitude) than that on the other A ARE containing genes. This could be explained by the three tandem copies of the C/EBP ATF sequence in the TRB3 promoter, whereas most other AARE genes have only one. CHOP is a well known inducer of cell cycle arrest and apoptosis during stress condition s (Zinszner et al., 1998) The current study compared the cellular responses to amino acid deprivation and ER stress in terms of the activation of apoptosis. In the HepG2 cells, CHOP protein induction was weak and transient during amino acid deprivation, but was much greater and sustained during ER stress. Consistent with this result, no apoptosis was observed up to 48 h of amino aci d deprivation, whereas ER stress triggered apoptosis within 24 h.

PAGE 77

62 Collectively, the results from this study document the transcriptional activity of CHOP as an ATF4 interacting partner on the ASNS gene, and extend the analysis to other amino acid responsiv e genes. Further investigation of the mechanisms by which CHOP regulates transcription will provide additional insight into the transcription factor network involved in the cellular responses to nutrient stress conditions.

PAGE 78

63 Figure 3 1. Schematics of di fferent ATF4 fragments tested as bait in the yeast two hybrid screen. The diagram shows the entire domain structure of the 351 amino acids that make up the ATF4 protein and the fragments tested as bait for yeast two hybrid screening. For the fragments, t he stippled sections indicate the ATF4 regions that exhibited self activation when tested. The black bar (aa 270 351), corresponding to the basic leucine zipper domain, was used as bait in the screen.

PAGE 79

64 Figure 3 2. Self activation test of the full leng th ATF4 protein as bait. AH109 cells were transformed with no plasmid (mock transformation), pGBKT7 vector, or hATF4 FL/pGBKT7 plasmid, as described in Chapter 2, and then plated on SD Trp, SD Trp/ His, and SD Trp/ His/ Ade agar plates. Growth on the Tr p plates indicates successful transformation of the bait plasmid. Growth on the His or His/ Ade selection medium indicates the activation of corresponding reporter gene, HIS3 or HIS3/ADE2, respectively.

PAGE 80

65 Figure 3 3. Self activation test of differen t fragments from the ATF4 activation domain as bait. AH109 cells were transformed with pGBKT7 vector, hATF4 AD (1 270)/pGBKT7, hATF4 AD (1 200)/pGBKT7, hATF4 AD (1 130)/pGBKT7, or hATF4 AD (130 270)/pGBKT7 plasmid, as described in Chapter 2, and then plat ed on SD Trp, SD Trp/ His, and SD Trp/ His/ Ade agar plates. Growth on the Trp plates indicates successful transformation of the bait plasmid. Growth on His or His/ Ade selection medium indicates the activation of corresponding reporter gene, HIS3 or HIS3/ADE2, respectively.

PAGE 81

66 Figure 3 4. Self activation test of 45 aa fragments from the ATF4 activation domain as bait. AH109 cells were transformed as described in Chapter 2, with pGBKT7 vector, hATF4 AD (1 45)/pGBKT7, hATF4 AD (44 92)/pGBKT7, or hAT F4 AD (91 135)/pGBKT7 plasmid, and then plated on SD Trp, SD Trp/ His, and SD Trp/ His/ Ade agar plates. Growth on the Trp plates indicates successful transformation of the bait plasmid. Growth on His or His/ Ade selection medium indicates the activat ion of corresponding reporter gene, HIS3 or HIS3/ADE2, respectively.

PAGE 82

67 Figure 3 5. Yeast two hybrid screen with ATF4 bZIP domain as bait. AH109 cells were transformed with pGBKT7 vector or hATF4 bZIP (270 351)/pGBKT7 plasmid, as described in Chapter 2 and then plated on SD Trp, SD Trp/ His, and SD Trp/ His/ Ade agar plates (Panel A). Growth on the Trp plates indicates successful transformation of the bait plasmid. Growth on His or His/ Ade selection medium indicates the activation of correspondin g reporter gene, HIS3 or HIS3/ADE2, respectively. The yeast two hybrid screening was performed by transforming AH109 cells with hATF4 bZIP (270 351)/pGBKT7, pGADT7 Rec, and HepG2 cDNA library, and then plating the cells on SD Trp, SD Trp/ Leu, SD Trp/ Leu / His, and SD Trp/ Leu/ His/ Ade agar plates (Panel B). Growth on the Trp or Leu plates indicates successful transformation of the bait plasmid or the prey plasmid, respectively. Growth on His or His/ Ade selection medium indicates the activation of corresponding reporter gene, HIS3 or HIS3/ADE2, respectively. Panel B shows the plates from a typical screen.

PAGE 83

68 Table 3 1. Genes identified from the yeast two hybrid screening. Pubmed entry Gene name Alias No. of clones NM_003345.3 Homo sapiens ubiquitin conjugating enzyme E2I (UBC9 homolog, yeast) C358B7.1, P18, UBC9 7 NM_004083 Homo sapiens DNA damage inducible transcript 3 (DDIT3) CEBPZ, CHOP, CHOP10, GADD153, MGC4154 5 NM_003017.4 Homo sapiens splicing factor, arginine/serine rich 3 (SFRS3) SRp20 4 NM_001675.2 Homo sapiens activating transcription factor 4 CREB 2, CREB2, TAXREB67, TXREB 3 NM_001102667.1 Homo sapiens proteasome (prosome, macropain) subunit, alpha type, 4 (PSMA4) HC9, HsT17706, MGC111191, MGC12467, MGC24813 2 NM_000146.3 Homo s apiens ferritin, light polypeptide (FTL) MGC71996 2 NM_001030287.2 Homo sapiens activating transcription factor 3 (ATF3) ATF3 1 NM_001806.2 Homo sapiens CCAAT/enhancer binding protein (C/EBP), gamma (CEBPG), GPE1BP, IG/EBP 1 1 NM_006164.3 Homo sapien s nuclear factor (erythroid derived 2) like 2 (NFE2L2), Nrf2 NRF2 1 NM_018360.1 chromosome X open reading frame 15 (FIAT) RP11 716A19.4, FIAT, FLJ11209, LSR5, MGC126621, MGC126625 1 NM_004497.2 Homo sapiens forkhead box A3 (FOXA3) FKHH3, HNF3G, MGC1017 9, TCF3G 1 NM_003542.3 Homo sapiens histone cluster 1, H4c (HIST1H4C) H4/g, H4FG, dJ221C16.1 1 NM_005324.3 Homo sapiens H3 histone, family 3B (H3.3B) (H3F3B) RP11 396C23.1, H3.3B, H3F3A 1 NM_001111112.1 Homo sapiens huntingtin interacting protein 2 (H IP2) E2 25K, HIP 2, HYPG, LIG 1 NM_033027.2 Homo sapiens AXIN1 up regulated 1 (AXUD1) DKFZp566F164, FAM130B, TAIP 3, URAX1 1 NM_001177.3 Homo sapiens ADP ribosylation factor like 1 (ARL1) ARFL1 1 NM_020642.3 Homo sapiens chromosome 11 open reading fram e 17 (C11orf17) A kinase interacting protein (AKIP), BCA3 1 NM_004134.5 Homo sapiens heat shock 70kDa protein 9 (mortalin) (HSPA9) CSA, GRP75, HSPA9B, MGC4500, MOT, MOT2, MTHSP75, PBP74, mot 2 1 NM_002026.2 Homo sapiens fibronectin 1 (FN1) CIG, FINC, FN LETS, MSF 1 NM_000976.2 Homo sapiens ribosomal protein L12 (RPL12) RPL12 1 NM_001861.2 Homo sapiens cytochrome c oxidase subunit IV isoform 1 (COX4I1) COX4, COXIV, MGC72016 1 NM_000039.1 Homo sapiens apolipoprotein A I (APOA1) Apo AI, ApoA I, MGC117 399 1 NM_001813.2 Homo sapiens centromere protein E, 312kDa (CENPE) KIF10 1 NM_002815.2 Homo sapiens proteasome (prosome, macropain) 26S subunit, non ATPase, 11 (PSMD11) MGC3844 Rpn6 S9 p44.5 1 NM_002804.4 Homo sapiens proteasome (prosome, macro pain) 26S subunit, ATPase, 3 (PSMC3) TBP 1, TBP1 1 NM_017518.5 Homo sapiens 26S proteasome associated UCH interacting protein 1 (UIP1) HSXQ28ORF, UIP1 1 NM_001012661.1 sapiens solute carrier family 3 (activators of dibasic and neutral amino acid transp ort), member 2 (SLC3A2) 4F2, 4F2HC, 4T2HC, CD98, CD98HC, MDU1, NACAE 1

PAGE 84

69 Pubmed entry Gene name Alias No. of clones NM_001128431.1 Homo sapiens solute carrier family 39 (zinc transporter), member 14 (SLC39A14) KIAA0062, LZT Hs4 ZIP 14 ZIP14 cig19 1 NM_005116.5 Homo sapiens solute carrier family 23 (nucleobase transporters), member 2 (SLC23A2) KIAA0238, NBTL SLC23A1 SVCT2 YSPL2 hSVCT2 1 NM_020857.2 Homo sapiens vacuolar protein sorting protein 18 (VPS18) KIAA1475, PEP3 1 NM_001030.3 Homo sapiens ribosomal protein S27 (metallopanstimulin 1) (RPS27) MPS 1, MPS1 1 NM_001002.3 Homo sapiens ribosomal protein, large, P0 (RPLP0) L10E, MGC111226, MGC88175, P0, PRLP0, RPP0 1 NM_000308.2 Homo sapiens protective protein for beta galactosida se (galactosialidosis) (PPGB) N/A 1 NM_001428.2 Homo sapiens enolase 1, (alpha) (ENO1) ENO1L1, MBP 1, MPB1, NNE, PPH 1 NM_007126.3 Homo sapiens valosin containing protein (VCP) IBMPFD, MGC131997, MGC148092, MGC8560, TERA, p97 1 NM_016504.2 Homo sapiens mitochondrial ribosomal protein L27 (MRPL27) L27mt, MGC23716 1 NM_001932.3 Homo sapiens membrane protein, palmitoylated 3 (MAGUK p55 subfamily member 3) (MPP3) DLG3 1 NM_001015.3 Homo sapiens ribosomal protein S11 (RPS11) N/A 1 Homo sapiens chromos ome 12 genomic contig, reference assembly N/A 1 NM_021019.3 Homo sapiens myosin, light chain 6, alkali, smooth muscle and non muscle (MYL6) ESMLC, LC17, LC17 G, LC17 NM LC17A, LC17B, MLC 3 MLC1SM, MLC3NM, MLC3SM 1 NM_018256.2 Homo sapiens WD repeat domain 12 (WDR12) FLJ10881 FLJ12719 FLJ12720 YTM1 1 NM_020449.2 Homo sapiens THO complex 2 (THOC2) CXorf3 THO2 Tho2 dJ506G2.1 1 NM_014887.2 Homo sapiens phosphonoformate immuno associated protein 5 (PFAAP5) 92M18.3, CG005 CG016 FLJ361 95 FLJ41089 FLJ43077, PFAAP5 1 NM_003860.2 Homo sapiens barrier to autointegration factor 1 (BANF1), mRNA BAF BCRG1 BCRP1 D14S1460 MGC111161 1 N/A PREDICTED: Homo sapiens hypothetical protein LOC257407 (LOC257407) NA 1 NM_006836.1 Homo sa piens GCN1 general control of amino acid synthesis 1 like 1 (yeast) (GCN1L1) GCN1 GCN1L HsGCN1 KIAA0219 1 NM_003093.1 Homo sapiens small nuclear ribonucleoprotein polypeptide C (SNRPC) FLJ20302, U1 C 1 NM_006710.4 Homo sapiens COP9 constitutive ph otomorphogenic homolog subunit COP9, CSN8, MGC1297 MGC43256 SGN8 hCOP9 1 NM_004960.2 Homo sapiens fusion (involved in t(12;16) in malignant liposarcoma) (FUS) FUS CHOP FUS1 POMp75, TLS TLS/CHOP, hnRNP P2 1

PAGE 85

70 Figure 3 6. Identification of CHOP as an ATF4 interacting partner by yeast two hybrid screen. Panel A shows the domain structure of CHOP protein and the black bars below indicate the amino acid sequence of different CHOP clones identified in the yeast two hybrid screening. The indica ted N terminal (aa 1 100) and C terminal (aa 100 169) are the CHOP fragments over expressed to obtain the data in Figure 3 12. A self activation test for CHOP was performed by transforming AH109 cells with CHOP FL/pGADT7 Rec plasmid and plating on the SD Leu, SD Leu/ His, and SD Leu/ His/ Ade agar plates (Panel B). Growth on the Leu plates indicates successful transformation of the bait plasmid. A confirmation test for ATF4/CHOP interaction was performed by co transforming the AH109 cells with hATF4 FL/ pGBKT7 and CHOP FL/pGADT7 Rec plasmids, and plating on SD Trp/ Leu, SD Trp/ Leu/ His, and SD Trp/ Leu/ His/ Ade agar plates (Panel C).

PAGE 86

71 Figure 3 7. Interaction of CHOP and ATF4 in vivo HA ATF4 and CHOP were expressed either alone or in combination in HEK293T cells. Whole cell extracts were collected and subjected to immunoprecipitation using HA antibody conjugated agarose beads. Bound protein was eluted and subjected to immunoblot analysis to detect the presence of ATF4 and CHOP (Panel A). In a reci procal experiment, extracts were subjected to immunoprecipitation using CHOP antibody or normal mouse IgG antibody (IgG). Bound protein was eluted and subjected to immunoblot analysis to detect the presence of ATF4 and CHOP (Panel B).

PAGE 87

72 Figure 3 8. Effe ct of CHOP on ASNS promoter driven transcription activated by HisOH or ATF4. A diagram showing the structure of the ASNS 173/+51 promoter/Luc reporter (Panel A). HepG2 cells were co transfected wit ASNS 173/+51 promoter/Luc reporter plasmid and the CHOP pcDNA3.1 expression plasmid at the amounts indicated. To activate ATF4 dependent transcription, half of the cells were treated with 2 mM HisOH for 12 h (Panel B) and the othe r half were co transfected with 10 ng/well of ATF4 pcDNA3.1 expression plasmid (Panel C). The total amount of transfected DNA was kept constant among experimental groups by the addition of pcDNA3.1 plasmid. Cell extracts were assayed for luciferase activ ity as described in Chapter 2. Each experiment was repeated with three different batches of cells, and the results shown represent the means S.E.M.

PAGE 88

73 Figure 3 9. Effect of CHOP over expression on endogenous gene expression. HEK293T cells were transfe cted with GFP pcDNA3.1 or CHOP pcDNA3.1 plasmids as described in GFP transfected cells (2 g/60 mm dish). At 24 h post transfection, cells were incubated for 8 h in MEM or i n MEM containing 2 mM HisOH and then RNA and protein extracts were isolated. Protein extracts from GFP or CHOP transfected cells were subjected to immunoblot analysis to detect the presence of CHOP protein (Panel B). The RNA was subjected to qRT PCR ana lysis for the mRNA content of indicated genes, as well as GAPDH mRNA as an internal control. The results are shown as the means S.E.M. for three independent experiments (Panel C). An asterisk indicates that the induction of expression was significantly changed by CHOP expression (p < 0.05).

PAGE 89

74 Figure 3 10. Time course of CHOP expression and apoptosis following amino acid limitation and ER stress. HepG2 cells were incubated in MEM, MEM containing 2 mM HisOH, or MEM containing 300 nM Tg. At the times in dicated, RNA was isolated and analyzed by qRT PCR analysis for CHOP or GAPDH mRNA content (Panel A). The graph illustrates the means S.E.M. for three independent experiments. Where not shown, the error bars are contained with the symbol. In a second s eries of experiments, protein extracts were collected and subjected to imm unoblot analysis for actin content (Panel B). The data shown are representative of three independent experiments. To collect the data for Panel C, HepG2 cells were incubated for 0 48 h in MEM, MEM containing 2 mM HisOH, or MEM containin g 300 nM Tg. At the times indicated, protein extracts were collected and subjected to actin content. The blot shown is representative of two independent experiments.

PAGE 90

75 Figure 3 11. Effect of siRNA mediated CHOP knock down on endogenous ASNS expression. HepG2 cells were transfected with either control siRNA or CHOP siRNA. At 36 h post transfection, cells were incubated in MEM, MEM containing 2 mM HisOH, or MEM containing 300 nM Tg f or 4 h. Total RNA was isolated and subjected to qRT PCR analysis for ASNS hnRNA content to assess ASNS transcriptional activity (Panel A). The data are presented as the ratio of ASNS hnRNA to the GAPDH control. The CHOP mRNA and GAPDH mRNA content were also measured (Panel B). The graph illustrates the means S.E.M. for three independent experiments. actin content (Panel C).

PAGE 91

76 Figure 3 12. E ffect of truncated CHOP proteins on tr anscription driven by the ASNS promoter. A diagram showing the constructs that express full length or truncated CHOP proteins labeled with a V5 epitope tag (Panel A). The CHOP constructs were transfected into HEK293T cells and 12 h later protein extracts were sub jected to immunoblot analysis for V5 actin (Panel B). The CHOP constructs were also transfected into HepG2 cells along with the ASNS 173/+51 promoter/Luc reporter plasmid in combination with either 2 mM HisOH treatment or ATF4 over e xpression, as indicated (Panel C). Cell extracts were assayed for 3 assays and each experiment was repeated with two different batches of cells. The results shown repre sent the means S.E.M. An asterisk indicates that the value is significantly different (p < 0.05) from the control (0 ng/well CHOP).

PAGE 92

77 Figure 3 13. A ssociation of CHOP with genes containing a C/EBP ATF composite site. HepG2 cells were incubated in ME M control medium or treated with either 2 mM HisOH or 300 nM Tg for 4 h. Chromatin immunoprecipitation (ChIP) assays were performed with CHOP antibody and the enrichment of CH OP protein at the indicated genes was analyzed by qPCR, using primer sets specific for the regions of interest. Data were plotted as the ratio to the value obtained with a 1:20 dilution of input DNA. Each condition was analyzed in triplicate and each poi nt represents the S.E.M. for two independent experiments. An asterisk indicates that the value is significantly different from the MEM control (p < 0.05).

PAGE 93

78 Figure 3 14. C o occupancy of CHOP and ATF4 at C/EBP ATF composite sites. HepG2 cells were inc ubated in MEM control medium or treated with either 2 mM HisOH or 300 nM Tg for 4 h. Double eluted, and then subjected to the second immunoprecipitation with ATF4 antibody or normal rabbit IgG antibody. The enrichment of CHOP/ATF4 proteins at the indicated gene was analyzed by qPCR, using primer sets specific for the regions of interest. Data are plotted as the ratio to the value obtained with a 1:200 dilution of input DNA. Each point represents the S.E.M. for three independent experiments. An asterisk indicates that the value is significantly different from both the MEM control and the IgG control (p < 0.05).

PAGE 94

79 CHAPTER 4 REGULATION OF THE ASPARAGINE SYNTHETASE (ASNS) GENE BY ASPARAGINASE (ASNase) TREATMENT IN MOLT 4 LEUKEMIA CELLS AND ITS CORRELATION WITH ASNase RESISTANCE Introduction ASNase Therapy for ALL Patients As described in Chapter I, Asparaginase (ASNase) is a popular chemotherapeutic agent against acute lymphoblastic leukemia (ALL). The E. coli enzyme is a 141 kD protein that is comprised of four identical subunits, each containing an active site (Jerlstrm et al., 1989; Maita and Matsuda, 1980) ASNase is found in both prokaryotic and eukaryotic organisms, such as bacteria, yeast, plants, and vertebrates. The discovery that the ASNase isolated from E. coli was equally as effective as that in the guinea pig serum in promoting tumor regression (Mashburn an d Wriston, Jr., 1964) led to the large scale production of this enzyme, which is used for anti leukemia agent in human. The most popularly used ASNase in the clinical treatment of ALL is the Type II isozyme located in the periplasmic space of E. coli (Wriston, Jr. and Yellin, 1973; Wriston, Jr., 1985) ASNase catalyzes hydrolysis of the amino acid asparagine to aspartate and ammonia leading to rapid depletion of asparagine from the serum (and to a lesser extent glutamine ), and a subsequent decline in intracellular asparagine due to efflux (Cooney et al., 1970) This depletion of extracelluar asparagine appears to be the basis of the anti leukemic activity of ASNase. In mammals, the asparagine level is controlled by two functionally opposing enzymes, asparaginase (ASNase) and asparagine synthetase (ASNS). All normal tissues contain a measurable level of ASNS activity, which can be induced during asparagine depletion by ASNase. However, for unknown reasons, some lymphoblastic leukemia cells have an extremely low level of ASNS activity, therefore they are sensitive to ASNase induced asparagine depletion.

PAGE 95

80 ASNase treatment also results in a decrease in protein synthesis in its target cells (Ellem et al., 1970) consequently reducing the DNA, rRNA and tRNA synthesis. Besides inhibiting protein synthesis, ASNase also causes cell cycle blockade, which then triggers apoptosis (Story et al., 1993; Aslanian et al., 2001) Furthermore, ASNase treatment was also shown to cause decreased glycine levels in 6C3HED tumors, which may affect purine synthesis (Ryan and Dworak, 1970) The intracellular glutamine and glutamate levels are also affected by ASNase treatment (Bussolati et al., 1995) Collectively, ASNase can cause severe perturbation of metabolic pathways that are essential to normal cellular functions, therefore effectively kill the target cells. Besides its function to hydrolyze asparagine, the E. coli ASNase also has glutaminase activity. A higher glutaminase activity causes glutamine deprivation of the body, which is as sociated with immunosuppressive side effects. Therefore, the E. coli ASNase is more effective in the ALL treatment compared to the ASNase prepared from E. chrysanthemi, which has higher glutaminase activity (0.10 vs. 0.03 glutaminase/asparaginase ratio). During ASNase therapy, circulating asparagine and glutamine are eliminated from the plasma immediately after drug administration. However, the glutamine level typically rises after five days of treatment, presumably due to the up regulation of intracellu lar glutamine synthetase (Miller et al., 1969) Interestingly, the glutaminase activity of ASNase may be a useful feature in the treatment of the asparaginase resistant patient, as Rotoli et al. (Rotoli et al., 2005) reported that inhibition of glutamine synthetase activity with L methioninesulfoximine (MSO), in conjunction with ASNase treatment, triggered apoptosis in asparaginase resistant cells, due to the depletion of both asparagine and glutamine from the plasma.

