Histone Deacetylase 3 and the Epigenomic Regulation of Brown Adipose Physiology

University of Pennsylvania ScholarlyCommons

Publicly Accessible Penn Dissertations

2017

Histone Deacetylase 3 And The Epigenomic Regulation Of Brown Adipose Physiology

Matthew Joseph Emmett University of Pennsylvania, [email protected]

Follow this and additional works at: https://repository.upenn.edu/edissertations

Part of the Genetics Commons, Molecular Biology Commons, and the Physiology Commons

Recommended Citation Emmett, Matthew Joseph, "Histone Deacetylase 3 And The Epigenomic Regulation Of Brown Adipose Physiology" (2017). Publicly Accessible Penn Dissertations. 2851. https://repository.upenn.edu/edissertations/2851

This paper is posted at ScholarlyCommons. https://repository.upenn.edu/edissertations/2851 For more information, please contact [email protected]. Histone Deacetylase 3 And The Epigenomic Regulation Of Brown Adipose Physiology

Abstract Brown adipose tissue (BAT) is a thermogenic organ, which helps to protect body temperature against cold and can counteract metabolic disease. Numerous pathways modulate the activity of BAT in a highly regulated and coordinated manner to exert tight control over its immense calorie burning capacity. However, less is known regarding the transcriptional and epigenomic mechanisms controlling the thermogenic gene program of brown adipocytes. Here, we show that histone deacetylase 3 (HDAC3) is required for the activation of BAT enhancers to ensure thermogenic aptitude, thus enabling survival of dangerously cold environmental temperature. Loss of HDAC3 leads to deficient expression of uncoupling protein 1 (UCP1) and oxidative phosphorylation genes, resulting in an inability to survive severely cold temperatures. Mechanistic studies demonstrate BAT HDAC3 functions as a coactivator of the nuclear receptor estrogen-related receptor α (ERRα), through deacetylation of PGC-1α to promote enhancer activation and gene transcription. Remarkably, this coactivator function of HDAC3 is required for the basal expression of the thermogenic gene program and occurs independently of adrenergic signaling. Thus, HDAC3 primes Ucp1 and the thermogenic transcriptional program to maintain a critical capacity for BAT to enable survival of dangerously cold temperature.

We further demonstrate that short environmental exposure to a moderate cold temperature circumvents the genetic requirement of HDAC3 for Ucp1 transcriptional activation. Environmental exposure facilitates a persistent epigenomic rescue mechanism permitting the prolonged expression of thermogenic genes and an acquired BAT aptitude to avert hypothermia. Notably, we discovered that moderate cold activates BAT ERRα enhancers in an HDAC3-independent manner. Together, these studies identify for the first time, that BAT HDAC3 coactivation of ERR enhancers promotes a basal thermogenic transcriptional program ensuring cold survival, while HDAC3-independent mechanisms, signaling from a moderate cold environment to genes, imparts epigenomic changes to influence the BAT thermogenic capacity against probable and impending dangerously cold temperatures.

Degree Type Dissertation

Degree Name Doctor of Philosophy (PhD)

Graduate Group Cell & Molecular Biology

First Advisor Mitchell A. Lazar

Keywords brown adipose, enhancer RNAs, epigenomic, HDAC3, metabolism, PGC-1a

Subject Categories Genetics | Molecular Biology | Physiology

This dissertation is available at ScholarlyCommons: https://repository.upenn.edu/edissertations/2851 HISTONE DEACETYLASE 3 AND THE EPIGENOMIC REGULATION OF BROWN ADIPOSE

PHYSIOLOGY

Matthew J. Emmett

A DISSERTATION

Cell and Molecular Biology

Presented to the Faculties of the University of Pennsylvania

Partial Fulfillment of the Requirements for the

Degree of Doctor of Philosophy

2017

Supervisor of Dissertation

______

Mitchell A. Lazar, M.D., Ph.D.

Willard and Rhoda Ware Professor in Diabetes and Metabolic Diseases

Graduate Group Chairperson

______

Daniel S. Kessler, Ph.D.

Associate Professor of Cell and Developmental Biology

Dissertation Committee

Patrick Seale, Ph.D. (Chair), Associate Professor of Cell and Developmental Biology

Kathryn E. Wellen, Ph.D., Associate Professor of Cancer Biology

M. Celeste Simon, Ph.D., Arthur H. Rubenstein, MBBCh Professor

Daniel J. Rader, M.D., Seymour Gray Professor of Molecular Medicine

HISTONE DEACETYLASE 3 AND THE EPIGENOMIC REGULATION OF BROWN ADIPOSE

PHYSIOLOGY

2017

Matthew Joseph Emmett

This work is licensed under the Creative Commons Attribution- NonCommercial-ShareAlike 3.0 License

To view a copy of this license, visit https://creativecommons.org/licenses/by-nc-sa/3.0/us/ To my wife, Jaclyn.

iii

ACKNOWLEDGMENT

To Mitch Lazar: First and foremost, I would like to thank you for providing outstanding mentorship since the first day I arrived in your lab. Thank you for creating an unparalleled environment in which to learn, explore, and dive deeply into research. You continually answered questions, proposed and challenged ideas, and provided immense support. It has been a true pleasure to learn from you and have you as my PhD thesis advisor. I look forward to your continued mentorship through the exciting future stages of my education and career.

To my PhD Thesis Committee: Patrick Seale, Kathryn Wellen, Celeste Simon, and Dan

Rader. Thank you for all being excellent scientific role models and for your willingness to both support and challenge me. Our thesis committee meetings facilitated many insightful discussions, which helped shape the direction of my work. To Patrick- thank you for being an excellent committee chair. Your advice, and countless discussions regarding experiments, ideas, and science, has been outstanding. Your positivity and encouragement will stick with me far beyond my years in graduate school!

To Zachary Gerhart-Hines: Thank you for the unwavering support and intellectual discussions. I’ve greatly enjoyed both your mentorship and friendship in the Lazar lab and from abroad in Copenhagen.

To David Steger, Daniel Cohen, and Sean Armour: Thank you for our daily interactions that have furthered my development. I cannot thank each of you enough for the countless scientific questions and discussions, which have increased and broadened my understanding of science. To Dave, thank you for all of the personal support and

iv discussions over the past few years. Your mentorship and advice was terrific, as well as your occasional joke.

To Lim: Thank you for an excellent collaboration spanning several years and the achievements that we’ve had together. Your approach to bioinformatics and biology has informed my thinking, future research, and approach to collaborating.

To Erika and Lindsey: Thank you for the unwavering support and friendship. We’ve had a lot of fun along the way! It has been a long road and I couldn’t have done it without your wonderful help and expert abilities.

To Joe: Thank you for facilitating so much of the work that I was able to complete in the lab. Whenever I needed your help or assistance, you were there without question.

To Hannah: Thank you for being an excellent learner, collaborator, and friend. It was a pleasure to teach you in the lab!

To the remaining Lazar lab members, past and present: Thank you for the memories, fun times, and for sharing intellectual challenges and ideas.

To the Penn MSTP: Many thanks to Skip Brass, Maggie Krall and my MSTP classmates, for always promoting each other and for being an inspirational group of people.

To my past scientific mentors: Leisha Emens (Johns Hopkins), Elissa Lei (NIH), Ben

Stanger (Penn), and Andrew Rhim (MD Anderson). Thank you for cultivating my scientific curiosity, potential, my growth, as well as the support along the way.

To my family: Most importantly, thank you to my wife, Jaclyn, for the unending love, support, encouragement, patience, and constant fun. Thank you to my parents, for

v always encouraging me to pursue ideas, aspirations, and for supporting my endeavors.

Krista, Mark, and Saxby- thank you for your understanding, encouragement, and support. To the Bassett family, thank you for your unwavering support, encouragement, and sense of humor.

ABSTRACT

HISTONE DEACETYLASE 3 AND THE EPIGENOMIC REGULATION OF BROWN ADIPOSE

PHYSIOLOGY

Matthew Joseph Emmett

Mitchell A. Lazar

Brown adipose tissue (BAT) is a thermogenic organ, which helps to protect body temperature against cold and can counteract metabolic disease. Numerous pathways modulate the activity of BAT in a highly regulated and coordinated manner to exert tight control over its immense calorie burning capacity. However, less is known regarding the transcriptional and epigenomic mechanisms controlling the thermogenic gene program of brown adipocytes. Here, we show that histone deacetylase 3 (HDAC3) is required for the activation of BAT enhancers to ensure thermogenic aptitude, thus enabling survival of dangerously cold environmental temperature. Loss of HDAC3 leads to deficient expression of uncoupling protein 1 (UCP1) and oxidative phosphorylation genes, resulting in an inability to survive severely cold temperatures. Mechanistic studies demonstrate BAT HDAC3 functions as a coactivator of the nuclear receptor estrogen- related receptor α (ERRα), through deacetylation of PGC-1α, to promote enhancer activation and gene transcription. Remarkably, this coactivator function of HDAC3 is required for the basal expression of the thermogenic gene program and occurs independently of adrenergic signaling. Thus, HDAC3 primes Ucp1 and the thermogenic

vii transcriptional program to maintain a critical capacity for BAT to enable survival of dangerously cold temperature.

BAT aptitude to avert hypothermia. Notably, we discovered that moderate cold activates

BAT ERRα enhancers in an HDAC3-independent manner. Together, these studies identify for the first time, that BAT HDAC3 coactivation of ERR enhancers promotes a basal thermogenic transcriptional program ensuring cold survival, while HDAC3- independent mechanisms, signaling from a moderate cold environment to genes, imparts epigenomic changes to influence the BAT thermogenic capacity against probable and impending dangerously cold temperatures.

viii

TABLE OF CONTENTS

Abstract ...... vii LIST OF TABLES ...... x LIST OF FIGURES ...... xi LIST OF EXTENDED FIGURES ...... xi Manuscripts ...... xiii Manuscripts discussed ...... xiii Manuscripts not discussed ...... xiv Chapter 1 ...... 1 General Introduction: Histone Deacetylase 3 and the Epigenomic Regulation of Integrative Physiology ...... 1 Abstract ...... 2 Introduction ...... 2 The HDAC family ...... 4 HDAC3 functions in nuclear receptor corepressor complexes ...... 5 HDAC3 demonstrates catalytic-dependent and catalytic independent functions ...... 7 Tissue-specific functions of HDAC3 ...... 8 Discussion and Perspective ...... 19 Chapter 2 ...... 22 Methods ...... 22 Chapter 3 ...... 49 Histone deacetylase 3 prepares brown adipose tissue for acute thermogenic challenge ...... 49 Abstract ...... 50 Introduction ...... 51 Results ...... 52 Discussion ...... 59 Chapter 4 ...... 87 Brown adipose tissue HDAC3-dependent cold survival is circumvented by environmentally induced epigenomic memory...... 87 Abstract ...... 88 Introduction ...... 89 Results ...... 91 Discussion ...... 102 Chapter 5 ...... 119 Summary and Future Directions ...... 119 ix

Future Directions: ...... 124 Chapter 6 ...... 129 Bibliography ...... 129

LIST OF TABLES Table 1: Primers...... 48

LIST OF FIGURES

Figure 3.1: HDAC3 controls BAT thermogenesis...... 60 Figure 3.2: HDAC3 is required for expression of UCP1 and OXPHOS genes in BAT...... 62

Figure 3.3.: HDAC3 functions as an ERRα coactivator in BAT...... 64

Figure 3.4: HDAC3 coactivation of ERRα is mediated by PGC-1α deacetylation...... 66 Figure 4.1: Moderate environmental cold exposure circumvents HDAC3- dependent Ucp1 expression...... 105 Figure 4.2: HDAC3-independent thermogenic pathways facilitate sustained BAT thermogenesis and cold survival...... 106 Figure 4.3: HDAC3-independent thermogenesis requires Ucp1 for survival...... 108 Figure 4.4: Environmental cold exposure imparts an epigenetic memory promoting thermogenic gene expression...... 109 Figure 4.5: Moderate cold activates estrogen-related receptor a enhancers independent of HDAC3...... 111

Figure 4.6: Cold induces expression of the alternative Pgc-1α4 isoform to promote Ucp1 expression...... 113

LIST OF EXTENDED FIGURES

Extended Data Figure 3.1: Ablation of HDAC3 in adipose tissue depots. ... 68 Extended Data Figure 3.2: BAT HDAC3 is required for cold-mediated induction of Ucp1 expression, and HDAC3 expression is not altered by acute cold...... 70 Extended Data Figure 3.3: HDAC3 is neither induced nor required for brown adipogenesis, but required for cell-autonomous Ucp1 expression...... 72

Extended Data Figure 3.4: HDAC3 is required for expression of mitochondrial OXPHOS and TCA cycle genes...... 74 Extended Data Figure 3.5: Metabolic studies of Adiponectin-Cre and Ucp1- Cre HDAC3 KO mouse models ...... 76 Extended Data Figure 3.6: Effect of high fat diet on Adiponectin-Cre and Ucp1-Cre HDAC3 KO mouse models...... 78 Extended Data Figure 3.7: Effect of high fat diet on Adiponectin-Cre and Ucp1-Cre HDAC3 KO mouse models housed at thermoneutrality...... 80

Extended Data Figure 3.8: Transcriptional role of HDAC3 and ERRα in BAT...... 81

Extended Data Figure 3.9: Role of HDAC3 on PGC-1α acetylation and function...... 83

Extended Data Figure 3.10: HDAC3 and ERRα activate Ppargc1b enhancers and transcription...... 85 Extended Data Figure 4.1: Moderate environmental cold rescues Ucp1 expression in Adipoq-cre HDAC3 KO mice...... 115 Extended Data Figure 4.2: Moderate cold acclimation and recovery protocol for control and HDAC3 KO mice...... 116 Extended Data Figure 4.3: Unbiased hierarchical gene clustering identifies sustained HDAC3-independent induction of Ucp1...... 117

Extended Data Figure 4.4: BAT-specific ablation of PGC-1α downregulates Ucp1 expression...... 118

xii

Manuscripts

Manuscripts discussed

Emmett MJ*, Lim HW*, Jager J, Richter HJ, Adlanmerini M, Peed LC, Briggs ER, Steger

DJ, Ma T, Sims CA, Baur JA, Pei L, Won KJ, Seale P, Gerhart-Hines Z, Lazar MA.

Histone deacetylase 3 prepares brown adipose tissue for acute thermogenic challenge.

Nature. 2017; 546(7659):544-548. *Notes co-first authorship

Emmett MJ, Lim HW, Richter HJ, Peed LC, Briggs ER, Won KJ , Lazar MA. Brown adipose tissue HDAC3-dependent cold survival is circumvented by environmentally induced epigenomic memory. In preparation.

Emmett MJ, Lazar MA. Histone Deacetylase 3 and the Epigenomic Regulation of

Integrative Physiology. Nature Reviews Molecular Cell Biology. In preparation.

xiii

Manuscripts not discussed

Gerhart-Hines Z, Feng D*, Emmett MJ*, Everett LJ, Loro E, Briggs ER, Bugge A, Hou C,

Ferrara C, Seale P, Pryma DA, Khurana TS, Lazar MA. The nuclear receptor Rev-erbα controls circadian thermogenic plasticity. Nature. 2013; 503(7476):410-3. *Notes co- second authorship

Koren S, DiPilato LM, Emmett MJ, Shearin AL, Chu Q, Monks B, Birnbaum MJ. The role of mouse Akt2 in insulin-dependent suppression of adipocyte lipolysis in vivo.

Diabetologia. 2015; 58(5):1063-70.

Zhang Y*, Fang B*, Emmett MJ, Damle M, Sun Z, Feng D, Armour SM, Remsberg JR,

Jager J, Soccio RE, Steger DJ, Lazar MA. Discrete functions of nuclear receptor Rev- erbα couple metabolism to the clock. Science. 2015; 348(6242):1488-92.

Soccio RE, Chen ER, Rajapurkar SR, Safabakhsh P, Marinis JM, Dispirito JR, Emmett

MJ, Briggs ER, Fang B, Everett LJ, Lim HW, Won KJ, Steger DJ, Wu Y, Civelek M,

Voight BF, Lazar MA. Genetic Variation Determines PPARγ Function and Anti-diabetic

Drug Response In Vivo. Cell. 2015; 162(1):33-44.

xiv

Teng X, Emmett MJ, Lazar MA, Goldberg E, Rabinowitz JD. Lactate Dehydrogenase C

Produces S-2 Hydroxyglutarate in Mouse Testis. ACS Chemical Biology. 2016;

11(9):2420-7.

Soccio RE, Li Z, Chen ER, Foong YH, Benson KK, Dispirito JR, Mullican SE, Emmett

MJ, Briggs ER, Peed LC, Dzeng RK, Medina CJ, Jolivert JF, Kissig M, Rajapurkar SR,

Damle M, Lim HW, Won KJ, Seale P, Steger DJ, Lazar MA. Targeting PPARγ in the epigenome rescues genetic metabolic defects in mice. The Journal of Clinical

Investigation. 2017; 127(4):1451-1462.

Chapter 1

General Introduction: Histone Deacetylase 3 and the Epigenomic Regulation of Integrative Physiology

In Preparation, Nature Reviews Molecular Cell Biology.

Emmett and Lazar

Abstract

Histone deacetylase 3 (HDAC3) is an important epigenomic regulator governing mammalian metabolic pathways, physiology, and development through tissue specific mechanisms. Unlike other histone deacetylases, HDAC3 is the sole HDAC that modulates nuclear receptor mediated transcription across tissues. Here, we review recent advances in our understanding of epigenomic mechanisms by which HDAC3 regulates cell-type specific enhancers to control gene expression and important effects on metabolism and physiology. Recent studies integrating genetics, genomics, transcriptomics, metabolomics, and proteomics to interrogate the effects of HDAC3 in different tissues have revealed unanticipated HDAC3 mechanisms and loss-of-function phenotypes that have revealed new insights into metabolic physiology, disease states, and mammalian development. Further, these findings highlight an important cautionary tale for extrapolation of protein function amongst different tissues, the need for careful cell-type specific studies in vivo, and the complexity of informed drug design.

Introduction

Regulation of gene expression is essential for the coordination and initiation of complex mammalian biological processes. This coordination occurs basally to support homeostatic the functions of a cell, or under the influence of a milieu of highly dynamic cues stemming from physiological signals, the environment, or by neighboring cells and tissue, amongst others, to impact and change gene expression. Changes in gene expression are controlled primarily through cis-regulatory enhancer elements and bound by sequence-specific transcription factors. However, the local chromatin environment at 2 enhancers and gene bodies also influences gene regulation. Epigenomic modifications of histone tails also influence chromatin structure as well as the binding of transcriptional regulators thereby influencing gene expression. Transcriptional coregulators act as intermediaries bridging transcription factor binding to the recruitment of chromatin- modifying enzymes to facilitate transcriptional changes.

Since Allis and colleagues discovered a transcription-associated Tetrahymena histone acetyltransferase (HAT) with homology to the yeast transcriptional adaptor protein GCN5

(Brownell and Allis, 1995; Brownell et al., 1996), histone acetylation has been linked to gene transcription. Many coactivators have since been shown to have HAT activity, including p300, CBP, pCAF, and TAFII250 (Bannister and Kouzarides, 1996; Chen et al., 1997; Mizzen et al., 1996; Ogryzko et al., 1996; Spencer et al., 1997; Yang, XJ;

Ogryzko, VV; Nishikawa, J; Howard, BH; Nakatani, 1996) and pronounced effects on gene transcription. The acetylation of histone tails can recruit proteins with bromo domains (Dhalluin et al., 1999; Tamkun et al., 1992), including BRD4, that can facilitate the scaffolding of protein complexes on chromatin, can act as transcription factors, and may contain intrinsic catalytic functions, all serving as mechanisms to promote and regulate transcription (Fujisawa and Filippakopoulos, 2017; LeRoy et al., 2008).

Moreover, histone acetylation can compete with other post-translational modifications

(PTMs) with different transcriptional consequences, such as methylation, and may also affect PTM of neighboring histone residues (Allis and Jenuwein, 2016; Berger, 2002).

Thus, is it critical that histone acetylation is reversible due to the action of histone deacetylases (HDACs) (Seto and Yoshida, 2014).

The HDAC family

Schreiber and colleagues were first to isolate and characterize and HDAC, which is now known as HDAC1 (Taunton et al., 1996). HDAC1 was found to be structurally related to yeast transcription factor rpd3 (Kadosh and Struhl, 1998; Rundlett et al., 1996; Taunton et al., 1996; Yang and Seto, 2008), and it is the founding member of 11 highly conserved mammalian genes comprising the HDAC family (Gregoretti et al., 2004). Those most closely related to HDAC1 are referred to as Class I HDACs, which consists of HDACs 1,

2, 3, 8. Class II HDACs are further subdivided into Class IIa (HDAC 4, 5, 7, 9), class IIb

(6, 10), and class IV (HDAC 11). All of these proteins contain a conserved deacetylase domain, although enzymatic activity has been difficult to demonstrate for some, particulary class IIa HDACs (Haberland et al., 2009; Lahm et al., 2007). HDACs comprising class I, II, and IV are all metalloenzymes require zinc (Zn) metal for enzymatic function (Yang and Seto, 2008). Whereas, a third class of HDACs called

Sirtuins are less related to HDAC1, and catalyze histone deacetylation by a mechanism that consumes NAD (Chalkiadaki and Guarente, 2012; Chang and Guarente, 2014;

Schwer and Verdin, 2008). There is biological interest in all of the HDACs, which have been reviewed elsewhere (Haberland et al., 2009; Yang and Seto, 2008). The purpose of this review is to provide a deep dive into HDAC3, whose molecular and physiological functions are quite distinct from those of other HDACs. Indeed, HDAC3 unlike other

HDACs, is conserved only from flies to mammals (Yang and Seto, 2008), where it uniquely regulates different genes and physiological processes in different tissues, and this review will focus on the understanding this pleiotropic biology.

HDAC3 functions in nuclear receptor corepressor complexes

HDAC3 is the predominant HDAC associated with nuclear receptor corepressors.

HDAC3 was initially discovered based on its homology to HDAC1 (Emiliani et al., 1998).

The intimate relationship between HDAC3 and nuclear receptor corepressors became apparent in 2000, when HDAC3 was independently identified as a stoichiometric component of complexes containing Silencing Mediator of Retinoid and Thyroid

Receptors (SMRT, also called NCOR2) (Guenther et al., 2000) and Nuclear Receptor

Corepressor (NCOR, also called NCOR1) (Li et al., 2000). Both NCOR and SMRT complexes also contain the WD40-containing proteins TBL1 (Guenther et al., 2000) and

TBL1R(Yoon et al., 2003), as well as GPS2 (Zhang et al., 2002). Together, the molecules comprise the core components of the NCOR/SMRT complex. Importantly, although earlier studies had demonstrated in vitro interactions between NCoR or SMRT and other Class I HDACs including HDAC1 (Heinzel et al., 1997; Kao et al., 2000; Nagy et al., 1997), HDAC3 was the only Class I HDAC found in endogenous NCOR and

SMRT complexes. Class II HDACs were also shown to interact with NCOR and SMRT in vitro and when overexpressed in mammalian cells (Huang et al., 2000; Kao et al.,

2000) but they have not been identified in endogenous NCOR or SMRT complexes.

However, HDAC3 has been found in association with Class IIa HDACs, where it has been suggested to provide the deacetylase activity of those complexes (Fischle et al.,

2002).

HDAC3 enzyme activity is dependent upon interaction with nuclear receptor corepressors

Not only is HDAC3 a core stoichiometric component of the NCOR/SMRT complexes, but its catalytic activity relies on direct interaction with a conserved region of NCOR and

SMRT referred to as the Deacetylase Activating Domain (DAD) (Guenther et al., 2001;

Ishizuka and Lazar, 2005). This has been confirmed by multiple groups (Codina et al.,

2005; Zhang et al., 2002), and the crystal structure of the SMRT DAD bound to HDAC3 revealed the atomic details of the activating interaction, including both protein-protein interactions as well as an inositol tetraphosphate (IP4) molecule acting as an ionic bridge between the two polypeptides (Codina et al., 2005; Oberoi et al., 2011; Watson et al., 2012a). Together, the DAD and IP4, interacts with HDAC3 to generate a conformational change to facilitating access of substrates into the catalytic channel

(Arrar et al., 2013; Watson et al., 2012b). The critical role of the DAD domain in the biological activation of HDAC3 was demonstrated in mutant mice harboring DAD domain mutations in both alleles of NCOR and SMRT (Alenghat et al., 2008; You et al., 2013).

The structure of HDAC3 is most similar to HDAC8 (Gregoretti et al., 2004; Hu et al.,

2000; Marmorstein, 2001), and both require an active site phenylalanine for catalysis

(Gregoretti et al., 2004; Marmorstein, 2001). Embryonic and adult tissue of these mice had wildtype levels of HDAC3, but the HDAC3 protein had no detectable catalytic activity

(You et al., 2013). Most cellular HDAC3 exists NCOR/SMRT complexes, and any excess is found together with the TRiC chaperone complex (Guenther et al., 2002), whose role as HDAC3 chaperone in the assembly of the corepressor-HDAC3 complex is reminiscent of its role in the assembly of other multi-protein enzyme complexes,

6 including VHL and cyclin E-Cdk2 complex (Boudiaf-Benmammar et al., 2013; Dunn et al., 2001; Feldman et al., 1999; Yam et al., 2008).

HDAC3 demonstrates catalytic-dependent and catalytic- independent functions

HDAC3 is essential for life (Bhaskara et al., 2010; Lewandowski et al., 2014;

Montgomery et al., 2008); however, its intrinsic catalytic activity may be dispensable for survival (Alenghat et al., 2008; You et al., 2013). Recent work has demonstrated a striking discrepancy in the correlation between histone acetylation and gene activation, as loss of acetylation does not always impair gene transcriptional activity (Kasper et al.,

2010). Conversely, treatment of cells with histone deacetylase inhibitors (HDIs) does not necessarily increase the transcription of HDAC (Lopez-Atalaya et al., 2013) or more specifically, HDAC3 target genes (Mullican et al., 2011). In fact, a catalytic-dead mutant of HDAC3 can rescue gene repression genome-wide while failing to decrease histone acetylation levels (Sun et al., 2013). This finding sheds light on the biology of HDAC3, where at some genomic sites it may serve a primary role as a scaffold, much like the class IIa HDACs which demonstrate no catalytic activity (Lahm et al., 2007), for other proteins to bind and regulate transcription (Mihaylova et al., 2011). These interesting observations must be taken into account when evaluating the biological and physiological roles of HDAC3 throughout different tissues, as well as in the design of therapeutic strategies against HDAC3 and other chromatin regulators.

Tissue-specific functions of HDAC3

Germline deletion of HDAC3 is lethal early in embryogenesis (Bhaskara et al., 2008;

Lewandowski et al., 2015; Montgomery et al., 2008), illustrating that HDAC3 function is non-redundant with that of other HDACs. Advances in mouse genetics have made it possible to elucidate the tissue-specific roles of HDAC3 in mammalian development, physiology, and metabolism.

