The Regulation and Membrane Topology of DHCR24, a Key Enzyme in Cholesterol Synthesis

The regulation and membrane topology of DHCR24, a key enzyme in cholesterol synthesis

Eser J. Zerenturk

A thesis submitted in fulfilment of the requirements for the Degree of Doctorate of Philosophy (Biochemistry and Molecular Genetics)

School of Biotechnology and Biomolecular Sciences The University of New South Wales

Submitted: December 2013 Revised: April 2014

THE UNIVERSITY OF NEW SOUTH WALES Thesis/Dissertation Sheet

Surname or Family name: Zerenturk

First name: Eser Other name/s: Jane

Abbreviation for degree as given in the University calendar: PhD

School: Biotechnology and Biomolecular Sciences Faculty: Science

Title: The regulation and membrane topology of DHCR24, a key enzyme in cholesterol synthesis

Abstract 350 words maximum:

Cholesterol is necessary for mammalian life, as an essential component in cell membranes, foetal development, and a precursor for steroid hormones. Hence, cholesterol levels must be tightly regulated. Previous research has focused on an early step in cholesterol synthesis: 3- hydroxy-3-methylglutaryl-coenzyme A reductase (HMGCR), target of the cholesterol-lowering statin drugs. However, less is known about other steps in the pathway. We investigated 3β-hydroxysterol Δ24-reductase (DHCR24); involved in the last step of cholesterol synthesis, and implicated in inflammation, oxidative stress and hepatitis C virus infection. There is a paucity of fundamental information on the structure of DHCR24 and how it interacts with cellular membranes, as well as how this critical enzyme is regulated.

We found that DHCR24 is an integral endoplasmic reticulum (ER) membrane protein, with multiple atypical membrane associated regions. We present biochemical evidence that the majority of the enzyme is associated with the ER membrane, contrary to published membrane topology models and in silico predictions. This has important consequences for the many functions attributed to DHCR24. In particular, those that suggest DHCR24 alters its localisation within the cell should be reassessed in light of this new information. We show that DHCR24 is regulated at multiple levels, with potent effects on cholesterol synthesis. Our findings demonstrate feedback regulation at the transcriptional and post‑translational level. Transcriptional regulation occurs through sterol regulatory element binding protein 2 (SREBP-2), mediated by dual sterol regulatory elements (SREs) within the DHCR24 promoter, which work cooperatively to regulate expression. DHCR24 activity is potently inhibited at the post‑translational level by endogenous side-chain oxysterols, in particular 24(S),25‑epoxycholesterol (24,25EC), which was independent of DHCR24 protein levels. This observation was extended to another structurally similar byproduct of the cholesterol synthesis pathway, progesterone. Another mode of regulation is through signalling. We found that phosphorylation affects DHCR24 activity, at a known phosphorylation site, T110. We found that protein kinase C (PKC) also ablated DHCR24 activity, through an unknown phosphorylation site.

These findings provide fundamental new insights into DHCR24 and its regulation, indicating it may be an important regulatory step in cholesterol synthesis and maintaining cellular cholesterol homeostasis.

Declaration relating to disposition of project thesis/dissertation

I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or in part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all property rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation.

I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstracts International (this is applicable to doctoral theses only).

16/05/2014 …………………………………………………………… ……………………………………..……………… ……….……………………...…….… Signature Witness Date

The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research.

FOR OFFICE USE ONLY Date of completion of requirements for Award:

ORIGINALITY STATEMENT ‘I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.’

Signed …………………………………………… Date ……………………………………………

COPYRIGHT STATEMENT ‘I hereby grant the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all proprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation. I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstract International (this is applicable to doctoral theses only). I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted I have applied/will apply for a partial restriction of the digital copy of my thesis or dissertation.'

Signed …………………………………………… Date ……………………………………………

AUTHENTICITY STATEMENT ‘I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the conversion to digital format.’

Signed …………………………………………… Date ……………………………………………

ABSTRACT

Cholesterol is necessary for mammalian life, as an essential component in cell membranes, fetal development, and a precursor for steroid hormones. Hence, cholesterol levels must be tightly regulated. Previous research has focused on an early step in cholesterol synthesis: 3-hydroxy-3-methylglutaryl-coenzyme A reductase (HMGCR), target of the cholesterol-lowering statin drugs. However, less is known about other steps in the pathway. We investigated 3β-hydroxysterol Δ24-reductase (DHCR24); involved in the last step of cholesterol synthesis, and implicated in inflammation, oxidative stress and hepatitis C infection infection. There is a paucity of fundamental information on the structure DHCR24 and how it interacts with cellular membranes, as well as how this critical enzyme is regulated. We found that DHCR24 is an integral endoplasmic reticulum (ER) membrane protein, with multiple atypical membrane associated regions. Prediction programs and previous studies have shown conflicting results regarding which regions of DHCR24 are associated with the membrane, although there was general agreement that this was limited to only the N-terminal portion. Here, we present biochemical evidence that in fact the majority of the enzyme is associated with the ER membrane. This has important consequences for the many functions attributed to DHCR24. In particular, those that suggest DHCR24 alters its localisation within the cell should be reassessed in light of this new information. We show that DHCR24 is regulated at multiple levels, with potent effects on cholesterol synthesis. Our findings demonstrate feedback regulation at the transcriptional and post-translational level. Transcriptional regulation occurs through dual sterol regulatory elements (SREs) within the DHCR24 promoter, which work cooperatively to regulate expression. DHCR24 is potently inhibited at the post-translational level by endogenous side-chain oxysterols, in particular 24(S),25-epoxycholesterol, and the steroid hormone, progesterone. Another mode of regulation is through signalling. We found that phosphorylation affects DHCR24 activity, at a known phosphorylation site, T110. We found that protein kinase C (PKC) also ablated DHCR24 activity, through an unknown phosphorylation site. These findings indicate DHCR24 may be an important regulatory step in cholesterol synthesis and maintaining cellular cholesterol homeostasis.

iii

LIST OF PUBLICATIONS

Zerenturk E.J.*, Kristiana I.*, Gill S., Brown A.J. (2012) The endogenous regulator 24(S),25-epoxycholesterol inhibits cholesterol synthesis at DHCR24 (Seladin-1). Biochim Biophys Acta 1821: 1269-1277.

Zerenturk E.J.*, Sharpe L.J.*, Brown A.J. (2012) Sterols regulate 3β-hydroxysterol Δ24-reductase (DHCR24) via dual sterol regulatory elements: cooperative induction of key enzymes in lipid synthesis by Sterol Regulatory Element Binding Proteins. Biochim Biophys Acta 1821: 1350-1360.

Zerenturk E.J., Sharpe L.J., Ikonen, E., Brown A.J. (2013) Desmosterol and DHCR24: Unexpected new directions for a terminal step in cholesterol synthesis. Prog Lipid Res, 52: 666-680

Luu, W.*, Zerenturk E.J.*, Kristiana I., Sharpe L.J., Brown A.J. (2013) Signalling regulates activity of DHCR24, the final enzyme in cholesterol synthesis, DHCR24. J Lipid Res, 55(3): 410-420.

Zerenturk E.J., Sharpe L.J., Brown A.J. (2013) DHCR24 associates strongly with the Endoplasmic Reticulum beyond predicted membrane domains: Implications for the Activities of this Multi-Functional Enzyme. Biosci Rep, 34(2): 107-117.

*These authors contributed equally to this work.

PRESENTATIONS

Oral Presentations

Zerenturk, E.J., Kristiana, I., Gill, S., Brown, A.J. The endogenous regulator 24(S),25-epoxycholesterol inhibits cholesterol synthesis at DHCR24 (Seladin-1). XVI International Symposium on Atherosclerosis, Sydney, Australia, March 2012

Poster Presentations

Sharpe, L.J., Zerenturk, E.J., Brown, A.J. Sterol-dependent transcriptional regulation of DHCR24: a key enzyme in Alzheimer’s Disease and cholesterol synthesis. XVI International Symposium on Atherosclerosis, Sydney, Australia, March 2012

Zerenturk, E.J., Sharpe, L.J., Brown, A.J. Sterols Regulate 3β-Hydroxysterol Δ24-Reductase (DHCR24) via Dual Sterol Regulatory Elements: Cooperative Induction of Key Enzymes in Lipid Synthesis by Sterol-regulatory Element-binding Proteins. Frontiers in Lipid Biology (American Society of Biochemistry and Molecular Biology (ASBMB), International Conference of the Bioscience of Lipids (ICBL)) Conference, Banff, Alberta, Canada, September 2012.

Zerenturk, E.J., Kristiana, I., Gill, S., Brown, A.J. The Endogenous Regulator 24(S),25-Epoxycholesterol Inhibits Cholesterol Synthesis at DHCR24 (Seladin-1). Frontiers in Lipid Biology (ASBMB, ICBL) Conference, Banff, Alberta, Canada, September 2012.

Zerenturk, E.J., Kristiana, I., Sharpe, L.J., Brown, A.J. DHCR24: A new regulatory step in cholesterol synthesis. Experimental Biology (American Society of Biochemistry and Molecular Biology) Conference, Boston, MA, USA, April 2013

Zerenturk, E.J., Kristiana, I., Sharpe, L.J., Brown, A.J. DHCR24: A new regulatory step in cholesterol synthesis. Australian Society for Medical Research (ASMR) Conference, Sydney, Australia, June 2013.

ACKNOWLEDGEMENTS

Many people need to be acknowledged in this thesis, so, I start off with the Brown lab. My time in room 257 has been very memorable, and I am sure I will be tearing up when I eventually have to leave the nest. Andrew, you are the best supervisor and mentor a student could ask for. I could go into lots of detail about your support and guidance, but I don’t want to embarrass/bore you. Suffice it to say, I will miss your lame dad jokes. Laura, leaving the annex to sit across from you was the best decision ever. I would call you mother duck, but I don’t want to get bashed. But your help and advice has been much appreciated over the years. And we have had fun too, what with writing our paper - cloning is the best! To James, sitting behind a genius was very rewarding – you know everything! Even your non-geniusy side was great. Anytime I wanted to talk about lame things, and everyone ran away, you always listened and offered advice. To Winnie, team EW was a great collaboration, and one we should have started earlier . Can we talk about nails and fashion now that we are finished? To all the newbs (Anika, Lisa, Vicky, Cyril) you all brought a breath of fresh air to the lab, which I appreciated more than you will know. It was great working (and not working) with you all. Anika, I will miss your smartass comments (they are better than mine), and Lisa, I miss your endless cheerfulness, even though E. coli destroyed your soul. Can you please come back and train me now that I am fat from writing? To Julian, why did you destroy Lisa’s soul – she’s never coming back! And who am I going to argue with when I leave? It makes me a little sad? Cyril, you are lame (x 1000) – have fun setting up your Twerking and Jazzercise Academy in Canberra. To those who have left, Ika and Saloni, thanks for your all your help. My TLCs never looked as good as both of yours . Thanks to the Yang and Dawes lab, for all their yeasty help, including yeast advice and foodsies for the critters. Special thanks to Jack, Anita, Martin, Ian and Robert. Too bad none of that work ever saw the light of day – at least I can say I know how to work with yeast now! To the Bioanalytical Mass Spectrometry Facility, especially Martin Bucknall, thanks for all the help and advice and patience to teach me GC-MS. Machines are awesome. These experiments really saved the day! To my review panel, Louise and Marc, you have both been so kind and helpful as my PhD progressed.

vii

To BABS, past and present staff and students. These past few years have been memorable. I will be sad when I leave this place. Although it feels like I have been here for ages, I never got to see the Jacaranda tree again. To my family. You have had so much patience with me these past few years, whilst I have been extremely absent. So many babies were born – now I can see them (I am talking about you Marley!). To my mum, you are so supportive; I couldn’t have done this without you. You’re the best!!! To Alev, sorry for being poor and the worst maid of honour. Promise I will quickly become the best maid of honour ever once I submit this bad boy!!! To Auda, my time as an honours and PhD student would have been very different (and lonely) without you. Having my best friend just across the hall guaranteed I was always happy. Insert mushiness. To my friends, sorry for being lame. Now I can come out of hibernation and the Dogbox can once more be reunited! Lorenzo, you’re the best doggy anyone could ask for. Thanks for sitting by me all the times I have had to chain myself to my desk and write, even though you ate all my socks. Thanks for the help with the statistical analysis. Can I dedicate my thesis to my dog? Yes, I can - love you to bits Renny!!!

viii

TABLE OF CONTENTS

Contents

1 GENERAL INTRODUCTION ...... 3

1.1 INTRODUCTION ...... 3

1.2 CHOLESTEROL ...... 3

1.3 CHOLESTEROL HOMEOSTASIS ...... 4

1.4 CHOLESTEROL SYNTHESIS ...... 5 3-HYDROXY-3-METHYLGLUTARYL-COENZYME A REDUCTASE ...... 5

1.5 REGULATION OF CHOLESTEROL SYNTHESIS ...... 7 1.5.1 STEROL-MEDIATED TRANSCRIPTIONAL REGULATION ...... 7

1.5.2 POST-TRANSLATIONAL REGULATION AT RATE-LIMITING ENZYMES ...... 10 1.6 DHCR24 ...... 14

1.6.1 FUNCTIONS OF DHCR24 ...... 14

1.6.2 DHCR24 IN DISEASE ...... 15

1.7 DHCR24 CHARACTERISATION ...... 16 1.7.1 DISCOVERY OF DHCR24 ...... 16

1.7.2 DHCR24 HOMOLOGS ...... 16

1.7.3 HUMAN DHCR24 ...... 17

1.8 DHCR24 REGULATION ...... 18 1.8.1 TRANSCRIPTIONAL REGULATION OF DHCR24 ...... 18

1.8.2 POST-TRANSLATIONAL REGULATION OF DHCR24 ...... 21

1.9 AIMS AND HYPOTHESES ...... 22

2 MATERIALS AND METHODS ...... 25

2.1 GENERAL MATERIALS ...... 25

2.1.1 EQUIPMENT ...... 30

2.1.2 SOFTWARE ...... 30

2.1.3 BUFFER RECIPES ...... 31

2.1.4 POLYACRYLAMIDE GEL RECIPES ...... 33

2.2 CELL CULTURE ...... 33 2.2.1 CELL-LINES ...... 34 ix

2.2.2 GENERATION OF THE CHO OVEREXPRESSING CELL-LINE ...... 35

2.3 PLASMIDS ...... 36

2.3.1 CLONING ...... 39

2.3.2 TRANSFORMATION USING XL-10 GOLD CELLS ...... 40

2.3.3 COLONY PCR ...... 41

2.3.4 MINIPREP ...... 41

2.3.5 SEQUENCING ...... 41

2.3.6 MIDIPREP ...... 42

2.3.7 PLASMID TRANSFECTION ...... 42

2.4 STEROL SYNTHESIS ASSAY ...... 42

2.4.1 LIPID EXTRACTION ...... 42

2.4.2 ARGENTATION THIN LAYER CHROMATOGRAPHY ...... 43

2.5 WESTERN BLOTTING ...... 43 2.5.1 PREPARATION OF CELL LYSATE FOR SDS-PAGE ...... 43 2.5.2 SDS-PAGE ...... 44

2.5.3 WESTERN BLOTTING ...... 44

2.6 QUANTITATIVE REAL-TIME REVERSE TRANSCRIPTION PCR (QRT-PCR)...... 45

2.7 DATA PRESENTATION ...... 47

3 THE MEMBRANE TOPOLOGY OF DHCR24 ...... 51

3.1.1 INTRODUCTION ...... 51

3.2 MATERIALS AND METHODS ...... 54

3.2.1 MATERIALS ...... 54

3.2.2 CELL FRACTIONATION...... 54

3.2.3 MEMBRANE ISOLATION ...... 54

3.2.4 DIFFERENTIAL SOLUBILISATION ...... 55

3.2.5 PROTEASE PROTECTION ASSAY ...... 55

3.2.6 BIOINFORMATICS TOOLS ...... 55

3.3 RESULTS ...... 56 3.3.1 DHCR24 HAS AN EXTREMELY HYDROPHOBIC N-TERMINUS, SUGGESTING A

CANDIDATE REGION FOR TRANSMEMBRANE DOMAINS ...... 56

3.3.2 DHCR24 IS NOT PREDICTED TO ASSOCIATE WITH THE MEMBRANE VIA

POST-TRANSLATIONAL MODIFICATIONS ...... 57

3.3.3 PREDICTED MEMBRANE TOPOLOGY OF DHCR24 CONSISTS OF N-TERMINAL TMDS 58

3.3.4 A SECRETORY SIGNAL PEPTIDE IS PREDICTED FOR DHCR24 ...... 60

3.3.5 PUTATIVE TMDS OF LOW PROBABILITY ARE ELIMINATED DUE TO LOCATION ...... 61

3.3.6 MEMBRANE ORIENTATION OF THE N- AND C- TERMINI OF DHCR24 ...... 61

3.3.7 THE PUTATIVE TMD IS NOT ESSENTIAL FOR DHCR24 MEMBRANE ASSOCIATION 64

3.3.8 DHCR24 ASSOCIATES STRONGLY WITH MEMBRANES ...... 65

3.3.9 OTHER CANDIDATE TMDS ARE NOT ESSENTIAL FOR DHCR24 MEMBRANE

ASSOCIATION ...... 67

3.4 DISCUSSION...... 69

4 STEROL-MEDIATED TRANSCRIPTIONAL REGULATION OF DHCR24 77

4.1 INTRODUCTION ...... 77

4.2 MATERIALS AND METHODS ...... 79 4.2.1 MATERIALS ...... 79

4.2.2 CELL CULTURE MEDIA ...... 79

4.2.3 PLASMIDS ...... 79

4.2.4 LUCIFERASE ASSAY ...... 81

4.2.5 ELECTROPHORETIC MOBILITY SHIFT ASSAY ...... 82

4.2.6 QUANTIFYING COOPERATIVITY BY CALCULATING THE HILL SLOPE ...... 83

4.3 RESULTS ...... 84 4.3.1 DHCR24 EXPRESSION IS REGULATED BY STEROLS IN HUMAN CELL-LINES ...... 84

4.3.2 DHCR24 REGULATION IS DEPENDENT ON SREBP-2 IN CHO CELLS ...... 86

4.3.3 TRANSCRIPTIONAL REGULATION OF DHCR24 PROMOTER-REPORTER GENE BY

CO-EXPRESSED SREBP-2 ...... 88

4.3.4 IDENTIFICATION OF PUTATIVE REGULATORY ELEMENTS RESPONSIBLE FOR

SREBP-2-MEDIATED TRANSCRIPTION OF DHCR24 ...... 89

4.3.5 IDENTIFICATION OF TWO SRES IN DHCR24 ...... 94

4.3.6 SREBP-2 BINDS TO DHCR24 SRES ...... 96

4.3.7 NF-Y BINDING SITES IN THE DHCR24 PROMOTER ALSO PLAY A ROLE IN ITS

REGULATION ...... 99

4.3.8 THE SPATIAL ARRANGEMENT OF THE TWO SRES IN THE DHCR24 PROMOTER IS

IMPORTANT ...... 99

4.3.9 DUAL SRES IN THE DHCR24 PROMOTER RESULT IN SREBP-2 HOMOTYPIC

COOPERATIVITY ...... 102

4.4 DISCUSSION...... 103

5 STEROL-MEDIATED POST-TRANSLATIONAL REGULATION OF DHCR24 BY INHIBITION OF ACTIVITY ...... 109

5.1 INTRODUCTION ...... 109

5.2 MATERIALS AND METHODS ...... 112 5.2.1 24,25EC SYNTHESIS ASSAY ...... 112

5.3 RESULTS ...... 113 5.3.1 24,25EC REDUCES CHOLESTEROL AND INCREASES DESMOSTEROL LEVELS SIMILAR

TO DHCR24 INHIBITORS ...... 113

5.3.2 COMPACTIN PRE-TREATMENT INCREASES SENSITIVITY OF ARG-TLC ASSAY ..... 114

5.3.3 THE CHOLESTEROL TO DESMOSTEROL RATIO IS THE BEST INDICATOR FOR DHCR24

ACTIVITY ...... 116

5.3.4 24,25EC REDUCES THE CHOLESTEROL TO DESMOSTEROL RATIO WITHOUT

AFFECTING DHCR24 PROTEIN LEVELS ...... 117

5.3.5 EFFECT OF VARIOUS OXYSTEROLS ON THE CHOLESTEROL TO DESMOSTEROL RATIO 119

5.3.6 INCREASED SYNTHESIS OF ENDOGENOUS 24,25EC INHIBITS THE CHOLESTEROL TO

DESMOSTEROL RATIO ...... 121

5.3.7 MANIPULATION OF ENDOGENOUS 24,25EC MODULATES THE CHOLESTEROL TO

DESMOSTEROL RATIO ...... 123

5.3.8 ENDOGENOUS 24,25EC APPEARS TO REGULATE DHCR24 ACTIVITY ...... 125

5.3.9 EFFECT OF DHCR24 OVEREXPRESSION ON 24,25EC’S ABILITY TO REDUCE THE

CHOLESTEROL TO DESMOSTEROL RATIO ...... 127

5.3.10 PROGESTERONE INHIBITS DHCR24 ACTIVITY ...... 128

5.4 DISCUSSION...... 130

6 REGULATION OF DHCR24 BY POST-TRANSLATIONAL MODIFICATIONS ...... 135

6.1 INTRODUCTION ...... 135

6.2 MATERIALS AND METHODS ...... 137

xii

6.2.1 MATERIALS ...... 137

6.2.2 SIRNA TRANSFECTION ...... 137

6.2.3 GAS CHROMATOGRAPHY MASS SPECTROMETRY ...... 137

6.3 RESULTS ...... 139

6.3.1 PROTEIN KINASE C INHIBITION DECREASES DHCR24 ACTIVITY ...... 139

6.3.2 PKC INHIBITION RESULTS IN DESMOSTEROL ACCUMULATION ...... 141

6.3.3 DHCR24-T110 PHOSPHOMUTANT DOES NOT ALTER EXPRESSION ...... 144

6.3.4 DHCR24-T110 PHOSPHOMUTANT DECREASES DHCR24 ACTIVITY ...... 146

6.3.5 PKC INHIBITION DECREASES DHCR24 ACTIVITY INDEPENDENTLY OF T110 ..... 148

6.3.6 THE BIM INHIBITORY EFFECT IS NOT MEDIATED BY PKC EPSILON ...... 150

6.4 DISCUSSION...... 151

7 GENERAL DISCUSSION ...... 155

7.1 MEMBRANE TOPOLOGY OF DHCR24: AN UNORTHODOX PROTEIN ...... 157 7.1.1 LACK OF COMMON RECOGNITION SEQUENCES ...... 157

7.1.2 OTHER CHOLESTEROL SYNTHETIC ENZYMES ...... 157

7.1.3 OTHER FUNCTIONS OF DHCR24 ...... 158

7.1.4 FUTURE EXPERIMENTS ...... 158

7.2 TRANSCRIPTIONAL REGULATION OF DHCR24 BY STEROLS ...... 159 7.2.1 LXR VS SREBP-2 ...... 159

7.2.2 DIRECT BINDING OF SREBP-2 TO DHCR24 ...... 160

7.2.3 RELEVANCE OF DHCR24 BEING SREBP-2 REGULATED ...... 161

7.2.4 RELEVANCE TO FUTURE WORK...... 161

7.2.5 OTHER MODES OF TRANSCRIPTIONAL REGULATION OF DHCR24? ...... 161

7.2.6 IMPLICATIONS FOR CARDIOVASCULAR DISEASE ...... 162

7.3 DHCR24 ACTIVITY ...... 163

7.3.1 DHCR24 MAY BE A CONTROL POINT IN THE MEVALONATE PATHWAY ...... 163

7.3.2 PREDICTED REGULATORY MECHANISMS FOR DHCR24 ...... 164

7.3.3 DEVELOPMENT OF AN ACTIVITY ASSAY ...... 164

7.4 POST-TRANSLATIONAL MODIFICATIONS IN DHCR24 REGULATION ...... 166

7.4.1 KNOWN PHOSPHORYLATION SITES IN DHCR24 ...... 166

7.4.2 THE INVOLVEMENT OF PKC ...... 167

7.4.3 UBIQUITINATION AND THE POSSIBILITY OF REGULATED TURNOVER ...... 168

xiii

7.5 LIMITATIONS OF IN SILICO PREDICTION PROGRAMS ...... 169

7.6 FUTURE DIRECTIONS ...... 170

7.6.1 FURTHER DHCR24 CHARACTERISATION ...... 170

7.6.2 KEY PLAYERS IN THE POST-LANOSTEROL PATHWAY ...... 171

7.6.3 EPIGENETIC REGULATION OF DHCR24 ...... 173

7.7 CONCLUDING REMARKS ...... 173

8 APPENDICES ...... 177

9 REFERENCES ...... 187

xiv

LIST OF FIGURES

FIGURE 1.1. THE CHEMICAL STRUCTURE OF CHOLESTEROL WITH THE CARBONS

NUMBERED...... 4

FIGURE 1.2. OVERVIEW OF CHOLESTEROL HOMEOSTASIS...... 5

FIGURE 1.3. CHOLESTEROL IS THE END PRODUCT OF THE MEVALONATE PATHWAY. ... 7

FIGURE 1.4. STEROL-MEDIATED TRANSCRIPTIONAL REGULATION OF CHOLESTEROL

SYNTHESIS BY SREBP-2 PROCESSING...... 9

FIGURE 1.5. POST-TRANSLATIONAL REGULATION OF CHOLESTEROL SYNTHESIS BY

END PRODUCT INHIBITION...... 11

FIGURE 1.6. CHOLESTEROL SYNTHESIS VIA ALTERNATE PATHWAYS...... 14

FIGURE 1.7. PROMOTER MAP OF HUMAN DHCR24...... 20

FIGURE 3.1. PREDICTED AND KNOWN FUNCTIONAL FEATURES OF DHCR24 PROTEIN...... 52

FIGURE 3.2. DHCR24 MEMBRANE TOPOLOGY MODELS...... 53

FIGURE 3.3. IDENTIFICATION OF CANDIDATE TMDS...... 57

FIGURE 3.4. DHCR24 MEMBRANE TOPOLOGY PREDICTIONS...... 59

FIGURE 3.5. DHCR24 CONTAINS A PUTATIVE SIGNAL SEQUENCE...... 61

FIGURE 3.6. MEMBRANE ORIENTATION OF THE N- AND C-TERMINI OF DHCR24 AS

DETERMINED BY TRYPSIN PROTEOLYSIS...... 63

FIGURE 3.7. THE N-TERMINUS OF DHCR24 IS STRONGLY MEMBRANE ASSOCIATED,

FOLLOWED BY A LUMINAL LOOP...... 64

FIGURE 3.8. MEMBRANE ASSOCIATION OF DHCR24 AND Δ56 DHCR24...... 65

FIGURE 3.9. DHCR24 ASSOCIATES STRONGLY WITH THE MEMBRANE...... 66

FIGURE 3.10. MEMBRANE ASSOCIATION OF DHCR24 N-TERMINAL TRUNCATIONS. .. 68

FIGURE 3.11. COMPARISON OF CURRENT DHCR24 MEMBRANE TOPOLOGY MODELS...... 72

FIGURE 3.12. HYPOTHETICAL MEMBRANE TOPOLOGY MODEL OF DHCR24 ...... 73

FIGURE 4.1. DHCR24 EXPRESSION UPON ALTERED STEROL STATUS IN HUMAN CELLS...... 86

FIGURE 4.2. REGULATION OF DHCR24 EXPRESSION WITH ALTERED SREBP-2 LEVELS

IN CHO CELLS...... 87

FIGURE 4.3. CO-EXPRESSION OF SREBP-2 UP-REGULATES A DHCR24 PROMOTER

REPORTER...... 89

FIGURE 4.4. STEROL RESPONSE ELEMENT MOTIF AND SEQUENCE OF THE PROMOTER

OF HUMAN DHCR24...... 91

FIGURE 4.5. SREBP-2 BINDS TO BOTH -196INV SRE AND -160 SRE IN THE PROMOTER

OF THE HUMAN DHCR24 GENE...... 95

FIGURE 4.6. EMSA OPTIMISATION...... 96

FIGURE 4.7. EMSA OPTIMISATION WITH SALMON SPERM DNA...... 97

FIGURE 4.8. SREBP-2 BINDS TO BOTH -196 INVSRE AND -160 SRE IN THE PROMOTER

OF THE HUMAN DHCR24 GENE...... 98

FIGURE 4.9. -196 INVSRE AND -160 SRE IN THE PROMOTER OF THE HUMAN DHCR24

GENE ARE EVOLUTIONARILY CONSERVED...... 99

FIGURE 4.10. NF-Y SITES AND SPATIAL ARRANGEMENTS AFFECT STEROL-

RESPONSIVENESS OF THE DHCR24 PROMOTER...... 101

FIGURE 4.11. SREBP-2 ACTS COOPERATIVELY TO ACTIVATE GENES WITH DUAL

SRES...... 102

FIGURE 5.1. 24(S),25-EPOXYCHOLESTEROL IS PRODUCED IN A SHUNT OF THE

CHOLESTEROL SYNTHETIC PATHWAY...... 110

FIGURE 5.2. 24,25EC TREATMENT RESULTS IN ACCUMULATION OF DESMOSTEROL AT

THE EXPENSE OF CHOLESTEROL...... 114

FIGURE 5.3. DESMOSTEROL AND CHOLESTEROL ARE THE MAIN STEROLS LABELLED BY 14 [ C]-ACETATE, AND ARE PREDOMINANTLY PRESENT IN CELLS RATHER THAN

MEDIA...... 115

FIGURE 5.4. THE CHOLESTEROL TO DESMOSTEROL RATIO IS A REASONABLE

SURROGATE FOR DHCR24 ACTIVITY...... 117

FIGURE 5.5. 24,25EC TREATMENT REDUCES THE CHOLESTEROL TO DESMOSTEROL

RATIO RAPIDLY AND SENSITIVELY WITHOUT AFFECTING DHCR24 PROTEIN

LEVELS...... 118

FIGURE 5.6. CERTAIN SIDE-CHAIN OXYSTEROLS CAN REDUCE THE CHOLESTEROL TO

DESMOSTEROL RATIO...... 121

FIGURE 5.7. MODULATION OF ENDOGENOUS 24,25EC SYNTHESIS BY

PHARMACOLOGICAL MANIPULATION DECREASES THE CHOLESTEROL TO

DESMOSTEROL RATIO...... 122

xvi

FIGURE 5.8. MODULATION OF ENDOGENOUS 24,25EC BY GENETIC MANIPULATION

ALTERS THE CHOLESTEROL TO DESMOSTEROL RATIO...... 124

FIGURE 5.9. ENDOGENOUS 24,25EC APPEARS TO REGULATE DHCR24 ACTIVITY. .. 127

FIGURE 5.10. DHCR24 OVEREXPRESSION BLUNTS THE INHIBITORY EFFECT OF

24,25EC ON THE CHOLESTEROL TO DESMOSTEROL RATIO...... 128

FIGURE 5.11. DHCR24 OVEREXPRESSION BLUNTS THE INHIBITORY EFFECT OF

PROGESTERONE ON DHCR24 ACTIVITY...... 129

FIGURE 6.1. EFFECT OF KINASE INHIBITORS ON DHCR24 ACTIVITY...... 140

FIGURE 6.2. BISINDOLYLMALEIMIDE I TREATMENT RESULTS IN ACCUMULATION OF

DESMOSTEROL...... 142

FIGURE 6.3. BIM INHIBITS DHCR24 ACTIVITY...... 143

FIGURE 6.4. CHARACTERISATION OF THE CHO-DHCR24-T110A/E STABLE CELL-

LINES...... 146

FIGURE 6.5. HAMSTER SPECIFIC KNOCKDOWN OF DHCR24 MRNA...... 147

FIGURE 6.6. EFFECT OF T110 MUTANTS ON DHCR24 ACTIVITY...... 148

FIGURE 6.7. BIM DOES NOT AFFECT DHCR24-T110A ACTIVITY...... 149

FIGURE 6.8. PKC EPSILON DOES NOT AFFECT DHCR24 ACTIVITY...... 150

FIGURE 7.1. MEMBRANE TOPOLOGY AND REGULATORY MECHANISMS OF DHCR24...... 156

FIGURE 7.2. KEY ENZYMES IN THE BLOCH AND KANDUTSCH-RUSSELL PATHWAYS. 172

APPENDIX 8.1. THE DHCR24 PROTEIN AND ITS HOMOLOGS ARE UBIQUITOUS ACROSS

THE DOMAINS OF LIFE...... 177

APPENDIX 8.2. ALIGNMENT OF DHCR24 WITH HOMOLOGS...... 178

APPENDIX 8.3. THE MAJOR ENZYMATIC REACTIONS IN THE POST-LANOSTEROL

CHOLESTEROL SYNTHESIS PATHWAY...... 179

APPENDIX 8.4. THE MAJOR STEROL INTERMEDIATES FORMED IN THE POST-

LANOSTEROL CHOLESTEROL SYNTHESIS PATHWAY...... 180

APPENDIX 8.5. DHCR24 HOMOLOGS AND THEIR ACCESSION NUMBERS...... 181

APPENDIX 8.6. A COMPARISON OF KNOWN AND PUTATIVE DHCR24 INHIBITORS. ... 182

APPENDIX 8.7. PROGESTERONE, AND NOT ANDROSTENEDIONE, INHIBIT DHCR24. . 183

xvii

xviii

LIST OF TABLES

TABLE 2.1. CELL CULTURE MEDIA...... 34

TABLE 2.2. CELL CULTURE CONDITIONS...... 35

TABLE 2.3. EXPRESSION PLASMIDS...... 37

TABLE 2.4. PROMOTER PLASMIDS IN FIREFLY AND RENILLA LUCIFERASE VECTORS.. 38

TABLE 2.5. ANTIBODIES USED FOR WESTERN BLOTTING IN THIS THESIS...... 45

TABLE 2.6. PRIMERS FOR QRT-PCR...... 46

TABLE 4.1. LUCIFERASE PLASMIDS USED IN THIS CHAPTER...... 80

TABLE 4.2. SREBP-2 EXPRESSION PLASMIDS USED IN THIS CHAPTER...... 81

TABLE 4.3. EMSA PROBES...... 83

TABLE 4.4. CHARACTERISATION OF KNOWN SRES...... 92

TABLE 4.5. PUTATIVE DHCR24 SRES...... 93

TABLE 6.1. ION M/Z VALUES MONITORED DURING SELECTIVE ION MONITORING

ANALYSIS OF THE TRIMETHYLSILIYL DERIVATIVES...... 138

TABLE 7.1. KNOWN PHOSPHORYLATION SITES IN DHCR24...... 167

xix

ABBREVIATIONS

4HC 4β-hydroxycholesterol 5α,6αEC 5α,6α-epoxycholesterol 5β,6βEC 5β,6β- epoxycholesterol 7αHC 7α-hydroxycholesterol 7βHC 7β-hydroxycholesterol 7KC 7-ketocholesterol 19HC 19-hydroxycholesterol 24,25EC 24(S),25-epoxycholesterol 20HC 20α-hydroxycholesterol 22HC 22(R)-hydroxycholesterol 24HC 24(S)-hydroxycholesterol 25HC 25-hydroxycholesterol 27HC 27-hydroxycholesterol aa Amino acid ABCA1 ATP-binding cassette transporter A1 ABCG1 ATP-binding cassette transporter G1 ACC Acetyl-CoA carboxylase ACSL1 Acyl-CoA synthase 1 ARE Androgen response element Arg-TLC Argentation thin layer chromatography BCA Bicinchoninic acid BIM Bisindolylmaleimide BSA Bovine serum albumin CAR Constitutive androstane receptor CD Methyl-β-cyclodextrin cDNA Complementary DNA CHO Chinese hamster ovary CMV Cytomegalovirus CPN Compactin DWF1 Dwarf 1 DHCR24 3β-hydroxysterol Δ24-reductase

xxi

DHCR7 7-dehydrocholesterol reductase DMEM Dulbecco’s Modified Eagle’s Medium DMSO Dimethyl sulfoxide DOS 2,3(S):22(S),23-dioxidosqualene DTT Dithiothreitol ECL Enhanced chemiluminescence EDTA Ethylenediaminetetraacetic Acid EGTA Ethylene glycol-bis(2-aminoethylether)-N,N,N′,N′- tetraacetic acid EMSA Electrophoretic mobility shift assay ER Endoplasmic reticulum ERE Estrogen response element F-12 Ham’s Nutrient Mixture F-12 FAD Flavin adenine dinucleotide FASN Fatty acid synthase FCLPDS Lipoprotein-deficient fetal calf serum FCS Fetal calf serum FDFT1 Farnesyl-diphosphate farnesyltransferase 1 FDPS Farnesyl diphosphate synthase FRT Flp recombinase target GC Gas chromatography GRAVY Grand average of hydropathicity HCV Hepatitis C virus HEK Human embryonic kidney HDL High density lipoprotein CR 3-hydroxy-3-methylglutaryl-coenzyme A reductase HMGCS 3-hydroxy-3-methylglutaryl-coenzyme A synthase HMM Hidden Markov model HSD17B7 Hydroxysteroid (17β) dehydrogenase 7 Insig Insulin-induced gene LDL Low density lipoprotein LDLR Low density lipoprotein receptor LPDS Lipoprotein-deficient serum

xxii

LXR Liver X receptor Mdm2 Mouse double minute 2 homolog MOS 2,3(S)-monooxidosqualene MS Mass spectrometry MSA Multiple sequence alignment Mut Mutated NADPH Nicotinamide adenine dinucleotide phosphate NBS Newborn bovine serum NF-Y Nuclear factor Y nPAGE Native polyacrylamide gel electrophoresis nSREBP-2 Mature N-terminal form of the sterol regulatory element binding protein OSC 2,3-oxidosqualene cyclase Oxysterol Oxygenated sterol PAGE Polyacrylamide gel electrophoresis PBGD Porphobilinogen deaminase PBS Phosphate buffered saline PCR Polymerase chain reaction PKC Protein kinase C PXR Pregnane X receptor qRT-PCR Quantitative real-time PCR RPMI Roswell Park Memorial Institute 1640 medium Seladin-1 Selective Alzheimer's disease indicator-1 S1P Site 1 protease S2P Site 2 protease Scap SREBP-cleavage activating protein SD Standard deviation SDS Sodium dodecyl sulphate SEM Standard Error of the Mean SM Squalene monooxygenase Sp1 Specificity protein 1 SQLE Squalene epoxidase SRE Sterol regulatory element

xxiii

SREBP Sterol regulatory element binding protein TH Thyroid hormone TLC Thin layer chromatography TM7SF2 Transmembrane 7 superfamily member 2 TM Transmembrane TMD Transmembrane domain UNSW University of New South Wales WT Wild-type

xxiv

Chapter 1

General Introduction

1 GENERAL INTRODUCTION

1.1 Introduction

This thesis investigates the ultimate step in cholesterol synthesis via the Bloch pathway: the reduction of desmosterol by 3β-hydroxysterol Δ24-reductase (DHCR24). In particular, the regulation of this enzymatic step is explored, and DHCR24’s topology within the endoplasmic reticulum (ER) membrane is characterised.

1.2 Cholesterol

Cholesterol is derived from the Greek words chole (bile) and stereos (solid), after its discovery in bile and gallstones, first by François Poulletier de la Salle, and later by Michel Eugène Chevreul [1]. The chemical structure was elucidated by Wieland and

Windaus, for which they were awarded a Nobel Prize in 1928: a C27 sterol, composed of a tetracyclic ring containing a hydroxyl (OH) group positioned at C-3, and a saturated, aliphatic side-chain (Figure 1.1). Cholesterol is the major mammalian sterol; the building block of many essential molecules such as bile acids, vitamin D, and steroid hormones. Cholesterol has a vital role in cellular membranes, being critical for membrane fluidity and permeability, as well as the organisation of lipid rafts, for signalling [2]. Disruption to cholesterol levels is related to developmental disorders [3], Alzheimer’s disease [4], cardiovascular disease [5], hepatitis C viral (HCV) infection [6], and in some instances, cancer [7]. Therefore, the regulation of cholesterol is critical as levels must be maintained at a stable equilibrium to achieve homeostasis [8].

Figure 1.1. The chemical structure of cholesterol with the carbons numbered.

Cholesterol, or (3β)-cholest-5-en-3-ol, is a C27 sterol, composed of a tetracyclic ring containing a hydroxyl (OH) group positioned at C-3 and a saturated, aliphatic side-chain.

1.3 Cholesterol homeostasis

Cellular cholesterol homeostasis is modulated at three levels: uptake, synthesis, and efflux. Cholesterol is taken up by the cell in the form of lipoproteins, with low-density lipoprotein (LDL) being the major lipoprotein. LDL is internalised by the LDL receptor (LDLR). Alternatively, cholesterol can be synthesised de novo using acetyl CoA as the precursor, and involving over 25 enzymatic steps, collectively known as the mevalonate pathway. Lastly, cholesterol can be eliminated from the cell via ATP-binding cassette transporters A1 and G1 (ABCA1, ABCG1) through high-density lipoprotein (HDL) particles. Each of these steps is controlled by two master transcriptional factors: the sterol regulatory element-binding proteins (SREBPs) and liver X receptor (LXR) (Figure 1.2). The cholesterol specific isoform, SREBP-2, promotes cholesterol accumulation, by stimulating cholesterol uptake and synthesis; whereas LXR promotes cholesterol reduction, by increased cholesterol efflux and inhibition of cholesterol uptake (Figure 1.2). Both transcription factors are subject to feedback regulation. This thesis focuses primarily on the homeostatic mechanisms for cholesterol synthesis.

Figure 1.2. Overview of cholesterol homeostasis. Cellular cholesterol is maintained by regulation of uptake, synthesis, and efflux. Two master transcription factors, SREBP-2 and LXR, regulate each of these processes. Adapted from [9].