PAGE 96

81 ASNase Resistance and AS NS Up Regulation Although ASNase is an effective chemotherapeutic agent against ALL cells, and has been used for more than 30 years as a key component of all induction protocols for treatment of childhood ALL, the ASNase resistance developed in certain r elapsed patients is one of the major drawbacks of ASNase therapy, as these patients typically have very poor prognosis. The ASNase resistance phenotype has been linked to elevated ASNS expression in ASNase resistant leukemia cells. Using extracts isolate d from 6c3HED tumors, Broome (Broome, 1968) discovered that the expression level of ASNS is higher in the ASNase resistant cells, compared to that in the sensitive tumors. The same phenomenon was subsequently discov ered in many other cell lines, including U937 lymphocytes (Kiriyama et al., 1989) HT 1080 fibroblast (Andrulis et al., 1990) Chinese hamster ovary (CHO) cells (Andrulis et al., 1979) and MOLT 4 leukemia cells (Hutson et al., 1997) In the Kilberg laboratory, Aslanian et al. (Aslanian et al., 2001) demonstrated that over expression of ASNS in the ASNase sensitive leukemia cell line MOLT 4 resulted in protection from ASNase induced cytotoxicity. Despite obvious correlation between ASNS expres sion and ASNase sensitivity in various cell lines, studies using patient samples have yielded conflicting information. In a 1969 study, 18 of 21 children with ALL cells that were dependent on extracellular asparagine to grow in vitro achieved remission; w hereas remission was observed in only 1 of 9 patients whose cells grew independently of extracellular asparagine (Oettgen and Schulten, 1969) Haskell and Canellos (Haskell and Canellos, 1969) observed higher ASNS enzymatic activity in five ASNase resistant versus four drug sensitive patients. More recently, Dbbers et al. (Dubbers et al., 2000) reported similar median values of ASNS activity in blasts from children with AML or ALL, but there was a large range in the activity levels between high and low samples. However, Fine et al. (Fine et al. 2005) observed no strong correlation in primary ALL cells when tested for ASNase

PAGE 97

82 resistance in vitro Many genetic abnormalities are associated with ALL, the most common one in childhood ALL is the cryptic t(12; 21)(p13;q22). This translocation fuses t he TEL gene on chromosome 12, a member of the ETS family of transcription factors to AML 1, which encodes to the AML 1/CBFB (core binding factor) transcription factor complex on chromosome 21 (Rubnitz et al., 1997) Stams et al. (Stams et al., 2003) hypothesized that TEL AML1 inhibition of ASNS expression might explain the increased in vitro sensitivity to ASNase of TEL AML1 positive leukemias, but their study showed that TEL AML1 positive ALL cells expressed five fold more ASNS mRNA than did TEL AML1 negative cells or normal controls. They also observed no difference between patients, positive or negative for TEL AML1, in the capacity to up regulate ASNS mRNA after an acute in vitro exposure to ASNase. Krejci et al. (Krejci et al., 2004) also observed increased ASNS mRNA in TEL AML1 positive cells, despite the increased ASNase sensitivity relative to TEL AML1 negative cells. Interestingly, although 10 of 15 TEL AML1 positive patients with ASNS expression below the m edian relapsed, none of the patients with elevated ASNS relapsed. However, Stams et al. (Stams et al., 2004) did not see this difference in relapse and survival relati ve to ASNS expression levels. Holleman et al. (Holleman et al., 2004) tested the ASNase sensitivity of leukemia cells from 173 pediatric patients and observed that ASNS mRNA was significantly elevated in ALL cells resistant to ASNase, consistent with a previously reported inverse relationship between the level of ASNS expression and ASNase sensitivity in a National Cancer Institute panel of 60 human cancer cell lines (Scherf et al., 2000) The major limitation of these recent studies is that only ASNS mRNA was analyzed by either arrays or RT PCR. Neither protein content nor en zymatic activity was assayed. Whether the relative ASNS mRNA level reflects ASNS protein expression is subjected to investigation in this chapter.

PAGE 98

83 Cell Lines Used in the Laboratory Studies of ALL Primary isolates of acute lymphoblastic leukemia (ALL) cel ls are not adaptable to long term culture, so a number of ALL derived cell lines have been established for laboratory research. Our laboratory has employed MOLT 4 cells, a human T cell derived cell line, to investigate the metabolic and gene expression ch anges that occur in ALL cells following ASNase treatment (Aslanian et al., 2 001) To extend these studies to B cell derived ALL cells, and to test cell lines with or without TEL AML1 expression, a TEL AML1( ) cell line, NALM 6, and a TEL ALM(+) cell line, REH, were chosen for comparison to the parental MOLT 4 cells (MOLT 4P) and to a drug selected ASNase resistant MOLT 4 line (MOLT 4R) (Aslanian et al., 2001) As described previously (Aslanian et al., 2001) a third MOLT 4 c ell line was generated by transducing MOLT 4P cells with a retrovirus that encodes human ASNS and then selecting for stable ASNS expression (MOLT 4P/ASNS). The current study utilized cultured ALL cell lines to study the correlation between ASNS mRNA and pr otein expression with ASNase resistance. The results provide explanations to the controversial observations made by different groups and add to our understanding of the mechanism for ASNase resistance of ALL patients. Results Characterization of ASNS a nd Transcription Factor Expression in the MOLT 4 Cells Aslanian et al. (Asla nian et al., 2001) in the Kilberg laboratory demonstrated that the ASNS expression is higher in the MOLT 4R cells, and exogenous expression of ASNS in the MOLT 4P cells alone causes ASNase resistance. To see if the regulation of ASNS expression by ASNase treatment is the same in the MOLT 4P and MOLT 4R cells, the expression of ASNS and some transcription factors involved in the AAR pathway were investigated. MOLT 4P cells

PAGE 99

84 were cultured in RPMI or RPMI containing 1U/mL ASNase for 12 h. MOLT 4R cells were maintained in RPMI containing 1U/mL ASNase. Total RNA and whole cell extracts were collected and subjected to qRT PCR and immunoblotting analysis. As shown in Figure 4 1A, the MOLT 4P cells expressed a very low basal level of ASNS mRNA, whereas the MOLT 4R cells expressed more than 15 fold ASNS mRNA compared to the MOLT 4P cells. However, after 12 h of ASNase treatment, the ASNS mRNA level in the MOLT 4P cells was dramatically induced to a level comparable to that in the MOLT 4R cells. As for the expre ssion of ATF4 and ATF3, which are both induced by amino acid deprivation at the transcriptional level and translational level, the basal mRNA expression of ATF4 and ATF3 in the MOLT 4P cells is similar to that in the MOLT 4R cells. Upon ASNase treatment, the mRNA level of ATF4 and ATF3 was significantly induced, and is even higher than that in the MOLT 4R cells which were maintained in the presence of ASNase (Figure 4 1A). The protein level of ASNS, ATF4 and ATF3 was also measured, as shown in Figure 4 1B The MOLT 4P cells did not express detectable ASNS protein, even after 12 h of ASNase treatment. In the contrast, the MOLT 4R cells expressed a high level of ASNS protein, when cultured in the medium containing 1U/mL ASNase. The ATF4 and ATF3 protein e xpression was induced by ASNase treatment in the MOLT 4P cells, to a similar level as that in the MOLT 4R cells. Interestingly, the pattern of ATF4 band in the MOLT4 P and MOLT 4R cells appeared different. This could be due to the differential post trans lational modification of the ATF4 protein, as ATF4 was shown to be subjected to phophorylation (Lassot et al., 2001) ubiquitylat ion (Lassot et al., 2001) and acetylation (Lassot et al., 2005) This difference in post translational modification of ATF4 in MOLT 4P and MOLT 4R cells may have physiological significance, as ATF4 has a pro apoptotic function, and a prolonged expression of ATF4 protein in the MOLT 4R cells could

PAGE 100

85 pose a threat to cell survival. Therefore, a mechanism involving post translational modification may alter ATF4 function and make it more specific for mediating adaptive mechanisms. These results suggested that the amino acid respons e (AAR) pathway is intact in both MOLT 4P and MOLT 4R cells. Assembly of the Transcription Machinery on the ASNS Promoter in the MOLT 4 Cells Although the transcription factors that are involved in the regulation of the ASNS gene, such as ATF4 and ATF3, ar e present at similar levels in the MOLT 4P and MOLT 4R cells during ASNase treatment, it is still possible that the regulation of ASNS is different in these two cell lines due to changes in the ASNS gene itself. To test that possibility, we investigated t he assembly of the transcription machinery on the ASNS promoter in the MOLT 4P and MOLT 4R cells. A chromatin immunoprecipitation (ChIP) assay was performed to measure the enrichment of components in the transcription machinery on the ASNS promoter (Figur e 4 2 and Figure 4 3). To measure the acute response of both cell lines to ASNase treatment, the MOLT 4R cells were pre incubated in normal RPMI medium and cultured for 96 h to eliminate any prolonged effect from ASNase treatment, which allowed the cells 4P and MOLT 4R cells were then treated with 1U/mL ASNase for 12h, followed by a 12 h drug (Figure 4 2). The basal binding o f all four factors was higher in the MOLT 4R cells compared to that in the MOLT 4P cells. Upon ASNase treatment, the binding of Pol II and all of the three transcription factors increased, in both MOLT 4P and MOLT 4R cells, and the binding in the resistan t cells was significantly higher than that in the parental cells. After 12 h withdraw of ASNase from the culture media, the binding of all four factors returned to the basal level, indicating that the induction of ASNS by amino acid deprivation is an acut e response. As

PAGE 101

86 negative a control, immunoprecipitation with a normal rabbit IgG showed minimal binding 2). To further investigate the enrichment of several other markers for trans criptional activation on the ASNS promoter, ChIP analysis was performed to measure the level of acetylation of histone H3 and H4, and the binding of several general transcription factors (GTFs), including TBP, TAF250, and TFIIB, on the ASNS promoter. As s hown in Figure 4 3, the H3 and H4 acetylation was significantly induced in both MOLT 4P and MOLT 4R cells upon ASNase treatment, and the level of acetylation was significantly higher in the resistant cells compared to that in the parental cells in both bas al and activated states. Consistent with the increased histone acetylation, the binding of TAF1, which is a subunit of the TFIID complex and has histone acetyl transferase (HAT) activity, was increased upon ASNase treatment, in both parental and resistant cells. The binding of TBP and TFIIB to the ASNS promoter was weak, possibly due to the poor quality of the antibodies. However, the binding of both GTFs was significantly increased in the MOLT 4R cells upon ASNase treatment. Collectively, these results demonstrate that the assembly of the transcription machinery on the ASNS promoter is generally stronger in the MOLT 4R cells than in the MOLT 4P cells. However, the fact that the transcription machinery does assemble on the ASNS promoter in the parental cells indicates that the mechanism through which the ASNS gene is regulated by amino acid deprivation in the MOLT 4P cells is similar to that in the MOLT 4R cell, though to different levels. These data are also consistent with the observation that the ASN S mRNA level was induced in the MOLT 4P cells by ASNase treatment, but was lower than that in the MOLT 4R cells (Figure 4 1). Irreversibility of the ASNase Resistance of the MOLT 4R Cells The MOLT 4R cells were derived from the MOLT 4P cells by treating them with low concentration of ASNase (Hutson et al., 1997) a process that mimics the clinical development of

PAGE 102

87 ASNase resistance in the ALL patients. Therefore, the investigation of the mechanism through which the ALL cells become resistance, and whether or not the resistance can be reversed, is important for the clinical treatment of ALL. To test if the MOLT 4R cell can naturally reverse to its ASNase sensitive state after ASNase withdraw, I maintained the MOLT 4R cells in normal RPMI medium without ASN ase for 6 weeks and named them MOLT 4R0. After 6 weeks of ASNase withdraw, the MOLT 4R0 cells retained its ASNase resistance, as treating these cells with ASNase did not cause any cell death from the trypan blue analysis (data not shown). We then investi gated and compared the ASNS expression in MOLT 4R0 and MOLT 4P cells. As shown in Figure 4 4A, the ASNS mRNA expression in the MOLT 4R0 cells was still significantly higher than that in the MOLT 4P cells, and upon ASNase treatment, the ASNS mRNA level was further induced. Consistent with the mRNA expression, the MOLT 4R0 cells also expressed a basal level of ASNS protein, which was further induced by ASNase treatment to a level that was comparable to that in the MOLT 4R cells. In the contrast, the MOLT 4 P cells did not express detectable ASNS protein with or without ASNase treatment (Figure 4 4B). The protein expression of ATF4 was also investigated, as shown in Figure 4 4B. The MOLT 4R0 cells maintained a basal level of ATF4 protein, which was lower th an that in the ASNase treated MOLT 4P or MOLT 4R cells but was higher than the basal ATF4 level in the MOLT 4P cells. This higher basal level of ATF4 protein may be responsible for maintaining the ASNS mRNA expression in the MOLT 4R0 cells in the absent o f ASNase. Moreover, ChIP analysis showed 4R0 cells compared to that in the MOLT 4P cells (Figure 4 5), indicating that the ASNS gene is actively transcribed in the MOLT 4R0 cells even without co ntinued ASNase challenge. These results demonstrate that the ASNS expression remains elevated in the MOLT 4R cells even after

PAGE 103

88 a prolonged drug withdraw, and this elevated ASNS expression may be responsible for the sustained ASNase resistance. Sensitivity of Different ALL Cell Lines to ASNase Treatment. Previous studies by Aslanian et al. (Aslanian et al., 2001) in the Kilberg laboratory suggested that there is a correlation between the ASNS expression and the ASNase resistance in ALL cells. However, clinical studies using microarray technology argue against this theory, as in th e ALL patients, the ASNS mRNA expression does not correlate with ASNase resistance, which was particularly obvious in those patients with TEL AML1 tranlocation. Although the TEL AML1 positive patients have higher ASNS mRNA expression than the TEL AML1 neg ative patients, they tend to have much better prognosis from ASNase treatment (Stams et al., 2003) To invest igate if the ASNS expression is correlated with ASNase resistance in cultured cells, I expanded my study to the following ALL cell lines: NALM6, a B cell derived TEL AML1 negative ALL cell line; REH, a B cell derived TEL AML1 positive ALL cell line; and MO LT 4P/ASNS, a cell line generated by Aslanian et al. (Aslanian et al., 2001) in the Kilberg laboratory with constitutive ASNS over expression following stable transduction with a retrovirus containing the cDNA of ASNS. To document the growth behavior of these ALL cells under a constant level of ASNase challenge, all the five ce ll lines were cultured in RPMI 1640 medium for 12 h, then transferred to culture medium containing 1 U/mL ASNase and incubated for 0 96 h (Figure 4 6). Total viable cell number and percent viability were analyzed at specific intervals during the 4 d perio d, using trypan blue assay (as described in Chapter 2). The sensitivity of MOLT 4P cells was confirmed by a halt in cell growth from 0 48 h of ASNase treatment and a rapid decrease in viable cell number after 48 h (Figure 4 6A). The cytotoxic effect of A SNase on the MOLT 4P cells took place as early as 12 h post administration (Figure 4 6B). In the contrast, neither of the B cell

PAGE 104

89 lines, NALM 6 and REH, exhibited substantial cell death in response to the 1 U/mL ASNase treatment (Figure 4 6B), but growth w as slower than that observed for the MOLT 4R cells (Figure 4 6A), illustrating the cytostatic nature of ASNase for these particular cell lines. The fifth cell line MOLT 4P/ASNS that constitutively expresses exogenous ASNS (Aslanian et al., 2001) showed a significant difference from the MOLT P cells in ASNase sensitivity, confirm ing that elevated expression of ASNS alone is sufficient to induce a more drug resistant phenotype (Aslanian et al., 2001) However, the MOLT 4P/ASNS cells were still more sensitive than the MOLT 4R, NALM6, and REH cells, as 1U/mL ASNase caused significant growth retardation (Figure 4 6A), and approximately 20% cell death after 48 h of treatment (Figure 4 6B). To determine the relative sensitivity of the five cell lines to ASNase, cells were incubated for 72 h in media containing a gradient concentration of ASNase, ranging from 0.01 to 5.0 U/mL. Cell growth (Figure 4 7) and viab ility (Figure 4 8) were analyzed by trypan blue assay. For a clearer demonstration, the data for the ASNase concentrations from 0.001 to 0.1 are shown in the inserts, whereas those from 0.1 to 5.0 are illustrated in the larger graphs. Given that the cell number to begin the 72 h incubation period was the same (5 x 10 4 /plate) for each cell line, when the cells were culture at 0 U/mL ASNase, the viable cell number counted at different times would represent their relative growth rates. As calculated from F igure 4 7, different ALL cell lines exhibited distinct growth rate under normal culture conditions, with an order of MOLT 4R and NALM 6 > MOLT 4P and MOLT 4P/ASNS >> REH. As shown in Figure 4 7, a low concentration (0.01 0.1 U/mL) of ASNase caused apparen t growth retardation even in cells with the most resistant phenotype (MOLT 4R), which further confirmed the cytostatic function of ASNase.

PAGE 105

90 In order to get a quantitative measurement of the ASNase sensitivity for different ALL cells, the IC 50 values, the concentration at which 50% of the cells are killed, were calculated based on the percentage cell viability at different ASNase concentration (Figure 4 8). The MOLT 4P cells were the most ASNase sensitive of all the cell lines tested with an IC 50 of 0.0005 U/mL, and a near to complete cell death at 0.02 U/mL (Figure 4 8). Conversely, the MOLT 4R cells represented the opposite end of the spectrum with less than 20% cell death, even at 5 U/mL ASNase, and had an IC 50 of 23 U/mL ASNase, which was beyond the ra nge of experimental concentration, and was purely calculated based on a mathematical trend line. The NALM 6, REH, and MOLT 4P/ASNS cells were intermediate in their sensitivity with IC 50 values of 2.7, 5.4, and 0.04 U/mL, respectively. It is important to note that the MOLT 4P/ASNS cells were more sensitive than either of the B cell derived lines, though it was much more resistant than the MOLT 4P cells. Analysis of the ASNS mRNA and Protein Expression in ALL Cell Lines To investigate if there is a correlat ion between the ASNase sensitivity and the ASNS expression in these ALL cell lines, I measured the ASNS mRNA expression during a time course of ASNase treatment (Figure 4 9). The MOLT 4P cells expressed a very low level of ASNS mRNA under normal culture c ondition. In the contrast, the NALM6, REH, and MOLT 4R cells expressed a much higher ASNS mRNA level at this basal condition, and the levels in these three cell lines were similar. The MOLT 4P/ASNS cells expressed an enormous level of ASNS mRNA, approxim ately four fold the level in MOLT 4R cells, which had the highest ASNS mRNA expression among the other four cell lines (Figure 4 9). Upon ASNase treatment, the ASNS mRNA expression was induced in all the cell lines. At 12 h, the ASNS mRNA expression in t he MOLT 4P, MOLT 4R, NALM6, and REH cells was at a comparable level (within 30% variation). The induction of ASNS mRNA in the MOLT 4P/ASNS cells was solely

PAGE 106

91 from the induction of the endogenous ASNS gene, as the promoter in the ASNS expression construct wa s constitutive, and not subject to the control by amino acid deprivation. The ASNS mRNA expression in MOLT 4R, NALM6, and REH cells was sustained throughout the entire time course. However the data points of the MOLT 4P cells were not available after 48 h of ASNase treatment, which was due to significant cell death after 24 h (Figure 4 8). As for the ASNS protein expression, the basal ASNS protein expression varied significantly among the five ALL cell populations, as shown by immunoblotting analysis (Fig ure 4 10A). The ASNS protein level in the MOLT 4P cells was almost undetectable at 0 and 12 h. An induction was observed only after 24 h of ASNase treatment, and that increase was modest. In the contrast, the MOLT 4P/ASNS cells exhibited a greater amou nt of ASNS protein expression, an expected result and consistent with the observation documented previously (Aslanian et al., 2001) Similar to the MOLT 4P cells, the ASNS protein level was also induced after 48 h in the MOLT 4P/ASNS cells, and this induction was due to the increased mRNA expression from the endogenous ASNS gene The MOLT 4R cells had the highest level of ASNS protein expression, whereas the B cell derived cell lines, NALM6 and REH, contained ASNS protein levels that were intermediate in amount between the MOLT 4P and MOLT 4R, with the NALM 6 cells containing mo re than the REH cells, but still less than the level in the MOLT 4 R cells (Figure 4 10A). In contrast to the obvious protein induction in the MOLT 4P and MOLT 4P/ASNS cells in response to acute ASNase treatment, the MOLT 4R, NALM 6, and REH cells exhibit ed little or no change in ASNS protein from 0 to 96 h (Figure 4 10B). In order to compare the ASNS protein content in the five cell populations, the bands from the immunoblotting analysis were quantified and plotted by setting the MOLT P value to 1.0 (Fig ure 4 10B). The highest value (15.7 4.7) for the induced ASNS protein content in the MOLT P

PAGE 107

92 cells (48 h) was still only equal to or less than the lowest basal value for the remaining cell types (REH cells, 20.3 4.6). These data show that there is a s ignificant lag between the time of ASNase exposure and the production of ASNS protein in the MOLT 4P cells. This also holds true for the other ALL cell lines, as no significant change in ASNS protein level was observed during ASNase treatment in the MOLT 4R, NALM6 or REH cells. This lag of protein induction is due to a lack of de novo protein synthesis, as the turn over rate of the ASNS protein is extremely low, which is evidenced by a sustained ASNS protein level in the MOLT 4P cells after 48 h of ASNase treatment, despite a nearly complete lethality in the cell population. Furthermore, even the highest ASNase induced ASNS protein expression (at 48 h) in the MOLT P cells is an amount just approaching that in the REH cells, which are still sensitive to th e drug (Figure 4 10). These results demonstrated a lack of correspondence between the ASNS mRNA and protein level in the ALL cell lines tested. One interpretation is that there is translational control of the ASNS mRNA such that the rate of ASNS protein synthesis does not reflect the abundance of the mRNA. For example, the MOLT 4P/ASNS cells express a high basal level of ASNS mRNA (Figure 4 9) as the result of stable integration of the CMV promoter driven ASNS expression construct. However, its protein production for ASNS is extremely low, such that the protein level is only moderately increased. Furthermore, in the MOLT 4R and the B cell lines, the ASNS mRNA was indeed induced by ASNase treatment, but their protein level barely exhibited any change (F igure 4 9 and Figure 4 10). These results support the interpretation that there is a widespread disconnection between ASNS mRNA and protein expression in the ALL cells as a result of translational control. In conjunction of the ASNase sensitivity data, t hese

PAGE 108

93 observations also confirmed the proposal that the ASNS mRNA level does not reflect ASNase sensitivity in ALL patients. Correlation of ASNase resistance and ASNS protein expression The fact that MOLT 4P cells express an extremely low level of ASNS pr otein, and are sensitive to ASNase treatment, prompted me to investigate the correlation between ASNase resistance and ASNS protein expression in the ALL cell lines. The IC 50 values obtained from the ASNase sensitivity assay were plotted against the relat ive ASNS mRNA (Figure 4 11A) and protein (Figure 4 11B) levels in the MOLT 4P, NALM6, and REH cells, the three naturally derived ALL cell lines. No correlation was found between the IC 50 value and the ASNS mRNA level (Figure 4 11A), but an apparent linear correlation was observed between the relative ASNS protein level and the IC 50 value, or in another words, the ASNase resistance (Figure 4 11B). Given the fact that ASNS is ubiquitously expressed in almost all tissues, the ASNase treatment has no major e ffect on normal tissue functions. To test if the ASNS is the only protective mechanism against ASNase treatment, Dr. Yuanxiang Pan in the Kilberg laboratory utilized an siRNA approach to knock down ASNS expression, and then test the sensitivity to ASNase treatment (Figure 4 12). Given the fact that lymphocytes, including established ALL cell lines, are difficult to transfect, human HepG2 hepatoma cells were used alternatively for the siRNA analysis. HepG2 cells contain a high level of basal ASNS protein and a high degree of ASNase resistance, with an estimated IC 50 of greater than 5 U/mL and therefore, they are an ideal target to test the effect of ASNS knock down on ASNase resistance. For the experiment, HepG2 cells were treated with either control (GFP ) siRNA or siRNA against ASNS for 48 h, then transferred to either normal MEM medium or MEM containing 2U/mL of ASNase and incubated for additional 48 h. ASNS mRNA content was measured by RT qPCR and sensitivity to the drug was monitored by apoptosis usin g Annexin V binding, as described in Chapter 2.