Liver. The mammalian liver is a central metabolic hub simultaneously coordinating many branches of intermediary metabolism to achieve physiological and systemic homeostasis. Pathological states such as obesity and nutrient overload, inflammation, and environmental influences can disrupt hepatic homeostasis and lead to development of insulin resistance, type 2 diabetes mellitus, non-alcoholic fatty liver disease (NAFLD), non-alcoholic steatohepatitis (NASH), and hepatocellular carcinoma (HCC) (Anstee et al., 2013; Byrne and Targher, 2015; Tilg et al., 2016). Initial indications that Hdac3 played a major role in coordinating hepatic intermediary metabolism came from liver- specific genetic loss of function studies. Loss of hepatic HDAC3 from early in embryogenesis revealed significant differences in lipid, cholesterol, and carbohydrate metabolism (Knutson et al., 2008). Deletion of HDAC3 in adult liver leads to an even more steatotic liver, due to both increased lipogenesis and decreased fatty acid oxidation (Feng et al., 2011; Sun et al., 2012). Mice with a steatotic liver due to loss of

HDAC3 are not insulin resistant, but actually more insulin sensitive as intermediary metabolites are directed towards de novo lipogenesis and lipid storage pathways and away from gluconeogenesis (Sun et al., 2012). Chromatin immunoprecipitation (ChIP) of

HDAC3 followed by next generation sequencing (Feng et al., 2011) revealed diurnal

8 recruitment of HDAC3 to sites of binding of the circadian nuclear receptor Rev-erbα, which binds NCOR1 (Feng et al., 2011). Recruitment of HDAC3 through NCOR1 to Rev- erbα binding sites imparts an epigenomic rhythm of histone acetylation at genes regulating hepatic lipogenesis, and genetic ablation of Rev-erbα results in the formation of fatty liver (Bugge et al., 2012; Feng et al., 2011; Zhang et al., 2015). Nevertheless, the HDAC3 KO livers are far more steatotis than mice lacking Rev-erbα, suggesting that

HDAC3 is recruited to lipid metabolic genes by other factors. An in vivo proteomics screen for HDAC3 interaction partners on liver chromatin identified a module consisting of nuclear receptor HNF4α as well as PROX1, which recruit HDAC3 and contribute to its control of liver lipid metabolism (Armour, S.M., et al., 2017).

Adipose. Brown adipose tissue is a specialized mammalian organ with the unique capacity to oxidize stored energy substrates to generate heat through mitochondria containing uncoupling protein 1 (Ucp1). Whereas white adipose tissue, functions predominately to safely store excess energy as lipid. In interscapular brown adipose tissue, HDAC3 is critical for ensuring expression of the thermogenic gene program to maintain mammalian body temperature from lethal hypothermia during acute environmental cold (Emmett et al., 2017). ChIP-seq studies of HDAC3 in brown fat integrated with enhancer RNA (eRNA) analysis (Fang et al., 2014; Lam et al., 2014) to investigate functional enhancers revealed HDAC3 serves as a coactiviator at a subset of genomic sites to facilitate the basal transcription of thermogenic genes. Notably, coactivator function of HDAC3 occurred through the deacetylation and activation PGC-

1α to promote estrogen-related receptor α (ERRα) transcription (Emmett et al., 2017).

Interestingly, the coactivator mechanism of HDAC3 on the nuclear receptor

ERRα in brown adipose, was dominant over Rev-erbα utilization of the NCOR/HDAC3 complex to regulate thermogenesis (Emmett et al., 2017; Gerhart-Hines et al., 2013).

Further insight into HDAC3 and adipocyte enhancer activation was shed by chromatin looping studies, which demonstrated extensive recruitment of the NCOR/HDAC3 complex to sites of activated enhancer-promoter loops (Siersbæk et al., 2017). Studies of HDAC3 in the white adipose tissue of aged mice, demonstrated regulation of a futile cycle of de novo fatty acid synthesis and beta-oxidation promoted the browning of the inguinal adipose depots (Ferrari et al., 2017).

Heart. Congential heart defects account for nearly 1% of live births and the unique functional contributions of Hdac3 among class I histone deacetylases in cardiogenesis are of much interest (Haberland et al., 2009; Vincent and Buckingham, 2010). Loss of

Hdac3 in cardiac progenitor cells of the first and second heart fields using Nkx2.5-Cre results in complete lethality as early as E11.5 (Lewandowski et al., 2015). In cardiac progenitors, Hdac3 physically interacts with the transcription factor Tbx5, implicated in human Holt-Oram syndrome, where Hdac3 deacetylates Tbx5 to repress transcriptional activation (Lewandowski et al., 2015). Hdac3 knockout in progenitors leads to the early differentiation of cardimyocytes and results in ventricular and septal defects. In pre- migratory neural crest cells, genetic ablation of Hdac3 using Wnt1-Cre or Pax3-Cre also causes in perinatal lethality due to severe cardiovascular abnormalities, despite intact neural crest specification and migration to the forming heart (Singh et al., 2011).

Specifically, these cardiac defects were comprised of aortic arch hypoplasia, ventricular septal defects, double-outlet right ventricles, and interrupted aortic arches. Heart defects followed from cardiac neural crest cells failing to differentiate into smooth muscle, due to

10 a cell autonomous defect in expression of the Notch signaling ligand, Jagged1. Loss of

Hdac3 in the second heart field progenitor cells using Isl1-Cre and Mef2C-Cre, exhibited perinatal lethality due to severe abnormalities of second heart field structures, which are regulated in an Hdac3-dependent catalytic-independent manner through recruitment of the PRC2 complex to suppress aberrant TGF-β signaling (Lewandowski et al., 2015).

Conditional deletion in cardiac muscle was achieved using the α-MHC-Cre transgenic line and resulted in lethality by 16-weeks of age (Montgomery et al., 2008).

Mice deficient for Hdac3 demonstrated severe cardiac hypertrophy in the left ventrical and septum as well as interstitial fibrosis, which collectively impaired cardiac contractility and ventricular function (Montgomery et al., 2008). Hdac3 depleted hearts demonstrated gene expression profiles suggesting increased fatty acid oxidation, decreased glucose utilization, and oxidative phosphorylation. Despite this enhanced bioenergetic profile, knockout hearts exhibited mitochondrial dysfunction perhaps due to mitotoxicity from fatty acid oxidation byproducts (Montgomery et al., 2008). Whereas, ablation of Hdac3 with the muscle creatine kinase promoter (MCK-Cre) in both skeletal and cardiac muscle, resulted in mild increases of heart weight and thickening of ventricular walls, interventricular septae, with enlarged atria, but demonstrated preserved contractility and no impact on survival (Sun et al., 2011). Surprisingly, the Hdac3 knockout mice displayed downregulation of mitochondrial bioenergetics processes such as oxidative phosphorylation and ATP synthesis, TCA cycle, and fatty acid metabolism genes.

Challenge of mice on high-fat diet resulted in severe hypertrophic cardiomyopathy and cardiac contractile dysfunction, ultimately leading to a lethal heart failure in all mice due to myocardial energy starvation and failure to oxidize fatty acids (Sun et al., 2011). In this model, cardiomyopathy in Hdac3 knockouts is reminiscent of phenotypes and gene

11 regulation changes found in mice deficient for PPAR-β/δ, PGC-1α/β, ERRα or

ERRγ, suggesting HDAC3 might modulate these pathways (Emmett et al., 2017; Sun et al., 2011).

Skeletal Muscle. Metabolically, skeletal muscle consumes a vast amount of the body’s glucose supply, through both insulin-dependent and insulin-independent exercise-induced glucose utilization pathways. Moreover, muscle glucose-utilization or lack thereof is thought to be a major contributor to the pathological insulin resistance of type 2 diabetes. Conditional deletion of Hdac3 in skeletal muscle by MLC-Cre leads to pronounced physiological changes: systemic insulin resistance, impaired insulin and glucose tolerance, diminished skeletal muscle glucose uptake, with no effect on insulin signaling all while improving exercise endurance (Hong et al., 2016). In vivo metabolomics and tracing studies, coupled to ChIP-seq and RNA-seq datasets, demonstrated Hdac3 controls a fuel switch from glucose to amino acid catabolism, where it partners with the circadian nuclear receptor Rev-erbα to anticipate and control the circadian rhythm of skeletal glucose versus amino acid fuel consumption(Hong et al.,

2016).

Macrophages. In macrophages, Hdac3 binds to the genome together with the macrophage lineage-determining transcription factor PU.1 (Mullican et al., 2011). In the absence of HDAC3, macrophage gene expression is reminiscent of the anti- inflammatory program of alternative activation, and the response to alternative activators

12 such as IL4 is amplified (Mullican et al., 2011). Reciprically, the M1 classical inflammatory activation pathway requires Hdac3 for gene expression (Chen et al., 2012).

Highlighting the potential importance of HDAC3 for the normal balance of macrophage polarization, transplant of LysM-Cre, Hdac3 bone marrow into atherosclerosis prone

LDLR-/- mice induced stable atherosclerotic plaques due to collagen deposition and decreased lipid accumulation and foam cells, due to the alternative activation and anti- inflammatory nature of HDAC3 deficient macrophages (Hoeksema et al., 2014).

Pancreas. Inducible ablation of Hdac3 in insulin-produced pancreatic β-cells using the MIP-CreERT transgenic mouse line revealed markedly improved glucose clearance following glucose challenge on both regular chow and high fat diet (Oropeza et al., 2015;

Remsberg et al., 2017). Despite improved glucose tolerance, loss of Hdac3 did not increase insulin transcription or total insulin content, but rather enhanced insulin secretion from β- cells both in vivo and ex vivo in isolated pancreatic islets, perhaps in part through a higher sensitivity to glucose (Remsberg et al., 2017). Global cistromic and transcriptomic studies suggest that Hdac3 controls multiple genes which influence glucose-stimulated insulin secretion (Remsberg et al., 2017). The findings from MIP-

CreERT Hdac3 β- cell knockout mice are consistent with Hdac3-specific inhibition promoting glucose metabolism in diabetic rats (Lundh et al., 2015), however, mice lacking β-cell Hdac3 in Rip-Cre mice (under control of the rat insulin 2 gene promoter) were reported to have decreased pancreatic insulin content and impaired glucose- stimulated insulin secretion (Chen et al., 2016). However, conclusions drawn from the

Rip-Cre conditional model may be confounded by aberrant Cre excision in the brain and central nervous system. Thus, future studies will need to address Hdac3 function in the endocrine pancreas. 13

Brain. Unlike other Hdacs, Hdac3 plays a significant and non-redundant role in neurodevelopment, as conditional loss of Hdac3 displays major brain abnormalities

(Morris and Monteggia, 2013). Ablation of Hdac3 in neural progenitor cells using Nes-

Cre, which deletes throughout the developing central nervous system, results in perinatal lethality by 16 hours of birth, due to organizational, structural, and cellular abnormalities in the cortex, cerebellar cortex, and neocortex (Norwood et al., 2014).

Thy1-Cre mediated deletion in the forebrain and cerebellum permits life for upwards of 6 weeks, but major locomotor deficits exist from progressive paralysis and ataxia, consistent with Purkinje neuron abnormalities. SImilar phenotypes were observed using a CaMKII-Cre (T50 line) model deleting in forebrain and cerebellum (Norwood et al.,

2014).

Oligodendrocyte progenitor cells (OPCs) are precursor cells to both oligodendrocytes and astrocytes, which assure axonal myelination and neural repair.

Deletion of Hdac3, by Olig1-Cre or Olig2-Cre in the OPCs, leads to the pathological development of tremors and optic nerve impairment due to reduced oligodendrocytes and severe myelination defects (Zhang et al., 2016). Following loss of function and lineage tracing experiments, Hdac3 mechanistically emerged as a molecular brake to guide OPCs towards an oligodendrocyte cell fate over undertaking astrogliogenesis

(Zhang et al., 2016). OPC Hdac3 ChIP-seq and RNA-seq, revealed a mechanism whereby Hdac3 antagonizes STAT3 acetylation and pro-astrogliogenic genes, while cooperating with the p300 coactivator to activate gene expression and promote oligodendrocyte formation (Zhang et al., 2016).

Bilateral injection of AAV-Cre recombinase into the CA1 hippocampal region to ablate Hdac3, supported a major role of Hdac3 repressing long-term memory formation and memory duration through repression of Nr4a2 expression in a NCoR-dependent manner (McQuown et al., 2011). In a process thought similar to long-term memory formation, Hdac3 in the nucleus accumbens negatively regulates cocaine-context- associated memory formation, again through regulation of Nr4a2 and Fos expression.

Cocaine was found to decrease Hdac3 occupancy at Nr4a2 and Fos genes, and in a similar fashion to Hdac3 knockout mice, cocaine may increase the negative reinforcing properties of drug abuse (Rogge et al., 2013).

Brain Hdac3 associates with MeCP2, which is mutated in the human neurodevelopmental disorder Rett syndrome. MeCP2 binds to methylated DNA and recruits the Sin3a/Hdac1/Hdac2 and NCoR/Hdac3 complex to repress gene transcription. Missense mutations in MeCP2 cause Rett syndrome impair recruitment with the NCoR/Hdac3 complex at brain enhancers (Ebert et al., 2013; Lyst et al., 2013).

Upon deletion of Hdac3 with the αCaMKII-Cre (CW2 transgenic line) in forebrain excitatory neurons of the hippocampus, a phenotype similar to MeCP2 KO models of

Rett syndrome occurs featuring impaired sociability, cognitive deficits, and irregular locomotor coordination (Nott et al., 2016). Detailed Hdac3 ChIP-seq studies in the brain coupled to RNA-seq, suggest Hdac3 may act to positively regulate a subset of brain genes via recruitment to MeCP2 binding sites where HDAC3 deacetylates FOXO3, to activate gene expression (Nott et al., 2016).

Lung. Conditional deletion of Hdac3 under control of Shh-Cre, which drives recombination in the lung endodermal epithelium at E8.75, demonstrated a requirement for Hdac3 in lung sacculation and early alveologenesis, as newborn knockout mice died between P2 to P10 (Wang et al., 2016b). This was mediated in part by derepression of miR-17-92 and Dlk-Dio3 miRNA clusters and subsequent cell-autonomous repression of

TGF-β signaling required to achieve cell flattening of lung alveolar type I (AT1) cells for efficient gas exchange (Wang et al., 2016b). Genetic depletion of Hdac3 in developing lung mesenchyme with Dermo1-Cre resulted in complete lethality at birth due to respiratory failure (Wang et al., 2016a). Unlike deletion in lung endoderm, depletion of

Hdac3 in the underlying mesoderm results in lung hypoplasia and loss of AT1 cell differentiation in the lung epithelium due to decreased Wnt/β-catenin signaling cues from the mesenchyme to epithelium (Wang et al., 2016a).

Hematoepoetic Cells. Deletion of Hdac3 in the hematopoietic compartment using Vav-Cre produces anemia, an afunctional thymus, and mice devoid of lymphoid progenitor cells leading to a subsequent block in T- and B- cell development, due to defects in DNA replication and genomic instability (Bhaskara et al., 2010; Summers et al., 2013). Hdac3 is also required for the proliferation and renewal of hematopoietic stem cells, as mice receiving Hdac3-depeleted bone marrow transplant failed to repopulate the HSC cells and perished (Summers et al., 2013).

Lymphocytes. Conditional loss of Hdac3 using Mb1-Cre to ablate Hdac3 expression in early B-cell progenitors impedes mature B-cell development due to failure

16 to induce efficient VHDJH recombination. Hdac3 deacetylase activity is required for B-cell development and VHDJH rearrangements, as catalytic-dead Hdac3 cannot rescue the loss-of-function phenotype (Stengel et al., 2017). HDAC3 deletion reduced nucleosome compaction, suggesting that altered chromatin structure could impair long-distance looping and recombination events needed for VHDJH recombination. Moreover, early loss of Hdac3 aberrantly promoted Th17 cells triggering inflammatory bowel disease (Philips et al., 2016). Ablation of Hdac3 using Lck-Cre, which deletes during double-negative stages of T-cell development, led to sparse abundance of double or singly CD4+ or

CD8+ T-cells (Stengel KR, et al., 2015). Intriguingly, during these developmental steps knockout of Hdac3 downregulated 2-3-fold more genes than were induced (Stengel KR, et al., 2015), which is counterintuitive for a transcriptional repressor. Deletion of Hdac3 at the double positive CD4+/CD8+ T-cell stage, by CD4-Cre mediated recombination, results in a peripheral T-cell maturation defect (Hsu et al., 2015). Moreover, the lineage determination factor for T reg cells, FOXP3, is regulated post-transcriptionally by histone deacetylases and can interact with Hdac3 (Wang et al., 2015b). Loss of Hdac3 in mature

Tregs by Foxp3-Cre disrupts Hdac3-dependent immunosuppressive functions and activates NF-kB signaling, triggering lethal autoimmunity in animals by 6 weeks of age

(Chen Lf et al., 2001; Wang et al., 2015b). Deletion of Hdac3 in the thymic epithelium by

Foxn1-Cre recombination resulted in thymic hypoplasia and fails to induce immunological tolerance (Goldfarb et al., 2016).

Skeletal system. Hdac3 has become one of the most well studied and characterized epigenetic regulators in bone development and growth (Bradley et al.,

2015). Deletion of HDAC3 using Wnt-Cre and Pax3-Cre to target pre-migratory neural crest cells that provide the main source of craniofacial mesenchyme causes severe defects in the bones of the viscerocranium and calvaria resulting in microcephaly, micrognathia, and cleft palate, and increased perinatal lethality (Singh et al., 2013).

Depletion of Hdac3 later in development using Osx1-Cre, which is expressed in osteochondral progenitor cells after osteoblast specification, resulted in a less severe intramembraneous bone phenotype although these mice exhibited delayed parietal bone formation and impaired calvarial bone development, with a thin and porous bone structure (Razidlo et al., 2010). Depletion of Hdac3 with Prrx1-Cre in mesenchymal progenitors of the limb bones led to severely shortened fore and hind limbs, resulting in perinatal lethality (Feigenson M, et al., 2017). Loss of Hdac3 impairs and delays chondrocyte and osteoblast differentiation through increased expression of both matrix metalloproteinases and Fgf21, and the cells exhibit increased apoptosis (Feigenson M, et al., 2017). HDAC3 deletion later in endochondral bone development with type II collagen α1-Cre (Col2α1-Cre) demonstrates HDAC3 dependency for growth plate chondrocyte maturation and long bone development (Carpio et al., 2016a). Genome- wide RNA-seq and ChIP-seq studies uncovered an HDAC3-dependent IL-6 autocrine loop impeding chondrocyte function and bone formation, as well as increased levels acetylation of NF-κB and increased Erk1/2 signaling (Carpio et al., 2016a, 2016b).

Finally, deletion strategies in mature osteoblasts and osteocytes with OCN-Cre, demonstrate age-related loss of trabecular and cortical bone mass, as well as decreased bone mineralization causing regular fractures (McGee-Lawrence et al., 2013).

Discussion and Perspective

HDAC3 is unique among the HDAC family as it is the sole histone deacetylase in the nuclear receptor corepressor complex (Bhaskara et al., 2008; Montgomery et al., 2008).

In this complex, it regulates a broad array of developmental, physiological, and metabolic processes vital to sustaining life. Throughout development, HDAC3 serves an important role in integrating to complex signals from the environment, to instruct cellular decisions and events of organogenesis. Whereas in mature animals, HDAC3 regulates the response to physiological challenges and controls the metabolic pathways and processes of cells. Further, the unique requirement for HDAC3 is best exemplified by the lack of functional redundancy amongst the eleven other classic HDACs. This requirement is evident by the many pleiotropic roles discovered amongst different tissue types as a result from careful genetic dissection of HDAC3 in mouse loss-of-function genetic studies.

Technological advances have made it possible to not only explore the tissue-specific functions of HDAC3, but also to ask new questions regarding its mechanism of action among different cell types. In liver, brown adipose, skeletal muscle, brain, and bone, identification of the HDAC3 cistromes have enabled the identification of interacting transcription factors through analysis of genomic binding sites putatively recruiting

HDAC3. Additionally, these techniques coupled to RNA-seq and differential gene analyses have allowed HDAC3 target genes to be identified in tissue-specific fashions.

Furthermore, knowledge of target genes has helped connect phenotype-to-epigenotype, or more specifically, to discover mechanisms behind loss-of-function HDAC3 phenotypes by knowing the gene regulatory pathways altered in its absence. The integration of multiple datasets in tissue-specific fashion has proved even more power into discerning yet unrecognized molecular mechanisms governed by HDAC3 function but also unappreciated elements of general mammalian physiology. In liver, ChIP-seq studies first uncovered a powerful role for HDAC3 in regulating circadian hepatic metabolism (Feng et al., 2011). Later ChIP-mass spectrometry technologies revealed unrecognized HDAC3 interacting proteins and provided a new wealth of insight into factors important for modulating hepatic metabolism (Armour, S.M., et al., 2017).

Proteomics and genome-wide ChIP-seq and RNA-seq in skeletal muscle, provided insights into circadian amino acid catabolism and the existence of transcriptionally regulated fuel switches which had been previously unappreciated (Hong et al., 2016).

Cistromics and functional enhancer analysis by GRO-seq techniques in brown adipose tissue facilitated discovery of unrecognized HDAC3 coactivator activity and facilitated the subsequent identification of an unappreciated HDAC3 non-histone target (Emmett et al.,

2017).

While the numerous and interesting phenotypes of HDAC3 loss-of-function studies provide insight into our greater understanding of HDAC3 biology, we know relatively little about HDAC3 catalytic-independent functions (Sun et al., 2013). As discussed previously, the fatty liver phenotype can be repressed by expression of an HDAC3 catalytic-dead molecule. Therefore, much of the liver HDAC3 knockout gross phenotype can be dissociated from the actual enzymatic activity. Past studies of HDAC3 provide the historical insight to understand the necessity for uncovering catalytic-independent

20 effects in a tissue-specific fashion. As any approach to pharmacological targeting of

HDAC3 enzymatic activity will retain the HDAC3 catalytic-independent effects, rational drug design would require an understanding of the two arms of HDAC3 action (catalytic and catalytic-independent) in a given tissue. Further, catalytic-independent effects may be potentially targeted through small-molecule based pharmacotherapy approaches designed to disrupt specific HDAC3 protein-protein interactions.

Chapter 2

Methods

Statistical methods

Data are presented as mean ± s.e.m. unless otherwise stated. Graphpad Prism software was used for graphing and statistical analysis. All statistical tests are fully described in the figure legends and met criteria for normal distribution with similar variance. No statistical methods were used to pre-determine sample size. For comparison between two groups, a Student’s t-test was used. For assessment between more than two groups, one-way ANOVA with multiple comparisons was used, and for assessment between two independent variables, two-way ANOVA with multiple comparisons was used, followed by a Tukey or Holm–Šidák post-test. For survival analysis, a log-rank

(Mantel–Cox) test was performed. FDR for RNA-seq analysis was calculated by EdgeR bioinformatics software.

Animal studies

Animal experiments were performed according to ethical regulations and protocols approved by the Institutional Animal Care and Use Committee of the Perelman

School of Medicine at the University of Pennsylvania and the Children’s Hospital of

Philadelphia. All experiments used male age-matched littermates group housed at four or five animals per cage. Mice were group housed with enrichment in a temperature- and humidity-controlled, specific-pathogen free animal facility at 22 °C under a 12:12-h light:dark cycle with free access to standard chow (LabDiet, 5010) and water. High-fat diet studies were conducted by feeding mice a purified-ingredient diet composed of

60:20:20 kcal percentage of fat:carbohydrate:protein (Research Diets, D12492) at 12 weeks of age.

Gene expression and protein studies were performed on 10- to 12-week-old male mice at zeitgeber time 10. Thermoneutrality and cold-exposure experiments were performed in climate-controlled rodent incubators (Powers Scientific) maintained at 29

°C or 4–5 °C, respectively, with interior temperature monitoring by digital thermometers.

For thermoneutrality experiments, mice were allowed to acclimate to 29 °C for 2 weeks.

No animals were excluded from studies and no randomization or blinding was performed.

Hdac3 loxP/loxP mice maintained on a C57BL/6 background were bred to

Adiponectin-cre mice maintained on a C57BL/6J background (Mullican et al., 2013) (The

Jackson Laboratory, B6;FVB-Tg(Adipoq-Cre)1Evdr/J, Stock 010803), to Ucp1-cre mice maintained on a C57BL/6J background (The Jackson Laboratory, B6.FVB-Tg(Ucp1-

Cre)1Evdr/J, Stock 024670)(Kong et al., 2014), and to Rosa26-creERT2 mice maintained on a C57BL/6J background (The Jackson Laboratory, B6.129-

Gt(Rosa)26SorTm1(cre/ERT2)Tyj/J, Stock 008463). Ucp1−/+ mice maintained on a

C57BL/6J background (The Jackson Laboratory, B6.129-Ucp1tm1Kz/J, Stock 003124) were used to generate Ucp1−/− and Ucp1+/+ mice. ERRα−/+ mice maintained on a

C57BL/6 background were used to generate ERRα−/− and ERRα+/+ mice as previously described (Luo et al., 2003). Ppargc1α loxp mice maintained on a C57BL/6NJ background (The Jackson Laboratory, B6N.129(FVB)-Ppargc1atm2.1Brsp/J, Stock

009666) were used to generate Ucp1-cre Ppargc1α and Hdac3/ Ppargc1α double floxed mice.

Tissue-specific HDAC3 KO mice were generated by mating males heterozygous for Adipoq-cre or Ucp1-cre and homozygous for the floxed HDAC3 alelle to females homozygous for the HDAC3 allele. Male littermate cre-positive (HDAC3 KO) and cre- negative (littermate controls) mice were then group housed upon weaning.

Tissue-specific HDAC3/UCP1 DKO and control littermate mice were generated by mating mice heterozygous for Adipoq-cre and homozygous for the floxed HDAC3, and heterozygous for Ucp1 −/+ to cre-negative mice homozygous for the floxed HDAC3, and heterozygous for Ucp1 −/+. Male littermate mice were then group housed upon weaning at room temperature.

Tissue-specific Ppargc1α KO mice were generated by mating males heterozygous for Ucp1-cre and homozygous for the floxed Ppargc1α alelle to females homozygous for the Ppargc1α allele. Male littermate cre-positive (Ppargc1α KO) and cre-negative (littermate controls) mice were then group housed upon weaning at room temperature.

Tissue-specific HDAC3/Ppargc1α DKO mice were generated by mating males heterozygous for Ucp1-cre and homozygous for floxed HDAC3 and Ppargc1α alelles to females homozygous for the HDAC3 and Ppargc1α alleles. Male littermate cre-positive

(HDAC3/Ppargc1α DKO) and cre-negative (littermate controls) mice were then group housed upon weaning at room temperature.

Core body temperature measurements and cold-tolerance testing

Baseline measurements of internal core-body temperatures were recorded from

10-week-old male mice maintained at 22 °C (starting at zeitgeber time 4) using a digital

25 thermometer (Oakton Instruments, Temp 10T Thermocouple) and rectal thermocouple probe (Physitemp, RET-3 probe). Cold-exposure studies were performed within climate- controlled rodent incubators. Mice were placed in pre-chilled cages at 4–5 °C with bedding, a cotton nestlet, free access to standard chow and water, and cage lid partly open. Rectal temperatures were recorded every 60 min. Individual mice were removed from the study and euthanized if core body temperature fell ≥10 °C from baseline measurement. Core temperatures ≥10 °C from baseline measurement were not included in core temperature analysis and scored as an event for survival analysis.

Moderate cold acclimation and cold recovery

HDAC3 KO and control littermate mice were exposed to 15 °C for 24-hours in temperature controlled rodent incubators. Mice were singly housed at 15 °C in pre- chilled cages to uniformly challenge mice with cold and present huddling for warmth.

During the time at 15 °C, cage lids remained cracked open to ensure temperature did not rise above 15 °C from body heat. Following 24-hours of exposure, mice were placed back into group housing with the same mice they were previously group housed with.

Mice were monitored for potential fighting incidents once returned to group housing.

Mice were allowed to recover at 22 °C for either 24-hours or 7-days before conducting cold tolerance physiological testing or gene expression analysis.