1.4 Cholesterol synthesis

Cholesterol is produced at the end of the mevalonate pathway, with multiple enzymatic steps required to transform acetyl CoA (C2) into cholesterol (C27) (Figure 1.3), an energy demanding process requiring the reducing power of nicotinamide adenine dinucleotide phosphate (NADPH) at multiple steps [10]. 3-hydroxy-3-methylglutaryl-coenzyme A reductase (HMGCR), an early step in the pathway, is the pharmacological target of the highly successful statins, currently the most potent cholesterol-lowering drug family (Figure 1.3) [11]. HMGCR controls flux through the pathway, with mevalonate, its product, being the precursor for not only cholesterol, but non-sterol products including ubiquinone and heme A (cellular respiration), dolichols (glycosylation), isopentyl adenine (present in some transfer RNAs), and isoprenoids (farnesyl and geranyl groups for prenylation/membrane attachment) (Figure 1.3) [12]. Squalene monooxygenase (SM) is another enzyme that our laboratory has found to be important [13], and will be discussed in Section 1.5.2. SM catalyses the first oxygenation step in the pathway, converting squalene into

2,3(S)-monooxidosqualene (MOS), which is the precursor for lanosterol, the first sterol produced in the pathway. A further 19 enzyme catalysed reactions occur post-lanosterol to reach cholesterol, with the final step catalysed by DHCR24 or 7-dehydrocholesterol reductase (DHCR7) (detailed in Section 1.5.2.1). Steroid hormones, bile acids and most oxysterols are derived from cholesterol. One notable exception is 24(S),25-epoxycholesterol (24,25EC), which is produced in a shunt of the mevalonate pathway (Figure 1.3). Numerous sterol intermediates and by-products are also formed, which have important roles in cholesterol homeostasis [14].

Figure 1.3. Cholesterol is the end product of the mevalonate pathway. Schematic of the mevalonate pathway, showing key enzymes and products. Acetyl-CoA is the precursor for cholesterol synthesis. HMGCR is the major flux-controlling enzyme in the pathway, and the target of the statin drugs. The mevalonate pathway also generates non-sterol products. SM catalyses the first oxygenation step in the pathway, converting squalene into MOS, which is the precursor for lanosterol, the first sterol produced in the pathway. The end product, cholesterol, is the precursor for steroid hormones, bile acids, and oxysterols, with the notable exception of 24,25EC, which is produced in a shunt of the pathway. Adapted from our recent publication [15].

1.5 Regulation of cholesterol synthesis

Cholesterol synthesis is a highly conserved and tightly regulated metabolic pathway that maintains homeostasis through multivalent feedback regulation. Long- and short-term regulation occur at the transcriptional and post-translational level, respectively. Here, we will discuss some of these mechanisms

1.5.1 Sterol-mediated transcriptional regulation

Cholesterol homeostasis is maintained with exquisite control by SREBPs and LXR, which work in a concerted fashion to increase and decrease cellular sterol status, respectively. Both undergo extensive negative feedback regulation. Most of the enzymes involved in cholesterol synthesis are thought to be under the control of the SREBP transcription factors [16, 17], yet a large proportion of these have not been characterised as such [17]. Three isoforms of SREBP exist in mammals: -1a, −1c and −2 [8], where each isoform activates a specific set of genes. SREBP-2 activates genes for cholesterol synthesis (e.g. HMGCR; [18]) and uptake (e.g. LDLR; [19]); SREBP-1c activates genes for fatty acid synthesis (e.g. acetyl-CoA carboxylase, ACC; [20]); and SREBP-1a regulates genes for both. SREBPs also activate genes required to generate NADPH, which is consumed at multiple stages of these lipid synthetic pathways [21]. SREBPs are synthesised as inactive precursors bound to the ER membrane (Figure 1.4). Low cholesterol levels are detected by insulin-induced gene (Insig) and SREBP-cleavage activating protein (Scap); by direct sensing of cholesterol, sterol

7 intermediates and oxysterol by-products of the pathway [14, 22]. Scap then escorts SREBP to the Golgi for proteolytic cleavage of the N-terminus of SREBP [23]. This active transcription factor then enters into the nucleus to bind to sterol response elements (SREs) within the promoters of target genes, which induces the recruitment of cofactors specificity protein 1 (Sp1) and nuclear factor Y (NF-Y), to synergistically activate target gene expression [24]. For SREBP-2, this results in an increase in cellular cholesterol from uptake and synthesis [12]. Increased flux though the pathway creates sterol and oxygenated sterol (oxysterol) intermediates, which also feedback to signal cholesterol overload. When cholesterol, or oxysterols, accumulate within the cell, Insig binds to Scap, which prevents SREBP processing, and therefore target gene expression [25]. Oxysterols can also act as ligands for the nuclear receptor, liver X receptor (LXR), inducing the expression of genes involved in sterol efflux (e.g. ABCA1; [26]) and degradation of LDLR (e.g. Idol [27]), thereby reducing sterol levels and maintaining homeostasis.

Figure 1.4. Sterol-mediated transcriptional regulation of cholesterol synthesis by SREBP-2 processing. When cellular cholesterol levels are low, SREBP-2 is escorted by Scap to the Golgi membrane, where it is cleaved sequentially by Site-1 Protease and Site-2 Protease. The active N-terminal transcription factor migrates to the nucleus, and binds to SREs of genes involved in cholesterol synthesis and uptake, inducing the recruitment of cofactors Sp1 and NF-Y, which bind directly to SREBP-2, and synergistically activate target gene expression. As cholesterol accumulates within the cell, the Insig retention protein binds to Scap, retaining the SREBP-Scap complex in the ER membrane, preventing SREBP-2 processing, and therefore down-regulating target genes, such as HMGCR and LDLR. Adapted from [28].

1.5.2 Post-translational regulation at rate-limiting enzymes

Feedback regulation of cholesterol synthesis also occurs at the post-translational level by inhibition of key rate-limiting steps of the mevalonate pathway. The conversion of HMG CoA to mevalonate by HMGCR is the first committed step in cholesterol synthesis. This is a key point for flux control of the pathway, as it controls mevalonate production, the precursor for cholesterol, as well as non-sterol bioactive molecules (Figure 1.3). HMGCR has, until recently, been considered the only control point. HMGCR undergoes feedback regulation by multiple mechanisms (Figure 1.5). HMGCR undergoes Insig-dependent degradation, stimulated by various products of the mevalonate pathway. Oxysterols and the cholesterol synthetic intermediates lanosterol and 24,25-dihydroxylanosterol, trigger Insig to bind directly to HMGCR, and promote its ubiquitination and proteasomal degradation (Figure 1.5) [8, 29, 30]. Insig-dependent HMGCR degradation is also stimulated by the isoprenoid, geranylgeraniol [31]. HMGCR is also affected by metabolic state via post-translational modifications. AMP-activated protein kinase (AMPK) controls energy expenditure during metabolic stress, and inhibits human HMGCR activity by phosphorylation at S872, a conserved serine in the HMGCR active site [32, 33]. Dephosphorylation by protein phosphatase 2A (PP2A) restores enzyme activity [34]. These sterol and non-sterol mechanisms allow for rapid inhibition of cholesterol synthesis in times of cholesterol overload or metabolic stress [35]. Our laboratory has recently discovered a second flux control point further down the pathway at squalene monooxygenase (SM), which catalyses the first oxygenation step in cholesterol synthesis, converting squalene into MOS (Figure 1.5). Cholesterol accumulation signals degradation of SM via an Insig-independent mechanism, which, unlike HMGCR, is signalled by two sterols: cholesterol and its immediate precursor, desmosterol [13]. This recent discovery suggests the potential for multiple flux- controlling points in the cholesterol synthetic pathway, which may occur after lanosterol production, at any of the 19 enzyme catalysed reactions that occur before cholesterol synthesis (Figure 1.5).

Figure 1.5. Post-translational regulation of cholesterol synthesis by end product inhibition. Flux-controlling steps catalysed by HMGCR and SM are subject to feedback inhibition by cholesterol itself (solid line) or secondary signals, including cholesterol derivatives and pathway intermediates (dashed line). Potential flux controlling steps may occur after lanosterol production, at any of the 19 enzyme catalysed reactions that occur before cholesterol synthesis.

1.5.2.1 DHCR24 as a potential control point in cholesterol synthesis

Synthesis of the first sterol, lanosterol, occurs after both HMGCR and SM catalysed reactions, with a further 19 enzyme reactions required before cholesterol synthesis - yet these steps have been little characterised. One of these enzymes, 3β-hydroxysterol Δ24-reductase (DHCR24) catalyses the ultimate step in cholesterol synthesis and is a likely target for feedback regulation, as it would provide the most specific means to switch cholesterol synthesis on or off, making this step a potential control point in the cholesterol synthetic pathway. DHCR24 catalyses an important step in cholesterol synthesis: the Δ24 reduction of the side-chain (Figure 1.6 B). After the production of lanosterol, the mevalonate 11 pathway bifurcates into the Bloch and Kandutsch-Russell pathways (Figure 1.6 A). These include the same enzymes, but the timing of their action is different. In the Bloch pathway, desmosterol is converted into cholesterol by DHCR24 in the final step. In the Kandutsch-Russell pathway, lanosterol is converted to 24,25-dihydroxylanosterol by DHCR24 in the first step. Thus, DHCR24 controls the levels of key feedback regulators in cholesterol homeostasis: desmosterol, cholesterol, and 24,25-dihydroxylanosterol. The conversion of desmosterol into cholesterol is described as the final step in the mevalonate pathway, but the DHCR24 enzyme can act upon any intermediate containing the C-24,25 double bond. One in vitro study suggests that other sterol intermediates are actually preferred substrates over desmosterol [36]. However, this is not substantiated from cell based [37] and in vivo studies, since when DHCR24 is inhibited [38, 39] or non-functional [40-43], desmosterol accumulates. Moreover, molecular docking simulations support the concept that desmosterol is the preferred substrate compared to earlier intermediates [44]. Therefore, this thesis will refer to the DHCR24 catalysed reaction via the Bloch pathway, with desmosterol and cholesterol being the precursor and product respectively (Figure 1.6 B).

A Acetyl CoA I "" I I I I I I I I (Bloch) ( Kandutsch-Russell ) Lanosterol• ~ ~24-reductase ' (DHCR24)

24 ,25-d ihyd rolanosterol

FF-MAS dihydro-FF-MAS t-14-reductase -~- T-MAS"" dihydro-"" T-MAS C4-demethylase -~-- zymostero"" l zymostenol"" t.Bt-7-isomerase - -- 24-dehydrolathosterol"" lathosterol"" t-5-desatu rase 7 -dehydrodes"" mosterol 7 -dehydrocholesterol"" t.7 -reductase Desmosterol

I ~24-reductase • ( Cholesterol ) ------' (DHCR24)

B NADPHv NADP+

Desmosterol Cholesterol

Figure 1.6. Cholesterol synthesis via alternate pathways. A, Post-lanosterol biosynthesis of cholesterol occurs via two alternate pathways. In the Bloch pathway, desmosterol is converted into cholesterol by DHCR24 in the final step. In the Kandutsch-Russell pathway, lanosterol is converted to 24,25-dihydrolanosterol by DHCR24 in the first step. Enzymatic reactions and enzymes involved: C14- demethylase (CYP51), Δ14-reductase (TM7SF2), C4-demethylase (SC4MOL, NSDHL, HSD17B7), Δ8Δ7-isomerase (EBP), Δ5-desaturase (SC5DL), Δ7-reductase (DHCR7), Δ24-reductase (DHCR24). Systematic names for enzymes and chemical names of intermediates are in Appendix 8.3 and 8.4. A complete post-lanosterol pathway is presented in [45]. B, Reduction of desmosterol by DHCR24, which requires the cofactor NADPH, to form cholesterol. Taken from our recent review [46].

1.6 DHCR24

DHCR24 has been implicated in a number of other cellular processes besides cholesterol synthesis. The DHCR24 catalysed reaction is crucial, with complete loss of function of DHCR24 having catastrophic effects during development, and dysregulation often resulting in disease states.

1.6.1 Functions of DHCR24

As a major constituent of cellular membranes, cholesterol is essential in providing the membrane with its mechanical and structural properties, but also is critical in the formation of membrane nanodomains. DHCR24 is crucial in the formation and stability of lipid rafts, and therefore in cell signalling [47]. Loss of DHCR24 impaired lipid raft function due to accumulated membrane desmosterol [47, 48], inhibiting downstream signal transduction pathways, such as raft-dependent insulin signalling [49-51], plasmin activation, and amyloid β precursor protein processing [47]. DHCR24 has been shown to respond to stress stimuli, with overexpression of DHCR24 protecting cells against hydrogen peroxide, and DHCR24 knockdown increasing susceptibility to hydrogen peroxide [52-55]. DHCR24 has also been reported to have antioxidant properties in a variety of tissues and disease settings, including bone formation [56] and HCV-infected hepatocytes [55], as well as the ability to directly scavenge hydrogen peroxide [57]. In response to both oxidative and oncogenic stress, DHCR24 can bind to p53, which has downstream effects on apoptosis [58]. 14

Reflecting the importance of DHCR24 in a variety of cellular functions, DHCR24 has been implicated in a number of diseases.

1.6.2 DHCR24 in disease

Initial studies identified DHCR24 as Selective Alzheimer’s Disease Indicator-1 (Seladin-1), as it was initially identified to be down-regulated in affected regions of Alzheimer’s disease brains [52, 59]. However, this name was later suggested to be erroneous [60], and more recently, our laboratory has provided evidence that DHCR24 is not down-regulated in affected regions of Alzheimer’s disease brains [61]. Despite this, the evidence that DHCR24 is neuroprotective is more robust [47, 52, 62]. In addition, single nucleotide polymorphisms (SNPs) have been identified in association studies for Alzheimer’s disease risk [63-65], with one SNP, rs600491 (T allele), significantly correlating with Alzheimer’s disease risk in men [63, 65]. Overexpression of DHCR24 is a hallmark of prostate cancer, with high levels observed in low-grade prostate cancer, which diminish as the cancer progresses to a higher grade [66]. This down-regulation is associated with an increase in cell proliferation and androgen secretion [66, 67], and has been linked to prostate cancer relapse after radical prostatectomy, thus labelling DHCR24, like prostate-specific antigen, a biomarker for prostate cancer relapse [68]. Lowered gene expression is also related to an increased incidence of metastasis [69]. HCV exploits host lipids for replication and further infection, such as for providing a raw material for the viral envelope. DHCR24 is augmented in HCV infection, in vivo and in vitro [55, 70, 71], with low expression suppressing viral replication [70]. DHCR24 expression paralleled HCV pathogenicity and subsequent hepatocarcinogenesis, through impairing apoptotic-signalling pathways by suppression of p53 activity [55]. Emerging studies implicate DHCR24 in cardiovascular disease. DHCR24 has been reported to mediate the anti-inflammatory effects of HDL [72] and have protective effects in atherosclerosis through an increase in its substrate, desmosterol [73].

1.7 DHCR24 characterisation

Although considerable research has been conducted on DHCR24, there have been few attempts to characterise the protein, including its regulation and membrane topology.

1.7.1 Discovery of DHCR24

DHCR24 was first identified in 1995 as a potential gene involved in human steroid synthesis due to its similarity to DIMINUTO/DWARF1 (DWF1) in Arabidopsis thaliana (A. thaliana). At this time, the exact function of DWF1 itself was unclear, but it was known to be essential in normal growth and development, as mutations in the gene led to reduced growth and fertility (dwarfism) in A. thaliana [74-76]. DWF1 was eventually characterised as an important protein in plant sterol (phytosterol) and steroid (brassinosteroid) synthesis, by catalysing the isomerisation of the Δ24(28) bond, and the subsequent reduction of the Δ24(25) bond in various sterol precursors [74]. The precise function of DHCR24 was not discovered until 2001, when it was identified by Waterham et al. as 3β-hydroxysterol Δ24-reductase (DHCR24), a flavin adenine dinucleotide (FAD)-dependent oxidoreductase in humans that reduces the Δ24(25) bond of desmosterol to yield cholesterol (Figure 1.6 B) [77].

1.7.2 DHCR24 homologs

DHCR24 homologs in plants catalyse a different but analogous reaction in sterol/steroid biosynthesis. Invertebrates such as Bombyx mori (B. mori, silkworm), which cannot perform de novo cholesterol synthesis, also possess a DHCR24 homolog [78]. This protein provides the organism with cholesterol by transformation of dietary phytosterols through a dealkylation reaction [78], similar to the DWF1 catalysed reaction [74]. Furthermore, this reaction is FAD and NADPH dependent. Similarly, yeast do not produce cholesterol, but synthesise an alternate sterol, ergosterol, which differs from cholesterol by a double bond at C-22 and an additional methyl group at C-24. Although yeast lack a DHCR24 homolog, they do have a C-24 methyl transferase, ERG6, which catalyses a similar reaction to DHCR24 and DWF1, and is essential in the final steps of ergosterol synthesis.

1.7.3 Human DHCR24

DHCR24 is almost ubiquitously expressed, with high levels in the adrenal gland, brain (in particular the spinal cord and medulla oblongata), liver, lung, spleen, and prostate [52]. DHCR24 is expressed in all cells/tissues that synthesise cholesterol, with the highest expression in cholesterogenic (brain and liver) and steroidogenic tissues (endocrine glands like adrenal, testes, and ovaries). It is highly conserved amongst mammals (e.g. mouse), lower organisms (e.g. Caenorhabditis elegans), and plants (e.g. A. thaliana), with 96%, 48%, and 39% amino acid identity, respectively (Appendix 8.1 and 8.2). Rare defects in the DHCR24 gene result in the rare autosomal recessive disease desmosterolosis, causing elevated levels of desmosterol in plasma and tissues, multiple congenital anomalies, and sometimes resulting in death [40-43]. Specifically, seven missense mutations have been described in desmosterolosis: N294T, K306N, Y471S, E191K, R94H, E480K, and R103C [40-43, 77].

1.7.3.1 DHCR24 cellular localisation and membrane topology

DHCR24, like its homologs in A. thaliana and B. mori [74, 78], is localised to ER membranes [52]. However, under certain conditions, DHCR24 has been observed to translocate to the nucleus [58, 67, 79-82]. Cellular stress (both oxidative and oncogenic, but not UV or ionising radiation) induced nuclear localisation, which was reversed upon removal of the stress stimulus [58]. How DHCR24 interacts with the membrane has been less well characterised. Using molecular docking and dynamics simulations, the N-terminus of DHCR24 is predicted to be partially embedded in the membrane as a stem or “peduncle” (meaning “little foot”) with the C-terminus cytoplasmic [44]. However, other studies suggest the N-terminus of DHCR24 traverses the whole lipid bilayer, as a single transmembrane domain [81]. One of the aims of this thesis is to characterise the cellular membrane topology of DHCR24 (Chapter 3).

1.8 DHCR24 regulation

Since its discovery, few regulatory mechanisms have been characterised for DHCR24 at the transcriptional and post-translational levels.

1.8.1 Transcriptional regulation of DHCR24

The DHCR24 proximal promoter is extremely GC rich (~80%) and lies within a CpG island, a region of DNA with a high G and C content and a high frequency of CpG dinucleotides [83, 84]. This region is important in the epigenetic regulation of DHCR24 by both DNA methylation and histone acetylation, with pronounced methylation occurring in cell types with low DHCR24 levels [83]. Four SNPs have been identified in the DHCR24 promoter (at -200, -488, -1420, and -1453, relative to the ATG start site) which correlate with HCV induced hepatocellular carcinoma and cirrhosis (Figure 1.7) [85]. Also, specificity protein 1 (Sp1) regulates DHCR24 in responsive to oxidative stress in HCV infection [71]. Hormonal regulation of DHCR24 has been extensively characterised. Sex steroids, such as estrogens and androgens, are positive regulators of DHCR24, increasing gene expression via activation of their respective nuclear receptors: estrogen receptor and androgen receptor [66, 86-89]. Putative binding sites have been located within the DHCR24 promoter (see Figure 1.7 for a summary of known transcription factor binding sites): estrogen receptor elements (ERE; -4148 and -3789) [87, 88], and an androgen receptor element (ARE; -2901) [66, 86]. Non-steroid hormones are also implicated in transcriptional regulation of DHCR24, such as adrenocorticotropic hormone (ACTH) [79, 80, 90, 91] and thyroid hormone (TH) [92-94]. The constitutive androstane receptor (CAR) plays a pivotal role in xenobiotic-induced expression of genes involved in drug metabolism and excretion [95]. CAR activation increased DHCR24 expression by inducing binding of CAR (−1593) to the promoter [96]. Similarly, the pregnane X receptor (PXR) was able to transactivate DHCR24 by binding to the same element [96]. A recent study has implicated DHCR24 as an LXR target gene [97]; however, most studies support a link between SREBP activity and DHCR24 expression [16, 36, 98-100], with one study observing a 40-fold induction of DHCR24 in rats treated with lovastatin and cholestyramine [36]. As SREBP and LXR have opposing effects on cellular sterol levels, it would be paradoxical if DHCR24 were regulated by both 18 transcription factors. One of the aims of this thesis is to determine which transcription factor regulates DHCR24 expression upon altered sterol status (Chapter 4).

Figure 1.7. Promoter map of human DHCR24. The DHCR24 gene (-5000/+1000) with its known regulatory sites marked. Specificity protein 1 (Sp1), estrogen response element (ERE), androgen response element (ARE), thyroid hormone receptor element (TRE), constitutive androstane receptor (CAR), pregnane X receptor (PXR). Modified from our recent review [46].

1.8.2 Post-translational regulation of DHCR24

Plant sterols reduce cholesterol synthesis by inhibiting DHCR24 [38]. Phytosterols are the plant equivalent of cholesterol, that differ by the presence of a double bond at C-22 and/or an additional alkyl group on C-24. They are known cholesterol-lowering agents, reducing cholesterol absorption in human intestines [101-103]. C-22 unsaturated sterols (phytosterols, stigmasterol and brassicasterol; yeast sterol, ergosterol) competitively inhibit DHCR24 enzyme activity, with no inhibition observed by phytosterols with a saturated side chain (β-sitosterol and campesterol) [38]. A desmosterol isomer, 5,22-cholestedien-3β-ol, was similarly able to inhibit DHCR24 [38]. The physiological relevance of this is unclear as these foreign sterols are unlikely to accumulate to sufficient concentrations to inhibit DHCR24, apart from in the rare disease sitosterolemia/phytosterolemia [89]. Progesterone, the major gestational steroid hormone, is a C21 steroid with a keto group at C-20. In cultured cells, progesterone and other similar progestins inhibit cholesterol synthesis, accompanied by accumulation of desmosterol [104-106]. One of the aims of this thesis is to determine if DHCR24 undergoes negative feedback regulation at the activity level by mammalian physiological regulators (sterol and oxysterol) of similar structure to the phytosterols inhibitors (Chapter 5). Another aim of this thesis is to investigate if the effect of progesterone on cholesterol synthesis is due to inhibition of DHCR24 activity (Chapter 5). Global proteomic studies have identified various post-translational modifications on DHCR24, including multiple ubiquitination [107-109] and phosphorylation sites [110, 111]. Ubiquitination and phosphorylation are important regulatory mechanisms in cholesterol homeostasis, acting upon key flux controlling enzymes in cholesterol synthesis (Section 1.5.2). Therefore, another aim of this thesis is to investigate if DHCR24 post-translational modifications, like phosphorylation, regulate its activity, and to determine their effect on cholesterol synthesis (Chapter 6).

1.9 Aims and hypotheses

Although there is increasing interest in DHCR24, there is little work on its role in cholesterol homeostasis. There is also a deficit of fundamental information on DHCR24, such as its structure and interaction with cellular membranes, and also how it is regulated. Therefore, the central aim of this thesis is to examine if DHCR24 has the potential as a control point in cholesterol synthesis beyond HMGCR and SM. We hypothesise that cholesterol levels are attenuated by multiple feedback mechanisms acting upon DHCR24. This leads to the specific aims: 1. To examine how DHCR24 interacts with membranes and to characterise its topology (Chapter 3). 2. To elucidate if DHCR24 is subject to sterol feedback mechanisms, at the transcriptional (Chapter 4) and post-translational level (Chapters 5). 3. To investigate the role of progesterone as a negative feedback regulator of steroidogenesis at DHCR24 (Chapter 5). 4. To determine if phosphorylation affects DHCR24 function (Chapter 6).

Chapter 2

Materials and Methods

2 MATERIALS AND METHODS

This chapter describes the materials and methods generally used throughout this thesis. Materials and methods specific to only one research chapter are described in the chapter materials and methods.

2.1 General Materials

Company Materials Acros Organics (Geel, Belgium) Hepes AGFA (Mortsel, Belgium) Developer Fixer Agilent Technologies (Santa Clara, XL10-Gold ultracompetent Escherichia CA, USA) coli cells Ajax FineChem (Taren Point, NSW, Acetic acid AU) Boric acid Diethyl ether Dimethyl sulfoxide (DMSO) Ethanol Glycerol Glycine Heptane Hexane Hydrochloric acid (HCl) Isopropanol Methanol Potassium chloride (KCl) Sodium chloride (NaCl) Sodium hydroxide (NaOH) Sucrose Tris-base (Tris)

APS Chemicals (Seven Hills, NSW, Magnesium chloride (MgCl2) AU) Sodium carbonate (Na2CO3) Avanti Polar Lipids (Alabaster, AL, [2H ]-desmosterol USA) 6 Bio-Rad Laboratories (Hercules, CA, 40% Acrylamide/bis-acrylamide USA) solution (37.5:1) Precision Plus Protein Kaleidoscope 25

Standards Trans-Blot nitrocellulose transfer medium Bioline (London, UK) Deoxynucleotide triphosphates (dNTPs) MangoTaq DNA polymerase 5 × MangoTaq Reaction Buffer SensiMix SYBR No-Rox Bovogen (East Keilor, Vic, AU) Bovine serum albumin (BSA) Fetal calf serum (FCS) Coles Supermarket (Randwick, NSW, Skim milk powder (Diploma brand) AU) Enzo Life Sciences (Farmingdale, NY, 24(S),25-Epoxycholesterol (24,25EC) USA) GE Healthcare (Buckinghamshire, Amersham Hyperfilm ECL England) Glaxo-Smith Kline (Middlesex, UK) OSC inhibitor (GW534511X) Jackson ImmunoResearch Horseradish peroxidase-conjugated Laboratories, Inc. (West Grove, PA, AffiniPure donkey anti-mouse and USA) anti-rabbit IgG antibodies Life Technologies (Carlsbad, CA, 5 × sequencing buffer USA) BigDye v3.1 DH5α (competent E. coli cells) Dithiothreitol (DTT) Dulbecco’s Modified Eagle’s Medium (DMEM) high glucose DMEM: Ham’s Nutrient Mixture F-12 (1:1) (DMEM/F-12) GlutaMAX Hygromycin B Lipofectamine LTX transfection reagent Lipofectamine RNAiMAX transfection reagent Flp-In Complete System Luria-Bertani (LB) powder Newborn calf serum (NCS) Opti-MEM I Reduced Serum Medium pcDNA3.1-V5-His TOPO TA Cloning Kit Penicillin-Streptomycin Roswell Park Memorial Institute 1640 medium (RPMI) 26

Salmon sperm DNA SOC bacterial growth medium SuperScript III First Strand cDNA Synthesis Kit TrypLE Express UltraPure Agarose Zeocin Macherey-Nagel (Düren, Germany) NucleoBond Xtra Midi Plus Merck (Darmstadt, Germany) Akt inhibitor VIII 0.22 μm filters Silica Gel 60 F254 plates Millipore (Billerica, MA, USA) Immobilon Western Chemiluminescent horse radish peroxidase substrate enhanced chemiluminescent detection system Mirus Bio (Madison, WI, USA) TransIT-2020 transfection reagent New England Biolabs (Ipswich, MA, DpnI USA) 5 × Phusion GC Buffer (for high GC content DNA) 5 × Phusion High-Fidelity Buffer Phusion Hot Start II High-Fidelity Polymerase Perkin Elmer (Waltham, MA) [14C]-Acetate sodium salt (specific radioactivity: 56.8 mCi/mmol, 2.1 GBq/mmol) Promega (Madison, WI, USA) Dual Luciferase Reporter Assay System pGEM-T-easy vector pGL3-basic luciferase reporter vector RBC Bioscience (Taiwan, China) HiYield Plasmid Mini kit Restek (Waltham, MA, USA) Rxi-5Sil MS w/Integra-Guard capillary GC column (30 m x 0.25 mm, 0.25 μm film thickness) Sigma-Aldrich (St. Louis, MO, USA) 1-bromo-3-chloropropane 5β,6β-epoxycholesterol (5β,6βEC) 20α-hydroxycholesterol (20HC) 22(R)-hydroxycholesterol (22HC) 25-hydroxycholesterol (25HC) Agar Ammonium persulfate (APS) Ampicillin ATP 27

β-mercaptoethanol Bisindolylmaleimide (BIM) Bromophenol blue BSA (essentially fatty acid free) BSTFA (N-O-bis-(trimethylsilyl) trifluoroacetamide) Butylated hydroxytoluene Compactin (mevastatin) Cyclohexamide Ethidium bromide Ethylenediaminetetraacetic acid (EDTA) Ethylene glycol-bis(2-aminoethylether)- N,N,N′,N′-tetraacetic acid (EGTA) Desmosterol Dulbecco’s phosphate-buffered saline (PBS) H89 Methyl-β-cyclodextrin (CD) Mevalonate MG132 (Z-Leu-Leu-Leu-al) Nonidet P-40 Oligonucleotide primers (including Cy5) Progesterone Protease inhibitor cocktail Ro-318220 Silver Nitrate siRNA Sodium deoxycholate Sodium dodecyl sulfate (SDS) Sodium oleate TEMED (N,N,N′,N′- Tetramethylethylenediamine) TO-901317 TRI Reagent Triparanol Triton X-100 Trizma HCl Trypsin Tween20 U18666A

Steraloids (Newport, RI, USA) 7α-hydroxycholesterol (7αHC) 7β-hydroxycholesterol (7βHC) 7-ketocholesterol (7KC) 19-hydroxycholesterol (19HC) 27-hydroxycholesterol (27HC) Cholesterol Thermo Fisher Scientific (Waltham, Bicinchoninic acid protein (BCA) assay MA, USA) kit

5α,6α-epoxycholesterol (5α,6αEC), 4-hydroxycholesterol (4HC), and 24(S)-hydroxycholesterol (24HC) were kind gifts from the Centre for Vascular Research (UNSW). The PKC epsilon peptide inhibitor (εV1-2) was a kind gift from Dr Carsten Schmitz-Peiffer (Garvan Institute of Medical Research, Australia) [112].

2.1.1 Equipment

Company Equipment Biolab (Scoresby, Vic, AU) Nanodrop ND-1000 Bio-Rad Laboratories (Hercules, CA, USA) C1000 Thermal Cycler Corbett Research (Mortlake, NSW, AU) Rotor-Gene 3000 FLA-5100 phosphorimager and Fujifilm (Tokyo, Japan) fluorescence scanner Gelaire (Seven Hills, NSW, AU) Laminar flow cabinet Heraeus (Hanau, Germany) Incubator BB 15 Sigma-Aldrich (St. Louis, MO, USA) Kontes motorised micropestle Thermo Trace gas chromatograph (Thermo Fisher Scientific; Waltham, MA, (GC) USA) Thermo DSQII mass spectrometer

(MS)

Thermo Triplus Autosampler Turner BioSystems (Sunnyvale, CA, USA) Veritas microplate luminometer

2.1.2 Software

Company Software Biomatters (Auckland, NZ) Geneious Basic (version 6.0.6) Corbett Researc (Doncastor, Vic) Rotor Gene 6.1 (Build 81) Science lab Image Gauge 2001 Fujifilm (Tokyo, Japan) (version 4.0) Mekentosj/Nucleobytes (Aalsmeer, The 4peaks (version 1.7) Netherlands) GraphPad (La Jolla, CA, USA) Prism 6 (version 6.02) Microsoft (North Ryde, NSW, AU) Excel 2010 PowerPoint 2010 Word 2010 NIH (Bethesda, MD, USA) ImageJ (version 1.47t) Thermo Xcalibur Software Thermo Fisher Scientific (Waltham, MA, USA) (version 2.1.0.1140)

2.1.3 Buffer recipes

Buffer Recipe 5 × SDS-PAGE loading buffer 250 mM Tris-HCl (pH 6.8) 10% (w/v) SDS 25% (v/v) glycerol 0.2% (w/v) bromophenol blue 5% (v/v) β-mercaptoethanol 5 × nPAGE loading buffer 250 mM Tris-HCl (pH 6.8) 25% (v/v) glycerol Buffer A 10 mM Hepes-KOH (pH 7.4) 10 mM KCl

1.5 mM MgCl2 5 mM EDTA 5 mM EGTA 250 mM sucrose Buffer B 20 mM Hepes-KOH (pH 7.6) 25% (v/v) glycerol 0.42 M NaCl

1.5 mM MgCl2 5 mM EDTA 5 mM EGTA Buffer C Buffer A containing 100 mM NaCl. Buffer D 10 mM Tris-HC1 (pH 7.4) 100 mM NaC1 1% (w/v) SDS Running Buffer 25 mM Tris 192 mM glycine 0.1% (w/v) SDS Transfer Buffer 18 mM Tris 150 mM glycine 10% (v/v) methanol Stripping Buffer 25 mM glycine 1.5% (w/v) SDS pH 2.0 PBST (Phosphate-buffered saline/Tween20) 0.1% (v/v) Tween20 PBS Resuspension Buffer 10 mM Hepes-KOH (pH 7.4) 10 mM KCl

1.5 mM MgCl2 5 mM EDTA

5 mM EGTA 250 mM sucrose 5 mM DTT Nuclear Extract Buffer 20 mM HEPES-KOH (pH 7.6) 25% (v/v) glycerol 0.42 M NaCl

1.5 mM MgCl2 1 mM EDTA 1 mM EGTA Binding Buffer 10 mM Tris-HCl (pH 7.5) 50 mM KCl

5 mM MgCl2 1 mM DTT 1 mM EDTA 6% (v/v) glycerol 0.05% nonidet P-40 0.25 mg/mL salmon sperm DNA TBE Buffer (Tris/Borate/EDTA) 0.1 M Tris 90 mM boric acid 1 mM EDTA Electrophoretic mobility shift assay (EMSA) 50% (v/v) TBE Buffer Running Buffer 10 × Native PAGE mix 50% (v/v) TBE Buffer 25% (v/v) glycerol LB broth (and agar) per 1 L 20 g LB powder 5 g NaCl (15 g Agar)

2.1.4 Polyacrylamide gel recipes

Components 7.5% 10% 4% 1.5 M Tris-HCl (pH 8.8) 2 mL 2 mL - 0.5 M Tris-HCl (pH 6.8) - - 0.63 mL Acrylamide 1.5 mL 2 mL 250 μL 10% (w/v) SDS 80 μL 80 μL 25 μL 10% (w/v) APS 85 μL 85 μL 12.5 μL TEMED 5 μL 5 μL 2.5 μL Water 4.33 mL 3.83 mL 1.6 mL

For electrophoretic mobility shift assays (EMSAs; Section 4.2.5), 6% (w/v) native polyacrylamide gels were used: 1 mL 10 × Native PAGE mix, 1.5 mL Acrylamide, 7.34 mL water, 150 μL 10% (w/v) APS, 10 μL TEMED.

2.2 Cell culture

Cell culture was performed in a laminar flow biological safety cabinet. Cells were grown as a monolayer at 37 °C in a 5% CO2 atmosphere, and stored in liquid nitrogen in the appropriate culture media supplemented with 5% DMSO. Cell culture media (Table 2.1) was sterilised by filtration using a 0.22 μm filter. Lipoprotein-deﬁcient serum was prepared from NCS (LPDS) or FCS (FCLPDS) as described [113, 114]. Base medium for cell culture was DMEM/F-12, or DMEM (high glucose). Cells were treated with test agents in ethanol or DMSO as stated in figure legends, with solvent concentrations remaining constant between treatments, and not exceeding 0.2% (v/v). Sterols were delivered either in ethanol or complexed with methyl-β-cyclodextrin (CD; prepared as described previously [115]) as indicated.

Table 2.1. Cell culture media.

Medium Composition A DMEM/F-12 supplemented with 5% (v/v) LPDS B Medium A supplemented with 1 μg/mL 25HC DMEM/F-12 supplemented with 5% (v/v) NBS, 5 μg/mL C cholesterol, 1 mM sodium mevalonate, 20 μM sodium oleate D RPMI 1640 medium supplemented with 10% (v/v) FCS E DMEM (high glucose) supplemented with 10% (v/v) FCS F RPMI 1640 medium supplemented with 10% (v/v) FCLPDS G Medium A supplemented with 500 μg/mL zeocin H Medium A supplemented with 150 μg/mL hygromycin B

2.2.1 Cell-lines

The following cells were all kind gifts from various researchers and institutions, as indicated. Chinese hamster ovary (CHO)-7 [116], SRD-1 (overexpress the nuclear form of SREBP-2, without the sterol regulated proteolytic step, resulting in high basal expression of SREBP-2 target genes [116]), and SRD-13A cells (lack the SREBP cleavage activating protein (Scap), resulting in no proteolytic activation of SREBP [117]) were from Drs. Michael S. Brown and Joseph L. Goldstein (UT Southwestern Medical Center, TX, USA). HeLaT and BE(2)C cells were from Drs. Noel Whitaker and Louise Lutze-Mann (UNSW). HepG2 and J774 cells were from The Centre for Vascular Research (UNSW). The following stable cell-lines were all prepared in the laboratory by various students. The Flp-In-CHO-7 cell-line was generated from CHO-7 cells by a previous student of the laboratory, Gorjana Mitic. The CHO-7 cell-lines stably expressing: the pcDNA5/FRT/TO empty expression plasmid, CHO-EV [118]; pCMV4-OSC-Myc, CHO-OSC [119]; pCMV-SM-myc, CHO-SM were generated by previous and present PhD students, Julian Stevenson, Jenny Wong, and Saloni Gill, respectively. CHO-7 cell-lines stably expressing DHCR24 (CHO-DHCR24), and mutated DHCR24 (CHO- DHCR24-T110A-V5, CHO-DHCR24-T110E-V5) were generated by Eser Zerenturk (Chapter 2.2.2).

Table 2.2. Cell culture conditions. Cell-line Medium Maintain Treat CHO-7 A A SRD-1 B A SRD-13A C A HeLaT D F BE(2)C E F HepG2 E F J774 D F CHO-SM-myc G A CHO-OSC-myc H A CHO-EV H A CHO-DHCR24 H A CHO-DHCR24-T110A H A CHO-DHCR24-T110E H A

2.2.2 Generation of the CHO overexpressing cell-line

The Flp-In system was used to generate stable DHCR24 expression in CHO-7 cells. Flp-In-CHO-7 cells contain a single Flp recombinase target (Frt) site integrated into the genome. These cells were cotransfected with pOG44 (a Flp recombinase expression vector) and a Flp-In expression construct [containing the DHCR24 coding sequence (WT, T110A, or T110E), a FRT site, and a hygromycin resistance gene] in Medium A, according to the manufacturer’s instructions, using Lipofectamine LTX for 24 h. The Flp recombinase mediates a homologous recombination event between the Frt sites, thus allowing integration of the Flp-In construct into the genome. Cells were washed with PBS then refreshed with Medium A for 24 h. The cells were then split at ~25% confluence and 150 µg/mL hygromycin B (Medium H) was added to apply selection pressure to maintain stable expression; cells were refreshed with Medium H every 3-4 days. When distinct colonies formed, the cells were washed with PBS, and 5% (v/v) 35

TrypLE/PBS was added to allow colonies to be picked up, and transferred to individual wells of a 12-well plate containing Medium H. The cells were refreshed with Medium H every 3-4 days until they were sufficiently confluent to split into a 6-well plate. Stable transfectants were then screened for DHCR24-V5 expression by Western blotting (Section 2.5).

2.3 Plasmids

The human DHCR24 coding sequence was amplified from human prostate epithelial cell cDNA and cloned into pcDNA3.1-V5-His TOPO TA, according to the manufacturer’s instructions. This coding sequence was also subcloned into a Flp-In expression construct to generate DHCR24 overexpressing cells. Truncations and/or mutations were made to both (Table 2.3).

Table 2.3. Expression plasmids.

Plasmid Description Source1 DHCR24 with a C-terminal V5 tag DHCR24-V5 Created in house (JS) (pcDNA3.1) DHCR24 with an N-terminal V5 tag V5-DHCR24 Created in house (EZ) (pcDNA3.1) DHCR24 lacking the first 56 aa, with a DHCR24-V5-Δ56 Created in house (EZ) C-terminal V5 tag (pcDNA3.1) DHCR24-V5- DHCR24 lacking the first 160 aa, with a Created in house (LS) Δ160 C-terminal V5 tag (pcDNA3.1) DHCR24-V5- DHCR24 lacking the first 240 aa, with a Created in house (LS) Δ240 C-terminal V5 tag (pcDNA3.1) Gift from Drs. Michael Brown and Joseph Insig-1 with a C-terminal 6×myc tag Insig-myc Goldstein, University (pCMV) of Texas Southwestern, USA DHCR24-FRT- DHCR24 with a C-terminal V5 tag Created in house (JS) V5 containing a single FRT site (pcDNA5) DHCR24-FRT-V5 with the threonine at DHCR24-T110A- 110 (T110) mutated to an alanine to Created in house (EZ) FRT-V5 prevent phosphorylation DHCR24-FRT-V5 with T110 mutated DHCR24-T110E- to a glutamic acid to mimic Created in house (EZ) FRT-V5 phosphorylation A CMV driven expression plasmid Gift from Dr John encoding the mature N-terminal form nSREBP-2 Shyy, University of (aa 14-480) of human SREBP-2 California, USA [120] (nSREBP-2) A thymidine kinase driven expression TK-nSREBP-2 Created in house (JK) plasmid encoding human nSREBP-2 nSREBP-2 plasmid with the tyrosine at nSREBP-2(Mut)] aa 342 mutated to an arginine residue to Created in house (JW) prevent DNA binding [121]

1 Plasmids were generated by various people in the laboratory. PhD students: EZ, Eser Zerenturk; JK, James Krycer; JS, Julian Stevenson; JW, Jenny Wong. Research associate: LS, Laura Sharpe.

The promoter of the human DHCR24 gene (~1.2 kb upstream from the ATG start site) was prepared by polymerase chain reaction (PCR) amplification from human

fibroblast genomic DNA and cloned into pGEM-T-easy. This fragment was then subcloned into the Firefly luciferase pGL3-basic vector, with truncations and mutations made (Table 2.4).