PAGE 109

94 The HepG2 cells were relatively resistant to ASNase as illustrated by minimal apoptosis after delivery of the control siRNA. The dependence of cell viability on ASNS expression, even in the presence of the asparagine in the culture medium, was illustrated by the 40% cell death by the ASNS siRNA in the absence of added ASNase (Figure 4 12). Cells treated with the combination of ASNase and the ASNS specific siRNA exhibited a decrease in viable cells by about 70%. These results demonstrate that reducing ASNS expression can lead to enhanced sensitivity to ASNase and are consistent with the relative correlation between ASNS protein level and ASNase sensitivity in the five ALL cell lines. ASNS mRNA and protein ex pression in patient samples From the studies described above, I concluded that it is important to measure the ASNS protein expression in ALL patient samples in order to predict the prognosis from ASNase therapy. To conduct an example study, Dr. Mi Zhou fr laboratory, utilized RT PCR to examine ASNS mRNA levels in a panel of 40 cryo preserved ALL specimens that had been obtained at the time of initial diagnosis. Although the mRNA varied between patients, none of the p atient samples had a level of ASNS mRNA comparable to that expressed in MOLT 4R cells (data not shown). ASNS protein levels were measured in a subset of the patient samples that represented the spectrum of ASNS mRNA content, and according to their ASNS mR NA expression, these patient samples were numbered 1 to 10, with #1 having the highest and #10 having the lowest ASNS mRNA level (Figure 4 13A). Immunoblotting analysis revealed that except patient #3, 9 of 10 samples had non detectable ASNS protein expre ssion (Figure 4 13B). The protein level in patient #3 was higher than that in the MOLT 4P cells, but still much lower than that in the MOLT 4R cells. Also noteworhty, is that the ASNS mRNA levels in half of the patient samples were higher than that in th e MOLT 4P cells, however, most of these patients expressed even less ASNS protein. These results

PAGE 110

95 suggest that most ALL patients exhibit low ASNS protein expression due an unknown mechanism that translationally suppresses the ASNS protein synthesis, and th erefore, they are sensitive to ASNase therapy. However, once this mechanism is dismissed, the resistance to ASNase is likely to be irreversible. Conclusions and Discussions The major effect known for ASNase treatment is depletion of serum asparagine, alth ough the intrinsic glutaminase activity of ASNase may also cause a minor decrease of the glutamine level. ASNS catalyzes the synthesis of asparagine, therefore directly counteracts the effect of ASNase, and consequently, the up regulation of ASNS in the A LL cells was thought to be the mechanism for ASNase resistance. However, after years of studies, controversial results were obtained and this basic question, that is whether or not ASNase resistance is correlated with ASNS up regulation, still remained un answered. Previous studies using cultured ALL cell lines showed that the parental human ALL cell line MOLT 4P expresses ASNS mRNA and protein at extremely low levels, and therefore like most patients, is sensitive to ASNase treatment. This sensitivity c an be overcome by exogenous over expression of ASNS in the cells (Aslanian e t al., 2001) Moreover, prolonged treatment of the MOLT 4P cells with ASNase induces the resistant phenotype and led to the generation of an ASNase resistant cell line, MOLT 4R, which has much higher level of ASNS mRNA and protein expression (Hutson et al ., 1997) These results suggest that there is a correlation between ASNase resistance and up regulation of the ASNS gene. However, clinical studies using patient samples generated controversial data. Microarray studies showed that ALL patients have elev ated ASNS mRNA expression upon ASNase treatment regardless of incident of relapse (Holleman et al., 2004) It was also shown that the TEL AML1 positive patients have better prognosis from ASNase treatment despite higher ASNS

PAGE 111

96 mRNA expression compared to TEL AML1 negative patients (Krejci et al., 2004) The current study coordinates those observations by investigating the inter connection between ASNS mRNA expression, ASNS protein expression and ASNase resistance. Expression profiling assay first confirmed that the parental MOLT 4 cells hav e a lower level of ASNS mRNA and protein expression compared to the resistant cells. Mechanistic studies demonstrated that the ASNS gene is up regulated through the amino acid response (AAR) pathway by ASNase treatment in both parental and resistant MOLT 4 cells, and the MOLT 4P cells are able to express the ASNS mRNA to a level comparable to that in the MOLT 4R cells. Despite the high level of ASNS mRNA in the MOLT 4P cells, there is an apparent gap between the ASNS mRNA induction and protein expression during ASNase treatment, as the induction of ASNS protein was not observed until 24 48 h post administration. Moreover, the MOLT 4P/ASNS cells, which have an identical genetic background as the MOLT 4P cells but constitutively express a high level of ASNS mRNA, do not have comparable ASNS protein content. These results indicate that either there is a defect or a control mechanism that suppresses the translation of ASNS mRNA in the MOLT 4P or even all the ALL cells. This proposal is supported by a lack of protein induction during ASNase treatment in the other ALL cell lines, although those cells have elevated ASNS mRNA expression. Although the ASNS protein expression is not induced in the B cell derived ALL cell lines REH and NALM 6, they do have a basal l evel of ASNS mRNA and protein that is considerably greater than the T cell derived parental MOLT 4 cells. The REH, NALM 6, and MOLT 4P/ASNS cells have a protein content that is intermediate between the parental and resistant MOLT 4 cells. Consistent with these observations, only the MOLT 4P cells exhibit a significant amount of cell death during an ASNase challenge, whereas the ASNase action is primarily

PAGE 112

97 cytostatic, not cytotoxic, against the REH, NALM 6, MOLT 4P/ASNS, and MOLT 4R cells. By comparing the ASNS protein and mRNA expression in the five independent cell lines, I observed a non linear correlation between ASNS mRNA and ASNS protein content indicating that there are cell specific differences in ASNS mRNA translation efficiencies among these cell populations. Microarray analysis in patient samples had led to the interpretation that ASNS mRNA expression does not correlate with ASNase sensitivity (Holleman et al., 2004; Krejci et al., 2004) This is consistent with the present observations that neither ASNS protein expression nor ASNase sensitivity necessarily correlates with ASNS mRNA content. An ASNase sensitivity assay illustrated a correlation, although not entirely linear, between the ASNase sensitivity and the ASNS protein abundance in the ALL cell lines studied. Furthermore, the current study not only demonstrates the correlation between ASNS protein expression and ASNS sensitivity, but also suggests that the enzyme ASNS may be the only protective mechanism against ASNase in all tissues, as siRNA mediated knock down of ASNS in the HepG2 human hepatoma cells resulte d in an ASNase sensitive phenotype. Given that most newly diagnosed ALL children are ASNase sensitive, they are most likely reflected by the MOLT 4P cells with little or no ASNS protein expression. This is at least partially demonstrated by expression pro filing in some ALL patient samples. This is a positive message for clinical ASNase therapy, because even though primary ALL cells have been shown to up regulate ASNS mRNA in response to ASNase treatment in vitro ALL patients may be sensitive to ASNase be cause of the significant delay and deficiency of ASNS protein expression after ASNase treatment, as observed in the MOLT 4P cells. Unfortunately, once resistant happens in the patients, it is likely to be an irreversible process as the repressive mechanis m on

PAGE 113

98 the ASNS protein production is permanently removed. This is demonstrated by the MOLT 4R cells, which retained its resistant phenotype even after 6 weeks of drug withdraw. Therefore, it is recommended that a different therapeutic approach be employed for the treatment of relapse patients from ASNase therapy. Expression of the fusion protein TEL AML1 has been shown to be a positive factor for ASNase therapy (Krejci et al., 2004; Stams et al., 2003) despite the fact that the level of ASNS mRNA content in these patients can be equal to or greater than that in the TEL AML1 negative patients who are less sensitive. Once again, A SNS protein content has not been monitored in these two populations, but it is possible that translational control of ASNS synthesis or ASNS protein stability is differentially affected by the TEL AML1 fusion and the ASNS mRNA does not provide an accurate measurement of ASNS activity, and consequently, ASNase sensitivity. Collectively, these results demonstrated that the ASNS mRNA level should not be used as an indicator for ASNS protein expression, and neither should it be used as a marker for ASNase se nsitivity in clinical treatment for ALL patients. However, the ASNS protein content/activity, if can be accurately measured in ALL patients, can serve as an indicator for the prognosis from ASNase treatment, or for the prediction of relapse due to ASNase resistance. The conclusion from my work has received recent confirmation by the report that the ASNase sensitivity of 60 NCI ovarian cancer cell lines correlates with ASNS protein, not ASNS mRNA (Lorenzi et al., 2008)

PAGE 114

99 Figure 4 1. Expression of ASNS, ATF4, and ATF3 in the parental and resistant MOLT 4 cells. The MOLT 4P cells were cultured in RPMI 1640 or RPMI 1640 containing 1U/mL ASNase for 12 h. MOLT 4R cells were maintained in RPMI 1640 containing 1U/mL ASNase. Total RNA was collected and subjected to qRT PCR analysis for ASNS, ATF4, ATF3, and GAPDH mRNA contents (Panel A). The graph illustrates the means S.E.M. for three independent experiments. Whole cell extracts were also c ollected and subjected to immunoblotting analysis for the ASNS, ATF4, and ATF3 protein content (Panel B). The blot shown is representative of two independent experiments.

PAGE 115

100 Figure 4 2. A parental and resistant MOLT 4 cells. MOLT 4P cells were cultured in RPMI 1640 or RPMI 1640 containing 1U/mL ASNase for 12 h. In order to study the acute response to ASNase treatme nt, MOLT 4R cells were first cultured in RPMI 1640 without ASNase for 24 h, and then transferred to RPMI 1640 containing 1U/mL ASNase for 12 h. After 12 h of ASNase treatment, both MOLT 4R and MOLT 4P cells were transferred to RPMI 1640 media, and culture d for 12 h (12h + withdraw) Chromatin immunoprecipitation (ChIP) assays were performed as described in Chpater 2. DNA fragments were immunoprecipitated with Pol II, ATF4, ATF3, rabbit IgG antibodies and the enrichment of these proteins at the ASNS promoter was analyzed by qPCR, using primer sets to amplify a region (nt 87 to 22) containing the AARE element. Data were plotted as the ratio to the value obtaine d with a 1:20 dilution of input DNA. The graph illustrates the means S.E.M. for three independent experiments. An asterisk indicates that the value is significantly different from the no ASNase control (p < 0.05).

PAGE 116

101 Figure 4 3. Association of GTFs an d acetylated histones with the ASNS promoter in the parental and resistant MOLT 4 cells. MOLT 4P cells were cultured in RPMI 1640 or RPMI 1640 containing 1U/mL ASNase for 12 h. In order to study the acute response to ASNase treatment, MOLT 4R cells were first cultured in RPMI 1640 without ASNase for 24 h, and then transferred to RPMI 1640 containing 1U/mL ASNase for 12 h. Chromatin immunoprecipitation (ChIP) assays were performed as described in Chpater 2. DNA fragments were immunoprecipitated with Acet yl H4, Acetyl H3, TBP, TAF1, TFIIB, and normal rabbit IgG antibodies, and the enrichment of these proteins at the ASNS promoter was analyzed by qPCR, using primer sets to amplify a region (nt 87 to 22) containing the AARE element. Data were plotted as t he ratio to the value obtained with a 1:20 dilution of input DNA. The graph illustrates the means S.E.M. for three independent experiments. An asterisk indicates that the value is significantly different from the no ASNase control (p < 0.05).

PAGE 117

102 Figure 4 4. Characterization of the ASNS expression in the ASNase resistant MOLT 4 cells after a prolonged drug withdraw. To investigate the reversibility of the resistant phenotype of the MOLT 4R cells, the cells were cultured in RPMI 1640 without ASNase for 6 weeks, frozen and named MOLT 4R0. The MOLT 4P and MOLT 4R0 cells were cultured in RPMI 1640 or RPMI 1640 containing 1U/mL ASNase for 12 h. In order to study the acute response to ASNase treatment, MOLT 4R cells were first cultured in RPMI 1640 without ASNase for 24 h, and then transferred to RPMI 1640 containing 1U/mL ASNase for 12 h. Total RNA was collected and subjected to qRT PCR analysis for ASNS and GAPDH mRNA contents (Panel A). The graph illustrates the means S.E.M. for three independent expe riments. Whole cell extracts were also collected and subjected to immunoblotting analysis for the ASNS, ATF4, or actin protein content (Panel B). The blot shown is representative of two independent experiments.

PAGE 118

103 Figure 4 5. Association of Pol II, ATF4 ATF3, and C/E MOLT 4P and MOLT 4R0 cells. The MOLT 4P and MOLT 4R0 cells were cultured in RPMI 1640 media. Chromatin immunoprecipitation (ChIP) assays were performed as described in Chpater 2. DNA fragments were immunoprecipitated wi th Pol II, ATF4, ATF3, antibodies, and the enrichment of these proteins at the ASNS promoter was analyzed by qPCR, using primer sets to amplify a region (nt 87 to 22) containing the AARE element. Data were plotted as the ra tio to the value obtained with a 1:20 dilution of input DNA. The graph illustrates the means S.E.M. for three independent experiments. An asterisk indicates that the value is significantly different from that in the MOLT 4P cells (p < 0.05).

PAGE 119

104 Figure 4 6. The ALL cell lines show differential growth patterns and sensitivities to ASNase treatment. MOLT 4P, MOLT 4R, MOLT 4P/ASNS, NALM 6 and REH cells fed with complete RPMI 1640 media for 24 h were placed at 0.5 X 10 6 /mL and treated with 1U/mL of ASNase for indicated times. The medium was changed and cells were treated with 1U/mL of ASNase 36 h post treatment. Total cell numbers and viable cell numbers per ml were determined using a trypan blue assay. (Panel A) Viable cell numbers were plotted versus time of ASNase treatment for each cell line. (Panel B) Percentage cell viabilities were plotted versus time of ASNase treatment for each cell line. The results shown are the averages standard deviations from three separate experiments.

PAGE 120

105 Figure 4 7. Growth inhibition effect of different concentration of ASNase on ALL cell lines. MOLT 4P, MOLT 4R, MOLT 4P/ASNS, NALM 6 and REH cells fed with complete RPMI 1640 media for 24 h were placed at 5 X 10 4 /mL in 100 l medium in 96 well plate and treated with different concentration of ASNase as indicated for 72 h. Viable cell numbers were determined using a trypan blue assay and plotted versus ASNase concentration. For clarity, curves representing treatments with no more than 0.1 U/mL of ASNase were shown a s small panels on the top right corner of each figure. The results shown are the averages standard deviations from three separate experiments.

PAGE 121

106 Figure 4 8. Effect of ASNase concentration on cell death of ALL cell lines. MOLT 4P, MOLT 4R, MOLT 4P/ASN S, NALM 6 and REH cells fed with complete RPMI 1640 media for 24 h were placed at 5 X 10 4 /mL in 100 L medium in 96 well plate and treated with different concentration of ASNase as indicated for 72 h. Both total cell numbers and viable cell numbers were counted using a trypan blue assay and cell viabilities represented by the ratios of viable cell number to total cell number were plotted versus ASNase concentration. For clarity, curves representing treatments with no more than 0.1 U/mL of ASNase were sho wn as small panels on the top right corner of each panel. The results shown are the averages standard deviations from three separate experiments.

PAGE 122

107 Figure 4 9. mRNA expression of ASNS in ALL cell lines during ASNase treatment. MOLT 4P, MOLT 4R, MOLT 4P/ASNS, NALM 6 and REH cells fed with complete RPMI 1640 media for 24 h were placed at 0.5 X 10 6 /mL and treated with 1U/mL of ASNase for indicated times. The medium was changed and cells were treated with 1U/mL of ASNase 36 h post treatment. RNA was co llected at each time point and subjected to quantitative RT PCR using primers specific for ASNS and GAPDH. ASNS mRNA levels were normalized to the GAPDH level and the fold induction relative to MOLT 4P cells at time 0 were plotted versus time of ASNase tr eatment. The results shown are the averages standard deviations from three separate experiments.

PAGE 123

108 Figure 4 10. Protein expression of ASNS in ALL cell lines during ASNase treatment. MOLT 4P, MOLT 4R, MOLT 4P/ASNS, NALM 6 and REH cells fed with comple te RPMI 1640 media for 24 h were placed at 0.5 X 10 6 /mL and treated with 1U/mL of ASNase for indicated times. The medium was changed and cells were treated with 1U/mL ASNase for 36 h. Whole cell extracts were collected and subjected to immunoblotting fo r ASNS or actin (Panel A) and the results were quantified by densitometry (Panel B). ASNS levels were normalized to actin levels and shown as fold induction relative to the MOLT 4P cells at time 0. The results shown are the averages standard deviations from three separate experiments.

PAGE 124

109 Figure 4 11. Relationship between ASNase IC 50 and ASNS protein or mRNA expression. For the three parental cell lines (MOLT 4P, NALM 6, and REH), the estimated IC 50 values for ASNase sensitivity from the data of Figure 4 8 were compared to the relative amount of ASNS mRNA (Panel A) or ASNS protein (Panel B), for which values were acquired from Figures 4 9 and Figure 4 10.

PAGE 125

110 Figure 4 12. Effect of siRNA mediated ASNS knock down on the sensitivity to ASNase induced apo ptosis in HepG2 cells. An siRNA targeting ASNS was introduced into HepG2 cells by retrovirus as described in the Chapter 2. After continued culture for two days, the cells were incubated in medium containing 2 IU/mL of ASNase for 48 h and then analyzed f or apoptosis as described in the Chapter 2. The numbers in each quadrant indicate the percentage of live cells from a total of 10,000 cells analyzed and the data presented are representative of three independent experiments. (Data courtesy of Dr. Yuanxian g Pan)

PAGE 126

111 Figure 4 13. Protein expression of ASNS in childhood ALL primary patient samples. Whole cell extracts were prepared from ALL patient samples obtained at the time of initial diagnosis. ASNS or actin protein levels. As a reference for relative abundance, cell extract from MOLT 4P and MOLT 4R cells was also included. (Data courtesy of Dr. Mi Zhou)

PAGE 127

112 CHAPTER 5 REGULATION OF THE FOXA FAMILY OF G ENES BY NUTRIENT STR ESS AND THEIR INVOLVEMENT IN THE C ONTROL OF AMINO ACID RESPONSIVE GENES Introduction Liver is a critical organ for metabolic control. A complex gene network functions delicately to keep the homeostasis of a numbe r of nutritional molecules, including glucose and amino acids. One group of such genes is the FOXA family, which was previously known as the hepatic nuclear factor 3 (HNF3) family (Kaestner et al., 2000) The F OXA protein family was initially discovered in liver nuclear extract because of th eir ability to bind to the promoters of the antitrypsin (Serpina1), and albumin (Alb1) genes (Herbst et al., 1991; Lai et al., 1990) Th erefore, these genes were originally named hepatocyte nuclear factor 3 (HNF 3) helix proteins are termed FOX proteins, for forkhead box (Kaestner et al., 2000) As a result, the and FOXA3 (HNF3G), respectively, in humans, and Foxa1 (Hnf3a), Foxa2 (Hnf3b) and Foxa3 (Hnf3g) in mice (Kaestner et al., 2000) All the FOXA proteins contain a forkhead motif, which is flanked by sequences required for nuclear localization, in the middle of the protein. The forkhead motifs of the FOXA proteins share 95% similarity in sequence, and therefore are proposed to bind to the same group of genes. The sequence homology outside the forkhead box is relatively weaker. However, the N and C termini of the proteins are highly conserved (Friedman and Kaestner, 2006) Function of FOXA Proteins in Activating Transcription More than 10 0 genes were shown to have HNF3 binding sites, these genes are primarily expressed in the liver, pancreas, intestine and lung, as well as during early embryogenesis (Cereghini, 1996) Two mechanisms have been proposed for the transcriptional activation

PAGE 128

113 activities of the FOXA proteins. The first mechanism is involved in direct activation of the target genes throu gh their trans activation domains (Pani et al., 1992) However, the molecular mechanism for this process has not been fully unveiled, as no co activators have been identified for these trans activation domains. The second mechanism suggests that the FOXA proteins have chromatin remodeling like activities. Given the fact that the forkhead box structure is similar to that of the linker histone H1 (Clark et al., 199 3) they may compete with histone H1 for DNA binding. Indeed, Cirillo and colleagues (Cirillo et al., 2002; Holmqvist et al., 2005) demonstrated that the FOXA proteins are able to open highly compacted chromatin in vitro in a ma nner not requiring the SWI/SNF chromatin remodeling complex. They also proposed that the mechanism of this activity is due to the ability of the C terminal domains of FOXA proteins to interact with the core histones H3 and H4 (Cirillo et al., 2002) Based on their chromatin remodeling like activities, the FOXA proteins have been pro transcription factors, which function to displace the linker histones from compacted chromatin and facilitate the binding of other transcription factors (Friedman and Kaestner, 2006) Function of FOXA Proteins in Development and Organogenesis The FOXA proteins are expressed during embryonic development, (Ang et al., 1993; Monaghan et al., 1993; Sasak i and Hogan, 1993) and play important roles in organogenesis. Foxa1 mutant mice develop severe hypoglycemia and die shortly after birth (Shih et al., 1999) Foxa2 mutant mice have severe defects in liver and pancreas development, and d ie during early embryo genesis. Interestingly, deletion of Foxa2 in hepatocytes during late fetal development has essentially no effect on metabolism and gene expression (Sund et al., 2000) Unlike Foxa1 and Foxa2, although the expression of several Foxa target genes were reduced in Foxa3 / mice, they did not show a discernible phenotype (Kaestner et al., 1998) which can be partly explained by up regulation of Foxa1 and Foxa2 (Kaestner et al., 1998) However, Foxa3 has been shown