Whole-animal oxygen-consumption rate

Oxygen-consumption rates of 10-week-old male mice were measured using

Comprehensive Laboratory Animal Monitoring System (CLAMS, Columbus Instruments) metabolic cages housed within environment-controlled rodent incubators. Following previously described protocols, mice maintained at 22 °C were anaesthetized with 75 mg per kg (body weight) pentobarbital (Nembutal) by intraperitoneal injection and placed into CLAMS cages pre-acclimated to 33 °C to maintain mouse body temperature while under anaesthesia (Cannon and Nedergaard, 2011). Before noradrenaline-induced oxygen consumption readings, baseline oxygen consumption measurements were recorded for several cycles after pentobarbital administration until stable readings were recorded. A subcutaneous injection of 1 mg per kg (body weight) L-(−)-Norepinephrine

(+)-bitartrate salt monohydrate (Sigma, A9512) dissolved in sterile 0.9% NaCl (Sigma,

S8776) was given into the dorsal nuchal region. After injection, individual mice were immediately placed into the CLAMS cages and oxygen consumption rates were recorded until rates began to decline.

In vivo metabolic phenotyping

Whole-body energy metabolism was evaluated using a CLAMS (Columbia

Instruments). Mice were singly housed and acclimated in metabolic chambers for 48 h before data collection. Each mouse was continuously monitored for physical activity and food intake. CO2/O2 levels were collected four or five times per hour per mouse over the duration of the experiment. Analysis of covariance (ANCOVA) of VO2 versus body mass during light, dark, and 24 h periods was performed through the NIDDK Mouse Metabolic

Phenotyping Centers Energy Expenditure Analysis website

(http://www.mmpc.org/shared/regression.aspx). Body composition analyses by NMR were performed using Echo-MRI (Echo Medical Systems) to measure fat and lean mass.

BAT mitochondrial respiration assays

Mitochondrial respiration rates were determined using an O2K high-resolution respirometer (Oroboros Instruments, Austria). To preserve all populations of mitochondria, whole-tissue homogenates were prepared from BAT minced on ice and dounced at 4 °C (150 r.p.m. × eight strokes) in 15 ml of ice-cold mitochondrial isolation buffer (210 mM mannitol, 70 mM sucrose, 10 mM HEPES, 1 mM EGTA, adjusted to pH

7.2 with KOH) (Houstek et al., 1975; Pecinová et al., 2011; Picard et al., 2011).

Homogenates were centrifuged at 8,500g for 10 min at 4 °C. Lipid and supernatant were discarded. Pellet was re-suspended in 1 ml of ice-cold mitochondrial isolation buffer, passively filtered (100 µm), and protein concentration determined by BCA assay.

Respirometry measurements were performed using homogenates (0.25 mg) re- suspended in respiration buffer and maintained at 37 °C (110 mM mannitol, 0.5 mM,

EGTA, 3 mM MgCl2, 20 mM taurine 10 mM KH2PO4, 60 mM K lactobionate, 0.3 mM

DTT, and 0.1% BSA (fatty acid free), adjusted to pH 7.1). Samples were vigorously stirred by magnetic stir bar and the O2 concentration maintained between 100 and 200 nmol ml−1. UCP1-dependent respiration and individual complex-dependent respiration rates were determined by a standard substrate/inhibitor titration protocol. After stabilization, real-time oxygen concentration and flux data were collected continuously

(DatLab software 4.3, Oroboros Instruments, Austria). Baseline respiration following the

28 addition of substrate (palmatoylcarnitine (4 mM), pyruvate (10 mM), and malate (5 mM)) was measured. UCP1-dependent respiration was assessed by addition of 2 mM guanosine diphosphate. Independent complex-dependent respiration was then determined in the presence of ADP (1 mM). To assess complex-II-dependent respiration, rotenone (0.5 µM) was added to selectively inhibit complex I followed by succinate (10 mM), a complex II substrate. Antimycin A (5 µM) was then added to inhibit complex III followed by addition of TMPD (0.5 mM) and ascorbate (2 mM) as artificial electron donors for complex IV. Sodium azide (2.5 mM) was used to assess the contribution of leak respiration independent of complex IV(Porter et al., 2016; Wang et al., 2015a).

Histological analysis

Adipose tissues were fixed in 4% paraformaldehyde/1× PBS overnight at 4 °C and dehydrated through sequential ethanol washes. Tissue was embedded in paraffin before sectioning and then stained with haematoxylin and eosin. Stained sections were visualized and photographed under bright-field microscopy.

RNA isolation and gene expression analysis (RT–qPCR)

Total RNA was isolated from snap-frozen adipose tissues, which were mechanically homogenized in 500 µl TRIzol (Life Technologies) in a TissueLyser

(Qiagen) for 5 min, at a frequency of 20 s−1. After homogenization, samples were centrifuged at 4 °C and homogenates transferred to a fresh tube for RNA extraction with chloroform. Aqueous chloroform fractions were mixed with equal volumes of 70% 29 ethanol/30% water and RNA further purified with RNeasy Mini spin columns (Qiagen) and on-column DNase-digestion (Qiagen). TRIzol phases were saved for further isolation of genomic and mitochondrial DNA.

One and a half micrograms of total RNA was used for complementary DNA synthesis using a High-Capacity cDNA Reverse Transcription kit (Applied Biosystems).

Complementary DNAs (cDNAs) were analysed by qPCR using a Power SYBR Green

PCR Master mix on a Quant Studio 6 Flex Real-Time PCR system (Applied

Biosystems). All qPCR data were analysed using a standard curve and normalized to

36B4 (Arbp) or 18s expression. Specific primer sequences are listed in primer table.

Quantification of mitochondrial content

Total DNA was precipitated from TRIzol tissue extraction by ethanol precipitation and centrifugation, followed by sodium citrate/ethanol solution (0.1 M sodium citrate/10% ethanol, pH 8.5) washes of DNA pellet. DNA pellet was allowed to air dry before re- suspension. DNA samples were analysed by qPCR using a standard curve to determine absolute mitochondrial DNA abundance normalized to nuclear genomic DNA abundance. Specific primer sequences are listed in a primer table in the Supplementary

Information.

Immunoblotting

Interscapular BAT samples were homogenized in radioimmunoprecipitation assay buffer (150 mM NaCl, 1.0% NP-40, 0.5% sodium deoxycholate, 0.1% SDS, 50

30 mM Tris-HCl pH 8.0, 5 mM EDTA, 5 mM NaF, 1 mM Na3VO4) with 1 mM phenylmethylsulfonylflouride, supplemented with complete protease inhibitor (Roche), using a TissueLyser (Qiagen) for 3 min at a frequency of 20 s−1 at 4 °C followed by sonication with a Biorupter (Diagenode) for 30 s. For nuclear extracts, nuclei were prepared by dounce homogenization of five frozen BAT pads thawed in swelling buffer plus protease inhibitor (10 mM Tris-HCl, 2 mM MgCl2, 3 mM CaCl2), dounced with pestle A/B, filtered with 100 µm cell strainer, pelleted at 600g, and re-suspended in swelling buffer with cells lysed in an equal volume of lysis buffer (swelling buffer + 10% glycerol + 1% NP-40). Nuclei were isolated by centrifugation at 600g and soluble protein extracted by re-suspending in 4 pellet volumes of modified nuclear extract buffer (20 mM

HEPES, 0.42 M KCl, 3 mM MgCl2, 10% glycerol, 3 mM 2-mercaptoethanol, 0.5%

Trition-X, 1 mM EDTA, and 1 mM EGTA) plus protease inhibitors and incubated at 4 °C with rotation for 2 h. Protein concentration was determined using a Direct Detect

Spectrometer (EMD Millipore) and protein loaded onto 10% Criterion TGX gels (Bio-

Rad) by SDS–polyacrylamide gel electrophoresis, and transferred to a polyvinylidene difluoride (PVDF) membrane. After antibody incubation, immunoblots were visualized using enhanced chemiluminescence substrates (Western Lightning Plus-ECL,

PerkinElmer; or SuperSignal West Dura, Thermo) by autoradiography film or Bio-Rad

ChemiDoc Imaging.

Antibodies

The following antibodies were used for western blotting: HDAC3 (GeneTex,

GTX113303, Lot 40436), HDAC3 (Abcam, 76295), β-Actin (mAbcam 8226) (ABCAM,

20272), ERRα (ABCAM, 16363), HSP90 (Cell Signaling, 4874), Vinculin (Sigma,

V9264), Acetylated Lysine Rabbit Polyclonal (Cell Signaling, 9441, Lot 12), UCP1 (R&D,

MAB6158), HA Tag (GeneTex, 115044-01), Myc Tag (GeneTex, 21261), V5 Tag

(ThermoFisher, R960-25), Pol II Antibody (N-20, Santa Cruz 899 X, Lot K1215), GCN5

(H-75, Santa Cruz 20698, Lot K0112), PGC-1α Mouse mAb (EMD Millipore, 4C.13,

ST1202, Lot TE0349482), Anti-Rabbit IgG-HRP (Cell Signaling, 7074), Anti-mouse IgG-

HRP (Cell Signaling, 7076), Mouse monoclonal SB62a Anti-Rabbit IgG light chain (HRP)

(ABCAM, 99697). The following antibodies were used for ChIP: HDAC3 (ABCAM, 7030,

Lot GR121157-6), H3K27ac (ABCAM, 4729, Lot GR251958-1), ERRα (ABCAM, 16363,

Lot GR263385-1), NCoR (previously described, raised in rabbit against amino acids

1944–2453, affinity purified) (Feng et al., 2011), H3K4me1 (ABCAM 8895), and H3

(ABCAM 1791).

Cell culture

NIH/3T3 cells (ATCC, CRL1658) were maintained at 37 °C in 5% CO2 in high- glucose Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% Bovine

Calf Serum (Sigma) and 1% penicillin–streptomycin (Gibco); 293FT cells (Thermo,

R700-07) were maintained at 37 °C in 5% CO2 in high-glucose DMEM supplemented with 10% fetal bovine serum (Tissue Culture Biologicals) and 1% penicillin–streptomycin 32

(Gibco). All lines were passaged at approximately 80–85% confluence. Immortalized brown pre-adipocytes were maintained at 37 °C in 5% CO2 in DMEM/F-12 Glutamax,

10% FBS, 1% penicillin–streptomycin (Gibco) (Harms et al., 2014). NIH/3T3 and 293FT cell lines were low passage and not tested for mycoplasma. Immortalized brown pre- adipocytes tested negative for mycoplasma using a MycoAlert Mycoplasma Detection kit

(Lonza).

Primary brown adipocyte isolation and differentiation

Matings of Rosa26-CreER-positive, HDAC3F/F × HDAC3F/F mice were used to generate Rosa26-CreER-postive, HDAC3F/F, and littermate control HDAC3F/F pups.

Pre-adipocytes were isolated from BAT depots of postnatal day 1–3 mice and tails saved for genotyping. Individual depots were placed into 200 µl DMEM/F-12 Glutamax

(Invitrogen) and minced finely with spring scissors (Roboz) before digestion in 1 ml 1.5 U ml−1 collagenase D (Roche) and 2.4 U ml−1 Dispase II (Roche) at 37 °C in a thermocycler for 40 min at 1,200 r.p.m. Cells were purified through 100 µm filters

(Millipore) into 5 ml of DMEM/F-12 Glutamax with 10% FBS, pelleted at 700g, and re- suspended in growth media (DMEM/F-12 Glutamax, 10% FBS, 1% penicillin– streptomycin (Gibco), and primocin (Invivogen)).

Pre-adipocytes of the same genotype were pooled, re-plated, and, upon confluence, adipocyte differentiation was initiated with induction media (growth media supplemented with 500 nM dexamethasone, 125 µM indomethacin, 0.5 mM IBMX, 1 µM

Rosi, 1 nM T3, 20 nM insulin) for 48 h. After induction, cells were cultured in maintenance media (growth media plus 1 nM T3, 20 nM insulin) and media refreshed

33 every 48 h. For inducible HDAC3 deletion experiments, 2 µM 4-hydroxytamoxifen

(Sigma) was added to induction media of both Rosa26-CreER-positive and Rosa26-

CreER-negative cells from days 0 to 2. All analysis was performed on day 8 cultured brown adipocytes.

Retrovirus

WT Pgc-1α DNA and a non-acetylatable Pgc-1α mutant (R13) cDNA were cloned into a murine stem cell virus (MSCV retroviral construct. Retrovirus was produced in

HEK-293FT cells by co-transfection of MSCV and pCL-Eco vectors with FuGENE 6

(Promega), and supernatants were harvested 72 h after transfection. Supernatants were passed through a 0.20 µm filter (Corning) before viral concentration with PEG-it Virus

Precipitation Solution (System Biosciences) at 4 °C. Twenty-four-well plates were seeded with 6 × 104 brown pre-adipocytes per well and MSCV retroviruses were added upon cell adherence in media containing 2 µg ml−1 polybrene and allowed to reach confluence before adipocyte differentiation.

Dual luciferase assays

The mouse Ucp1 (-6k) enhancer (389 base pairs (bp)) containing ERRα and

HDAC3 binding sites, was PCR-amplified from C57BL/6 genomic DNA using primers listed in primer table to add XhoI and BglII sites for cloning into the pGL4.24 luciferase vector (Promega). The mouse Pgc-1α (-38k) enhancer (998 bp) containing identified binding sites was PCR-amplified from C57BL/6 genomic DNA using the primers listed in

34 the table in Supplementary Information and Gibson cloned between the NheI and BglII sites of the pGL4.24 luciferase vector. Luciferase assays were performed by transient transfection of NIH/3T3 cells seeded at a density of 1.25 × 105 cells per well (24-well plate) using Lipofectamine 2000 (3:1 ratio, Lipofectamine:DNA). Optimized transfections were performed using 1,200 ng DNA to 500 ng pGL4.24 luciferase vector and 2.0 ng

Renilla luciferase (for normalization) plus factor expression vectors and/or empty pcDNA3.1. Twenty-four hours after transfection, cell lysates were harvested using a

Dual Luciferase kit (Promega) to measure luciferase activities on a Synergy HT plate reader (Biotek).

Immunoprecipitation assays

For cellular acetylation experiments, 293FT cells were co-transfected with combinations of pcDNA3.1-GFP, pcDNA3.1-HA-PGC-1α, pcDNA3.1-HDAC3-myc, pcDNA3.1-Flag-GCN5, and/or empty pcDNA3.1. Forty-eight hours after transfection, cells were harvested in ice-cold NP40 IP lysis buffer (20 mM Tris-HCl, 150 mM NaCl,

10% glycerol, 2 mM EDTA, 0.1% NP40, 10 mM NaF, with 1 mM DTT, 1 mM phenylmethylsulfonylflouride), and supplemented with complete protease inhibitor cocktail (Roche), 4 µM trichostatin A (Cell Signaling), 5 mM nicotinamide (Sigma), and 1

µM Bortezomib (Cell Signaling) on ice. Cell lysates were freeze/thawed in liquid nitrogen twice, and lysates cleared by centrifugation at 4 °C for 10 min. Two hundred and fifty milligrams of protein was used for each immunoprecipitation and the volume adjusted to

1 ml with lysis buffer. Washed anti-haemagglutinin (HA) Agarose beads (Pierce) were incubated with samples and rotation at 4 °C for 4 h, followed by three brief washes with 35 lysis buffer. Immunoprecipitated proteins were eluted in 2× loading buffer with boiling at

95 °C for 5 min.

Co-immunoprecipitation

For endogenous co-immunoprecipitations, protein lysates were made from day 8 differentiated brown adipocytes in IPLS buffer (50 mM Tris-HCl pH7.4, 150 mM NaCl,

0.5% NP-40) with protease and phosphatase inhibitor (Roche). Lysates were sonicated, lipid removed by centrifugation at 4 °C, and 0.5 mg protein lysate was pre-cleared with anti-rabbit Trueblot beads (Rockland) for 3 h at 4 °C with rotation. One per cent of total pre-cleared lysate was saved for input. Pre-cleared lysate was then incubated with anti-

HDAC3 (Abcam, 7030) at 4 °C for 2 h with rotation, followed by immunoprecipitation with conjugated anti-rabbit goat IgG Trueblot beads for 1 h at 4 °C. Beads were washed four times with IPLS buffer plus inhibitors and samples eluted in 2× loading buffer at 55 °C for

20 min with agitation, supernatant isolated from beads, 2-mercaptoethanol (BME) and

1,3-dithiothreitol (DTT) added, and boiled at 95 °C for 5 min. Eluted proteins were resolved by immunoblotting and PVDF membrane incubated with anti-ERRα (Abcam,

16363) overnight at 4 °C, a conformation-specific secondary mouse anti-rabbit IgG–

HRP, and then developed. The blot was stripped (Restore Plus, Thermo), blocked, and incubated overnight at 4 °C with HDAC3 (Abcam, 76295) followed by incubation with conformation-specific secondary mouse anti-rabbit-HRP.

For overexpression co-immunoprecipitation experiments, V5-tagged ERRα and/or HA-tagged PGC-1α and myc-tagged HDAC3 were co-transfected into HEK-

293FT cells for 48 h. Forty-eight hours after transfection, cells were harvested in ice-cold

IPLS buffer supplemented with complete protease inhibitor cocktail, 1 mM phenylmethylsulfonylflouride, and 1 µM bortezomib on ice. Cell lysates were freeze/thawed in liquid nitrogen twice, sonicated, and lysates cleared by centrifugation at

4 °C for 10 min. One hundred milligrams of lysate was pre-cleared with Mouse IgG

Sepharose Bead Conjugate (Cell Signaling, 3420) before immunoprecipitation with V5- agarose beads (Sigma, A7345) at 4 °C for 2 h. Beads were washed four times in IPLS buffer and eluted in 2× loading buffer at 95 °C for 5 min, the supernatant isolated from the beads, and BME added and boiled at 95 °C for 5 min. Immunoblot for myc-HDAC3 was performed using goat anti-myc HRP (GeneTex), stripped, re-developed, and then immunoblotted for V5-ERRα with mouse monoclonal anti-V5 (ThermoFisher).

Protein purification

HA-tagged PGC-1α was co-transfected into HEK-293FT cells with GCN5 for 48 h and the cells were lysed in radioimmunoprecipitation assay lysis buffer supplemented with complete protease inhibitor cocktail (Roche), 1 mM phenylmethylsulfonylflouride, 4

µM trichostatin A (Cell Signaling), 5 mM nicotinamide, and 1 µM Bortezomib (Cell

Signaling) on ice. Lysate was sonicated, cleared by centrifugation, and immunoprecipitation of acetylated PGC-1α performed with anti-HA agarose beads

(Pierce) at 4 °C with rotation in the presence of inhibitors. Sequential washes (with inhibitors) were performed: six washes with radioimmunoprecipitation assay buffer, two washes with high salt wash buffer (10 mM Tris-HCl pH 8.0, 250 mM LiCl, 0.5% NP-40,

0.5% sodium-deoxycholate, 1 mM EDTA), followed by two washes in 1× PBS. The

37 purified protein remained immobilized on HA-beads in the presence of inhibitors before de-acetylation assay.

In vitro de-acetylation assay

Immobilized pre-acetylated PGC-1α was washed three times with HDAC assay buffer (Enzo). Thirty microlitres of the bead slurry was added to 100 µl HDAC assay buffer containing buffer alone, 100 ng recombinant human HDAC3/NCoR-DAD (Enzo), or 100 ng recombinant human HDAC3/NCoR-DAD plus 4 µM trichostatin A, and incubated at 37 °C with agitation for 1 h. After the reaction, the beads were boiled in 2× loading buffer with BME at 95 °C and acetylation of PGC-1α was assessed by western blot.

Acute siRNA knockdown in immortalized brown adipocytes

Acute siRNA knockdown was performed via reverse-transfection. Sixty thousand differentiated brown adipocytes (day 5) were seeded into 96-well plates containing

Lipofectamine RNAiMax and siRNAs (pre-incubated for 25 min in wells) or into no transfection control. Total RNAi concentration was held constant at 50 nM. Seventy-two hours after transfection, day 8 brown adipocytes were harvested in Trizol for analysis by

RT–qPCR.

ON-Target Plus SMARTpool siRNAs targeting mouse ERRα (Dharmacon, L-

040772-00, target sequence 5′-GCUGAAAGCUCUGGCCCUU-3′, 5′-

GCGGAGGACGGCAGAAGUA-3′, 5′-CUGAGAAGCUGUACGCCAU-3′, 5′-

GCAUCGAGCCUCUCUACAU-3′), Ppargc1α (Dharmacon, L-040773-01, 5′- 38

GAACAAGACUAUUGAGCGA, 5′-UUACGCAGGUCGAACGAAA-3′, 5′-

ACAAUGAGCCUGCGAACAU-3′, 5′-CAGCCGAGGACACGAGGAA-3′), Ppargc1β

(Dharmacon, L-040905-01, 5′-UGGUACAGCUCAUUCGCUA-3′, 5′-

GGGAAAAGCAAGUACGAAG-3′, 5′-GCUUUGAGGUGUUCGGUGA-3′, 5′-

GGAAAAGGCCAUCGGUGAA-3′) or non-targeting control (Dharmacon, D-001810-10,

5′-UGGUUUACAUGUCGACUAA-3′, 5′-UGGUUUACAUGUUGUGUGA-3′, 5′-

UGGUUUACAUGUUUUCUGA-3′, 5′-UGGUUUACAUGUUUUCCUA-3′) were used.

RNA-seq library preparation

Total RNA was extracted from interscapular BAT pads of 12-week-old male mice adapted to thermoneutrality for 2 weeks. One microgram of purified DNase-treated total

RNA from biological replicates (four WT control HDAC3F/F, four KO Adipoq-Cre;

HDAC3F/F) was processed with a Ribo-Zero Magnetic rRNA removal kit (Epicentre,

MRZH11124). The RNA libraries were prepared using a TruSeq RNA Sample Prep kit v2 (Illumina, RS-122-2001) and standard Illumina protocol. RNA-seq libraries were sequenced at single-end 100 bp read length on an Illumina HiSeq 2000.

RNA-seq data analysis

RNA-seq reads were aligned to the University of California, Santa Cruz (UCSC) mm9 genome browser using Tophat2(Kim et al., 2013). For differential gene expression 39 analysis, raw read counts were measured within RefSeq genes using featureCounts(Liao et al., 2014), and an exact test was performed using the edgeR pipeline(Robinson et al., 2010). Genes were included in differential expression analysis if their expression levels were >0.5 RPKM in at least three samples. Genes having a fold change of >1.5 (either up or down) and FDR of <0.01 were selected as HDAC3-KO- regulated genes.

Gene Ontology analysis was performed used Enrichr, where the top-ranked

KEGG pathway, Biological Process, and Cell Component terms were selected by

Enrichr combined score(Chen et al., 2013; Kuleshov et al., 2016). Representative downregulated genes identified by Gene Ontology were presented with their log2(fold change) as a heatmap. As an extended figure, all TCA cycle and mitochondrial respiratory chain complex genes retrieved from the HUGO gene nomenclature database were presented by heatmap to show quantitative trends identified in Gene

Ontology(Eyre et al., 2006).

For genome browser visualization, bigwig files were generated for individual replicates in a 1-bp-resolution RPM scale using genomeCoverageBed with ‘-split’ option in bedtools48 and bedGraphToBigWig in the SRA toolkit49 (Kent et al., 2010; Quinlan and Hall, 2010; 2011). A representative track was generated for each condition from the replicate bigwig files. All the genome browser snapshots were taken via IGV (Robinson et al., 2011; Thorvaldsdóttir et al., 2013).

ChIP

HDAC3 ChIP was performed as previously described with modification(Feng et al., 2011; Mullican et al., 2011). Immediately after euthanasia, the entire interscapular

BAT pad (two lobes) was snap-frozen. Tissue was later thawed and minced directly in cross-linking solution (10 ml, 1× PBS/1% formaldehyde) and allowed to rock for 20 min at room temperature. For epididymal white adipose tissue (eWAT) ChIP, two fat pads per were processed in an identical fashion to BAT.

Cross-linking was quenched by addition of 0.5 ml of 2.5 M glycine and allowed to mix for 5 min, followed by three ice-cold 1× PBS washes. Cross-linked tissue pieces were re-suspended in ice-cold ChIP dilution buffer (50 mM HEPES pH7.5, 155 mM

NaCl, 1 mM EDTA, 1.1% Triton X-100, 0.11% sodium deoxycholate, 0.1% SDS) with protease inhibitors (Complete Protease Inhibitor, Roche) and placed on ice. Chromatin fragmentation was performed using probe sonication at 4 °C (Fischer Scientific, FB705

Sonic Dismembrator) for three cycles of 10 s at 10% amplitude, followed by three cycles of 10 s at 15% amplitude, with a 30 s pause on ice between cycles. Chromatin lysates were cleared of lipid by centrifugation at 4 °C.

Chromatin lysates were brought to 1 ml with ChIP dilution buffer, input saved, and incubated with antibodies overnight at 4 °C, then immunoprecipitated with BSA blocked Protein A Sepharose CL-4B Beads (GE Healthcare) with rotation for 2 h at 4 °C.

Immunoprecipitations were washed: one quick wash 1 ml ChIP dilution buffer, one 5 min wash ChIP dilution buffer, one 5 min wash in 1 ml ChIP dilution buffer with 500 mM

NaCl, one 5 min wash in ChIP wash buffer (10 mM Tris-HCl pH 8.0, 250 mM LiCl, 0.5%

NP-40, 0.5% sodium deoxycholate, 1 mM EDTA), followed by one 5 min wash in Tris-

EDTA (10 mM Tris-HCl pH 8.0, 1 mM EDTA). Cross-linking was reversed overnight at

65 °C in elution buffer (50 mM Tris-HCl pH 8.0, 10 mM EDTA, 1% SDS), removed from beads, incubated with proteinase K, and DNA was isolated using phenol/chloroform 41 extraction in 2 ml Phase Lock Gel tubes (5 Prime) followed by NaCl/EtOH-mediated

DNA precipitation overnight at −20 °C with 20 µg glycogen carrier (Roche).

ChIP-seq library preparation

For BAT ChIP-seq of HDAC3, ERRα, and NCoR, five biological replicate ChIPs were pooled for sequencing. For BAT ChIP-seq of H3K27ac, replicate ChIPs were pooled into two libraries (three samples per library) for sequencing and library reads pooled into one data set. Enzymes from New England Biolabs were used to generate libraries according to the ChIP Sequencing Sample Preparation Guide provided by

Illumina. Adaptor oligonucleotides and primer sequences from Illumina were used for library construction and amplification. Before PCR library amplification, size selection of adaptor-ligated DNA was performed using Agencourt AMPure XP Beads (Beckman

Coulter). PCR Purification and MinElute (Qiagen) kits were used for library purification steps.

ChIP-seq data analysis

All ChIP-seq libraries were sequenced on HiSeq 2500 (single-end 50 bp or 100 bp) or NextSeq 500 (single-end 75 bp or 76 bp) systems. Before alignment, all reads were adjusted to 50 bp by trimming the 3′ end to avoid any bias due to different read lengths. Sequencing reads were aligned to the UCSC mm9 using Bowtie (Langmead et al., 2009). All peak calling was performed with the findPeaks command in the Homer package(Heinz et al., 2010). For HDAC3 peak calling, WT and HDAC3-KO aligned 42 reads were down-sampled to 20 million reads to adjust read-depth bias and avoid peak saturation, and duplicate reads were removed except for one. Stringent HDAC3 peaks were called in the WT sample against the KO sample using option ‘-F 3’ (FDR < 0.001).

ERRα peaks in WT and HDAC3-KO mice and NCoR peaks in WT mice were called against an input using default options (FDR < 0.001).

After initial peak calling, all peaks were resized into 200 bp followed by 1 RPM cutoff, and peaks within the ENCODE blacklist regions were discarded(Dunham et al.,

2012). When visualizing HDAC3, ERRα, and NCoR ChIP-seq profiles (line plots and heat maps), we used control-subtracted signal: KO sample for HDAC3, input sample for

ERRα and NCoR. All bigwig files were generated using Homer and the bedGraphToBigWig command. HDAC3 and ERRa ChIP-seq signals were compared by

RPM-normalized tag counts between KO-repressed and unchanged eRNAs within a 400 bp window, where the P value was calculated by a Wilcoxon test.