Table 2.4. Promoter plasmids in Firefly and Renilla luciferase vectors. Various human promoter sequences fused to a Firefly luciferase gene, or unless stated, a Renilla luciferase gene. Numbering is relative to the ATG start codon. Plasmid Promoter Description Source1 pGL3-basic empty - Promega vector -1200luc DHCR24 promoter Created in house (EZ and LS) -900luc DHCR24 promoter truncations -650luc Created in house (LS) -300luc Created in house (EZ and LS) -100luc -275mut -300luc with SRE mutations -254mut -244mut -240mut -196mut -160mut -119mut -109mut -196/-160mut -196LDLR -300luc with the SRE swapped with the LDLR SRE -160LDLR -160↔-196inv -300luc with the -160 invSRE and -196 invSRE swapped 3×(-196inv) 3 × tandem -196 invSRE 3×(-160) 3 × tandem -160 SRE -225mut -300luc with NF-Y mutations

-174mut -131mut -176mut -300luc Sp1 mutation SREs 5del -300luc with progressive deletions between -160 invSRE and -196 SREs 10del invSRE (5, 10, 15, 35 bp) SREs 15del SREs 35del LDLRluc The human LDLR promoter [122] LDLR-mutSRE LDLRluc with the SRE mutated Created in house (JK) [114] FDFT1luc The human FDFT1 promoter Gift from Dr Ryuichiro Sato [123] phRL-PBGD A PBGD (-460/+5) driven Created in house expression plasmid of Renilla (JK) [114] luciferase.

1 Plasmids were generated by various people in the laboratory. PhD students: EZ, Eser Zerenturk; JK, James Krycer; JS, Julian Stevenson; JW, Jenny Wong. Research associate: LS, Laura Sharpe.

2.3.1 Cloning

To create the plasmids used in this thesis, two PCR based cloning methods were utilised: two step megaprimed site-directed mutagenesis and polymerase incomplete primer extension (PIPE). As a general rule, 45 s/kb of was used to determine the time required for DNA extension, and the melting temperature of the primers + 5 °C was used for the annealing temperature in the first instance. Following amplification, 5 µL PCR product was visualised on an agarose gel using ethidium bromide, and if no product formed, a temperature gradient PCR was performed.

2.3.1.1 Two step megaprimed site-directed mutagenesis

To generate the Δ56 DHCR24 truncation construct (Chapter 3), and DHCR24 phosphomutant and phosphomimetic Flp-In constructs (Chapter 6), two step megaprimed site-directed mutagenesis was performed [124], with minor modifications. Mutagenic megaprimer was created in the first PCR using 10 ng template plasmid, 5 µL 39

5 × Phusion GC Buffer, 0.4 U Phusion Hot Start II High-Fidelity DNA polymerase, 5% (v/v) DMSO, 0.5 µM each of forward and reverse primers and 200 µM dNTPs in a total volume of 20 µL [98 °C 3 min, 5 × (98 °C 30 s, 55 °C 1 min, 68 °C 2 min)]. Megaprimer extension occurred in the second reaction by linear amplification of 10 ng template plasmid using 1.5 μL of the megaprimer generated from first PCR, using 5 µL 5 × Phusion GC Buffer, 0.4 U Phusion Hot Start II High-Fidelity DNA polymerase, 5% (v/v) DMSO and 200 µM dNTPs in a total volume of 20 µL [98 °C 3 min, 20 × (98 °C 30 s, 68 °C 4 min)]. Products were digested by DpnI (20 U) at 37 °C overnight to degrade methylated (template) DNA, and then the mutated/truncated plasmid was transformed into XL10-Gold Ultracompetent cells (Section 2.3.2).

2.3.1.2 Polymerase incomplete primer extension (PIPE)

To generate V5-DHCR24, and the Δ160 and Δ240 DHCR24 truncation constructs (Chapter 3), DHCR24 promoter constructs (Chapter 4), and the DHCR24 Flp-In construct (Chapter 5), polymerase incomplete primer extension (PIPE) was performed [125], with minor modifications. PIPE was performed using 5 ng template plasmid, 5 µL 5 × Phusion GC Buffer, 0.4 U Phusion Hot Start II High-Fidelity DNA polymerase, 5% (v/v) DMSO, 125 nM each of forward and reverse primers, and 200 µM dNTPs in a total volume of 20 µL [98 °C 3 min, 30 × (98 °C 30 s, 72 °C 45 s, 72 °C 2.5 min), 98 °C 30 s, 85 °C 2 min, 50 °C 15 min]. Products were digested by DpnI (20 U) at 37 °C overnight to degrade methylated (template) DNA, and then the mutated/truncated plasmid was transformed into XL10-Gold Ultracompetent cells (Section 2.3.2).

2.3.2 Transformation using XL-10 Gold cells

To increase transformation efficiency, 1 µL β-mercaptoethanol was added to 25 µL XL10-Gold cells and incubated on ice for 10 min, gently shaking every 2 min. 5 µL plasmid was added to the cells on ice for 30 min prior to heat-shocking them at 42 °C for 30 s, followed by incubation on ice for a further 2 min, which was added to 475 µL SOC medium, and incubated at 37 °C for 1 h, with vigorous shaking. The cell suspension was spread-plated on LB agar plates containing ampicillin (100 µg/mL), and incubated overnight at 37 °C. Cloning was verified by colony PCR (Section 2.3.3) for large deletions or insertions, or miniprep (Section 2.3.4) followed by sequencing (Section 2.3.5) for small deletions and insertions, or mutations.

2.3.3 Colony PCR

A small amount of bacteria was picked up from a single colony into 20 µL sterile water, vortexed and heated at 95 °C for 5 min to lyse the cells and release DNA. Cellular debris was pelleted by centrifugation of samples at 13,000 × g for 2 min, and the supernatant was used as a source of DNA. A colony PCR was performed using 2 µL of this DNA, 2 µL 5 × Mango Taq Reaction Buffer, 1.5 U Mango Taq DNA polymerase,

5% (v/v) DMSO, 200 µM dNTPs, 1.5 mM MgCl2, 125 nM each of forward and reverse primers, and in a final volume of 10 µL per reaction [95 °C 3 min, 35 × (95 °C 30 s, 53 °C 45 s, 68 °C 1.5 min), 68 °C 7 min]. Following amplification, 5 µL PCR product was visualised on an agarose gel using ethidium bromide. Once cloning was verified, large quantities of plasmid DNA were generated by midiprep (Section 2.3.6).

2.3.4 Miniprep

A single colony was used to inoculate 3.5 mL LB broth supplemented with 100 µg/mL ampicillin, and incubated at 37 °C with vigorous shaking overnight. The cell suspension was harvested by centrifugation at 20,000 × g for 5 min. The supernatant was removed, and the plasmid DNA was extracted from the cell pellet using HiYield Plasmid Mini kit, according to the manufacturer’s instructions.

2.3.5 Sequencing

Sequencing reactions contained ~200-300 ng DNA, 3.2 pmol primer, 4 μL5 × Sequencing Buffer, 1 μL BigDye v3.1 and 1 μLDMSO in a final volume of 20 μL. Cycling conditions for sequencing were 25 × (96 °C 10 s, 50 °C 5 s, 60 °C 4 min). Sequencing products were precipitated by the addition of 5 μL 125 mM EDTA and 60 μL 100% ethanol, incubated at room temperature for 45 min, and centrifuged at 16,000 × g for 20 min. The supernatant was removed and the pellet was washed with 250 μL 70% (v/v) ethanol and centrifuged for a further 10 min. The supernatant was removed and the pellet air dried for 10 min. DNA sequencing reactions were performed by the Ramaciotti Centre for Gene Function Analysis, UNSW. Once cloning was verified, large quantities of plasmid DNA were generated by midiprep.

2.3.6 Midiprep

Large quantities of plasmid DNA was prepared by adding 10 ng plasmid to 25 µL DH5α cells and incubating on ice for 30 min. The cells were then heat-shocked at 37 °C for 20 s, and incubated on ice for a further 2 min. 475 µL SOC medium was then added to the cells, and incubated at 37 °C for 1 h, with vigorous shaking. The cell suspension was spread-plated on LB agar plates containing 100 µg/mL ampicillin, and incubated overnight at 37 °C. A single colony was selected and inoculated in 3.5 mL LB broth supplemented with 100 µg/mL ampicillin, and was incubated at 37 °C with vigorous shaking for 6 h before adding to 100 mL LB broth 100 µg/mL ampicillin. The cells were incubated overnight 37 °C with vigorous shaking. The cell suspension was harvested by centrifugation at 3,000 × g for 5 min. The supernatant was removed, and the plasmid DNA was extracted from the cell pellet using the NucleoBond Xtra Midi Plus, according to the manufacturer’s instructions.

2.3.7 Plasmid transfection

Plasmids (Table 2.3 and 2.4) were transiently transfected into CHO cells using Lipofectamine LTX Reagent using a ratio of 1 μg DNA: 4 μL transfection reagent, according to the manufacturer’s instructions, for 24 h. HeLaT cells were transiently transfected using TransIT-2020 using a ratio of 1 μg DNA: 2 μL transfection reagent, according to the manufacturer’s instructions, for 24 h.

2.4 Sterol Synthesis Assay

2.4.1 Lipid extraction

The effect of various treatments on cholesterol and desmosterol synthesis were investigated using various chromatographic methods. Cells were seeded in 6 well plates, and treated the following day. For Chapter 5, cells were statin pre-treated overnight to deplete sterols with the HMGCR inhibitor, compactin (5 μM), and a low concentration of mevalonate (50 μM), that allows the synthesis of essential non-sterol isoprenoids but not cholesterol. Cells were treated and metabolically labelled as indicated in the figure legends for 1-4 h. Cells were then lysed in 500 µL 0.1 M NaOH

and protein concentrations measured by the BCA assay. Lysates, and in some experiments, media, were saponified with 1 mL 100% ethanol, 500 µL 75% (w/v) KOH, 1 µL 20 mM butylated hydroxytoluene, and 20 µL 20 mM EDTA at 70ºC for 1 h. After cooling, non-saponifiable lipids were extracted into 2.5 mL hexane for cholesterol and desmosterol analysis (Sections 2.4.2 and 6.2.3) or 2.5 mL hexane:diethyl ether (1:1, v/v) for 24,25EC analysis (Section 5.2.1), and evaporated to dryness.

2.4.2 Argentation Thin layer chromatography

For measurement of desmosterol and cholesterol synthesis, lipid extracts were re-dissolved in 60 μL hexane, and aliquots corresponding to equivalent amounts of protein were separated by argentation thin layer chromatography (Arg-TLC) using 4%

(w/v) silver-coated Silica Gel 60 F254 plates with a mobile phase of heptane:ethyl acetate (2:1, v/v), run four times. The bands corresponding to cholesterol and desmosterol were visualised using the FLA-5100 Phosphorimager. The relative intensities of bands were quantified using Science lab Image Gauge.

2.5 Western blotting

Western blotting was used to determine protein content in whole cell lysates and cellular fractions.

2.5.1 Preparation of cell lysate for SDS-PAGE

For cell fractionation and membrane isolation (Chapter 3), cells were lysed in Buffer A and passed through an 18 gauge needle 50 times. For preparation of whole cell lysates (Chapters 4, 5 and 6), cells were lysed in 10% (w/v) SDS supplemented with 5 μL (v/v) protease inhibitor cocktail, and homogenised using a Kontes motorised pestle. Protein concentrations were measured by the BCA assay, and equal amounts of protein were mixed with 5 × SDS-PAGE loading buffer, and boiled for 5 min.

2.5.2 SDS-PAGE

Samples were resolved by SDS-PAGE with a 7.5% or 10% separating gel, and a 4% stacking gel (Section 2.1.4). Samples (and molecular weight standards) were loaded and electrophoresed in Running Buffer at 200 V and 400 mA for 45-80 min. After electrophoresis, the proteins were transferred to a nitrocellulose membrane in Transfer Buffer and electrophoresed at 200 V and 400 mA for 1.5 h.

2.5.3 Western blotting

Membranes were incubated in 5% (w/v) skim milk/PBST for 1 h, with gentle agitation; then incubated with primary antibody (Table 2.5) for 1 h at RT, or overnight at 4 °C, with gentle agitation. Membranes were washed 3 × 10 min with PBST before incubation with secondary antibody (anti-mouse or anti-rabbit conjugated to horseradish peroxidase, all raised in donkeys). Membranes were washed 3 × 10 min with PBST, and visualised using the Immobilon Western Chemiluminescence horse radish peroxidase ubstrate, then developed by Hyperfilm ECL. Protein band intensities from Western blots were quantified by densitometry using ImageJ in Chapter 6. Prior to blocking and incubating with a different primary antibody, membranes were incubated 2 × 15 min with Stripping Buffer. The observed protein bands migrated according to their predicted molecular weight of approximately: 60 kDa for DHCR24; 50 kDa for α-tubulin; 37 kDa for Insig-1, which appears as a doublet; 30 kda kDa for IgG-R139)

Table 2.5. Antibodies used for Western Blotting in this thesis. Antibody Dilution Diluent Secondary Company V5 1:5,000 5% skim milk/PBST Mouse Life Technologies DHCR241 1:2,500 5% skim milk/PBST Rabbit Abcam Santa Cruz c-Myc 1:10,000 5% BSA/PBST Mouse Biotechnology α-tubulin 1:200,000 5% skim milk/PBST Mouse Sigma-Aldrich Gift from Drs. Michael Brown and IgG-R1392 1:10,000 5% skim milk/PBST Rabbit Joseph Goldstein, University of Texas Southwestern, USA IgG-7D43 - 5% BSA/PBST Mouse Prepared in-house

1 Rabbit polyclonal antibody against hamster DHCR24. 2 Rabbit polyclonal antibody against the trypsin-protected hamster SREBP-cleavage activating protein (Scap) fragment (aa 54-277 and 540-707) [126]. 3 Mouse monoclonal antibody against the N-terminus of hamster SREBP-2 (aa 32-250) [127].

2.6 Quantitative real-time reverse transcription PCR (qRT-PCR)

Cells were seeded in triplicate in 12-well plates, and treated for 24 h as described in figure legends. Total RNA (1μg) was harvested using TRI Reagent according to the manufacturer’s instructions, and then reverse transcribed into cDNA using the SuperScript III First Strand cDNA Synthesis Kit on a C1000 Thermal Cycler., according to the manufacturer’s instructions, with slight modifications. The reactions were heated to 42ºC for 50 min for the reverse transcription reaction, and at 72ºC for 15 min to terminate the reaction. mRNA levels were determined by qRT-PCR using SYBR Green and the primers in Table 2.6 [98ºC 10 min, 40 × (95 ºC 15s, 55-60ºC 60s)]. The mRNA expression levels were normalised to the housekeeping control porphobilinogen deaminase (PBGD) and made relative to the vehicle condition using the ΔΔCt method. qRT-PCR experiments were performed by Laura Sharpe (Chapter 4) and Winnie Luu (Chapter 6).

Table 2.6. Primers for qRT-PCR.

Gene1 Direction Primer Sequence (5’3’) Reference F GCACTGAGGAAGATGCTGAAA hABCA1 [128] R AGTTCCTGGAAGGTCTTGTTCAC F ATAGCAGGCTCCAACCCTGAC haABCA1 [129] R GGTACTGAAGCATGTTTCGATGTT F ACGCAGTTCTGCATCCTCTTC hABCG1 [130] R TGTCAGAACAGTAGGCATGAG F GCAGAAGCTGTACACACTGCAG hFASN [131] R CAGGATGGGCACCTGCTGCT

F TCTTGGGGAGAAAACAGAGA Present hACC R TCGCTCAGCCTGTACTTTTT study F AGTTACATCATCACTGGTGG haACC [132] R ACGTTCTGCCTGCACTTTTT F TTGGTGATGGGAGCTTGCTGTG hHMGCR [133] R AGTCACAAGCACGTGGAAGACG F TGGTGATGGGAGCTTGCTGTG haHMGCR [134] R AATCACAAGCACGAGGAAGACG F TTCGAGTTCCACTGCCTAAG hLDLR [130] R TAACGCAGCCAACTTCATCG

F AGGCAGCTGGAGAAGTTTGT Present h, haDHCR24 R CCTCGCGGTTCATATAGCAATC study F GAGTGATTCGCGTGGGTACC hPBGD [128] R GGCTCCGATGGTGAAGCC F AGATTCTTGATACTGCACTC haPBGD [134] R TGAAAGACAACAGCATCACA

1 h, human; ha, hamster.

2.7 Data Presentation

Results shown are representative of at least three separate experiments, unless otherwise indicated. Data are presented as mean + SEM, unless otherwise indicated, and values are normalised to the control condition and/or cell-line as indicated. For Figures 4.1, 5.6, 5.11, and 6.6, statistical differences were determined using a paired t-test (two tailed) versus the vehicle condition, * p<0.05, and ** p<0.01 were considered statistically significant.

Chapter 3

The membrane topology of DHCR24

Protease protection assays, cell fractionation, differential solubilisation, Western blots and bioinformatics analysis were performed by Eser Zerenturk.

The following work has been submitted to Bioscience Reports for publication:

The following figure panels have been presented in:

Zerenturk EJ (2009) The regulation and membrane topology of Seladin-1 (Honours thesis): Figure 3.8 B Figure 3.9 A and B

3 THE MEMBRANE TOPOLOGY OF DHCR24

3.1.1 Introduction

Cholesterol synthesis enzymes are typically localised to the endoplasmic reticulum (ER), and as membrane-associated proteins, little has been done to determine their exact topology and structure. Cholesterol synthesis occurs in the membranes of the ER, thus most of the cholesterogenic machinery resides in this subcellular compartment. Indeed, early investigations into cholesterol synthesis used rat liver microsomes as a source of many cholesterogenic enzymes, including DHCR24 [135]. Similarly, DHCR24 homologs in A. thaliana [74] and B. mori [78] are localised to ER membranes (Section 1.7.2). Initial studies characterising DHCR24 by Greeve et al. [52] found it was targeted to the ER, and to a lesser degree, the Golgi, which was subsequently confirmed by Wu et al. [58]. Since the discovery of DHCR24, numerous binding partners and cofactors have been identified and characterised. DHCR24 contains a highly conserved FAD binding domain [44, 76, 136], which includes a stretch of 10 amino acids that are identical in the diverse species examined (Figure 3.1 and Appendix 8.2). Reduction of desmosterol is dependent on FAD [77], suggesting functionality of the conserved domain. In addition, DHCR24 contains conserved binding domains for both p53 and Mdm2, required for mediating cellular responses to oncogenic and oxidative stress [58] (Figure 3.1). DHCR24 is reported to be proteolytically cleaved during apoptosis, at two highly conserved caspase recognition motifs located in the FAD and p53 binding domains, presumably destroying them (Figure 3.1), producing a soluble, 40 kDa peptide [52]. How DHCR24 interacts with the membrane is less well known.

Figure 3.1. Predicted and known functional features of DHCR24 protein. The human sequence of DHCR24 (Q15392) is presented with the following features indicated: the region matching the Pfam [136] entry for an FAD-binding domain is overlined. Binding domains for p53 and Mdm2 are boxed. Caspase cleavage sites (Casp) are marked with a dotted line. Yellow shaded residues are strictly conserved in DHCR24 homologs from mouse, chicken, turkey, Chinese softshell turtle, anole lizard, zebrafish, chichlid fish, C. elegans (DIMINUTO), tomato, and A. thaliana (DWARF1); accession numbers are listed in Appendix 8.5.

Using simulated DHCR24 membrane models with and without substrate and cofactors, Pedretti et al. [44] predicted DHCR24 as a monotopic membrane protein; with the N-terminus partially embedded in the membrane as a stem or “peduncle” (meaning “little foot”), rather than traversing the membrane bilayer (bitopic) (Figure 3.2, Ped). This “peduncle” firmly anchors the protein to the membrane, with the C-terminus protruding into the cytoplasm, allowing access to substrates and cofactors [44]. DWF1 in A. thaliana has a similar predicted structure, based on hydropathy predictions: strong membrane association and a cytoplasmic C-terminus [74].

However, Lu et al. [81] published experimental work in support of a bitopic, single transmembrane domain (TMD) model of DHCR24: a luminal N-terminus, followed by a single TMD (also at the N-terminus), and cytoplasmic C-terminus (Figure 3.2, TMD). The N-terminus was essential for DHCR24 membrane association, as deletion of this region (1-58, Δ58 DHCR24) translocated DHCR24 to the cytoplasm. The orientation of the N- and C- termini were deduced from protease protection assays of fluorescent fusion constructs; an N-terminal Ds-Red fluorescent fusion protein was preserved, indicating that it was luminal, whereas a C-terminal green fluorescent protein was degraded by trypsin, indicating cytoplasmic localisation. By contrast, Pedretti et al.’s “peduncle” model [44] predicts that N-terminus is buried in the membrane, and the C-terminus is cytoplasmic (Figure 3.2). However, fusion of the large (~28 kDa), soluble, Ds-Red fluorescent protein to the N-terminus could feasibly draw the membrane–embedded N-terminus into the lumen, that is, interfere with the native membrane topology of DHCR24. In this chapter, we aim to determine the validity of the two different published models [44, 81] (Figure 3.2), investigating how DHCR24 interacts with membranes and its membrane topology.

Figure 3.2. DHCR24 membrane topology models. Peduncle (Ped) model proposed by Pedretti et al., with the N-terminus of DHCR24 embedded within the lipid bilayer (black), and a cytoplasmic C-terminus [44]; transmembrane domain (TMD) model proposed by Lu et al., with one TMD and a luminal N-terminus (grey) and cytoplasmic C-terminus [81]. FAD, Mdm2, and p53 domains as indicated. Not to scale.

3.2 Materials and Methods

3.2.1 Materials

The antibody against the trypsin Scap fragment was a generous gift of Drs. Michael Brown and Joseph Goldstein (UT Southwestern Medical Center, TX, USA): IgG-R139, a rabbit polyclonal antibody against hamster Scap (aa 54-277 and 540-707 [126]).

3.2.2 Cell fractionation

The cell fractionation protocol was adapted from Feramisco et al. [137]. Cells were washed and scraped into PBS on ice and centrifuged at 1,000 × g. The cellular pellet was resuspended in 400 μL Buffer A [10 mM Hepes-KOH (pH 7.4), 10 mM KCl, 1.5 mM MgCl2, 5 mM EDTA, 5 mM sodium EGTA, and 250 mM sucrose] and passed through an 18 gauge needle 50 times. Protein concentration of cell lysate was determined using the BCA assay. Cell lysate was centrifuged at 1,000 × g for 5 min at 4 °C. The 1,000 × g pellet was resuspended in 100 μL Buffer B [20 mM Hepes-KOH

(pH 7.6), 25% (v/v) glycerol, 0.42 M NaCl, 1.5 mM MgCl2, 5 mM EDTA, 5 mM sodium EGTA], rotated at 4 °C for 1 h, and centrifuged at 100,000 × g for 30 min at 4 °C. The supernatant from this centrifugation was collected and designated the nuclear fraction. The supernatant from the 1,000 × g centrifugation was centrifuged at 100,000 × g for 30 min at 4 °C. The supernatant was collected and designated the cytosolic fraction. The pellet was resuspended in 100 μL Buffer D [10 mM Tris-HC1 (pH 7.4), 100 mM NaC1, 1% (w/v) SDS], and designated the membrane fraction.

3.2.3 Membrane isolation

To prepare membranes for differential solubilisation (Section 3.2.4) and protease protection assays (Section 3.2.5), cells were washed and scraped in PBS and centrifuged at 1,000 × g for 5 min at 4 °C. The cellular pellet was resuspended in Buffer A or Buffer C (Buffer A containing 100 mM NaCl; as indicated in figure legends), passed through an 18 gauge needle 50 times, and the cell lysate was centrifuged at 1,000 × g for 5 min at 4 °C. The supernatant was then centrifuged at 20,000 × g for at least 15 min at 4 °C, and the resulting membrane pellet was resuspended in 65 μL Buffer A or C. Membrane protein concentration was determined using the BCA assay.

3.2.4 Differential solubilisation

To determine the strength of the protein-membrane interaction, equivalent amounts of membrane were treated with 200 μL 1% SDS, Buffer A, 0.1 M Na2CO3 (pH 11.5), or 1 M NaCl, and rotated at 4 °C for 30 min. Treated membranes were centrifuged at 100,000 × g for 30 min at 4 °C; the supernatant was designated the cytosolic fraction (C), and the pellet was resuspended in 200 μL Buffer C and designated the membrane fraction (M).

3.2.5 Protease protection assay

To determine the membrane orientation of the N- and C- termini, equivalent amounts of membrane were treated with trypsin as described [137]. Briefly, membranes were treated with the indicated amount of trypsin in the absence or presence of Triton X-100 in Buffer A, for 30 min at 30 °C. Reactions were stopped by the addition of nPAGE loading buffer and heat inactivation at 95 °C for 10 min.

3.2.6 Bioinformatics tools

For in silico analysis, the complete human amino acid sequence of DHCR24 (Q15392) was used. A signal peptide was predicted using SignalP v4.1 [138] and a hydrophobicity profile was returned by ProtScale [139]. Transmembrane domains and membrane topology were predicted using TMHMM 2.0 [140], TOPCONS [140], and ΔG predictor server v1.0 [141]. Myristoylator [142], NMT [143], big-PI predictor [144], PrePS [145], CSS-Palm 2.0 [146] were used to predict post-translational modifications involved in membrane attachment. To model the membrane topology, the LaTeX package TeXtopo [147] was employed, with a minor modification to the source code to enable a “half-loop” to extend beyond 14 amino acids.

3.3 Results

3.3.1 DHCR24 has an extremely hydrophobic N-terminus, suggesting a candidate region for transmembrane domains

To investigate the membrane topology of DHCR24, we first used an in silico approach. Transmembrane (TM) proteins are hydrophobic by nature, and can therefore be predicted based on this characteristic. The human DHCR24 sequence from UniProt (Q15392) was analysed for putative TM domains (TMDs) based on hydrophobicity by the method of Kyte and Doolittle (Figure 3.3 A) [139]. Two potential TMDs were identified at peak regions of hydrophobicity at the extreme N-terminus of DHCR24 (*, **; Figure 3.3 A), with scores above the recommended threshold by Kyte and Doolittle (dotted red line, 1.6) [139]. As DHCR24 is not an extremely hydrophobic protein overall, having a grand average of hydropathicity (GRAVY) score of -0.061 (blue line), other hydrophobic peaks are discernible above this value, but do not reach the Kyte and Doolittle threshold of 1.6 (Figure 3.3 A). A more specific prediction of regions that lie within the ER membrane is the free energy calculation for membrane insertion by means of the Sec61 translocon

(ΔGmi): negative ΔGmi values are indicative of potential TMDs of strong hydrophobicity and sufficient length; positive ΔGmi values are less likely to be TMDs [148]. Both potential TMDs identified by the method of Kyte and Doolittle have a negative ΔGmi, further indicating them as putative TMDs (*, **; Figure 3.3 B). Three additional regions were identified as putative TMDs based on their ΔGmi values (#,##, ###;; Figure

3.3 B); however, their ΔGmi values were positive, and these regions did not reach the threshold for detection of TMDs in the hydropathy plot. Therefore, based on their hydrophobicity and their ΔGmi values, they are less likely to form TMDs.

Predicted ΔG Rank order of Predicted TMD Position mi (kcal/mol) TMD Probability * 3-21 -1.147 2 ** 31-53 -2.034 1 # 136-156 2.177 4 ## 210-230 0.645 3 ### 349-369 2.926 5

Figure 3.3. Identification of candidate TMDs. A, A hydropathy profile of DHCR24 using the method of Kyte and Doolittle [139], returned by ProtScale with a window size of 19. The recommended threshold for the detection of TMDs is indicated by the dotted red line (1.6) [139], with peak regions of hydrophobicity above this marked in red (* and **). The grand average of hydropathicity (GRAVY) score is indicated by the solid blue line (-0.061). B, Putative TMDs and their probability of forming based on the apparent free energy difference for

ER membrane insertion by the Sec61 translocon (ΔGmi) [148]. Predicted TMDs based on ΔGmi below the 1.6 threshold are marked in blue (#, ##, ###).

3.3.2 DHCR24 is not predicted to associate with the membrane via post-translational modifications

Hydrophobic post-translational modifications can facilitate membrane attachment; and common groups include the irreversible attachment of various lipid groups, such as prenyl, myristoyl, or glycophosphatidylinositol (GPI) groups, or the reversible attachment of a palmitoyl group. The prediction programs Myristoylator [142], NMT

[143], big-PI predictor [144], PrePS [145], and CSS-Palm 2.0 [146] predict that DHCR24 is unlikely to associate with the ER membrane via PTMs; therefore, these means of membrane attachment were not considered for DHCR24.

3.3.3 Predicted membrane topology of DHCR24 consists of N-terminal TMDs

For advanced DHCR24 TMD prediction, TOPCONS and TMHMM membrane topology prediction programs were employed. Both of these programs use hidden Markov models (HMM) and/or neural network algorithms to integrate sequence information, and multiple sequence alignments (MSAs). TOPCONS uses multiple TMD prediction programs: SCAMPI-seq and SCAMPI-msa [149], PRODIV and PRO [150] and OCTOPUS [151]. The TOPCONS global prediction (TOPCONS consensus)

[149] is a consensus of these five predictions, as well as the ΔGmi [148] and ZPRED

[152] algorithms. Both regions identified by their hydrophobicity and ΔGmi as putative

TMDs (*, **; Figure 3.3) and the three regions with positive ΔGmi values (#,##, ###; Figure 3.3) were predicted by at least one of the TOPCONS prediction algorithms (Figure 3.4 A); however, only the two N-terminal hydrophobic regions (*, **; Figure 3.3) were uniformly predicted by all five of the individual algorithms, and the global TOPCONS prediction (TOPCONS consensus; Figure 3.4 A). Furthermore, the self-assessed reliability of the TOPCONS consensus is lowest in the region of the three other putative TMDs (#,##, ###). The TMHMM prediction [140], which also includes information on evolutionary conservation from MSAs, agreed with the TOPCONS consensus, predicting two N-terminal TMDs with high probability. However, TOPCONS predicted a cytosolic C-terminus, whereas TMHMM predicted it to be luminal (Figure 3.4 B).

Figure 3.4. DHCR24 membrane topology predictions. A, TMD predictions by the five TOPCONS algorithms: SCAMPI-seq and SCAMPI-msa [149], PRODIV and PRO [150] and OCTOPUS [151] and the TOPCONS global prediction (TOPCONS consensus) [149]. Potential TMDs are boxed grey (cytosol → lumen) or white (lumen → cytosol). B, DHCR24 membrane topology as predicted by the TMHMM server [140], with the probability of cytosol, membrane or luminal location in blue, red or pink, respectively.

3.3.4 A secretory signal peptide is predicted for DHCR24

DHCR24 contains a putative TMD at its extreme N-terminus (*; Figures 3.3-3.4), which is usually the location of a secretory signal sequence Signal sequences allow translocation of a protein to, or across the ER membrane, which is cleaved for targeting beyond the ER membrane (signal peptide, SP), or retained within the membrane for ER membrane localisation (signal anchor, SA) [153]. Both SPs and SAs contain a stretch of hydrophobic residues, and therefore can be incorrectly predicted as TMDs by many prediction algorithms [154]. Most TMD prediction programs, such as TMHMM cannot discriminate N-terminal TMDs, SPs and SAs [154]. Prediction algorithms, such as SignalP, use a neural network to distinguish N-terminal TMDs from SPs based on a combined scoring system for each position at the N-terminus [155]. The most N-terminal putative TMD for DHCR24 (*; Figures 3.3-3.4) was predicted to be a SP (Figure 3.5), with the cleavage site at 22-23 based on all scoring systems: predicted cleavage site (C-score), signal peptide length (S-score), and the combined cleavage site score (Y-score; a combination of the C-score and the slope of the S-score), which is a better cleavage site prediction than the raw C-score alone. To test whether the putative SP is required for ER membrane localisation, truncated DHCR24 (Δ23 DHCR24) was transfected into CHO-7 cells, and the cellular localisation examined. Through cell fractionation, Δ23 DHCR24 was smaller than WT DHCR24 and, like WT DHCR24, Δ23 DHCR24 localised primarily to the membrane fraction (data not shown). This demonstrates that the putative SP is not cleaved off in WT DHCR24, and is not necessary for membrane retention.

Figure 3.5. DHCR24 contains a putative signal sequence. Signal sequence prediction by SignalP v4.1, with the output given for the first 70 residues [155]. C-score, raw cleavage site score; S-score, signal peptide score; Y-score, combined cleavage site score.

3.3.5 Putative TMDs of low probability are eliminated due to location

The discounted SP and putative TMD lie in one of the least conserved regions of DHCR24. The C-terminal region, however, is highly conserved and contains multiple protein binding/recognition sites for FAD [44, 76, 136], caspase 3 [52], p53 and Mdm2 [58] (Figure 3.1 and Appendix 8.2). The three predicted TMDs that did not have negative ΔGmi values all reside close to functional domains of DHCR24 (136-156, FAD; 210-230, Mdm2; 349-369, p53) that by definition should be cytosolic. Furthermore, a single TMD model is fitting, as it would allow the C-terminus to reside in the cytoplasm, making the multiple binding sites in DHCR24 accessible to soluble binding partners.

3.3.6 Membrane orientation of the N- and C- termini of DHCR24

To determine the localisation of the N- and C-termini of DHCR24 with respect to the ER membrane, CHO-7 cells were transfected with DHCR24 with a V5 epitope tag at either the N-terminus (V5-DHCR24) or C-terminus (DHCR24-V5), and co-transfected with Insig-1, an integral membrane protein with a myc epitope on the cytoplasmic

C-terminus (Insig-myc) [137] (Figure 3.6 B). Intact ER membrane vesicles were isolated and treated with trypsin in the presence or absence of Triton X-100, subjected to SDS-PAGE and immunoblotted for V5 (Figure 3.6 A). As controls, samples were also immunoblotted with myc for Insig-1, and R139 for Scap, which recognises a luminal loop (aa 540-707) in hamster Scap [156] (Figure 3.6 B). Therefore, when trypsin was added to microsomes, the myc epitope on Insig-1 was digested, and the R139 epitope of Scap remained intact, as it is located within the lumen (Figure 3.6 A and C, lanes 1-4). Likewise, DHCR24-V5 was also digested. However, V5-DHCR24 remained stable (Figure 3.6 A, lanes 1-4). When Triton X-100 was added to partially solubilise the membrane, allowing trypsin access to luminal peptides, the R139 epitope of Scap was digested. However, V5-DHCR24 remained relatively stable (Figure 3.6 A, lanes 5-8), indicating that the N-terminus was protected from trypsin digestion by the membrane. Although there is no obvious change in the size of the V5-DHCR24 band with trypsin and Triton X-100 treatment, the band intensity slightly decreased (Figure 3.6 A, lanes 5-8), and a small fragment was discernible with increasing amounts of Triton X-100 (* ~11 kDa, Figure 3.7, lanes 6-9). This smaller band was not visible in non-treated conditions (lanes 1-4) or when trypsin was inactivated by SDS (lane 10). Furthermore, the band appeared with or without Triton X-100 treatment, indicating a partially accessible cytoplasmic loop following a membrane embedded N-terminus. The calculated molecular weight of this digested fragment indicates that the loop is located between aa 70-100, which contains six possible arginines/lysines (Figure 3.1) that could be cleaved by trypsin (dotted red line; Figure 3.7 B). Altogether, this data is indicative of a cytoplasmic C-terminus and a non-accessible N-terminus, which is either embedded within the membrane, or luminal but inaccessible, due to strong interactions with the ER membrane (Figure 3.7 B).

Figure 3.6. Membrane orientation of the N- and C-termini of DHCR24 as determined by trypsin proteolysis. A, CHO-7 cells were transfected with 4 μg of DHCR24-V5 or DHCR24-V5, and co-transfected with 1 μg Insig-myc in a 10 cm dish for 24 h. Cell lysate was fractionated and membranes were isolated and digested with the indicated amount of trypsin in the absence or presence of Triton X-100 as indicated. Membranes were separated by 10% SDS-PAGE and immunoblotted with antibodies against the V5 epitope for DHCR24, myc epitope for Insig-1, and R139 epitope for Scap. Data from at least n=2 experiments. B, Schematics of V5-DHCR24, DHCR24-V5, Insig-myc and Scap with the membrane orientation of epitopes provided.

Figure 3.7. The N-terminus of DHCR24 is strongly membrane associated, followed by a luminal loop. A, CHO-7 cells were transfected with 4 μg V5-DHCR24 and 1 μg Insig-myc in a 10 cm dish for 24 h. Cell lysate was fractionated and membranes were isolated and treated with Triton X-100 or 1% SDS in the absence or presence of trypsin as indicated. Membranes were run on a 10% SDS-PAGE and immunoblotted with antibodies against the V5 epitope for DHCR24. Data from at least n=2 experiments. B, A schematic of V5-DHCR24 with the membrane orientation of the V5 epitope and trypsin cleavage site (dotted red line) indicated in relation to the putative TMD.

3.3.7 The putative TMD is not essential for DHCR24 membrane association

To determine if the hydrophobic N-terminus containing the putative TMD of DHCR24 is required for ER membrane attachment, truncated DHCR24 (Δ56; Figure 3.8 A) was transfected into CHO-7 cells, and the cellular localisation examined. DHCR24-V5 (WT), like Insig-myc, localised only to membranes (Figure 3.8 B). The truncation lacking the putative TMD, Δ56 DHCR24-V5, also localised specifically to membranes (Figure 3.8 B). Therefore, contrary to the predictions in Figures 3.3 and 3.4, these very hydrophobic regions are not necessary for DHCR24 membrane association.

Figure 3.8. Membrane association of DHCR24 and Δ56 DHCR24. A, The schematics of DHCR24-V5 and Δ56 DHCR24-V5, given in relation to the putative TMD, and Insig-myc. B, CHO-7 cells were transfected with either 8 μg DHCR24-V5 (WT) or Δ56 DHCR24-V5 (Δ56), and co-transfected with 2 μg Insig-myc in a 14.5 cm dish for 24 h. Cell lysate was fractionated, and the nuclear (N), membrane (M), and cytoplasmic (C) fractions were separated by 7.5% SDS-PAGE and immunoblotted with antibodies against V5 (DHCR24) and myc (Insig-1). Data from n=2 experiments.

3.3.8 DHCR24 associates strongly with membranes

To determine if DHCR24 is an integral membrane protein, differential solubilisation was performed on membranes to determine the manner in which DHCR24 associates with it. The ability of buffers (detergent, high pH, high salt) to dissociate DHCR24 from the membrane (M) into the cytoplasm (C) indicates the strength of its association with the membrane (ionic or hydrophobic). When exposed to aqueous buffer, DHCR24 was located in the membrane fraction, but was released into the cytoplasmic fraction upon treatment with 1% (w/v)

SDS (Figure 3.9 A). Treatment with 0.1 M Na2CO3 and 1 M NaCl, reagents known to disrupt peripheral associations by altering pH and electrostatic interactions respectively,

were not able to solubilise DHCR24, as it remained in the membrane fraction (Figure 3.9 A). For all treatments, Insig-1 displayed similar results, indicating DHCR24 is associated with membranes, and that it is an integral membrane protein (Figure 3.9 A). Similarly, ΔDHCR24-V5 behaved like an integral membrane protein (Figure 3.9 A). Although Δ56 DHCR24-V5 remained in the membrane fraction after treatment with 0.1

M Na2CO3, the presence of a weaker signal in the cytoplasmic fraction indicated that Δ56 DHCR24-V5 might be solubilised with this treatment. However, performing the treatment and centrifugation process a second time showed that Δ56 DHCR24-V5 remained membrane associated (Figure 3.9 B), suggesting that regions beyond the first 56 residues are integral for strong membrane association.

Figure 3.9. DHCR24 associates strongly with the membrane. A and B, CHO-7 cells were transfected with either 8 μg DHCR24-V5 or Δ56 DHCR24-V5, and co-transfected with 2 μg Insig-myc in a 14.5 cm dish for 24 h. Cell lysate was fractionated and membranes were isolated and resuspended in 1% (w/v) SDS

(strong detergent), Buffer C (aqueous buffer), 0.1 M Na2CO3 pH 11.5 (high pH) or 1 M NaCl (high salt), and rotated for 30 min at 4 °C, then ultracentrifuged at 100,000 × g and the supernatant, representing cytoplasm (C) and pellet, representing membrane (M) were collected. B, The process was repeated using the supernatant for the high pH treatment (C1), with the resulting 100,000 × g supernatant designated C2. Cytoplasmic and membrane fractions were separated by 7.5% SDS-PAGE and immunoblotted with antibodies against V5 and myc. Data from at least n=3 experiments.

3.3.9 Other candidate TMDs are not essential for DHCR24 membrane association

Having established that the most likely TMD (**; Figure 3.3) was not essential for membrane localisation, we next examined the other predicted TMDs. Based on the

ΔGmi values, and the TMHMM probability score (Figures 3.3-3.4), the two next best TMD candidates were examined (#, 136-156; ##, 210-230). Truncated DHCR24 (Δ160, Δ240; Figure 3.10 A) were transfected into CHO-7 cells, and the cellular localisation examined (Fig. 3.10 B). Similar to WT DHCR24-V5, both truncations were found only in the membrane fraction (Fig. 3.10 B), demonstrating that the first 240 residues are also not necessary for membrane attachment.

Figure 3.10. Membrane association of DHCR24 N-terminal truncations. A, The schematics of DHCR24-V5, Δ160 DHCR24-V5 and Δ240 DHCR24-V5 shown in relation to the putative TMD and other lower scoring putative TMDs (#,##, ###), and Insig-myc. B, CHO-7 cells were transfected with either 4 μg DHCR24-V5 (WT), Δ160 DHCR24-V5 (Δ160) or Δ240 DHCR24-V5 (Δ240), and co-transfected with 1 μg Insig-myc in a 10 cm dish for 24 h. Cell lysate was fractionated, and the nuclear (N), membrane (M), and cytoplasmic (C) fractions were separated by 10% SDS-PAGE and immunoblotted with antibodies against V5 (DHCR24) and myc (Insig-1). Data from n=2 experiments. N not shown for WT due to technical difficulties.