PAGE 129

114 to be the dominant regulator of GLUT2 expression in the liver, an d is required for the maintenance of glucose homeostasis during a prolonged fast (Shen et al., 2001) Foxa1 and Foxa2 are required for the induction of liver specification by inductive signals (Lee et al., 2005) The Foxa1/Foxa2 deficient mouse is the only known model of a completely vertebrate. The Foxa proteins are also important for the pancreas as Foxa1 and Foxa2 are expressed in foregut endoderm, from which the pancreas develops, prior to the onset of pancreatic specification (Monaghan et al., 1993) Furthermore, Foxa1 and Foxa2 regulate several pancreas specific genes, i ncluding Pdx 1 (Ipf1), which is required for expansion of the pancreatic buds and maintenance of differentiated pancreatic cell types (Ashizawa et al., 2004) Although the pancreas develops normally in neonates with Foxa1/Foxa3 or Foxa2/Foxa3 double knockout, presumably due to the functional redundancy of the Foxa proteins, the pancreas function in these mutant mice are not normal (Friedman and Kaestner, 2006; Lee et al., 2005) The Foxa family proteins are also important for the prostate development and function. They regulate a numbe r of prostate specific genes, including prostate specific antigen (Klkb1) and probasin (Pbsn) (Gao et al., 2003) Foxa1 / mouse prostate lacks normal duct structures and the epithelium is arrested in a poorly differentiated embryonic stage (Mirosevich et al., 2005) and neither Foxa2 nor Foxa3 can substitute for Foxa1 function. Furthermore, given the distinct phenotypes of the male reproductive system in conditional Foxa2 mutants and Foxa3 / mice, the Foxa proteins may have specific functions in prostate development. Function of FOXA Proteins in Hormone Control of Gene Expression and Glucose Metabolism The maintenance of glucose homeostasis is crucial to mammalian survival, and disruptions in this homeosta tic status are associated with a number of diseases, such as diabetes. The liver is the major contributor to the control of glucose metabolism. It stores glucose by converting it to

PAGE 130

115 glycogen or triglycerides when there is an excess amount of glucose in t he blood, such as after a carbohydrate rich meal. Conversely, under the conditions of hypoglycemia, the liver activates gluconeogenesis to produce glucose from, amino acids and glycerol. These two processes are orchestrated by insulin, glucagons, or gluc ocorticoids. The FOXA proteins play a central role in coordinating the hormonal and metabolic control of gene expression in the liver. The high homology in their protein sequences and the presumable functional redundancy are consistent with the critical role of the FOXA proteins in protecting mammalians from hypoglycemia (Shen et al., 2001; Lee et al., 2005; Kaestner et al., 1998; Lantz et al., 2004) The Foxa proteins are involved in direct regulation of hormones involved in the control of glucose homeostasis. For example, Foxa 1 regulates the expression of the proglucagon gene (Kaestner et al., 1999) and mice lacking Foxa1 have severe hypoglycemia. Furthe rmore, both Foxa2 (Kaestner et al., 1999) and Foxa3 also bind to and activate the rat proglucagon gene promoter, though Foxa3 is no t essential for proglucagon gene expression (Liu et al., 2002) Many of the Foxa target genes in the liver are involved in glucose homeostasis, particularly the response to fasting. These genes include the gluconeogenic enzymes phosphoenolpyruvate c arboxykinase (Pepck), glucose 6 phosphatase (G6pc), and tyrosine aminotransferase (Tat). Both Foxa2 and Foxa3 have been shown to regulate these genes in the liver. Tissue specific knock down of Foxa2 in hepatocytes leads to failure of full activation of Pepck, Tat, and Igfbp1 in response to fasting (Zhang et al., 2005) In Foxa3 / mice, the expression of gluconeogenic enzymes is diminished as is the expression of Glut2, which is a transport protein that exports glucose from hepatocytes. As a result, Foxa3 / mice display hypo glycemia during a prolonged fast (Shen et al., 2001; Kaestner et al., 1998)

PAGE 131

116 Regulation of the FOXA Genes Compared to their functions in regulating other genes, the information on the regulation of the FOXA gene themselves is relatively limited. FOXA1 was shown to be up regulated in breast cancer samples (Laganiere et al., 2005) through an unknown mechanism. The transcriptional activity of Foxa2 is blocked by treatment with insulin, which results from the phosphorylation of threonine 156 by the ins ulin activated kinase AKT2/ PKB. This phosphorylation leads to nuclear exclusion of Foxa2 (Wolfrum et al., 2003; Wolfrum et al., 200 4) FOXA3 is specifically expressed in organs derived from the endoderm such as liver, gut and pancreas. This tissue specific expression involves the liver specific transcription factor FOXA2, and possibly other transcription factors (Hiemisch et al., 1997) The three Foxa family members seem to form a transcription factor hierarchy with F oxa2 at the top and Foxa3 at the bottom. This is evidenced by the fact that during the formation of endoderm, Foxa2 is activated first, followed by Foxa1 and then Foxa3 (Ang et al., 1993; Monaghan et al., 1993) Furthermore, the expression of FOXA1 is eliminated in a visceral endoderm derived from FOXA2 / embryoid bodies (Duncan et al. 1998) and Foxa1 directly regulates Foxa3 expression by binding to a 3' region of the Foxa3 gene (Boj et al., 2001) Although mouse Foxa proteins are well known for their function in metabolic control of gene expression, the information on their regulation by nutrient stresses is limited. The expression of Foxa3 was shown to be induced in the mice deprived of protein (Imae et al., 2000) However, no simila r study has been done in human, nor is there a systematic study of the regulation of all three FOXA genes by nutrient stresses in either mouse or human. In the present study, the regulation of any of the FOXA family genes by nutrient stresses, including a mino acid deprivation and ER stress, was investigated. The potential role of FOXA proteins in the AAR pathway was also studied.

PAGE 132

117 Results Regulation of the FOXA Genes by Amino Acid Limitation and ER Stress The FOXA proteins are winged helix transcription fa ctors, featured by a winged helix DNA binding domain on their N termini (Figure 5 1). The three FOXA proteins, FOXA1, 2, and 3 share high similarities in their amino acid sequences, especially in their DNA binding domains which have more than 90% identity (Figure 5 1). Due to this high homology, the FOXA proteins were proposed to have a high degree functional redundancy (Kaestner et al., 2000) However, the fact that the human FOXA genes are located on different chromosom es, with FOXA1 on 14q12 q13, FOXA2 on 20p11 and FOXA3 on 19q13.2 q13.4 (Mincheva et al., 1997) raised the possibility that these genes are differentially regulated. Imae et al. (Imae et al., 2000) reported that the Foxa3 gene is up regulated in the mice deprived of protein. Given the function of Foxa3 in regulating glucose homeostasis, we decided to investigate the regulation of FOXA3 as well as other FOXA family member s by nutrient limitation of HepG2 human hepatoma cells, and their potential roles in the nutrient deprivation induced gene regulation. Given the fact that the FOXA genes are primarily expressed in liver, pancreas, and GI track, the HepG2 cells were used t o determine if the expression of FOXA family genes is regulated by amino acid limitation or ER stress. The cells were cultured in MEM, MEM containing 2mM HisOH, or MEM containing 300nM Tg for 8h, total RNA was collected and subjected to qRT PCR to measure the steady state mRNA level of FOXA1, FOXA2, and FOXA3. As shown in Figure 5 2A, the mRNA expression of FOXA1 was decreased by 25% following HisOH treatment and by 60% following Tg treatment. The FOXA2 mRNA level was induced by both HisOH and Tg treatme nt, with approximately 2.5 fold induction in HisOH treated cells and 1.5 fold induction in Tg treated cells. FOXA3 mRNA was significantly induced to about 6 fold by HisOH treatment, but stayed largely unchanged in Tg

PAGE 133

118 treated cells. To investigate if the expression of the FOXA genes are also regulated in normal liver cells, primary human hepatocytes were cultured in MEM, MEM containing 2mM HisOH, or MEM containing 300nM Tg for 8h, and the mRNA expression of FOXA1, 2, and 3 were measured by qRT PCR. As sho wn in Figure 5 2B, FOXA3 was induced by approximately 1.5 fold by amino acid deprivation, whereas FOXA1 and FOXA2 remained largely unchanged. The small increase in FOXA3 expression can be explained by the fact that the primary hepatocytes do not divide in culturing medium, therefore are less sensitive to nutrient deprivation. However, the increase in mRNA expression indicates that FOXA3 may play a role during amino acid deprivation both in cultured cells and normal liver tissue. To investigate the regula tion of the FOXA genes in vivo by protein limitation, mice at day 5 of pregnancy were pair fed with normal protein diet (19.39% protein) or with low protein diet (8% protein) until day 18.5 of gestation. The mothers were sacrificed and liver collected for RNA isolation. The mRNA content of FOXA1, FOXA2, FOXA3, and GAPDH was analyzed by qRT PCR. As a control, the mRNA content of ASNS and SNAT2, which are induced by amino acid deprivation in cultured cells (Kilberg et al., 2005) was also in vestigated. As shown in Figure 5 3, the ASNS expression was dramatically (~20 fold) induced in the mice fed with low protein diet (LPD) compared to that in the mice fed with normal protein diet (NPD). The expression of SNAT2 gene was also moderately induce d, to about 1.5 fold. As for the FOXA genes, the FOXA1 expression was not changed between the NPD and LPD mice. However, both FOXA2 and FOXA3 were slightly induced (both to ~1.6 fold) in the LPD mice compared to the NPD mice (Figure 5 3). Collectively, the se results demonstrate that FOXA2 and FOXA3 are induced by amino acid or protein limitation in both cultured cells and whole animals.

PAGE 134

119 To address the expression pattern of the FOXA genes during a time course of amino acid deprivation or ER stress, HepG2 ce lls were cultured in MEM, MEM containing 2mM HisOH, or MEM containing 300nM Tg for 0 24 h, and the steady state mRNA level of FOXA1, FOXA2, and FOXA3 was measured by qRT PCR. As shown in Figure 5 4, the FOXA1 mRNA level was decreased rapidly from 0 4 h fo llowing either HisOH (35% decrease) or Tg (50% decrease) treatment and exhibited little change for the rest of the time (Figure 5 4A). Interestingly, the FOXA1 mRNA level also decreased over time in the cells cultured in MEM medium and to a similar level as that in the HisOH treated cells at 24 h (Figure 5 4A). The mRNA level of FOXA2 stayed unchanged within 4 h following HisOH or Tg treatment, but was induced abruptly from 4 to 8 h, the increasing trend persisted throughout the entire time course in HisO H treated cells, but stopped at 8 h in Tg treated cells (Figure 5 4B). The delayed induction of FOXA3 mRNA level was rapidly induced within 2 h following HisOH trea tment, peaked at 16 h, and gradually went down from 16 to 24 h. In the contrast, Tg treatment hardly caused any change of the FOXA3 mRNA level (Figure 5 4C). These data demonstrated that not only are the three FOXA genes differentially regulated by nutri ent stresses, including amino acid deprivation and ER stress, but they also have different expression kinetics during stress responses. Generally speaking, there are two mechanisms accounting for the mRNA induction of a gene, the increase of de novo mRNA synthesis (transcription), mRNA stabilization, or a combination of both. For example, the mRNA induction of the ASNS gene by amino acid limitation is due to increased transcription (Chen et al., 2004) whereas the induction of p21 and p27 is mainly due to increased mRNA stability (Leung Pineda et al., 2004) To determin e which mechanism is responsible for the increase of FOXA2 and FOXA3 mRNA level after HisOH

PAGE 135

120 treatment, HepG2 cells were first cultured in MEM medium containing 2mM HisOH for 12 h, so lls were then (ActD) to block de novo transcription. Total RNA was collected at different intervals from 0 9 h, the mRNA content of FOXA2 and FOXA3 was measured and plotted again time to estimate the turnover rate of their mRNA under normal or the amino acid deprived condition (Figure 5 5A and Figure 5 5B). As a control, the turnover rate of p21 mRNA was also investigated (Figure 5 5C). The turnover rate of both FOXA2 and FOXA3 mRNA was slightly decreased following HisOH treatment, whereas the p21 mRNA was significantly stabilized, which is consistent with a previous observation (Leung Pineda et al., 2004) This result indicates that mRNA stabilization may contribute to the increase of FOXA2 and FOXA3 mRNA during amino acid deprivation. To determine whether the slight increase in mRNA half life accounts for the induction of mRNA observed in previous experiments (Figure 5 4) for FOXA2 and FOXA3 HepG2 cells were treated with HisOH in the presence or absence of the transcription inhibitor actinomycin D (Figure 5). The FOXA2 (Figure 5A) and FOXA3 (Figure 5B) mRNA content was induced by HisOH treatment, as expected from the previous experiments. However, in the presence of actinomycin D, the basal level of mRNA was significantly reduced and the HisOH ind uced mRNA expression for both genes was blocked. These results demonstrate that the increased expression of FOXA2 and FOXA3 mRNA following activation of the AAR pathway requires de novo transcription, and that there may also be a minor contribution by mRNA stabilization. To investigate the protein expression patterns of the FOXA genes during amino acid deprivation or ER stress, HepG2 cells were cultured in MEM, MEM containing 2mM HisOH, or MEM containing 300nM Tg for 0 24 h. Total cellular protein was col lected and

PAGE 136

121 subjected to immunoblotting (Figure 5 7A). Quantified data from film were plotted against times, as shown in Figure 5 7B D. Consistent with the mRNA expression pattern, the FOXA1 protein level was decreased following either amino acid deprivat ion or ER stress (Figure 5 7B). Although the FOXA2 mRNA level was induced by both amino acid deprivation and ER stress, the protein level remained largely unchanged during amino acid deprivation, and was slightly induced (2 fold) during ER stress (Figure 5 7C). The protein level of FOXA3 was increased by about 2 to 2.5 fold during amino acid deprivation from 8 24 h, though its mRNA level was induced by around 6 to 8 fold (Figure 5 4). The FOXA3 mRNA level was unchanged during ER stress (Figure 5 4), but the protein content was slightly decreased (Figure 5 7D). The lack of correlation between the mRNA and protein expression of the FOXA genes can be explained by (Hershey, 1991) The increase of FOXA2 mRNA synthesis is more likely to maintain rather than raising its protein level during amino acid deprivation, and therefore, keep normal cellular functions. However, the increase of both mRNA and protein suggested an important role of FOXA3 in regulating target genes during amino acid deprivation. Amino Acid Deprivation During amino acid limitation, a number of genes are induced through the amino acid response (AAR) pathway. As one of the most important terminal regula tors, ATF4 is involved in the regulation of all the genes containing an AARE element, which is also known as C/EBP ATF composite site (Fawcett et al., 1999) Besides ATF4, the ATF5 protein expression is also induced by amino acid limitation, through the same mechanism by which ATF4 is induced (Vattem and Wek, 2004) ATF5 wa s shown to play similar function as ATF4 on some AARE containing genes such as ASNS, in vitro (Al et al., 2005) To investigate if the induction of

PAGE 137

122 FOXA2 and FOXA3 by amino acid deprivation is dependent on ATF4 and/or ATF5, an siRNA approach was employed to knock down ATF4 and/or ATF5 expression in human HepG2 cells. For the experiment, cells were treated with control siRNA or siRNA against ATF4 or A TF5 either alone or in combination. At 36 h post transfection, cells were transferred into MEM, MEM containing 2mM HisOH, or MEM containing 300nM Tg, and cultured for additional 8 h. Total RNA was collected and subjected to qRT PCR to measure the mRNA co ntent of FOXA2 and FOXA3 (Figure 5 8C). As controls, the mRNA levels of ASNS and SNAT2, which are known to be regulated by ATF4, were also investigated (Figure 5 8A). The siRNA treatment successfully knocked down the mRNA expression of ATF4 or ATF5 (Figu re 5 8B). It is important to note that the ATF5 mRNA expression was also affected by ATF4 knock down, which is consistent with a previous study showing that the ATF5 mRNA level was significantly lower in ATF4 knockout MEF cells (Zhou et al., 2008) The induction of the ASNS gene by either amino acid deprivation or ER stress was completely blocked by ATF4 knock down, but was largely unaffected by ATF5 knock down (Figu re 5 8A). This result indicated that although ATF5 was able to activate the transcription from the ASNS promoter in vitro it is not required for the ASNS induction by amino acid deprivation of intact cells. Given the fact that ATF5 expression was also a ffected by ATF4 knock down, although ATF4 knock down resulted in a complete block of the ASNS induction, it remained unanswered whether or not ATF5 alone is sufficient to activate the ASNS gene in the absence of ATF4. The induction of the SNAT2 gene by am ino acid deprivation was also partially repressed by ATF4 knock down, but was not affected by ATF5 knock down (Figure 5 8A). In contrast to ASNS and SNAT2, the induction of FOXA2 by amino acid deprivation was slightly inhibited by ATF4 knock down(Figure 5 8C), however, given that ATF4/ATF5 double knock down did not significantly affect the induction of

PAGE 138

123 FOXA2, no conclusion can be made. Similar to FOXA2, the induction of FOXA3 was nearly unaffected by ATF4 and/or ATF5 knock down (Figure 5 8C). These resul ts suggest that the induction of FOXA2 and FOXA3 does not require ATF4 or ATF5, although additional evidence using alternative approaches (i.e., knockout cells) would be helpful. Friedman et al. (Friedman et al., 2004) reported that FOXA3 expression is decreas ed in 9C). The induction of FOXA2 was also inhibited, but to a lesser extent compared to FOXA3 (Figure 5 9B). These results demonstrated that the induction of human FOXA3 gene by amino aci d deprivation is at maximum induction of FOXA2. Genomic analysis of the FOXA3 gene Many amino acid responsive genes contain at least one amino acid responsive eleme nt (AARE), which directs the binding of ATF4 and C/EBP proteins (Kilberg et al., 2005) Given that the FOXA3 gene is the most induced by amino acid deprivation among the three FOXA family members, I decided to investigate whether the induct ion of FOXA3 is through an AARE element. Computer based genomic analysis revealed four potential AARE sequences in the TGATGCAAC 7855 / TGATGAAAA 2113 / TGAT GCAGC 210 / TGATGCAAG 10). Because all functional AAREs are bound by (Kilberg et al., 2005) a chromatin immunoprecipitation (ChIP) assay was performe d to investigate the association of ATF4 or

PAGE 139

124 containing a known AARE was also investigated. Surprisingly, although all four potential AARE sequences are highly similar o r even identical to the consensus of known AAREs, none of and the binding was not induced by HisOH treatment (Figure 5 11A). Although the binding of FOXA3 promoter region was significantly increased after HisOH treatment, the overall value was still very low as compared to that on the ASNS promoter (Figure 5 11A). pected, amino acid deprivation (Figure 5 11A). The values of binding to ASNS were dramatically higher (more than 50 fold) than those on the FOXA3 AARE like sequences. This lack of not serve as functional AARE elements. We continued our effort to identify the genomic sequence that is responsible for the induction of the FOXA3 gen e during amino acid deprivation, I first cloned a genomic fragment covering the proximal promoter region ( 1000 / +151) of the FOXA3 gene, and inserted it into the pGL3 luciferase reporter plasmid to drive the expression of Firefly luciferase (FOXA3 promot er Luc). HepG2 cells were transfected with this reporter plasmid, and cultured in MEM, MEM containing 2mM HisOH, or MEM containing 300nM Tg for 12h. The 1.15 kb fragment indeed had promoter activity as it drove the expression of luciferase compared to th e non promoter plasmid (data not shown). However, the luciferase activity was slightly inhibited by HisOH or Tg treatment (Figure 5 11B), suggesting that this promoter fragment does not contain an enhancer that can mediate the amino acid response of the F OXA3 gene. One of the four

PAGE 140

125 AARE like sequences ( 7855 / 7864) is identical to the AARE sequence in the HERP gene (Lenz et al., 2006) (Figure 5 10 decided to test its function using this in vitro reporter system. A second reporter (FOXA3 8kb AARE pro/Luc) was made by cloning a 360 bp fragment ( nt 7977/ 7617 ) that include the AARE like sequen ce ( 7855 / 7864) and upstream of the 1.15 kb promoter in the FOXA3 promoter Luc plasmid. As shown in Figure 5 11B, the addition of this AARE like sequence did not cause transcriptional activation from the FOXA3 1.15kb promoter during amino acid deprivat ion, which was consistent with the proposal that a lack of ATF4 binding equals non functionality as an AARE element. Hiemisch et al. (Hiemisch et al., 1997) demonstrated that the FOXA3 gene contains a tissue specific enhancer element located at +16kb, which is downstream of the transcriptional termination site, and is sufficient to drive the expression of a reporter gene in liver, pancreas, stomach and small intestine. To test if that enhancer also mediates the induction of FOXA3 by amino acid deprivation, a 1.7 kb genomic fragment ( nt + 15552 /+ 17280 ) around nt +16kb was cloned, and placed i t upstream of the FOXA3 promoter in the FOXA3 promoter Luc plasmid to make the FOXA3 enh promoter Luc reporter. The response of this reporter to amino acid deprivation or ER stress was then investigated using the luciferase assay. As shown in Figure 5 11 B, the addition of the enhancer element did not grant the FOXA3 promoter the ability to respond to amino acid deprivation. However, when the +16kb enhancer was present, over expression of FOXA3 promoter (Figure 5 11B). This result demonstrates that although the +16kb enhancer does not mediate the amino acid response of the FOXA3 gene, it may be an important cis regulatory element gi

PAGE 141

1 26 acid deprivation (Figure 5 growth control (Greenbau m et al., 1998; Buck and Chojkier, 2003) Collectively, these results demonstrate that the induction of the FOXA3 gene by amino acid deprivation is likely to be through a non conventional AAR pathway, which requires the participation of the transcription fa Function of FOXA2 and FOXA3 on Amino Acid Responsive Genes The induction of FOXA2 and FOXA3 by amino acid deprivation prompted me to investigate if they are involved in the regulation of amino acid responsive genes during nutrient stress. G iven that FOXA3 is the only FOXA gene for which protein expression was significantly induced during amino acid deprivation, I started by testing the function of FOXA3 on two well studied amino acid responsive genes, ASNS and SNAT2, using a luciferase repor ter analysis. Two reporter constructs were generated previously in the Kilberg laboratory by inserting the promoter regions of ASNS ( 173/+51) and SNAT2 ( 512/+770) into the pGL3 luciferase vector. Both promoter fragments contain a functional AARE sequen ce and were shown to be able to mediate the induction of these genes by both amino acid deprivation and ER stress in vitro (Zhong et al., 2003; Palii et al., 2004) Over expression of FOXA3 slightly induced the basal transcription from the ASNS promoter by about 2 fold at the highest concentration of plasmid transfected (100ng/well) (Figure 5 12A). The induction of the ASNS promoter driven luciferase expression by amino acid deprivation or E R stress was also enhanced by FOXA3 over expression, by approximately 30% in each condition, at the highest concentration of plasmid transfected (100ng/well) (Figure 5 12A). In contrast to its effect on the ASNS promoter, FOXA3 over expression slightly in hibited both basal and induced transcription from the SNAT2 promoter, by either amino acid limitation or ER stress (Figure 5 12B). It is worth noting that although the changes in the luciferase activity reached statistical significance relative to