Two biological replicates of H3K27ac ChIP-seq were prepared from both WT and

HDAC3-KO BAT. Duplicate reads were condensed and replicates were pooled for visualization in a genome browser. H3K27ac levels at HDAC3-bound enhancer nearby

KO-regulated genes were compared between WT and KO mice by RPM-normalized tag counts within a 2 kb window around HDAC3 peaks, and P values were calculated by a

Wilcoxon test.

Super-enhancers were called on the basis of the H3K27ac data using a ROSE pipeline with default options. H3K4me1 and H3 ChIP-seq data in WT BAT and WT eWAT were generated to prepare an eRNA filter for GRO-seq analysis. Together with previously published adipose H3K27ac data (GSE63964) (Harms et al., 2015), histone modification peaks were called against corresponding H3 ChIP-seq samples in each depot by Homer with an option ‘-style histone -F 3’ (FDR <0.001). 43

De novo motif searches of ChIP-seq peaks were performed using the findMotifsGenome command in Homer within a 200 bp window with default options.

Discovered motifs were ranked by P values calculated on the basis of a binomial test against GC%-matched background, and the top four motifs presented.

GRO-seq library preparation

GRO-seq was performed and analysed as previously described, with minor modifications for mouse interscapular BAT samples (Fang et al., 2014). Ten interscapular BAT pads from genotype-confirmed Adipoq-Cre, HDAC3 KO, and control littermates were quickly dissected, immediately pooled and minced in ice-cold nuclei isolation buffer A, and dounce homogenized on ice to isolate ~40 million nuclei. Aliquots of nuclei preparations were inspected visually for quality under a microscope with DAPI

(4′,6-diamidino-2-phenylindole) staining, and nuclei counted using a Countess

Automated Cell Counter (Invitrogen). The nuclear run-on reaction was performed for 7 min and RNA hydrolysis was allowed to proceed for 10 min. All GRO-seq library preparations were performed in parallel to avoid batch effects.

GRO-seq data analysis

Before alignment, sequencing reads were cleaned by trimming off low-quality base, adaptor, and poly-A tailing sequences using cutadapt (Martin, 2011). Trimmed reads of at least 25 bp were selected and aligned to UCSC mm9 using Bowtie with an 44 option ‘–best–strata -m 1 -v 3’ (Langmead et al., 2009). Aligned reads within extremely high signal regions (such as ribosomal RNA, small nucleolar RNA, small nuclear RNA, and transfer RNA) were removed before downstream analysis to minimize read-depth bias. For genome browser visualization, bigwig files were generated separately for (+) and (−) strands, and all reads were extended to 150 bp for smooth profile. All (−) strand signals were presented as negative values.

Gene body transcription levels were quantitated in all conditions for comparative analysis. First, all reads were extended to 50 bp towards the 3′ end, then RPKM- normalized read counts were measured within RefSeq gene bodies excluding the 500 bp region at the 5′ end. However, for genes with total gene length <1 kb, the entire gene body was used. Gene transcription levels were further normalized among conditions for global bias correction using loess normalization in the limma package(Ritchie et al.,

2015). A gene was classified as ‘transcribed’ in a given condition if the transcription level was >0.3 RPKM. Genes that were not transcribed throughout all conditions were eliminated before downstream analysis. A gene was defined as ‘differential’ between a given pair of conditions if it was transcribed in either condition and the fold change was greater than 1.5 (either up or down).

Ucp1 and Ppargc1α nascent transcription levels were compared between genotypes and between temperatures via an exact test in EdgeR using a common dispersion estimated from housekeeping genes(Robinson and Smyth, 2008; Robinson et al., 2010).

Hierarchical clustering was performed for nascent gene transcription level at 22

°C and 29 °C in WT and HDAC3 KO mice to identify an HDAC3-dependent functional module. First, putatively highly variable genes were selected by fold change >1.5 between maximum and minimum transcription levels. Next, clustering was done for log2- 45 transformed gene transcription level by Ward’s criterion using a Pearson correlation coefficient as a similarity measure.

One cluster of genes was identified that displayed similar transcriptional changes to Ucp1 or Ppargc1α genes. Overall gene transcription trend was visualized as a boxplot in z-score scale. Next, Gene Ontology analysis was performed for this cluster using

Enrichr; top-ranked Biological Process terms and KEGG pathways were selected and their relative gene transcription levels were visualized by a heatmap in z-score scale.

eRNA calling was performed as previously described with minor optimization

(Fang et al., 2014; Step et al., 2014). First, eRNA peaks were called separately on each strand for all samples via Homer peak calling using the option ‘-style factor -separate - centre -fragLength 150 -size 200 -minDist 400 -tbp 0 -L 3 -localSize 10000 -fdr 0.001’.

Next, another round of peak calling was performed using the same option for intergenic reads only to rescue promoter-proximal eRNA peaks skipped by local background effect of highly transcribed genes. All eRNA peaks, excluding those falling into TSS 2kb and

TTS 2kb windows, were then pooled and merged among biological conditions to prepare a master eRNA peak set. Next, a series of filters was applied: (1) peaks within ENCODE blacklist regions were discarded; (2) small peaks (<0.2 RPM) or any peaks embedded within same-stranded gene transcription were discarded; (3) an epigenomic filter was applied using active enhancer marks, H3K27ac and H3K4me1 from BAT and eWAT, to broadly define and capture adipose eRNAs. Remaining peaks were then paired between opposite strands if their distance was <1kb. For paired eRNA peaks, their centre was defined as an enhancer; for unpaired eRNA peaks, their 5′ upstream 180 bp positions were defined as enhancers.

eRNA transcription levels were measured in RPKM within a 1 kb window around enhancers. If an enhancer was intergenic, both strands were measured; if it was within a 46 transcribed gene body, only the anti-sense signal was measured. After measuring eRNA levels in all conditions, loess normalization was applied for global bias correction as previously performed in the gene transcription analysis. For differential analysis comparing two different conditions, eRNAs with >0.5 RPKM in at least one condition were selected, then eRNAs with a fold change >2 (either up or down) were defined as differential eRNAs. De novo motif search for eRNAs was performed within a 400 bp window using Homer, and the top four motifs selected by P value were presented.

Table 1: Primers.

Chapter 3

Histone deacetylase 3 prepares brown adipose tissue for acute thermogenic challenge

Published in part in Nature, 2017, 546(7659):544-548.

Abstract

Brown adipose tissue is a thermogenic organ that dissipates chemical energy as heat to protect animals against hypothermia and to counteract metabolic disease

(Cannon and Nedergaard, 2004). However, the transcriptional mechanisms that determine the thermogenic capacity of brown adipose tissue before environmental cold are unknown. Here we show that histone deacetylase 3 (HDAC3) is required to activate brown adipose tissue enhancers to ensure thermogenic aptitude. Mice with brown adipose tissue-specific genetic ablation of HDAC3 become severely hypothermic and succumb to acute cold exposure. Uncoupling protein 1 (UCP1) is nearly absent in brown adipose tissue lacking HDAC3, and there is also marked downregulation of mitochondrial oxidative phosphorylation genes resulting in diminished mitochondrial respiration. Remarkably, although HDAC3 acts canonically as a transcriptional corepressor, it functions as a coactivator of estrogen-related receptor α (ERRα) in brown adipose tissue (Guenther et al., 2001). HDAC3 coactivation of ERRα is mediated by deacetylation of PGC-1α and is required for the transcription of Ucp1, Ppargc1α

(encoding PGC-1α), and oxidative phosphorylation genes. Importantly, HDAC3 promotes the basal transcription of these genes independently of adrenergic stimulation.

Thus, HDAC3 uniquely primes Ucp1 and the thermogenic transcriptional program to maintain a critical capacity for thermogenesis in brown adipose tissue that can be rapidly engaged upon exposure to dangerously cold temperature.

Introduction

Brown adipose tissue (BAT) is a major site of mammalian non-shivering thermogenesis mediated through UCP1-dependent respiration. Cold temperature triggers fuel oxidation and UCP1-mediated dissipation of the mitochondrial proton gradient to rapidly generate heat in BAT, and C57BL/6J (B6) mice acclimated to room temperature (22 °C) survive acute exposure to 4 °C through this mechanism (Cannon and Nedergaard, 2004; Chouchani et al., 2016; Enerbäck et al., 1997). While UCP1- deficient mice can utilize other thermogenic mechanisms upon gradual acclimation to cold, UCP1 is required to prevent lethal hypothermia upon rapid decreases in ambient temperature, such as from 22 °C to 4 °C (Bal et al., 2012; Enerbäck et al., 1997; Kazak et al., 2015; Long et al., 2016; Ukropec et al., 2006). Although much is known about brown adipose commitment and differentiation, the transcriptional mechanisms that ensure readiness of mature BAT for immediate heat production remain unclear(Harms and Seale, 2013; Loft et al., 2016).

The ubiquitously expressed class I histone deacetylase HDAC3 is an epigenomic modulator of nuclear receptor controlled gene expression, functioning as a stoichiometric component of the nuclear receptor co-repressor (NCoR) complex to modulate deacetylation of histones as well as non-histone targets (Guenther et al., 2000, 2001).

Global HDAC3 deletion is embryonic lethal, but studies of its tissue-specific functions link HDAC3 to hepatic steatosis, macrophage function, atherosclerosis, bone density, intestinal homeostasis, and cardiac energy metabolism (Alenghat et al., 2013; Bhaskara et al., 2008; Hoeksema et al., 2014; Montgomery et al., 2008; Mullican et al., 2011;

Razidlo et al., 2010; Sun et al., 2011, 2012). However, its physiological role in BAT is not known.

Results

We bred B6 mice with a floxed HDAC3 allele to B6 mice harboring the pan adipose adiponectin-cre (Adipoq-cre) and the BAT-specific Ucp1-cre for conditional pan- adipose and BAT-specific knockout (KO) (Extended Data Fig. 3.1a), and challenged adults with a drop in ambient temperature from 22 °C to 4 °C. As expected, control littermates maintained their core body temperature in the face of the acute environmental change (Fig. 3.1a). Strikingly, both the Adipoq-cre and Ucp1-cre HDAC3

KO mice exhibited a rapid loss of core body temperature, becoming severely hypothermic within just a few hours of moving to 4 °C (Fig. 3.1a). The inability to maintain core body temperature was lethal for every mouse lacking HDAC3 in BAT, whereas all the control littermates survived (Fig. 3.1b). Notably, the severe cold susceptibility of the HDAC3 KO mice were similar to that observed in congenic Ucp1 KO mice (Fig. 3.1a, b).

The requirement for HDAC3 in regulating BAT-thermogenic capacity was examined by measuring noradrenaline-induced wholebody oxygen consumption in anaesthetized mice. Control littermates exhibited a rapid and robust increase in oxygen consumption following noradrenaline treatment whereas HDAC3 KO mice had a blunted response comparable to that observed in Ucp1 KO mice (Cannon and Nedergaard,

2004) (Fig. 3.1c). Despite severely impaired BAT metabolic respiration, loss of HDAC3

52 had little effect on interscapular BAT mass, BAT mitochondrial content, or total body mass (Extended Data Fig. 3.1b–d).

BAT mitochondrial function was tested by high-resolution respirometry. As expected, palmitoylcarnitine and pyruvate induced respiration in control BAT mitochondria and, consistent with UCP1-dependence, this was inhibited by guanosine diphosphate (Cannon and Nedergaard, 2004) (Fig. 3.1d). Remarkably, mitochondria from HDAC3 KO BAT exhibited impaired substrate-induced respiration, with reduced function of complexes I, II, and IV of the electron transport chain (Fig. 3.1d). Consistent with this mitochondrial dysfunction, histological analysis revealed the presence of larger lipid droplets in HDAC3 KO BAT (Fig. 3.1e). By contrast, the histology of inguinal white adipose tissue (iWAT) lacking HDAC3 was similar to that of wild-type (WT) mice

(Extended Data Fig. 3.1e).

To elucidate the molecular basis through which HDAC3 controls the thermogenic capacity of BAT, we next performed RNA sequencing (RNA-seq) on BAT from control and KO mice housed at thermoneutrality (29 °C) to avoid the potentially confounding influence of cold-stress. Consistent with the well-known function of HDAC3 as a corepressor, many genes were induced by loss of HDAC3, including classic repression targets of the nuclear receptor Rev-erbα such as Bmal1 (also known as Arntl), E4bp4

(also known as Nfil3), Elovl5, and Plin2 (Fig. 3.2a). However, many genes were also decreased in HDAC3 KO BAT, including a near-complete loss of Ucp1 in the BAT of

HDAC3 KO animals (Fig. 3.2a). UCP1 protein expression was nearly undetectable in

HDAC3 KO BAT, at both 29 °C and 22 °C (Fig. 3.2b). Ucp1 expression remained low in

HDAC3 KO BAT after acute cold exposure (Extended Data Fig. 3.2a, b) although it was induced in iWAT (Extended Data Fig. 3.2c, d), suggesting a BAT-selective requirement 53 for HDAC3. BAT HDAC3 messenger RNA (mRNA) and protein levels were not appreciably affected by acute cold exposure (Extended Data Fig. 3.2e, f), and HDAC3 was neither induced nor required for primary brown adipocyte differentiation (Extended

Data Fig. 3.3a–d). However, primary brown adipocytes lacking HDAC3 exhibited reduced basal Ucp1, which was minimally responsive to isoproterenol (Extended Data

Fig. 3.3e–g), demonstrating a cell-autonomous effect of HDAC3.

Analysis of the genes repressed by HDAC3 KO in BAT revealed a strong enrichment for thermogenic pathways, including mitochondrial oxidative phosphorylation, tricarboxylic acid cycle (TCA) enzymes, and other metabolic and energy producing processes, with no signal for inflammatory pathways (Fig. 3.2c). Many components of the TCA cycle and complexes I–V of the electron transport chain were diminished in the absence of HDAC3 (Fig. 3.2d and Extended Data Fig. 3.4a), as confirmed by reverse- transcription–quantitative PCR (RT–qPCR) at both 29 °C and 22 °C in pan-adipose as well as BAT-specific HDAC3 KO models (Extended Data Fig. 3.4b, c).

Relative to controls, BAT HDAC3 KO animals had similar body weight and lean mass, with slightly less body fat but indistinguishable basal energy expenditure, activity, and food intake at 22 °C (Extended Data Fig. 3.5). Moreover, after 12 weeks of high-fat feeding at room temperature, pan-adipose and BAT-specific HDAC3 KO mice gained similar amounts of weight and body fat relative to controls (Extended Data Fig. 3.6a–d), whose BAT HDAC3 expression was indistinguishable from mice fed with regular chow

(Extended Data Fig. 3.6e). Additionally, mice acclimated to thermoneutrality and subsequently challenged with fed high-fat diet for 12 weeks exhibited no significant differences in weekly body weights (Extended Data Fig. 3.7a-b). The findings from room temperature are similar to the phenotype of Ucp1 KO mice, and suggestive of Ucp1- 54 independent mechanisms that may control energy expenditure and body weight (Bal et al., 2012; Enerbäck et al., 1997; Kazak et al., 2015; Long et al., 2016; Ukropec et al.,

2006). The thermoneutrality findings on high-fat diet, suggests loss of adipose HDAC3 may control other energy-wasting programs, which compensate for defective Ucp1 expression in diet-induced thermogenesis (Feldmann et al., 2009). Thus, HDAC3 is uniquely required for priming the mitochondrial oxidative and thermogenic energy- expending gene programs in classic BAT to defend against environmental cold.

To determine the mechanism by which HDAC3 drives expression of thermogenic gene pathways, we performed chromatin immunoprecipitation followed by sequencing

(ChIP–seq) in BAT for HDAC3 and for acetylated histone H3 lysine 27 (H3K27ac) as a marker of active enhancers(Rada-Iglesias et al., 2011). Loss of HDAC3 led to reduced

H3K27ac at many sites of HDAC3 binding in control mice, often near genes that were repressed by HDAC3 depletion (Fig. 3.3a). These results imply that, at these sites, the usual function of HDAC3 as a histone deacetylase was overridden by its function as a coactivator. By contrast, H3K27ac was increased at HDAC3 bound sites near genes that were induced in HDAC3 KO BAT (Fig. 3.3a).

Steady-state mRNA levels correlated well with nascent transcription at gene bodies as assessed by global run-on sequencing (GRO-seq) (Core et al., 2008)

(Extended Data Fig. 3.8a). Mining of the GRO-seq data for non-coding enhancer RNA

(eRNA) transcription to map functional enhancers revealed pronounced reduction in activity of multiple enhancers at the Ucp1 locus (Fang et al., 2014; Wang et al., 2011)

(Fig. 3.3b). Genome-wide, HDAC3 binding at eRNAs that were downregulated upon

HDAC3 KO was enriched in BAT of control mice, adding to the evidence that HDAC3 functioned as a coactivator at these enhancers (Fig. 3.3c). Since Rev-erbα (also known 55 as Nr1d1) deletion has the opposite effect on Ucp1, we reason that the ability of HDAC3 to function as Rev-erbα corepressor must be overcome by HDAC3 coactivation of a different transcription factor (Gerhart-Hines et al., 2013).

To identify the transcription factor(s) coactivated by HDAC3, we used genome- wide de novo motif analysis at sites of eRNA downregulation in HDAC3 KO BAT, revealing enrichment of the binding motif for ERR at both 29 °C and 22 °C (Fig. 3.3d and

Extended Data Fig. 3.8b). ERRs are potent activators of mitochondrial metabolism, oxidative phosphorylation (OXPHOS) pathways, and energy metabolism (Giguère,

2008). Other enriched motifs included those of known adipose lineage factors including the BAT-specific EBF2 (Rajakumari et al., 2013; Soccio et al., 2015). Confirming the motif prediction, genome-wide ERRα binding in BAT was strongest at sites where eRNAs were downregulated upon HDAC3 KO (Fig. 3.3e). On a genome-wide level,

HDAC3 and ERRα binding strongly co-localized along with NCoR at downregulated enhancers (Fig. 3.3f), suggesting a novel function of HDAC3. Moreover, endogenous

ERRα and HDAC3 co-immunoprecipitated in mature brown adipocytes (Extended Data

Fig. 3.8c).

Inspection of the Ucp1 locus revealed a super-enhancer (encompassing the eRNAs highlighted in Fig. 3.3b) at sites of co-localized HDAC3 and ERRα (Fig. 3.3g).

Absence of HDAC3 led to marked attenuation of eRNA transcription and decreased levels of H3K27ac (Fig. 3.3g and Extended Data Fig 3.8d, e), along with pronounced reduction of Ucp1 mRNA at 29 °C and 22 °C in both HDAC3 KO models (Fig. 3.3g and

Extended Data Fig. 3.8f). Loss of HDAC3 had little or no effect on ERRα binding at the

Ucp1 locus (Fig. 3.3g and Extended Data Fig. 3.8 g). Genetic ablation of ERRα

56 decreased Ucp1 expression and eRNA transcription in BAT (Extended Data Fig. 3.8h, i) as well as in mature brown adipocytes (Extended Data Fig. 3.8j), albeit to a lesser degree than HDAC3 KO perhaps because of redundancy with ERRγ (Dixen et al., 2013).

These data strongly suggest that HDAC3 is a critical coactivator of ERR-driven thermogenic gene transcription in BAT.

Transcriptional activation by ERRα is highly dependent upon the coactivator peroxisome proliferator-activated receptor gamma coactivator 1 alpha (Pgc-1α, also known as Ppargc1α) (Giguère, 2008; Rodgers et al., 2008). Notably, the transcriptional function of PGC-1α is activated by lysine deacetylation, suggesting the possibility that

HDAC3 may activate ERR-dependent transcription via deacetylation of PGC-1α

(Rodgers et al., 2008). Indeed, HDAC3 markedly increased the transcriptional activity of

ERRα at a major HDAC3-bound enhancer of the Ucp1 gene in the presence of WT

PGC-1α but not a mutant form of PGC-1α that is unable to directly interact with ERRα

(Fig. 3.4a). Both PGC-1α and HDAC3 co-immunoprecipitated with ERRα (Extended

Data Fig. 3.9a), and HDAC3 robustly deacetylated PGC-1α that had been acetylated by the histone acetyltransferase GCN5, both in cells and in vitro, and in a manner that was prevented by the class I HDAC inhibitor trichostatin A (Rodgers et al., 2008) (Fig. 3.4b, c). HDAC3 also reversed GCN5-mediated repression of PGC-1α coactivator function

(Extended Data Fig. 3.9b). Moreover, ectopic expression of non-acetylatable but not WT

PGC-1α in primary brown adipocytes lacking HDAC3 (Extended Data Fig. 3.9c) markedly induced expression of Ucp1 but not the adipocyte differentiation marker Fasn

(Extended Data Fig. 3.9d).

In addition to the post-translational regulation of PGC-1α, the Pgc-1α genomic locus was bound by HDAC3 and ERRα at sites of potent enhancer activity that were lost in the absence of HDAC3 (Fig. 3.4d), and HDAC3 functioned as an activator at this enhancer (Fig. 3.4e). Indeed, Pgc-1α mRNA and protein were markedly reduced in

HDAC3 KO BAT (Fig. 4f, g). Of note, combined deficiency of Pgc-1α and Pgc-1β (also known as Ppargc1β) leads to loss of BAT Ucp1 expression and impaired mitochondrial function, reminiscent of the BAT HDAC3 KO phenotype (Uldry et al., 2006). Enhancers at the gene encoding PGC-1β were also bound by HDAC3 and ERRα (Extended Data

Fig. 3.10a), and Pgc-1β mRNA was markedly reduced in HDAC3 KO BAT (Extended

Data Fig. 3.10b). Moreover, eRNAs at these sites were dependent upon HDAC3

(Extended Data Fig. 3.10c, d). Further, depletion of PGC-1α and β and/or ERRα in mature brown adipocytes decreased Ucp1 transcription (Extended Data Fig. 3.10e).

Thus, the ability of HDAC3 to activate PGC-1α both transcriptionally and post- translationally amplifies its coactivator function at ERR family-regulated enhancers in a feed-forward manner (Fig. 3.4h).

Intriguingly, the transcription of Ucp1 and Pgc-1α was even lower in the HDAC3

KO fat at thermoneutrality than at room temperature (Extended Data Fig. 3.10f), indicating that the effect of loss of HDAC3 on basal expression of these genes is more potent than eliminating sympathetic stimulation of BAT. This pattern of expression

(HDAC3 KO < thermoneutrality < room temperature) was identified in many additional genes (Fig. 3.4i) critical for oxidative phosphorylation and energy metabolism (Fig. 3.4i, j). Thus, a critical function of HDAC3 in BAT is to maintain a minimal basal expression of

Ucp1 and other genes regulating energy metabolism.

Discussion

The maintenance and protection of core body temperature through tight control of metabolism is a defining element of mammalian physiology. Our finding that BAT

HDAC3 transcriptionally maintains the basal activity of thermogenic gene networks provides a physiological basis for the ability to rapidly generate heat in response to acute exposure to dangerously cold environmental temperatures. HDAC3 has a unique role in priming the brown adipocyte to support immediate thermogenic respiration and heat production without delay. Thus, distinct from adaptive (inducible) thermogenic mechanisms that require time to develop and respond, HDAC3 sets a basal thermogenic tone which establishes the facultative capacity of BAT to turn on rapidly, on-demand, to ensure readiness against acute and life-threatening cold exposure.

Figure 3.1: HDAC3 controls BAT thermogenesis.

Figure 3.1: HDAC3 controls BAT thermogenesis. a, b, Effect of acute cold exposure from standard housing at 22 °C to 4 °C on Adipoq-cre HDAC3 KO (Hdac3f/f) mice versus control littermates (n = 15, 8), Ucp1-cre HDAC3 KO mice versus control littermates

(n = 15, 7), and Ucp1-/- mice (n = 15): a, core body temperature; b, survival. c, Oxygen consumption rates of anaesthetized Adipoq-cre HDAC3 KO mice versus control littermates (n = 12, 5), Ucp1-cre HDAC3 KO mice versus control littermates (n = 6, 5), and Ucp1-/- mice (n = 5) after injection of 1 mg per kg (body weight) noradrenaline. d,

Mitochondrial respiration of purified BAT homogenates from Adipoq-cre HDAC3 KO mice versus control littermates (n = 5, 6), and Ucp1-cre HDAC3 KO mice versus control littermates (n = 6, 5), after brief acclimation to thermoneutrality. Mitochondria were provided palmitoylcarnitine and pyruvate substrates. UCP1-dependent respiration was assessed upon addition of guanosine diphosphate (GDP), and coupled respiration rates of complexes I, II, and IV (CI, CII, CIV) were determined in the presence of adenosine diphosphate (ADP). e, Representative haematoxylin and eosin staining of interscapular brown adipose from 10- to 12-week-old mice at 22 °C. Scale bars, 100 mm. *P < 0.05,

**P < 0.01, ***P < 0.001, ****P < 0.0001 as analysed by two-way analysis of variance

(ANOVA) and a Tukey’s post-test (a, c), log-rank (Mantel–Cox) test (b), or two-tailed

Student’s t-test (d). Data are represented as mean ± s.e.m.

Figure 3.2: HDAC3 is required for expression of UCP1 and OXPHOS genes in BAT.

Figure 3.2: HDAC3 is required for expression of UCP1 and OXPHOS genes in BAT. a, Scatterplot of RNA-seq data showing HDAC3-regulated BAT genes from Adipoq-Cre

HDAC3 KO versus control littermates (n = 4, 4) adapted to thermoneutrality (fold change

>1.5 up (red) or down (blue) and false discovery rate (FDR) < 0.01). RPKM, reads per kilobase per million. b, Immunoblot from BAT of Adipoq-Cre HDAC3 KO, control littermates, or Ucp1-/- adapted to 29 °C or maintained at 22 °C (n = 3, 3; n = 3, 3; n = 2). c, Gene Ontology (GO) and pathway analysis of downregulated genes identified by

RNA-seq and selected by Enrichr combined score (KP, KEGG Pathway; BP, Biological

Process; CC, Cellular Component). d, Heat map depicting downregulated genes identified in Gene Ontology analysis.

Figure 3.3.: HDAC3 functions as an ERRα coactivator in BAT.

Figure 3.3: HDAC3 functions as an ERRα coactivator in BAT. a, Average H3K27ac

ChIP-seq profiles of Adipoq-Cre HDAC3 KO mice versus control littermates (n = 3, 3; pooled biological replicates per library) at enhancers bound by HDAC3 within 100 kilobases (kb) of transcription start sites of HDAC3 KO-regulated genes by GRO-seq

(fold change >1.5 or <0.5). Decreased H3K27ac at HDAC3 sites near repressed genes

(n = 1,085, P = 1.0 × 10-123) and increased H3K27ac at HDAC3 sites near induced genes

(n = 897, P = 9.0 × 10-56) upon loss of HDAC3. RPM, reads per million. b, Scatter plot of enhancer RNAs (eRNAs) measured by GRO-seq in Adipoq-Cre HDAC3 KO mice versus control littermates (n = 10, 10; pooled biological replicates per library) at 29 °C, highlighting induced and repressed eRNAs (red, fold change >2; blue, fold change <0.5). c, Average HDAC3 ChIP-seq profile of control littermates (n = 5, pooled biological replicates per library) at enhancers with repressed or unchanged eRNAs in HDAC3 KO.