3.4 Discussion

To further elucidate the membrane topology of DHCR24, we predicted how it interacts with the ER membrane through the use of numerous membrane insertion algorithms, as well as considering the current literature on DHCR24 topology. We then characterised this interaction using biochemical methods. To create a DHCR24 topology model, we first characterised DHCR24 based on its hydrophopathy using the method of Kyte and Doolittle [139]. Two regions of peak hydrophobicity were identified as candidates for TMDs (Figure 3.3 A), which were validated by numerous TMD prediction algorithms (ΔGmi; Figure 3.3 B) and programs (TOPCONS and TMHMM; Figure 3.4). DHCR24 was predicted to contain an SP (Figure 3.5), which generally targets proteins to or across the ER membrane, and this was located at the first putative TMD. Therefore, the final predicted DHCR24 topology model tested included a SP followed by one TMD. The DHCR24 homolog in A. thaliana, DWF1, has a similar predicted structure, based on hydropathy predictions: the N-terminal domain as a potential membrane anchor, and a cytoplasmic C-terminus [74]. The efficacy of a DHCR24 SP was tested first by the creation of a truncation construct, which did not contain the putative SP. This was smaller than its WT DHCR24 counterpart, and still associated with the membranes. Secondly, the N-terminal V5 tagged DHCR24 construct (V5-DHCR24) was the same size as C-terminal V5-tagged DHCR24 (DHCR24-V5), indicating no cleavage of the N-terminus, which would be predicted for a true SP. These data show that the N-terminus is not cleaved, nor is it required for membrane targeting, demonstrating that DHCR24 does not contain an N-terminal SP. This is similar to cytochrome P450s, which is also an integral ER membrane protein that does not require an N-terminal SP for ER targeting, but contains complex N- and C-terminal retention signals [157]. Upon deletion of the N-terminus (including the putative TMD; Δ56), DHCR24 still associated with the membrane (Figure 3.8), and could not be solubilised by various treatments (Figure 3.9). Again, this is similar to cytochrome P450, which contains a hydrophobic N-terminus predicted to associate with the ER membrane, which retained strong membrane attachment after truncation experiments and differential solubilisation [158, 159]. Further N-terminal truncations in DHCR24 that deleted lower scoring TMD candidates also associated exclusively with the membrane (Figure 3.10). These data

discounts the findings of Lu et al. [81]: that the N-terminus is essential for membrane attachment, and that deletion of the N-terminus (Δ58) resulted in altered cellular localisation of DHCR24. In contrast, our findings demonstrate that the membrane association of DHCR24 is not reliant on the putative TMD at the N-terminus, but contains membrane associated region(s) beyond the hydrophobic N-terminus. We also elucidated the membrane orientation of the N- and C-termini of DHCR24, for which the TMD prediction programs gave different results (Figure 3.3). Using protease protection assays, we determined the C-terminus was cytoplasmic, due to its accessibility to trypsin. The N-terminus, however, was strongly protected, even when trypsin was accessible to the lumen by partial solubilisation of the membrane (Figure 3.6). This suggests that the N-terminus is “hidden” from trypsin; either due to being embedded within the ER membrane, which would confirm (in part) Pedretti et al.’s “peduncle” model [44], or due to peripheral interactions with the ER membrane. Both of these possible models disagree with the findings by Lu et al. [81], that the N-terminus is luminal, which is most likely due to the difference in epitopes used in the protease protection assay [81]. As mentioned, Lu et al. [81] used a large fusion protein (~28 kDa), whereas we used a small V5 epitope (~1 kDa). Testing the membrane association of a C-terminal truncation using a differential solubilisation assay would determine the type of interaction the N-terminus has with the membrane (integral or peripheral). Lastly, the identification of a digested N-terminal ~11 kDa fragment suggests that the membrane associated N-terminus is followed by a sterically-hindered cytosolic loop at aa ~70-100 (Figure 3.7). Overall, our findings demonstrate that DHCR24 has a strong affinity with the ER membrane, due to multiple membrane associated regions, including beyond the N-terminus (Δ240; Figure 3.10). The extreme hydrophobic nature of the N-terminus suggests that it is also membrane associated, but this needs to be confirmed experimentally by C-terminal truncations. We hypothesise that there are multiple membrane associated regions throughout the protein or re-entrant loops that only associate with one leaflet of the membrane. Re-entrant loops are typically rich in small residues such as glycines and alanines [160], and one of the lower scoring putative TMDs (#, Figure 3.3) is rich in these amino acids, and therefore a prime candidate for a re-entrant loop. Our proposed model may contain classical TMD(s), as although this study ruled out the most likely putative TMD, one low scoring putative TMDs was

untested (###, Figure 3.3). As this region is beyond the first 240 residues, it is likely to explain the strong membrane localisation of Δ240 DHCR24 (Figure 3.10). Our data strongly suggest that DHCR24 relocation to other cellular compartments, proposed to occur in response to stress signals, is highly unlikely [58]. This conclusion should be re-examined in light of our results. Altogether, these findings are at odds with prediction algorithms and published topology models (Figure 3.11, [44, 81]). This chapter demonstrates that DHCR24 does not agree with either published topology models; being neither monotopic, like Pedretti et al. suggest, nor bitopic, as Lu et al. propose, but polytopic, with multiple hydrophobic domains passing through and/or anchoring DHCR24 to the hydrophobic core of the lipid bilayer of the ER. To create a hypothetical membrane topology model of DHCR24, we have also utilised in addition to our experimental findings our knowledge of known posttranslational modifications on DHCR24 [161]. Phosphorylation and ubiquitination sites must be accessible to cytosolic kinases and ubiquitin ligases respectively, and therefore we have used these data to refine our topology model of DHCR24 (Figure 3.12). To the best of our knowledge, we are the first to propose that the expanding databases of posttranslational modifications provide an unlikely but invaluable resource for membrane topology mapping, including for other atypical cholesterogenic membrane proteins [17]. To refine our membrane topology model of DHCR24 (Figure 5.12), further experimentation is required to precisely define membrane, cytoplasmic and luminal regions, including whether the final putative membrane region (###) is bona fide.

Figure 3.11. Comparison of current DHCR24 membrane topology models. The predicted and published models for DHCR24 topology are presented with features indicated, compared to our hypothesised DHCR24 membrane topology model.

Figure 3.12. Hypothetical membrane topology model of DHCR24 This model integrates the known PTM sites to reﬁne our working model of DHCR24 membrane topology. As the focus of this work has been on the N-terminus, this hypothetical model does not attempt to encapsulate the membrane topology of the C-terminus, which is likely to snorkel in and out of the membrane similar to the N-terminus, and is therefore protected from trypsin digestion (Figure 3.6 A, lanes 1–4 versus 6–9). Cofactor-binding sites marked: FAD (111–203), p53 (358–425), Mdm2 (203–215). Caspage cleavage sites (122–127, 383–387).

Chapter 4

Sterol-mediated transcriptional regulation of DHCR24 qRT-PCR was performed by Laura Sharpe. Promoter and cooperativity assays were performed by Eser Zerenturk and Laura Sharpe. EMSAs were performed by Eser Zerenturk. SRE prediction method was created by Laura Sharpe.

The following work has been published in:

Zerenturk EJ*, Sharpe LJ*, Brown AJ (2012) Sterols regulate 3β-hydroxysterol Δ24-reductase (DHCR24) via dual sterol regulatory elements: Cooperative induction of key enzymes in lipid synthesis by Sterol Regulatory Element Binding Proteins. Biochimica et Biophysica acta, 1821 (10): 1350-1360. * Equal first author.

4 STEROL-MEDIATED TRANSCRIPTIONAL REGULATION OF DHCR24

4.1 Introduction

DHCR24 is crucial in multiple processes beyond cholesterol synthesis (Section 1.6.1), and has therefore been implicated in numerous diseases (Section 1.6.2). However, comparatively little is known about how this important, multifunctional enzyme is regulated. This chapter explores the transcriptional regulation of DHCR24. Gene expression of DHCR24 is responsive to numerous factors, such as hormones and epigenetic regulation (Section 1.8.1). The Sterol Regulatory Element Binding Proteins (SREBPs), which control the expression of most cholesterol synthetic genes [162], have also been implicated in the transcriptional regulation of DHCR24 [163-165]. SREBPs are major regulators of cholesterol and fatty acid homeostasis [8], and three isoforms exist in mammals: SREBP-1a, -1c and -2 (see Section 1.5.1 and Figure 1.4 for more detail). SREBPs activate genes by heterotypic activation, by first binding as a homodimer to the cis-acting sterol regulatory elements (SREs) in the promoters of its target genes, which resemble the classical (LDLR) SRE sequence aTCACcCCAC [166, 167]. Binding of SREBPs induces the recruitment of specificity protein 1 (Sp1) and nuclear factor Y (NF-Y) cofactors to nearby sites, which bind directly to SREBP, and synergistically activate target gene expression [24]. SREBPs can also bind and activate palindromic E-boxes, in a sterol-independent manner [121]. When stimulated with the growth hormone PDGF, human foreskin fibroblast cells activated SREBP-1c and -2, and induced the expression of known SREBP targets, as well as DHCR24 [163]. DHCR24 expression increased in these cells upon transfection of an adenovirus encoding SREBP-1c [165]. Also, the involvement of SREBP-2 in DHCR24 expression has been indicated in livers from SREBP-2 transgenic mice [164]. Furthermore, a statin (simvastatin), which activates SREBP, has been shown to up-regulate DHCR24 in human neuronal cells and murine brains [168]. Considering that most other cholesterol biosynthetic enzymes are controlled by SREBP-2 [162], it is reasonable that DHCR24 may be similarly regulated, although to no SREs have previously been identified in the DHCR24 promoter.

The Liver X Receptor (LXR) has also been implicated in the regulation of DHCR24 [169]. LXR is a nuclear receptor activated by oxysterol ligands that up-regulates genes involved in sterol efflux, such as the ATP-Binding Cassette transporter A1 (ABCA1) [170-172]. Thus, LXR decreases cell cholesterol content. A genome wide screen identified LXR occupancy within the second intron of DHCR24, which conferred activity in an LXR reporter gene assay in HEK293 and HepG2 cells [169]. However, expression studies using LXRβ -/- mice showed that LXR regulation of DHCR24 was not ubiquitous, occurring in skin, but not brain tissue [169]. Based on these studies, both SREBP and LXR may transcriptionally up-regulate DHCR24. However, as both tend to have differing effects on cholesterol homeostasis, with SREBP-2 generally increasing cell cholesterol status, and LXR decreasing it, this is apparently contradictory. Furthermore, other cholesterol biosynthetic enzymes are mainly under the control of SREBP-2, rather than LXR [162]. In this chapter, we set out to investigate the relative contribution of SREBP-2 and LXR on DHCR24 gene expression, and characterise the sterol-mediated regulation of DHCR24 through the identification of functional binding sites within DHCR24.

4.2 Materials and Methods

4.2.1 Materials

An antibody to SREBP-2 was prepared in-house: IgG-7D4, a mouse monoclonal antibody against hamster SREBP-2 (aa 32-250 [173].

4.2.2 Cell culture media

Medium Composition A DMEM/F-12, supplemented with 5% (v/v) LPDS D RPMI 1640 medium supplemented with 10% (v/v) FCS F RPMI 1640 medium supplemented with 10% (v/v) FCLPDS

4.2.3 Plasmids

The promoter of the human DHCR24 gene (~1.2 kb upstream from the ATG start site) was prepared by PCR amplification from human fibroblast genomic DNA and cloned into pGEM-T-easy. This fragment was then subcloned into the Firefly luciferase pGL3-basic vector, with truncations and mutations made (Table 4.1). For details on ownership or construction of plasmids, see Sections 2.3.

Table 4.1. Luciferase plasmids used in this chapter. Various human promoter sequences fused to a Firefly luciferase gene, or unless stated, a Renilla luciferase gene. Numbering is relative to the ATG start codon. Plasmid Promoter Description Usage pGL3-basic - Figures 4.3, 4.5, empty vector 4.10 -1200luc DHCR24 promoter truncations Figure 4.5 -900luc -650luc -300luc Figures 4.3, 4.5, 4.10, 4.11 -100luc Figure 4.5 -275mut -300luc with SRE mutations Figure 4.5 -254mut -244mut -240mut -196mut -160mut -119mut -109mut -196/-160mut -196LDLR -300luc with the SRE swapped with the LDLR SRE -160LDLR -160↔-196inv -300luc with the -160 invSRE and -196 Figure 4.10 invSRE swapped 3×(-196inv) 3 × tandem -196 invSRE 3×(-160) 3 × tandem -160 SRE -225mut -300luc with NF-Y mutations -174mut -131mut -176mut -300luc Sp1 mutation

SREs 5del -300luc with progressive deletions between -160 invSRE and -196 invSRE SREs 10del (5, 10, 15, 35 bp) SREs 15del SREs 35del LDLRluc The human LDLR promoter Figures 4.5, 4.11 LDLR-mutSRE LDLRluc with the SRE mutated Figure 4.5 FDFT1luc The human FDFT1 promoter Figure 4.11 phRL-PBGD A PBGD (-460/+5) driven expression Figures 4.3, 4.5, plasmid of Renilla luciferase. 4.10, 4.11

Table 4.2. SREBP-2 expression plasmids used in this chapter. Plasmid Description Usage A CMV driven expression plasmid encoding the nSREBP-2 mature N-terminal form (aa 14-480) of human Figure 4.3 SREBP-2 (nSREBP-2) A thymidine kinase driven expression plasmid TK-nSREBP-2 Figure 4.11 encoding human nSREBP-2 nSREBP-2 plasmid with the tyrosine at aa 342 nSREBP-2(Mut)] mutated to an arginine residue to prevent DNA Figure 4.3 binding

4.2.4 Luciferase Assay

HeLaT cells were grown in triplicate in 24-well plates and transiently transfected with a Firefly luciferase plasmid (250 ng, Table 4.1) and phRL-PBGD Renilla luciferase plasmid (25 ng, Table 4.1), with or without the expression plasmids nSREBP-2 or nSREBP-2(Mut) (25 ng, Table 4.2), using TransIT-2020 for 24 h. For cooperativity studies, TK-nSREBP-2 (0 – 16 ng, Table 4.2) was used, with TK-driven empty vector up to 16 ng as required. After transfection, cells were refreshed with or treated in lipoprotein deficient media (Medium F) for 24 h, then lysed and assayed for Firefly and Renilla luciferase activities using the Dual Luciferase Reporter Assay System. Relative luciferase activity was determined as a ratio of Firefly to Renilla luciferase activity for each individual sample, and normalised as stated in figure legends.

4.2.5 Electrophoretic mobility shift assay

HeLaT cells were plated in 10 cm dishes, and transfected with 8 μg nSREBP-2 (Table 4.2) per dish using TransIT-2020 for 24 h, and refreshed with Medium D for 24 h. SRD-1, CHO-7 and SRD-13A cells were plated in 10 cm dishes, refreshed with Medium A for 24 h. All cells were and treated with 10 μM MG132 in the last 4 h, to minimise SREBP-2 protein degradation. Cells were harvested on ice; washed with cold PBS, scraped, and pelleted by centrifugation at 1,000 × g for 5 min at 4 °C. Cell pellets were resuspended in 500 μL of Resuspension Buffer [10 mM HEPES-KOH (pH 7.4),

10 mM KCl, 1.5 mM MgCl2, 5 mM EDTA, 5 mM EGTA, 250 mM sucrose, 5 mM DTT, supplemented with 2% (v/v) protease inhibitor mixture]. The cell suspension was passed through a 21 gauge needle 30 times and centrifuged 1,000 × g for 7 min at 4 °C. The pellet was resuspended in 100 μL of Nuclear Extract Buffer [20 mM HEPES-KOH

(pH 7.6), 25% (v/v) glycerol, 0.42 M NaCl, 1.5 mM MgCl2, 1 mM EDTA, 1 mM EGTA, supplemented with 2% (v/v) protease inhibitor mixture] and rotated at 4 °C overnight, and supernatant was isolated after centrifugation at 30,000 × g for 30 min at 4 °C. Protein concentration was measured by the BCA assay For the Electrophoretic mobility shift assay (EMSA), 20 μg nuclear extract was incubated with Binding Buffer [10 mM Tris-HCl (pH 7.6), 50 mM KCl, 5 mM MgCl2, 1 mM DTT, 1 mM EDTA, 6% (v/v) glycerol, 0.05% (v/v) nonidet P-40, and 0.05 mg/mL (or 0.025 mg/mL when using HeLaT cell lysate) salmon sperm DNA] in a 20 μL reaction for 5 min at 25 °C. For the competition assay, 100 pmol unlabelled probe (Table 4.3) was added, and the mixture was incubated for 5 min at 25 °C. 1 pmol Cy5 labelled probe (Table 4.3), double-stranded with the corresponding reverse complements) was then added, and the mixture was incubated for 30 min at 15 °C. For the supershift assay, 7D4 antibody (1-2 μL) was added after incubation with 1 pmol Cy5 labelled probe. 5 × nPAGE loading buffer was added to mixtures, and a separate sample containing 5 × nPAGE loading buffer with 5% β-mercaptoethanol was prepared. All samples were then subjected to 6% (w/v) native PAGE (Section 2.1.4) in EMSA Running Buffer (Sections 2.1.3), which was stopped once the dye in 5 × nPAGE loading buffer with 5% β-mercaptoethanol sample had reached the bottom of the gel. Samples were visualised using an FLA-5100 fluorescence scanner with a R665 LPR filter at 635 nm excitation.

Table 4.3. EMSA probes. Both Cy5 and unlabelled DNA probes, with the SRE underlined. EMSA probes Sequence (5’ 3’) LDLR CATTTGAAAATCACCCCACTGCAAACTCC -196 invSRE CTGCGGGTTCCTGGTCCGATCCCCGGCGCG -160 SRE CCGCCCGGGTCTCGGCCCACCGAACCTCGG

4.2.6 Quantifying cooperativity by calculating the Hill Slope

To determine the level of cooperativity between SREBPs, the slope/steepness of the dose response curve generated from the cooperativity assays (Section 4.2.4), which is quantified by the Hill Slope, was calculating using Prism 6. A Hill Slope of 1 indicates 1 binding site, and therefore no cooperativity; a Hill Slope >1 is indicative of cooperativity, with the calculated slope correlating with the number of binding sites.

4.3 Results

4.3.1 DHCR24 expression is regulated by sterols in human cell-lines

DHCR24 has been observed to respond as both an SREBP- [163-165, 169] and LXR-target gene [169], which is paradoxical considering that these transcription factors have opposing effects on cell cholesterol status. We investigated this apparent dichotomy by comparing the mRNA expression pattern of DHCR24 with known SREBP- and LXR- target genes across a variety of conditions and cell types. Beginning with the commonly-used HeLaT cell-line, we measured, in addition to DHCR24, the mRNA levels for several genes involved in lipid metabolism. These were two LXR targets: ATP-binding cassette transporter A1 (ABCA1) [174], ABCG1 [175]; two SREBP-1c targets: fatty acid synthase (FASN) [176] and acetyl-CoA carboxylase (ACC) [162, 177], which are also under the direct and indirect control of LXR, respectively [178-180]; and two SREBP-2 targets: HMG-CoA reductase (HMGCR) [162, 181] and LDL receptor (LDLR) [166]. We manipulated gene expression with a statin (compactin), which increases SREBP-2 activation and decreases LXR activation. Conversely, the oxysterols 25HC and 24,25EC stimulate LXR-mediated expression [182], but inhibit SREBP-2 activation [183, 184]. Lastly, treatment with the synthetic LXR ligand TO-901317 increases LXR-mediated expression independently of SREBP-2 [185]. Thus, a comparison of the mRNA expression patterns should indicate the relative contribution of SREBP and/or LXR on DHCR24 gene expression. The gene expression pattern for DHCR24 did not resemble that of ABCA1 and ABCG1, which both increase with 25HC, 24,25EC, and TO-901317 treatment, but decrease with statin (Figure 4.1). Nor did it resemble FASN or ACC, which exhibit the influence of LXR by increasing with TO-901317, as well as SREBP effects, increasing with statin in the majority of cell lines, although most did not reach statistical significance, and decreasing with 25HC and 24,25EC. However, DHCR24 closely resembled LDLR and HMGCR, which show SREBP-2 regulation and no influence of LXR: increasing with statin, decreasing with 25HC and 24,25EC, and no significant change with TO-901317. This pattern was also reflected in cell-lines of neuronal [BE(2)C] and hepatic (HepG2) origin, consistent with DHCR24 being transcriptionally

regulated by SREBP-2, with little or no contribution of LXR in the cell-lines studied (Figure 4.1).

HeLaT Be(2)C HepG2

150 200 3 1.5 1.5 *

1.0 * 150 1.0 100 2 0.5 ** 0.5 ** ** 100 0.0 0.0 50 1 ABCA1 50 * * 0 0 ** 0

100 15 500 1.5 1.5 1.5 ** ** ** 80 1.0 1.0 400 1.0 ** 10 * 60 0.5 0.5 300 0.5 * ** ** LXR target LXR genes 40 0.0 0.0 * 200 0.0 ** 5 ABCG1 20 100 ** * 0 ** 0 0

6 4 8 * 3 6 4 2 4

FASN 2 1 * 2 * * * ** 0 0 0

2.5 3 5 * 1c target genes

2.0 4 - 2 1.5 * 3

1.0 2 ACC * 1 0.5 1 * SREBP 0.0 0 0

3 * 8 4 * * 6 3 2 4 2 1 Relative mRNA levels Relative mRNA 2 1 HMGCR ** ** ** ** ** ** 0 0 0

2.5 * 4 3 * 2 target genes 2.0 - 3 2 1.5 2 1.0 1 LDLR 1 SREBP 0.5 ** * * ** ** ** 0.0 0 0

2.5 * 5 2.5 * 2.0 4 ** 2.0

1.5 3 1.5 1.0 2 1.0 *

DHCR24 0.5 ** 1 0.5 ** ** ** 0.0 0 0.0 25HC 25HC 25HC Statin Statin Statin 901317 901317 901317 Control Control Control - - - 24,25EC 24,25EC 24,25EC

TO TO TO 85

Figure 4.1. DHCR24 expression upon altered sterol status in human cells. HeLaT, BE(2)C, and HepG2 cells were treated with 5 M compactin (statin), 10 M 25-hydroxycholesterol (25HC), 10 M 24(S),25-epoxycholesterol (24,25EC) or 10 M TO-901317 for 24 h. Total RNA was harvested and mRNA levels for ATP-binding cassette transporter A1 (ABCA1), ABCG1, fatty acid synthase (FASN), acetyl-CoA carboxylase (ACC), HMG-CoA reductase (HMGCR), LDL receptor (LDLR) and DHCR24 were measured using qRT-PCR and normalised to porphobilinogen deaminase (PBGD). mRNA levels are relative to the control condition, which was set to 1. Data are presented as mean + SEM from 3 separate experiments, where each experiment was performed with triplicate cultures. * p≤0.05 and **, p≤0.01, using a paired t-test.

4.3.2 DHCR24 regulation is dependent on SREBP-2 in CHO cells

To further investigate the effect of SREBP-2 on DHCR24 expression, a number of mutant CHO cell-lines with altered SREBP activities were employed. SRD-1 cells overexpress the nuclear form of SREBP-2, without the sterol regulated proteolytic step, resulting in high basal expression of SREBP-2 target genes [186]. On the other hand, SRD-13A cells lack the SREBP cleavage activating protein (Scap), resulting in no proteolytic activation of SREBP [117], and thus have low basal expression of SREBP target genes (see Section 1.5.1 for SREBP processing) [187]. CHO-7 cells were used as a control CHO cell-line that exhibits regulated cholesterol homeostasis [186]. Cells were treated with statin or 25HC, and ABCA1, ACC, and HMGCR mRNA levels were measured as typical genes for LXR, SREBP-1c, and SREBP-2, respectively. Similarly to the human cell-lines in Figure 4.1, DHCR24 displayed up-regulation with statin treatment, and down-regulation with 25HC treatment in wild-type cells (CHO-7) (Figure 4.2). DHCR24 mRNA levels correlated with SREBP-2 activity, being higher in SRD-1 cells, with virtually no expression in SRD-13A cells relative to CHO-7 cells, irrespective of treatments (Figure 4.2). This expression profile closely resembles that of the SREBP-2 target gene HMGCR, and not the SREBP-1c target ACC or the LXR target ABCA1. The absence of DHCR24 regulation in these mutant cell-lines confirms that SREBP-2 regulates DHCR24 expression.

Figure 4.2. Regulation of DHCR24 expression with altered SREBP-2 levels in CHO cells. SRD-13A, CHO-7 and SRD-1 cells were treated with 5 M compactin (statin) or 10 M 25-hydroxycholesterol (25HC) for 24 h. Total RNA was harvested and mRNA levels of ABCA1, ACC, HMGCR, and DHCR24 were measured using qRT-PCR and normalised to PBGD. mRNA levels are relative to the control condition in CHO-7 cells, which was set to 1. Data are presented as mean + SEM from n=3 experiments, where each experiment was performed with triplicate cultures. 87

4.3.3 Transcriptional regulation of DHCR24 promoter-reporter gene by co-expressed SREBP-2

The SREBP-2 responsiveness of DHCR24 was investigated at the promoter level using a cell-based luciferase reporter assay. Reporter constructs were created, containing progressive deletions of the promoter upstream of the ATG translational start codon of human DHCR24. These constructs were then co-transfected with expression plasmids encoding the mature N-terminal form of SREBP-2 (nSREBP-2) or a mutated version that cannot bind to SREs [nSREBP-2(Mut)]. The -1200, -900, -650, and -300 fragments of the DHCR24 promoter were SREBP-2 responsive, with relative luciferase activity increasing with nSREBP-2, and decreasing with nSREBP-2(Mut) (Figure 4.3 A), indicating the presence of functional SREs. As an indication of relative SREBP-2 responsiveness, the ratio of luciferase activities between nSREBP-2 and nSREBP-2(Mut) was used (Figure 4.3 B). The ratio of the -100 fragment of the DHCR24 promoter (-100luc) was lower than all other fragments and similar to the pGL3-basic empty luciferase vector control, indicating that the location of the sterol responsive region was between -300/-100 of the DHCR24 promoter.

Figure 4.3. Co-expression of SREBP-2 up-regulates a DHCR24 promoter reporter. A, HeLaT cells were transfected with 250 ng Firefly plasmid: pGL3-basic empty vector, -1200luc, -900luc, -650luc, -300luc or -100luc in the presence of 25 ng phRL-PBGD Renilla luciferase construct. Cells were co-transfected with 25 ng either wild-type N-terminal SREBP-2 (nSREBP-2, ■) or mutated N-terminal SREBP-2 [nSREBP-2(Mut), □] expression plasmid for 24 h, and treated with fresh media for a further 24 h. Data are presented as mean + SEM from at least n=3 experiments and are relative to the -1200luc and nSREBP-2 condition, which has been set to 1. B, The ratio of nSREBP-2 to nSREBP-2(Mut) relative luciferase activities of each luciferase construct.

4.3.4 Identification of putative regulatory elements responsible for SREBP-2-mediated transcription of DHCR24

Next we sought to determine the location of the SREBP-2 response elements (SREs) within the DHCR24 promoter. Three putative SREs were predicted far from the ATG start codon, at -4281, -2387 and -1177, using Matinspector [188], which upon mutational analysis did not indicate functionality (data not shown), consistent with our truncation studies (Figure 4.3). Furthermore, SREs are not usually located far from the

translational start site, but located within the proximal promoter of target genes (Table 4.4). The statistical basis for the SREBP-2 predictions in Matinspector are very low, with the SREBP-2 specific prediction matrix based on 8 known SREs; therefore, we predicted potential SREs in the region -300/-100 using the Transcriptional Regulatory Element Database (TRED) [189] sequence matrix search page, which enables predictions using our in-house positional weight matrix (Figure 4.4 A) based on twenty known human SRE sequences (Table 4.4). Putative SREs in -300/-100 that scored at least as highly as the lowest-scoring known SRE using the same prediction method (2.44), were chosen for further analysis, as was one that scored highly when considering a more SREBP-2-specific matrix (SREs predominantly responsive to SREBP-2, cut-off 2.2) (Figure 4.4 B and Table 4.5).

Figure 4.4. Sterol response element motif and sequence of the promoter of human DHCR24. A, A matrix based on known sterol response elements (SREs) was produced by counting the number of SREs containing a particular base in each position, and presented using WebLogo [190]. B, The 320 nucleotide sequence upstream of the ATG translation start codon. Predicted SREs are underlined and labelled green, and their position and orientation stated (inv: inverse). Matinspector-predicted NF-Y and Sp1 sites, and a recently-published Sp1 site [191] are overlined and labelled blue and red, respectively. In subsequent experiments, the NF-Ys and SREs in bold type were found to be functional.

Table 4.4. Characterisation of known SREs. Known SREs sequence and position relative to the ATG translational start site of SREBP gene targets, except for HMGCR, which is relative to the transcriptional start site. SRE sequences were submitted to the Transcriptional Regulatory Element Database (TRED) using the positional weight matrix in Figure 4.4 A and their scores recorded. ABCA7: ATP-binding cassette, sub-family A (ABC1), member 7; ACLY: ATP citrate lyase; APOA2: apolipoprotein A-II; CYP51A1: cytochrome P450, family 51, subfamily A, polypeptide 1; FDFT1: farnesyl-diphosphate farnesyltransferase 1; FDPS: farnesyl diphosphate synthase; HMGCR: HMG-CoA reductase; HMGCS1: 3- hydroxy-3-methylglutaryl-CoA synthase 1 (soluble); LDLR: LDL receptor; LRP1: LDLR-related protein 1; NPC1L1: NPC1 (Niemann-Pick disease, type C1, gene)-like 1; PCSK9: proprotein convertase subtilisin/kexin type 9; PRSS8: protease, serine, 8; PTGS2: prostaglandin-endoperoxide synthase 2 (prostaglandin G/H synthase and cyclooxygenase); SREBF2: sterol regulatory element binding transcription factor 2; STARD4: StAR-related lipid transfer domain containing 4.

TRED Gene Position SRE Sequence Ref. score(s) ABCA7 -189 G CCAC C CTAC 6.63 [192] ACLY -115 C TCAG G CTAG 4.82 [193] APOA2 -61 G TCAC C TGAC 5.93 [194] CYP51A1 -232 A TCAC C TCAG 7.53 [195] -138 C TCAC A CTAG 6.64 FDFT1 [196] -189 A TCAC G CCAG 8.43 FDPS -234 C TCAC A CGAC 7.03 [197] -147 A CCGC A CCAT 4.25 HMGCR [181] -157 C TCTC A CCAC 7.34 -404 G CCAC C TCAC 6.50 HMGCS1 [123] -386 C TCAC A CCAC 8.95 LDLR -65 A TCAC C CCAC 9.97 [198] LRP1 +235 T TCAC C CCAC 8.67 [199] -649 G TCAC C CCTA 5.28 NPC1L1 [200] -25 T CCTC C CCTT 2.44 PCSK9 -337 A TCAC G CCAC 9.19 [201] PRSS8 +107 G TCTG G CCAC 5.52 [202] PTGS2 -422 A TCAG T CCCG 3.56 [203] SREBF2 -128 A TCAC C CCAC 9.97 [204] STARD4 -201 A TCAT T CCAT 4.75 [205] Table 4.5. Putative DHCR24 SREs. The eight SREs predicted within the -300/-100 region of the human DHCR24 promoter, and their scores. If the score calculated was below 2.44 or 2.2 for the SREBP or SREBP-2 specific (excluding HMGCR, ACLY and CYP51A1) matrices respectively, they were not listed.

TRED Score SRE Sequence Putative SRE SREBP SREBP-2 Specific 5’ 3’ -275 SRE - 2.51 G GCAC C CCCG -254 invSRE 2.82 3.32 T TCTC C CCGC

-244 invSRE 2.50 2.75 T CCAC C CTTT -240 invSRE 4.65 3.41 A CCGC T CCAC -196 invSRE 4.46 2.26 A TCGG A CCAG -160 SRE 5.34 3.69 C TCGG C CCAC -119 invSRE 2.63 2.59 G CCTC G CGAT -109 invSRE 2.86 - A TCGC C CGCC

4.3.5 Identification of two SREs in DHCR24

To evaluate which of these putative SREs confers sterol regulation to DHCR24, each of the sites was mutated individually in -300luc, and transfected into HeLaT cells. The previous assay (Figure 4.3) used ectopic SREBP-2 to determine the sterol responsiveness of DHCR24 promoter truncations; however, this assay was not able to detect subtle changes in sterol responsiveness with the SRE mutations. Therefore, we employed a more physiological method to detect changes in sterol responsiveness: endogenous SREBP-2 was manipulated by treating with statin (up-regulation) or sterol (down-regulation). Responsiveness was then determined as the ratio of statin to sterol relative luciferase activities (Figure 4.5). Mutation of -196 invSRE or -160 SRE resulted in the greatest reduction in the sterol responsiveness of -300luc (by 85 and 70%, respectively). The combined mutation of -196 invSRE and -160 SRE lowered the statin to sterol ratio further (almost 100%). Individually mutating each SRE to the known human LDLR SRE enhanced their sterol responsiveness, to a similar level as the LDLR promoter itself (Figure 4.5). This is probably due to the LDLR SRE being a stronger sequence for SREBP-2 binding (a score of 9.97 cf. 4.46 and 5.34 for -196 invSRE and -160 SRE, respectively). Together, these data indicate that both -196 invSRE and -160 SRE are required for the regulation of DHCR24 by sterols.

Figure 4.5. SREBP-2 binds to both -196inv SRE and -160 SRE in the promoter of the human DHCR24 gene. HeLaT cells were transfected with 250 ng Firefly plasmid: pGL3-basic empty vector, -100luc, -300luc, -275mut, -254mut, -244mut, -240mut, -196mut, - 160mut, -119mut, -109mut, -196/-160mut, -196LDLR, -160LDLR, LDLRluc, or LDLR-mutSRE in the presence of 25 ng phRL-PBGD Renilla luciferase construct for 24 h. Cells were then treated with 5 μM compactin (statin) or 10 μM 25-hydroxycholesterol (sterol) for 24 h. Values presented as the ratio of statin to sterol relative luciferase activities of each luciferase construct. Data are presented as mean + SEM from at least n=3 experiments, performed with triplicate cultures.

4.3.6 SREBP-2 binds to DHCR24 SREs

To confirm that SREBP-2 is capable of binding to the putative SREs, electrophoretic mobility shift assays (EMSAs) were performed. Nuclear extract from SRD-1 cells was used as a source of SREBP-2, which produced two major bands when combined with a Cy5 labelled -196 invSRE probe (Figure 4.6, lane 2). The bands were competed out with unlabelled probes (lane 3); however, only the bottom band reappeared upon the addition of an unlabelled probe containing mutated -196 invSRE (lane 4). Furthermore, the addition of an SREBP-2 specific antibody did not produce the expected supershift of either of the bands (lanes 5-6). Therefore, to determine which of these bands represented the true shift band, a salmon sperm DNA dose curve was used. Only the top band remained altered with changes in salmon sperm DNA concentrations, indicating this was the bona fide shift band (Figure 4.7 A).

Figure 4.6. EMSA optimisation. Nuclear extracts from SRD-1 cells were incubated with Cy5 labelled -196 invSRE oligonucleotides as indicated, with or without a 100-fold molar excess of the corresponding unlabelled competitor oligonucleotide, or SREBP-2 antibody (ab, 1-2 µL), and run on an nPAGE, and visualised by fluorescence. *, Two major bands.

Figure 4.7. EMSA optimisation with salmon sperm DNA. Nuclear extracts from A, SRD-1 or B, HeLaT cells over-expressing active human SREBP-2, were incubated with Cy5 labelled -196 invSRE oligonucleotides, with or without a 100-fold molar excess of the corresponding unlabelled competitor oligonucleotide, with 0.025-0.25 mg/mL salmon sperm DNA as indicated. Samples were run on a native PAGE and visualised by fluorescence.

This experiment was repeated using HelaT cells over-expressing active human SREBP-2 containing an HA epitope (Figure 4.7 B). HeLaT nuclear extracts produced a band shift when combined with Cy5 labelled -196 invSRE or -160 SRE probes, which was competed out with unlabelled probes, similar to the positive control LDLR (Figure 4.8 A). However, we were still unable to produce a supershift band, and therefore used nuclear extracts from CHO cell-lines that contain high (CHO-7) or low (SRD-13A) levels of SREBP-2. These extracts were also able to produce a band shift when combined with a Cy5 labelled -196 invSRE probe (Figure 4.8 B). The intensities of the

shift bands correlated with the levels of SREBP-2, indicating the binding of SREBP-2 to the SRE.

Figure 4.8. SREBP-2 binds to both -196 invSRE and -160 SRE in the promoter of the human DHCR24 gene. A, Nuclear extracts from HeLaT cells over-expressing active human SREBP-2 were incubated with Cy5-labelled SRE oligonucleotides as indicated, with or without a 100-fold molar excess of the corresponding unlabelled competitor oligonucleotide. B, Nuclear extracts from SRD-13A or CHO-7 cells were incubated with Cy5-labelled -196 invSRE oligonucleotides. A and B, Sample were run on a native PAGE and visualised by fluorescence. B, Extracts were also subjected to SDS-PAGE and Western blotting to confirm mature SREBP-2 content. Darker exposures show a faint SREBP-2 band in the SRD-13A cells. Loading control shows a representative band from total protein staining on the membrane.

These results demonstrate that human SREBP-2 can specifically bind to the two SRE motifs (-196 invSRE and -160 SRE) in the promoter of DHCR24. Furthermore, both of these SREs are conserved amongst mammals (Figure 4.9), indicating regulation is most likely also conserved.

Figure 4.9. -196 invSRE and -160 SRE in the promoter of the human DHCR24 gene are evolutionarily conserved. Sequences of the DHCR24 promoter from the species indicated were aligned, and the region corresponding to -215/-143 in human is presented.

4.3.7 NF-Y binding sites in the DHCR24 promoter also play a role in its regulation

To determine the role of other transcription factor binding sites in SREBP-2 activation of DHCR24, we searched for NF-Y and Sp1 binding sites in close proximity to the SREs. Three NF-Y (-225 NF-Y, -174 NF-Y, -131 NF-Y) and two Sp1 (-269 invSp1, -176 Sp1) putative binding sites were identified using Matinspector [188], which has a high statistical basis for the prediction of both these sites. One Sp1 binding site (-266 Sp1) was identified from a recent publication [191]. One predicted Sp1 binding site (-269 invSp1) and the recently published site (-266 Sp1) [191], have already been tested by the SRE mutations -275 SRE and -254 invSRE, respectively, and were shown not to affect SREBP-2-mediated DHCR24 expression (Figure 4.5). Thus, we mutated the remaining three NF-Y and one Sp1 site in -300luc, as indicated in Figure 4.4 B. We found that two NF-Y (-225 and -174 inv) but no Sp1 sites were necessary for sterol-regulated activation of the DHCR24 promoter (Figure 4.10 A). -225 NF-Y is located near -196 invSRE, and -174 invNF-Y is located between the two SREs.

4.3.8 The spatial arrangement of the two SREs in the DHCR24 promoter is important

We next determined whether the relative locations of our two SREs was critical for their functionality. First, we constructed a luciferase reporter plasmid driven by three tandem copies of one of the two SREs, and another where their positions in the -300luc had been switched, whilst maintaining their orientation. This indicated that the tandem SREs were weakly responsive above empty luciferase promoter levels, and that 99

interestingly, the switching of the SREs increased the sterol responsiveness of the promoter (Figure 4.10 B). This could be explained by the NF-Y results, in that moving the higher scoring SRE (-160 SRE, Table 4.5) into the proximity of both NF-Y sites may enhance overall sterol responsiveness of the construct. To investigate this further, we created deletion constructs of -300luc, with either the entire section between the SREs deleted (35 bp deletion), or 5, 10, or 15 bp between -196 invSRE and -174 invNF-Y deleted. This indicated that indeed the distance between the SREs plays a large role in determining their relative sterol-responsiveness (Figure 4.10 C). Deleting 5, 10, or 15 bp between our two SREs greatly decreased the sterol-responsiveness of the -300luc. Surprisingly, deleting the entire 35 bp between them restored the sterol response of the construct. Again, this could be explained by the NF-Y sites, as moving the -160 SRE closer to the -225 NF-Y should enable it to help activate both SREs.

100

Figure 4.10. NF-Y sites and spatial arrangements affect sterol-responsiveness of the DHCR24 promoter. HeLaT cells were transfected with 250 ng Firefly plasmid: pGL3-basic empty vector, or A, -300luc, -225mut, -176mut, -174mut, -131mut, B, -300luc, [3×(-196 inv)], [3×(-160)], (-160↔-196 inv), C, -300luc (with 35 bp between the two SREs), or 5, 10, 15 or 35 bp deletions, in the presence of 25 ng phRL-PBGD Renilla luciferase construct for 24 h. Cells were then treated with 5 μM compactin (statin) or 10 μM 25-hydroxycholesterol (sterol) in Medium F for 24 h. Values presented as the ratio of statin to sterol relative luciferase activities of each luciferase construct. Data are presented as mean + SEM from at least n=3 experiments, performed with triplicate cultures.

101

4.3.9 Dual SREs in the DHCR24 promoter result in SREBP-2 homotypic cooperativity

Having determined that the spatial arrangement of the dual SREs was critical, we next investigated the dose response of SREBP-2 activating DHCR24. To do this, we compared the relative luciferase activities of the DHCR24 -300luc with those of the LDLRluc [122], which contains one SRE, and the farnesyl-diphosphate farnesyltransferase 1 (FDFT1)luc [123], with two SREs. Luciferase activities were determined over a range of transfected SREBP-2 concentrations, with endogenous SREBP suppressed by treatment with 25HC (Figure 4.11). The shape of the curve for LDLRluc was linear (Figure 4.11 A), whereas the -300luc and FDFT1luc curves were sigmoidal (Figure 4.11 B). Accordingly, the Hill slope coefficients reflected the number of SREs in each gene promoter (Figure 4.11 C), suggesting that SREBP-2 functions cooperatively to up-regulate genes with dual SREs.

Figure 4.11. SREBP-2 acts cooperatively to activate genes with dual SREs. A, HeLaT cells were transfected with 250 ng Firefly plasmid: DHCR24 -300luc (●), LDLRluc (∆), or FDFT1luc (□), and the indicated amount (0 – 16 ng) active TK-nSREBP-2, in the presence of 25 ng phRL-PBGD Renilla luciferase construct for 24 h. Cells were then treated with 10 µM 25-hydroxycholesterol in Medium F for 24 h to suppress endogenous SREBP-2. For each Firefly luciferase construct, the relative luciferase value at 16 ng TK-nSREBP-2 was set to 1, and 0 ng was set to 0. Data are presented as mean + SEM from 16 to 24 data points from at least n=2 experiments, performed with triplicate cultures. C, Hill slopes of the curves.