PAGE 142

127 inducti on of 50 to 100 fold by ATF4 (Pan et al. 2003) the relative small percentage of increase or decrease of the luciferase activity mediated by FOXA3 over expression indicated that FOXA3 is not a major regulator of ASNS or SNAT2, at least not through the promoter fragments investigated in this ex periment. To extend the study to other AARE containing genes, I utilized an siRNA approach to simultaneously knock down endogenous FOXA2 and FOXA3 expression. This double knock down assay, compared to knock down of a single transcription factor, avoided possible functional redundancy between FOXA2 and FOXA3, therefore allowed a more comprehensive understanding of their function on the amino acid responsive genes. HepG2 cells were treated with either control siRNA or FOXA2 and FOXA3 siRNA, 36 h post trans fection, cells were transferred to MEM, MEM containing 2mM HisOH, or MEM containing 300nM Tg, and cultured for additional 8 h. Total RNA was isolated and subjected to qRT PCR to measure the mRNA abundance of specific amino acid responsive genes, including ASNS, SNAT2, CHOP, TRB3, VEGF, and CAT 1 (Figure 5 13 and Figure 5 14). As a control, the mRNA abundance of GLUT2, a known FOXA3 target gene, was also investigated (Figure 5 13B). The siRNA against FOXA2 and FOXA3 specifically knocked down the mRNA expr ession of these two genes, as evidenced by qRT PCR analysis (Figure 5 13A). As expected, the expression of GLUT2 was decreased following FOXA2 and FOXA3 knock down, with approximately 40% reduction under amino acid deprivation and 20% reduction under ER s tress (Figure 5 13B). Interestingly, the GLUT2 expression increased slightly after 8 h of amino acid deprivation but decreased by approximately 50% during ER stress. This decrease of the glucose transporter expression during ER stress, which can be trigg ered by glucose starvation (Patil and Walter, 2001) is likely to be part of an adaptive mechanism to prevent further loss of glucose from the cells. It is worth

PAGE 143

128 noting that during ER stress, the FOXA3 expression was la rgely unaffected, whereas the GLUT2 expression decreased dramatically, which indicates that there are additional factors involved in the regulation of GLUT2. In contrast to the GLUT2 gene, the mRNA expression of many amino acid responsive genes, including ASNS, TRB3, VEGF, and CAT 1, was not affected by FOXA2 and FOXA3 knock down (Figure 5 14). Surprisingly, both the basal and the induced expression of SNAT2 were significantly decreased after FOXA2/3 knock down (Figure 5 14). Moreover, the induction of C HOP by ER stress, but not by amino acid deprivation, was also suppressed by FOXA2 and FOXA3 knock down. Given the pro apoptotic function of CHOP (Zinszner et al., 1998) t he functional connection between FOXA2/FOXA3 and CHOP will be of interest for further investigation. Collectively, these results indicated that FOXA2 and FOXA3 mRNA are induced by amino acid deprivation, likely throug h an ATF4 independent pathway. Their target genes fall into a different group than those for which expression is induced through ATF4 during amino acid deprivation, though one of the ATF4 target genes, SNAT2, requires FOXA2 and FOXA3 for both basal expres sion and the induction by amino acid deprivation. Given the function of FOXA2 and FOXA3 in maintaining glucose homeostasis in liver, this study suggests potential cross talk between the amino acid response pathway and glucose metabolism. Conclusions and D iscussions Only a small group of genes are up regulated by nutrient deprivation. Some of the genes are directly involved in metabolic control, such as ASNS, some are membrane transporters, such as SNAT2 and CAT 1, and many are transcription factors (Kilberg et al., 2005) Identification and characterization of transcription factors induced during nutrient stress will greatly add to our knowledge of the nutrient control of gene expression.

PAGE 144

129 The results described in the present study documen t the differential regulation of the forkhead box A (FOXA) family of genes by amino acid deprivation and ER stress. The FOXA family consists of three members, FOXA1, 2, and 3. They are primarily expressed in tissues such as liver, pancreas, and gastrio i ntestinal (GI) track, and play important roles during development and metabolic control (Friedman and Kaestner, 2006) The mRNA expression of FOXA2 and FOXA3 are increased during amino acid deprivation, whereas the expression of FOXA1 is decreased. The increase in mRNA abundance is likely due to elevated transcription rather than mR NA stabilization. As opposed to amino acid deprivation, ER stress dramatically down regulates the FOXA1 and FOXA3 genes, but slightly increases the expression of FOXA2. Although both FOXA2 and FOXA3 mRNA are induced by amino acid deprivation, only FOXA3 exhibits significant increase in protein abundance, whereas FOXA2 has an increased protein expression during ER stress. The siRNA analysis demonstrated that the induction of FOXA2 and FOXA3 by amino acid deprivation was not through a conventional amino ac id response pathway, which is dependent on the bZIP transcription factor ATF4, but rather their induction are at least partially dependent on another bZIP transcription factor, also supported by the fact that the induction of FOXA3 by amino acid deprivation is not mediated by an AARE element within its gene locus. FOXA2 and FOXA3 are not involved in the regulation of most other AARE containing genes. Ho wever, one of the amino acid responsive genes, SNAT2 requires FOXA2 and FOXA3 for its basal expression and its induction by amino acid deprivation. Mammalian cells respond rapidly to a change of nutrient availability in the environment. Deprivation of cel ls with essential nutritional molecules, such as amino acids and glucose, initiates either adaptive mechanisms or programmed cell death by changing the gene expression

PAGE 145

130 profile. Although the level of responsiveness varies among different cell types, cell l ines derived from hepatocytes usually exhibit strong response to a variety of nutrient stress conditions, therefore several hepatoma cell lines, such as HepG2 and Huh7, have been popularly used for such studies. Several universal nutrient response pathway s have been established, such as the amino acid response (AAR) pathway initiated by amino acid limitation (Kilberg et al., 2005) and the unfolded protein response (UPR) pathway initiated by glucose starvation (Patil and Walter, 2001) However, as a major organ for metabolic control, the liver holds a number of tissue specific pathways involving many liver specific transcription factors, such as the Hepatic Nuclear factor (HNF) super family. It is importan t to study how liver integrates these universal and tissue specific pathways to delicately maintain the metabolic homeostasis of the body. regulated in mice deprived of protein (Imae et al., 2000) inspire d me to investigate the regulation of FOXA3 as well as other FOXA family members by nutrient limitation. The three human FOXA genes, FOXA1, FOXA2, and FOXA3 have been mapped to chromosomes14q12 q13, 20p11 and 19q13.2 q13.4, respectively (Mincheva et al., 1997) Consistent with their difference in loca tion, a differential response of the FOXA genes to amino acid deprivation or ER stress was observed. Although all three FOXA proteins share similarity in their DNA binding domains (more than 90% identity), and are proposed to have functional redundancy in regulating FOXA target genes (Kaestner et al., 2000) their differential expression under stress conditions may indicate their relative importance in regulating certain genes. Many genes for which expression is induced b y amino acid deprivation harbor at least one amino acid response (AARE) element which provides a binding site for ATF4 and other bZIP transcription factors (Kilberg et al., 2005) Besides its critical role in the AAR

PAGE 146

131 pathway, ATF4 also mediates the induction of many genes by other stress conditions, such as ER stress (Costa Mattioli et al., 2007) the presence of long double strand RNA (Barber, 2005) and heme deficiency (Yoneda et al., 2001) The observation that the induction of FOXA3 by amino acid deprivation was neither dependent on ATF4 nor on a functional AARE element, brought forth the possibility of a non conventional AAR pathway that activates the FOXA3 gene during amino acid deprivation. Such a pathway may involve the bZIP tran scriptio which is also induced by amino acid deprivation. This proposal is supported by the fact that the the FOXA3 promoter through a known FOXA3 enhancer (Hiemisch et al., 1997) although thi s enhancer did not mediate the amino acid response of FOXA3 (Figure 5 11B). Shen et al. (Shen et al., 2001) demonstrated that mouse Foxa3 regulates the expression of the glucose transporter gene GLUT2 during a prolonged fast. This is confirmed by our siRNA knock down assay (Figure 5 13B). Interestingly, the GLUT2 expression was dramatically inhibited by ER stress (Figure 5 13B). The down regulation of the glucose transporter expression under ER stress may be important to reduce glucose efflux from the hepatocytes therefore prevents further glucose loss in a potential glucose deprivation condition. FOXA family proteins are monomeric transcription factors (Weigel et al., 1989; W eigel and Jackle, 1990) therefore are unlikely to mediate transcriptional activation through an AARE element, which requires the binding of a heterodimer of two bZIP transcription factors. This was confirmed by the lack of function of FOXA3 on the ASNS or SNAT2 promoter, which contains an AARE element (Figure 5 12). Furthermore, we have demonstrated that FOXA2 and FOXA3 are not required for many AARE containing genes (Figure 5 14). However, the fact that the

PAGE 147

132 SNAT2 expression was significantly hindered after FOXA2 and FOXA3 knock down suggests a functional involvement of the FOXA genes in the amino acid response pathway. A complete screen of FOXA2/3 target genes by ChIP on CHIP or ChIP/sequencing analysis will greatly help our understanding of amino aci d control of gene expression.

PAGE 148

133 Figure 5 1. Functional domains in human FOXA1, 2 and 3 proteins. All three family members share 95% identity within the forkhead domain. Outside of the forkhead domain, Foxa3 is only weakly similar to FOXA1 and FOXA2, with the greatest homology in the N terminal and C terminal transactivation domains. The C terminal region has also been shown to interact with the core histones H3 and H4. TA, transactivation domain; HI, histone interaction domain; NL, nuclear localiza tion.

PAGE 149

134 Figure 5 2. Regulation of the FOXA family of genes by amino acid deprivation and ER stress. HepG2 cells (A) or human primary hepatocytes (B) were incubated in MEM, MEM containing 2 mM HisOH, or MEM containing 300 nM Tg for 8 h. Total RNA was collected and subjected to qRT PCR analysis for FOXA1, FOXA2, FOXA3 or GAPDH mRNA content. The ratio of FOXA m RNA to the GAPDH control in HepG2 cells (Panel A) or human primary hepatocytes (Panel B) was calculated and the data are presented as the fold in duction relative to T=0 control for each condition. The graph illustrates the means standard error of the mean (S.E.M.) for three independent experiments. (Human hepatocyte RNA was provided by Dr. Michelle M. Thaville.)

PAGE 150

135 Figure 5 3. The expression of ASNS, SNAT2, FOXA1, FOXA2, and FOXA3 in the pregnant mice fed with normal or low protein diet. At day 5 of pregnancy, the C57BL/6J mice were pair fed with normal protein diet (19.39% protein), named NPD mice, or low protein diet (8% protein), named LPD mice. The mice were sacrificed at day 18.5 of gestation, and the livers were isolated, frozen and stored at 80C. Total RNA was collected and subjected to qRT PCR analysis for ASNS, SNAT2, FOXA1, FOXA2, FOXA3 or GAPDH mRNA content. The ratio of specif ic m RNA to the GAPDH control was calculated and the data are presented as the fold induction relative to the NPD mice. The values illustrated in the graph are the average of two mice in each experimental group and individual data point is represented as b lack dot. (Mouse liver samples kindly provided by Mr. Jason Brant)

PAGE 151

136 Figure 5 4. Time couse expression of FOXA genes during amino acid deprivation and ER stress. HepG2 cells were incubated in MEM, MEM containing 2 mM HisOH, or MEM containing 300 nM Tg. At the times indicated, RNA was isolated and analyzed by qRT PCR analysis for FOXA1, FOXA2, FOXA3 or GAPDH mRNA content. The ratio of FOXA m RNA to the GAPDH control was calculated and the data are presented as the fold induction relative to control (Time = 0 h) for each condition. The graph illustrates the means S.E.M. for three independent experiments. Individual panels present the expression patterns of FOXA1 (Panel A), FOXA2 (Panel B), and FOXA3 (Panel C).

PAGE 152

137 Figure 5 5. Change of FOXA2, FOXA3, an d p21 mRNA stability following amino acid depriva tion. HepG2 cells were pre incubated in MEM containing 2 mM HisOH for 12 h, then transferred to MEM or MEM containing 2 mM HisOH, each supplemented analyzed by qRT PCR analysis for FOXA1, FOXA2, p21 or GAPDH mRNA content. The ratio of specific m RNA to the GAPDH control was calculated. The data are presented by plotting on a semi logarithmic graph the fold induction relative to control (Time = 0 h) versus the time, for FOXA2 (Pa nel A), FOXA3 (Panel B), and p21 (Panel C). The graph illustrates the means S.E.M. for three independent experiments.

PAGE 153

138 Figure 5 6. Effect of blockage of de novo transcription on the induction of FOXA2 and FOXA3 by amino acid deprivation. HepG2 cells were divided into two experimental groups (ActD). After a 30 min pre incubation, the cells within each experimental group were further treated without (MEM) or with 2 mM HisOH for 8 h in the continued presence of ActD. Total RNA was isolated and analyzed by qRT PCR analysis for FOXA2, FOXA3 or GAPDH mRNA content. The ratio of FOXA m RNA to the GAPDH control was calculated. The data are presented as the fold induction relative to the MEM value in the control group (no ActD) for FOXA2 and FOXA3. Note the difference in the values in the presence or absence of ActD. The graph illustrates the means S.E.M. for three independent experiments.

PAGE 154

139 Figure 5 7. Time couse expression of FOXA protein during amino acid deprivation and ER stress. HepG2 cells were incubated in MEM, MEM containing 2 mM HisOH, or MEM containing 300 nM Tg. At the times indicated, protein extracts were collected and subjected to immunoblot analysis for FOXA1, F actin content (Panel A). The blots shown are the representative of three independent experiments. Quantified immunoblot data (as described in Materials and Methods) for FOXA1, FOXA2, and FOXA3 are plotted relative to the control value (time = 0 h) for each condition (Panel B D). The graph illustrates the means S.E.M. for three independent experiments.

PAGE 155

140 Figure 5 8. Effect of ATF4 and/or ATF5 knock down on the expression of FOXA2 and FOXA3. HepG2 cells were transfected with eithe r control siRNA or ATF4 and/or ATF5 siRNA. At 36 h post transfection, cells were incubated in MEM, MEM containing 2 mM HisOH, or MEM containing 300 nM Tg for 8 h. Total RNA was isolated and subjected to qRT PCR analysis for ATF4, ATF5 (Panel B), ASNS, SN AT2 (Panel A), FOXA2, FOXA3 (Panel C), and GAPDH mRNA content. The data are presented by plotting the fold induction relative to control (MEM, control siRNA). The graph illustrates the means S.E.M. for three independent experiments. Those values that are significantly different (p < 0.05) from the corresponding control are indicated with asterisks.

PAGE 156

141 Figure 5 9 down on the expression of FOXA2 and FOXA3. HepG2 transfection, cells were incubated in MEM or MEM containing 2 mM HisOH for 8 h. Total RNA was isola ted and subjected to qRT (B), FOXA3 (C), and GAPDH mRNA content. The data are presented by plotting the fold induction relative to control (MEM, control siRNA). The graph illustrates the means S.E.M. for three indepen dent experiments. Those values that are significantly different (p < 0.05) from the corresponding control are indicated with asterisks.

PAGE 157

142 Figure 5 10. Genomic analysis of the FOXA3 gene. The diagram shows a computer based genomic analysis of the FOXA g ene. Four potential AARE sequences, as well as a known enhancer are shown along the gene. A consensus sequence (TGATGXAAX) from the known AARE elements of amino acid responsive genes is also shown.

PAGE 158

143 Figure 5 11. Promoter and enhancer analysis of the F OXA3 gene. HepG2 cells were incubated in MEM control medium or treated with 2 mM HisOH 8 h. Chromatin immunoprecipitation (ChIP) assays were performed as described in Chapter 2. DNA fragments were immunoprecipitated with ATF4 or antibody and the enrichment of ATF4 or protein at the indicated gene region was analyzed by qPCR, using primer sets specific for the regions of interest. Data were plotted as the ratio to the value obtained with a 1:20 dilution of input DNA. Each condition was ana lyzed in triplicate and each point represents the S.E.M. for two independent experiments (Panel A). For the luciferase reporter assay, FOXA3 promoter/Luc, FOXA3 promoter/ 8kb AARE/Luc, and FOXA3 promoter/+16kb Enh/Luc constructs were transfected into He pG2 cells along either 2 mM HisOH or 300nM Tg treatment. Cell extracts were assayed for luciferase activity as described in Chapter 2 (Panel B). Each value represents three assays and each experiment was repeated with two different batches of cells. The results shown represent the means S.E.M. An asterisk indicates that the value is significantly different (p < 0.05) from the MEM control.

PAGE 159

144 Figure 5 12. Effect of FOXA3 on the transcription from the ASNS or SNAT2 promoter. The FOXA3 expression plasmi d was transfected into HepG2 cells along with the ASNS 173/+51 promoter/Luc or SNAT2 512/+770 promoter/Luc reporter plasmid in combination with either 2 mM HisOH or 300nM Tg treatment, as indicated. Cell extracts were assayed for luciferase activity a s described in Chapter 2. Each value represents three assays and each experiment was repeated with two different batches of cells. The results shown represent the means S.E.M. An asterisk indicates that the value is significantly different (p < 0.05) f rom the control (0 ng/well FOXA3).

PAGE 160

145 Figure 5 13. Effect of FOXA2 and FOXA3 knock down on the expression of GLUT2. HepG2 cells were transfected with either control siRNA or FOXA2 and FOXA3 siRNA. At 36 h post transfection, cells were incubated in MEM, MEM containing 2 mM HisOH, or MEM containing 300 nM Tg for 8 h. Total RNA was isolated and subjected to qRT PCR analysis for FOXA2, FOXA3 (Panel A), GLUT2 (Panel B), and GAPDH mRNA content. The data are presented by plotting the fold induction relative t o control (MEM, control siRNA). The graph illustrates the means S.E.M. for three independent experiments. Those values that are significantly different (p < 0.05) from the corresponding control are indicated with asterisks.

PAGE 161

146 Figure 5 14. Effect of FOXA2 and FOXA3 double knock down on the expression of AARE containing genes. HepG2 cells were transfected with either control siRNA or FOXA2 and FOXA3 siRNA. At 36 h post transfection, cells were incubated in MEM, MEM containing 2 mM HisOH, or MEM cont aining 300 nM Tg for 8 h. Total RNA was isolated and subjected to qRT PCR analysis for ASNS, SNAT2, CHOP, TRB3, VEGF, and CAT 1, and GAPDH mRNA content. The data are presented by plotting the fold induction relative to control (MEM, control siRNA). The graph illustrates the means S.E.M. for three independent experiments. Those values that are significantly different (p < 0.05) from the corresponding control are indicated with asterisks.

PAGE 162

147 Figure 5 15. Effect of FOXA2 or FOXA3 single knock down on t he expression of SNAT2 and CHOP. HepG2 cells were transfected with control siRNA, FOXA2, or FOXA3 siRNA. At 36 h post transfection, cells were incubated in MEM, MEM containing 2 mM HisOH, or MEM containing 300 nM Tg for 8 h. Total RNA was isolated and s ubjected to qRT PCR analysis for SNAT2, CHOP (Panel A), FOXA2, FOXA3 (Panel B), and GAPDH mRNA content. The data are presented by plotting the fold induction relative to control (MEM, control siRNA). The graph illustrates the means S.E.M. for three ind ependent experiments. Those values that are significantly different (p < 0.05) from the corresponding control are indicated with asterisks.

PAGE 163

148 CHAPTER 6 CONCLUSIONS AND FUTU RE DIRECTIONS Conclusions ATF4 is the best characterized terminal regulator in the amino acid response (AAR) pathway (Kilberg et al., 2005) It is also up regulated by a number of other stress response pathways, including unfolded protein response (UPR), double stranded RNA response, and heme deprivation response (Kaufman, 2004; Barber, 2005; Lu et al., 2001) The study of ATF4 function in the Kilberg laboratory is car ried out mainly in connection with regulation of the ASNS and SNAT2 genes. During an effort to search for co activators of ATF4 function on these two genes using chromatin immunoprecipitation (ChIP) analysis, many well known co activator or co activator c omplexes, such as CBP, p300, pCAF, the Mediator complex, and the SWI/SNF chromatin remodeling complex, were demonstrated not to be associated with the regulation of the ASNS and SNAT2 genes (Kilberg laboratory, unpublished data). These observations, in co njunction with the potency of ATF4 in mediating transcription in yeast without the facilitation of other mammalian factors (as demonstrated in Chapter 3), and the fact that ATF4 is associated with several components of the general transcription machinery, such as RPB3 subunit of RNA Pol II, TFIIB, and TBP (Liang and Hai, 1997), generated the hypothesis that ATF4 may not need co activators to activate the transcription of some stress responsive genes. Nevertheless, identification of novel ATF4 interacting p artners will greatly add to our understanding of the mechanism of ATF4 dependent transcriptional regulation during stress responses. Besides ATF4, many other transcription factors are also induced by amino acid deprivation, and regulate genes involved in different cellular processes. Identification and characterization of these transcription factors will help unveil the interconnection and cross talk among different cellular processes.