Significant HDAC3 binding found at HDAC3 KO-repressed eRNAs relative to unchanged eRNAs (P = 2.4 × 10-215). d, De novo motif search at eRNA sites repressed in HDAC3

KO at 29 °C (ranked by P value). e, Average ERRα ChIP-seq profile in control littermates (n = 5, pooled biological replicates per library) at enhancers with differential eRNAs in HDAC3 KO. Significant ERRα found at HDAC3 KO-repressed eRNAs relative to unchanged eRNAs (P = 1.0 × 10-288). f, Heat map depicting HDAC3, ERRα, and

NCoR co-localization at enhancers with repressed eRNAs in HDAC3 KO mice. g,

Genome browser tracks of the Ucp1 super-enhancer locus highlighting GRO-seq (22 °C,

29 °C), RNA-seq (29 °C), and ChIP-seq (29 °C) data (y axis scales in brackets: reads per million; eRNA tracks feature adjusted y axis scale). P values for ChIP-seq or motif search determined by Wilcoxon or binomial test, respectively.

Figure 3.4: HDAC3 coactivation of ERRα is mediated by PGC-1α deacetylation.

Figure 3.4: HDAC3 coactivation of ERRα is mediated by PGC-1α deacetylation. a,

Luciferase reporter assay of transcription driven by an identified Ucp1 enhancer (−6 kb), demonstrating effects of ERRα, HDAC3, WT PGC-1α, and/or a PGC-1α LXXLL mutant

(L1/2/3A) unable to interact with ERRα (n = 3 replicates per condition). b, Immunoblot analysis of PGC-1α lysine acetylation after immunoprecipitation from co-transfected

293FT cells. c, Immunoblot analysis of an in vitro deacetylation reaction of purified acetylated-PGC-1α by recombinant human HDAC3, with or without trichostatin A (TSA). d, Genome browser tracks of the Ppargc1a locus highlighting GRO-seq and RNA-seq data, and eRNAs at HDAC3, ERRα, and NCoR co-bound distal enhancers (boxed). e,

Luciferase reporter assay as in a for identified Ppargc1a distal enhancer (−38 kb), (n = 3 replicates per condition). f, BAT Pgc-1α mRNA from Adipoq-Cre HDAC3 KO mice versus control littermates (n = 9, 6) and Ucp1-Cre HDAC3 KO mice versus control littermates (n = 5, 6). g, Immunoblot of PGC-1α in BAT nuclear extract of Adipoq-Cre

HDAC3 KO versus control littermates (n = 5, pooled biological replicates per lane). h,

BAT HDAC3 deacetylates PGC-1α to coactivate an ERR-driven transcriptional loop of

Ppargc1a/b, Ucp1, and OXPHOS genes. i, Flow chart depicting unbiased hierarchical gene clustering of GRO-seq gene transcription by temperature and genotype. Boxplot displays gene cluster requiring HDAC3 coactivator function, where line denotes median, top/bottom of boxes represents first/third quartiles, and whiskers denote 1.5 times the interquartile range from first/third quartiles. j, Heatmap of thermogenic and OXPHOS genes identified in i. ***P < 0.001, ****P < 0.0001 as determined by one-way ANOVA with multiple comparisons and a Tukey’s post-test (a, e) or two-tailed Student’s t-test (f).

Data are represented as mean ± s.e.m.

Extended Data Figure 3.1: Ablation of HDAC3 in adipose tissue depots.

Extended Data Figure 3.1: Ablation of HDAC3 in adipose tissue depots. a,

Immunoblot analysis of interscapular BAT, iWAT, and eWAT of Adipoq-Cre HDAC3 KO and control littermates, or Ucp1-Cre HDAC3 KO and control littermates maintained at

22 °C (n = 2, all groups) demonstrating tissue-specific conditional KO of HDAC3. b,

Interscapular BAT mass, (c) Relative BAT mitochondrial number, and (d) total body mass from Adipoq-Cre HDAC3 KO and Ucp1-Cre HDAC3 KO versus control littermates maintained at 22 °C (n = 13 Adipoq-Cre, n = 9 control; n = 9 Ucp1-Cre, n = 10 control). e, Representative haematoxylin and eosin (H&E) staining of inguinal white adipose from

10- to 12-week-old Adipoq-Cre HDAC3 KO, Ucp1-Cre HDAC3 KO, Ucp1 KO, or control mice housed at 22 °C. Scale bars, 100 mm. Data are represented as mean ± s.e.m.

Extended Data Figure 3.2: BAT HDAC3 is required for cold-mediated induction of Ucp1 expression, and HDAC3 expression is not altered by acute cold.

Extended Data Figure 3.2: BAT HDAC3 is required for cold-mediated induction of

Ucp1 expression, and HDAC3 expression is not altered by acute cold. a, b, BAT

Ucp1 mRNA levels following a 3 h exposure to 4 °C (from 22 °C) versus control littermates maintained at 22 °C in (a) Adipoq-Cre HDAC3 KO versus control (n = 5, 5, per temperature) and (b) Ucp1-Cre HDAC3 KO versus control (n = 5, 5, per temperature). c, d, iWAT Ucp1 mRNA levels following 3 h exposure to 4 °C, versus control littermates maintained at 22 °C in (c) Adipoq-Cre HDAC3 KO versus control

(n = 5, 5, per temperature) and (d) Ucp1-Cre HDAC3 KO versus control (n = 5, 5, per temperature). e, BAT HDAC3 mRNA expression levels following a 3 h exposure to 4 °C

(from 22 °C) versus control littermates maintained at 22 °C (n = 5, 5, per temperature). f,

BAT HDAC3 protein levels following 3 h acute cold exposure at 4 °C (from 22 °C) versus control littermates maintained at 22 °C. VCL, vinculin. NS, not significant, *P < 0.05,

**P < 0.01, ***P < 0.001, ****P < 0.0001 as determined by a two-way ANOVA and Holm–

Šidák’s post-test (a–d) or two-tailed Student’s t-test (e). Data are represented as mean ± s.e.m.

Extended Data Figure 3.3: HDAC3 is neither induced nor required for brown adipogenesis, but required for cell-autonomous Ucp1 expression.

Extended Data Figure 3.3: HDAC3 is neither induced nor required for brown adipogenesis, but required for cell-autonomous Ucp1 expression. a, Gene expression spanning differentiation of cultured WT primary brown adipocytes (n = 5 replicates per time point). b, Depletion of HDAC3 in day 8 cultured mature brown adipocytes after addition of 2 mm 4-hydroxytamoxifen (4-OHT) during days 0–2 of differentiation to Rosa26-CreER-positive (HDAC3 KO) and Rosa26-CreER-negative

(control) cells derived from littermates (n = 3, 3). c, Adipocyte-specific gene expression in cultured primary brown adipocytes after depletion of HDAC3 versus control (n = 3, 3). d, Assessment of lipid accumulation (evaluated by Oil Red-O staining) in cultured

HDAC3 KO versus control primary brown adipocytes. e, Ucp1 mRNA expression in cultured primary brown adipocytes after depletion of HDAC3 versus control (n = 3, 3). f,

UCP1 protein expression in cultured primary brown adipocytes after depletion of HDAC3 versus control. (n = 3, 3). g, Ucp1 mRNA expression in cultured primary brown adipocytes after depletion of HDAC3 versus control and treated with vehicle (ethanol) or isoproterenol (1 mm) for 3 h (n = 4 per group). *P < 0.05, **P < 0.01, ***P < 0.001,

****P < 0.0001, as determined by a two-tailed Student’s t-test (b, c, e) or by a two-way

ANOVA and Holm–Šidák’s post-test (g). Data are represented as mean ± s.e.m.

Extended Data Figure 3.4: HDAC3 is required for expression of mitochondrial OXPHOS and TCA cycle genes.

Extended Data Figure 3.4: HDAC3 is required for expression of mitochondrial

OXPHOS and TCA cycle genes. a, Bioinformatic extension of identified Gene Ontology categories (Fig. 2c.) to all oxidative phosphorylation and TCA cycle genes as retrieved by the HUGO gene nomenclature database. *Gene expression change in RNA-seq data set with an FDR <0.01. b, c, RT–qPCR verification of gene expression changes highlighted in Fig. 2d: b, Adipoq-Cre HDAC3 KO versus control littermates at 29 °C

(upper, n = 9, 6) and 22 °C (lower, n = 9, 7); c, Ucp1-Cre HDAC3 KO versus control littermates at 29 °C (upper, n = 5, 6) and 22 °C (lower, n = 5, 7). *P < 0.05, **P < 0.01,

***P < 0.001, as determined by a two-tailed Student’s t-test. Data are represented as mean ± s.e.m.

Extended Data Figure 3.5: Metabolic studies of Adiponectin-Cre and Ucp1-Cre HDAC3 KO mouse models

Extended Data Figure 3.5: Metabolic studies of Adiponectin-Cre and Ucp1-Cre

HDAC3 KO mouse models. a, b, NMR analysis of body composition: a, Adipoq-Cre mice versus control littermates (n = 8, 11); b, Ucp1-Cre mice versus control littermates

(n = 7, 9). c–n, CLAMS metabolic cage analysis. c, d, Oxygen consumption (VO2): c,

Adipoq-Cre HDAC3 KO versus control littermates (n = 6, 5); d, Ucp1-Cre HDAC3 KO versus control littermates (n = 6, 6). e, f, ANCOVA of VO2 (linear regression analysis of total body mass and oxygen consumption): e, Adipoq-Cre HDAC3 KO versus control littermates (n = 6, 5); f, Ucp1-Cre HDAC3 KO versus control littermates (n = 6, 6). g, h,

Respiratory exchange ratio (RER): g, Adipoq-Cre HDAC3 KO versus control littermates

(n = 6, 5); h, Ucp1-Cre HDAC3 KO versus control littermates (n = 6, 6). i, j, Heat measurements (kcal h−1): i, Adipoq-Cre HDAC3 KO versus control littermates (n = 6, 5); j, Ucp1-Cre HDAC3 KO versus control littermates (n = 6, 6). k, l, Food intake: k, Adipoq-

Cre HDAC3 KO versus control littermates (n = 6, 5); l, Ucp1-Cre HDAC3 KO versus control littermates (n = 6, 6). m, n, Physical activity: m, Adipoq-Cre HDAC3 KO versus control littermates (n = 6, 5); n, Ucp1-Cre HDAC3 KO versus control littermates (n = 6,

6). P values shown in italics. CLAMS data are graphed as rolling averages. NS, not significant, *P < 0.05 as determined by a two-tailed Student’s t-test (a–d, g–n) or

ANCOVA (e, f). Data are represented as mean ± s.e.m.

Extended Data Figure 3.6: Effect of high fat diet on Adiponectin-Cre and Ucp1- Cre HDAC3 KO mouse models.

Extended Data Figure 3.6: Effect of high fat diet on Adiponectin-Cre and Ucp1-Cre

HDAC3 KO mouse models. Twelve-week-old weight-matched HDAC3 KO and control littermates were fed high-fat diet (HFD) for 12 weeks. a, Weekly body weights, (n = 8

Adipoq-Cre, n = 10 control). b, Body composition analysis by NMR (n = 8 Adipoq-Cre, n = 10 control). c, Weekly body weights, (n = 7 Ucp1-Cre, n = 7 control). d, Body composition analysis by NMR (n = 7 Ucp1-Cre, n = 7 control). e, RT–qPCR of BAT

HDAC3 mRNA expression after 12 weeks high-fat diet versus regular chow fed controls

(n = 7, 5, respectively). Data are represented as mean ± s.e.m.

Extended Data Figure 3.7: Effect of high fat diet on Adiponectin-Cre and Ucp1- Cre HDAC3 KO mouse models housed at thermoneutrality.

Twelve-week-old weight-matched HDAC3 KO and control littermates were fed high-fat diet (HFD) for 12 weeks at thermoneutrality. a, Weekly body weights, (n = 8 Adipoq-Cre, n = 12 control). b, Weekly body weights, (n = 10 Ucp1-Cre, n = 10 control). Data are represented as mean ± s.e.m.

Extended Data Figure 3.8: Transcriptional role of HDAC3 and ERRα in BAT.

Extended Data Figure 3.8: Transcriptional role of HDAC3 and ERRα in BAT. a, Heat map demonstrating correlation of RNA-seq and GRO-seq data. Differentially expressed genes in RNA-seq or GRO-seq data were sorted by log2(fold change) in

RNA-seq. b, De novo motif enrichment at repressed eRNAs in Adipoq-Cre, HDAC3 KO mice versus control littermates (n = 10, 10; pooled biological replicates per library) maintained at 22 °C and ranked by P value. c, Endogenous HDAC3 co- immunoprecipitation of ERRα in differentiated mature brown adipocytes. d, e, RT– qPCRs of BAT Ucp1 eRNA expression and (f) Ucp1 mRNA at 22 °C and 29 °C in

Adipoq-Cre and Ucp1-Cre HDAC3 KO mice versus control littermates, 29 °C (n = 9

Adipoq-Cre, 6 control; n = 5 Ucp1-Cre, 6 control) and 22 °C (n = 9 Adipoq-Cre, 7 control; n = 5 Ucp1-Cre, 7 control). g, ChIP–qPCR of ERRα at Ucp1 enhancers in Adipoq-Cre

HDAC3 KO versus control littermates (n = 3, 3) adapted to 29 °C. h, RT–qPCR of

ERRα and Ucp1 mRNA expression and (i) Ucp1 eRNA expression in ERRα KO BAT versus control littermates (n = 8, 7). j, RT–qPCR analysis of ERRα and Ucp1 mRNA expression in mature brown adipocytes after siRNA-mediated knockdown of

ERRα versus scramble 72 h after transfection (n = 3, 3). *P < 0.05, **P < 0.01,

***P < 0.001 as determined by a two-tailed Student’s t-test (d–e, g–j), two-way ANOVA with Holm–Šidák’s post-test (f). P values for motif enrichment as determined by binomial test (b). Data are represented as mean ± s.e.m.

Extended Data Figure 3.9: Role of HDAC3 on PGC-1α acetylation and function.

Extended Data Figure 3.9: Role of HDAC3 on PGC-1α acetylation and function. a, Co-immunoprecipitation of HDAC3 and PGC-1α with ERRα from 293FT cells. b, Luciferase reporter assay of transcription driven by the major Ucp1 enhancer

(−6 kb) after transfection of ERRα, PGC-1α, GCN5, and/or HDAC3 (n = 3 replicates per condition). c, d, Primary brown pre-adipocytes from Rosa26-CreER-positive HDAC3F/F and HDAC3F/F control littermates transduced with MSCV retroviruses: control, PGC-1α

WT, or non-acetylatable PGC-1α R13 mutant, treated with 2 mm 4-hydroxytamoxifen during days 0–2 of differentiation to deplete HDAC3, and studied at day 8 of differentiation. c, Immunoblot analysis of exogenous PGC-1α expression in primary brown adipocytes (n = 2 replicates pooled per lane). d, RT–qPCR analysis of Ucp1 and

Fasn expression in control and HDAC3 KO primary brown adipocytes following transduction with MSCV-Control (n = 4 control, 3 HDAC3 KO), MSCV-PGC-1α WT

(n = 4 control, 4 HDAC3 KO), or MSCV-PGC-1α R13 (non-acetylatable mutant) (n = 3

Control, 4 HDAC3 KO). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, as determined by a one-way ANOVA and a Tukey’s post-test (b, d). Data are represented as mean ± s.e.m.

Extended Data Figure 3.10: HDAC3 and ERRα activate Ppargc1b enhancers and transcription.

Extended Data Figure 3.10: HDAC3 and ERRα activate Ppargc1b enhancers and transcription. a, Genome browser tracks of the Ppargc1b locus highlighting

GRO-seq and ChIP-seq data from HDAC3 KO and control BAT (y axis scales, normalized reads, reads per million) demonstrating co-binding of HDAC3, ERRα, and

NCoR at functional enhancers. b, BAT PGC-1β mRNA levels in Adipoq-Cre HDAC3 KO

BAT versus control littermates (29 °C: n = 9, 6; 22 °C: n = 9, 7). c, d, RT–qPCR of eRNAs found at HDAC3 and ERRα enhancers in Adipoq-Cre HDAC3 KO BAT versus control littermates (29 °C: n = 9, 6; 22 °C: n = 9, 7) and Ucp1-Cre HDAC3 KO BAT versus control littermates (29 °C: n = 5, 6; 22 °C: n = 5, 7). e, RT–qPCR analysis of

Ucp1 mRNA expression in mature brown adipocytes after combinatorial siRNA knockdown of Pgc-1α, Pgc-1β, and/or ERRα versus scramble siRNA (n = 5 replicates per condition). Statistical analysis performed among groups transfected with siRNAs. f,

Quantification of Ucp1 and Pgc-1α nascent gene body transcription (GRO-seq) at 22 °C and 29 °C in Adipoq-Cre HDAC3 KO BAT versus control littermates (n = 10, 10, pooled biological replicates per library). *P < 0.05, **P < 0.01, ***P < 0.001, as determined by a two-tailed Student’s t-test (b–d), one-way ANOVA and a Holm–Šidák post-test (e), or an exact test (performed in EdgeR). Data are represented as mean ± s.e.m.

Chapter 4

Brown adipose tissue HDAC3-dependent cold survival is circumvented by environmentally induced epigenomic memory.

In preparation.

Abstract

Brown adipose tissue liberates stored chemical energy as heat through substrate oxidation and uncoupled respiration to protect against hypothermia (Cannon and

Nedergaard, 2004; Harms and Seale, 2013). The basal thermogenic capacity of brown adipose tissue is established transcriptionally by histone deacetylase 3 (HDAC3) mediated enhancer activation (Emmett et al., 2017). However, the existence of alternative HDAC3-independent thermogenic pathways promoting cold survival are unknown. Here, we demonstrate that brief environmental exposure to moderate cold circumvents the genetic requirement of HDAC3 for Ucp1 activation and cold survival, signifying an epigenomic rescue. Brown adipose genetic ablation of HDAC3 results in deficient Ucp1 expression and mice briefly exposed to a cooler environment rescue

Ucp1. Remarkably, knockout mice briefly exposed to moderate cold and subsequently returned to ambient temperature, demonstrate an acquired physiological aptitude to avert hypothermia upon acute exposure to dangerously low temperatures.

Environmental exposure imposes an HDAC3-independent epigenomic memory supporting Ucp1 and oxidative phosphorylation gene expression lasting days. Notably, moderate cold activates BAT enhancers bound by Estrogen-related receptor α (ERRα) requiring the PGC-1α coactivator. Moderate cold induced PGC-1α isoforms bypass the requirement for HDAC3, as BAT-specific depletion of HDAC3 and PGC-1α largely abrogate moderate cold induced Ucp1 transcription. Thus, while brown adipose HDAC3 primes the basal thermogenic tone, past experience of moderate cold imparts an epigenomic memory to anticipate and ensure survival of probable impending dangerously cold temperatures.

Introduction

Brown adipose tissue (BAT) is a major site of mammalian thermogenesis and contributes to the facultative and adaptive responses to cold( Cannon and Nedergaard,

2004; Harms and Seale, 2013). As BAT is a primary tissue responsible for the generation of heat upon cold stimulus, brown adipocytes are dense in mitochondria and express a unique gene program to facilitate substrate oxidation required for thermogenesis. The critical effector of brown adipocyte thermogenic function is uncoupling protein 1 (UCP1), which resides in the inner mitochondrial membrane and dissipates the proton motive force upon activation by fatty acid and reactive oxygen species to generate heat (Chouchani et al., 2016; Fedorenko et al., 2012). In addition to the thermogenic function of brown fat, recent evidence suggests that brown and ‘brown- like’ adipocytes may additionally provide beneficial effects to human metabolism through the combustion of stored calories to counteract diabetes, obesity, and insulin resistance

(Cypess et al., 2009; Van Marken Lichtenbelt et al., 2009; Nedergaard et al., 2007; Saito et al., 2009; Virtanen et al., 2009). While BAT may counter the harmful metabolic effects of adult humans, there exists great variation of both BAT mass and activity levels amongst individuals (Ouellet et al., 2011; Persichetti et al., 2013). The reasons and mechanisms underlying BAT heterogeneousness among humans are unknown, but they are likely to involve genetic, epigenetic, and environmental influences. Thus, the mechanisms that influence brown adipose tissue function in thermogenesis and metabolism are of much interest.

HDAC3 has emerged as an important regulator of the brown fat thermogenic program (Emmett et al., 2017). As the sole histone deacetylase of the nuclear receptor corepressor (NCoR) complex, HDAC3 physically interacts with the nuclear receptor 89 corepressor molecules, NCOR and SMRT, which bind to and modulate nuclear receptor transcriptional activity through the deacetylation of histone and non-histone targets

(Guenther et al., 2001). HDAC3 has demonstrated a plethora of vital and pleiotropic roles in mammalian physiology and metabolism, such as cardiac energy metabolism(Montgomery et al., 2008; Sun et al., 2011), skeletal muscle exercise endurance(Hong et al., 2016), liver hepatosteatosis (Feng et al., 2011; Sun et al., 2012), and macrophage alternative activation (Hoeksema et al., 2014). In BAT, HDAC3 surprisingly functions as a coactivator of the nuclear receptor estrogen-related receptor

α (ERRα) function, through the deacetylation and activation of peroxisome proliferator- activated receptor gamma coactivator 1-alpha (PGC-1α) to prime the basal transcription of thermogenic genes (Emmett et al., 2017; Rodgers et al., 2005; Villena et al., 2007).

Brown adipose tissue-specific loss-of-function studies, demonstrated a genetic requirement for HDAC3 in activating the basal transcription of Ucp1, Pgc-1α, and oxidative phosphorylation genes, to ensure survival of acute life-threatening cold temperatures (Emmett et al., 2017).

Cold survival is Ucp1-dependent in C57BL/6 mice acclimated to the standard housing temperatures of 22 °C (Enerbäck et al., 1997; Golozoubova et al., 2006).

Despite the critical importance for Ucp1 gene expression in surviving acute cold from this ambient temperature, it has been demonstrated that mice genetically deficient for the Ucp1 gene may be gradually acclimated to a cold environment, and multiple Ucp1- independent pathways facilitate this metabolic adaptation (Bal et al., 2012; Kazak et al.,

2015; Long et al., 2016; Ukropec et al., 2006). Despite this ability, cold adaptation in

Ucp1 knockout C57BL/6 mice may take weeks to develop, while some mice succumb to cold during the acclimation period (Ukropec et al., 2006), highlighting both the critical

90 importance of Ucp1-dependent thermogenesis as well as the complexity of Ucp1- independent pathways. Nevertheless, both pathways prove important to our understanding of metabolism as well as the interaction of genes-to-environment, and vice-versa. The vast variety of mechanisms uncovered using the Ucp1 knockout mouse model to study cold adaptation, suggests that in other models of impaired brown fat thermogenesis, there may yet exist unappreciated regulatory and compensatory mechanisms contributing to mammalian physiology.

Thus, we sought to study the ability of the extremely cold-intolerant BAT-specific

HDAC3 knockout mice, to adapt and survive environmental cold (Emmett et al., 2017).

Specifically, we utilized previously described mouse models using Ucp1-cre or

Adiponectin-cre (Adipoq-cre) to deplete HDAC3 in C57BL/6 mice in a brown-specific or adipose-specific fashion, respectively. Here, we report that brief environmental exposure of BAT-specific HDAC3 knockout mice to moderately cold temperature restores defective Ucp1 expression through ERRα enhancer activation and improves the thermogenic capacity of mice, thus enabling survival of extreme cold. The restoration of defective Ucp1 expression and its lasting duration of several days are indicative of a cold-dependent epigenomic memory, which occurs in both control and HDAC3-deficient mice to prepare against future bouts of cold. This provides a unique insight into the molecular gene-environment interactions that occur in brown adipose and have physiological significance.

Results

Moderate Cold Temperature Initiates an HDAC3-Independent Thermogenic Pathway in Brown Adipose Tissue.

Ucp1-cre and Adipoq-cre mice crossed to mice with a conditional HDAC3 allele on the C57BL/6 background, results in profound hypothermia upon extreme cold from a thermogenic deficiency of brown adipose. These BAT-specific HDAC3 knockout (KO) mice were utilized to uncover putative mechanisms of adaptation to harsh environmental cold. Since, HDAC3 knockout mice succumb to hypothermia upon moving from 22 °C to

4 °C, similar to Ucp1 knockout mice (Enerbäck et al., 1997), we hypothesized that

HDAC3 knockouts may also survive and adapt to a more modest decrease in ambient environmental temperature. To test this hypothesis, we moved mice from 22 °C to 15 °C, representing a 7-degree drop in housing temperature. To our surprise, all BAT-specific

HDAC3 KO mice survived this 24-hour exposure to lower temperature and exhibited no physical signs of cold distress.

Quantification of mRNA yielded the expected ablation of HDAC3 mRNA in Ucp1- cre-positive animals as compared to their matched Ucp1-cre-negative control littermates at 22 °C and following exposure to 15 °C for 24 hours (Figure 4.1a). Surprisingly, Ucp1 mRNA was robustly induced in HDAC3 KO mice briefly housed at 15 °C for 24 hours, whereas littermate HDAC3 KO mice remaining at room temperature exhibited nearly absent Ucp1 mRNA (Figure 4.1b). We next assessed whether the rescued Ucp1 transcription in HDAC3 knockout mice normally deficient for Ucp1 mRNA, resulted in increased UCP1 at the protein level. Indeed, assessment of UCP1 protein levels in mice following 24 hours at 15 °C resulted in a complete rescue of defective UCP1 protein expression in HDAC3 KO BAT, while littermate knockouts exhibited a complete loss of

UCP1 protein at 22 °C (Figure 4.1c). We further confirmed these results in a second model of brown fat HDAC3 ablation using Adipoq-cre (Supplementary Figure 4.1a),

92 which demonstrated the same observations at the mRNA and protein levels (Extended

Data Figure 4.1b and c).

Consistent with impaired thermogenic gene expression and substrate oxidation,

HDAC3 KO animals displayed larger lipid droplets under standard housing conditions, which decreased upon exposure to moderate cold (Figure 4.1d). These results suggest that moderate environmental cold can initiate and circumvent defective HDAC3 KO Ucp1 expression as well as re-activate pathways needed for the oxidation of stored lipid substrate. We thus hypothesized that a brief exposure of HDAC3 KO mice to moderate cold, may exhibit physiological impact on cold tolerance and survival.

Brief Exposure to Moderate Environmental Cold Imparts a Lasting Thermogenic Aptitude for Cold-Survival in HDAC3 Knockout Mice.

We tested our hypothesize using the moderate cold exposure regimen as a starting point for physiological studies of brown adipose tissue function in the HDAC3

KO mouse models. Previous studies demonstrating the severe cold intolerance and lethal hypothermia in HDAC3 KO mice were performed starting from standard housing conditions of 22 °C, where mice were acutely transferred to 4 °C for acute cold tolerance testing(Emmett et al., 2017). Notably, C57BL/6 mice acclimated to 22 °C require Ucp1 for cold survival upon exposure to 4 °C (Enerbäck et al., 1997; Golozoubova et al., 2001,

2006). Therefore, we developed a regimen to assess the physiological impact of prior moderate cold exposure on HDAC3 KO mice and control littermates following a return of the mice to 22 °C (Extended Data Figure 4.2a).

Control animals and HDAC3 KO littermates, for Ucp1-cre and Adipoq-cre mouse lines, acclimated to 22 °C were challenged with a precipitous drop in ambient

93 temperature to 4 °C. As anticipated, control littermates demonstrated a remarkable ability to preserve core body temperature upon challenge at 4 °C (Figure 4.2a).

However, as expected based on defective Ucp1 expression (Enerbäck et al., 1997), the

Ucp1-cre and Adipoq-cre HDAC3 KO mice became severely hypothermic following merely 1-2 hours of physiologic challenge at 4 °C (Figure 4.2a)(Emmett et al., 2017).

The profound inability of all HDAC3 KO mice to protect core body temperature was lethal for all mice, whereas control littermates thrived upon cold challenge (Figure 4.2b).