102

4.4 Discussion

Our major finding is the direct role for SREBP-2 in the regulation of DHCR24, and identification of two functional SREs within the promoter. Through a comparison of genes regulated by SREBP-2, -1c, or LXR, we found that DHCR24 responded most similarly to the SREBP-2 target genes (Figures 4.1-4.2). We found no evidence that LXR has a major influence on DHCR24 expression, at least in the three human cell-lines we examined. However, as this study did not include cells derived from skin tissue, which was reported to display LXR-mediated regulation of DHCR24 [169], we cannot exclude the possibility of tissue-specific LXR regulation. The essential requirement for SREBP in DHCR24 expression was further demonstrated through the use of mutant CHO cell-lines, with no DHCR24 expression in SRD-13A cells, which lack an active SREBP pathway (Figure 4.2). SREBP binding sites are highly enriched in proximal promoter regions, unlike other transcription factors, which are more widely distributed [206, 207]. Our experiments concur with this observation, with the region of the human DHCR24 promoter required for sterol regulation (-300/-100) containing two separate binding sites for SREBP-2 (-196 invSRE and -160 SRE). Of eight putative SREs initially identified by sequence similarity to known SREs, only two were shown to be functional through mutagenesis (Figure 4.5). A subsequent study by Daimiel et al. [208] has confirmed our findings: that DHCR24 is regulated transcriptionally by SREBP-2. This study demonstrated SREBP-2 occupancy on DHCR24 in vivo using chromatin immunoprecipitation (ChIP), which confirms and strengthens our EMSA findings. Daimiel et al. confirmed that SREBP-2 regulation of DHCR24 occurs through binding to -160 SRE. However, the authors did not identify the second SRE, most likely due to the use of visual inspection and not a matrix based prediction program for identification of putative SREs. Our identification of a second SRE led to our proposal that sterol regulation of DHCR24 requires the cooperativity of SREBP-2 by binding to both SREs. Since an 11.7% overlap exists between SREBP-2 and -1 targets as determined using genome-wide ChIP-seq [206], there is the possibility that -1a or -1c SREBP isoforms may also bind the SREs in the DHCR24 promoter. The transcriptional activities of each of the SREBP isoforms for the two DHCR24 SREs could be determined in vivo; however, the use of ChIP would not be informative here, due to the 103

close proximity of the SREs in the DHCR24 promoter (~ 40 bp). This could instead be tested using these isoforms in similar promoter and EMSA experiments as described in this study. We predict that in a physiological setting, the -2 isoform of SREBP would preferentially occupy the DHCR24 SRE sites, as it has been shown to have the greatest transcriptional activity for SREs, compared with the isoforms -1a and -1c, which activate SREs to a lesser degree [121]. Moreover, DHCR24 was shown to behave most similarly to SREBP-2 gene targets (Figures 4.1-4.2). The SREBP family of transcription factors are relatively weak activators of gene expression alone, and commonly require the cooperation of the co-factors Sp1 and NF-Y. Binding sites for these transcription factors are regions of high GC content and CCAAT boxes respectively, and are located in close proximity to SRE(s) [209]. Putative Sp1 and NF-Y sites were mutated in the -300/-100 region of the DHCR24 gene, and we found that two NF-Y sites were functional in our promoter, and no functional Sp1 sites (Figure 4.10 A). Although no functional Sp1 binding site was identified in DHCR24, this is not unusual; similar arrangements of these cis-acting elements are found in other SREBP-2 target genes, including many cholesterogenic genes such as ACSL1, HMGCS, FDPS, FDFT1, SQLE, TM7SF2, DHCR7, and HSD17B7 which all contain SRE(s) flanked by Sp1 and/or NF-Y sites in their proximal promoter regions [123, 196, 197, 210-214]. Furthermore, the spacing between SREs and NF-Y sites has previously been shown to be critical [123], with the optimal spacing being between 16 – 20 bp. Our deletion studies in Figure 4.10 C, where spacing between -196 invSRE and -174 invNF-Y was incrementally decreased, strongly support this. The native spacing is 17 bp, and decreasing this by 5, 10, or 15 bp greatly inhibited sterol responsiveness. Deleting the entire section between our SREs (including -174 invNF-Y) restored sterol responsiveness, probably because this placed the -160 SRE near enough to the -225 NF-Y for it to help activate both SREs. Mutation of the SREs in the DHCR24 promoter individually resulted in minimal sterol responsiveness (Figure 4.5), consistent with the idea that these dual SREs, which are conserved amongst mammals (Figure 4.9), are cooperative. Although the idea of heterotypic cooperation of SREBPs with cofactors is well known (e.g. [123]), little has been reported on the homotypic cooperation between multiple SREBPs. The importance of the spatial arrangement of SREs was initially proposed for ACC, the rate-controlling enzyme of fatty acid biosynthesis, whose promoter contains dual SREs

104

for both SREBP-1 and -2 [215]. Similar to our results, both SREs of the ACC promoter must be intact for normal sterol regulation. Furthermore, we have clearly demonstrated that the spatial arrangement of our dual SREs and their NF-Ys is critical (Figure 4.10). It should be noted that both our SREs are located on the same side of the nucleosome (within 74 bp), which has been shown to enhance cooperativity between transcription factor binding sites [216]. The two cooperative SREs in FDFT1 are also located on the same side of the nucleosome (~50 bp); and it would be interesting to determine the effect of inserting additional bp to place the two SREs on opposite sides of the nucleosome for both DHCR24 and FDFT1. We have extended this idea of spatial arrangement of SREs in the current study, and propose the idea that homotypic cooperation exists between SREs in close proximity to each other, which enables synergistic gene activation by SREBPs. This, along with Sp1 and NF-Y sites, would contribute to the robustness of the SRE(s) [217]. Energetically, cholesterol synthesis is an expensive process, and cooperativity would ensure that key genes must be strongly activated to commit to cholesterol synthesis, and for ACC, fatty acid synthesis. This hypothesis is supported by the finding that whilst two cholesterol synthesis genes, DHCR24 and FDFT1 (squalene synthase), contain two SREs and displayed a sigmoidal SREBP-2 response curve (with a Hill slope of ~2, clearly indicating cooperativity), the LDLR gene with only one SRE gave a linear response (with a Hill slope of ~1). Since importing cholesterol into the cell through the LDLR requires less energy than synthesis, it would follow that its expression does not need to be as tightly controlled as DHCR24 and FDFT1. This is the first study to specifically observe the role of SREBP-2 homotypic cooperativity in regulating expression of genes with dual SREs (DHCR24 and FDFT1) compared to a gene with only one SRE (LDLR). This finding provides new insights into the complexities of transcriptional regulation involved in cellular cholesterol homeostasis.

105

106

Chapter 5

Sterol-Mediated Post-Translational Regulation of DHCR24 by Inhibition of Activity

TLCs and Western blots in this chapter were performed by Eser Zerenturk and Ika Kristiana. Densitometry and data analysis were performed by Eser Zerenturk.

The following figures presented in this chapter have been published in:

Zerenturk EJ*, Kristiana I*, Gill S, Brown AJ (2012) The endogenous regulator 24(S),25-epoxycholesterol inhibits cholesterol synthesis at DHCR24 (Seladin-1). Biochimica et Biophysica acta, 1821 (9): 1269-1277. * Equal first author. Figures 5.1-5.10

Jansen M, Wang W, Greco D, Bellenchi GC, di Porzio U, Brown AJ, Ikonen E (2013) What dictates the accumulation of desmosterol in the developing brain? FASEB journal, 27: 865-870. Figure 5.11

107

108

5 STEROL-MEDIATED POST-TRANSLATIONAL REGULATION OF DHCR24 BY INHIBITION OF ACTIVITY

5.1 Introduction

Cholesterol synthesis undergoes end product inhibition, with cholesterol feeding back on its own synthesis. However, numerous sterol and oxygenated sterol (oxysterol) intermediates also feed back to signal cholesterol overload (Section 1.5.1). According to current concepts, certain oxysterols may be significant regulators of cholesterol homeostasis [218, 219]. These cholesterol-like molecules contain one or more additional oxygen atoms, usually as a hydroxyl, carbonyl, and/or epoxide group. First discovered in human liver [220], 24(S),25-epoxycholesterol (24,25EC) is one of the most potent oxysterol regulators [221]. Rather than being derived from cholesterol like other oxysterols, 24,25EC has a unique origin, as its synthesis parallels that of cholesterol (Figure 5.1 A). Produced in a shunt of the cholesterol synthesis pathway, it is synthesised in all cells that make cholesterol [221]. 24,25EC works at multiple points to maintain cellular cholesterol homeostasis. Cells control their cholesterol levels by three main mechanisms, notably by regulating synthesis, uptake (especially via low-density lipoprotein, LDL), and cholesterol efflux (Section 1.3). Synthesis and uptake are largely governed by the Sterol Regulatory Element Binding Protein (SREBP) family of transcription factors, whilst genes involved in cholesterol efflux are under the control of the oxysterol-activated liver X receptor (LXR) [8, 219] (Section 1.5.1). Synthesis is also controlled by post-translational degradation of a key rate-controlling enzyme, 3-hydroxy-3-methylglutaryl-coenzyme A reductase (HMGCR) [29, 222] (Section 1.5.2). 24,25EC reduces cell cholesterol levels at all three points [221]: inhibiting synthesis (by stimulating HMGCR degradation) and uptake (both via suppressing SREBP activation), whilst stimulating export (by serving as a potent ligand for LXR). Employing a genetic approach to selectively inhibit endogenous 24,25EC synthesis (2,3-oxidosqualene cyclase (OSC) overexpression), our laboratory found that 24,25EC plays an important role in the acute regulation of cholesterol synthesis, serving as both a monitor and modulator of cholesterol synthesis [119]. Specifically, 24,25EC restrains SREBP activation, fine-tuning the cell’s

109

cholesterol homeostatic response to newly-synthesised cholesterol. Without 24,25EC, cholesterol homeostatic responses are erratic and exaggerated [119].

Figure 5.1. 24(S),25-epoxycholesterol is produced in a shunt of the cholesterol synthetic pathway. A, Schematic pathway of cholesterol synthesis and the production of 24(S),25-epoxycholesterol (24,25EC) via the shunt pathway, and a chemical structure of cholesterol with the carbons numbered. B, A comparison between the chemical structures of 24,25EC, desmosterol and cholesterol highlights their similarity, differing only at C-24,25, by the presence of an epoxide group, double-bond, or single-bond (circled) respectively.

110

DHCR24 catalysed reaction occurs after the branching of the shunt pathway, catalysing the reduction of desmosterol to form cholesterol (Figure 5.1 A). Like many other enzymes in the cholesterol synthetic pathway [16], DHCR24 is under the transcriptional control of SREBP (Chapter 4). Gene expression of DHCR24 is also stimulated by various hormones (androgens, estrogen, thyroid hormones), and growth factors [60, 66, 87, 92] and epigenetics [83]. However, the regulation of DHCR24 activity beyond the transcriptional level remains poorly understood. In this chapter, we investigated if 24,25EC influences DHCR24’s activity, considering that this oxysterol is produced in the endoplasmic reticulum where DHCR24 resides (Chapter 3), plays a key role in fine-tuning cholesterol synthesis [119], and is structurally similar to desmosterol (Figure 5.1 B), the bona fide substrate. This could provide another mechanism by which 24,25EC fine-tunes cholesterol homeostasis.

111

5.2 Materials and Methods

5.2.1 24,25EC synthesis assay

Lipids were extracted (Section 2.4.1) and re-dissolved in 60 μL hexane:diethyl ether (1:1, v/v). Aliquots corresponding to equivalent amounts of protein were separated by thin layer chromatography (TLC) using Silica Gel 60 F254 plates with a mobile phase of hexane:diethyl ether:acetic acid (60:40:1, v/v/v). These TLCs and the corresponding Western blots were performed by research assistant, Ika Kristiana. The bands corresponding to cholesterol and 24,25EC were visualised using the FLA-5100 Phosphorimager. The relative intensities of bands were quantified using Science lab Image Gauge.

112

5.3 Results

5.3.1 24,25EC reduces cholesterol and increases desmosterol levels similar to DHCR24 inhibitors

To measure DHCR24 activity in a cell based system, a metabolic labelling method was applied where cells were labelled for 4 h with [14C]-acetate, which feeds into the beginning of the cholesterol biosynthetic pathway. Cells were pre-treated with a statin (compactin) to reduce cholesterol status and increase the expression of cholesterogenic genes. Thus, when the statin was washed away, there was augmented incorporation of [14C]-acetate into sterols [13, 119], thereby increasing the sensitivity of this assay. Cholesterol and desmosterol were resolved by argentation thin layer chromatography (Arg-TLC) and the radioactive bands were visualised by phosphorimaging. Only desmosterol was present in untreated J774 murine macrophage-like cells that are defective in DHCR24 (Figure 5.2 A, lane 1). Treatment of CHO-7 cells with two known inhibitors of DHCR24, triparanol [223] and U18666A [224], reduced cholesterol and increased desmosterol levels (lanes 3 and 4 versus lane 2). Addition of excess desmosterol resulted in accumulation of [14C]-desmosterol (lane 5), indicating competition for the same enzyme. Therefore, this method was able to detect changes in synthesised desmosterol and cholesterol levels, and therefore was a sufficient method for measuring DHCR24 activity. Incubation with 24,25EC (10 M) also resulted in accumulation of desmosterol at the expense of cholesterol (lane 6), consistent with inhibition of DHCR24 activity. We confirmed that this effect of 24,25EC was generalisable, increasing desmosterol and reducing cholesterol in a number of human cell types, including those of hepatic (HepG2) and neuronal (BE(2)C) origin (Figure 5.2 B).

113

Figure 5.2. 24,25EC treatment results in accumulation of desmosterol at the expense of cholesterol. Cells were pre-treated with mevalonate (50 M) and compactin (5 M) overnight. A, J774 and CHO-7 cells were radiolabelled with [14C]-acetate and treated for 4 h with or without 10 M Triparanol, 0.15 g/mL U18666A, 20 g/mL desmosterol/CD (Desm/CD) or 10 M 24,25EC. B, CHO-7, HeLaT, HepG2, and BE(2)C cells were radiolabelled with [14C]-acetate for 4 h with or without 24,25EC (10 M). Lipid extracts were separated by Arg-TLC, and bands corresponding to cholesterol and desmosterol were visualised by phosphorimager. Phosphorimages are representative of at least n=2 experiments.

5.3.2 Compactin pre-treatment increases sensitivity of Arg-TLC assay

To address the argument that pre-treatment using compactin could create an artificial situation that is prone to rapid suppression by exogenous oxysterols, we examined the effect of 24,25EC addition with and without statin pre-treatment. Notably, cholesterol and desmosterol were the major sterols detected (Figure 5.3 A). Furthermore, the vast majority of [14C]-labelled sterols were found in the cells rather than effluxed into the media (Figure 5.3 A, lanes 5-8). As expected, statin pre-treatment increased [14C]-isotope incorporation into the sterols (Figure 5.3 B). However, regardless of whether cells were statin pre-treated or not, 24,25EC addition reduced incorporation into cholesterol and increased incorporation into desmosterol (Figure 5.3 B). In both cases, 24,25EC appeared to inhibit DHCR24 activity, based on the reduced ratio of the product (cholesterol) to the substrate (desmosterol) (Figure 5.3 C).

114

Figure 5.3. Desmosterol and cholesterol are the main sterols labelled by [14C]-acetate, and are predominantly present in cells rather than media. A, CHO-7 cells were pre-treated with or without mevalonate (50 M) and compactin (5 M) overnight, then treated with or without 2.5 M 24,25EC and radiolabelled with [14C]-acetate for 4 h. Lipid extracts from cell lysate and treatment media were separated by Arg-TLC, and bands corresponding to cholesterol and desmosterol were visualised by phosphorimager. B and C, Densitometry was performed from n=2 experiments presented as mean + half range. Values were expressed relative to the vehicle-treated control condition (lane 1), which was set to 1.

115

5.3.3 The cholesterol to desmosterol ratio is the best indicator for DHCR24 activity

As a proof of concept experiment that the cholesterol to desmosterol ratio is a reasonable surrogate for DHCR24 activity, we incubated cells with increasing concentrations of unlabelled desmosterol substrate (Figure 5.4 A). A mutant CHO cell- line, SRD-1, was used to exclude transcriptional regulation by sterol-independent expression of SREBP-2 target genes [116], which include the majority of cholesterol biosynthetic enzymes [16]. Although the relative [14C]-isotope incorporation into cholesterol decreased in a concentration-dependent manner, and incorporation into desmosterol tended to increase, these changes were not always reciprocal (Figure 5.4 B). By contrast, the relative product (cholesterol) to substrate (desmosterol) ratio was a better indicator of DHCR24 activity, in that it decreased consistently with increasing desmosterol concentration (Figure 5.4 C). Therefore, the cholesterol to desmosterol ratio was used as a primary indicator of DHCR24 activity in this thesis.

116

Figure 5.4. The cholesterol to desmosterol ratio is a reasonable surrogate for DHCR24 activity. A, SRD-1 cells were pre-treated with mevalonate (50 M) and compactin (5 M) overnight, then treated with Desm/CD (0.025-1.6 g/mL) and radiolabelled with [14C]-acetate for 4 h. Lipid extracts were separated by Arg-TLC, and bands corresponding to cholesterol and desmosterol were visualised by phosphorimager. B and C, Densitometry was performed from n=3 experiments presented as mean + SEM, expressed relative to the vehicle-treated control condition (lane 1), which was set to 1, so that for B, incorporation into [14C]-cholesterol+[14C]-desmosterol = 1 at 0 g/mL Desm/CD, and C, incorporation into [14C]-cholesterol/[14C]-desmosterol = 1 at 0 g/mL Desm/CD.

5.3.4 24,25EC reduces the cholesterol to desmosterol ratio without affecting DHCR24 protein levels

To determine the time- and concentration-dependence of 24,25EC, this assay was applied and the cholesterol to desmosterol ratio measured. In CHO-7 cells, 24,25EC reduced the cholesterol to desmosterol ratio rapidly (within 1 h) (Figure 5.5 A). The effect was dose-dependent and already evident at a concentration of 1 M 24,25EC (Figure 5.5 B, lane 2 versus lane 1). Notably, DHCR24 protein levels, as determined by Western blotting, were unaffected (Figure 5.5 B). Furthermore, the effect of 24,25EC was observed in SRD-1 cells (Figure 5.5 C), excluding any potential SREBP-2-mediated transcriptional effects. Collectively, these data support the concept that 24,25EC may directly inhibit DHCR24.

117

Figure 5.5. 24,25EC treatment reduces the cholesterol to desmosterol ratio rapidly and sensitively without affecting DHCR24 protein levels. All cells were pre-treated with mevalonate (50 M) and compactin (5 M) overnight. A: CHO-7 cells were treated with or without 24,25EC (2.5 M) for 1-4 h, and radiolabelled with [14C]-acetate 1 h prior to harvesting. B and C, CHO-7 and SRD-1 cells were treated with 24,25EC (1-10 M) or cholesterol/CD (Chol/CD) (10, 50 M), and for Arg-TLC, radiolabelled with [14C]-acetate, for 4 h. For Western blotting, cells were co-treated with mevalonate (50 M) and compactin (5 M). Cell lysate was subjected to immunoblotting with antibodies against DHCR24 and -tubulin. Lipid extracts were separated by Arg-TLC and bands corresponding to cholesterol and desmosterol were visualised by phosphorimager and densitometry performed. Data from at least n=3 experiments are presented as mean + SEM. Values were expressed relative to the vehicle-treated control condition (lane 1) in each panel, which was set to 1.

118

5.3.5 Effect of various oxysterols on the cholesterol to desmosterol ratio

Cholesterol treatment did not reduce the cholesterol to desmosterol ratio in either CHO-7 or SRD-1 cells (Figure 5.5 B and C), suggesting that the effect may be selective for certain oxysterols like 24,25EC. Further experiments were conducted in CHO-7 cells to investigate the specificity of this effect in terms of the oxygen-containing group (hydroxyl, carbonyl, or epoxide) and position on the sterol backbone. When added at a concentration of 2.5 M (equivalent to 1 g/mL), none of the ring or side-chain oxygenated sterols decreased the cholesterol to desmosterol ratio (Figure 5.6 A and B), except 25-hydroxycholesterol (25HC) (Figure 5.6 B). 22(R)-hydroxycholesterol (22HC) had a less pronounced effect which did not quite attain statistical significance (p=0.05). To determine if these effects were independent of SREBP-2-mediated transcription, side-chain oxysterols were further tested in SRD-1 cells (Figure 5.6 C). Again, 24,25EC and 25HC had the greatest effect in decreasing the cholesterol to desmosterol ratio. Thus, oxysterols oxygenated particularly at C-25 (24,25EC and 25-HC) appear to be effective in modulating DHCR24 activity.

119

120

Figure 5.6. Certain side-chain oxysterols can reduce the cholesterol to desmosterol ratio. CHO-7 (A and B) and SRD-1 (C) cells were pre-treated with mevalonate (50 M) and compactin (5 M) overnight. Cells were radiolabelled with [14C]-acetate for 4 h and treated with 2.5 M of the indicated oxysterols (see table below). Lipid extracts were separated by Arg-TLC, and bands corresponding to cholesterol and desmosterol were visualised by phosphorimager and densitometry performed. Data from at least n=4 experiments are presented as mean + SEM. Values were expressed relative to the vehicle-treated control condition (lane 1) in each panel, which was set to 1. * p<0.05 and ** p<0.01, using a paired t-test versus the control condition (lane 1). Figure 5.6 Sterol Abbreviation A, B, C 24(S),25-epoxycholesterol 24,25EC 4-hydroxycholesterol 4HC 5,6-epoxycholesterol 5,6EC 5β,6β-epoxycholesterol 5β6βEC A 7-hydroxycholesterol 7HC 7-ketocholesterol 7KC 19-hydroxycholesterol 19HC

20-hydroxycholesterol 20HC 22(R)-hydroxycholesterol 22HC B, C 24(S)-hydroxycholesterol 24HC 25-hydroxycholesterol 25HC 27-hydroxycholesterol 27HC

5.3.6 Increased synthesis of endogenous 24,25EC inhibits the cholesterol to desmosterol ratio

To determine if this effect on DHCR24 occurs with endogenous 24,25EC, CHO-7 cells were treated with low concentrations of the OSC inhibitor, GW534511X. This causes the intermediate 2,3(S)-monooxidosqualene (MOS) to accumulate, giving squalene monooxygenase (SM) a greater chance to act again to form 2,3(S):22(S),23- dioxidosqualene (DOS) (Figure 5.7 A). Because OSC favours DOS over MOS, this ensures a greater flux through the shunt pathway to produce more 24,25EC [119, 225, 121

226]. As 24,25EC levels increased (Figure 5.7 B), the cholesterol to desmosterol ratio decreased (Figure 5.7 C).

Figure 5.7. Modulation of endogenous 24,25EC synthesis by pharmacological manipulation decreases the cholesterol to desmosterol ratio. A, 24,25EC synthesis is favoured over cholesterol when there is partial inhibition of 2,3-oxidosqualene cyclase (OSC). B and C, CHO-7 cells were pre-treated with mevalonate (50 M) and compactin (5 M) overnight, then treated with the OSC inhibitor GW534511X (0, 0.1, 1 nM) and radiolabelled with [14C]-acetate for 4 h. Lipid extracts were separated by two different TLC systems: B; regular silica TLC to separate cholesterol and 24,25EC; and C, Arg-TLC to separate cholesterol and desmosterol. Bands were visualised by phosphorimager and densitometry performed. Data from at least n=3 experiments are presented as mean + SEM. Values were expressed relative to the vehicle-treated control condition, which was set to 1.

122

5.3.7 Manipulation of endogenous 24,25EC modulates the cholesterol to desmosterol ratio

Just as partial OSC inhibition increases 24,25EC, our laboratory has previously shown that a genetic approach involving OSC overexpression specifically inhibits 24,25EC production [119] (Figure 5.8 A). Conversely, we reasoned that overexpression of SM would stimulate 24,25EC production, since more MOS is converted to DOS, which is favoured by OSC, again ensuring more flux through the shunt pathway (Figure 5.8 B). CHO-7 cells stably transfected with human SM (CHO-SM) and OSC (CHO-OSC) [130] were used to further test if endogenous 24,25EC affects DHCR24 activity. DHCR24 protein levels were comparable between the cell-lines (Figure 5.8 C), whilst apparent DHCR24 activity decreased as 24,25EC increased (Figure 5.8 C).

123

Figure 5.8. Modulation of endogenous 24,25EC by genetic manipulation alters the cholesterol to desmosterol ratio. A and B, 24,25EC synthesis is altered by the enzymes OSC and SM. A, Cholesterol synthesis is favoured over 24,25EC when OSC is overexpressed; and B, 24,25EC synthesis is favoured over cholesterol when SM is overexpressed. C, A CHO-7 control cell-line containing the empty vector (EV), and two cell-lines expressing either human OSC or SM were pre-treated with mevalonate (50 M) and compactin (5 M) overnight, and for Arg-TLC, radiolabelled with [14C]-acetate for 4 h.. Cell lysate was subjected to immunoblotting with antibodies against DHCR24 and -tubulin. Lipid extracts were separated by Arg-TLC, and bands corresponding to cholesterol and desmosterol were visualised by phosphorimager and densitometry performed. Data from at least n=3 experiments are presented as mean + SEM. Values were expressed relative to EV cell-line which was set to 1.

124

5.3.8 Endogenous 24,25EC appears to regulate DHCR24 activity

Partial inhibition of OSC increased relative 24,25EC levels and DHCR24 activity in these CHO-SM and CHO-OSC to varying degrees (Figure 5.9 A and B). Plotting these two variables against one another shows an inverse relationship, where small increases in relative 24,25EC levels are associated with a precipitous fall in the cholesterol to desmosterol ratio (Figure 5.9 C). Together, these data indicate that small changes in endogenous 24,25EC levels, by pharmacological and genetic methods, are able to regulate DHCR24 activity.

125

126

Figure 5.9. Endogenous 24,25EC appears to regulate DHCR24 activity. Cells overexpressing EV, OSC and SM were pre-treated with mevalonate (50 M) and compactin (5 M) overnight then treated with the OSC inhibitor GW534511X (0, 0.1, 1 nM) and radiolabelled with [14C]-acetate for 4 h. Lipid extracts were separated by two different TLC systems: A; regular silica TLC to separate cholesterol and 24,25EC; and B, Arg-TLC to separate cholesterol and desmosterol. Bands were visualised by phosphorimager and densitometry performed. Data from at least n=3 experiments are presented as mean + SEM. Values were expressed relative to the vehicle-treated control condition in the EV cell-line (lane 1), which was set to 1. C, A scattergram of the average data from A and B yielded the following equation of the line: log(y) = -0.38*log(x) + 0.17 (R=0.89; p=0.001).

5.3.9 Effect of DHCR24 overexpression on 24,25EC’s ability to reduce the cholesterol to desmosterol ratio

If inhibition of DHCR24 is the reason for the reduced cholesterol to desmosterol ratio observed in response to 24,25EC treatment, then we would predict that this effect would be blunted if DHCR24 is overexpressed. To test this hypothesis, CHO-7 cells stably overexpressing human DHCR24 were created. CHO-DHCR24 had substantially higher DHCR24 protein levels than the empty vector stable cells (CHO-EV). Untreated CHO-DHCR24 cells also had a higher cholesterol to desmosterol ratio (~1.3 times greater than in CHO-EV cells) (Figure 5.10). In the CHO-DHCR24 cells, more 24,25EC (at least 5 M) was required to attain the same cholesterol to desmosterol ratio as the CHO-EV cells (1 M) (Figure 5.10, lanes 2 and 8). Thus, with DHCR24 overexpression, the effect of 24,25EC on the cholesterol to desmosterol ratio was indeed blunted, consistent with 24,25EC’s effect occurring by direct inhibition of DHCR24.

127

Figure 5.10. DHCR24 overexpression blunts the inhibitory effect of 24,25EC on the cholesterol to desmosterol ratio. CHO-EV and CHO-DHCR24 cells were pre-treated with mevalonate (50 M) and compactin (5 M) overnight, then treated with 24,25EC (1-5 M), and for Arg-TLC, radiolabelled with [14C]-acetate for 4 h. For Western blotting, cells were co-treated with mevalonate (50 M) and compactin (5 M). Cell lysates were subjected to immunoblotting with antibodies against DHCR24, V5 and -tubulin. Lipid extracts were separated by Arg-TLC and bands corresponding to cholesterol and desmosterol were visualised by phosphorimager and densitometry performed. Data from n=3 experiments are presented as mean + SEM. Values were expressed relative to the vehicle-treated control condition in CHO-EV (not shown), which was set to 1.

5.3.10 Progesterone inhibits DHCR24 activity

Similarly structured products beyond the cholesterol synthesis pathway may also modulate DHCR24 activity, such as the steroid hormone, progesterone. Like 24,25EC, progesterone treatment resulted in an accumulation of desmosterol at the expense of cholesterol in CHO-EV cells, which was blunted in CHO-DHCR24 cells (Figure 5.11). Notably, progesterone did not decrease either ectopic or endogenous DHCR24 protein levels (Figure 5.11). These results indicate that DHCR24 activity is reduced in the presence of progesterone and that more enzyme necessitates more progesterone for inhibition. Because DHCR24 protein levels were not reduced by progesterone, and

128

progesterone inhibited enzyme activity, the inhibition of DHCR24 by progesterone is most probably post-transcriptional, possibly via direct enzyme inhibition.

Figure 5.11. DHCR24 overexpression blunts the inhibitory effect of progesterone on DHCR24 activity. CHO-EV and CHO-DHCR24 cells were treated with progesterone (0-50 nM), and for Arg-TLC, radiolabelled with [14C]-acetate for 4 h. Cell lysates were subjected to immunoblotting with antibodies against DHCR24, V5, and -tubulin. Lipid extracts were separated by Arg-TLC and bands corresponding to cholesterol and desmosterol were visualised by phosphorimager and densitometry performed. Data from n=8 experiments are presented as mean + SEM. * p<0.001, using a paired t-test.

129

5.4 Discussion

The oxysterol 24,25EC can affect cholesterol homeostasis in a variety of ways. As a potent ligand for LXR, it can induce cholesterol efflux [119] and potentially decrease cholesterol uptake via the LDL receptor [227]. By binding to Insig, it can suppress SREBP activation [22], thereby down-regulating genes involved in cholesterol uptake and synthesis. 24,25EC can also decrease cholesterol synthesis by accelerating the degradation of HMGCR [228]. Our laboratry previously proposed that 24,25EC has an especially significant role in fine-tuning cholesterol synthesis, since it closely tracks cholesterol production under a variety of conditions [225], and without it, acute cholesterol synthesis is exaggerated [119]. In this chapter, we reveal a novel mode by which 24,25EC can regulate cholesterol synthesis, by interfering with the reaction catalysed by DHCR24. We metabolically labelled cells with [14C]-acetate which was incorporated into two main sterols: [14C]-cholesterol and [14C]-desmosterol (Figure 5.3), the product and substrate of DHCR24, respectively. Cholesterol and desmosterol are the major bands evident in our cell systems; however, DHCR24 can also act on other intermediates of the cholesterol pathway. These may represent the minor bands evident in a full phosphorimage which accumulate on 24,25EC addition (Figure 5.3). Most of the [14C]-sterol was observed in the cells, with negligible [14C]-sterol lost to the media during the assay (Figure 5.3). Experiments using unlabelled desmosterol to compete for DHCR24 helped to justify our use of the cellular cholesterol to desmosterol ratio as a proxy measure for DHCR24 enzyme activity (Figure 5.4). This ratio was judged to be superior to incorporation of the [14C]-isotope into the individual sterols, because it tended to be more consistent within and between repeat experiments. Moreover, this ratio helped to circumvent the lack of an obvious reciprocal change in incorporation of [14C]-isotope into these two sterols observed in some experiments (e.g. Figures 5.2 B, 5.3, 5.7-5.9). This lack of reciprocity was probably due to other effects of 24,25EC on SREBP activation and HMGCR degradation. We showed that 24,25EC rapidly reduces the product to precursor ratio of DHCR24, without affecting DHCR24 protein levels, in line with this oxysterol interfering with the activity of this enzyme. This effect was observed in multiple cell-lines, including mutant CHO cells (SRD-1) that lack SREBP transcriptional regulation (Figure 5.5). Furthermore, overexpression of DHCR24 blunted the effect of 130

24,25EC (Figure 5.10); supporting a direct inhibitory effect of 24,25C on the enzyme. This effect was relatively specific in that it was not seen with high doses of cholesterol (Figure 5.5) and was restricted to certain side-chain oxygenated sterols, particularly 24,25EC and 25HC (Figure 5.6). 24,25EC production occurs in the endoplasmic reticulum where DHCR24 resides. Importantly, endogenous levels of 24,25EC, manipulated both pharmacologically and genetically, appeared to be sufficient to reduce DHCR24 activity (Figures 5.7-5.9). Moreover, there was no detectable accumulation of desmosterol in the absence of 24,25EC, consistent with this being a physiological mechanism of controlling cholesterol synthesis. Other side-chain oxysterols 20α-, 22(R), 25-, and 27- hydroxycholesterol and the steroid hormone progesterone, were observed to mimic the inhibitory effect of 24,25EC, at least to some extent (Figures 5.6, 5.12). Similarly, C-22 unsaturated plant and fungal sterols reduce the product to precursor ratio of DHCR24 in HL-60 and Caco-2 cells at similar concentrations tested in this study [38]. Furthermore, these sterols were shown to competitively inhibit DHCR24 in microsome preparations [38]. The substrate desmosterol, 24,25EC and the C-22 unsaturated sterols (ergosterol, brassicasterol, stigmasterol and 5,22-cholestedien-3β-ol), are all structurally similar, with the presence of a more electron-rich aliphatic side-chain compared to cholesterol. Therefore, 24,25EC is also likely to directly interfere with the activity of DHCR24. However, further work is required to complement our cell culture-based approaches, by testing 24,25EC and other oxysterols in a DHCR24 enzyme assay in vitro. In addition, such experiments would allow the mode of inhibition to be determined, which we would predict should be competitive, based on structural similarities with reported competitive inhibitors [38]. In conclusion, this chapter introduces a novel putative role for 24,25EC in its multivalent control of cholesterol homeostasis, paring back cholesterol synthesis by inhibiting the ultimate step catalysed by DHCR24.

131

132

Chapter 6

Regulation of DHCR24 by Post-translational Modifications

DHCR24 stable cell-lines were created by Eser Zerenturk. Arg-TLCs were performed by Eser Zerenturk and Ika Kristiana. GC-MS was performed by Eser Zerenturk. Western blots and qRT-PCR were performed by Winnie Luu and Ika Kristiana. Densitometry and data analysis were performed by Eser Zerenturk and Winnie Luu.

The following figures in this chapter have also been presented in Winnie Luu’s PhD thesis (2013) The role of signalling and sensing in cellular cholesterol homeostasis. Figures 6.1, 6.4, and 6.7.

The following work has been published in:

Luu, W*., Zerenturk E.J*., Kristiana I., Bucknall, M., Sharpe L.J., Brown A.J. (2013) Signalling regulates activity of DHCR24, the final enzyme in cholesterol synthesis. J Lipid Res, 55(3): 410-420. * Equal first author.

133

134

6 REGULATION OF DHCR24 BY POST-TRANSLATIONAL MODIFICATIONS

6.1 Introduction

Gene expression of DHCR24 is responsive to numerous factors, such as sterols (Chapter 4), as well as hormones, xenobiotics, and epigenetic regulation (Section 1.8.1). In contrast, relatively little is known about the post-translational regulation of DHCR24 activity. In this thesis, we have found that the oxysterol regulator, 24(S),25-epoxycholesterol (24,25EC), inhibits DHCR24 activity with potent effects on cellular cholesterol levels (Chapter 5). Plant sterols (Section 1.8.2), progesterone (Chapter 5) and similar progestins [39, 106] similarly inhibit DHCR24 activity. However, the role of post-translational modifications in DHCR24 regulation, such as phosphorylation, have yet to be investigated. Large-scale proteomic studies have identified phosphorylation sites on DHCR24; T110 [110] and Y321 [111], and there is evidence for additional phosphorylation sites identified by large-scale unpublished proteomics studies [PhosphoSitePlus; 161]. The function of these phosphorylation sites and the kinases/phosphatases involved are currently unknown due to a lack of mechanistic studies. Cell signalling by phosphorylation is a major mode of regulating cellular processes, and is implicated in various facets of cholesterol homeostasis: cholesterol synthesis and uptake at SREBP-2 and HMGCR by Akt and AMP-activated protein kinase (AMPK) (Section 1.5.2) respectively; cholesterol efflux at ABCA1 and ABCG1 by protein kinase A (PKA) [reviewed in 229]. There is emerging evidence that other enzymes in the cholesterol synthetic pathway besides HMGCR (the classic rate-limiting step) are potential control points for rapid regulation of cholesterol synthesis [17], such as SM [13]. DHCR24 is a prime candidate [17] due to its position in the Bloch pathway as the final enzyme. Thus, regulating DHCR24 by phosphorylation would allow for a rapid means for attenuating cholesterol synthesis, with the previous chapter in this thesis demonstrating its ability for rapid inhibition by multiple feedback mechanisms (Chapter 5). Furthermore, modulating DHCR24 activity alters levels of desmosterol, a strong LXR ligand [73, 230], further reducing cellular cholesterol status. In this chapter, we examined the

135

effect of inhibition of major protein kinases, and a known phosphorylation site, on DHCR24 activity. This could provide a novel regulatory mechanism for cholesterol synthesis at DHCR24 by kinases and phosphorylation.

136

6.2 Materials and Methods

6.2.1 Materials

Lipofectamine RNAiMAX transfection reagent was from Life Technologies (Carlsbad, CA, USA). Akt inhibitor VIII was from Merck (Darmstadt, Germany). N-O-bis-(trimethylsilyl) trifluoroacetamide containing 1% trimethylchlorosilane (BSTFA), bisindolylmaleimide (BIM), Ro-318220 and H89 were from Sigma-Aldrich 2 (St. Louis, MO, USA). [ H6]-desmosterol was from Avanti Polar Lipids (Alabaster, AL, USA).

6.2.2 siRNA transfection

Cells were transfected with 25 nM control or hamster specific DHCR24 siRNA (target sequence: GAGAGCCACGTGTGAAGCA; designed by Sigma Aldrich) for 24 h using Lipofectamine RNAiMAX transfection reagent, according to the manufacturer’s instructions. The cells were washed with PBS, and then refed fresh media and labelled with 1 μCi/well [14C]-acetate for 4 h. Cells were harvested, lipids were extracted and separated by Arg-TLC as described in Section 2.4.

6.2.3 Gas chromatography mass spectrometry

Cells were treated as indicated in the figure legends with or without metabolic labelling 2 using 1 μg/mL [ H6]-desmosterol complexed to methyl-β-cyclodextrin 2 ([ H6]-desmosterol/CD) for 4 h. Cells were harvested and lipid extracts were prepared as described in Section 2.4.1. Lipids were derivatised with BSTFA for 1 h at 60 °C. Derivatised samples were analysed using a Thermo Trace gas chromatograph (GC) coupled with a Thermo DSQII mass spectrometer (MS) and Thermo Triplus Autosampler. Samples (1 μL) were injected via a heated (290 °C) splitless inlet into an Rxi-5Sil MS w/Integra-Guard, 30 m × 0.25 mm, 0.25 μm film thickness, capillary GC column. The column oven was initially held at 80 °C for 1 min, then heated to 260 °C at 80 °C min-1, then to 280 °C at 10 °C min-1 and then to 295 °C at 2 °C min-1. Finally, the oven was increased to 305 °C at 10 °C min-1 and held for 1 min. Helium was used as a carrier gas at a constant flow (1.3 mL min-1, with vacuum compensation on).

137

Mass spectrometry conditions: electron energy 70 eV, ion source temperature 200 °C, transfer line temperature 305 °C. The emission current was set to 130 μA and the detector gain to 3.0 × 105. Samples were analysed either in scan mode (35-520 Da, 2.5 scans s-1) to obtain mass spectra for peak identification, or in single ion monitoring (SIM) mode to measure DHCR24 activity (m/z ions monitored in Table 6.1). Thermo Xcalibur Software was used to acquire and process the data. 5α-cholestane (0.1 μg added during cell harvesting) was used as an internal standard. Peak identification was based on comparison of retention times with those of standards and comparison of mass spectra with the Wiley 9/NIST 2011 combined mass spectral library. The SIM ions for 5α-cholestane were monitored from 8.20 to 10.50 min (width 2 0.1 Da, dwell 80 ms); for cholesterol and [ H6]-cholesterol from 10.50 to 11.45 min 2 (width 0.1 Da, dwell 30 ms); for desmosterol and [ H6]-desmosterol from 11.45 to 14.7 min (width 0.1 Da. dwell 30 ms). Quantification was based on chromatographic peak areas as measured relative to that of the internal standard. SIM chromatograms were smoothed using 15 point Gaussian smoothing and chromatographic peaks were integrated using the following parameters: baseline window 80, area noise factor 40, peak noise factor 20, multiplet threshold 5 and noise method set to “Incos noise”.

Table 6.1. Ion m/z values monitored during selective ion monitoring analysis of the trimethylsiliyl derivatives. The presence of coincident chromatographic peaks in the confirmation ion chromatograms was used to verify analyte identity. Retention Quantification ion Confirmation ion Compound time (min) m/z value m/z value 5α-cholestane (IS) 8.54 217.20 372.35 Cholesterol 11.27 368.33 458.38

2 [ H6]-Cholesterol 11.18 374.38 464.42 Desmosterol 11.69 327.20 456.36

2 [ H6]-Desmosterol 11.60 333.29 462.33

138

6.3 Results

6.3.1 Protein kinase C inhibition decreases DHCR24 activity

To investigate the possibility of phosphorylation playing a role in DHCR24 activity, we first inhibited some common protein kinases and measured the effects on DHCR24 activity. We used inhibitors of protein kinases A (PKA; H89 [231]), B (PKB, also called Akt; AktiVIII [232]), and C (PKC; BIM [233] and Ro-318220 [234]). CHO-7 cells stably transfected with human DHCR24 were radiolabelled with [14C]-acetate, which feeds into the beginning of the cholesterol synthetic pathway. Cholesterol and desmosterol were resolved by Arg-TLC, and radioactive bands were visualised by phosphorimaging. The cholesterol to desmosterol ratio was used as an indicator of DHCR24 activity, as in Chapter 5. As shown previously (Figure 5.2 A), treatment with unlabelled desmosterol decreased the cholesterol to desmosterol ratio, demonstrating competitive inhibition of DHCR24 enzyme. Inhibition of PKA (H89) or PKB (AktiVIII) did not have a selective effect on DHCR24 activity, although H89 affected cholesterol synthesis on a global level (Figure 6.1 A). In contrast, the PKC inhibitors BIM and Ro-318220 reduced cholesterol levels and accumulated desmosterol within 4 h, indicating decreased DHCR24 activity. BIM had the most robust effect, and ablated DHCR24 activity in a dose-dependent manner (Figure 6.1 B). Conversely, PKC stimulation with phorbol myristate acetate increased cholesterol synthesis by ~28% ± 5%, p≤0.01 using a paired t-test (two-tailed). Since inhibition did not affect DHCR24 protein levels (Figure 6.1 A), these results suggest PKC may be influencing DHCR24 activity through modulating its phosphorylation state.