PAGE 164

149 In my studies, a yeast two hybrid screening assay was performed to ide ntify ATF4 interacting proteins. The transcription factor CHOP was identified as one of the strongest positives in the screen, and was demonstrated to interact with ATF4 in mammalian cells by co IP analysis. CHOP negatively regulates ATF4 function on the transcription from an ASNS promoter driven reporter, and also inhibits the endogenous ASNS expression. Prior to this study, The discovery of CHOP as another ATF4 a ntagonist may have physiological significance other phase of ASNS activation, CHOP appears to be an acute respondent upon activation of stress. The robust and s imultaneous induction of ATF4 and CHOP during stress responses may keep the cells at a checkpoint for cell fate determination. While a number of adaptive mechanisms are activated by ATF4, CHOP antagonizes ATF4 function, and if necessary, initiates program med cell death. Intriguingly, when the cells have successfully adapted to the stress, such as deprivation for a certain amino acid, CHOP expression is quickly restored to a very low basal level to avoid activation of apoptotic pathway. CHOP was originall y demonstrated to negatively regulate the transcriptional activity of the C/EBP family proteins by sequestering them from binding to their DNA recognition sites (Ron and Habener, 1992) but CHOP was shown later to have DNA binding activities on specific DNA sequences (Ohoka et al., 2005) In my study, the bZIP domain of CHOP, wh ich is enough to bind to ATF4, was not sufficient to inhibit ATF4 activity on the ASNS gene. This led me to hypothesize that the CHOP forms a functional heterodimer with ATF4 and regulates ATF4 activity on the ASNS gene. This hypothesis was supported by the association of both CHOP and ATF4 to the ASNS promoter region containing the AARE element. The DNA binding activity

PAGE 165

150 of CHOP also holds true for the AARE regions of several other amino acid responsive genes, including TRB3, SNAT2, and VEGF. Interestin gly, the outcomes from CHOP binding to these different genes are quite different, as CHOP activates the TRB3 gene, which is consistent with a previous report (Ohoka et al., 2005) inhibits the ASNS gene, and has largely no effect on the SNAT2 and the VEGF gene. The mechanisms by which CHOP differentially regulates stress responsive genes are of great interest for future investigation. The expression of ASNS is transcriptionally regulated through the AAR pathway during a mino acid deprivation. However, I demonstrated that an increase of the ASNS mRNA level does not necessarily reflect a proportional increase in its protein abundance. A ChIP analysis was performed using two ALL cell lines, the MOLT 4 parental and resistan t cells, which are from the same origin, but dramatically different in ASNase sensitivity. The results demonstrated that the transcription of the ASNS gene is almost identically regulated through the AAR pathway though to a different extent. However, the protein expression is extremely inefficient and delayed in the ASNase sensitive cells, which leads to the sensitive phenotype. By investigating the ASNS mRNA and protein expression in several ALL cell lines, I observed that different ALL cell lines have different basal ASNS mRNA expression. Upon ASNase treatment, all these cells are able to induce the ASNS mRNA expression to a comparable level. However, the ASNS protein levels vary dramatically among different ALL cell lines. My studies confirm the obs ervations in the previous reports that the ASNS mRNA expression does not correlate with ASNase sensitivity (Holleman et al., 2004; Krejci et al., 2004) However, I also demonstrated that the ASNS protein expression is the ultimate indicator for ASNase sensitivity, as cells with higher ASNS protein abundance are more resistant t o ASNase treatment. As for the ALL patients, expression profiling in some patient samples revealed that different ALL patients have

PAGE 166

151 different level of ASNS mRNA expression. However, most patients do not express detectable ASNS protein, and therefore, are sensitive to ASNase therapy. This study may provide a guide for clinical prediction of the prognosis for the chemotherapies involving ASNase. In my study, the transcription factor FOXA3 was found to be induced by amino acid deprivation in the HepG2 hu man hepatoma cells, which is consistent with a previous report that FOXA3 is induced in protein deprived mice (Imae et al ., 2000) Investigation of all three members of the FOXA family, FOXA1, 2, and 3, revealed that both FOXA2 and FOXA3 are induced upon activation of the AAR pathway, whereas FOXA1 was slightly down regulated. In the contrast, activation of the UPR pathwa y dramatically decreases the expression of FOXA1, but has little or no effect on FOXA2 and FOXA3. The up regulation of FOXA2 and FOXA3 by amino acid limitation was also confirmed in vivo as livers isolated from mice fed with a low protein diet have highe r FOXA2 and FOXA3 mRNA levels compared to those in the mice fed with normal diet. The induction of FOXA2 and FOXA3 by amino acid deprivation seems to be independent of ATF4, as siRNA knock down of ATF4 did not affect their induction. Furthermore, althoug h the FOXA3 gene locus has four AARE like sequences, none of them was able to direct the binding of ATF4 upon activation of the AAR pathway. In contrast to ATF4, knocked down by siRNA, the induction of both FOXA2 and FOXA3 was partially suppressed, and FOXA3 during amino acid deprivation. The FOXA family proteins play an important role in maintaining glucose homeostasis. Consistent with a previous report that FOXA3 directly regulates the GLUT2 gene (Shen et al., 2001) which encodes for a glucose transporter, siRNA mediated knock down of FOXA3

PAGE 167

152 resulted in decreased expression of GLUT2. In the contrast, although both FOXA2 and FOXA3 are induced by amino acid deprivation, simultaneous knock down of their expression by siRNA had essentially no effect on the expression of many amino acid responsive genes. Interestingly, the expression of the amino acid transporter SNAT2 was decreased when FOXA2 an d FOXA3 were knocked down, and this inhibition effect was demonstrated to be mainly due to FOXA2. Collectively, these results suggest the existence of an ATF4 independent amino acid response pathway, and the transcription factors induced by this pathway, such as FOXA2 and FOXA3, may mediate the cross talk between amino acid deprivation and the regulation of other metabolic processes. Future Direction The yeast two hybrid screen generated a list of potential ATF4 interacting proteins, which may be subjec ts for future studies. The highest scored positive gene identified from the screen is human hUBC9, which is a key component in the SUMOylation pathway (Muller et al., 2001) Although hUBC9 over expression had no obvious effect on the transcription from the ASNS promoter (Figure C 1), we can not exclude the possibility tha t hUBC9 needs other components in the SUMOylation machinery to modify ATF4 and alter its activity. Indeed, Lys X Glu) in the ATF4 protein sequence (KKXE). hUBC9 directly interacts wi th its target proteins through the SUMOylation sites (Melchior, 2000; Melchi or et al., 2003) The first step to study hUBC9 function in the amino acid response will be using GST pull down and co immunoprecipitation assays to confirm the interaction between hUBC9 and ATF4 both in vitro and in vivo A mutant ATF4 with its potenti al SUMOylation site mutated should also be investigated to see if this sequence is responsible for hUBC9/ATF4 interaction. SUMOylation of many transcription factors, such as p53, c jun, and many C/EBP proteins, has been shown to reduce their

PAGE 168

153 transcription al activities (Melchior and Hengst, 2002; Poukka et al., 2000; Bies et al., 2002; Eloranta and Hurst, 2002; Ross et al., 2002; Sapetschnig et al., 2002) Consistent with these observations, over expression of hUBC9 at high concentration slightly decreased the ATF4 activity on the ASNS promoter (Figure C 1). In future studies, simultaneous expression of all the components of the SUMOylation pathway will generate more convincing data on the function of ATF4 SUMOylation. Ove r expression of wild type and mutant ATF4 in ATF4 / MEF cells in luciferase reporter assays could also provide useful information on the regulation of ATF4 activity by hUBC9 mediated SUMOylation. The Kilberg laboratory has been focusing on the transcrip tional regulation of gene expression. The ASNS gene is one of the best characterized amino acid responsive genes, and Chen et al. (Chen et al., 2004) proposed a two phase model for the transcriptional regulation of the ASNS gene during amino a cid limitation. In the current study, I demonstrated that the expression of ASNS seems to be regulated at the translational level as well. This is particularly important when it comes to the question that how is the regulation of the ASNS gene related to the clinical treatment of the ALL patients with the chemotherapeutic agent ASNase. The mechanism through which the translation of the ASNS mRNA is differentially regulated in different ALL cell lines is not yet understood. It may be due to the restricti on of mRNA export from the nucleus to the cytoplasm. This possibility can be tested by fractionating the nuclear and cytosolic RNA and detecting for ASNS mRNA in each fraction. The other possible mechanism is the inhibition of mRNA translation by micro R NA (miRNA). One such example is the translation inhibition of the CAT 1 mRNA by the miRNA miR UTR region of CAT 1 mRNA during normal culture condition, and is released during stress (Bhattacharyya et al., 2006) To investigate whether or not the translation of the ASNS mRNA

PAGE 169

154 is also regulated by miRNA, a screen of known miRNA against the ASNS mRNA sequence using databases such as miRBase ( http://microrna.sanger.ac.uk ), will be the initial step. An interesting study is to see if the ASNase mediated apoptosis in the ALL cells is a CHOP or p53 dependent process. Although CHOP has been shown to be involved in the ER stress mediated apoptosis, no similar study has been carried out in the context of amino acid deprivation. The fact that HisOH mediated amino acid deprivation did not lead to apoptosis in the HepG2 cells for up to 48 h suggests that either HisOH is no t as strong as ASNase in terms of amino acid starvation effect or ASNase has other pro apoptotic functions than simply depleting the extracellular asparagine. In either case, documenting the CHOP function in ASNase mediated apoptosis will greatly add to o ur understanding of CHOP function during nutrient stresses. I demonstrated that the induction of FOXA3, the most induced gene in the FOXA family, by amino acid deprivation is not dependent on ATF4, nor is it dependent on an AARE element in the FOXA3 gen e locus. It will be of great excitement to unveil this non conventional, ATF4 independent amino acid response pathway. Continuation of the genomic analysis is the first step in future investigations. This can be carried out by two approaches. The first one is cloning a larger piece from the FOXA3 promoter and testing its amino acid responsiveness in a luciferase reporter assay, because the 1.15kb promoter I cloned does not respond to amino acid deprivation. The second approach is to use DNase I hyperse nsitivity assays to identify potential amino acid regulated enhancer regions along the FOXA3 gene locus. The role of FOXA3 in regulating transcription during amino acid deprivation was not fully investigated in the current study. It has been shown that FO XA3 regulates many genes involved in glucose metabolism or transport (Friedman an d Kaestner, 2006) Its induction by

PAGE 170

155 amino acid deprivation suggests that FOXA3 may have more downstream targets. To identify these genes, two approaches can be employed. The first one is to use ChIP on CHIP or ChIP/sequencing to identify all the genes that are bound by FOXA3. The second approach is to use siRNA to knock down FOXA3 expression and subject the RNA samples to microarray analysis to identify all OF the down regulated genes, which are likely to be the targets of FOXA3.

PAGE 171

156 APPENDIX A EFFECT OF AMINO ACID DEPRIV ATION AND ER STRESS DOUBLE STRESS ON THE EXPRESSION OF THE FO XA3 GENE Figure A 1. Regulation of the FOXA3 gene by amino acid deprivation and ER stress double stress. HepG2 cells were divided into two groups, with one group culture d with 300nM Tg, whereas the other group not. Within each group, cells were incubated in MEM, MEM lacking histidine, or MEM containing 2mM HisOH for 8 h. Total RNA was collected and subjected to qRT PCR analysis for FOXA3 or GAPDH mRNA content. The rati o of FOXA m RNA to the GAPDH control was calculated and the data are presented as the fold induction relative to T=0 control for each condition. The graph illustrates the means standard error of the mean (S.E.M.) for three independent experiments. The e xpression of FOXA3 was slightly inhibited by Tg treatment under MEM condition. While histidine starvation induced the expression of FOXA3, Tg almost completely blocked this induction. In the contrast, HisOH treatment also induced FOXA3 mRNA expression, a nd this induction was only partially affected by Tg treatment.

PAGE 172

157 APPENDIX B INVOLVEMENT OF SEVER AL HISTONE ACETYLTRA NSFEREASES IN THE AM INO ACID DEPRIVATION MEDIATED TRANSCRIPTIONAL ACTI VATION Figure B 1. Effect of over expression of several histone acet yltransferases on the ATF4 dependent transcription from the ASNS promoter. The plasmids expressing CBP, pCAF, or p300 were transfected into HepG2 cells along with the ASNS 173/+51 promoter/Luc reporter plasmid in combination with over expression of 10ng/w ell ATF4 plasmid, as indicated. Cell extracts were assayed for luciferase activity as described in Chapter 2. Each value represents three assays in a single experiment. The results shown represent the means standard deviation. The histone acetyltrans ferases CBP, pCAF, and p300 function as transcriptional co activators. Over expression of these factors alone did not cause increased transcription from the ASNS promoter. ATF4 over expression significantly induced transcription from the ASNS promoter. However, over expression of CBP, pCAF, or p300 resulted in little or no further enhancement of the transcription induction.

PAGE 173

158 Figure B 2. Effect of E1A over expression on the transcription from the ASNS promoter. The plasmids expressing E1A was transf ected at different concentrations into HepG2 cells along with the ASNS 173/+51 promoter/Luc reporter plasmid in combination with over expression of 10ng/well ATF4 plasmid or 2mM HisOH treatment, as indicated. Cell extracts were assayed for luciferase acti vity as described in Chapter 2. Each value represents 3 assays in a single experiment. The results shown represent the means standard deviation. E1A is a dominant negative regulator of the histone acetyltransferases CBP, pCAF, and p300 (Jones, 1995). Over expression of E1A at low concentration (5ng/well) significantly repressed (~40%) the transcription activation from the ASNS promoter, by either ATF4 over expression or HisOH treatment. However increasing the concentration of E1A plasmid did not caus e a concentration dependent inhibition.

PAGE 174

159 APPENDIX C INVOLVEMENT OF HUMAN UBC9 IN THE ATF4 MED IATED TRANSCRIPTIONA L ACTIVATION FROM THE ASNS PROMOTER Figure C 1. Effect of hUBC9 over expression on the transcription from the ASNS promoter. The plasmid ex pressing hUBC9 was transfected at different concentrations into HepG2 cells along with the ASNS 173/+51 promoter/Luc reporter plasmid in combination with over expression of 10ng/well ATF4 plasmid, as indicated. Cell extracts were assayed for luciferase ac tivity as described in Chapter 2. Each value represents three assays in a single experiment. The results shown represent the means standard deviation. The human SUMO E2 conjugating enzyme hUBC9 was identified as an ATF4 interacting partner in the yeas t two hybrid screen, as described in Chapter 3. Only when expressed at a very high concentration (400ng/well), did hUBC9 show a slight repress ive (~30%) effect on the transcription activation from the ASNS promoter by ATF4 over expression.

PAGE 175

160 APPENDIX D I NTERACTION BETWEEN A TF4 AND TBP IN VIVO Figure D 1. Test for Interaction of TATA binding protein (TBP) and TFIIB with ATF4 in vivo HA ATF4 was over expressed in HEK293T cells. Whole cell extracts were collected and subjected to co immunoprecipitation using HA antibody conjugated agarose beads. Bound protein was eluted and subjected to immunoblot analysis to detect the presence of TBP and TFIIB. TBP but not TFIIB specifically co immunoprecipitated with HA ATF4.

PAGE 176

161 APPENDIX E IDENTIFICATION OF AT F4 IN TERACTING PROTEINS B Y CO IMMUNOPRECIPITATION Figure E 1. Identification of ATF4 interacting proteins by co IP analysis. HA ATF4 was over expressed in HEK293T cells. Whole cell extracts were collected and subjected to immunoprecipit ation using HA antibody conjugated agarose beads. As control, immunoprecipitation was also performed using HEK293T whole cell extracts without HA ATF4 over expression. Bound protein was eluted and separated by SDS PAGE, followed by silver staining. A sp ecific protein band at MW ~45kD (as indicated by the asterisk) was observed in the HA ATF4 immunoprecipitates. 250 150 100 75 50 37 25 20 15

PAGE 177

162 LIST OF REFERENCES Abcouwer,S.F., Marjon,P.L., Loper,R.K., and Vander Jagt,D.L. (2002). Response of VEGF expression to amino acid deprivation and inducers of endoplasmic reticulum stress. Invest Ophthalmol. Vis. Sci. 43 2791 2798. Al,S.J., Vinson,C., and Thiel,G. (2005). Regulation of asparagine synthetase gene transcription by the basic region leucine zipper transcripti on factors ATF5 and CHOP. Biol. Chem. 386 873 879. Ameri,K. and Harris,A.L. (2008). Activating transcription factor 4. Int. J. Biochem. Cell Biol. 40 14 21. Andrulis,I.L., Argonza,R., and Cairney,A.E.L. (1990). Molecular and genetic characterization of h uman cell lines resistant to L asparaginase and albizziin. Somat. Cell Mol. Gen. 16 59 65. Andrulis,I.L., Hatfield,G.W., and Arfin,S.M. (1979). Asparaginyl tRNA aminoacylation levels and asparagine synthetase expression in cultured Chinese hamster ovary c ells. J. Biol. Chem. 254 10629 10633. Ang,S.L., Wierda,A., Wong,D., Stevens,K.A., Cascio,S., Rossant,J., and Zaret,K.S. (1993). The formation and maintenance of the definitive endoderm lineage in the mouse: involvement of HNF3/forkhead proteins. Developme nt 119 1301 1315. Ashizawa,S., Brunicardi,F.C., and Wang,X.P. (2004). PDX 1 and the pancreas. Pancreas 28 109 120. Aslanian,A.M., Fletcher,B.S., and Kilberg,M.S. (2001). Asparagine synthetase expression alone is sufficient to induce l asparaginase resist ance in MOLT 4 human leukaemia cells. Biochem. J. 357 321 328. Aslanian,A.M. and Kilberg,M.S. (2001). Multiple adaptive mechanisms affect asparagine synthetase substrate availability in asparaginase resistant MOLT 4 human leukemia cells. Biochem. J. 358 59 67. Averous,J., Bruhat,A., Jousse,C., Carraro,V., Thiel,G., and Fafournoux,P. (2004). Induction of CHOP Expression by Amino Acid Limitation Requires Both ATF4 Expression and ATF2 Phosphorylation. J. Biol. Chem. 279 5288 5297. Bain,P.J., LeBlanc Chaffin ,R., Chen,H., Palii,S.S., Leach,K.M., and Kilberg,M.S. (2002). The mechanism for transcriptional activation of the human ATA2 transporter gene by amino acid deprivation is different than that for asparagine synthetase. J. Nutr. 132 3023 3029. Barber,G.N. (2005). The dsRNA dependent protein kinase, PKR and cell death. Cell Death. Differ. 12 563 570.

PAGE 178

163 Barbosa Tessmann,I.P., Chen,C., Zhong,C., Schuster,S.M., Nick,H.S., and Kilberg,M.S. (1999a). Activation of the unfolded protein response pathway induces human asparagine synthetase gene expression. J. Biol. Chem. 274 31139 31144. Barbosa Tessmann,I.P., Chen,C., Zhong,C., Siu,F., Schuster,S.M., Nick,H.S., and Kilberg,M.S. (2000). Activation of the human asparagine synthetase gene by the amino acid response and the endoplasmic reticulum stress response pathways occurs by common genomic elements. J. Biol. Chem. 275 26976 26985. Barbosa Tessmann,I.P., Pineda,V.L., Nick,H.S., Schuster,S.M., and Kilberg,M.S. (1999b). Transcriptional regulation of the human asparagin e synthetase gene by carbohydrate availability. Biochem. J. 339 151 158. Bernales,S., Papa,F.R., and Walter,P. (2006). Intracellular signaling by the unfolded protein response. Annu. Rev. Cell Dev. Biol. 22 487 508. Bhattacharyya,S.N., Habermacher,R., Ma rtine,U., Closs,E.I., and Filipowicz,W. (2006). Relief of microRNA Mediated Translational Repression in Human Cells Subjected to Stress. Cell 125 1111 1124. Bies,J., Markus,J., and Wolff,L. (2002). Covalent attachment of the SUMO 1 protein to the negative regulatory domain of the c Myb transcription factor modifies its stability and transactivation capacity. J. Biol. Chem. 277 8999 9009. Boj,S.F., Parrizas,M., Maestro,M.A., and Ferrer,J. (2001). A transcription factor regulatory circuit in differentiated pancreatic cells. Proc. Natl. Acad. Sci. U. S. A 98 14481 14486. Broker,L.E., Kruyt,F.A., and Giaccone,G. (2005). Cell death independent of caspases: a review. Clin. Cancer Res. 11 3155 3162. Broome,J.D. (1961). Evidence that the L asparaginase activity of guinea pig serum is responsible for its antilymphoma effects. Nature 191 1114 1115. Broome,J.D. (1968). Studies on the mechanism of tumor inhibition by L asparaginase. J. Exp. Med. 127 1055 1072. Bruhat,A., Jousse,C., Carraro,V., Reimold,A.M., Ferrara ,M., and Fafournoux,P. (2000). Amino acids control mammalian gene transcription: activating transcription factor 2 is essential for the amino acid responsiveness of the CHOP promoter. Mol. Cell Biol. 20 7192 7204. Bruhat,A., Jousse,C., Wang,X. Z., Ron,D., Ferrara,M., and Fafournoux,P. (1997). Amino acid limitation induces expression of CHOP, a CCAAT/enhancer binding protein related gene, at both transcriptional and post transcriptional levels. J. Biol. Chem. 272 17588 17593. Buck,M. and Chojkier,M. (2003) Signal transduction in the liver: C/EBPbeta modulates cell proliferation and survival. Hepatology 37 731 738.

PAGE 179

164 Bussolati,O., Belletti,S., Uggeri,J., Gatti,R., Orlandini,G., Dall'Asta,V., and Gazzola,G.C. (1995). Characterization of apoptotic phenomena in duced by treatment with L asparaginase in NIH3T3 cells. Exper. Cell Res. 220 283 291. Carlson,S.G., Fawcett,T.W., Bartlett,J.D., Bernier,M., and Holbrook,N.L. (1993). Regulation of the C/EBP related gene, gadd 153 by glucose deprivation. Mol. Cell Biol. 13 4736 4744. Cenatiempo,Y. (1986). Prokaryotic gene expression in vitro: transcription translation coupled systems. Biochimie 68 505 515. Cereghini,S. (1996). Liver enriched transcription factors and hepatocyte differentiation. FASEB J. 10 267 282. Chan, S.W. and Egan,P.A. (2005). Hepatitis C virus envelope proteins regulate CHOP via induction of the unfolded protein response. FASEB J. 19 1510 1512. Chen,A., Muzzio,I.A., Malleret,G., Bartsch,D., Verbitsky,M., Pavlidis,P., Yonan,A.L., Vronskaya,S., Grody,M .B., Cepeda,I., Gilliam,T.C., and Kandel,E.R. (2003). Inducible enhancement of memory storage and synaptic plasticity in transgenic mice expressing an inhibitor of ATF4 (CREB 2) and C/EBP proteins. Neuron 39 655 669. Chen,B.P., Wolfgang,C.D., and Hai,T. ( 1996). Analysis of ATF3, a transcription factor induced by physiological stresses and modulated by gadd153/Chop10. Mol. Cell Biol. 16 1157 1168. Chen,H., Pan,Y.X., Dudenhausen,E.E., and Kilberg,M.S. (2004). Amino acid deprivation induces the transcription rate of the human asparagine synthetase gene through a timed program of expression and promoter binding of nutrient responsive bZIP transcription factors as well as localized histone acetylation. J. Biol. Chem. 279 50829 50839. Chevray,P.M. and Nathans,D (1992). Protein interaction cloning in yeast: identification of mammalian proteins that react with the leucine zipper of Jun. Proc. Natl. Acad. Sci. U. S. A 89 5789 5793. Cirillo,L.A., Lin,F.R., Cuesta,I., Friedman,D., Jarnik,M., and Zaret,K.S. (2002). Opening of compacted chromatin by early developmental transcription factors HNF3 (FoxA) and GATA 4. Mol. Cell 9 279 289. Clark,K.L., Halay,E.D., Lai,E., and Burley,S.K. (1993). Co crystal structure of the HNF 3/fork head DNA recognition motif resembles hi stone H5. Nature 364 412 420. Cooney,D.A., Capizzi,R.L., and Handshumacher,R.E. (1970). Evaluation of L asparagine metabolism in animals and man. Cancer. Res. 30 929 935. Costa Mattioli,M., Gobert,D., Stern,E., Gamache,K., Colina,R., Cuello,C., Sossin,W. Kaufman,R., Pelletier,J., Rosenblum,K., Krnjevic,K., Lacaille,J.C., Nader,K., and Sonenberg,N. (2007). eIF2alpha phosphorylation bidirectionally regulates the switch from short to long term synaptic plasticity and memory. Cell 129 195 206.