Following 24 hours of environmental exposure at 15 °C, HDAC3 KO mice and control littermates were returned to standard housing conditions at 22 °C for 24 hours of re-acclimation and recovery (designated the ‘24-hour recovery’ group) (Extended Data

Figure 4.2a). The 24-hour recovery group was then challenged with the acute environmental challenge at 4 °C. Amazingly, both the Adipoq-cre and Ucp1-cre HDAC3

KO mice thrived at 4 °C and vigorously defended their core body temperatures in the absence of any visible signs of physiological stress (Figure 4.2c). Notably, both HDAC3

KO mice and control littermates survived the duration of cold challenge (Figure 4.2d) unlike mice previously tested from a constant 22 °C environment.

Next, a third group of HDAC3 KO mice and control littermates where exposed to

15 °C for 24 hours and allowed to recover and acclimate back to room temperature for 7 days (designated the ‘7 day recovery’ group). Remarkably, HDAC3 KO mice, like their littermate controls, were again capable of vigorously defending their core body temperature and demonstrated no propensity towards the development of hypothermia

(Figure 4.2e). Prior moderate cold enabled all HDAC3 KO mice tested to survive the six- hour severe cold challenge (Figure 4.2f), and this thermogenic aptitude remained one week following a brief exposure to a mild environmental change. 94

In contrast to mice housed at constant ambient temperatures, HDAC3 KO mice, subjected to a modest shift in temperature change, demonstrate a remarkable and enduring physiological change in thermogenic brown fat capacity, despite the absence of

HDAC3. As HDAC3 is genetically required for Ucp1 expression under standard housing conditions, and is the sole genetic difference between the mice studied here, our results suggest that a unique environment-genome reprogramming occurs to facilitate the restoration of Ucp1 expression and perhaps additional Ucp1-independent components of the thermogenic program. To test the hypothesis that moderate cold may induce both

Ucp1-dependent and Ucp1-independent pathways (Bal et al., 2012; Golozoubova et al.,

2001; Granneman et al., 2003; Kazak et al., 2015; Long et al., 2016; Ukropec et al.,

2006), we genetically deleted the Ucp1 gene using a congenic C57BL/6 knockout allele crossed to the tissue specific HDAC3 KO mice, generating HDAC3/UCP1 double knockout (DKO) mice.

Acquired Thermogenic Aptitude in HDAC3 Knockout Mice is UCP1- dependent.

As Ucp1 knockout mice are capable of adaptation to cold through Ucp1- independent pathways, we tested the genetic requirement for Ucp1 expression in the phenotypic rescue of the HDAC3 KO model following exposure to 15 °C. While attempting to generate DKO mice, we were only able to generate Adipoq-cre, Hdac3f/f,

Ucp1-/- mice, presumably due to integration of the Ucp1-cre BAC transgene onto the same chromosome as the Ucp1 null allele (Ch. 8). Therefore, we subjected the Adipoq- cre, Hdac3f/f, Ucp1-/- DKO mice housed at 22 °C to 24-hours at 15 °C, and observed that mice of all generated genotypes survived. Following a 24-hour recovery period at 22 °C, we subjected the HDAC3/UCP1 DKO mice to acute environmental challenge at 4 °C. To 95 our surprise, all mice genetically lacking Ucp1 became severely from hypothermic

(Figure 4.3a), and perished regardless of HDAC3 expression in adipose from hypothermia (Figure 4.3b). Thus, we conclude that Ucp1 is genetically required for the phenotypic rescue of HDAC3 KO mice at 15 °C, as defective Ucp1 is activated in these mice. Further, we conclude that 24 hours of exposure to moderate cold does not meet either the temporal and/or metabolic requirement(s) for the induction of Ucp1- independent pathways. Therefore, we focused on identifying the transcriptional changes that occur in both HDAC3 KO mice and control littermates, following moderate cold.

Environmental Exposure to Moderate Cold Mediates an Epigenetic Rescue of HDAC3 Knockout BAT Thermogenic Genes.

To elucidate the transcriptional changes that facilitate the persistent thermogenic aptitude of HDAC3 KO mice to defend core body temperature against cold, we performed RNA sequencing (RNA-seq) on Ucp1-cre HDAC3 KO and control littermate mice housed at 22 °C and following 24 hour and 7 day recovery periods. We performed an unbiased principal component analysis (PCA) using 24 brown fat transcriptomes

(biological replicates) to determine the relationships between genotype and environmental exposure. Identified relationships were projected and plotted as a function of the two principal components identified in the analysis (PC1 and PC2). Interestingly, the PCA analysis revealed four distinct populations of brown fat, demarcated by genotype and environment (Figure 4.4a). However, while HDAC3 KO and littermate controls remain distinct in all conditions, these populations display a common transition following cold exposure (Figure 4.4a). Stated differently, the control and HDAC3 KO BAT becomes more similar following cold, suggestive of a common cold-induced regulatory mechanism exercising functions independent of genotype. To further analyze this point, 96 we generated RNA-seq scatterplots to visualize differential gene expression between control littermate and HDAC3 KO BAT (genotype) before and after cold (environment).

Consistent with PCA analysis, we observed fewer differential genes between control and

HDAC3 KO BAT following cold exposure. Notably, we uncovered that Ucp1 transcription that is nearly absent in the HDAC3 KO BAT at room temperature and is activated similarly in both WT and HDAC3 KO mice in the 24-hour and 7-day recovery groups

(Figure 4.4b), which was confirmed by reverse-transcription–quantitative PCR (RT– qPCR) (Extended Data Figure 4.3a).

To further discover transcriptional changes, like Ucp1, which are commonly regulated in an environment-dependent fashion, we undertook an unbiased hierarchical gene clustering approach to identify co-regulated genes. Hierarchical clustering analysis revealed two distinct gene clusters (labeled 1 and 2), which exhibited expression patterns mostly induced and persisting moderate cold exposure (Extended Data Figure

4.3b). Interestingly, these gene clusters were highly enriched for genes down-regulated in the HDAC3 KO BAT and were transcriptionally rescued following moderate cold exposure (Extended Data Figure 4.3b). Analysis of the genes falling within these clusters, revealed a dramatic enrichment for mitochondrial processes such as the mitochondrial electron transport chain and respiration (oxidative phosphorylation), as well as mitochondrial organization and fatty acid oxidation (Figure 4.4c-d). Remarkably, many of the genes identified through clustering analysis are oxidative phosphorylation genes previously identified as dependent upon HDAC3 basal co-activator function for transcriptional activation demonstrated prolonged expression (Figure 4.4e) (Emmett et al., 2017).

As functional mitochondria are a requirement for brown adipose mediated thermogenesis, and HDAC3 KO mice have impaired mitochondrial respiration(Emmett et al., 2017), we next evaluated mitochondrial structure of HDAC3 KO mice and control littermates by electron microscopy. Notably, in mice lacking HDAC3, there were significant structural changes in mitochondrial morphology (Figure 4.4f). Electron micrographs of HDAC3 KO BAT demonstrated striking change, as many mitochondria appeared much smaller, and appeared more rounded and varied in size, as compared to control littermates (Figure 4.4f). Ultrastructural analysis also demonstrated irregular and decreased density of cristae in the HDAC3 KO. Interestingly, following moderate cold and 24-hours of recovery at room temperature, the abnormal size and structures of BAT

HDAC3 KO mitochondria were mostly abrogated as the HDAC3 KO resembled control

(Figure 4.4f). Consistent with the observed gene ontology enrichment in mitochondrial processes and mitochondrial organization, we conclude these transcriptional changes have physiological significance and contribute to the rescued cold intolerance phenotype of HDAC3 KO mice. Thus, we conclude that moderate cold must induce a persistent epigenomic change in brown fat, which circumvents the requirement of HDAC3 coactivator activity for the transcription of thermogenic gene programs.

Moderate Cold Exposure Activates BAT Enhancers in an HDAC3- Independent Manner.

To determine the mechanism of how brief exposure to moderate cold both rescues the defective gene expression in HDAC3 KO mice and induces a pro- thermogenic gene expression profile in BAT, we performed global run-on sequencing

(GRO-seq) (Core et al., 2008) to map nascent transcription. From the GRO-seq datasets, we bioinformatically identified non-coding enhancer RNAs (eRNAs) to define 98 and map functional enhancers (Emmett et al., 2017; Fang et al., 2014; Kim et al., 2010;

Li et al., 2013) in BAT from HDAC3 KO animals and control littermates at 22 °C and following 24-hours at 15 °C. We reasoned that since transcriptional changes persisted in control and HDAC3 KO tissue following recovery at 22 °C, mapping eRNAs transcription directly dependent upon cold signals would be most informative for uncovering the underlying mechanism. Thus, we performed GRO-seq on BAT during exposure to 15 °C.

Analysis of the differential BAT eRNAs from control littermates and HDAC3 KO at

22 °C versus 15 °C, demonstrated the remarkable shared induction of eRNAs from both genotypes upon a modest decrease in environmental temperature (Figure 4.5a, red highlighted box). Having previously identified sustained induction of thermogenic genes in our RNA-seq dataset, we hypothesized that the 3,963 identified cold induced eRNAs in control and HDAC3 KO BAT, may mark the functional enhancers responsible for mediating an epigenomic rescue of Ucp1 and brown fat aptitude. To identify the putative transcription factor, which may modulate enhancer function at these eRNA sites, we performed de novo motif analysis of the induced eRNAs (Figure 4.5b). Interestingly, we discovered significant enrichment for the estrogen-related receptor motif at sites of cold- induced transcription in both HDAC3 KO and control littermates (Figure 4.5b).

Previously, we identified HDAC3 as a critical coactivator of ERRα- mediated transcriptional activity (Emmett et al., 2017). Taken together, these data suggest moderate cold exposure may molecularly bypass the requirement of HDAC3 mediated ERRα enhancer coactivation under basal conditions, and strengthens the emerging roles for ERRs as regulators of mitochondrial energy metabolism and brown fat thermogenesis (Giguère and Gigue, 2008; Luo et al., 2003; Villena and Kralli, 2008;

Villena et al., 2007). Additionally, we identified strong motif enrichment of the important

99 brown adipose lineage factors (Soccio et al., 2015) C/EBP (Kajimura et al., 2009;

Lefterova et al., 2010) and EBF2 (Rajakumari et al., 2013; Shapira et al., 2017) at sites of induced eRNA activity, which may enhance ERR function at brown fat enhancers.

To confirm the de novo induced ERRα motif at eRNA sites, we performed ChIP- seq of ERRα in littermate control mice housed at 22 °C to determine true ERRα bound enhancers. In fact, ERRα binding was strongest at sites of eRNA downregulation in

HDAC3 KO BAT at room temperature (Figure 4.5c), similar to results obtained from animals maintained at thermoneutrality (Emmett et al., 2017). We next quantitatively examined eRNA transcriptional changes at ERRα binding sites HDAC3 KO BAT following cold, where eRNA transcription in downregulated at 22 °C. Strikingly, the average eRNA transcriptional profile at ERRα bound enhancers was significantly increased in the HDAC3 KO following cold exposure (Figure 4.5d), supporting a mechanism whereby the genetic requirement for HDAC3 is circumvented by moderate cold. Since, ERRα is a very weak transcriptional activator and relies upon physical recruitment of the strong transcriptional coactivator PGC-1α for nearly all of its function

(Giguère and Gigue, 2008; Schreiber et al., 2003; Villena and Kralli, 2008) we next inspected the PGC-1α locus.

Activation of Ucp1 Transcription is Partially Abrogated by Genetic Depletion of PGC-1α

Due to the strong genomic evidence that ERRα sites may be mediating induced enhancer activity in control and HDAC3 KO BAT, we inspected the Pgc-1α locus in

GRO-seq data from the BAT of mice exposed to moderate cold. To our surprise, cold temperature strongly stimulated transcription from an alternative upstream promoter, 100 known to Pgc-1α isoforms in nearby alternative first exons (exons 1b or 1b’) from nearly

14kb upstream of the proximal promoter (Figure 4.6a).

Isoform expression from this alternative promoter has been demonstrated in muscle upon exercise (Ruas et al., 2012) and in brown adipose at 4 °C (Martínez-

Redondo et al., 2015; Zhang et al., 2009); however, here we demonstrate expression at

15 °C. qPCR results confirm that basal Pgc-1α1 transcription from the proximal promoter is HDAC3-dependent (Figure 4.6b) consistent with previous results (Emmett et al., 2017), in contrast to HDAC3-independent induction in the cold-stimulated state following 24 hours at 15 °C (Figure 4.6b). We next confirmed the increased expression of Pgc-1α isoforms from the alternative promotoer in BAT of control and HDAC3 KO mice exposed to 15 °C versus mice maintained at room temperature (Figure 4.6c).

Notably, whole-body knockout mice with deletion of all Pgc-1α gene product are cold-intolerant and expire to acute cold exposure (Lin et al., 2004), whereas a different genetic whole-body knockout of Pgc-1α, retained a shorter form of Pgc-1α (Chang et al.,

2012) resulted in complete cold tolerant (Leone et al., 2005), suggesting the Pgc-1α isoforms may contribute adaptive thermogenesis.

We next screened by the isoforms by qPCR to see if expression persisted during the 24-hour and 7-day recovery period. We found the Pgc-1α1, transcript from the proximal promoter decreased to near basal levels upon removal from cold (Figure 6D).

However, the Pgc-1α4 isoform from generated from alternative upstream exons e1b remained elevated (Figure 4.6e), albeit at lower levels than during cold, at 24-hours and

7-days recovery times. We next compared PGC-1α4 to PGC-1α1 coactivation ability of

ERRα using an ERR enhancer from the Ucp1 locus (Martínez-Redondo et al., 2015). 101

Notably, PGC-1α4 was able to coactivate ERRα (Emmett et al., 2017), with comparable magnitude to the major Ppargc1a gene product, PGC-1α1 (Figure 4.6f). We next evaluated the levels of Ucp1 using a custom siRNA designed to target the alternative first exon, exon 1b, of the Pgc-1α4 transcript, which led to a significant reduction in Ucp1 transcription. Consistent with a role for PGC-1α enhancing Ucp1 transcription, BAT- specific Ucp1-cre mediated PGC-1α knockouts (Extended Data Figure 4.4a) reduced

Ucp1 transcription.

Therefore, we next tested the hypothesis that BAT-specific Ucp1-cre HDAC3,

PGC-1α DKO mice, would demonstrate an impaired rescue of Ucp1 expression following moderate cold exposure at 15 °C. Consistent with the loss of HDAC3, Ucp1- cre, HDAC3/PGC-1α DKO mice displayed nearly absent Ucp1 expression at room temperature (Figure 4.6f). However, following moderate cold exposure and 24-hours recovery at 22 °C, the HDAC3/PGC-1α DKO mice displayed defective restoration of

Ucp1 mRNA expression.

Discussion

Brown adipose tissue is important in the mammalian response against cold to defend core body temperature and ensure homeostasis and survival. Therefore, mechanisms to unsure euthermia have evolutionary, biological, and physiological importance. Additionally, mechanisms that enhance energy expenditure and regulate brown fat thermogenic capacity are of great interest for understanding the ways brown fat may work to counteract obesity and metabolic disease, and even to explain the

102 heterogenous occurance of active brown fat in adult humans. Previous work (Chapter 3) demonstrated that the class I histone deacetylase HDAC3, functions as a transcriptional coactivator for the nuclear receptor ERRα. HDAC3 activates brown adipose thermogenic enhancers to ensure thermogenic aptitude and cold survival, through priming the thermogenic transcription program.

As models of cold-intolerance and brown adipocyte dysfunction have demonstrated the capability to adapt to cold, we studied the effect of moderate cold on

BAT-specific HDAC3 KO mice. We found that a brief exposure to 15 °C activated defective Ucp1 expression. HDAC3 knockout mice exposed to moderate cold temperatures and subsequently returned to ambient housing conditions, were imparted with a lasting thermogenic aptitude for cold-survival, thus overcoming the effects stemming from genetic loss of HDAC3, to ensure cold survival. Transcriptional profiling of control and HDAC3 KO BAT demonstrated a convergence of gene expression towards a more common population, indicative of a cold-induced epigenetic program influencing brown adipose thermogenesis, as these pathways were significantly induced independent of genotype.

Our study shows that moderate cold increases the transcriptional activities of enhancers bound by ERRα in an HDAC3-independent manner throughout the BAT genome. These transcriptional changes are additionally accompanied by increased transcriptional activation at brown fat enhancers enriched for C/EBP, which may contribute to the verified activity at sites of ERR binding. In addition, as ERRα transcription is dependent upon the transcriptional coactivator PGC-1α, we have demonstrated robust moderate cold induced transcription of PGC-1α including isoforms

103 from an upstream alternative exon. Moderate cold promotes enduring expression of the

PGC-1α4 isoform in BAT, which works to promote the retained and enduring expression of Ucp1 and OXPHOS genes. However, it is worth noting that other factors, such as enhancers bound by C/EBP, may further contribute to the pro-thermogenic phenotype of brown fat.

Our work uncovers a cold-induced epigenetic memory that facilitates an enduring pro-thermogenic phenotype in BAT, which overcomes the genetic deficiency of HDAC3 deficiency in BAT and promotes an enduring transcriptional response in wild-type BAT.

Evolutionarily, the presence of a cold-dependent epigenomic memory, would serve to interconnect an animal’s past environmental exposure to the genome, to better prepare against future bouts of cold. Thus, brown adipose exhibits unique molecular gene- environment interactions of physiological significance to survive dangerous cold temperatures.

104

Figure 4.1: Moderate environmental cold exposure circumvents HDAC3- dependent Ucp1 expression.

a, b, BAT mRNA levels in Ucp1-cre HDAC3 KO versus control littermates (22 °C: n = 5,

6; 15°C, 24 h 22 °C: n = 5, 8): a, Hdac3; b, Ucp1. c, BAT UCP1 protein levels in Ucp1- cre HDAC3 KO versus control littermates (22 °C: n = 3, 3; 15°C, 24 h 22 °C: n = 3, 3). d,

H&E histology of BAT from Ucp1-cre HDAC3 KO versus control littermates housed at

22 °C or exposed to 15°C for 24.

105

Figure 4.2: HDAC3-independent thermogenic pathways facilitate sustained

BAT thermogenesis and cold survival.

106

Figure 4.2: HDAC3-independent thermogenic pathways facilitate sustained

BAT thermogenesis and cold survival.

a, b, Effect of acute cold exposure from standard housing at 22 °C to 4 °C on Adipoq-cre

HDAC3 KO (Hdac3f/f) mice versus control littermates (n = 14, 13), Ucp1-cre HDAC3 KO mice versus control littermates (n = 13, 13): a, core body temperature; b, survival. c, d,

Effect of acute cold exposure from standard housing at 22 °C to 4 °C, following 24 hours at 15 °C and 24 hours of recovery at 22 °C, on Adipoq-cre HDAC3 KO mice versus control littermates (n = 10, 10), Ucp1-cre HDAC3 KO mice versus control littermates

(n = 7, 10): c, core body temperature; d, survival. e, f, Effect of acute cold exposure from standard housing at 22 °C to 4 °C, following 24 hours at 15 °C and 7 days of recovery at

22 °C, on Adipoq-cre HDAC3 KO mice versus control littermates (n = 7, 5), Ucp1-cre

HDAC3 KO mice versus control littermates (n = 6, 6): e, core body temperature; f, survival.

107

Figure 4.3: HDAC3-independent thermogenesis requires Ucp1 for survival. a, b, Effect of acute cold exposure from standard housing at 22 °C to 4 °C, following 24 hours at 15 °C and 24 hours of recovery at 22 °C, on Adipoq-cre HDAC3 KO, Ucp1+/+;

Adipoq-cre HDAC3 KO, Ucp1-/-; and HDAC3f/f, Ucp1-/- mice versus control HDAC3f/f,

Ucp1+/+ littermates (n = 5, 7, 6, 8 respectively): a, core body temperature; b, survival.

108

Figure 4.4: Environmental cold exposure imparts an epigenetic memory promoting thermogenic gene expression.

109

Figure 4: Environmental cold exposure imparts an epigenetic memory to promote thermogenic gene expression. a, Unbiased principal component analysis

(PCA) plot, visualizing the relatedness of whole transcriptomes from the RNA-seq of

Ucp1-cre HDAC3 KO versus control littermates (n = 4, 4) at 22 °C, Ucp1-cre HDAC3 KO versus control littermates (n = 4, 4) exposed to 15 °C for 24 hours (h) and allowed 24 h of recovery at 22 °C, and Ucp1-cre HDAC3 KO versus control littermates (n = 4, 4) exposed to 15 °C and allowed 7 days of recovery at 22 °C b, Scatterplot of RNA-seq data showing HDAC3-regulated BAT genes from Ucp1-cre HDAC3 KO versus control littermates (n = 4, 4) samples in a) (fold change >1.5 up (red) or down (blue) and false discovery rate (FDR) < 0.01). RPKM, reads per kilobase per million. c-d, Gene Ontology

(GO) biological processes of gene clusters 1 and 2 from hierarchical clustering analysis of RNA-seq data (Ext. Data 3b) and selected by adjusted P-value score. e, Heat map depicting oxidative phosphorylation genes discovered in clustering and gene ontology analysis and visualized by average z-score of RNA-seq replicates. f, Electron micrographs of brown adipose mitochondria from Ucp1-cre HDAC3 KO versus control littermates at 22 °C and following exposure to 15 °C for 24 h and allowed 24 h of recovery at 22 °C at 15,000x magnification. Scale bar, 500nm.

110

Figure 4.5: Moderate cold activates estrogen-related receptor α enhancers independent of HDAC3.

111

Figure 4.5: Moderate cold activates estrogen-related receptor α enhancers independent of HDAC3. a, Differential enhancer RNAs (eRNAs) measured by GRO- seq in Adipoq-Cre HDAC3 KO mice versus control littermates (n = 10, 10; pooled biological replicates per library) at 22 °C and following 24 h at 15 °C. Outlined red box highlights cold induced eRNAs common to both control and HDAC3 KO BAT, whereas blue box highlights cold repressed eRNAs. b, De novo motif search at cold induced eRNA sites in control and HDAC3 KO BAT (ranked by P value). c, Average ERRα ChIP- seq profile in control littermates performed at 22 °C (n = 5, pooled biological replicates per library) at enhancers with differential eRNAs in HDAC3 KO at 22 °C. Significant

ERRα found at HDAC3 KO-repressed eRNAs relative to unchanged eRNAs. d, Average eRNA transcriptional profile at ERRα bound enhancers in HDAC3 KO BAT at 22 °C

(dashed line) and in HDAC3 KO BAT following 24 h at 15 °C (solid line), demonstrating significantly increased eRNA transcription upon moderate cold (P = 6.8e−28).

112

Figure 4.6: Cold induces expression of the alternative Pgc-1α4 isoform to promote Ucp1 expression.

113

Figure 4.6: Cold induces expression of the alternative Pgc-1α4 isoform to promote Ucp1 expression. a, Genome browser tracks of the Ppargc1a locus highlighting GRO-seq data (22 °C,

15 °C, [y axis scales in brackets: reads per million]) from control and HDAC3 KO BAT, data displaying nascent gene body transcription from the proximal promoter and alternative promoter (encoding alternative exons 1b and 1b’ for used in isoform splicing). b, c, BAT mRNA levels in Ucp1-cre HDAC3 KO versus control littermates (22 °C: n = 5,

6; 15°C, 24 h 22 °C: n = 5, 8): b, Pgc-1α1 mRNA transcript transcribed from the proximal promoter; c, Pgc-1α2-4 isoform mRNA transcript transcribed from the alternative promoter. d, e, BAT mRNA levels in Ucp1-cre HDAC3 KO versus control littermates

(22 °C: n = 13, 10; 24 h recovery at 22 °C: n = 9, 11; and 7 day recovery at 22 °C: n = 10, 11): d, Pgc-1α1 mRNA transcript transcribed from the proximal promoter; e, Pgc-

1α4 isoform mRNA transcript transcribed from the alternative promoter. f, Luciferase reporter assay of transcription driven by Ucp1 enhancer (−6 kb) bound by ERRα, demonstrating effects of ERRα coactivated by PGC-1α1 or PGC-1α4 (n = 3 replicates per condition). e, RT–qPCR analysis of Ucp1 mRNA expression in mature brown adipocytes after siRNA knockdown of Pgc-1α transcripts containing the alternative exon

1b versus scramble siRNA (n = 3, 3). f, BAT mRNA levels in Ucp1-cre HDAC3/PGC-1α

DKO versus control littermates following 24 h recovery at 22 °C (22 °C: n = 5, 6; 24 h recovery at 22 °C: 3, 5).

114

Extended Data Figure 4.1: Moderate environmental cold rescues Ucp1 expression in Adipoq-cre HDAC3 KO mice. a, b, BAT mRNA levels in Adipoq-cre HDAC3 KO versus control littermates (22 °C: n = 3, 4; 15°C, 24 h 22 °C: n = 3, 4): a, Hdac3; b, Ucp1. c, BAT UCP1 protein levels in

Adipoq-cre HDAC3 KO versus control littermates (22 °C: n = 3, 3; 15°C, 24 h 22 °C: n = 3, 3).

115

Extended Data Figure 4.2: Moderate cold acclimation and recovery protocol for control and HDAC3 KO mice.

a, Diagram of moderate cold exposure paradigm. (1) Mice housed at 22 °C are subjected to acute cold tolerance testing (CTT) directly from 22 °C to 4 °C, or (2) from 22

°C following 24 hours exposure to 15 °C and a 24 hour recovery back at 22 °C, or (3) from 22 °C following 24 hours exposure to 15 °C and a 24 hour recovery back at 22 °C.

Gene expression analysis performed and designated as (4) “24 hour Recovery” or (5) “7 day Recovery”, where exposed to 15 °C and subsequently returned to housing at 22 °C for designated time. All mice used throughout these protocols were male mice 10-12 weeks of age

116

Extended Data Figure 4.3: Unbiased hierarchical gene clustering identifies sustained HDAC3-independent induction of Ucp1.

a, BAT Ucp1 mRNA levels in Ucp1-cre HDAC3 KO versus control littermates (22 °C: n = 13, 10; 24 h recovery: n = 9, 11; 7 d recovery n = 10, 11). b, Unbiased hierarchical gene clustering analysis (n = 4, 4 biological replicates per condition and pooled into one average of replicate for visualization) identification of gene cluster 1 and 2, which contain

Ucp1 and OXPHOS genes, and display prolonged transcription increases in following moderate cold in both control and HDAC3 KO.

117

Extended Data Figure 4.4: BAT-specific ablation of PGC-1α downregulates

Ucp1 expression. a, b, BAT mRNA levels in Ucp1-cre PGC-1α KO versus control littermates (22 °C: n = 7,

4): a, Pgc-1α ; b, Ucp1.

118

Chapter 5

Summary and Future Directions

119

Summary and Discussion:

In Chapter 3, we investigated the role of HDAC3, the sole histone deacetylase of the nuclear receptor corepressor complex, in brown adipose tissue through the development of two novel mouse models. Using tissue specific models of pan-adipose and BAT-specific HDAC3 ablation, we demonstrated a critical role for HDAC3 in promoting the thermogenic program of brown adipose tissue. Using a variety of physiological and phenotypic assays, we demonstrated that HDAC3 is necessary for cold survival and the prevention of environmental cold-induced hypothermia. Further, we demonstrated that HDAC3 is required for the transcription of Ucp1 and mitochondrial oxidative phosphorylation genes.