139

Figure 6.1. Effect of kinase inhibitors on DHCR24 activity. A, CHO-DHCR24 cells were treated for 4 h with desmosterol/CD (Desm/CD, 2 µg/mL), H89 (10 µM), Akt inhibitor VIII (AktiVIII, 5 µM), bisindolylmaleimide I (BIM, 5 µM), or Ro-318220 (10 µM). B, CHO-DHCR24 cells were pre-treated in NBS, and then treated for 4 h with BIM at the indicated concentrations. A and B, Cells were either harvested for Western blotting, or radiolabelled with [14C]-acetate during the treatment for Arg-TLC. For Western blotting, whole cell lysates were subjected to SDS PAGE and Western blotting for DHCR24 (V5) and α-tubulin. Western blots are representative of at least n=2 experiments. For Arg-TLC, lipid extracts were separated by Arg-TLC, and bands corresponding to cholesterol and desmosterol were visualised by phosphorimaging. Arg-TLC is representative of at least n=2 experiments.

140

6.3.2 PKC inhibition results in desmosterol accumulation

To confirm the identity of the desmosterol band accumulating with BIM treatment using Arg-TLC, samples were re-analysed using gas chromatography mass spectrometry (GC-MS, Figures 6.2-6.3). Two peaks were identified (Figure 6.2 A), where the major peak was cholesterol and a second minor peak, which increased with BIM treatment, was identified as desmosterol based on the mass spectra (Figure 6.2 B). To measure 2 DHCR24 activity, samples were treated with BIM and [ H6]-desmosterol. As observed with Arg-TLC (Figure 6.1 A), competition with unlabelled desmosterol inhibited the 2 2 conversion of [ H6]-desmosterol to [ H6]-cholesterol, resulting in an increase in 2 [ H6]-desmosterol (Figure 6.3). BIM treatment had similar effects, strongly inhibiting 2 DHCR24 activity, as evidenced by the [ H6]-cholesterol to desmosterol ratio (Figure 6.3).

141

Figure 6.2. Bisindolylmaleimide I treatment results in accumulation of desmosterol. CHO-7 cells stably overexpressing DHCR24-V5 were treated with desmosterol/CD (Desm/CD, 2 µg/mL) or bisindolylmaleimide I (BIM, 5 µM) for 4 h. Lipid extracts and a desmosterol standard were analysed by GC-MS. A and B show total ion chromatograms, and B also shows mass spectra obtained for the peak labelled ‘D’ and used to verify its identity as desmosterol. GC-MS is representative of at least n=3 experiments.

142

Figure 6.3. BIM inhibits DHCR24 activity. CHO-7 cells stably overexpressing DHCR24-V5 were treated with desmosterol/CD (Desm/CD, 2 µg/mL) or bisindolylmaleimide I (BIM, 5 µM), and labelled with 2 [ H6]-desmosterol/CD (1 μg/mL) for 4 h. Lipid extracts and a desmosterol standard 2 were analysed by GC-MS. SIM ion traces were used to quantify [ H6]-cholesterol (C) 2 and [ H6]-desmosterol (D) relative to the internal standard, (IS) 5α-cholestane. Data are mean + SEM from at least n=3 experiments.

143

6.3.3 DHCR24-T110 phosphomutant does not alter expression

From the panel of kinase inhibitors tested, only PKC inhibitors affected DHCR24 activity. Since PKC phosphorylates serine and threonine residues, it is possible that BIM inhibits DHCR24 activity through the only known serine/threonine phosphorylation site, T110 [110]. Thus, we mutated this to alanine (T110A; phosphomutant) or glutamic acid (T110E; phosphomimetic), and stably overexpressed these constructs in CHO-7 cells, as we did previously with the generation of the CHO-DHCR24-WT cells (Section 5.3.9). DHCR24 mRNA and protein levels between the CHO-DHCR24 cell-lines were comparable (Figure 6.4).

144

145

Figure 6.4. Characterisation of the CHO-DHCR24-T110A/E stable cell-lines. CHO-7 cells stably overexpressing the empty vector (EV), or V5-tagged wild-type (WT) or mutant (T110A, T110E) DHCR24 were tested for mRNA and protein levels. A, Total RNA was harvested and mRNA levels for DHCR24 was measured using qRT-PCR and normalised to porphobilinogen deaminase (PBGD). Data are presented relative to DHCR24-WT, which has been set to 1. Data are from n=3 experiments, each triplicate cultures, and are presented as mean + SEM. B, Whole cell lysates were subjected to SDS-PAGE and Western blotting for DHCR24 (V5) and α-tubulin. Western blots are representative of n=3 experiments, and the relative expression represents the densitometric quantification of DHCR24, where the DHCR24-WT expression has been set to 1. C, Histograms represent ectopic DHCR24 protein/mRNA (mean + SEM), from n=3 experiments.

6.3.4 DHCR24-T110 phosphomutant decreases DHCR24 activity

To measure the specific effects of phosphorylation mutations on human DHCR24, endogenous hamster DHCR24 expression was reduced using RNA interference. Transfection of hamster-specific DHCR24 siRNA ablated endogenous hamster DHCR24 mRNA without affecting human DHCR24 expression (Figure 6.5). Thus, transfecting CHO-DHCR24 cells with hamster siRNA reduced background activity in our stable cells, with measured DHCR24 activity attributed to the stably expressed human DHCR24 only.

146

Figure 6.5. Hamster specific knockdown of DHCR24 mRNA. CHO-7 (hamster) and HeLaT (human) cells were transfected with control or hamster-specific DHCR24 siRNA (25 nM) for 24 h. The cells were washed with PBS and refed fresh media overnight. Total RNA was harvested and mRNA levels for DHCR24 were measured using qRT-PCR and normalised to porphobilinogen deaminase (PBGD). Data are presented relative to the hamster control siRNA condition which has been set to 1, and are mean + SD performed in triplicate cultures per condition.

Using this approach, we examined whether DHCR24-T110 mutants affect DHCR24 activity. As indicated in Figure 6.6, transfecting DHCR24 siRNA into EV cells ablated endogenous DHCR24 activity. Mutating T110 so that it cannot be phosphorylated (T110A) significantly decreased DHCR24 activity by ~60% (Figure 6.6). Furthermore, DHCR24 activity of the phosphomimetic (T110E) mutant was increased relative to T110A, with activity not statistically different from WT cells (p=0.06). These results suggest that phosphorylation at T110 modulates DHCR24 activity.

147

Figure 6.6. Effect of T110 mutants on DHCR24 activity. CHO-7 cells stably overexpressing EV or DHCR24 (WT, T110A, T110E) were transfected with control or hamster-specific DHCR24 siRNA (25 nM) for 24 h. The cells were washed with PBS and refed fresh media overnight, and then radiolabelled with [14C]-acetate in fresh media. Lipid extracts were separated by Arg-TLC, and bands corresponding to cholesterol and desmosterol were visualised by phosphorimaging. Arg-TLC is representative of n=4 experiments, and the relative cholesterol to desmosterol ratio was quantified by densitometry, normalised to protein expression in Figure 6.4 B, and then normalised to the DHCR24 siRNA transfected CHO-DHCR24-WT condition, which has been set to 1. **, p≤0.01 Data are mean + SEM (n=4).

6.3.5 PKC inhibition decreases DHCR24 activity independently of T110

Next, we determined whether BIM inhibition of DHCR24 activity occurs at T110. Cells were treated in low (LPDS) or high (NBS) cholesterol conditions to exclude any potential effects of cholesterol status. BIM had similar effects in DHCR24-T110A as WT cells: a decrease in DHCR24 activity, with no effect on DHCR24 protein levels (Figure 6.7). This indicates that BIM targets a different residue in DHCR24, or that the effect is indirect.

148

Figure 6.7. BIM does not affect DHCR24-T110A activity. CHO-7 cells stably overexpressing DHCR24 (WT, T110A) were pre-treated in either LPDS or NBS containing media (A or C) overnight, and treated in fresh media, with or without BIM (5 µM) for 4 h. Cells were either harvested for Western blotting, or radiolabelled with [14C]-acetate during the treatment for Arg-TLC. Whole cell lysates were subjected to SDS-PAGE and Western blotting for DHCR24 (V5) and α-tubulin. Western blots are representative of n=2 experiments. Lipid extracts were separated by Arg-TLC, and bands corresponding to cholesterol and desmosterol were visualised by phosphorimaging. Arg-TLC is representative of n=4 experiments, and the relative cholesterol to desmosterol ratio was quantified by densitometry, and normalised to the non-treated control CHO-DHCR24-WT condition (in LPDS), which has been set to 1. Data are mean + SEM (n=4).

149

6.3.6 The BIM inhibitory effect is not mediated by PKC epsilon

As BIM does not target the only known serine/threonine phosphorylation site (T110; Figure 6.7), we attempted a different approach to identifying the PKC phosphorylation site. The two highest scoring putative PKC phosphorylation sites as predicted by a matrix based prediction program, Scansite [235] (T88 and S334) are both predicted to be PKCε (epsilon) substrates. To determine if PKCε, and therefore either of these putative sites are phosphorylated, cells were treated with a specific PKCε inhibitor. Unlike BIM, the PKCε inhibitor had no effect on DHCR24 activity (Figure 6.8), suggesting PKCε is not the isoform causing the inhibitory effect on DHCR24 upon BIM treatment.

Figure 6.8. PKC epsilon does not affect DHCR24 activity. CHO-7 cells stably overexpressing DHCR24 were treated with BIM (5 µM) or a specific PKC epsilon inhibitor (PKCε inh., εV1-2; 10 µM), and radiolabelled with [14C]-acetate for 4 h. Lipid extracts were separated by Arg-TLC, and bands corresponding to cholesterol and desmosterol were visualised by phosphorimaging. Arg-TLC is representative of n=2 experiments.

150

6.4 Discussion

In this chapter, we discovered a novel mode of regulation of a key protein in cholesterol synthesis, DHCR24. Recent studies on DHCR24, and previous chapters in this thesis (Chapters 4-5) have identified a number of regulatory mechanisms, primarily at the transcriptional level, particularly via feedback regulation by sterol/steroid molecules, with only a few at the post-translational level. Regulation by post-translational modifications such as phosphorylation has yet to be investigated, despite evidence that DHCR24 is phosphorylated at multiple sites [161]. One aim of this work was to identify protein kinase(s) involved in DHCR24 phosphorylation by looking at common protein kinases, using a panel of inhibitors. Whilst inhibiting serine/threonine kinases did not affect DHCR24 protein expression, inhibition of PKC (using BIM and Ro-318220) markedly decreased desmosterol to cholesterol conversion. Indeed, BIM reduced DHCR24 activity in a dose-dependent manner, suggesting a specific action of PKC on DHCR24. We then tested if BIM inhibition of DHCR24 activity could be negated by mutating the only published serine/threonine phosphoresidue in DHCR24, T110. T110 is also predicted as a low stringency PKC substrate by Scansite [235]. However, the BIM inhibitory effect was still observed in the DHCR24 T110 phosphomutant cell-line (CHO-DHCR24-T110A), indicating that PKC does not phosphorylate DHCR24 at T110. This suggests that BIM may be affecting DHCR24 directly through an as yet unidentified phosphorylation site(s), or indirectly. We also examined the two highest scoring putative PKC phosphorylation sites as predicted by Scansite [235] (T88 and S334), which are both predicted to be PKCε substrates. However, a specific PKCε inhibitor did not alter DHCR24 activity like BIM. Future work should involve testing other putative sites as predicted by Scansite, for PKC phosphorylation and effect on DHCR24 activity, and to determine which of the remaining untested PKC isoforms is involved in modulating DHCR24 activity. Another aim of this study was to characterise the known phosphorylation sites in DHCR24, and examine their effects on DHCR24 activity. We developed a novel method for measuring DHCR24 activity in cell culture; by stably expressing human DHCR24 in hamster (CHO-7) cells, combined with metabolic labelling. To selectively examine the activity of stably expressed human DHCR24, hamster DHCR24 was

151

knocked down by hamster-specific DHCR24 siRNA (Figure 6.5). Hence, this assay measured cholesterol conversion from desmosterol by ectopic DHCR24 only. We examined T110, a published phosphoresidue [110], and created CHO cell-lines stably expressing DHCR24 T110 mutant (T110A) or a phosphomimetic (T110E). DHCR24-T110A significantly decreased DHCR24 activity relative to DHCR24-WT (by ~60%). Mutating T110 to a phosphomimetic (T110E) restored most of this activity (Figure 6.6), indicating that this is indeed a functional phosphorylation site. Although the physiological relevance of T110 phosphorylation needs to be elucidated, this study introduces a new mode of regulation for DHCR24 activity, and may explain settings where desmosterol accumulates, such as in cholesterol-laden macrophage foam cells [73], or in HCV infection [236]. As T110 lies within the predicted FAD binding domain (55-235) [44] and just outside of the highly conserved region for FAD binding (111-203) [76, 136], it is possible that phosphorylation of T110 affects DHCR24 enzyme activity by interfering with cofactor interaction. Amino acids located near the highly conserved region of the FAD binding domain have been shown to be crucial for DHCR24 function; mutation of arginine 94 (R94H), resulted in an almost complete loss of activity, and causes desmosterolosis [42]. R94 has also been predicted to be involved in specific interactions between FAD phosphate groups and DHCR24 [44]. T110 is preceded by three positively charged residues, which are all conserved in higher organisms (Appendix 8.2), and may also be involved in electrostatic interactions with the negatively-charged phosphate groups in FAD. Further work testing FAD binding affinity in DHCR24-T110A is required to consolidate this hypothesis. Together, this study has demonstrated that signalling and phosphorylation have functional effects on DHCR24. Further work is required to investigate the precise kinase(s) responsible for T110 phosphorylation, as well as the phosphorylation site(s) required for PKC’s action.

152

Chapter 7

General Discussion

153

154

7 GENERAL DISCUSSION

Since its discovery more than a decade ago, DHCR24 has been gaining interest not only in lipid research, but also in cancer, neuroscience, cardiovascular, and viral research, to name a few. DHCR24 and its substrate desmosterol play a pivotal role in cholesterol homeostasis, and we are only just beginning to appreciate the importance of these in a variety of cellular processes and disease settings. This thesis aimed to address the deficit of fundamental information on DHCR24, such as its topology and interaction with cellular membranes, and how it is regulated. We found that DHCR24 is an integral membrane protein, with multiple membrane associated regions, contrary to published topology models and in silico predictions (Chapter 3; Figure 7.1). We also found that multiple regulatory mechanisms act upon DHCR24 (Figure 7.1), specifically: 1. Sterol-mediated transcriptional regulation, mediated by cooperative binding of SREBP-2 and NF-Y transcription factors to the proximal promoter of DHCR24 (Chapter 4). 2. Post-translational regulation by the oxysterol 24,25EC and to a lesser degree other side-chain oxysterols, but not cholesterol, which feed back to inhibit DHCR24 enzyme activity (Chapter 5). 3. Post-translational regulation by the steroid hormone progesterone (Chapter 5). 4. The role of signalling, with phosphorylation affecting DHCR24 activity. We found evidence for PKC, and also a separate and distinct phosphorylation site at T110, both suggesting that DHCR24 is highly active in the phosphorylated state (Chapter 6). These findings indicate that DHCR24 has the potential to be a control point in cholesterol synthesis beyond HMGCR and SM, as it is highly regulated through multiple mechanisms. Together, this thesis provides fundamental new insights into DHCR24 and its regulation. In this chapter, we discuss the implications of these findings, with recommendations for future directions.

155

156

Figure 7.1. Membrane topology and regulatory mechanisms of DHCR24. DHCR24 is an integral membrane protein, with multiple re-entrant loops. Numerous feedback mechanisms regulate DHCR24: 1. transcriptional regulation by the nuclear form of SREBP-2 (N) and NF-Y; 2. Post-translational regulation by oxysterols, such as 24,25EC and 25HC, and 3. progesterone. 4. Signalling is involved in post-translational regulation, at T110, and by PKC.

7.1 Membrane topology of DHCR24: an unorthodox protein

7.1.1 Lack of common recognition sequences

DHCR24 is predicted to contain a signal peptide at the N-terminus, but we have shown that this region does not get cleaved, and is therefore not a signal peptide. This is similar to microsomal cytochrome P450s [237, 238]. Nuclear magnetic resonance studies show that the hydrophobic N-terminus of cytochrome P450 8A1 is not cleaved, but forms an anchor, interacting with only one leaflet of the ER membrane [239]. This is in agreement with our findings that the N-terminus of DHCR24 is not cleaved, but forms an anchor/peduncle in the ER membrane. Lastly, both proteins do not require the anchor for ER membrane targeting; therefore, both hydrophobic anchors appear not to have a signal function, but rather make up a membrane associated region. Other membrane bound proteins contain a di-arginine motif at the N- or C-terminus (RR, within first/last five amino acids), or a di-lysine motif at the C-terminus (which must be in the -3 and -4/-5 position relative to the C-terminus; KKxx or KxKxx) [240]. Whilst DHCR24 does not contain either motif at the N-terminus, it is rich in arginines and lysines in the C-terminus, although they do not conform to either classical motif [240- 242]. However, this region may consist of atypical retention signals, and therefore, future truncation and mutation studies are required to determine if they are bona fide. The DHCR24 catalysed reaction is strictly dependent on NADPH as the reducing agent [77, 135], and appears to require the cofactor FAD [77]. Unlike other NADPH dependent proteins, DHCR24 does not contain a consensus sequence for NADPH binding. Furthermore, DHCR24 does not contain the sequence binding motif common to most flavoproteins GxGxxG, where x is any amino acid [243]. However, DHCR24 does contain a highly conserved FAD binding domain, which contains a stretch of 10 amino acids that are identical in diverse species (Appendix 8.2) at the N-terminus (Figure 3.1).

7.1.2 Other cholesterol synthetic enzymes

Oxidosqualene cyclase (OSC) is a monotopic ER membrane protein, containing a membrane associated region, which resides in one leaflet of the membrane, and therefore does not span the bilayer [244]. OSC is one of two cholesterogenic enzymes that have their structure solved [244, 245], and the only enzyme with its membrane 157

topology elucidated [244]. Its catalytic domain resides in the membrane, forming a cavity, which is accessible to the extremely hydrophobic substrate, oxidosqualene. As the substrate binding domain of DHCR24 is as yet uncharacterised, but predicted to reside in the membrane associated C-terminus, and desmosterol is also hydrophobic, a membrane associated catalytic site is likely. However, both published topology models have the C-terminus, and therefore active site, residing in the cytosol [44, 81]. Our model of DHCR24 membrane topology, containing multiple membrane associated regions along the C-terminus, would be more fitting for such a structure. Solving the crystal structure of DHCR24 with desmosterol, similarly to lanosterol synthase, an integral ER membrane protein, with its substrate, oxidosqualene, would determine where the catalytic site is, and whether it is membrane bound. As most other cholesterogenic enzymes have their structure undefined, and interact with hydrophobic substrates, we predict that they may also contain atypical membrane topologies.

7.1.3 Other functions of DHCR24

DHCR24 has been observed to translocate to the nucleus under cellular stress, which was reversed upon removal of the stress stimulus [57, 58, 81]. Furthermore, smaller variants of DHCR24 (~40 kDa) have been observed during apoptosis, produced by caspase cleavage, which are cytosolic [52]. In light of our new topology model of DHCR24, these cellular relocalisations seem unlikely. Firstly, we have found that DHCR24 cannot be solubilised from the membrane; secondly, most of the protein is membrane associated, therefore, it is unlikely caspase cleavage would produce a soluble 40 kDa fragment. This has implications for other functions of DHCR24, which require accessibility to soluble substrates, cofactors, and binding proteins, such as p53 and Mdm2 [58]. Reports of DHCR24 translocation should be re-examined in light of our results. Potential mechanisms that allow binding of other proteins to DHCR24 might include a conformational change to DHCR24 in response to sterol and stress signals, allowing membrane protected regions access to cofactors and binding partners, respectively.

7.1.4 Future Experiments

To refine our membrane topology model of DHCR24, further experimentation is required to precisely define membrane, cytoplasmic and luminal loops. This commonly

158

requires modification of specific residues or introduction of epitopes and recognition sites. Cysteine derivitisation and glycosylation site mapping have previously been successful in elucidating the membrane topology of membrane bound proteins, such as Insig [77], and these techniques could similarly be applied here. Both techniques require the attachment of large molecules/residues to cysteine residues or glycosylation sites, and the membrane orientation of these sites is determined by the level of derivitisation or glycosylation, as determined by size changes of the protein using Western blotting. This thesis examined the membrane topology of ectopic DHCR24 using an overexpression system, not native DHCR24. Future investigations should characterise native DHCR24 membrane topology in a more physiological setting, as we have highlighted previously the potentially detrimental effect of epitopes on the native topology of membrane proteins. After the creation of a complete membrane topology map we can get a better understanding of the role of DHCR24 in various cellular functions.

7.2 Transcriptional regulation of DHCR24 by sterols

7.2.1 LXR vs SREBP-2

In Chapter 4, we characterised the transcriptional regulation of DHCR24 by sterols. Contrary to the only published finding on sterol regulation of DHCR24 at the time of this study [97], we found no involvement of LXR. We instead found that DHCR24 expression was responsive to SREBP-2. This previous study found LXR occupation on DHCR24 by chromatin immunoprecipitation (ChIP)-on-chip, and affinity for this site was demonstrated using EMSAs; however, this study lacked functional data demonstrating LXR regulation of DHCR24. Subsequently, a separate study has characterised another LXR site within the proximal promoter [93, 94]. This site is not conserved between mouse and humans, but is distinct in sequence and location. Furthermore, the demonstrated LXR regulation was modest (<2-fold in a range of cell types), which was not detectable in our hands using the same synthetic LXR ligand,

159

TO-901317. Therefore, sterol regulation of DHCR24 appears to occur predominately by SREBP-2, rather than LXR.

7.2.2 Direct binding of SREBP-2 to DHCR24

We characterised DHCR24 regulation in response to SREBP-2 at two distinct sites in the proximal promoter. To demonstrate direct interaction of SREBP-2 to these sites, multiple methods could be used. EMSAs demonstrate the in vitro binding affinity of proteins to specific nucleotide sequences, paralleling the binding of transcription factors to response elements in target gene promoters. Alternate methods are ChIP, which shows the direct binding of protein and DNA in vivo, and ChIP-seq, which can additionally identify the DNA sequence the protein binds to. ChIP based methods require antibodies against the DNA binding protein; as commercially available SREBP-2 antibodies at the time of the study were not sensitive or specific enough to pull-down SREBP-2 and bound DNA, these methods were not employed in this thesis, and therefore an EMSA was developed and optimised. However, there are multiple drawbacks with using this assay. It utilised fluorescent and not radiolabelled oligonucleotides; nuclear extracts and not recombinant protein as a source of SREBP-2. These features decreased the sensitivity and specificity of the assay respectively. We addressed these issues through the use of salmon sperm DNA, which improved specificity; however, identification of SREBP-2 as the bound protein still remained an issue, as nuclear extract contains many DNA binding proteins. Therefore, to demonstrate specific SREBP-2 recognition of our SREs, additional assays were required testing SREBP-2 specific antibodies; however, this was not successful, even with the use of ectopic antibodies. We eventually demonstrated specific SREBP-2 binding by using CHO cell-lines of varying levels of SREBP-2. A subsequent study by Damiel et al. has since demonstrated SREBP-2 occupancy on the DHCR24 promoter by ChIP [208], using a commercially available antibody (Abcam), suggesting that commercial antibodies have improved for use in ChIP assays. Furthermore, continuing use and optimisation of the EMSA protocol in our laboratory has since improved this method, through the introduction of recombinant SREBP-2, which has addressed the issues of sensitivity and specificity.

160

7.2.3 Relevance of DHCR24 being SREBP-2 regulated

This thesis strengthens previous findings [16] that DHCR24 is a bona fide SREBP-2 gene target, and may explain previous findings. For example, the observation that a statin up-regulates DHCR24 expression in human neuronal cells and murine brains [168] is likely mediated through SREBP-2, considering that statins activate SREBP-2. Similarly, high-density lipoprotein may increase DHCR24 expression [72] by removing cholesterol from cells and thus activating SREBP-2. In addition, HCV infection augments DHCR24 expression and facilitates viral replication [246, 247]. Considering that HCV infection also increases SREBP-2 signalling via Akt [248], we propose that increased DHCR24 levels may be explained by increased transcriptional activity of SREBP-2. Interestingly, the putative Sp1 site identified in DHCR24 that mediates HCV infection does not play a role in SREBP-2 regulation of DHCR24, despite being in close proximity of the SREs; DHCR24 augmentation by Sp1 was responsive to oxidative stress, and not sterols [71]. DHCR24 up-regulation may be an important factor for promoting not just HCV, but various other viral infections. Similar to observations in HCV studies, DHCR24 expression correlated with human immunodeficiency virus infection [249].

7.2.4 Relevance to future work

We predict this mechanism will be more relevant in the future as new findings emerge on the cooperativity of SREBP-2 in the regulation of other target genes that contain more than one SRE. This could identify other novel control points, like DHCR24, in the cholesterol synthetic pathway, and this concept of homotypic cooperativity to pass a set threshold may be extended to other transcription factors. Furthermore, as there is a distinct lack of reliable prediction programs for SREs, future publications identifying novel SREBP target genes may utilise our custom matrix. Again, this method could be applied to other transcription factors in predicting their binding sites, and therefore identifying novel target genes.

7.2.5 Other modes of transcriptional regulation of DHCR24?

Bile acids remain one of the byproducts of cholesterol untested in the feedback regulation of DHCR24. Bile acids form a negative feedback loop, activating the nuclear hormone receptor, FXR, which mediates feedback suppression of bile acid synthesis

161

[250]. DHCR24 has already demonstrated responsiveness to other nuclear hormone receptors, PXR and CAR (Section 1.8.1); therefore, it would be interesting to examine the effects of FXR on DHCR24 expression. The recent discovery of micro-RNAs (miR) and their role in cholesterol homeostasis has redefined our understanding of some of the feedback mechanisms. MiR-33 is located within intron 6 of SREBP-2, and complements the transcription factor, by strongly down-regulating expression of cholesterol exporters, ABCA1 and ABCG1 [251]. MiR-122 is involved in the regulation of cholesterol synthesis genes, in particular, phosphomevalonate kinase and HMGCR [252, 253], with evidence suggesting that DHCR24 is also affected [252]. Putative binding sites for miR-192 have been located in DHCR24 [254]. Therefore, it is plausible that DHCR24 is post-transcriptionally regulated by miRs, and future work should examine the effect of miR-122 and -192 on DHCR24 expression.

7.2.6 Implications for cardiovascular disease

This thesis has confirmed and extended DHCR24 as an SREBP-2 target gene, based on the observation that high and low sterol status decrease and increase DHCR24 expression, respectively. A recent study by Spann et al. showed that DHCR24 was the most suppressed transcript among the entire set of cholesterogenic transcripts evaluated [73]. This was associated with accumulation of desmosterol, which is particularly surprising because the entire cholesterol synthesis pathway should be shut down. Indeed, squalene epoxidase, whose gene product acts upstream of DHCR24, was down-regulated to almost the same extent as DHCR24. This invites the question as to whether or not the desmosterol accumulation can be solely ascribed to loss of DHCR24 transcript, or if other post-transcriptional modes of regulation may also be involved. For example, we found that side-chain oxysterols inhibit DHCR24 activity (Chapter 5), and they also accumulated along with desmosterol. However, desmosterol still accumulated in macrophages from mice with genetic ablation of three major enzymes forming side-chain oxysterols [73]. Therefore, the mode of inhibition is as yet uncharacterised, but may be at least partially explained with the two major findings that DHCR24 is inhibited by the distinct oxysterol, 24,25EC (Chapter 5), and by phosphorylation (Chapter 6).

162

7.3 DHCR24 activity

7.3.1 DHCR24 may be a control point in the mevalonate pathway

The introduction of the epoxide group across C-24,25 in 24,25EC saturates the double-bond which would normally be acted upon by DHCR24. This means that 24,25EC synthesis is independent of DHCR24, making 24,25EC-mediated inhibition of this enzyme an apt control point to shut down cholesterol synthesis. In addition, desmosterol, which accumulates when DHCR24 is inhibited, is more readily effluxed from cells than cholesterol [255, 256] and, like 24,25EC, can serve as a ligand for LXR [230]. An intermediate reaction that DHCR24 catalyses is conversion of lanosterol to 24,25-dihydrolanosterol, which is believed to be a major physiological degradation signal for HMGCR [30, 257]. Similarly, desmosterol is a physiological degradation signal for SM, a second control point in cholesterol synthesis [13]. The majority of our experiments were performed in CHO cells, which are commonly used for studies on cholesterol homeostasis. However, this effect was also evident in a number of human cell-lines, which is to be expected since 24,25EC is synthesised in all cells that produce cholesterol. Therefore, our finding that 24,25EC modulates DHCR24 activity has relevance for various tissues and disease states. For example, our laboratory has previously found that primary human brain cells can produce 24,25EC [258], and therefore this may be important for Alzheimer’s disease where DHCR24 is believed to play a significant role [52, 60]. Progesterone, the major gestational steroid hormone can also inhibit DHCR24 activity. In cultured cells, progesterone and other similar progestins inhibit cholesterol synthesis, accompanied by accumulation of desmosterol [104-106]. We have confirmed that this effect is due to potent inhibition of DHCR24 [39], and since progesterone synthesis declines during brain maturation [39], this may explain the rapid decrease in desmosterol levels from neuronal development to maturation [259]. As desmosterol is a superior substrate to cholesterol for progesterone formation [260-262], suppression of DHCR24 would ensure retention of sterols required for brain development [39].

163

7.3.2 Predicted regulatory mechanisms for DHCR24

Together, the structural similarity of known inhibitors of DHCR24 suggests the necessity of a side-chain, mostly bulky and aliphatic, for competitive inhibition. Based on these observations, we predict that other large classes of steroid hormones, such as the corticosteroids, mineralocorticoids and glucocorticoids (C21), also have the potential to inhibit cholesterol synthesis at DHCR24 (Appendix 8.6). This hypothesis is supported by the observation that the smaller steroid hormone, androstenedione (C19), derived from progesterone, does not inhibit DHCR24 (Appendix 8.7). However, androgens (C19) and estrogens (C18) do play a role in the transcriptional regulation of DHCR24, as discussed in Section 1.8.1. The findings from this thesis, combined with external studies on the regulation of DHCR24 indicate that this critical enzyme may serve as a novel control point in cholesterol synthesis, acting as a sensor and controlling flux at a final step in the pathway. In addition, DHCR24 may act as a control point for steroidogenesis, with complicated feedback (progesterone, potentially other C21 steroid hormones; Appendix

8.6) and feed-forward (sex steroids, smaller than C21) mechanisms, encompassing both transcriptional and post-translational regulation. Future work could investigate untested steroid hormones for their effects on the regulation of DHCR24, including mineralocorticoids and glucocorticoids (C21) on DHCR24 activity, and smaller steroid hormones on DHCR24 transcription. Regulatory mechanisms by steroid hormones may also be indirect; for example, it is known that testosterone and other androgens regulate DHCR24 via feedback on SREBP [263].

7.3.3 Development of an activity assay

To determine specific effects on DHCR24 activity, a cell based assay was performed using metabolic labelling. At the time this study was performed, there was only one commercially available radiolabelled desmosterol for use in an activity assay: [3H]-desmosterol. This would have been ideal, however, the purity of the product was questionable, with the batch sent to us turning out after testing to be [3H]-cholesterol. Therefore, we switched to an alternative label: [14C]-acetate, precursor to the mevalonate pathway. Whilst this was effective in demonstrating alterations in the relative flux through to desmosterol and cholesterol, it was not a direct measurement of DHCR24 activity, as the [14C] label passed through multiple intermediates before 164

reaching DHCR24 at the distal end of the pathway. Therefore, we attempted to generate [14C]-desmosterol in house, by labelling J774 cells (lacking DHCR24) with [14C]-acetate, separating lipids, and extracting [14C]-desmosterol. Although successful, we could not use the [14C]-desmosterol as the specific activity was extremely low. Thus, for Chapter 5, we developed an alternative method for measuring DHCR24 activity in cell culture: stably expressing human DHCR24 in hamster (CHO-7) cells, combined with [14C] metabolic labelling. The inhibitory effects of 24,25EC and progesterone were demonstrated by the decrease in cholesterol, and increase in desmosterol synthesis. Subsequent to this study, we attempted a DHCR24 activity assay using a yeast overxpression system [77]. Yeast do not contain DHCR24 nor synthesise cholesterol, thus making it a theoretically perfect system for stably overexpressing human DHCR24. Non-labelled desmosterol may be used, and desmosterol and cholesterol levels would be determined by GC-MS. Whilst we were able to express DHCR24 in yeast, we did not observe conversion of desmosterol into cholesterol. This was not due to desmosterol delivery, as we experimented with multiple detergent deliveries, as well as using desmosterol/CD. Thus, we concluded that the ectopic DHCR24 was not active, potentially due to misfolding. In Chapter 6, we required an assay that measured mutated DHCR24. To selectively examine the activity of stably expressed human DHCR24, hamster DHCR24 was knocked down by hamster-specific DHCR24 siRNA. Hence, this assay measured cholesterol conversion from desmosterol by ectopic DHCR24 only. 2 At this time, [ H6]-desmosterol became commercially available. Thus, we were 2 able to set up a cell based assay, and measure conversion of [ H6]-desmosterol to 2 [ H6]-cholesterol by GC-MS. This bypassed the previous issues of specificity, and allowed direct measurement of ectopic DHCR24 (WT, or mutated) activity.

165

7.4 Post-translational modifications in DHCR24 regulation

7.4.1 Known phosphorylation sites in DHCR24

Global proteomic studies have identified various post-translational modifications on DHCR24, but no studies to date have investigated the functional consequences of these modifications. This is the first work to demonstrate the importance of these modified residues on DHCR24 function and regulation. Whilst this study focused on one known phosphoresidue in DHCR24 (T110), there are five other known phosphorylation sites (Table 7.1). The kinase responsible for T110 phosphorylation remain elusive and requires further investigation. We have excluded the effects of PKA, PKB, and PKC, as well as ERK1/2, PI3K, mTOR, JAK2, SHP1/2, PTP1B (data not shown). Whilst this thesis did not look into the role of phosphatases, there is evidence for their involvement, as the T110 phosphorylation site was identified by treating cells with calyculin A, a specific serine/threonine phosphatase inhibitor of PP1 and PP2A [110]. Future studies on T110 should elucidate the kinase and/or phosphatase involved, and also determine if phosphorylation at the other known sites in DHCR24 have functional consequences.

166

Table 7.1. Known phosphorylation sites in DHCR24. Known phosphoresidues in human DHCR24 from literature and large-scale curated proteomics studies (Cell Signaling Technology, CST) (PhosphoSitePlus) [161]. Predicted kinases and scores from Scansite 2.0 [235], where <0.2% = high stringency, 0.2-1% = medium stringency, and 1-5% = low stringency. PKCζ, protein kinase C zeta; AblSH2, C-Abl oncogene 1, non-receptor tyrosine kinase; Fgr kinase, gardner-rasheed feline sarcoma viral (V-Fgr) oncogene homolog; PDGFR kinase, platelet-derived growth factor receptor kinase; PLCγ N-terminal SH2, phospholipase C, gamma 1 N-terminal; Lck, lymphocyte-specific protein tyrosine kinase; Src, V-Src sarcoma (Schmidt-Ruppin A-2) viral oncogene homolog (avian); Shc SH2, SHC (Src Homology 2 Domain Containing) transforming protein. Site PhosphoSitePlus Scansite predictions Literature CST Kinase Percentile Reference References T110 [110] 1 PKCζ 4.826 Y299 - 1 Abl SH2 4.664 Y300 - 1 - - Fgr kinase 4.620

Y321 [111] - PDGFR kinase 0.884 PLCγ N-terminal SH2 2.951 PKC α/β/γ 1.683 T373 - 1 PKCζ 1.167 Lck 2.591 Src SH2 3.277 Y507 - 12 Shc SH2 3.943 Src kinase 2.067

7.4.2 The involvement of PKC

This work relied on the use of pharmacological inhibitors to identify PKC as a kinase involved in DHCR24 regulation. A potential caveat with pharmacological inhibitors is the issue of specificity [264]. For example, BIM and Ro-318220 have been shown to exhibit comparable inhibition of 90 kDa and 70 kDa ribosomal S6 kinase and PKC, in vitro, although this did not translate into cell culture [265]. However, BIM and

167

Ro-318220 are relatively specific inhibitors of the PKC family [266], with BIM reducing DHCR24 activity in a dose-dependent manner, from concentrations as low as 0.1 µM, which suggest the involvement of PKC in moderating DHCR24 activity. We tested the highest scoring PKC isoform for DHCR24 as predicted by Scansite, PKCε, for two putative phosphoresidues: T88 and S334 [235]. However, PKCε inhibition had no effect on DHCR24 activity, ruling out the involvement of this isoform. Further experiments should determine which of the remaining untested PKC isoforms is involved in modulating DHCR24 activity, using more directed inhibitors. Lastly, identification of the residue phosphorylated by PKC should be determined; a primary candidate is T373 as it is the only remaining known threonine phosphorylation site [161], and it is predicted to be a PKC substrate [235]. Also, the two highest scoring putative PKC phosphorylation sites, T88 and S334 [235], should also be examined, in a similar fashion to T110 in this study.

7.4.3 Ubiquitination and the possibility of regulated turnover

Although this thesis focused on potential post-translational modification by phosphorylation, there is a vast amount of evidence that DHCR24 is ubiquitinated. At present, there are 11 identified ubiquitination sites (9 published [107-109, 267], and 2 from PhosphoSitePlus [161]), suggesting that DHCR24 may be regulated by proteasomal degradation. Furthermore, another study has characterised the binding of the E3 ubiquitin ligase, Mdm2, to DHCR24 under conditions of oncogenic and oxidative stress [58]. Indeed, more ubiquitination sites have been identified and are predicted than phosphorylation sites, similar to other cholesterogenic proteins [17], indicating the potential for regulated degradation of DHCR24, as well as other steps in the synthetic pathway besides HMGCR. However, we found that DHCR24 is extremely stable, and is not prone to proteasomal degradation (data not shown). Future work should focus on the role of regulated degradation under certain conditions; oxidative and oncogenic stress and cholesterol status, looking particularly at these known ubiquitination sites.

168

7.5 Limitations of in silico prediction programs

This thesis relied heavily on the use of in silico programs, to predict DNA and protein binding sites, as well as transmembrane domains. As detailed below, the success of these predictions was minimal. We initially used Matinspector, a matrix-based prediction program for the identification of putative SREs within DHCR24. Three sites were predicted in the promoter, which were not functional. These predicted SREs were not located in the proximal promoter, unlike all known SREs; thus, we inferred that Matinspector did not integrate positional information, only sequence information for predicting binding sites. Furthermore, the statistical basis for SRE predictions was extremely low, based on only 8 known SREs. Recognising this major deficit of Matinspector, we created our own custom matrix for the prediction of SREs. By researching all SREBP-2 target genes that had their SREs characterised and experimentally validated, we improved the statistical basis of the prediction (based on 20 known SREs). Using this custom matrix, we could then identify numerous putative SREs in the proximal promoter of DHCR24, of which two were functional. Other DNA binding proteins examined in this thesis were NF-Y and Sp1, which bind to CAATT (strict consensus) and GC (slight variability) boxes, respectively. SREs by comparison are extremely variable, with only one position in the recognition sequence strictly required, based on our research. This variability is reflected in the statistical basis for Matinspector predictions: extremely high for Sp1 and NF-Y binding sites. Thus, this program was utilised for these predictions, which found two functional NF-Y binding sites. A matrix based program, Scansite, was also used to predict phosphorylation sites on DHCR24. Since PKC phosphorylation was predicted for both experimentally validated and putative sites in DHCR24; we focused on one published site, T110 (Table 7.1). Although we saw functional effects on DHCR24 when PKC was inhibited using a general inhibitor (BIM), this was not through T110. We then focused on elucidating the PKC isoform responsible based on the two highest scoring PKC sites in Scansite, PKCε. Using a specific PKCε inhibitor, we could not reproduce the inhibitory effects observed when using the general PKC inhibitor, BIM. Thus, the predictions provided by Scansite were not informative in this study, neither supporting our experimental findings, nor as

169

a lead for the identification of specific kinases involved in DHCR24 post-translational modifications. Lastly, we utilised programs developed to predict membrane topology of proteins based on multiple parameters. SignalP identified a putative signal peptide in DHCR24; however, our experimental findings did not validate this. Although we have speculated on the potential for other atypical signal sequences in DHCR24 (Section 7.1.1), there is at present no bioinformatic approach to identifying putative signals. Lastly, we used multiple programs to predict TMDs in DHCR24, including TMHMM and Prodiv-TMHMM, hidden Markov models (HMM), which also use evolutionary information, known to increase the success rate of predicting TMDs [150]. Again, our experimental findings did not accord with the predictions from these programs, appearing to associate with the ER membrane by re-entrant loops, and not TMDs. There may be precedence for a re-entrant model of DHCR24, as recent studies predict that >10% of membrane proteins contain re-entrant loops [160]. However, although there are published bioinformatics models for prediction of re-entrant loops, they are not widely accessible to biochemists at present. Future application of such programs may be more successful in predicting membrane associated regions in DHCR24, as we predict it associated with the membrane by multiple re-entrant loops. Overall, this thesis cautions the use of bioinformatic approaches in biochemical studies, and emphasises the need to experimentally validate any predictions.

7.6 Future directions

In researching novel regulatory mechanisms for DHCR24, we have highlighted a lack of information on regulation and basic structural research on many cholesterogenic enzymes, with a particular focus on post-lanosterol enzymes. Future work should focus on addressing the questions raised and extending the findings to other cholesterogenic enzymes.

7.6.1 Further DHCR24 characterisation

Additional gaps in our knowledge include a detailed understanding of the domains and overall structure of DHCR24, its comprehensive interactome, and how these translate to

170

function. This could include a cholesterogenic enzyme complexe (cholestesome); looking at how DHCR24 interacts with other cholesterogenic enzymes and sterols, as well as a general interactome, looking at interactions with non-cholesterogenic enzymes, which would be insightful for its role in other cellular processes and diseases. In light of our novel findings on DHCR24 membrane topology, many questions arise: how important is caspase processing of DHCR24 and its translocation to non-ER compartments in various physiological and disease states? Also, if the desmosterol reductase function is not required for all of the roles ascribed to DHCR24 (e.g. some of its antioxidant properties), what other mechanism(s) of action are involved, and how generalisable are these in different cellular contexts?