PAGE 180

165 Dang,C.V. an d Semenza,G.L. (1999). Oncogenic alterations of metabolism. Trends Biochem. Sci. 24 68 72. Dubbers,A., Wurthwein,G., Muller,H.J., Schulze Westhoff,P., Winkelhorst,M., Kurzknabe,E., Lanvers,C., Pieters,R., Kaspers,G.J., Creutzig,U., Ritter,J., and Boos,J. (2000). Asparagine synthetase activity in paediatric acute leukaemias: AML M5 subtype shows lowest activity. Br. J. Haematol. 109 427 429. Duncan,S.A., Navas,M.A., Dufort,D., Rossant,J., and Stoffel,M. (1998). Regulation of a transcription factor network required for differentiation and metabolism. Science 281 692 695. Elefteriou,F., Benson,M.D., Sowa,H., Starbuck,M., Liu,X., Ron,D., Parada,L.F., and Karsenty,G. (2006). ATF4 mediation of NF1 functions in osteoblast reveals a nutritional basis for congenit al skeletal dysplasiae. Cell Metab 4 441 451. Ellem,K.A., Fabrizio,A.M., and Jackson,L. (1970). The dependence of DNA and RNA synthesis on protein synthesis in asparaginase treated lymphoma cells. Cancer Res. 30 515 527. Ellis,L.M. and Hicklin,D.J. (2008 ). VEGF targeted therapy: mechanisms of anti tumour activity. Nat. Rev. Cancer 8 579 591. Eloranta,J.J. and Hurst,H.C. (2002). Transcription factor AP 2 interacts with the SUMO conjugating enzyme UBC9 and is sumolated in vivo. J. Biol. Chem. 277 30798 30 804. Fawcett,T.W., Martindale,J.L., Guyton,K.Z., Hai,T., and Holbrook,N.J. (1999). Complexes containing activating transcription factor (ATF)/cAMP responsive element binding protein (CREB) interact with the CCAAT/enhancer binding protein (C/EBP) ATF compos ite site to regulate Gadd153 expression during the stress response. Biochem. J. 339 135 141. Fernandez,J., Lopez,A.B., Wang,C., Mishra,R., Zhou,L., Yaman,I., Snider,M.D., and Hatzolgou,M. (2003). Transcriptional control of the arginine/lysine transporter, cat 1, by physiological stress. J. Biol. Chem. 278 50000 50009. Fernandez,J., Yaman,I., Mishra,R., Merrick,W.C., Snider,M.D., Lamers,W.H., and Hatzoglou,M. (2001). Internal ribosome entry site mediated translation of a mammalian mRNA is regulated by amin o acid availability. J. Biol. Chem. 276 12285 12291. Fine,B.M., Kaspers,G.J., Ho,M., Loonen,A.H., and Boxer,L.M. (2005). A genome wide view of the in vitro response to l asparaginase in acute lymphoblastic leukemia. Cancer Res. 65 291 299. Fraumeni,J.F., Jr. and Miller,R.W. (1967). Leukemia mortality: downturn rates in the United States. Science 155 1126 1128. Friedman,J.R. and Kaestner,K.H. (2006). The Foxa family of transcription factors in development and metabolism. Cell Mol. Life Sci. 63 2317 2328.

PAGE 181

166 Friedman,J.R., Larris,B., Le,P.P., Peiris,T.H., Arsenlis,A., Schug,J., Tobias,J.W., Kaestner,K.H., and Greenbaum,L.E. (2004). Orthogonal analysis of C/EBPbeta targets in vivo during liver proliferation. Proc. Natl. Acad. Sci. U. S. A 101 12986 12991. Gac cioli,F., Huang,C.C., Wang,C., Bevilacqua,E., Franchi Gazzola,R., Gazzola,G.C., Bussolati,O., Snider,M.D., and Hatzoglou,M. (2006). Amino acid starvation induces the SNAT2 neutral amino acid transporter by a mechanism that involves eIF2alpha phosphorylatio n and cap independent translation. J. Biol. Chem. 281 17929 17940. Gachon,F., Gaudray,G., Thebault,S., Basbous,J., Koffi,J.A., Devaux,C., and Mesnard,J. (2001). The cAMP response element binding protein 2 (CREB 2) can interact with the C/EBP homologous pr otein (CHOP). FEBS Lett. 502 57 62. Gao,N., Zhang,J., Rao,M.A., Case,T.C., Mirosevich,J., Wang,Y., Jin,R., Gupta,A., Rennie,P.S., and Matusik,R.J. (2003). The role of hepatocyte nuclear factor 3 alpha (Forkhead Box A1) and androgen receptor in transcripti onal regulation of prostatic genes. Mol. Endocrinol. 17 1484 1507. Gjymishka,A., Palii,S.S., Shan,J., and Kilberg,M.S. (2008). Despite incr eased ATF4 binding at the C/EBP ATF composite site following activation of the unfolded protein response, system A transporter 2 (SNAT2) transcription activity is repressed in HepG2 cells. J. Biol. Chem. Gong,S.S., Guerrini,L., and Basilico,C. (1991). Regu lation of asparagine synthetase gene expression by amino acid starvation. Mol. Cell. Biol. 11 6059 6066. Greenbaum,L.E., Li,W., Cressman,D.E., Peng,Y., Ciliberto,G., Poli,V., and Taub,R. (1998). CCAAT enhancer binding protein is required for normal hepa tocyte proliferation in mice after partial hepatectomy. J. Clin. Invest 102 996 1007. Guerrini,L., Gong,S.S., Mangasarian,K., and Basilico,C. (1993). Cis and trans acting elements involved in amino acid regulation of asparagine synthetase gene expression Mol. Cell. Biol. 13 3202 3212. Hai,T. and Curran,T. (1991). Cross family dimerization of transcription factors Fos/Jun and ATF/CREB alters DNA binding specificity. Proc. Natl. Acad. Sci. U. S. A 88 3720 3724. Harding,H.P., Novoa,I., Zhang,Y., Zeng,H., Wek,R., Schapira,M., and Ron,D. (2000a). Regulated translation initiation controls stress induced gene expression in mammalian cells. Mol. Cell 6 1099 1108. Harding,H.P., Zhang,Y., Bertolotti,A., Zeng,H., and Ron,D. (2000b). Perk is essential for translat ional regulation and cell survival during the unfolded protein response. Mol. Cell 5 897 904. Harding,H.P., Zhang,Y., and Ron,D. (1999). Protein translation and folding are coupled by an endoplasmic reticulum resident kinase. Nature 397 271 274.

PAGE 182

167 Harding, H.P., Zhang,Y., Zeng,H., Novoa,I., Lu,P.D., Calfon,M., Sadri,N., Yun,C., Popko,B., Paules,R., Stojdl,D.F., Bell,J.C., Hettmann,T., Leiden,J.M., and Ron,D. (2003). An integrated stress response regulates amino acid metabolism and resistance to oxidative str ess. Mol. Cell 11 619 633. Haskell,C.M. and Canellos,G.P. (1969). L asparaginase resistance in human leukemia asparagine synthetase. Biochem. Pharm. 18 2578 2580. He,C.H., Gong,P., Hu,B., Stewart,D., Choi,M.E., Choi,A.M., and Alam,J. (2001). Identificati on of activating transcription factor 4 (ATF4) as an Nrf2 interacting protein. Implication for heme oxygenase 1 gene regulation. J. Biol. Chem. 276 20858 20865. Helmlinger,G., Yuan,F., Dellian,M., and Jain,R.K. (1997). Interstitial pH and pO2 gradients in solid tumors in vivo: high resolution measurements reveal a lack of correlation. Nat. Med. 3 177 182. Herbst,R.S., Nielsch,U., Sladek,F., Lai,E., Babiss,L.E., and Darnell,J.E., Jr. (1991). Differential regulation of hepatocyte enriched transcription fact ors explains changes in albumin and transthyretin gene expression among hepatoma cells. New Biol. 3 289 296. Hershey,J.W. (1991). Translational control in mammalian cells. Annu. Rev. Biochem. 60 717 755. Hiemisch,H., Schutz,G., and Kaestner,K.H. (1997). Transcriptional regulation in endoderm development: characterization of an enhancer controlling Hnf3g expression by transgenesis and targeted mutagenesis. EMBO J. 16 3995 4006. Hinnebusch,A.G. (2005). Translational regulation of GCN4 and the general amino acid control of yeast. Annu. Rev. Microbiol. 59 407 450. Hinnebusch,A.G. (1997). Translational regulation of yeast GCN4. J. Biol. Chem. 272 21661 21664. Hinnebusch,A.G. and Natarajan,K. (2002). Gcn4p, a master regulator of gene expression, is controlled at multiple levels by diverse signals of starvation and stress. Eukaryot. Cell 1 22 32. Hoffer,L.J. (1993). Are dietary protein requirements altered in diabetes mellitus. Can. J. Physiol. Pharmacol. 71 633 638. Holleman,A., Cheok,M.H., den Boer,M.L., Ya ng,W., Veerman,A.J., Kazemier,K.M., Pei,D., Cheng,C., Pui,C.H., Relling,M.V., Janka Schaub,G.E., Pieters,R., and Evans,W.E. (2004). Gene expression patterns in drug resistant acute lymphoblastic leukemia cells and response to treatment. N. Engl. J. Med. 35 1 533 542. Holmqvist,P.H., Belikov,S., Zaret,K.S., and Wrange,O. (2005). FoxA1 binding to the MMTV LTR modulates chromatin structure and transcription. Exp. Cell Res. 304 593 603.

PAGE 183

168 Hope,I.A. and Struhl,K. (1985). GCN4 protein, synthesized in vitro, binds HIS3 regulatory sequences: Implications for general control of amino acid biosynthetic genes in yeast. Cell 43 177 188. Hsu,W., Kerppola,T.K., Chen,P.L., Curran,T., and Chen Kiang,S. (1994). Fos and Jun repress transcription activation by NF IL6 through a ssociation at the basic zipper region. Mol. Cell Biol. 14 268 276. Hutson,R.G., Kitoh,T., Amador,D.A.M., Cosic,S., Schuster,S.M., and Kilberg,M.S. (1997). Amino acid control of asparagine synthetase: relation to asparaginase resistance in human leukemia c ells. Am. J. Physiol. 272 C1691 C1699. Hutson,R.G., Warskulat,U., and Kilberg,M.S. (1996). An example of nutrient control of gene expression: amino acid dependent regulation of asparagine synthetase. Clin. Nutr. 15 327 331. Imae,M., Inoue,Y., Fu,Z., Kato ,H., and Noguchi,T. (2000). Gene expression of the three members of hepatocyte nuclear factor 3 is differentially regulated by nutritional and hormonal factors. J. Endocrinol. 167 R1 R5. Inesi,G., Wade,R., and Rogers,T. (1998). The sarcoplasmic reticulum Ca2+ pump: inhibition by thapsigargin and enhancement by adenovirus mediated gene transfer. Ann. N. Y. Acad. Sci. 853 195 206. Izuishi,K., Kato,K., Ogura,T., Kinoshita,T., and Esumi,H. (2000). Remarkable tolerance of tumor cells to nutrient deprivation: p ossible new biochemical target for cancer therapy. Cancer Res. 60 6201 6207. Jerlstrm,P.G., Bezjak,D.A., Jennings,M.P., and Beacham,I.R. (1989). Structure and expression in Escherichia coli K 12 of the L asparaginase I encoding ansA gene and its flanking regions. Genomics 78 37 46. Jousse,C., Bruhat,A., Carraro,V., Urano,F., Ferrara,M., Ron,D., and Fafournoux,P. (2001). Inhibition of CHOP translation by a peptide encoded by an open reading frame localized in the chop 5'UTR. Nucleic Acids Res. 29 4341 43 51. Jousse,C., Deval,C., Maurin,A.C., Parry,L., Cherasse,Y., Chaveroux,C., Lefloch,R., Lenormand,P., Bruhat,A., and Fafournoux,P. (2007). TRB3 inhibits the transcriptional activation of stress regulated genes by a negative feedback on the ATF4 pathway. J. Biol. Chem. Kaestner,K.H., Hiemisch,H., and Schutz,G. (1998). Targeted disruption of the gene encoding hepatocyte nuclear factor 3gamma results in reduced transcription of hepatocyte specific genes. Mol. Cell Biol. 18 4245 4251. Kaestner,K.H., Katz,J., Li u,Y., Drucker,D.J., and Schutz,G. (1999). Inactivation of the winged helix transcription factor HNF3alpha affects glucose homeostasis and islet glucagon gene expression in vivo. Genes Dev. 13 495 504.

PAGE 184

169 Kaestner,K.H., Knochel,W., and Martinez,D.E. (2000). U nified nomenclature for the winged helix/forkhead transcription factors. Genes Dev. 14 142 146. Kaufman,R.J. (2004). Regulation of mRNA translation by protein folding in the endoplasmic reticulum. Trends Biochem. Sci. 29 152 158. Kawai,T., Matsumoto,M., Takeda,K., Sanjo,H., and Akira,S. (1998). ZIP kinase, a novel serine/threonine kinase which mediates apoptosis. Mol. Cell Biol. 18 1642 1651. KIDD,J.G. (1953). Regression of transplanted lymphomas induced in vivo by means of normal guinea pig serum. I. Co urse of transplanted cancers of various kinds in mice and rats given guinea pig serum, horse serum, or rabbit serum. J. Exp. Med. 98 565 582. Kilberg,M.S., Pan,Y.X., Chen,H., and Leung Pineda,V. (2005). Nutritional control of gene expression: How mammalia n cells respond to amino acid limitation. Annu. Rev. Nutr. 25 59 85. Kiriyama,Y., Kubota,M., Takimoto,T., Kitoh,T., Tanizawa,A., Akiyama,Y., and Mikawa,H. (1989). Biochemical characterization of U937 cells resistant to L asparaginase: the role of asparag ine synthetase. Leukemia 3 294 297. Kitzman,H.H., Jr., McMahon,R.J., Williams,M.G., and Frost,S.C. (1993). Effect of glucose deprivation on GLUT 1 expression in 3T3 L1 adipocytes. J. Biol. Chem. 268 1320 1325. Krejci,O., Starkova,J., Otova,B., Madzo,J., Kalinova,M., Hrusak,O., and Trka,J. (2004). Upregulation of asparagine synthetase fails to avert cell cycle arrest induced by L asparaginase in TEL/AML1 positive leukaemic cells. Leukemia 18 434 441. Laganiere,J., Deblois,G., Lefebvre,C., Bataille,A.R., R obert,F., and Giguere,V. (2005). From the Cover: Location analysis of estrogen receptor alpha target promoters reveals that FOXA1 defines a domain of the estrogen response. Proc. Natl. Acad. Sci. U. S. A 102 11651 11656. Lai,E., Prezioso,V.R., Smith,E., L itvin,O., Costa,R.H., and Darnell,J.E., Jr. (1990). HNF 3A, a hepatocyte enriched transcription factor of novel structure is regulated transcriptionally. Genes Dev. 4 1427 1436. Lantz,K.A., Vatamaniuk,M.Z., Brestelli,J.E., Friedman,J.R., Matschinsky,F.M., and Kaestner,K.H. (2004). Foxa2 regulates multiple pathways of insulin secretion. J. Clin. Invest 114 512 520. Lassot,I., Estrabaud,E., Emiliani,S., Benkirane,M., Benarous,R., and Margottin Goguet,F. (2005). p300 modulates ATF4 stability and transcriptio nal activity independently of its acetyltransferase domain. J. Biol. Chem. Lassot,I., Segeral,E., Berlioz Torrent,C., Durand,H., Groussin,L., Hai,T., Benarous,R., and Margottin Goguet,F. (2001). ATF4 degradation relies on a phosphorylation dependent intera ction with the SCF(betaTrCP) ubiquitin ligase. Mol. Cell Biol. 21 2192 2202.

PAGE 185

170 Lee,C.S., Friedman,J.R., Fulmer,J.T., and Kaestner,K.H. (2005). The initiation of liver development is dependent on Foxa transcription factors. Nature 435 944 947. Lenz,B., Blei ch,S., Beutler,S., Schlierf,B., Schwager,K., Reulbach,U., Kornhuber,J., and Bonsch,D. (2006). Homocysteine regulates expression of Herp by DNA methylation involving the AARE and CREB binding sites. Exp. Cell Res. 312 4049 4055. Leung Pineda,V. and Kilberg ,M.S. (2002). Role of Sp1 and Sp3 in the Nutrient Regulated Expression of the Human Asparagine Synthetase Gene. J. Biol. Chem. 277 16585 16591. Leung Pineda,V., Pan,Y., Chen,H., and Kilberg,M.S. (2004). Induction of p21 and p27 Expression By Amino Acid De privation of HepG2 Human Hepatoma Cells Involves mRNA Stabilization. Biochem. J. 379 79 88. Liang,G. and Hai,T. (1997). Characterization of human activating transcription factor 4, a transcriptional activator that interacts with multiple domains of cAMP r esponsive element binding protein (CREB) binding protein. J. Biol. Chem. 272 24088 24095. Liang,S.H., Zhang,W., McGrath,B.C., Zhang,P., and Cavener,D.R. (2006). PERK (eIF2alpha kinase) is required to activate the stress activated MAPKs and induce the expr ession of immediate early genes upon disruption of ER calcium homoeostasis. Biochem. J. 393 201 209. Lipson,K.E. and Baserga,R. (1989). Transcriptional activity of the human thymidine kinase gene determined by a method using the polymerase chain reaction and an intron specific probe. Proc. Natl. Acad. Sci. U. S. A 86 9774 9777. Liu,Y., Shen,W., Brubaker,P.L., Kaestner,K.H., and Drucker,D.J. (2002). Foxa3 (HNF 3gamma) binds to and activates the rat proglucagon gene promoter but is not essential for progluc agon gene expression. Biochem. J. 366 633 641. Lopez,A.B., Wang,C., Huang,C.C., Yaman,I., Li,Y., Chakravarty,K., Johnson,P.F., Chiang,C.M., Snider,M.D., Wek,R.C., and Hatzoglou,M. (2007). A feedback transcriptional mechanism controls the level of the argi nine/lysine transporter cat 1 during amino acid starvation. Biochem. J. 402 163 173. Lorenzi,P.L., Llamas,J., Gunsior,M., Ozbun,L., Reinhold,W.C., Varma,S., Ji,H., Kim,H., Hutchinson,A.A., Kohn,E.C., Goldsmith,P.K., Birrer,M.J., and Weinstein,J.N. (2008). Asparagine synthetase is a predictive biomarker of L asparaginase activity in ovarian cancer cell lines. Mol. Cancer Ther. 7 3123 3128. Lu,L., Han,A.P., and Chen,J.J. (2001). Translation initiation control by heme regulated eukaryotic initiation factor 2 alpha kinase in erythroid cells under cytoplasmic stresses. Mol. Cell Biol. 21 7971 7980. Lu,P.D., Harding,H.P., and Ron,D. (2004). Translation reinitiation at alternative open reading frames regulates gene expression in an integrated stress response. J. Cell Biol. 167 27 33.

PAGE 186

171 Maita,T. and Matsuda,G. (1980). The primary structure of L asparaginase from Escherichia coli. Hoppe Seylers. Z. Physiol Chem. 361 105 117. Markwell,M.A.K., Haas,S.M., Bieber,L.L., and Tolbert,N.E. (1978). A Modification of the Lowr y Procedure to Simplify Protein Determination in Membrane and Lipoprotein Samples. Anal. Biochem. 87 206 210. Mashburn,L.T. and Wriston,J.C., Jr. (1964). TUMOR INHIBITORY EFFECT OF L ASPARAGINASE FROM ESCHERICHIA COLI. Arch. Biochem. Biophys. 105 450 452 Masuoka,H.C. and Townes,T.M. (2002). Targeted disruption of the activating transcription factor 4 gene results in severe fetal anemia in mice. Blood 99 736 745. Matsumoto,M., Minami,M., Takeda,K., Sakao,Y., and Akira,S. (1996). Ectopic expression of CHO P (GADD153) induces apoptosis in M1 myeloblastic leukemia cells. FEBS Lett. 395 143 147. Maytin,E.V., Ubeda,M., Lin,J.C., and Habener,J.F. (2001). Stress inducible transcription factor CHOP/gadd153 induces apoptosis in mammalian cells via p38 kinase depen dent and independent mechanisms. Exp. Cell Res. 267 193 204. Mazumder,S., Plesca,D., and Almasan,A. (2008). Caspase 3 activation is a critical determinant of genotoxic stress induced apoptosis. Methods Mol. Biol. 414 13 21. McCullough,K.D., Martindale,J .L., Klotz,L.O., Aw,T.Y., and Holbrook,N.J. (2001). Gadd153 sensitizes cells to endoplasmic reticulum stress by down regulating Bcl2 and perturbing the cellular redox state. Mol. Cell Biol. 21 1249 1259. Melchior,F. (2000). SUMO -nonclassical ubiquitin. A nnu. Rev. Cell Dev. Biol. 16 591 626. Melchior,F. and Hengst,L. (2002). SUMO 1 and p53. Cell Cycle 1 245 249. Melchior,F., Schergaut,M., and Pichler,A. (2003). SUMO: ligases, isopeptidases and nuclear pores. Trends Biochem. Sci. 28 612 618. Miller,H.K., Salser,J.S., and Balis,M.E. (1969). Amino acid levels following L asparagine amidohydrolase (EC.3.5.1.1) therapy. Cancer Res. 29 183 187. Minchenko,A., Bauer,T., Salceda,S., and Caro,J. (1994). Hypoxic stimulation of vascular endothelial growth factor ex pression in vitro and in vivo. Lab Invest 71 374 379. Mincheva,A., Lichter,P., Schutz,G., and Kaestner,K.H. (1997). Assignment of the human genes for hepatocyte nuclear factor 3 alpha, beta, and gamma (HNF3A, HNF3B, HNF3G) to 14q12 q13, 20p11, and 19q13 .2 q13.4. Genomics 39 417 419. Mirosevich,J., Gao,N., and Matusik,R.J. (2005). Expression of Foxa transcription factors in the developing and adult murine prostate. Prostate 62 339 352.