Despite the failure to defend core body temperature upon transfer from room temperature to extreme cold, HDAC3 KO mice were able to adequately maintain core body temperature while at ambient room temperature. It is important to note, that at room temperature mice are below the threshold of thermoneutrality, where mice must increase their metabolism to defend against cold temperature (Golozoubova et al., 2006;

Nedergaard et al., 2001a, 2001b). Despite the dysfunction of interscapular brown fat, we observed no compensatory browning (Schulz et al., 2013) of the HDAC3 KO mice inguinal depots, which contain inducible beige fat cells. Further, despite impaired brown fat thermogenesis we observed no changes in body weights upon challenge with high-fat diet at both room temperature and thermoneutrality, indicating that alternative Ucp1- independent pathways may be activated in HDAC3 KO mice to defend core body temperature and maintain counteract body weight (Kazak et al., 2015; Long et al., 2016;

Ukropec et al., 2006).

120

We explored and determined the mechanism of HDAC3 function in brown adipose tissue, through combinatorial transcriptome, cistrome, and functional enhancers studies utilizing next generation high-throughput sequencing technologies. We took an innovative approach to bioinformatically integrate multiple layers of deep sequencing datasets to uncover a novel mechanism in brown fat thermogenesis. We utilized global run-on sequencing to quantitate nascent transcription genome-wide at gene bodies, but also to utilized bioinformatics to call non-coding enhancer RNAs (eRNAs) genome-wide at ~25,000 enhancer sites. We discovered that enhancers bound by HDAC3 have repressed eRNAs upon knockout of HDAC3 and these sites also display decreased enrichment of Histone H3K27ac which is a marker of active enhancers. These same enhancers also showed strong enrichment for the brown fat lineage factors C/EBP

(Manchado et al., 1994; Siersbaek et al., 2014) and EBF2 (Rajakumari et al., 2013;

Shapira et al., 2017), which raising interesting questions about their connection to

HDAC3 biology. However, mapping eRNAs gave a distinctive advantage for identifying important changes at enhancers genome-wide. First, the eRNA is derived from active transcriptional events and is not a proxy marker like H3K27ac. Second, while H3K27ac gives large swaths of genomic enrichment, eRNAs are generally within a few hundred bases of a transcription factor binding site. Therefore, using eRNAs, we could precisely map and perform de novo enhancer motif calling to gain a more precise understanding of what transcription factors are being modulated and/or working with HDAC3 at those specific sites.

Through these strategies, we were able to identify a novel mechanism for

HDAC3 in brown adipose, as well as identify an underappreciated role for the Estrogen-

Related Receptor family in BAT. We demonstrate functional interaction at the level of the

121 genome for HDAC3 and ERRα, and demonstrated using a variety of in vivo techniques and transcriptional assays that HDAC3 functions as a coactivator of ERRα.

ERRα activity requires PGC-1α for transcriptional activation, and in cellular transfection assays we demonstrated HDAC3 deacetylates PGC-1α to activate coactivator functions.

Despite demonstrating this result on cellular and biochemically purified pools of PGC-1α, we were not able to demonstrate this in vivo through co-immunoprecipitation approaches due to weak and non-specific antibodies. Our results demonstrate in principal that

HDAC3 may deacetylate other brown fat transcription factors and coregulator molecules.

Finally, we demonstrated through genome-wide nascent transcription assays that

HDAC3 is required for the basal transcription of the thermogenic genes Ucp1 and Pgc-

1α at both room temperature and thermoneutrality. This finding signifies that HDAC3 works independently of adrenergic-signaling and that it is a basal transcriptional coactivator of thermogenic genes. Recently, it has been demonstrated that a genentic deficiency of Ucp1 may influence the expression of mitochondrial oxidative phosphorylation genes (Kazak et al., 2017). However, it is important to highlight that the observed deficiency in mitochondrial oxidation gene transcription occur only following long-term chronic cold adaptation in Ucp1 knockout mice. Here, we report dramatic transcriptional changes in oxidative phosphorylation gene at both thermoneutrality and room temperature, which are dependent on HDAC3.

In Chapter 4, we made the surprising discover that the genetic requirement for

HDAC3 in brown fat for Ucp1 expression can be circumvented by brief acclimation to environmental cold. Following a 24-hour exposure at 15C, HDAC3 KO mice can reactivate Ucp1 expression and survive subsequent cold challenges, extending at least

122

7-days past the initial moderate cold experience. To address the mechanisms, which facilitate cold rescue, we performed RNA-seq and GRO-seq to map the transcriptional profiles of pre- an post- moderate cold-dependent rescue of the thermogenic gene program.

The change in thermogenic gene expression must be regulated at the level of chromatin to activate defective Ucp1 transcription. Therefore, there must be persistent changes to the epigenome that facilitate the enduring pro-thermogenic cold program in

HDAC3 KO BAT. Transcriptionally, we observed sustained gene expression and principal component analysis of the HDAC3 KO and control tissue demonstrated a convergence, or greater similarity, between samples that had experience cold. GRO-seq enhancer eRNA analysis also revealed that ERRα enhancers were upregulated upon exposure to cold temperature. Therefor, upon inspection of the PGC-1α, needed for

ERRα transcritional activity, we observed dramatic induction of a cold-induced alternative transcription driving expression of isoforms with little characterization in most tissues. Through a candidate-based approach, we showed that PGC-1α4 maintained a sustained induction for several days following cold exposure. Knockdown of this isoform in cells diminished Ucp1 expression, and upon double knockout of that PGC-1α and

HDAC3, we observed impaired induction of Ucp1 following moderate cold.

In light of Chapter 4, there are interesting ways forward to further address the hypothesis that BAT contains an epigenetic memory of cold exposure, which works to overcome brown fat deficiencies, which may occur in an organism to promote cold aptitude. Several approaches may be able to further and better characterize the changes occurring to chromatin during these time points. Gaining an understanding of how this

123 putative mechanism functions may help understand the brown fat paucity or abundance observed in humans.

Future Directions:

Identify the effect ageing on HDAC3 KO in beige and brown fat.

Following the publication of our work in Chapter 3, a paper utilizing a different allele of HDAC3 published findings from aged mice that differed from our discoveries in young adult mice (Ferrari et al., 2017). Notably, while we see impaired brown fat function, severe cold intolerance and no appreciable change in the browning of inguinal adipose depots by both qPCR and RNA-seq in 12-week old adult male mice, it was demonstrated in significantly aged mice that loss of HDAC3 led to increased browning of the inguinal depot and warmer body temperatures during prolonged cold exposure, with no appreciable thermogenic gene changes in the interscapular BAT. We bioinformatically analyzed our unpublished inguinal HDAC3 knockout data (using

Adiponectin-cre) from 12-week old mice to available GEO datasets (Ferrari et al., 2017) and we do not observe statistically significant changes in iWAT Ucp1 levels in either dataset. However, we do observe similar enrichment in gene ontology pathways, with enrichment of mitochondria processes, metabolic pathways, and oxidative phosphorylation. Mouse background strain effects, differences in recombination frequency between conditional alleles, diet, and environmental/housing differences may contribute to observed differences. Intriguingly, ageing may impart an epigenomic change within the adipose depots, which is regulated differently by HDAC3 in old versus

124 young mice. Therefore, using our BAT-specific HDAC3 KO mice, we propose to age mice to ~35 weeks and extensively characterize gene expression, UCP1 protein expression, mitochondrial content, adipose morphology, and in subsequent groups conduct physiological cold tolerance phenotyping.

Characterization and phenotyping of HDAC3 KO in beige adipocytes.

Our gene expression analysis of the inguinal adipose depots revealed no significant changes in browning genes. However, a major hurdle in identification of beige fat transcriptional programs is their physical embedding within the white adipose depots.

Development of a Ucp1-Cre, HDAC3f/f, ROSAmT/mG reporter mouse with a cell membrane-targeted, two-color fluorescent Cre-reporter allele, would allow for the cellular purification of Ucp1-Cre-positive and Ucp1-Cre-negative beige cells based upon Cre- mediate recombination. Such an approach would allow extensive characterization of the role of HDAC3 in these cells as compared to classical brown adipocytes. Further, isolated single cells could then be subjected to single-cell RNA-seq for comprehensive transcriptome analysis and comparison to HDAC3 KO BAT data. These data would provide great insight into the differing role of HDAC3 function between BAT and beige cells. Changes in metabolic pathways of the beige cells may exist contributing to the maintenance of core body temperature and metabolic inefficiency in mice lacking

HDAC3 in adipose. Putative changes in these pathways could then be further tested and interrogated in primary inguinal cell culture systems and subjected to cellular metabolic phenotyping assays. A parallel series of experiments could be performed to evaluate the effect of high-fat diet on the beige cells of HDAC3 control versus HDAC3 knockout.

125

Identification of the Brown Fat HDAC3 Interactome.

While we have learned much about the mechanism of HDAC3 coactivation of

ERR transcriptional activity, much remains to be learned about the HDAC3/NCOR complex in brown fat. Using a new CRISPR-generated HDAC3-FLAG eptitope tagged mouse, a high-confidence HDAC3 interactome in brown fat would help further elucidate its surprising activator function in brown fat. In liver, a similar approach identified a novel

HDAC3-PROX1 interaction important for controlling hepatic triglycerides (Armour, S.M., et al.2017). I hypothesize the HDAC3/NCOR complex in brown adipose tissue has unappreciated coregulatory interactors which help facilitate the unique and surprising roe of HDAC3 in brown fat. This approach would additionally be complemented by analysis of the acetylation post-translational modifications on identified proteins, as they may be modulated by HDAC3. Cistromic data for HDAC3 in BAT has revelaed dramatic enrichment for brown fat enhancers enriched with EBF2 and C/EBP-β binding.

Characterization of the HDAC3 proteome may reveal interesting interactions with other molecules that interact with EBF2 and CEBPs. It is possible that at these brown fat enhancers, there is significant co-regulation between multiple factors to control thermogenic function. Establishing functional interactions may increase our ability to develop small molecules to disrupt or promote complex function, rather than inhibition of specific molecules.

Identify and characterize HDAC3-catalytic-dependent and catalytic-independent functions in brown fat. 126

Research presented in chapter 3 assesses the role of HDAC3 loss of function in brown adipose tissue. However, work from liver and heart suggests (Lewandowski et al.,

2015; Sun et al., 2013) HDAC3 mediates significant biological effects on physiology through catalytic-independent functions. While our work establishes a role for HDAC3 deacetylation of PGC-1α, important gene and pathway regulation in brown fat may exist for which we currently lack appreciation. Utilizing a mouse model harboring the HDAC3 point mutation, Y298F (Sun et al., 2013) that leads to a catalytically dead enzyme we would extensively phenotype and transcriptionally profile the role of HDAC3 catalytic function dead in brown fat. These results may complement the HDAC3 interactome studies in above, if HDAC3 were found to have a subset of sites requiring its presence for scaffolding effect rather than its deacetylase activity.

Characterization of NCoR and SMRT in BAT.

As HDAC3 lead us to uncover a novel function in brown adipose, we would expect either NCOR and/or SMRT to exhibit a similar phenotype to HDAC3 KO mice.

However, NCOR and SMRT interact with nuclear receptors, so the loss of one or both of these factors may be especially complex. It remains unclear what the functional redundancies and differences are among NCOR and SMRT, as well as their tissue specific roles. Tissue specific depletion of either or both corepressor, will allow us to make insight into their contribution to the HDAC3 KO phenotype in brown adipose.

Interestingly, the SMRT first receptor interaction domain (mRID1) mouse model (Fang et al., 2011), suggests SMRT functions to promote expression of oxidative phosphorylation

127 genes, similar to the function of HDAC3 in adipose. However, much is to be learned from the conditional loss of NCOR and SMRT function in brown fat. We can further characterize the role of NCOR and SMRT in beige cells.

Moderate cold-induced epigenomic memory in BAT.

The existence of brown fat epigenetic memory is intriguing and may be rigorously characterized through GRO-seq of ambient, cold, and post-cold timepoints of brown fat, with attention paid to differential and persistent enhancer activation. Further, we can perform ChIP-seq for common histone modifications, and characterize the epigenomic marks induced by cold and those that gained or lost as a function of temperature and time. Further, most analysis may be performed to identify associated motifs and transcriptional regulatory molecules. Extensive physiological phenotyping of control mice would be performed to extensively characterize changes in body temperature, lipid profiling, and glucose metabolism.

128

Chapter 6

Bibliography

129

Bibliography

Alenghat, T., Meyers, K., Mullican, S.E., Leitner, K., Adeniji-Adele, A., Avila, J., Bucan, M., Ahima, R.S., Kaestner, K.H., and Lazar, M.A. (2008). Nuclear receptor corepressor and histone deacetylase 3 govern circadian metabolic physiology. Nature 456, 997–1000. Alenghat, T., Osborne, L.C., Saenz, S.A., Kobuley, D., Ziegler, C.G., Mullican, S.E., Choi, I., Grunberg, S., Sinha, R., Wynosky-Dolfi, M., et al. (2013). Histone deacetylase 3 coordinates commensal-bacteria-dependent intestinal homeostasis. Nature 504, 153–157. Allis, C.D., and Jenuwein, T. (2016). The molecular hallmarks of epigenetic control. Nat. Rev. Genet. 17, 487–500. Anstee, Q.M., Targher, G., and Day, C.P. (2013). Progression of NAFLD to diabetes mellitus, cardiovascular disease or cirrhosis. Nat. Rev. Gastroenterol. Hepatol. 10, 330–344. Armour, S.M., Remsberg, J.R., Damle, M., Sidoli, S., Ho, W.Y., Li, Z., Garcia, B.A., L.M.A. (2017). An HDAC3-PROX1 Corepressor Module Acts on HNF4α to Control Hepatic Triglycerides. Nat. Commun. Arrar, M., Turnham, R., Pierce, L., De Oliveira, C.A.F., and Andrew McCammon, J. (2013). Structural insight into the separate roles of inositol tetraphosphate and deacetylase-activating domain in activation of histone deacetylase 3. Protein Sci. 22, 83–92. Bal, N.C., Maurya, S.K., Sopariwala, D.H., Sahoo, S.K., Gupta, S.C., Shaikh, S.A., Pant, M., Rowland, L.A., Bombardier, E., Goonasekera, S.A., et al. (2012). Sarcolipin is a newly identified regulator of muscle-based thermogenesis in mammals. Nat. Med. 18, 1575–1579. Bannister, A.J., and Kouzarides, T. (1996). The CBP co-activator is a histone acetyltransferase. Nature 384, 641–643. Berger, S.L. (2002). Histone modifications in transcriptional regulation. Curr. Opin. Genet. Dev. 12, 142–148. Bhaskara, S., Chyla, B.J., Amann, J.M., Knutson, S.K., Cortez, D., Sun, Z.W., and Hiebert, S.W. (2008). Deletion of Histone Deacetylase 3 Reveals Critical Roles in S Phase Progression and DNA Damage Control. Mol. Cell 30, 61–72. Bhaskara, S., Knutson, S.K., Jiang, G., Chandrasekharan, M.B., Wilson, A.J., Zheng, S., Yenamandra, A., Locke, K., Yuan, J.L., Bonine-Summers, A.R., et al. (2010). Hdac3 is essential for the maintenance of chromatin structure and 130 genome stability. Cancer Cell 18, 436–447. Boudiaf-Benmammar, C., Cresteil, T., and Melki, R. (2013). The Cytosolic Chaperonin CCT/TRiC and Cancer Cell Proliferation. PLoS One 8. Bradley, E.W., Carpio, L.R., van Wijnen, A.J., McGee-Lawrence, M.E., and Westendorf, J.J. (2015). Histone Deacetylases in Bone Development and Skeletal Disorders. Physiol. Rev. 95, 1359–1381. Brownell, J.E., and Allis, C.D. (1995). An activity gel assay detects a single, catalytically active histone acetyltransferase subunit in Tetrahymena macronuclei. Proc. Natl. Acad. Sci. U. S. A. 92, 6364–6368. Brownell, J.E., Zhou, J., Ranalli, T., Kobayashi, R., Edmondson, D.G., Roth, S.Y., and Allis, C.D. (1996). Tetrahymena histone acetyltransferase A: A homolog to yeast Gcn5p linking histone acetylation to gene activation. Cell 84, 843–851. Bugge, A., Feng, D., Everett, L.J., Briggs, E.R., Mullican, S.E., Wang, F., Jager, J., and Lazar, M.A. (2012). Rev-erbα and Rev-erbβ coordinately protect the circadian clock and normal metabolic function. Genes Dev. 26, 657–667. Byrne, C.D., and Targher, G. (2015). NAFLD: A multisystem disease. J. Hepatol. 62, S47–S64. Cannon, B., and Nedergaard, J. (2004). Brown adipose tissue: function and physiological significance. Physiol. Rev. 84, 277–359. Cannon, B., and Nedergaard, J. (2011). Nonshivering thermogenesis and its adequate measurement in metabolic studies. J. Exp. Biol. 214, 242–253. Carpio, L.R., Bradley, E.W., McGee-Lawrence, M.E., Weivoda, M.M., Poston, D.D., Dudakovic, A., Xu, M., Tchkonia, T., Kirkland, J.L., van Wijnen, A.J., et al. (2016a). Histone deacetylase 3 supports endochondral bone formation by controlling cytokine signaling and matrix remodeling. Sci. Signal. 9, ra79-ra79. Carpio, L.R., Bradley, E.W., and Westendorf, J.J. (2016b). Histone Deacetylase 3 Suppresses Erk Phosphorylation and Matrix Metalloproteinase (Mmp)-13 Activity in Chondrocytes. Connect. Tissue Res. 8207, 03008207.2016.1236088. Chalkiadaki, A., and Guarente, L. (2012). Sirtuins mediate mammalian metabolic responses to nutrient availability. Nat. Rev. Endocrinol. 8, 287–296. Chang, H.C., and Guarente, L. (2014). SIRT1 and other sirtuins in metabolism. Trends Endocrinol. Metab. 25, 138–145. Chang, J.S., Fernand, V., Zhang, Y., Shin, J., Jun, H.J., Joshi, Y., and Gettys, T.W. (2012). NT-PGC-1α protein is sufficient to link β 3-adrenergic receptor 131 activation to transcriptional and physiological components of adaptive thermogenesis. J. Biol. Chem. 287, 9100–9111. Chen, E.Y., Tan, C.M., Kou, Y., Duan, Q., Wang, Z., Meirelles, G.V., Clark, N.R., and Ma’ayan, A. (2013). Enrichr: interactive and collaborative HTML5 gene list enrichment analysis tool. BMC Bioinformatics 14, 128. Chen, H., Lin, R.J., Schiltz, R.L., Chakravarti, D., Nash, A., Nagy, L., Privalsky, M.L., Nakatani, Y., and Evans, R.M. (1997). Nuclear receptor coactivator ACTR is a novel histone acetyltransferase and forms a multimeric activation complex with P/CAF and CBP/p300. Cell 90, 569–580. Chen, W., Gao, L., Wang, J., Wang, Y., and Dong, Z. (2016). Conditional ablation of HDAC3 in islet beta cells results in glucose intolerance and enhanced susceptibility to STZ-induced diabetes. Oncotarget 7, 57485–57497. Chen, X., Barozzi, I., Termanini, A., Prosperini, E., Recchiuti, A., Dalli, J., Mietton, F., Matteoli, G., Hiebert, S., and Natoli, G. (2012). Requirement for the histone deacetylase Hdac3 for the inflammatory gene expression program in macrophages. Proc. Natl. Acad. Sci. U. S. A. 109, E2865-74. Chen Lf, Fischle, W., Verdin, E., and Greene, W.C. (2001). Duration of nuclear NF-kappaB action regulated by reversible acetylation. Science 293, 1653–1657. Chouchani, E.T., Kazak, L., Jedrychowski, M.P., Lu, G.Z., Erickson, B.K., Szpyt, J., Pierce, K.A., Laznik-Bogoslavski, D., Vetrivelan, R., Clish, C.B., et al. (2016). Mitochondrial ROS regulate thermogenic energy expenditure and sulfenylation of UCP1. Nature 532, 112–116. Codina, A., Love, J.D., Li, Y., Lazar, M.A., Neuhaus, D., and Schwabe, J.W.R. (2005). Structural insights into the interaction and activation of histone deacetylase 3 by nuclear receptor corepressors. Proc. Natl. Acad. Sci. U. S. A. 102, 6009–6014. Core, L.J., Waterfall, J.J., and Lis, J.T. (2008). Nascent RNA sequencing reveals widespread pausing and divergent initiation at human promoters. Science 322, 1845–1848. Cypess, A.M., Lehman, S., Williams, G., Tal, I., Rodman, D., Goldfine, A.B., Kuo, F.C., Palmer, E.L., Tseng, Y.-H., Doria, A., et al. (2009). Identification and importance of brown adipose tissue in adult humans. N. Engl. J. Med. 360, 1509–1517. Dhalluin, C., Carlson, J.E., Zeng, L., He, C., Aggarwal, a K., and Zhou, M.M. (1999). Structure and ligand of a histone acetyltransferase bromodomain. Nature 399, 491–496. Dixen, K., Basse, A.L., Murholm, M., Isidor, M.S., Hansen, L.H.L., Petersen, 132

M.C.H., Madsen, L., Petrovic, N., Nedergaard, J., Quistorff, B., et al. (2013). ERRγ enhances UCP1 expression and fatty acid oxidation in brown adipocytes. Obesity 21, 516–524. Dunham, I., Kundaje, A., Aldred, S.F., Collins, P.J., Davis, C. a., Doyle, F., Epstein, C.B., Frietze, S., Harrow, J., Kaul, R., et al. (2012). An integrated encyclopedia of DNA elements in the human genome. Nature 489, 57–74. Dunn, A.Y., Melville, M.W., and Frydman, J. (2001). Review: Cellular Substrates of the Eukaryotic Chaperonin TRiC/CCT. J. Struct. Biol. 135, 176–184. Ebert, D.H., Gabel, H.W., Robinson, N.D., Kastan, N.R., Hu, L.S., Cohen, S., Navarro, A.J., Lyst, M.J., Ekiert, R., Bird, A.P., et al. (2013). Activity-dependent phosphorylation of MeCP2 threonine 308 regulates interaction with NCoR. Nature 499, 341–345. Emiliani, S., Fischle, W., Van Lint, C., Al-Abed, Y., and Verdin, E. (1998). Characterization of a human RPD3 ortholog, HDAC3. Proc. Natl. Acad. Sci. U. S. A. 95, 2795–2800. Emmett, M.J., Lim, H.-W., Jager, J., Richter, H.J., Adlanmerini, M., Peed, L.C., Briggs, E.R., Steger, D.J., Ma, T., Sims, C.A., et al. (2017a). Histone deacetylase 3 prepares brown adipose tissue for acute thermogenic challenge. Nature advance on. Enerbäck, S., Jacobsson, a, Simpson, E.M., Guerra, C., Yamashita, H., Harper, M.E., and Kozak, L.P. (1997). Mice lacking mitochondrial uncoupling protein are cold-sensitive but not obese. Nature 387, 90–94. Eyre, T.A., Ducluzeau, F., Sneddon, T.P., Povey, S., Bruford, E.A., and Lush, M.J. (2006). The HUGO Gene Nomenclature Database, 2006 updates. Nucleic Acids Res. 34, D319–D321. Fang, B., Everett, L.J., Jager, J., Briggs, E., Armour, S.M., Feng, D., Roy, A., Gerhart-Hines, Z., Sun, Z., and Lazar, M.A. (2014). Circadian enhancers coordinate multiple phases of rhythmic gene transcription in vivo. Cell 159, 1140– 1152. Fang, S., Suh, J.M., Atkins, A.R., Hong, S.-H., Leblanc, M., Nofsinger, R.R., Yu, R.T., Downes, M., and Evans, R.M. (2011). Corepressor SMRT promotes oxidative phosphorylation in adipose tissue and protects against diet-induced obesity and insulin resistance. Proc. Natl. Acad. Sci. U. S. A. 108, 3412–3417. Fedorenko, A., Lishko, P. V., and Kirichok, Y. (2012). Mechanism of fatty-acid- dependent UCP1 uncoupling in brown fat mitochondria. Cell 151, 400–413. Feigenson M, Shull LC, Taylor EL, Camilleri ET, Reister SM, van Wijnen AJ, Bradley EW, W.J. (2017). Histone deacetylase 3 deletion in mesenchymal 133 progenitor cells hinders long bone development. J. Bone Miner. Res. Feldman, D.E., Thulasiraman, V., Ferreyra, R.G., and Frydman, J. (1999). Formation of the VHL–Elongin BC Tumor Suppressor Complex Is Mediated by the Chaperonin TRiC. Mol. Cell 4, 1051–1061. Feldmann, H.M., Golozoubova, V., Cannon, B., and Nedergaard, J. (2009). UCP1 Ablation Induces Obesity and Abolishes Diet-Induced Thermogenesis in Mice Exempt from Thermal Stress by Living at Thermoneutrality. Cell Metab. 9, 203–209. Feng, D., Liu, T., Sun, Z., Bugge, A., Mullican, S.E., Alenghat, T., Liu, X.S., and Lazar, M.A. (2011). A Circadian Rhythm Orchestrated by Histone Deacetylase 3 Controls Hepatic Lipid Metabolism. Science (80-. ). 331, 1315–1319. Ferrari, A., Longo, R., Fiorino, E., Silva, R., Mitro, N., Cermenati, G., Gilardi, F., Desvergne, B., Andolfo, A., Magagnotti, C., et al. (2017). HDAC3 is a molecular brake of the metabolic switch supporting white adipose tissue browning. Nat. Commun. 8, 93. Fischle, W., Dequiedt, F., Hendzel, M.J., Guenther, M.G., Lazar, M.A., Voelter, W., and Verdin, E. (2002). Enzymatic activity associated with class II HDACs is dependent on a multiprotein complex containing HDAC3 and SMRT/N-CoR. Mol. Cell 9, 45–57. Fujisawa, T., and Filippakopoulos, P. (2017). Functions of bromodomain- containing proteins and their roles in homeostasis and cancer. Nat. Rev. Mol. Cell Biol. 18, 246–262. Gerhart-Hines, Z., Feng, D., Emmett, M.J., Everett, L.J., Loro, E., Briggs, E.R., Bugge, A., Hou, C., Ferrara, C., Seale, P., et al. (2013). The nuclear receptor Rev-erbα controls circadian thermogenic plasticity. Nature 503, 410–413. Giguère, V. (2008). Transcriptional control of energy homeostasis by the estrogen-related receptors. Endocr. Rev. 29, 677–696. Giguère, V., and Gigue, V. (2008). Transcriptional control of energy homeostasis by the estrogen-related receptors. Endocr. Rev. 29, 677–696. Goldfarb, Y., Kadouri, N., Levi, B., Sela, A., Herzig, Y., Cohen, R.N., Hollenberg, A.N., and Abramson, J. (2016). HDAC3 Is a Master Regulator of mTEC Development. Cell Rep. 15, 651–665. Golozoubova, V., Hohtola, E., Matthias, a, Jacobsson, a, Cannon, B., and Nedergaard, J. (2001). Only UCP1 can mediate adaptive nonshivering thermogenesis in the cold. FASEB J. 15, 2048–2050. Golozoubova, V., Cannon, B., and Nedergaard, J. (2006). UCP1 is essential for 134 adaptive adrenergic nonshivering thermogenesis. Am. J. Physiol. Endocrinol. Metab. 291, E350–E357. Granneman, J.G., Burnazi, M., Zhu, Z., and Schwamb, L. a (2003). White adipose tissue contributes to UCP1-independent thermogenesis. Am. J. Physiol. Endocrinol. Metab. 285, E1230-6. Gregoretti, I. V., Lee, Y.M., and Goodson, H. V. (2004). Molecular evolution of the histone deacetylase family: Functional implications of phylogenetic analysis. J. Mol. Biol. 338, 17–31. Guenther, M.G., Lane, W.S., Fischle, W., Verdin, E., Lazar, M.A., and Shiekhattar, R. (2000). A core SMRT corepressor complex containing HDAC3 and TBL1, a WD40-repeat protein linked to deafness. Genes Dev. 14, 1048– 1057. Guenther, M.G., Barak, O., and Lazar, M.A. (2001). The SMRT and N-CoR Corepressors Are Activating Cofactors for Histone Deacetylase 3. Mol. Cell. Biol. 21, 6091–6101. Guenther, M.G., Yu, J., Kao, G.D., Yen, T.J., and Lazar, M.A. (2002). Assembly of the SMRT-histone deacetylase 3 repression complex requires the TCP-1 ring complex. Genes Dev. 16, 3130–3135. Haberland, M., Montgomery, R.L., and Olson, E.N. (2009). The many roles of histone deacetylases in development and physiology: implications for disease and therapy. Nat. Rev. Genet. 10, 32–42. Harms, M., and Seale, P. (2013). Brown and beige fat: development, function and therapeutic potential. Nat. Med. 19, 1252–1263. Harms, M.J., Ishibashi, J., Wang, W., Lim, H.W., Goyama, S., Sato, T., Kurokawa, M., Won, K.J., and Seale, P. (2014). Prdm16 is required for the maintenance of brown adipocyte identity and function in adult mice. Cell Metab. 19, 593–604. Harms, M.J., Lim, H.-W., Ho, Y., Shapira, S.N., Ishibashi, J., Rajakumari, S., Steger, D.J., Lazar, M.A., Won, K.-J., and Seale, P. (2015). PRDM16 binds MED1 and controls chromatin architecture to determine a brown fat transcriptional program. Genes Dev. 29, 298–307. Heinz, S., Benner, C., Spann, N., Bertolino, E., Lin, Y.C., Laslo, P., Cheng, J.X., Murre, C., Singh, H., and Glass, C.K. (2010). Simple combinations of lineage- determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Mol. Cell 38, 576–589. Heinzel, T., Lavinsky, R.M., Mullen, T.M., Söderstrom, M., Laherty, C.D., Torchia, J., Yang, W.M., Brard, G., Ngo, S.D., Davie, J.R., et al. (1997). A complex 135 containing N-CoR, mSin3 and histone deacetylase mediates transcriptional repression. Nature 387, 43–48. Hoeksema, M. a, Gijbels, M.J., Van den Bossche, J., van der Velden, S., Sijm, A., Neele, A.E., Seijkens, T., Stöger, J.L., Meiler, S., Boshuizen, M.C., et al. (2014). Targeting macrophage Histone deacetylase 3 stabilizes atherosclerotic lesions. EMBO Mol. Med. 6, e201404170. Hong, S., Zhou, W., Fang, B., Lu, W., Loro, E., Damle, M., Ding, G., Jager, J., Zhang, S., Zhang, Y., et al. (2016). Dissociation of muscle insulin sensitivity from exercise endurance in mice by HDAC3 depletion. Nat. Med. 23, 223–234. Houstek, J., Cannon, B., and Lindberg, O. (1975). Glycerol-3-Phosphate Shuttle and Its function in intermediary metabolism of hamster brown adipose tissue. Eur. J. Biochem. 54, 11–18. Hsu, F.-C., Belmonte, P.J., Constans, M.M., Chen, M.W., McWilliams, D.C., Hiebert, S.W., and Shapiro, V.S. (2015). Histone Deacetylase 3 Is Required for T Cell Maturation. J. Immunol. 195, 1578–1590. Hu, E., Chen, Z., Fredrickson, T., Zhu, Y., Kirkpatrick, R., Zhang, G.F., Johanson, K., Sung, C.M., Liu, R., and Winkler, J. (2000). Cloning and characterization of a novel human class I histone deacetylase that functions as a transcription repressor. J. Biol. Chem. 275, 15254–15264. Huang, E.Y., Zhang, J., Miska, E.A., Guenther, M.G., Kouzarides, T., and Lazar, M.A. (2000). Nuclear receptor corepressors partner with class II histone deacetylases in a Sin3-independent repression pathway. Genes Dev. 14, 45–54. Ishizuka, T., and Lazar, M.A. (2005). The Nuclear Receptor Corepressor Deacetylase Activating Domain Is Essential for Repression by Thyroid Hormone Receptor. Mol. Endocrinol. 19, 1443–1451. Kadosh, D., and Struhl, K. (1998). Histone deacetylase activity of Rpd3 is important for transcriptional repression in vivo. Genes Dev. 12, 797–805. Kajimura, S., Seale, P., Kubota, K., Lunsford, E., Frangioni, J. V, Gygi, S.P., and Spiegelman, B.M. (2009). Initiation of myoblast to brown fat switch by a PRDM16-C/EBP-beta transcriptional complex. Nature 460, 1154–1158. Kao, H.Y., Downes, M., Ordentlich, P., and Evans, R.M. (2000). Isolation of a novel histone deacetylase reveals that class I and class II deacetylases promote SMRT-mediated repression. Genes Dev. 14, 55–66. Kasper, L.H., Lerach, S., Wang, J., Wu, S., Jeevan, T., and Brindle, P.K. (2010). CBP/p300 double null cells reveal effect of coactivator level and diversity on CREB transactivation. EMBO J. 29, 3660–3672.