7.6.2 Key players in the post-lanosterol pathway

This thesis highlights novel research on DHCR24, with a focus on its regulation. This information may shed light on potential regulatory mechanisms for DHCR7, which catalyses cholesterol synthesis in the alternate Kandutsch–Russell pathway. For example, if side-chain oxysterols feedback on DHCR24, would ring-oxygenated sterols affect DHCR7? This raises the intriguing question, are the final steps of both the Bloch and Kandutsch–Russell pathways equally strong regulatory steps? Future investigations into DHCR7 regulation will be needed for a fuller understanding of how flux through the entire cholesterol synthesis pathway is controlled. We describe DHCR24 primarily in the context of the Bloch pathway; however, this enzyme can also act upon lanosterol and control entry into the alternate Kandutsch-Russell pathway. CYP51 is a crucial enzyme in cholesterol synthesis [268], and competes with DHCR24 for lanosterol and therefore controls entry into the Bloch pathway. It may well be that in different tissues and/or under different physiological conditions, DHCR24 and CYP51 are differentially regulated, determining which pathway predominates. Future investigations should look into the differential regulation of these three important enzymes that control entry and exit into the post-lanosterol leg of the mevalonate pathway.

171

Figure 7.2. Key enzymes in the Bloch and Kandutsch-Russell pathways. Post-lanosterol biosynthesis of cholesterol occurs via two alternate pathways. In the Bloch pathway (in red), lanosterol is converted into FF-MAS by CYP51 in the first step, and desmosterol is converted into cholesterol by DHCR24 in the final step. In the Kandutsch-Russell pathway (in blue), lanosterol is converted to 24,25-dihydrolanosterol by DHCR24 in the first step, and 7-dehydrocholesterol is converted into cholesterol by DHCR7 in the final step. Adapted from our recent review [46].

172

7.6.3 Epigenetic regulation of DHCR24

The proximal promoter of DHCR24 lies within a CpG island, important in the epigenetic regulation of DHCR24 [83]. This is interesting, as many important regulatory sites lie within this region, of particular interest to this thesis, the sterol responsive region of DHCR24 (two SREs and and two NF-Y binding sites; Chapter 4). The high GC content as well as the repetitive nature of the DHCR24 promoter proved to be problematic, with cloning of this genomic region and subsequent mutagenesis of putative response elements requiring lengthy optimisation. This was eventually overcome through the development of PCR based cloning techniques [269], which involved optimising annealing temperatures, buffers and buffer components, such as dimethyl sulfoxide (DMSO). Epigenetic factors in DHCR24 regulation may explain why DHCR24 is differentially expressed in some cancers. Reduced expression of DHCR24 in adrenal cortical carcinoma, compared to adenomas, is inversely related to the methylation status of the promoter [270]. Similarly, epigenetic regulation may explain reduced DHCR24 levels upon rat parvoviral induced tumour suppression [271]. Future work should elaborate on these initial intriguing findings, and aim to identify signals for epigenetic regulation of DHCR24, and their role in the progression to various disease states.

7.7 Concluding remarks

Our findings highlight that DHCR24 is an apt control point for cholesterol levels, with regulation at multiple levels having potent effects on de novo cholesterol synthesis. In addition, this new control point is able to integrate external signals from other pathways. Future work on DHCR24 should continue exploring the leads from this work, with a particular focus on the potential for many avenues of post-translational modifications in acute regulation of DHCR24. This thesis has focused on one neglected enzyme in the cholesterol synthetic pathway; however, many more steps remain uncharacterised. As recently highlighted [17], many of these steps have the potential for potent regulation, and therefore there may be other control points within the pathway, regulated by feedback signals from

173

cholesterol synthesis, as well as from other cellular pathways. Future work should investigate these enzymatic steps using similar methods as presented here.

174

Chapter 8

Appendix

175

176

8 APPENDICES

Appendix 8.1. The DHCR24 protein and its homologs are ubiquitous across the domains of life. DHCR24 and homolog protein sequences from 43 taxa were retrieved from the NCBI and Ensembl protein databases, and then aligned using ClustalW with poorly aligned and divergent regions removed using Gblocks. A rooted phylogenetic tree was then produced using the Neighbour-Joining method and the Jukes-Cantor substitution model. The Mycobacterium sequence was used as an outgroup and the statistical support for nodes are shown as the proportion of 100 bootstrap replicates. The scale bar indicates the number of amino acid substitutions per site. Accession numbers for sequences are in Appendix 8.5.

177

178

Appendix 8.2. Alignment of DHCR24 with homologs. Amino acid sequence of human DHCR24 aligned with homologs from mouse, chicken, turkey, Chinese softshell turtle, anole lizard, zebrafish, chichlid fish, C. elegans (DIMINUTO), tomato, and A. thaliana (DWARF1) homologs. Accession numbers for sequences are in Appendix 8.5.

Appendix 8.3. The major enzymatic reactions in the post-lanosterol cholesterol synthesis pathway. Reaction Protein Symbol Enzyme name Uniprot ID EC Number (Uniprot) Gene Symbol* C14-demethylase CYP51 lanosterol-14α-demethylase Q16850 1.14.13.70 DHCR14, ∆14-reductase Δ14-sterol reductase/Transmembrane 7 superfamily member 2 O76062 1.3.1.70 TM7SF2 SC4MOL, methylsterol monooxygenase 1/ C-4 methylsterol oxidase Q15800 1.14.13.72 MSMO1

sterol-4α-carboxylate 3-dehydrogenase, decarboxylating/ C4-demethylase NSDHL Q15738 1.1.1.170 NAD(P) dependent steroid dehydrogenase-like

3-keto-steroid reductase HSD17B7 P56937 1.1.1.270 hydroxysteroid (17-beta) dehydrogenase 7 179 8 7

∆8∆7-isomerase EBP emopamil binding protein/ 3β-hydroxysteroid-Δ ,Δ -isomerase Q15125 5.3.3.5

∆5-desaturase SC5D sterol-C5-desaturase/ lathosterol oxidase O75845 1.14.21.6

∆7-reductase DHCR7 7-dehydrocholesterol reductase Q9UBM7 1.3.1.21

∆24-reductase DHCR24 Δ24-sterol reductase/3β-hydroxysterol Δ24-reductase Q15392 1.3.1.72

Appendix 8.4. The major sterol intermediates formed in the post-lanosterol cholesterol synthesis pathway. Common Name Systematic Name

lanosterol 4,4,14α-trimethyl-5α-cholesta-8(9),24-dien-3β-ol follicular fluid meiosis-activating sterol 4,4-dimethyl-5α-cholesta-8(9),14,24-trien-3β-ol (FF-MAS) testis meiosis-activating sterol 4,4-dimethyl-5α-cholesta-8(9),24-dien-3β-ol (T-MAS) zymosterol 5α-cholesta-8(9),24-dien-3β-ol 24-dehydrolathosterol 5α-cholesta-7,24-dien-3β-ol 7-dehydrodesmosterol 5α-cholesta-5,7,24-trien-3β-ol desmosterol 5α-cholesta-5,24-dien-3β-ol 180 24,25-dihydrolanosterol 4,4,14α-trimethyl-5α-cholest-8(9)en-3β-ol

dihydro-FF-MAS 4,4-dimethyl-5α-cholesta-8(9),14-dien-3β-ol dihydro-T-MAS 4,4-dimethyl-5α-cholest-8(9)-en-3β-ol zymostenol 5α-cholest-8(9)-en-3β-ol lathosterol 5α-cholest-7-en-3β-ol 7-dehydrocholesterol 5α-cholesta-5,7-dien-3β-ol cholesterol 5α-cholest-5-en-3β-ol

Appendix 8.5. DHCR24 homologs and their accession numbers. Organism Accession Number A. thaliana (DWARF1) Q39085.2 Anole lizard ENSACAP00000000133 C. elegans (DIMINUTO) F52H2.6 Chicken ENSGALP00000017539 Chimpanzee ENSPTRP00000001349 Chinese softshell turtle ENSPSIP00000003579 Cichlid fish ENSONIP00000003546 Cod ENSGMOP00000020672 Coelacanth ENSLACP00000007383 Corn NP_001105560.1 Cow ENSBTAP00000006153 Dog ENSCAFP00000028027 Elephant ENSLAFP00000004042 Ferret ENSMPUP00000007037 Fugu ENSTRUP00000041669 Gorilla ENSGGOP00000010590 Guinea pig ENSCPOP00000014098 Horse ENSECAP00000015780 Human ENSP00000360316 Lamprey ENSPMAP00000004155 Macaque ENSMMUP00000028412 M. oryzae XP_003716955.1 Marmoset ENSCJAP00000028837 Megabat ENSPVAP00000004171 Microbat ENSMLUP00000015411 Mouse ENSMUSP00000038063 Mycobacterium NP_218236.1 Orangutan ENSPPYP00000001525 Panda ENSAMEP00000010462 Pig ENSSSCP00000025201 Platyfish ENSXMAP00000011120 Rabbit ENSOCUP00000004499 Rat ENSRNOP00000009402 Squirrel ENSSTOP00000003642 Stickleback ENSGACP00000014741 Tetraodon ENSTNIP00000020771 Tomato NP_001234550.1 Turkey ENSMGAP00000011950

181

182

Appendix 8.6. A comparison of known and putative DHCR24 inhibitors. Inhibitors of DHCR24 (yellow); side-chain oxysterols, C-22 unsaturated phytosterols, and progesterone. Putative inhibitors (blue): mineralcorticosteroids, glucocorticosteroids, and progestins.

Appendix 8.7. Progesterone, and not androstenedione, inhibit DHCR24. CHO-7 cells were pre-treated with mevalonate (50 M) and compactin (5 M) overnight then treated with progesterone or androstenedione (0.025-2.5 M) for 4 h. Lipid extracts were separated by Arg-TLC, and bands corresponding to cholesterol and desmosterol were visualised by phosphorimager. Phosphorimages are representative of n=1 experiment.

183

184

Chapter 9

References

185

186

9 REFERENCES

[1] R.E. Olson, Discovery of the lipoproteins, their role in fat transport and their significance as risk factors, The Journal of nutrition, 128 (1998) 439S-443S. [2] D. Lingwood, K. Simons, Lipid rafts as a membrane-organizing principle, Science, 327 (2010) 46-50. [3] H.R. Waterham, Inherited disorders of cholesterol biosynthesis, Clin Genet, 61 (2002) 393-403. [4] L. Puglielli, R.E. Tanzi, D.M. Kovacs, Alzheimer's disease: the cholesterol connection, Nat Neurosci, 6 (2003) 345-351. [5] F.R. Maxfield, I. Tabas, Role of cholesterol and lipid organization in disease, Nature, 438 (2005) 612-621. [6] J. Ye, Reliance of host cholesterol metabolic pathways for the life cycle of hepatitis C virus, PLoS pathogens, 3 (2007) e108. [7] A.J. Brown, Cholesterol, statins and cancer, Clin Exp Pharmacol Physiol, 34 (2007) 135-141. [8] J.L. Goldstein, R.A. DeBose-Boyd, M.S. Brown, Protein sensors for membrane sterols, Cell, 124 (2006) 35-46. [9] J.R. Krycer, A.J. Brown, Cholesterol accumulation in prostate cancer: a classic observation from a modern perspective, Biochimica et biophysica acta, 1835 (2013) 219-229. [10] M.D. Siperstein, V.M. Fagan, Feedback control of mevalonate synthesis by dietary cholesterol, The Journal of biological chemistry, 241 (1966) 602-609. [11] I.C. Gelissen, A.J. Brown, Drug targets beyond HMG-CoA reductase: Why venture beyond the statins?, Frontiers in Biology, 6 (2011) 197-205. [12] J.L. Goldstein, M.S. Brown, Regulation of the mevalonate pathway, Nature, 343 (1990) 425-430. [13] S. Gill, J. Stevenson, I. Kristiana, A.J. Brown, Cholesterol-dependent degradation of squalene monooxygenase, a control point in cholesterol synthesis beyond HMG-CoA reductase, Cell Metab, 13 (2011) 260-273. [14] A.J. Brown, Cholesterol versus other sterols: How do they compare as physiological regulators of cholesterol homeostasis?, European Journal of Lipid Science and Technology, 114 (2012) 617-623. [15] E.J. Zerenturk, I. Kristiana, S. Gill, A.J. Brown, The endogenous regulator 24(S),25-epoxycholesterol inhibits cholesterol synthesis at DHCR24 (Seladin-1), Biochim Biophys Acta, 1821 (2012) 1269-1277. [16] J.D. Horton, N.A. Shah, J.A. Warrington, N.N. Anderson, S.W. Park, M.S. Brown, J.L. Goldstein, Combined analysis of oligonucleotide microarray data from transgenic and knockout mice identifies direct SREBP target genes, Proc Natl Acad Sci U S A, 100 (2003) 12027-12032. [17] L.J. Sharpe, A.J. Brown, Controlling Cholesterol Synthesis beyond 3-Hydroxy-3- methylglutaryl-CoA Reductase (HMGCR), The Journal of biological chemistry, 288 (2013) 18707-18715. [18] S.M. Vallett, H.B. Sanchez, J.M. Rosenfeld, T.F. Osborne, A direct role for sterol regulatory element binding protein in activation of 3-hydroxy-3-methylglutaryl coenzyme A reductase gene, The Journal of biological chemistry, 271 (1996) 12247- 12253.

187

[19] M.R. Briggs, C. Yokoyama, X. Wang, M.S. Brown, J.L. Goldstein, Nuclear protein that binds sterol regulatory element of low density lipoprotein receptor promoter. I. Identification of the protein and delineation of its target nucleotide sequence, The Journal of biological chemistry, 268 (1993) 14490-14496. [20] S.Y. Oh, S.K. Park, J.W. Kim, Y.H. Ahn, S.W. Park, K.S. Kim, Acetyl-CoA carboxylase beta gene is regulated by sterol regulatory element-binding protein-1 in liver, The Journal of biological chemistry, 278 (2003) 28410-28417. [21] I. Shimomura, H. Shimano, B.S. Korn, Y. Bashmakov, J.D. Horton, Nuclear sterol regulatory element-binding proteins activate genes responsible for the entire program of unsaturated fatty acid biosynthesis in transgenic mouse liver, The Journal of biological chemistry, 273 (1998) 35299-35306. [22] A. Radhakrishnan, Y. Ikeda, H.J. Kwon, M.S. Brown, J.L. Goldstein, Sterol- regulated transport of SREBPs from endoplasmic reticulum to Golgi: oxysterols block transport by binding to Insig, Proc Natl Acad Sci U S A, 104 (2007) 6511-6518. [23] J.D. Horton, J.L. Goldstein, M.S. Brown, SREBPs: activators of the complete program of cholesterol and fatty acid synthesis in the liver, J Clin Invest, 109 (2002) 1125-1131. [24] M.K. Bennett, T.F. Osborne, Nutrient regulation of gene expression by the sterol regulatory element binding proteins: increased recruitment of gene-specific coregulatory factors and selective hyperacetylation of histone H3 in vivo, Proceedings of the National Academy of Sciences of the United States of America, 97 (2000) 6340- 6344. [25] A. Radhakrishnan, L.P. Sun, H.J. Kwon, M.S. Brown, J.L. Goldstein, Direct binding of cholesterol to the purified membrane region of SCAP: mechanism for a sterol-sensing domain, Mol Cell, 15 (2004) 259-268. [26] P. Costet, Y. Luo, N. Wang, A.R. Tall, Sterol-dependent transactivation of the ABC1 promoter by the liver X receptor/retinoid X receptor, The Journal of biological chemistry, 275 (2000) 28240-28245. [27] N. Zelcer, P. Tontonoz, Liver X receptors as integrators of metabolic and inflammatory signaling, J Clin Invest, 116 (2006) 607-614. [28] L.J. Sharpe, W. Luu, A.J. Brown, Akt phosphorylates Sec24: new clues into the regulation of ER-to-Golgi trafficking, Traffic, 12 (2011) 19-27. [29] R.A. DeBose-Boyd, Feedback regulation of cholesterol synthesis: sterol- accelerated ubiquitination and degradation of HMG CoA reductase, Cell Res, 18 (2008) 609-621. [30] B.L. Song, N.B. Javitt, R.A. DeBose-Boyd, Insig-mediated degradation of HMG CoA reductase stimulated by lanosterol, an intermediate in the synthesis of cholesterol, Cell Metab, 1 (2005) 179-189. [31] N. Sever, B.L. Song, D. Yabe, J.L. Goldstein, M.S. Brown, R.A. DeBose-Boyd, Insig-dependent ubiquitination and degradation of mammalian 3-hydroxy-3- methylglutaryl-CoA reductase stimulated by sterols and geranylgeraniol, The Journal of biological chemistry, 278 (2003) 52479-52490. [32] R.V. Omkumar, B.G. Darnay, V.W. Rodwell, Modulation of Syrian hamster 3- hydroxy-3-methylglutaryl-CoA reductase activity by phosphorylation. Role of serine 871, The Journal of biological chemistry, 269 (1994) 6810-6814. [33] R. Sato, J.L. Goldstein, M.S. Brown, Replacement of serine-871 of hamster 3- hydroxy-3-methylglutaryl-CoA reductase prevents phosphorylation by AMP-activated kinase and blocks inhibition of sterol synthesis induced by ATP depletion, Proc Natl Acad Sci U S A, 90 (1993) 9261-9265.

188

[34] J.A. Friesen, V.W. Rodwell, The 3-hydroxy-3-methylglutaryl coenzyme-A (HMG- CoA) reductases, Genome biology, 5 (2004) 248. [35] J.S. Burg, P.J. Espenshade, Regulation of HMG-CoA reductase in mammals and yeast, Progress in lipid research, 50 (2011) 403-410. [36] S.H. Bae, Y.K. Paik, Cholesterol biosynthesis from lanosterol: development of a novel assay method and characterization of rat liver microsomal lanosterol delta 24- reductase, Biochem J, 326 ( Pt 2) (1997) 609-616. [37] I. Tabas, S.J. Feinmark, N. Beatini, The reactivity of desmosterol and other shellfish- and xanthomatosis-associated sterols in the macrophage sterol esterification reaction, J Clin Invest, 84 (1989) 1713-1721. [38] C. Fernandez, Y. Suarez, A.J. Ferruelo, D. Gomez-Coronado, M.A. Lasuncion, Inhibition of cholesterol biosynthesis by Delta22-unsaturated phytosterols via competitive inhibition of sterol Delta24-reductase in mammalian cells, Biochem J, 366 (2002) 109-119. [39] M. Jansen, W. Wang, D. Greco, G.C. Bellenchi, U. di Porzio, A.J. Brown, E. Ikonen, What dictates the accumulation of desmosterol in the developing brain?, FASEB journal : official publication of the Federation of American Societies for Experimental Biology, 27 (2013) 865-870. [40] D.R. FitzPatrick, J.W. Keeling, M.J. Evans, A.E. Kan, J.E. Bell, M.E. Porteous, K. Mills, R.M. Winter, P.T. Clayton, Clinical phenotype of desmosterolosis, American journal of medical genetics, 75 (1998) 145-152. [41] H.C. Andersson, L. Kratz, R. Kelley, Desmosterolosis presenting with multiple congenital anomalies and profound developmental delay, American journal of medical genetics, 113 (2002) 315-319. [42] C.P. Schaaf, J. Koster, P. Katsonis, L. Kratz, O.A. Shchelochkov, F. Scaglia, R.I. Kelley, O. Lichtarge, H.R. Waterham, M. Shinawi, Desmosterolosis-phenotypic and molecular characterization of a third case and review of the literature, Am J Med Genet A, 155A (2011) 1597-1604. [43] J. Zolotushko, H. Flusser, B. Markus, I. Shelef, Y. Langer, M. Heverin, I. Bjorkhem, S. Sivan, O.S. Birk, The desmosterolosis phenotype: spasticity, microcephaly and micrognathia with agenesis of corpus callosum and loss of white matter, Eur J Hum Genet, 19 (2011) 942-946. [44] A. Pedretti, E. Bocci, R. Maggi, G. Vistoli, Homology modelling of human DHCR24 (seladin-1) and analysis of its binding properties through molecular docking and dynamics simulations, Steroids, 73 (2008) 708-719. [45] J. Acimovic, D. Rozman, Steroidal triterpenes of cholesterol synthesis, Molecules, 18 (2013) 4002-4017. [46] E.J. Zerenturk, L.J. Sharpe, E. Ikonen, A.J. Brown, Desmosterol and DHCR24: Unexpected new directions for a terminal step in cholesterol synthesis, Progress in lipid research, 52 (2013) 666-680. [47] A. Crameri, E. Biondi, K. Kuehnle, D. Lutjohann, K.M. Thelen, S. Perga, C.G. Dotti, R.M. Nitsch, M.D. Ledesma, M.H. Mohajeri, The role of seladin-1/DHCR24 in cholesterol biosynthesis, APP processing and Abeta generation in vivo, EMBO J, 25 (2006) 432-443. [48] S.D. Gilk, D.C. Cockrell, C. Luterbach, B. Hansen, L.A. Knodler, J.A. Ibarra, O. Steele-Mortimer, R.A. Heinzen, Bacterial colonization of host cells in the absence of cholesterol, PLoS Pathog, 9 (2013) e1003107. [49] X. Lu, F. Kambe, X. Cao, T. Yoshida, S. Ohmori, K. Murakami, T. Kaji, T. Ishii, D. Zadworny, H. Seo, DHCR24-knockout embryonic fibroblasts are susceptible to

189

serum withdrawal-induced apoptosis because of dysfunction of caveolae and insulin- Akt-Bad signaling, Endocrinology, 147 (2006) 3123-3132. [50] S. Vainio, M. Jansen, M. Koivusalo, T. Rog, M. Karttunen, I. Vattulainen, E. Ikonen, Significance of sterol structural specificity. Desmosterol cannot replace cholesterol in lipid rafts, The Journal of biological chemistry, 281 (2006) 348-355. [51] J. Sánchez-Wandelmer, A. Dávalos, E. Herrera, M. Giera, S. Cano, G. de la Peña, M.A. Lasunción, R. Busto, Inhibition of cholesterol biosynthesis disrupts lipid raft/caveolae and affects insulin receptor activation in 3T3-L1 preadipocytes, Biochimica et Biophysica Acta (BBA)-Biomembranes, 1788 (2009) 1731-1739. [52] I. Greeve, I. Hermans-Borgmeyer, C. Brellinger, D. Kasper, T. Gomez-Isla, C. Behl, B. Levkau, R.M. Nitsch, The human DIMINUTO/DWARF1 homolog seladin-1 confers resistance to Alzheimer's disease-associated neurodegeneration and oxidative stress, J Neurosci, 20 (2000) 7345-7352. [53] D. Di Stasi, V. Vallacchi, V. Campi, T. Ranzani, M. Daniotti, E. Chiodini, S. Fiorentini, I. Greeve, A. Prinetti, L. Rivoltini, M.A. Pierotti, M. Rodolfo, DHCR24 gene expression is upregulated in melanoma metastases and associated to resistance to oxidative stress-induced apoptosis, Int J Cancer, 115 (2005) 224-230. [54] K. Kuehnle, A. Crameri, R.E. Kalin, P. Luciani, S. Benvenuti, A. Peri, F. Ratti, M. Rodolfo, L. Kulic, F.L. Heppner, R.M. Nitsch, M.H. Mohajeri, Prosurvival effect of DHCR24/Seladin-1 in acute and chronic responses to oxidative stress, Mol Cell Biol, 28 (2008) 539-550. [55] T. Nishimura, M. Kohara, K. Izumi, Y. Kasama, Y. Hirata, Y. Huang, M. Shuda, C. Mukaidani, T. Takano, Y. Tokunaga, H. Nuriya, M. Satoh, M. Saito, C. Kai, K. Tsukiyama-Kohara, Hepatitis C virus impairs p53 via persistent overexpression of 3beta-hydroxysterol Delta24-reductase, The Journal of biological chemistry, 284 (2009) 36442-36452. [56] R. Mirza, S. Qiao, K. Tateyama, T. Miyamoto, L. Xiuli, H. Seo, 3beta- Hydroxysterol-Delta24 reductase plays an important role in long bone growth by protecting chondrocytes from reactive oxygen species, J Bone Miner Metab, 30 (2012) 144-153. [57] X. Lu, F. Kambe, X. Cao, Y. Kozaki, T. Kaji, T. Ishii, H. Seo, 3beta- Hydroxysteroid-delta24 reductase is a hydrogen peroxide scavenger, protecting cells from oxidative stress-induced apoptosis, Endocrinology, 149 (2008) 3267-3273. [58] C. Wu, I. Miloslavskaya, S. Demontis, R. Maestro, K. Galaktionov, Regulation of cellular response to oncogenic and oxidative stress by Seladin-1, Nature, 432 (2004) 640-645. [59] S. Iivonen, M. Hiltunen, I. Alafuzoff, A. Mannermaa, P. Kerokoski, J. Puolivali, A. Salminen, S. Helisalmi, H. Soininen, Seladin-1 transcription is linked to neuronal degeneration in Alzheimer's disease, Neuroscience, 113 (2002) 301-310. [60] A. Peri, M. Serio, Neuroprotective effects of the Alzheimer's disease-related gene seladin-1, J Mol Endocrinol, 41 (2008) 251-261. [61] L.J. Sharpe, J. Wong, B. Garner, G.M. Halliday, A.J. Brown, Is seladin-1 really a selective Alzheimer's disease indicator?, Journal of Alzheimer's Disease, 30 (2012) 35- 39. [62] T. Sarajarvi, A. Haapasalo, J. Viswanathan, P. Makinen, M. Laitinen, H. Soininen, M. Hiltunen, Down-regulation of seladin-1 increases BACE1 levels and activity through enhanced GGA3 depletion during apoptosis, The Journal of biological chemistry, 284 (2009) 34433-34443.

190

[63] R. Lamsa, S. Helisalmi, M. Hiltunen, S.K. Herukka, T. Tapiola, T. Pirttila, S. Vepsalainen, H. Soininen, The association study between DHCR24 polymorphisms and Alzheimer's disease, Am J Med Genet B Neuropsychiatr Genet, 144B (2007) 906-910. [64] A. Tedde, E. Cellini, S. Bagnoli, S. Sorbi, A. Peri, Mutational screening analysis of DHCR24/seladin-1 gene in Italian familial Alzheimer's disease, Am J Med Genet B Neuropsychiatr Genet, 147B (2008) 117-119. [65] A. Feher, A. Juhasz, M. Pakaski, J. Kalman, Z. Janka, Gender dependent effect of DHCR24 polymorphism on the risk for Alzheimer's disease, Neurosci Lett, 526 (2012) 20-23. [66] L. Bonaccorsi, P. Luciani, G. Nesi, E. Mannucci, C. Deledda, F. Dichiara, M. Paglierani, F. Rosati, L. Masieri, S. Serni, M. Carini, L. Proietti-Pannunzi, S. Monti, G. Forti, G. Danza, M. Serio, A. Peri, Androgen receptor regulation of the seladin- 1/DHCR24 gene: altered expression in prostate cancer, Laboratory investigation; a journal of technical methods and pathology, 88 (2008) 1049-1056. [67] M.C. Battista, M.O. Guimond, C. Roberge, A.A. Doueik, L. Fazli, M. Gleave, R. Sabbagh, N. Gallo-Payet, Inhibition of DHCR24/seladin-1 impairs cellular homeostasis in prostate cancer, Prostate, 70 (2010) 921-933. [68] S.M. Henshall, D.E. Afar, J. Hiller, L.G. Horvath, D.I. Quinn, K.K. Rasiah, K. Gish, D. Willhite, J.G. Kench, M. Gardiner-Garden, P.D. Stricker, H.I. Scher, J.J. Grygiel, D.B. Agus, D.H. Mack, R.L. Sutherland, Survival analysis of genome-wide gene expression profiles of prostate cancers identifies new prognostic targets of disease relapse, Cancer Res, 63 (2003) 4196-4203. [69] P.J. Hendriksen, N.F. Dits, K. Kokame, A. Veldhoven, W.M. van Weerden, C.H. Bangma, J. Trapman, G. Jenster, Evolution of the androgen receptor pathway during progression of prostate cancer, Cancer research, 66 (2006) 5012-5020. [70] T. Takano, K. Tsukiyama-Kohara, M. Hayashi, Y. Hirata, M. Satoh, Y. Tokunaga, C. Tateno, Y. Hayashi, T. Hishima, N. Funata, M. Sudoh, M. Kohara, Augmentation of DHCR24 expression by hepatitis C virus infection facilitates viral replication in hepatocytes, Journal of hepatology, 55 (2011) 512-521. [71] M. Saito, M. Kohara, K. Tsukiyama-Kohara, Hepatitis C virus promotes expression of the 3beta-hydroxysterol delta24-reductase through Sp1, J Med Virol, 84 (2012) 733-746. [72] K.C. McGrath, X.H. Li, R. Puranik, E.C. Liong, J.T. Tan, V.M. Dy, B.A. DiBartolo, P.J. Barter, K.A. Rye, A.K. Heather, Role of 3beta-hydroxysteroid-delta 24 reductase in mediating antiinflammatory effects of high-density lipoproteins in endothelial cells, Arteriosclerosis, thrombosis, and vascular biology, 29 (2009) 877-882. [73] N.J. Spann, L.X. Garmire, J.G. McDonald, D.S. Myers, S.B. Milne, N. Shibata, D. Reichart, J.N. Fox, I. Shaked, D. Heudobler, C.R. Raetz, E.W. Wang, S.L. Kelly, M.C. Sullards, R.C. Murphy, A.H. Merrill, Jr., H.A. Brown, E.A. Dennis, A.C. Li, K. Ley, S. Tsimikas, E. Fahy, S. Subramaniam, O. Quehenberger, D.W. Russell, C.K. Glass, Regulated accumulation of desmosterol integrates macrophage lipid metabolism and inflammatory responses, Cell, 151 (2012) 138-152. [74] U. Klahre, T. Noguchi, S. Fujioka, S. Takatsuto, T. Yokota, T. Nomura, S. Yoshida, N.H. Chua, The Arabidopsis DIMINUTO/DWARF1 gene encodes a protein involved in steroid synthesis, Plant Cell, 10 (1998) 1677-1690. [75] T. Takahashi, A. Gasch, N. Nishizawa, N.H. Chua, The DIMINUTO gene of Arabidopsis is involved in regulating cell elongation, Genes Dev, 9 (1995) 97-107.

191

[76] A.R. Mushegian, E.V. Koonin, A putative FAD-binding domain in a distinct group of oxidases including a protein involved in plant development, Protein Sci, 4 (1995) 1243-1244. [77] H.R. Waterham, J. Koster, G.J. Romeijn, R.C. Hennekam, P. Vreken, H.C. Andersson, D.R. FitzPatrick, R.I. Kelley, R.J. Wanders, Mutations in the 3beta- hydroxysterol Delta24-reductase gene cause desmosterolosis, an autosomal recessive disorder of cholesterol biosynthesis, Am J Hum Genet, 69 (2001) 685-694. [78] L.F. Ciufo, P.A. Murray, A. Thompson, D.J. Rigden, H.H. Rees, Characterisation of a desmosterol reductase involved in phytosterol dealkylation in the silkworm, Bombyx mori, PLoS One, 6 (2011) e21316. [79] M.C. Battista, C. Roberge, A. Martinez, N. Gallo-Payet, 24-dehydrocholesterol reductase/seladin-1: a key protein differentially involved in adrenocorticotropin effects observed in human and rat adrenal cortex, Endocrinology, 150 (2009) 4180-4190. [80] M.C. Battista, C. Roberge, M. Otis, N. Gallo-Payet, Seladin-1 expression in rat adrenal gland: effect of adrenocorticotropic hormone treatment, J Endocrinol, 192 (2007) 53-66. [81] X. Lu, Y. Li, J. Liu, X. Cao, X. Wang, D. Wang, H. Seo, B. Gao, The membrane topological analysis of 3β-hydroxysteroid-Δ24 reductase (DHCR24) on endoplasmic reticulum, Journal of molecular endocrinology, 48 (2012) 1-9. [82] M. Otis, M.C. Battista, M. Provencher, S. Campbell, C. Roberge, M.D. Payet, N. Gallo-Payet, From integrative signalling to metabolic disorders, J Steroid Biochem Mol Biol, 109 (2008) 224-229. [83] J. Drzewinska, A. Walczak-Drzewiecka, M. Ratajewski, Identification and analysis of the promoter region of the human DHCR24 gene: involvement of DNA methylation and histone acetylation, Mol Biol Rep, 38 (2011) 1091-1101. [84] E.J. Zerenturk, L.J. Sharpe, A.J. Brown, Sterols regulate 3beta-hydroxysterol Δ24- reductase (DHCR24) via dual sterol regulatory elements: cooperative induction of key enzymes in lipid synthesis by Sterol Regulatory Element Binding Proteins, Biochim Biophys Acta, 1821 (2012) 1350-1360. [85] N.E. Salem, M. Saito, Y. Kasama, M. Ozawa, T. Kawabata, S. Harada, H. Suda, K. Asonuma, A. El‐Gohary, K. Tsukiyama‐Kohara, Genomic polymorphisms in 3β‐ hydroxysterol Δ24‐reductase promoter sequences, Microbiology and immunology, 57 (2013) 179-184. [86] P.S. Nelson, N. Clegg, H. Arnold, C. Ferguson, M. Bonham, J. White, L. Hood, B. Lin, The program of androgen-responsive genes in neoplastic prostate epithelium, Proc Natl Acad Sci U S A, 99 (2002) 11890-11895. [87] S. Benvenuti, P. Luciani, G.B. Vannelli, S. Gelmini, E. Franceschi, M. Serio, A. Peri, Estrogen and selective estrogen receptor modulators exert neuroprotective effects and stimulate the expression of selective Alzheimer's disease indicator-1, a recently discovered antiapoptotic gene, in human neuroblast long-term cell cultures, J Clin Endocrinol Metab, 90 (2005) 1775-1782. [88] P. Luciani, C. Deledda, F. Rosati, S. Benvenuti, I. Cellai, F. Dichiara, M. Morello, G.B. Vannelli, G. Danza, M. Serio, A. Peri, Seladin-1 is a fundamental mediator of the neuroprotective effects of estrogen in human neuroblast long-term cell cultures, Endocrinology, 149 (2008) 4256-4266. [89] H. Zu, J. Wu, J. Zhang, M. Yu, Z. Hong, Testosterone up-regulates seladin-1 expression by iAR and PI3-K/Akt signaling pathway in C6 cells, Neurosci Lett, 514 (2012) 122-126.

192

[90] D. Sarkar, T. Imai, F. Kambe, A. Shibata, S. Ohmori, A. Siddiq, S. Hayasaka, H. Funahashi, H. Seo, The human homolog of Diminuto/Dwarf1 gene (hDiminuto): a novel ACTH-responsive gene overexpressed in benign cortisol-producing adrenocortical adenomas, J Clin Endocrinol Metab, 86 (2001) 5130-5137. [91] P. Luciani, P. Ferruzzi, G. Arnaldi, C. Crescioli, S. Benvenuti, G. Nesi, A. Valeri, I. Greeve, M. Serio, M. Mannelli, A. Peri, Expression of the novel adrenocorticotropin- responsive gene selective Alzheimer's disease indicator-1 in the normal adrenal cortex and in adrenocortical adenomas and carcinomas, J Clin Endocrinol Metab, 89 (2004) 1332-1339. [92] S. Benvenuti, P. Luciani, I. Cellai, C. Deledda, S. Baglioni, R. Saccardi, S. Urbani, F. Francini, R. Squecco, C. Giuliani, G.B. Vannelli, M. Serio, A. Pinchera, A. Peri, Thyroid hormones promote cell differentiation and up-regulate the expression of the seladin-1 gene in in vitro models of human neuronal precursors, J Endocrinol, 197 (2008) 437-446. [93] E. Ishida, K. Hashimoto, S. Okada, T. Satoh, M. Yamada, M. Mori, Thyroid hormone receptor and liver X receptor competitively up-regulate human selective Alzheimer's disease indicator-1 gene expression at the transcriptional levels, Biochemical and biophysical research communications, 432 (2013) 513-518. [94] E. Ishida, K. Hashimoto, S. Okada, T. Satoh, M. Yamada, M. Mori, Crosstalk between Thyroid Hormone Receptor and Liver X Receptor in the Regulation of Selective Alzheimer's Disease Indicator-1 Gene Expression, PLoS One, 8 (2013) e54901. [95] Y.E. Timsit, M. Negishi, CAR and PXR: the xenobiotic-sensing receptors, Steroids, 72 (2007) 231-246. [96] K. Yoshinari, H. Ohno, S. Benoki, Y. Yamazoe, Constitutive androstane receptor transactivates the hepatic expression of mouse Dhcr24 and human DHCR24 encoding a cholesterogenic enzyme 24-dehydrocholesterol reductase, Toxicol Lett, 208 (2012) 185- 191. [97] Y. Wang, P.M. Rogers, K.R. Stayrook, C. Su, G. Varga, Q. Shen, S. Nagpal, T.P. Burris, The selective Alzheimer's disease indicator-1 gene (Seladin-1/DHCR24) is a liver X receptor target gene, Mol Pharmacol, 74 (2008) 1716-1721. [98] J.B. Demoulin, J. Ericsson, A. Kallin, C. Rorsman, L. Ronnstrand, C.H. Heldin, Platelet-derived growth factor stimulates membrane lipid synthesis through activation of phosphatidylinositol 3-kinase and sterol regulatory element-binding proteins, The Journal of biological chemistry, 279 (2004) 35392-35402. [99] A. Kallin, L.E. Johannessen, P.D. Cani, C.Y. Marbehant, A. Essaghir, F. Foufelle, P. Ferre, C.H. Heldin, N.M. Delzenne, J.B. Demoulin, SREBP-1 regulates the expression of heme oxygenase 1 and the phosphatidylinositol-3 kinase regulatory subunit p55 gamma, J Lipid Res, 48 (2007) 1628-1636. [100] M.C. Ramos, S. Sierra, C. Ramirez, J. Velasco, J.S. Burgos, Simvastatin modulates the Alzheimer's disease-related gene seladin-1, Journal of Alzheimer's disease : JAD, 28 (2012) 297-301. [101] I. Ikeda, K. Tanaka, M. Sugano, G.V. Vahouny, L.L. Gallo, Inhibition of cholesterol absorption in rats by plant sterols, J Lipid Res, 29 (1988) 1573-1582. [102] T.A. Miettinen, H. Gylling, Regulation of cholesterol metabolism by dietary plant sterols, Curr Opin Lipidol, 10 (1999) 9-14. [103] L. Normen, P. Dutta, A. Lia, H. Andersson, Soy sterol esters and beta-sitostanol ester as inhibitors of cholesterol absorption in human small bowel, Am J Clin Nutr, 71 (2000) 908-913.

193

[104] S.R. Panini, A. Gupta, R.C. Sexton, E.J. Parish, H. Rudney, Regulation of sterol biosynthesis and of 3-hydroxy-3-methylglutaryl-coenzyme A reductase activity in cultured cells by progesterone, The Journal of biological chemistry, 262 (1987) 14435- 14440. [105] J.E. Metherall, K. Waugh, H. Li, Progesterone inhibits cholesterol biosynthesis in cultured cells. Accumulation of cholesterol precursors, The Journal of biological chemistry, 271 (1996) 2627-2633. [106] B. Lindenthal, A.L. Holleran, T.A. Aldaghlas, B. Ruan, G.J. Schroepfer, Jr., W.K. Wilson, J.K. Kelleher, Progestins block cholesterol synthesis to produce meiosis- activating sterols, FASEB journal : official publication of the Federation of American Societies for Experimental Biology, 15 (2001) 775-784. [107] W. Kim, E.J. Bennett, E.L. Huttlin, A. Guo, J. Li, A. Possemato, M.E. Sowa, R. Rad, J. Rush, M.J. Comb, J.W. Harper, S.P. Gygi, Systematic and quantitative assessment of the ubiquitin-modified proteome, Mol Cell, 44 (2011) 325-340. [108] S.A. Wagner, P. Beli, B.T. Weinert, M.L. Nielsen, J. Cox, M. Mann, C. Choudhary, A proteome-wide, quantitative survey of in vivo ubiquitylation sites reveals widespread regulatory roles, Molecular & cellular proteomics : MCP, 10 (2011) M111 013284. [109] S.A. Wagner, P. Beli, B.T. Weinert, C. Scholz, C.D. Kelstrup, C. Young, M.L. Nielsen, J.V. Olsen, C. Brakebusch, C. Choudhary, Proteomic analyses reveal divergent ubiquitylation site patterns in murine tissues, Molecular & cellular proteomics : MCP, 11 (2012) 1578-1585. [110] H. Molina, D.M. Horn, N. Tang, S. Mathivanan, A. Pandey, Global proteomic profiling of phosphopeptides using electron transfer dissociation tandem mass spectrometry, Proc Natl Acad Sci U S A, 104 (2007) 2199-2204. [111] K.T. Rigbolt, T.A. Prokhorova, V. Akimov, J. Henningsen, P.T. Johansen, I. Kratchmarova, M. Kassem, M. Mann, J.V. Olsen, B. Blagoev, System-wide temporal characterization of the proteome and phosphoproteome of human embryonic stem cell differentiation, Sci Signal, 4 (2011) rs3. [112] C. Schmitz-Peiffer, D.R. Laybutt, J.G. Burchfield, E. Gurisik, S. Narasimhan, C.J. Mitchell, D.J. Pedersen, U. Braun, G.J. Cooney, M. Leitges, Inhibition of PKCɛ Improves Glucose-Stimulated Insulin Secretion and Reduces Insulin Clearance, Cell metabolism, 6 (2007) 320-328. [113] J.L. Goldstein, S.K. Basu, M.S. Brown, Receptor-mediated endocytosis of low- density lipoprotein in cultured cells, Methods in Enzymology, 98 (1983) 241-260. [114] J.R. Krycer, I. Kristiana, A.J. Brown, Cholesterol homeostasis in two commonly used human prostate cancer cell-lines, LNCaP and PC-3, PLoS One, 4 (2009) e8496. [115] A.J. Brown, L. Sun, J.D. Feramisco, M.S. Brown, J.L. Goldstein, Cholesterol addition to ER membranes alters conformation of SCAP, the SREBP escort protein that regulates cholesterol metabolism, Mol Cell, 10 (2002) 237-245. [116] J.E. Metherall, J.L. Goldstein, K.L. Luskey, M.S. Brown, Loss of transcriptional repression of three sterol-regulated genes in mutant hamster cells, The Journal of biological chemistry, 264 (1989) 15634-15641. [117] R.B. Rawson, R. DeBose-Boyd, J.L. Goldstein, M.S. Brown, Failure to cleave sterol regulatory element-binding proteins (SREBPs) causes cholesterol auxotrophy in Chinese hamster ovary cells with genetic absence of SREBP cleavage-activating protein, J. Biol. Chem., 274 (1999) 28549-28556.