PAGE 187

172 Mizock,B.A. (1999). Nutritional support in hepatic encephalopathy. N utrition 15 220 228. Monaghan,A.P., Kaestner,K.H., Grau,E., and Schutz,G. (1993). Postimplantation expression patterns indicate a role for the mouse forkhead/HNF 3 alpha, beta and gamma genes in determination of the definitive endoderm, chordamesoderm and neuroectoderm. Development 119 567 578. Mortimore,G.E. and Poso,A.R. (1988). Amino acid control of intracellular protein degradation. Methods Enzymol. 166 461 476. Muller,S., Hoege,C., Pyrowolakis,G., and Jentsch,S. (2001). SUMO, ubiquitin's mysterious cousin. Nat. Rev. Mol. Cell Biol. 2 202 210. Nagamatsu,S., Sawa,H., Inoue,N., Nakamichi,Y., Takeshima,H., and Hoshino,T. (1994). Gene expression of GLUT3 glucose transporter regulated by glucose in vivo in mouse brain and in vitro in neuronal cell culture s from rat embryos. Biochem. J. 300 ( Pt 1) 125 131. Nakagawa,H., Ohira,M., Hayashi,S., Abe,S., Saito,S., Nagahori,N., Monde,K., Shinohara,Y., Fujitani,N., Kondo,H., Akiyama,S., Nakagawara,A., and Nishimura,S. (2008). Alterations in the glycoform of cispl atin resistant human carcinoma cells are caused by defects in the endoplasmic reticulum associated degradation system. Cancer Lett. 270 295 301. Natarajan,K., Meyer,M.R., Jackson,B.M., Slade,D., Roberts,C., Hinnebusch,A.G., and Marton,M.J. (2001). Transcr iptional profiling shows that Gcn4p is a master regulator of gene expression during amino acid starvation in yeast. Mol. Cell Biol. 21 4347 4368. Nesbit,M.E., Ertel,I., and Hammond,G.D. (1981). L Asparaginase as a single agent in acute lymphocytic leukemi a: survey of studies form Childrens Cancer Study Group. Cancer Treat. Rep. 65 Suppl 4 101 107. Nishitoh,H., Matsuzawa,A., Tobiume,K., Saegusa,K., Takeda,K., Inoue,K., Hori,S., Kakizuka,A., and Ichijo,H. (2002). ASK1 is essential for endoplasmic reticulum stress induced neuronal cell death triggered by expanded polyglutamine repeats. Genes Dev. 16 1345 1355. Nishizawa,M. and Nagata,S. (1992). cDNA clones encoding leucine zipper proteins which interact with G CSF gene promoter element 1 binding protein. FEB S Lett. 299 36 38. Oettgen,H.F. and Schulten,H.K. (1969). Hemmung maligner neoplasien des menschen durch L asparagine. Klin Wochenschr 47 65. Ohoka,N., Yoshii,S., Hattori,T., Onozaki,K., and Hayashi,H. (2005). TRB3, a novel ER stress inducible gene, is i nduced via ATF4 CHOP pathway and is involved in cell death. EMBO J. 24 1243 1255. Okada,T., Yoshida,H., Akazawa,R., Negishi,M., and Mori,K. (2002). Distinct roles of activating transcription factor 6 (ATF6) and double stranded RNA activated protein kinase like endoplasmic reticulum kinase (PERK) in transcription during the mammalian unfolded protein response. Biochem. J. 366 585 594.

PAGE 188

173 Oyadomari,S. and Mori,M. (2004). Roles of CHOP/GADD153 in endoplasmic reticulum stress. Cell Death. Differ. 11 381 389. Oy adomari,S., Takeda,K., Takiguchi,M., Gotoh,T., Matsumoto,M., Wada,I., Akira,S., Araki,E., and Mori,M. (2001). Nitric oxide induced apoptosis in pancreatic beta cells is mediated by the endoplasmic reticulum stress pathway. Proc. Natl. Acad. Sci. U. S. A 98 10845 10850. Palii,S.S., Chen,H., and Kilberg,M.S. (2004). Transcriptional control of the human sodium coupled neutral amino acid transporter system A gene by amino acid availability is mediated by an intronic element. J. Biol. Chem. 279 3463 3471. Pali i,S.S., Thiaville,M.M., Pan,Y.X., Zhong,C., and Kilberg,M.S. (2006). Characterization of the amino acid response element within the human SNAT2 system A transporter gene. Biochem. J. 395 517 527. Pan,Y. X., Chen,H., Siu,F., and Kilberg,M.S. (2003). Amino acid deprivation and endoplasmic reticulum stress induce expression of multiple ATF3 mRNA species which, when overexpressed in HepG2 cells, modulate transcription by the human asparagine synthetase promoter. J. Biol. Chem. 278 38402 38412. Pan,Y.X., Chen, H., Thiaville,M.M., and Kilberg,M.S. (2007). Activation of the ATF3 gene through a co ordinated amino acid sensing response programme that controls transcriptional regulation of responsive genes following amino acid limitation. Biochem. J. 401 299 307. Pa ni,L., Overdier,D.G., Porcella,A., Qian,X., Lai,E., and Costa,R.H. (1992). Hepatocyte nuclear factor 3 beta contains two transcriptional activation domains, one of which is novel and conserved with the Drosophila fork head protein. Mol. Cell Biol. 12 3723 3732. Parker,R., Phan,T., Baumeister,P., Roy,B., Cheriyath,V., Roy,A.L., and Lee,A.S. (2001). Identification of TFII I as the endoplasmic reticulum stress response element binding factor ERSF: its autoregulation by stress and interaction with ATF6. Mol. C ell Biol. 21 3220 3233. Patil,C. and Walter,P. (2001). Intracellular signaling from the endoplasmic reticulum to the nucleus: the unfolded protein response in yeast and mammals. Curr. Opin. Cell Biol 13 349 355. Pierce,M.I., Borges,W.H., Heyn,R., Wolff,J .A., and Gilbert,E.S. (1969). Epidemiological factors and survival experience in 1770 children with acute leukemia. Treated by members of Children's Study Group A between 1957 and 1964. Cancer 23 1296 1304. Poukka,H., Karvonen,U., Janne,O.A., and Palvimo, J.J. (2000). Covalent modification of the androgen receptor by small ubiquitin like modifier 1 (SUMO 1). Proc. Natl. Acad. Sci. U. S. A 97 14145 14150. Pouyssgur,J., Shiu,R.P.C., and Pastan,I. (1977). Induction of two transformation sensitive membrane po lypeptides in normal fibroblasts by a block in glycoprotein synthesis or glucose deprivation. Cell 11 941 947.

PAGE 189

174 Ramji,D.P. and Foka,P. (2002). CCAAT/enhancer binding proteins: structure, function and regulation. Biochem. J. 365 561 575. Ries,L.A.G., Percy ,C.L., and Bunin,G.R. (1999). Introduction. In L.A.G.Ries, M.A.Smith, J.G.Gurney, M.Linet, T.Tamara, and J.T.Young, eds. ( : ), pp. 1 15. Roediger,W.E.W. (1995). New views on the pathogenesis of Kwashiorkor methionine and other amino acids. J. Pediatr. Gastroenterol. Nutr. 21 130 136. Rogers,T.B., Inesi,G., Wade,R., and Lederer,W.J. (1995). Use of thapsigargin to study C a2+ homeostasis in cardiac cells. Biosci. Rep. 15 341 349. Ron,D. and Habener,J.F. (1992). CHOP, a novel developmentally regulated nuclear protein that dimerizes with transcription factors C/EBP and LAP and functions as a dominant negative inhibitor of ge ne transcription. Gene. Devel. 6 439 453. Ross,S., Best,J.L., Zon,L.I., and Gill,G. (2002). SUMO 1 modification represses Sp3 transcriptional activation and modulates its subnuclear localization. Mol. Cell 10 831 842. Rotoli,B., Uggeri,J., Dall'Asta,V., Visigalli,R., Barilli,A., Gatti,R., Orlandini,G., Gazzola,G., and Bussolati,O. (2005). Inhibition of glutamine synthetase triggers apoptosis in asparaginase resistant cells. Cell Physiol Biochem. 15 281 292. Roybal,C.N., Hunsaker,L.A., Barbash,O., Jagt,D. L., and Abcouwer,S.F. (2005). The Oxidative Stressor Arsenite Activates Vascular Endothelial Growth Factor mRNA Transcription by an ATF4 dependent Mechanism. J. Biol. Chem. 280 20331 20339. Rubnitz,J.E., Downing,J.R., Pui,C.H., Shurtleff,S.A., Raimondi,S. C., Evans,W.E., Head,D.R., Crist,W.M., Rivera,G.K., Hancock,M.L., Boyett,J.M., Buijs,A., Grosveld,G., and Behm,F.G. (1997). TEL gene rearrangement in acute lymphoblastic leukemia: a new genetic marker with prognostic significance. J. Clin. Oncol. 15 1150 1157. Ryan,W.L. and Dworak,J.E. (1970). Amino acids of the 6C3HED lymphosarcoma following treatment with asparaginase. Cancer Res. 30 1206 1209. Saeki,K., Yuo,A., Isemura,M., Abe,I., Seki,T., and Noguchi,H. (2000). Apoptosis inducing activity of lipid der ivatives of gallic acid. Biol. Pharm. Bull. 23 1391 1394. Sapetschnig,A., Rischitor,G., Braun,H., Doll,A., Schergaut,M., Melchior,F., and Suske,G. (2002). Transcription factor Sp3 is silenced through SUMO modification by PIAS1. EMBO J. 21 5206 5215. Sasa ki,H. and Hogan,B.L. (1993). Differential expression of multiple fork head related genes during gastrulation and axial pattern formation in the mouse embryo. Development 118 47 59.

PAGE 190

175 Scherf,U., Ross,D.T., Waltham,M., Smith,L.H., Lee,J.K., Tanabe,L., Kohn,K. W., Reinhold,W.C., Myers,T.G., Andrews,D.T., Scudiero,D.A., Eisen,M.B., Sausville,E.A., Pommier,Y., Botstein,D., Brown,P.O., and Weinstein,J.N. (2000). A gene expression database for the molecular pharmacology of cancer. Nat. Genet. 24 236 244. Scheuner,D ., Song,B., McEwen,E., Liu,C., Laybutt,R., Gillespie,P., Saunders,T., Bonner Weir,S., and Kaufman,R.J. (2001). Translational control is required for the unfolded protein response and in vivo glucose homeostasis. Mol. Cell 7 1165 1176. Shen,W., Scearce,L.M ., Brestelli,J.E., Sund,N.J., and Kaestner,K.H. (2001). Foxa3 (hepatocyte nuclear factor 3gamma ) is required for the regulation of hepatic GLUT2 expression and the maintenance of glucose homeostasis during a prolonged fast. J. Biol. Chem. 276 42812 42817 Shih,D.Q., Navas,M.A., Kuwajima,S., Duncan,S.A., and Stoffel,M. (1999). Impaired glucose homeostasis and neonatal mortality in hepatocyte nuclear factor 3alpha deficient mice. Proc. Natl. Acad. Sci. U. S. A 96 10152 10157. Shimizu,M., Nomura,Y., Suzuki, H., Ichikawa,E., Takeuchi,A., Suzuki,M., Nakamura,T., Nakajima,T., and Oda,K. (1998). Activation of the rat cyclin A promoter by ATF2 and Jun family members and its suppression by ATF4. Exp. Cell Res. 239 93 103. Shweiki,D., Itin,A., Soffer,D., and Keshet ,E. (1992). Vascular endothelial growth factor induced by hypoxia may mediate hypoxia initiated angiogenesis. Nature 359 843 845. Sidrauski,C., Chapman,R., and Walter,P. (1998). The unfolded protein response: an intracellular signalling pathway with many surprising features. Cell Biol. 8 245 249. Siu,F., Bain,P.J., LeBlanc Chaffin,R., Chen,H., and Kilberg,M.S. (2002). ATF4 is a mediator of the nutrient sensing response pathway that activates the human asparagine synthetase gene. J. Biol. Chem. 277 24120 24127. Siu,F.Y., Chen,C., Zhong,C., and Kilberg,M.S. (2001). CCAAT/enhancer binding protein beta (C/EBPb) is a mediator of the nutrient sensing response pathway that activates the human asparagine synthetase gene. J. Biol. Chem. 276 48100 48107. Sood,R., Porter,A.C., Olsen,D.A., Cavener,D.R., and Wek,R.C. (2000). A mammalian homologue of GCN2 protein kinase important for translational control by phosphorylation of eukaryotic initiation factor 2 Genetics 154 787 801. Stams,W.A., den Boer,M.L., Beverloo,H .B., Meijerink,J.P., Stigter,R.L., Van Wering,E.R., Janka Schaub,G.E., Slater,R., and Pieters,R. (2003). Sensitivity to L asparaginase is not associated with expression levels of asparagine synthetase in t(12;21)+ pediatric ALL. Blood 101 2743 2747. Stams ,W.A., den Boer,M.L., Beverloo,H.B., Van Wering,E.R., and Pieters,R. (2004). Upregulation of asparagine synthetase and cell cycle arrest in t(12;21) positive ALL. Leukemia 19 318 319.

PAGE 191

176 Story,M.D., Voehringer,D.W., Stephens,L.C., and Meyn,R.E. (1993). L Asp araginase kills lymphoma cells by apoptosis. Cancer Chem. Pharm. 32 129 133. Sund,N.J., Ang,S.L., Sackett,S.D., Shen,W., Daigle,N., Magnuson,M.A., and Kaestner,K.H. (2000). Hepatocyte nuclear factor 3beta (Foxa2) is dispensable for maintaining the differe ntiated state of the adult hepatocyte. Mol. Cell Biol. 20 5175 5183. Tanaka,T., Tsujimura,T., Takeda,K., Sugihara,A., Maekawa,A., Terada,N., Yoshida,N., and Akira,S. (1998). Targeted disruption of ATF4 discloses its essential role in the formation of eye lens fibres. Genes Cells 3 801 810. Thiaville,M.M., Dudenhausen,E.E., Zhong,C., Pan,Y.X., and Kilberg,M.S. (2008). Deprivation of protein or amino acid induces C/EBPbeta synthesis and binding to amino acid response elements, but its action is not an absol ute requirement for enhanced transcription. Biochem. J. 410 473 484. Tirasophon,W., Welihinda,A.A., and Kaufman,R.J. (1998). A stress response pathway from the endoplasmic reticulum to the nucleus requires a novel bifunctional protein kinase/endoribonucle ase (Ire1p) in mammalian cells. Gene. Devel. 12 1812 1824. Ubeda,M. and Habener,J.F. (2000). CHOP gene expression in response to endoplasmic reticular stress requires NFY interaction with different domains of a conserved DNA binding element. Nucleic Acids Res. 28 4987 4997. Urano,F., Wang,X., Bertolotti,A., Zhang,Y., Chung,P., Harding,H.P., and Ron,D. (2000). Coupling of stress in the ER to activation of JNK protein kinases by transmembrane protein kinase IRE1. Science 287 664 666. Vallejo,M., Ron,D., Mi ller,C.P., and Habener,J.F. (1993). C/ATF, a member of the activating transcription factor family of DNA binding proteins, dimerizes with CAAT/enhancer binding proteins and directs their binding to cAMP response elements. Proc. Natl. Acad. Sci. U. S. A 90 4679 4683. Vattem,K.M. and Wek,R.C. (2004). Reinitiation involving upstream ORFs regulates ATF4 mRNA translation in mammalian cells. Proc. Natl. Acad. Sci. U. S. A 101 11269 11274. Wang,X.Z. and Ron,D. (1996). Stress induced phosphorylation and activatio n of the transcription factor CHOP (GADD153) by p38 MAP Kinase. Science 272 1347 1349. Wang,X. Z., Harding,H.P., Zhang,Y., Jolicoeur,E.M., Kuroda,M., and Ron,D. (1998a). Cloning of mammalian Ire1 reveals diversity in the ER stress responses. EMBO J. 17 5 708 5717. Wang,X. Z., Kuroda,M., Sok,J., Batchvarova,N., Kimmel,R., Chung,P., Zinszner,H., and Ron,D. (1998b). Identification of novel stress induced genes downstream of chop EMBO J. 17 3619 3630.

PAGE 192

177 Wei,M.C., Zong,W.X., Cheng,E.H., Lindsten,T., Panoutsakop oulou,V., Ross,A.J., Roth,K.A., MacGregor,G.R., Thompson,C.B., and Korsmeyer,S.J. (2001). Proapoptotic BAX and BAK: a requisite gateway to mitochondrial dysfunction and death. Science 292 727 730. Weigel,D. and Jackle,H. (1990). The fork head domain: a no vel DNA binding motif of eukaryotic transcription factors? Cell 63 455 456. Weigel,D., Jurgens,G., Kuttner,F., Seifert,E., and Jackle,H. (1989). The homeotic gene fork head encodes a nuclear protein and is expressed in the terminal regions of the Drosophi la embryo. Cell 57 645 658. Wertheimer,E., Sasson,S., Cerasi,E., and Ben Neriah,Y. (1991). The ubiquitous glucose transporter GLUT 1 belongs to the glucose regulated protein family of stress inducible proteins. Proc. Natl. Acad. Sci. U. S. A 88 2525 2529 Wolfgang,C.D., Chen,B.P., Martindale,J.L., Holbrook,N.J., and Hai,T. (1997). gadd153/Chop10, a potential target gene of the transcriptional repressor ATF3. Mol. Cell Biol. 17 6700 6707. Wolfrum,C., Asilmaz,E., Luca,E., Friedman,J.M., and Stoffel,M. (200 4). Foxa2 regulates lipid metabolism and ketogenesis in the liver during fasting and in diabetes. Nature 432 1027 1032. Wolfrum,C., Shih,D.Q., Kuwajima,S., Norris,A.W., Kahn,C.R., and Stoffel,M. (2003). Role of Foxa 2 in adipocyte metabolism and different iation. J. Clin. Invest 112 345 356. Wriston,J.C., Jr. (1985). Asparaginase. Methods Enzymol. 113 608 618. Wriston,J.C., Jr. and Yellin,T.O. (1973). L asparaginase: a review. Adv. Enzymol. Relat Areas Mol. Biol. 39 185 248. Xu,Y., Kim,S.O., Li,Y., and H an,J. (2006). Autophagy contributes to caspase independent macrophage cell death. J. Biol. Chem. 281 19179 19187. Yaman,I., Fernandez,J., Liu,H., Caprara,M., Komar,A.A., Koromilas,A.E., Zhou,L., Snider,M.D., Scheuner,D., Kaufman,R.J., and Hatzoglou,M. (20 03). The Zipper Model of Translational Control. A Small Upstream ORF Is the Switch that Controls Structural Remodeling of an mRNA Leader. Cell 113 519 531. Yang,X., Matsuda,K., Bialek,P., Jacquot,S., Masuoka,H.C., Schinke,T., Li,L., Brancorsini,S., Sasson e Corsi,P., Townes,T.M., Hanauer,A., and Karsenty,G. (2004). ATF4 is a substrate of RSK2 and an essential regulator of osteoblast biology; implication for Coffin Lowry Syndrome. Cell 117 387 398. Yoneda,T., Imaizumi,K., Oono,K., Yui,D., Gomi,F., Katayama, T., and Tohyama,M. (2001). Activation of caspase 12, an endoplastic reticulum (ER) resident caspase, through tumor necrosis factor receptor associated factor 2 dependent mechanism in response to the ER stress. J. Biol. Chem. 276 13935 13940.

PAGE 193

178 Yoon,S., Govi nd,C.K., Qiu,H., Kim,S.J., Dong,J., and Hinnebusch,A.G. (2004). Recruitment of the ArgR/Mcm1p repressor is stimulated by the activator Gcn4p: a self checking activation mechanism. Proc. Natl. Acad. Sci. U. S. A 101 11713 11718. Yoshida,H., Haze,K., Yanagi ,H., Yura,T., and Mori,K. (1998). Identification of the cis acting endoplasmic reticulum stress response element responsible for transcriptional induction of mammalian glucose regulated proteins. J. Biol. Chem. 273 33741 33749. Yoshida,H., Matsui,T., Yama moto,A., Okada,T., and Mori,K. (2001). XBP1 mRNA is induced by ATF6 and spliced by IRE1 in response to ER stress to produce a highly active transcription factor. Cell 107 881 891. Yoshida,H., Okada,T., Haze,K., Yanagi,H., Yura,T., Negishi,M., and Mori,K. (2000). ATF6 activated by proteolysis binds in the presence of NF Y (CBF) directly to the cis acting element responsible for the mammalian unfolded protein response. Mol. Cell Biol. 20 6755 6767. Yu,V.W., Ambartsoumian,G., Verlinden,L., Moir,J.M., Prud'ho mme,J., Gauthier,C., Roughley,P.J., and St Arnaud,R. (2005). FIAT represses ATF4 mediated transcription to regulate bone mass in transgenic mice. J. Cell Biol. 169 591 601. Yuan,J. and Yankner,B.A. (2000). Apoptosis in the nervous system. Nature 407 802 809. Zhang,L., Rubins,N.E., Ahima,R.S., Greenbaum,L.E., and Kaestner,K.H. (2005). Foxa2 integrates the transcriptional response of the hepatocyte to fasting. Cell Metab 2 141 148. Zhang,P., McGrath,B.C., Reinert,J., Olsen,D.S., Lei,L., Gill,S., Wek,S.A., Vattem,K.M., Wek,R.C., Kimball,S.R., Jefferson,L.S., and Cavener,D.R. (2002). The GCN2 eIF2alpha kinase is required for adaptation to amino acid deprivation in mice. Mol. Cell Biol. 22 6681 6688. Zhong,C., Chen,C., and Kilberg,M.S. (2003). Characterizatio n of the nutrient sensing response unit in the human asparagine synthetase promoter. Biochem. J. 372 603 609. Zhou,D., Palam,L.R., Jiang,L., Narasimhan,J., Staschke,K.A., and Wek,R.C. (2008). Phosphorylation of eIF2 Directs ATF5 Translational Control in R esponse to Diverse Stress Conditions. J. Biol. Chem. 283 7064 7073. Zimmerman,J.A., Malloy,V., Krajcik,R., and Orentreich,N. (2003). Nutritional control of aging. Exp. Gerontol. 38 47 52. Zinszner,H., Kuroda,M., Wang,X.Z., Batchvarova,N., Lightfoot,R.T., Remotti,H., Stevens,J.L., and Ron,D. (1998). CHOP is implicated in programmed cell death in response to impaired function of the endoplasmic reticulum. Gene. Devel. 12 982 995.

PAGE 194

179 BIOGRAPHICAL SKETCH Nan Su was was born in 1980 in Beijing, the capital ci the only child of Xuefei Su and Xiaoning Fu. In 1999, he graduated from high school, and was admitted by Tsinghua University as a recommended student. He enrolled in the Department of Biological Science and Biotechnol ogy, completed his graduation thesis under the supervition of Dr. Zihe Rao, and graduated with a Bachelor of Science degree in July 2003. In the same month of graduation, he married his college classmate Xiaolei Qiu. In 2004, he traveled from China to th e United States of America, and enrolled as a graduate student in the Interdisciplinary Program in Biomedical Sciences at the University of Florida. He joined the laboratory of Dr. Michael Kilberg in May 2005 and became an official PhD candidate in Novermb er 2006. He finished his thesis in November 2008 under the supervision of Dr. Kilberg, and graduated with a Doctor of Philosophy degree in December 2008.