194

[118] W. Luu, L.J. Sharpe, J. Stevenson, A.J. Brown, Akt acutely activates the cholesterogenic transcription factor SREBP-2, Biochim Biophys Acta, 1823 (2012) 458-464. [119] J. Wong, C.M. Quinn, I.C. Gelissen, A.J. Brown, Endogenous 24(S),25- epoxycholesterol fine-tunes acute control of cellular cholesterol homeostasis, The Journal of biological chemistry, 283 (2008) 700-707. [120] L. Zeng, H. Liao, Y. Liu, T.S. Lee, M. Zhu, X. Wang, M.B. Stemerman, Y. Zhu, J.Y. Shyy, Sterol-responsive element-binding protein (SREBP) 2 down-regulates ATP- binding cassette transporter A1 in vascular endothelial cells: a novel role of SREBP in regulating cholesterol metabolism, J. Biol. Chem., 279 (2004) 48801-48807. [121] M. Amemiya-Kudo, H. Shimano, A.H. Hasty, N. Yahagi, T. Yoshikawa, T. Matsuzaka, H. Okazaki, Y. Tamura, Y. Iizuka, K. Ohashi, J. Osuga, K. Harada, T. Gotoda, R. Sato, S. Kimura, S. Ishibashi, N. Yamada, Transcriptional activities of nuclear SREBP-1a, -1c, and -2 to different target promoters of lipogenic and cholesterogenic genes, Journal of Lipid Research, 43 (2002) 1220-1235. [122] T. Grand-Perret, A. Bouillot, A. Perrot, S. Commans, M. Walker, M. Issandou, SCAP ligands are potent new lipid-lowering drugs, Nat. Med., 7 (2001) 1332-1338. [123] J. Inoue, R. Sato, M. Maeda, Multiple DNA elements for sterol regulatory element-binding protein and NF-Y are responsible for sterol-regulated transcription of the genes for human 3-hydroxy-3-methylglutaryl coenzyme A synthase and squalene synthase, Journal of Biochemistry, 123 (1998) 1191-1198. [124] W.C. Tseng, J.W. Lin, T.Y. Wei, T.Y. Fang, A novel megaprimed and ligase-free, PCR-based, site-directed mutagenesis method, Anal Biochem, 375 (2008) 376-378. [125] H.E. Klock, S.A. Lesley, The Polymerase Incomplete Primer Extension (PIPE) method applied to high-throughput cloning and site-directed mutagenesis, Methods in Molecular Biology, 498 (2009) 91-103. [126] J. Sakai, A. Nohturfft, D. Cheng, Y. Ho, M.S. Brown, J.L. Goldstein, Identification of complexes between the COOH-terminal domains of sterol regulatory element-binding proteins (SREBPs) and SREBP cleavage-activating protein, J. Biol. Chem., 272 (1997) 20213-20221. [127] J. Yang, M.S. Brown, Y.K. Ho, J.L. Goldstein, Three different rearrangements in a single intron truncate sterol regulatory element binding protein-2 and produce sterol- resistant phenotype in three cell lines. Role of introns in protein evolution, The Journal of biological chemistry, 270 (1995) 12152-12161. [128] D. Kielar, W. Dietmaier, T. Langmann, C. Aslanidis, M. Probst, M. Naruszewicz, G. Schmitz, Rapid quantification of human ABCA1 mRNA in various cell types and tissues by real-time reverse transcription-PCR, Clinical Chemistry, 47 (2001) 2089- 2097. [129] F.J. Field, E. Born, S.N. Mathur, Stanol esters decrease plasma cholesterol independently of intestinal ABC sterol transporters and Niemann-Pick C1-like 1 protein gene expression, Journal of Lipid Research, 45 (2004) 2252-2259. [130] J. Wong, C.M. Quinn, A.J. Brown, Statins inhibit synthesis of an oxysterol ligand for the liver x receptor in human macrophages with consequences for cholesterol flux, Arteriosclerosis, thrombosis, and vascular biology, 24 (2004) 2365-2371. [131] J.R. Krycer, A.J. Brown, Cross-talk between the androgen receptor and the liver X receptor: implications for cholesterol homeostasis, J. Biol. Chem., 286 (2011) 20637- 20647.

195

[132] I. Kristiana, L.J. Sharpe, V.S. Catts, L.H. Lutze-Mann, A.J. Brown, Antipsychotic drugs upregulate lipogenic gene expression by disrupting intracellular trafficking of lipoprotein-derived cholesterol, Pharmacogenomics Journal, 10 (2010) 396-407. [133] J. Wong, C.M. Quinn, A.J. Brown, Synthesis of the oxysterol, 24(S), 25- epoxycholesterol, parallels cholesterol production and may protect against cellular accumulation of newly-synthesized cholesterol, Lipids in Health and Disease, 6 (2007) 10. [134] X.M. Du, I. Kristiana, J. Wong, A.J. Brown, Involvement of Akt in ER-to-Golgi transport of SCAP/SREBP: A link between a key cell proliferative pathway and membrane synthesis, Mol. Biol. Cell, 17 (2006) 2735-2745. [135] D. Steinberg, J. Avigan, Rat liver sterol Δ24-reductase, Methods in Enzymology, 15 (1969) 514-522. [136] R.D. Finn, J. Mistry, J. Tate, P. Coggill, A. Heger, J.E. Pollington, O.L. Gavin, P. Gunasekaran, G. Ceric, K. Forslund, L. Holm, E.L. Sonnhammer, S.R. Eddy, A. Bateman, The Pfam protein families database, Nucleic Acids Res, 38 (2010) D211-222. [137] J.D. Feramisco, J.L. Goldstein, M.S. Brown, Membrane topology of human insig- 1, a protein regulator of lipid synthesis, The Journal of biological chemistry, 279 (2004) 8487-8496. [138] O. Emanuelsson, S. Brunak, G. von Heijne, H. Nielsen, Locating proteins in the cell using TargetP, SignalP and related tools, Nat Protoc, 2 (2007) 953-971. [139] J. Kyte, R.F. Doolittle, A simple method for displaying the hydropathic character of a protein, Journal of molecular biology, 157 (1982) 105-132. [140] A. Krogh, B. Larsson, G. von Heijne, E.L. Sonnhammer, Predicting transmembrane protein topology with a hidden Markov model: application to complete genomes, Journal of molecular biology, 305 (2001) 567-580. [141] M. Hafner, P. Juvan, T. Rezen, K. Monostory, J.M. Pascussi, D. Rozman, The human primary hepatocyte transcriptome reveals novel insights into atorvastatin and rosuvastatin action, Pharmacogenetics and genomics, 21 (2011) 741-750. [142] G. Bologna, C. Yvon, S. Duvaud, A.L. Veuthey, N-Terminal myristoylation predictions by ensembles of neural networks, Proteomics, 4 (2004) 1626-1632. [143] S. Maurer-Stroh, B. Eisenhaber, F. Eisenhaber, N-terminal N-myristoylation of proteins: prediction of substrate proteins from amino acid sequence, Journal of molecular biology, 317 (2002) 541-557. [144] B. Eisenhaber, P. Bork, F. Eisenhaber, Sequence properties of GPI-anchored proteins near the omega-site: constraints for the polypeptide binding site of the putative transamidase, Protein Eng, 11 (1998) 1155-1161. [145] S. Maurer-Stroh, F. Eisenhaber, Refinement and prediction of protein prenylation motifs, Genome biology, 6 (2005) R55. [146] J. Ren, L. Wen, X. Gao, C. Jin, Y. Xue, X. Yao, CSS-Palm 2.0: an updated software for palmitoylation sites prediction, Protein Eng Des Sel, 21 (2008) 639-644. [147] E. Beitz, T(E)Xtopo: shaded membrane protein topology plots in LAT(E)X2epsilon, Bioinformatics, 16 (2000) 1050-1051. [148] T. Hessa, N.M. Meindl-Beinker, A. Bernsel, H. Kim, Y. Sato, M. Lerch-Bader, I. Nilsson, S.H. White, G. von Heijne, Molecular code for transmembrane-helix recognition by the Sec61 translocon, Nature, 450 (2007) 1026-1030. [149] A. Bernsel, H. Viklund, A. Hennerdal, A. Elofsson, TOPCONS: consensus prediction of membrane protein topology, Nucleic Acids Res, 37 (2009) W465-468.

196

[150] H. Viklund, A. Elofsson, Best α ‐ helical transmembrane protein topology predictions are achieved using hidden Markov models and evolutionary information, Protein Science, 13 (2004) 1908-1917. [151] H. Viklund, A. Elofsson, OCTOPUS: improving topology prediction by two-track ANN-based preference scores and an extended topological grammar, Bioinformatics, 24 (2008) 1662-1668. [152] E. Granseth, H. Viklund, A. Elofsson, ZPRED: predicting the distance to the membrane center for residues in alpha-helical membrane proteins, Bioinformatics, 22 (2006) e191-196. [153] T.A. Rapoport, B. Jungnickel, U. Kutay, Protein transport across the eukaryotic endoplasmic reticulum and bacterial inner membranes, Annual review of biochemistry, 65 (1996) 271-303. [154] A. Elofsson, G. von Heijne, Membrane protein structure: prediction versus reality, Annu Rev Biochem, 76 (2007) 125-140. [155] T.N. Petersen, S. Brunak, G. von Heijne, H. Nielsen, SignalP 4.0: discriminating signal peptides from transmembrane regions, Nature methods, 8 (2011) 785-786. [156] J. Sakai, A. Nohturfft, D. Cheng, Y.K. Ho, M.S. Brown, J.L. Goldstein, Identification of complexes between the COOH-terminal domains of sterol regulatory element-binding proteins (SREBPs) and SREBP cleavage-activating protein, The Journal of biological chemistry, 272 (1997) 20213-20221. [157] E. Szczesna-Skorupa, K. Ahn, C.D. Chen, B. Doray, B. Kemper, The cytoplasmic and N-terminal transmembrane domains of cytochrome P450 contain independent signals for retention in the endoplasmic reticulum, The Journal of biological chemistry, 270 (1995) 24327-24333. [158] S.J. Pernecky, J.R. Larson, R.M. Philpot, M.J. Coon, Expression of truncated forms of liver microsomal P450 cytochromes 2B4 and 2E1 in Escherichia coli: influence of NH2-terminal region on localization in cytosol and membranes, Proc Natl Acad Sci U S A, 90 (1993) 2651-2655. [159] S.J. Pernecky, N.M. Olken, L.L. Bestervelt, M.J. Coon, Subcellular localization, aggregation state, and catalytic activity of microsomal P450 cytochromes modified in the NH2-terminal region and expressed in Escherichia coli, Arch Biochem Biophys, 318 (1995) 446-456. [160] H. Viklund, E. Granseth, A. Elofsson, Structural classification and prediction of reentrant regions in alpha-helical transmembrane proteins: application to complete genomes, Journal of molecular biology, 361 (2006) 591-603. [161] P.V. Hornbeck, J.M. Kornhauser, S. Tkachev, B. Zhang, E. Skrzypek, B. Murray, V. Latham, M. Sullivan, PhosphoSitePlus: a comprehensive resource for investigating the structure and function of experimentally determined post-translational modifications in man and mouse, Nucleic acids research, 40 (2012) D261-270. [162] J.D. Horton, J.L. Goldstein, M.S. Brown, SREBPs: activators of the complete program of cholesterol and fatty acid synthesis in the liver, J. Clin. Invest., 109 (2002) 1125-1131. [163] J.B. Demoulin, J. Ericsson, A. Kallin, C. Rorsman, L. Ronnstrand, C.H. Heldin, Platelet-derived growth factor stimulates membrane lipid synthesis through activation of phosphatidylinositol 3-kinase and sterol regulatory element-binding proteins, J. Biol. Chem., 279 (2004) 35392-35402. [164] J.D. Horton, N.A. Shah, J.A. Warrington, N.N. Anderson, S.W. Park, M.S. Brown, J.L. Goldstein, Combined analysis of oligonucleotide microarray data from

197

transgenic and knockout mice identifies direct SREBP target genes, Proc. Natl. Acad. Sci. USA, 100 (2003) 12027-12032. [165] A. Kallin, L.E. Johannessen, P.D. Cani, C.Y. Marbehant, A. Essaghir, F. Foufelle, P. Ferre, C.H. Heldin, N.M. Delzenne, J.B. Demoulin, SREBP-1 regulates the expression of heme oxygenase 1 and the phosphatidylinositol-3 kinase regulatory subunit p55 gamma, J. Lipid Res., 48 (2007) 1628-1636. [166] M.R. Briggs, C. Yokoyama, X. Wang, M.S. Brown, J.L. Goldstein, Nuclear protein that binds sterol regulatory element of low density lipoprotein receptor promoter. I. Identification of the protein and delineation of its target nucleotide sequence, J. Biol. Chem., 268 (1993) 14490-14496. [167] A. Parraga, L. Bellsolell, A.R. Ferre-D'Amare, S.K. Burley, Co-crystal structure of sterol regulatory element binding protein 1a at 2.3 A resolution, Structure, 6 (1998) 661-672. [168] M.C. Ramos, S. Sierra, C. Ramirez, J. Velasco, J.S. Burgos, Simvastatin Modulates the Alzheimer's Disease-Related Gene seladin-1, J. Alzheimer's Dis., 28 (2012) 297-301. [169] Y. Wang, P.M. Rogers, K.R. Stayrook, C. Su, G. Varga, Q. Shen, S. Nagpal, T.P. Burris, The Selective Alzheimer's Disease Indicator-1 Gene (Seladin-1/DHCR24) Is a Liver X Receptor Target Gene, Mol. Pharmacol., 74 (2008) 1716-1721. [170] L.F. Michael, J.M. Schkeryantz, T.P. Burris, The pharmacology of LXR, Mini Rev Med Chem, 5 (2005) 729-740. [171] F.F. Wiebel, K.R. Steffensen, E. Treuter, D. Feltkamp, J.A. Gustafsson, Ligand- independent coregulator recruitment by the triply activatable OR1/retinoid X receptor- alpha nuclear receptor heterodimer, Molecular Endocrinology, 13 (1999) 1105-1118. [172] N. Zelcer, N. Khanlou, R. Clare, Q. Jiang, E.G. Reed-Geaghan, G.E. Landreth, H.V. Vinters, P. Tontonoz, Attenuation of neuroinflammation and Alzheimer's disease pathology by liver x receptors, Proceedings of the National Academy of Sciences of the United States of America, 104 (2007) 10601-10606. [173] J. Yang, R. Sato, J.L. Goldstein, M.S. Brown, Sterol-resistant transcription in CHO cells caused by gene rearrangement that truncates SREBP-2, Genes & development, 8 (1994) 1910-1919. [174] P. Costet, Y. Luo, N. Wang, A.R. Tall, Sterol-dependent transactivation of the ABC1 promoter by the liver X receptor/retinoid X receptor, J. Biol. Chem., 275 (2000) 28240-28245. [175] S.L. Sabol, H.B. Brewer, Jr., S. Santamarina-Fojo, The human ABCG1 gene: identification of LXR response elements that modulate expression in macrophages and liver, Journal of Lipid Research, 46 (2005) 2151-2167. [176] H. Shimano, J.D. Horton, I. Shimomura, R.E. Hammer, M.S. Brown, J.L. Goldstein, Isoform 1c of sterol regulatory element binding protein is less active than isoform 1a in livers of transgenic mice and in cultured cells, Journal of Clinical Investigation, 99 (1997) 846-854. [177] J.M. Lopez, M.K. Bennett, H.B. Sanchez, J.M. Rosenfeld, T.F. Osborne, Sterol regulation of acetyl coenzyme A carboxylase: a mechanism for coordinate control of cellular lipid, Proceedings of the National Academy of Sciences of the United States of America, 93 (1996) 1049-1053. [178] S.B. Joseph, B.A. Laffitte, P.H. Patel, M.A. Watson, K.E. Matsukuma, R. Walczak, J.L. Collins, T.F. Osborne, P. Tontonoz, Direct and indirect mechanisms for regulation of fatty acid synthase gene expression by liver X receptors, J. Biol. Chem., 277 (2002) 11019-11025.

198

[179] G. Liang, J. Yang, J.D. Horton, R.E. Hammer, J.L. Goldstein, M.S. Brown, Diminished hepatic response to fasting/refeeding and liver X receptor agonists in mice with selective deficiency of sterol regulatory element-binding protein-1c, J. Biol. Chem., 277 (2002) 9520-9528. [180] S. Talukdar, F.B. Hillgartner, The mechanism mediating the activation of acetyl- coenzyme A carboxylase-alpha gene transcription by the liver X receptor agonist T0- 901317, Journal of Lipid Research, 47 (2006) 2451-2461. [181] S.M. Vallett, H.B. Sanchez, J.M. Rosenfeld, T.F. Osborne, A direct role for sterol regulatory element binding protein in activation of 3-hydroxy-3-methylglutaryl coenzyme A reductase gene, Journal of Biological Chemistry, 271 (1996) 12247-12253. [182] J.M. Lehmann, S.A. Kliewer, L.B. Moore, T.A. Smith-Oliver, B.B. Oliver, J.L. Su, S.S. Sundseth, D.A. Winegar, D.E. Blanchard, T.A. Spencer, T.M. Willson, Activation of the nuclear receptor LXR by oxysterols defines a new hormone response pathway, J. Biol. Chem., 272 (1997) 3137-3140. [183] C.M. Adams, J. Reitz, J.K. De Brabander, J.D. Feramisco, L. Li, M.S. Brown, J.L. Goldstein, Cholesterol and 25-hydroxycholesterol inhibit activation of SREBPs by different mechanisms, both involving SCAP and Insigs, J. Biol. Chem., 279 (2004) 52772-52780. [184] B.A. Janowski, B. Shan, D.W. Russell, The hypocholesterolemic agent LY295427 reverses suppression of sterol regulatory element-binding protein processing mediated by oxysterols, J. Biol. Chem., 276 (2001) 45408-45416. [185] J.R. Schultz, H. Tu, A. Luk, J.J. Repa, J.C. Medina, L. Li, S. Schwendner, S. Wang, M. Thoolen, D.J. Mangelsdorf, K.D. Lustig, B. Shan, Role of LXRs in control of lipogenesis, Genes & Development, 14 (2000) 2831-2838. [186] J.E. Metherall, J.L. Goldstein, K.L. Luskey, M.S. Brown, Loss of transcriptional repression of three sterol-regulated genes in mutant hamster cells, Journal of Biological Chemistry, 264 (1989) 15634-15641. [187] J. Wong, C.M. Quinn, A.J. Brown, SREBP-2 positively regulates transcription of the cholesterol efflux gene, ABCA1 by generating oxysterol ligands for LXR, Biochemical Journal, 400 (2006) 485-491. [188] K. Cartharius, K. Frech, K. Grote, B. Klocke, M. Haltmeier, A. Klingenhoff, M. Frisch, M. Bayerlein, T. Werner, MatInspector and beyond: promoter analysis based on transcription factor binding sites, Bioinformatics, 21 (2005) 2933-2942. [189] F. Zhao, Z. Xuan, L. Liu, M.Q. Zhang, TRED: a Transcriptional Regulatory Element Database and a platform for in silico gene regulation studies, Nucleic acids research, 33 (2005) D103-107. [190] G.E. Crooks, G. Hon, J.M. Chandonia, S.E. Brenner, WebLogo: a sequence logo generator, Genome Research, 14 (2004) 1188-1190. [191] M. Saito, M. Kohara, K. Tsukiyama-Kohara, Hepatitis C virus promotes expression of the 3β-hydroxysterol δ24-reductase through Sp1, Journal of medical virology, 84 (2012) 733-746. [192] N. Iwamoto, S. Abe-Dohmae, R. Sato, S. Yokoyama, ABCA7 expression is regulated by cellular cholesterol through the SREBP2 pathway and associated with phagocytosis, Journal of Lipid Research, 47 (2006) 1915-1927. [193] R. Sato, A. Okamoto, J. Inoue, W. Miyamoto, Y. Sakai, N. Emoto, H. Shimano, M. Maeda, Transcriptional regulation of the ATP citrate-lyase gene by sterol regulatory element-binding proteins, J. Biol. Chem., 275 (2000) 12497-12502.

199

[194] H.Y. Kan, P. Pissios, J. Chambaz, V.I. Zannis, DNA binding specificity and transactivation properties of SREBP-2 bound to multiple sites on the human apoA-II promoter, Nucleic acids research, 27 (1999) 1104-1117. [195] D. Rozman, M. Fink, G.M. Fimia, P. Sassone-Corsi, M.R. Waterman, Cyclic adenosine 3',5'-monophosphate(cAMP)/cAMP-responsive element modulator (CREM)- dependent regulation of cholesterogenic lanosterol 14alpha-demethylase (CYP51) in spermatids, Molecular Endocrinology, 13 (1999) 1951-1962. [196] G. Guan, P.H. Dai, T.F. Osborne, J.B. Kim, I. Shechter, Multiple sequence elements are involved in the transcriptional regulation of the human squalene synthase gene, J. Biol. Chem., 272 (1997) 10295-10302. [197] K. Ishimoto, K. Tachibana, I. Hanano, D. Yamasaki, H. Nakamura, M. Kawai, Y. Urano, T. Tanaka, T. Hamakubo, J. Sakai, T. Kodama, T. Doi, Sterol-regulatory- element-binding protein 2 and nuclear factor Y control human farnesyl diphosphate synthase expression and affect cell proliferation in hepatoblastoma cells, Biochemical Journal, 429 (2010) 347-357. [198] J.R. Smith, T.F. Osborne, J.L. Goldstein, M.S. Brown, Identification of nucleotides responsible for enhancer activity of sterol regulatory element in low density lipoprotein receptor gene, J. Biol. Chem., 265 (1990) 2306-2310. [199] V. Llorente-Cortes, P. Costales, J. Bernues, S. Camino-Lopez, L. Badimon, Sterol regulatory element-binding protein-2 negatively regulates low density lipoprotein receptor-related protein transcription, Journal of Molecular Biology, 359 (2006) 950- 960. [200] W.A. Alrefai, F. Annaba, Z. Sarwar, A. Dwivedi, S. Saksena, A. Singla, P.K. Dudeja, R.K. Gill, Modulation of human Niemann-Pick C1-like 1 gene expression by sterol: Role of sterol regulatory element binding protein 2, American Journal of Physiology Gastrointestinal and Liver Physiology, 292 (2007) G369-376. [201] H.J. Jeong, H.S. Lee, K.S. Kim, Y.K. Kim, D. Yoon, S.W. Park, Sterol-dependent regulation of proprotein convertase subtilisin/kexin type 9 expression by sterol- regulatory element binding protein-2, Journal of Lipid Research, 49 (2008) 399-409. [202] M. Chen, L.M. Chen, K.X. Chai, Mechanisms of sterol regulatory element- binding protein-2 (SREBP-2) regulation of human prostasin gene expression, Biochemical and biophysical research communications, 346 (2006) 1245-1253. [203] L.H. Smith, M.S. Petrie, J.D. Morrow, J.A. Oates, D.E. Vaughan, The sterol response element binding protein regulates cyclooxygenase-2 gene expression in endothelial cells, Journal of Lipid Research, 46 (2005) 862-871. [204] R. Sato, J. Inoue, Y. Kawabe, T. Kodama, T. Takano, M. Maeda, Sterol- dependent transcriptional regulation of sterol regulatory element-binding protein-2, J. Biol. Chem., 271 (1996) 26461-26464. [205] R.E. Soccio, R.M. Adams, K.N. Maxwell, J.L. Breslow, Differential gene regulation of StarD4 and StarD5 cholesterol transfer proteins. Activation of StarD4 by sterol regulatory element-binding protein-2 and StarD5 by endoplasmic reticulum stress, J. Biol. Chem., 280 (2005) 19410-19418. [206] Y.K. Seo, T.I. Jeon, H.K. Chong, J. Biesinger, X. Xie, T.F. Osborne, Genome- wide localization of SREBP-2 in hepatic chromatin predicts a role in autophagy, Cell metabolism, 13 (2011) 367-375. [207] Y.K. Seo, H.K. Chong, A.M. Infante, S.S. Im, X. Xie, T.F. Osborne, Genome- wide analysis of SREBP-1 binding in mouse liver chromatin reveals a preference for promoter proximal binding to a new motif, Proceedings of the National Academy of Sciences of the United States of America, 106 (2009) 13765-13769.

200

[208] L.A. Daimiel, M.E. Fernandez-Suarez, S. Rodriguez-Acebes, L. Crespo, M.A. Lasuncion, D. Gomez-Coronado, J. Martinez-Botas, Promoter analysis of the DHCR24 (3beta-hydroxysterol Delta(24)-reductase) gene: characterization of SREBP (sterol- regulatory-element-binding protein)-mediated activation, Bioscience reports, 33 (2013) 57-69. [209] H. Shimano, Sterol regulatory element-binding proteins (SREBPs): transcriptional regulators of lipid synthetic genes, Progress in lipid research, 40 (2001) 439-452. [210] Y. Ikeda, J. Yamamoto, M. Okamura, T. Fujino, S. Takahashi, K. Takeuchi, T.F. Osborne, T.T. Yamamoto, S. Ito, J. Sakai, Transcriptional regulation of the murine acetyl-CoA synthetase 1 gene through multiple clustered binding sites for sterol regulatory element-binding proteins and a single neighboring site for Sp1, J. Biol. Chem., 276 (2001) 34259-34269. [211] J.H. Kim, J.N. Lee, Y.K. Paik, Cholesterol biosynthesis from lanosterol. A concerted role for Sp1 and NF-Y-binding sites for sterol-mediated regulation of rat 7- dehydrocholesterol reductase gene expression, J. Biol. Chem., 276 (2001) 18153-18160. [212] M. Nagai, J. Sakakibara, Y. Nakamura, F. Gejyo, T. Ono, SREBP-2 and NF-Y are involved in the transcriptional regulation of squalene epoxidase, Biochemical and biophysical research communications, 295 (2002) 74-80. [213] T. Ohnesorg, J. Adamski, Promoter analyses of human and mouse 17beta- hydroxysteroid dehydrogenase type 7, The Journal of steroid biochemistry and molecular biology, 94 (2005) 259-261. [214] G. Schiavoni, A.M. Bennati, M. Castelli, M.A. Fazia, T. Beccari, G. Servillo, R. Roberti, Activation of TM7SF2 promoter by SREBP-2 depends on a new sterol regulatory element, a GC-box, and an inverted CCAAT-box, Biochimica et biophysica acta, 1801 (2010) 587-592. [215] M.M. Magana, S.S. Lin, K.A. Dooley, T.F. Osborne, Sterol regulation of acetyl coenzyme A carboxylase promoter requires two interdependent binding sites for sterol regulatory element binding proteins, Journal of Lipid Research, 38 (1997) 1630-1638. [216] G. Moyle-Heyrman, H.S. Tims, J. Widom, Structural constraints in collaborative competition of transcription factors against the nucleosome, J Mol Biol, 412 (2011) 634-646. [217] Z. Zhu, J. Shendure, G.M. Church, Discovering functional transcription-factor combinations in the human cell cycle, Genome Research, 15 (2005) 848-855. [218] I. Bjorkhem, Are side-chain oxidized oxysterols regulators also in vivo?, J Lipid Res, 50 Suppl (2009) S213-218. [219] S. Gill, R. Chow, A.J. Brown, Sterol regulators of cholesterol homeostasis and beyond: the oxysterol hypothesis revisited and revised, Prog Lipid Res, 47 (2008) 391- 404. [220] T.A. Spencer, A.K. Gayen, S. Phirwa, J.A. Nelson, F.R. Taylor, A.A. Kandutsch, S.K. Erickson, 24(S),25-Epoxycholesterol. Evidence consistent with a role in the regulation of hepatic cholesterogenesis, The Journal of biological chemistry, 260 (1985) 13391-13394. [221] A.J. Brown, 24(S),25-epoxycholesterol: a messenger for cholesterol homeostasis, Int J Biochem Cell Biol, 41 (2009) 744-747. [222] F.R. Taylor, A.A. Kandutsch, A.K. Gayen, J.A. Nelson, S.S. Nelson, S. Phirwa, T.A. Spencer, 24,25-Epoxysterol metabolism in cultured mammalian cells and repression of 3-hydroxy-3-methylglutaryl-CoA reductase, The Journal of biological chemistry, 261 (1986) 15039-15044.

201

[223] G. Popjak, A. Meenan, E.J. Parish, W.D. Nes, Inhibition of cholesterol synthesis and cell growth by 24(R,S),25-iminolanosterol and triparanol in cultured rat hepatoma cells, The Journal of biological chemistry, 264 (1989) 6230-6238. [224] R.C. Sexton, S.R. Panini, F. Azran, H. Rudney, Effects of 3 beta-[2- (diethylamino)ethoxy]androst-5-en-17-one on the synthesis of cholesterol and ubiquinone in rat intestinal epithelial cell cultures, Biochemistry, 22 (1983) 5687-5692. [225] J. Wong, C.M. Quinn, A.J. Brown, Synthesis of the oxysterol, 24(S), 25- epoxycholesterol, parallels cholesterol production and may protect against cellular accumulation of newly-synthesized cholesterol, Lipids Health Dis, 6 (2007) 10. [226] A.H. Rowe, C.A. Argmann, J.Y. Edwards, C.G. Sawyez, O.H. Morand, R.A. Hegele, M.W. Huff, Enhanced synthesis of the oxysterol 24(S),25-epoxycholesterol in macrophages by inhibitors of 2,3-oxidosqualene:lanosterol cyclase: a novel mechanism for the attenuation of foam cell formation, Circ Res, 93 (2003) 717-725. [227] N. Zelcer, C. Hong, R. Boyadjian, P. Tontonoz, LXR regulates cholesterol uptake through Idol-dependent ubiquitination of the LDL receptor, Science, 325 (2009) 100- 104. [228] B.L. Song, R.A. DeBose-Boyd, Ubiquitination of 3-hydroxy-3-methylglutaryl- CoA reductase in permeabilized cells mediated by cytosolic E1 and a putative membrane-bound ubiquitin ligase, The Journal of biological chemistry, 279 (2004) 28798-28806. [229] W. Luu, L.J. Sharpe, I.C. Gelissen, A.J. Brown, The role of signalling in cellular cholesterol homeostasis, IUBMB Life, (2013). [230] C. Yang, J.G. McDonald, A. Patel, Y. Zhang, M. Umetani, F. Xu, E.J. Westover, D.F. Covey, D.J. Mangelsdorf, J.C. Cohen, H.H. Hobbs, Sterol intermediates from cholesterol biosynthetic pathway as liver X receptor ligands, The Journal of biological chemistry, 281 (2006) 27816-27826. [231] T. Chijiwa, A. Mishima, M. Hagiwara, M. Sano, K. Hayashi, T. Inoue, K. Naito, T. Toshioka, H. Hidaka, Inhibition of forskolin-induced neurite outgrowth and protein phosphorylation by a newly synthesized selective inhibitor of cyclic AMP-dependent protein kinase, N-[2-(p-bromocinnamylamino)ethyl]-5-isoquinolinesulfonamide (H-89), of PC12D pheochromocytoma cells, The Journal of biological chemistry, 265 (1990) 5267-5272. [232] S.F. Barnett, D. Defeo-Jones, S. Fu, P.J. Hancock, K.M. Haskell, R.E. Jones, J.A. Kahana, A.M. Kral, K. Leander, L.L. Lee, J. Malinowski, E.M. McAvoy, D.D. Nahas, R.G. Robinson, H.E. Huber, Identification and characterization of pleckstrin-homology- domain-dependent and isoenzyme-specific Akt inhibitors, Biochem J, 385 (2005) 399- 408. [233] D. Toullec, P. Pianetti, H. Coste, P. Bellevergue, T. Grand-Perret, M. Ajakane, V. Baudet, P. Boissin, E. Boursier, F. Loriolle, et al., The bisindolylmaleimide GF 109203X is a potent and selective inhibitor of protein kinase C, The Journal of biological chemistry, 266 (1991) 15771-15781. [234] P.D. Davis, C.H. Hill, E. Keech, G. Lawton, J.S. Nixon, A.D. Sedgwick, J. Wadsworth, D. Westmacott, S.E. Wilkinson, Potent selective inhibitors of protein kinase C, FEBS Lett, 259 (1989) 61-63. [235] J.C. Obenauer, L.C. Cantley, M.B. Yaffe, Scansite 2.0: Proteome-wide prediction of cell signaling interactions using short sequence motifs, Nucleic Acids Res, 31 (2003) 3635-3641.

202

[236] M.A. Rodgers, V.A. Villareal, E.A. Schaefer, L.F. Peng, K.E. Corey, R.T. Chung, P.L. Yang, Lipid metabolite profiling identifies desmosterol metabolism as a new antiviral target for hepatitis C virus, J Am Chem Soc, 134 (2012) 6896-6899. [237] S.J. Pernecky, J.R. Larson, R.M. Philpot, M.J. Coon, Expression of truncated forms of liver microsomal P450 cytochromes 2B4 and 2E1 in Escherichia coli: influence of NH2-terminal region on localization in cytosol and membranes, Proc. Natl. Acad. Sci. U.S.A., 90 (1993) 2651-2655. [238] S.J. Pernecky, N.M. Olken, L.L. Bestervelt, M.J. Coon, Subcellular localization, aggregation state, and catalytic activity of microsomal P450 cytochromes modified in the NH2-terminal region and expressed in Escherichia coli, Arch. Biochem. Biophys., 318 (1995) 446-456. [239] K.H. Ruan, S.P. So, W. Zheng, J. Wu, D. Li, J. Kung, Solution structure and topology of the N-terminal membrane anchor domain of a microsomal cytochrome P450: prostaglandin I2 synthase, Biochem J, 368 (2002) 721-728. [240] S. Pagny, P. Lerouge, L. Faye, V. Gomord, Signals and mechanisms for protein retention in the endoplasmic reticulum, Journal of Experimental Botany, 50 (1999) 157- 164. [241] M.R. Jackson, T. Nilsson, P.A. Peterson, Identification of a consensus motif for retention of transmembrane proteins in the endoplasmic reticulum, The EMBO Journal, 9 (1990) 3153. [242] M.-P. Schutze, P.A. Peterson, M.R. Jackson, An N-terminal double-arginine motif maintains type II membrane proteins in the endoplasmic reticulum, The EMBO Journal, 13 (1994) 1696. [243] O. Dym, D. Eisenberg, Sequence-structure analysis of FAD-containing proteins, Protein Sci, 10 (2001) 1712-1728. [244] R. Thoma, T. Schulz-Gasch, B. D'Arcy, J. Benz, J. Aebi, H. Dehmlow, M. Hennig, M. Stihle, A. Ruf, Insight into steroid scaffold formation from the structure of human oxidosqualene cyclase, Nature, 432 (2004) 118-122. [245] N. Strushkevich, S.A. Usanov, H.W. Park, Structural basis of human CYP51 inhibition by antifungal azoles, Journal of molecular biology, 397 (2010) 1067-1078. [246] T. Nishimura, M. Kohara, K. Izumi, Y. Kasama, Y. Hirata, Y. Huang, M. Shuda, C. Mukaidani, T. Takano, Y. Tokunaga, H. Nuriya, M. Satoh, M. Saito, C. Kai, K. Tsukiyama-Kohara, Hepatitis C virus impairs p53 via persistent overexpression of 3beta-hydroxysterol Delta24-reductase, J. Biol. Chem., 284 (2009) 36442-36452. [247] T. Takano, K. Tsukiyama-Kohara, M. Hayashi, Y. Hirata, M. Satoh, Y. Tokunaga, C. Tateno, Y. Hayashi, T. Hishima, N. Funata, M. Sudoh, M. Kohara, Augmentation of DHCR24 expression by hepatitis C virus infection facilitates viral replication in hepatocytes, J. Hepatol., 55 (2011) 512-521. [248] C.-Y. Park, H.-J. Jun, T. Wakita, J.H. Cheong, S.B. Hwang, Hepatitis C Virus Nonstructural 4B Protein Modulates Sterol Regulatory Element-binding Protein Signaling via the AKT Pathway, J. Biol. Chem., 284 (2009) 9237-9246. [249] A.B. van 't Wout, J.V. Swain, M. Schindler, U. Rao, M.S. Pathmajeyan, J.I. Mullins, F. Kirchhoff, Nef induces multiple genes involved in cholesterol synthesis and uptake in human immunodeficiency virus type 1-infected T cells, J Virol, 79 (2005) 10053-10058. [250] M. Makishima, A.Y. Okamoto, J.J. Repa, H. Tu, R.M. Learned, A. Luk, M.V. Hull, K.D. Lustig, D.J. Mangelsdorf, B. Shan, Identification of a nuclear receptor for bile acids, Science, 284 (1999) 1362-1365.

203

[251] K.J. Rayner, Y. Suarez, A. Davalos, S. Parathath, M.L. Fitzgerald, N. Tamehiro, E.A. Fisher, K.J. Moore, C. Fernandez-Hernando, MiR-33 contributes to the regulation of cholesterol homeostasis, Science, 328 (2010) 1570-1573. [252] C. Esau, S. Davis, S.F. Murray, X.X. Yu, S.K. Pandey, M. Pear, L. Watts, S.L. Booten, M. Graham, R. McKay, A. Subramaniam, S. Propp, B.A. Lollo, S. Freier, C.F. Bennett, S. Bhanot, B.P. Monia, miR-122 regulation of lipid metabolism revealed by in vivo antisense targeting, Cell Metab, 3 (2006) 87-98. [253] L. Larsen, M.W. Rosenstierne, L.W. Gaarn, A. Bagge, L. Pedersen, C.M. Dahmcke, J.H. Nielsen, L.T. Dalgaard, Expression and localization of microRNAs in perinatal rat pancreas: role of miR-21 in regulation of cholesterol metabolism, PloS one, 6 (2011) e25997. [254] W. Rui, F. Bing, S. Hai-Zhu, D. Wei, C. Long-Bang, Identification of microRNA profiles in docetaxel-resistant human non-small cell lung carcinoma cells (SPC-A1), Journal of cellular and molecular medicine, 14 (2010) 206-214. [255] S. Lusa, S. Heino, E. Ikonen, Differential mobilization of newly synthesized cholesterol and biosynthetic sterol precursors from cells, The Journal of biological chemistry, 278 (2003) 19844-19851. [256] J.E. Phillips, W.V. Rodrigueza, W.J. Johnson, Basis for rapid efflux of biosynthetic desmosterol from cells, J Lipid Res, 39 (1998) 2459-2470. [257] Y. Lange, D.S. Ory, J. Ye, M.H. Lanier, F.F. Hsu, T.L. Steck, Effectors of rapid homeostatic responses of endoplasmic reticulum cholesterol and 3-hydroxy-3- methylglutaryl-CoA reductase, The Journal of biological chemistry, 283 (2008) 1445- 1455. [258] J. Wong, C.M. Quinn, G. Guillemin, A.J. Brown, Primary human astrocytes produce 24(S),25-epoxycholesterol with implications for brain cholesterol homeostasis, J Neurochem, 103 (2007) 1764-1773. [259] R. Paoletti, R. Fumagalli, E. Grossi, P. Paoletti, Studies on brain sterols in normal and pathological conditions, Journal of the American Oil Chemists' Society, 42 (1965) 400-404. [260] D.S. Goodman, J. Avigan, H. Wilson, The in vitro metabolism of desmosterol with adrenal and liver preparations, J Clin Invest, 41 (1962) 2135-2141. [261] D.S. Goodman, J. Avigan, H. Wilson, The metabolism of desmosterol in human subjects during triparanol administration, J Clin Invest, 41 (1962) 962-971. [262] J.R. Arthur, H.A. Blair, G.S. Boyd, J.I. Mason, K.E. Suckling, Oxidation of cholesterol and cholesterol analogues by mitochondrial preparations of steroid- hormone-producing tissue, Biochem J, 158 (1976) 47-51. [263] J.V. Swinnen, W. Ulrix, W. Heyns, G. Verhoeven, Coordinate regulation of lipogenic gene expression by androgens: evidence for a cascade mechanism involving sterol regulatory element binding proteins, Proc Natl Acad Sci U S A, 94 (1997) 12975- 12980. [264] A.X. Wu-Zhang, A.C. Newton, Protein kinase C pharmacology: refining the toolbox, Biochem J, 452 (2013) 195-209. [265] N.A. Roberts, M.S. Marber, M. Avkiran, Specificity of action of bisindolylmaleimide protein kinase C inhibitors: do they inhibit the 70kDa ribosomal S6 kinase in cardiac myocytes?, Biochemical pharmacology, 68 (2004) 1923-1928. [266] S.F. Steinberg, Structural basis of protein kinase C isoform function, Physiological reviews, 88 (2008) 1341-1378. [267] J.M. Danielsen, K.B. Sylvestersen, S. Bekker-Jensen, D. Szklarczyk, J.W. Poulsen, H. Horn, L.J. Jensen, N. Mailand, M.L. Nielsen, Mass spectrometric analysis

204

of lysine ubiquitylation reveals promiscuity at site level, Molecular & Cellular Proteomics, 10 (2011). [268] R. Keber, H. Motaln, K.D. Wagner, N. Debeljak, M. Rassoulzadegan, J. Acimovic, D. Rozman, S. Horvat, Mouse knockout of the cholesterogenic cytochrome P450 lanosterol 14alpha-demethylase (Cyp51) resembles Antley-Bixler syndrome, The Journal of biological chemistry, 286 (2011) 29086-29097. [269] J. Stevenson, Krycer, J.R., Phan, L., Brown, A.J., A Practical Comparison of Ligation-Independent Cloning Techniques, PloS one, In press (2013). [270] L. Simi, F. Malentacchi, P. Luciani, S. Gelmini, C. Deledda, R. Arvia, M. Mannelli, A. Peri, C. Orlando, Seladin-1 expression is regulated by promoter methylation in adrenal cancer, BMC Cancer, 10 (2010) 201. [271] H. Iseki, R. Shimizukawa, F. Sugiyama, S. Kunita, A. Iwama, M. Onodera, H. Nakauchi, K. Yagami, Parvovirus nonstructural proteins induce an epigenetic modification through histone acetylation in host genes and revert tumor malignancy to benignancy, J Virol, 79 (2005) 8886-8893.

205