136

Kazak, L., Chouchani, E.T., Jedrychowski, M.P., Erickson, B.K., Shinoda, K., Cohen, P., Vetrivelan, R., Lu, G.Z., Laznik-Bogoslavski, D., Hasenfuss, S.C., et al. (2015). A Creatine-Driven Substrate Cycle Enhances Energy Expenditure and Thermogenesis in Beige Fat. Cell 163, 643–655. Kazak, L., Chouchani, E.T., Stavrovskaya, I.G., Lu, G.Z., Jedrychowski, M.P., Egan, D.F., Kumari, M., Kong, X., Erickson, B.K., Szpyt, J., et al. (2017). UCP1 deficiency causes brown fat respiratory chain depletion and sensitizes mitochondria to calcium overload-induced dysfunction. Proc. Natl. Acad. Sci. 114, 7981–7986. Kent, W.J., Zweig, a S., Barber, G., Hinrichs, a S., and Karolchik, D. (2010). BigWig and BigBed: enabling browsing of large distributed datasets. Bioinformatics 26, 2204–2207. Kim, D., Pertea, G., Trapnell, C., Pimentel, H., Kelley, R., and Salzberg, S.L. (2013). TopHat2: accurate alignment of transcriptomes in the presence of insertions, deletions and gene fusions. Genome Biol. 14, R36. Kim, T.-K., Hemberg, M., Gray, J.M., Costa, A.M., Bear, D.M., Wu, J., Harmin, D.A., Laptewicz, M., Barbara-Haley, K., Kuersten, S., et al. (2010). Widespread transcription at neuronal activity-regulated enhancers. Nature 465, 182–187. Knutson, S.K., Chyla, B.J., Amann, J.M., Bhaskara, S., Huppert, S.S., and Hiebert, S.W. (2008). Liver-specific deletion of histone deacetylase 3 disrupts metabolic transcriptional networks. EMBO J. 27, 1017–1028. Kong, X., Banks, A., Liu, T., Kazak, L., Rao, R.R., Cohen, P., Wang, X., Yu, S., Lo, J.C., Tseng, Y.H., et al. (2014). IRF4 is a key thermogenic transcriptional partner of PGC-1α. Cell 158, 69–83. Kuleshov, M. V., Jones, M.R., Rouillard, A.D., Fernandez, N.F., Duan, Q., Wang, Z., Koplev, S., Jenkins, S.L., Jagodnik, K.M., Lachmann, A., et al. (2016). Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res. gkw377. Lahm, A., Paolini, C., Pallaoro, M., Nardi, M.C., Jones, P., Neddermann, P., Sambucini, S., Bottomley, M.J., Lo Surdo, P., Carfi, A., et al. (2007). Unraveling the hidden catalytic activity of vertebrate class IIa histone deacetylases. Proc. Natl. Acad. Sci. 104, 17335–17340. Lam, M.T.Y., Li, W., Rosenfeld, M.G., and Glass, C.K. (2014). Enhancer RNAs and regulated transcriptional programs. Trends Biochem. Sci. 39, 170–182. Langmead, B., Trapnell, C., Pop, M., and Salzberg, S.L. (2009). Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol. 10, R25.

137

Lefterova, M.I., Steger, D.J., Zhuo, D., Qatanani, M., Mullican, S.E., Tuteja, G., Manduchi, E., Grant, G.R., Mitchell, A., and Lazar, M.A. (2010). Cell-Specific Determinants of Peroxisome Proliferator-Activated Receptor γ Function in Adipocytes and Macrophages Cell-Specific Determinants of Peroxisome Proliferator-Activated Receptor Function in Adipocytes and Macrophages §. Mol. Cell. Biol. 30, 2078–2089. Leone, T.C., Lehman, J.J., Finck, B.N., Schaeffer, P.J., Wende, A.R., Boudina, S., Courtois, M., Wozniak, D.F., Sambandam, N., Bernal-Mizrachi, C., et al. (2005). PGC-1?? deficiency causes multi-system energy metabolic derangements: Muscle dysfunction, abnormal weight control and hepatic steatosis. PLoS Biol. 3, 0672–0687. LeRoy, G., Rickards, B., and Flint, S.J. (2008). The Double Bromodomain Proteins Brd2 and Brd3 Couple Histone Acetylation to Transcription. Mol. Cell 30, 51–60. Lewandowski, S.L., Janardhan, H.P., Smee, K.M., Bachman, M., Sun, Z., Lazar, M.A., and Trivedi, C.M. (2014). Histone deacetylase 3 modulates Tbx5 activity to regulate early cardiogenesis. Hum. Mol. Genet. 23, 3801–3809. Lewandowski, S.L., Janardhan, H.P., and Trivedi, C.M. (2015). Histone Deacetylase 3 Coordinates Deacetylase-independent Epigenetic Silencing of Transforming Growth Factor-β1 (TGF-β1) to Orchestrate Second Heart Field Development. J. Biol. Chem. 290, 27067–27089. Li, J., Wang, J., Wang, J., Nawaz, Z., Liu, J.M., Qin, J., and Wong, J. (2000). Both corepressor proteins SMRT and N-CoR exist in large protein complexes containing HDAC3. EMBO J. 19, 4342–4350. Li, W., Notani, D., Ma, Q., Tanasa, B., Nunez, E., Chen, A.Y., Merkurjev, D., Zhang, J., Ohgi, K., Song, X., et al. (2013). Functional roles of enhancer RNAs for oestrogen-dependent transcriptional activation. Nature 498, 516–520. Liao, Y., Smyth, G.K., and Shi, W. (2014). FeatureCounts: An efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics 30, 923–930. Lin, J., Wu, P.H., Tarr, P.T., Lindenberg, K.S., St-Pierre, J., Zhang, C.Y., Mootha, V.K., Jäger, S., Vianna, C.R., Reznick, R.M., et al. (2004). Defects in adaptive energy metabolism with CNS-linked hyperactivity in PGC-1α null mice. Cell 119, 121–135. Loft, A., Forss, I., and Mandrup, S. (2016). Genome-Wide Insights into the Development and Function of Thermogenic Adipocytes. Trends Endocrinol. Metab.

138

Long, J.Z., Svensson, K.J., Bateman, L.A., Lin, H., Kamenecka, T., Lokurkar, I.A., Lou, J., Rao, R.R., Chang, M.R.R., Jedrychowski, M.P., et al. (2016). The Secreted Enzyme PM20D1 Regulates Lipidated Amino Acid Uncouplers of Mitochondria. Cell 166, 424–435. Lopez-Atalaya, J.P., Ito, S., Valor, L.M., Benito, E., and Barco, A. (2013). Genomic targets, and histone acetylation and gene expression profiling of neural HDAC inhibition. Nucleic Acids Res. 41, 8072–8084. Lundh, M., Galbo, T., Poulsen, S.S., and Mandrup-Poulsen, T. (2015). Histone deacetylase 3 inhibition improves glycaemia and insulin secretion in obese diabetic rats. Diabetes, Obes. Metab. 17, 703–707. Luo, J., Sladek, R., Carrier, J., Bader, J.-A., Richard, D., and Giguère, V. (2003). Reduced fat mass in mice lacking orphan nuclear receptor estrogen-related receptor α. Mol. Cell. Biol. 23, 7947–7956. Lyst, M.J., Ekiert, R., Ebert, D.H., Merusi, C., Nowak, J., Selfridge, J., Guy, J., Kastan, N.R., Robinson, N.D., de Lima Alves, F., et al. (2013). Rett syndrome mutations abolish the interaction of MeCP2 with the NCoR/SMRT co-repressor. Nat. Neurosci. 16, 898–902. Manchado, C., Yubero, P., Viñas, O., Iglesias, R., Villarroya, F., Mampel, T., and Giralt, M. (1994). CCAAT/enhancer-binding proteins alpha and beta in brown adipose tissue: evidence for a tissue-specific pattern of expression during development. Biochem. J. 302, 695–700. Van Marken Lichtenbelt, W.D., Vanhommerig, J.W., Smulders, N.M., Kemerink, G.J., Bouvy, N.D., Schrauwen, P., and Teule, G.J.J. (2009). Cold-activated brown adipose tissue in healthy men. N. Engl. J. Med. 360, 1500–1508. Marmorstein, R. (2001). Structure of histone deacetylases: Insights into substrate recognition and catalysis. Structure 9, 1127–1133. Martin, M. (2011). Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet.journal 17, 10. Martínez-Redondo, V., Pettersson, A.T., and Ruas, J.L. (2015). The hitchhiker’s guide to PGC-1α isoform structure and biological functions. Diabetologia 58, 1969–1977. McGee-Lawrence, M.E., Bradley, E.W., Dudakovic, A., Carlson, S.W., Ryan, Z.C., Kumar, R., Dadsetan, M., Yaszemski, M.J., Chen, Q., An, K.N., et al. (2013). Histone deacetylase 3 is required for maintenance of bone mass during aging. Bone 52, 296–307. McQuown, S.C., Barrett, R.M., Matheos, D.P., Post, R.J., Rogge, G.A., Alenghat, T., Mullican, S.E., Jones, S., Rusche, J.R., Lazar, M.A., et al. (2011). HDAC3 Is a 139

Critical Negative Regulator of Long-Term Memory Formation. J. Neurosci. 31, 764–774. Mihaylova, M.M., Vasquez, D.S., Ravnskjaer, K., Denechaud, P.D., Yu, R.T., Alvarez, J.G., Downes, M., Evans, R.M., Montminy, M., and Shaw, R.J. (2011). Class IIa histone deacetylases are hormone-activated regulators of FOXO and mammalian glucose homeostasis. Cell 145, 607–621. Mizzen, C.A., Yang, X.J., Kokubo, T., Brownell, J.E., Bannister, A.J., Owen- Hughes, T., Workman, J., Wang, L., Berger, S.L., Kouzarides, T., et al. (1996). The TAF(II)250 subunit of TFIID has histone acetyltransferase activity. Cell 87, 1261–1270. Montgomery, R.L., Potthoff, M.J., Haberland, M., Qi, X., Matsuzaki, S., Humphries, K.M., Richardson, J.A., Bassel-Duby, R., and Olson, E.N. (2008). Maintenance of cardiac energy metabolism by histone deacetylase 3 in mice. J. Clin. Invest. Morris, M.J., and Monteggia, L.M. (2013). Unique functional roles for class I and class II histone deacetylases in central nervous system development and function. Int. J. Dev. Neurosci. 31, 370–381. Mullican, S.E., Gaddis, C.A., Alenghat, T., Nair, M.G., Giacomin, P.R., Everett, L.J., Feng, D., Steger, D.J., Schug, J., Artis, D., et al. (2011). Histone deacetylase 3 is an epigenomic brake in macrophage alternative activation. Genes Dev. 25, 2480–2488. Mullican, S.E., Tomaru, T., Gaddis, C.A., Peed, L.C., Sundaram, A., and Lazar, M.A. (2013). A novel adipose-specific gene deletion model demonstrates potential pitfalls of existing methods. Mol. Endocrinol. 27, 127–134. Nagy, L., Kao, H.Y., Chakravarti, D., Lin, R.J., Hassig, C.A., Ayer, D.E., Schreiber, S.L., and Evans, R.M. (1997). Nuclear receptor repression mediated by a complex containing SMRT, mSin3A, and histone deacetylase. Cell 89, 373– 380. Nedergaard, J., Golozoubova, V., Matthias, A., Asadi, A., Jacobsson, A., and Cannon, B. (2001a). UCP1: The only protein able to mediate adaptive nonshivering thermogenesis and metabolic inefficiency. Biochim. Biophys. Acta - Bioenerg. 1504, 82–106. Nedergaard, J., Golozoubova, V., Matthias, a, Shabalina, I., Ohba, K., Ohlson, K., Jacobsson, a, and Cannon, B. (2001b). Life without UCP1: mitochondrial, cellular and organismal characteristics of the UCP1-ablated mice. Biochem. Soc. Trans. 29, 756–763. Nedergaard, J., Bengtsson, T., and Cannon, B. (2007). Unexpected evidence for

140 active brown adipose tissue in adult humans. Am. J. Physiol. Endocrinol. Metab. 293, E444-52. Norwood, J., Franklin, J.M., Sharma, D., and D’Mello, S.R. (2014). Histone deacetylase 3 is necessary for proper brain development. J. Biol. Chem. 289, 34569–34582. Nott, A., Cheng, J., Gao, F., Lin, Y.-T., Gjoneska, E., Ko, T., Minhas, P., Zamudio, A.V., Meng, J., Zhang, F., et al. (2016). Histone deacetylase 3 associates with MeCP2 to regulate FOXO and social behavior. Nat. Neurosci. 19, 1497–1505. Oberoi, J., Fairall, L., Watson, P.J., Yang, J.-C., Czimmerer, Z., Kampmann, T., Goult, B.T., Greenwood, J.A., Gooch, J.T., Kallenberger, B.C., et al. (2011). Structural basis for the assembly of the SMRT/NCoR core transcriptional repression machinery. Nat. Struct. Mol. Biol. 18, 177–184. Ogryzko, V. V., Schiltz, R.L., Russanova, V., Howard, B.H., and Nakatani, Y. (1996). The transcriptional coactivators p300 and CBP are histone acetyltransferases. Cell 87, 953–959. Oropeza, D., Jouvet, N., Budry, L., Campbell, J.E., Bouyakdan, K., Lacombe, J., Perron, G., Bergeron, V., Neuman, J.C., Brar, H.K., et al. (2015). Phenotypic characterization of MIP-CreERT1Lphi mice with transgene-driven islet expression of human growth hormone. Diabetes 64, 3798–3807. Ouellet, V., Routhier-Labadie, A., Bellemare, W., Lakhal-Chaieb, L., Turcotte, E., Carpentier, A.C., and Richard, D. (2011). Outdoor temperature, age, sex, body mass index, and diabetic status determine the prevalence, mass, and glucose- uptake activity of 18F-FDG-detected BAT in humans. J. Clin. Endocrinol. Metab. 96, 192–199. Pecinová, A., Drahota, Z., Nůsková, H., Pecina, P., and Houštěk, J. (2011). Evaluation of basic mitochondrial functions using rat tissue homogenates. Mitochondrion 11, 722–728. Persichetti, A., Sciuto, R., Rea, S., Basciani, S., Lubrano, C., Mariani, S., Ulisse, S., Nofroni, I., Maini, C.L., and Gnessi, L. (2013). Prevalence, Mass, and Glucose-Uptake Activity of 18F-FDG-Detected Brown Adipose Tissue in Humans Living in a Temperate Zone of Italy. PLoS One 8. Philips, R.L., Chen, M.W., McWilliams, D.C., Belmonte, P.J., Constans, M.M., and Shapiro, V.S. (2016). HDAC3 Is Required for the Downregulation of RORγt during Thymocyte Positive Selection. J. Immunol. 197, 541–554. Picard, M., Taivassalo, T., Ritchie, D., Wright, K.J., Thomas, M.M., Romestaing, C., and Hepple, R.T. (2011). Mitochondrial structure and function are disrupted

141 by standard Isolation methods. PLoS One 6. Porter, C., Herndon, D.N., Chondronikola, M., Chao, T., Annamalai, P., Bhattarai, N., Saraf, M.K., Capek, K.D., Reidy, P.T., Daquinag, A.C., et al. (2016). Human and Mouse Brown Adipose Tissue Mitochondria Have Comparable UCP1 Function. Cell Metab. 24, 246–255. Quinlan, A.R., and Hall, I.M. (2010). BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics 26, 841–842. Rada-Iglesias, A., Bajpai, R., Swigut, T., Brugmann, S. a, Flynn, R. a, and Wysocka, J. (2011). A unique chromatin signature uncovers early developmental enhancers in humans. Nature 470, 279–283. Rajakumari, S., Wu, J., Ishibashi, J., Lim, H.W., Giang, A.H., Won, K.J., Reed, R.R., and Seale, P. (2013). EBF2 determines and maintains brown adipocyte identity. Cell Metab. 17, 562–574. Razidlo, D.F., Whitney, T.J., Casper, M.E., McGee-Lawrence, M.E., Stensgard, B.A., Li, X., Secreto, F.J., Knutson, S.K., Hiebert, S.W., and Westendorf, J.J. (2010). Histone deacetylase 3 depletion in osteo/chondroprogenitor cells decreases bone density and increases marrow fat. PLoS One 5. Remsberg, J.R., Ediger, B.N., Ho, W.Y., Damle, M., Li, Z., Teng, C., Lanzillotta, C., Stoffers, D.A., and Lazar, M.A. (2017). Deletion of histone deacetylase 3 in adult beta cells improves glucose tolerance via increased insulin secretion. Mol. Metab. 6, 30–37. Ritchie, M.E., Phipson, B., Wu, D., Hu, Y., Law, C.W., Shi, W., and Smyth, G.K. (2015). Limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 43, e47. Robinson, M.D., and Smyth, G.K. (2008). Small-sample estimation of negative binomial dispersion, with applications to SAGE data. Biostatistics 9, 321–332. Robinson, J.T., Thorvaldsdóttir, H., Winckler, W., Guttman, M., Lander, E.S., Getz, G., and Mesirov, J.P. (2011). Integrative genomics viewer. Nat. Biotechnol. 29, 24–26. Robinson, M.D., McCarthy, D.J., and Smyth, G.K. (2010). edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics 26, 139–140. Rodgers, J.T., Lerin, C., Haas, W., Gygi, S.P., Spiegelman, B.M., and Puigserver, P. (2005). Nutrient control of glucose homeostasis through a complex of PGC-1alpha and SIRT1. Nature 434, 113–118. Rodgers, J.T., Lerin, C., Gerhart-Hines, Z., and Puigserver, P. (2008). Metabolic 142 adaptations through the PGC-1 alpha and SIRT1 pathways. FEBS Lett. 582, 46– 53. Rogge, G.A., Singh, H., Dang, R., and Wood, M.A. (2013). HDAC3 is a negative regulator of cocaine-context-associated memory formation. J. Neurosci. 33, 6623–6632. Ruas, J.L., White, J.P., Rao, R.R., Kleiner, S., Brannan, K.T., Harrison, B.C., Greene, N.P., Wu, J., Estall, J.L., Irving, B.A., et al. (2012). A PGC-1α isoform induced by resistance training regulates skeletal muscle hypertrophy. Cell 151, 1319–1331. Rundlett, S.E., Carmen, a a, Kobayashi, R., Bavykin, S., Turner, B.M., and Grunstein, M. (1996). HDA1 and RPD3 are members of distinct yeast histone deacetylase complexes that regulate silencing and transcription. Proc. Natl. Acad. Sci. U. S. A. 93, 14503–14508. Saito, M., Okamatsu-Ogura, Y., Matsushita, M., Watanabe, K., Yoneshiro, T., Nio-Kobayashi, J., Iwanaga, T., Miyagawa, M., Kameya, T., Nakada, K., et al. (2009). High incidence of metabolically active brown adipose tissue in healthy adult humans: Effects of cold exposure and adiposity. Diabetes 58, 1526–1531. Schreiber, S.N., Knutti, D., Brogli, K., Uhlmann, T., and Kralli, A. (2003). The transcriptional coactivator PGC-1 regulates the expression and activity of the orphan nuclear receptor estrogen-related receptor alpha (ERRalpha). J. Biol. Chem. 278, 9013–9018. Schulz, T.J., Huang, P., Huang, T.L., Xue, R., McDougall, L.E., Townsend, K.L., Cypess, A.M., Mishina, Y., Gussoni, E., and Tseng, Y.-H. (2013). Brown-fat paucity due to impaired BMP signalling induces compensatory browning of white fat. Nature 495, 379–383. Schwer, B., and Verdin, E. (2008). Conserved Metabolic Regulatory Functions of Sirtuins. Cell Metab. 7, 104–112. Seto, E., and Yoshida, M. (2014). Erasers of histone acetylation: The histone deacetylase enzymes. Cold Spring Harb. Perspect. Biol. 6. Shapira, S.N., Lim, H.W., Rajakumari, S., Sakers, A.P., Ishibashi, J., Harms, M.J., Won, K.J., and Seale, P. (2017). EBF2 transcriptionally regulates brown adipogenesis via the histone reader DPF3 and the BAF chromatin remodeling complex. Genes Dev. 31, 660–673. Siersbæk, R., Baek, S., Rabiee, A., Nielsen, R., Traynor, S., Clark, N., Sandelin, A., Jensen, O.N., Sung, M.H., Hager, G.L., et al. (2014). Molecular architecture of transcription factor hotspots in early adipogenesis. Cell Rep. 7, 1434–1442. Siersbæk, R., Madsen, J.G.S., Javierre, B.M., Nielsen, R., Bagge, E.K., Cairns, 143

J., Wingett, S.W., Traynor, S., Spivakov, M., Fraser, P., et al. (2017). Dynamic Rewiring of Promoter-Anchored Chromatin Loops during Adipocyte Differentiation. Mol. Cell 66, 420–435.e5. Singh, N., Trivedi, C.M., Lu, M., Mullican, S.E., Lazar, M.A., and Epstein, J.A. (2011). Histone deacetylase 3 regulates smooth muscle differentiation in neural crest cells and development of the cardiac outflow tract. Circ. Res. 109, 1240– 1249. Singh, N., Gupta, M., Trivedi, C.M., Singh, M.K., Li, L., and Epstein, J.A. (2013). Murine craniofacial development requires Hdac3-mediated repression of Msx gene expression. Dev. Biol. 377, 333–344. Soccio, R.E., Chen, E.R., Rajapurkar, S.R., Safabakhsh, P., Marinis, J.M., Dispirito, J.R., Emmett, M.J., Briggs, E.R., Fang, B., Everett, L.J., et al. (2015). Genetic Variation Determines PPARγ Function and Anti-diabetic Drug Response in Vivo. Cell 162, 33–44. Spencer, T.E., Jenster, G., Burcin, M.M., Allis, C.D., Zhou, J., Mizzen, C. a, McKenna, N.J., Onate, S. a, Tsai, S.Y., Tsai, M.J., et al. (1997). Steroid receptor coactivator-1 is a histone acetyltransferase. Nature 389, 194–198. Stengel, K.R., Barnett, K.R., Wang, J., Liu, Q., Hodges, E., Hiebert, S.W., and Bhaskara, S. (2017). Deacetylase activity of histone deacetylase 3 is required for productive VDJ recombination and B-cell development. Proc. Natl. Acad. Sci. U. S. A. Stengel KR, Zhao Y, Kaiser JF, Gordy LE, Joyce S, Hiebert SW, S.A. (2015). Histone Deacetylase 3 Is Required for Efficient T Cell Development. Mol Cell Biol 35, 3854–3865. Step, S.E., Lim, H.-W., Marinis, J.M., Prokesch, A., Steger, D.J., You, S.-H., Won, K.-J., and Lazar, M.A. (2014). Anti-diabetic rosiglitazone remodels the adipocyte transcriptome by redistributing transcription to PPAR-driven enhancers. Genes Dev. 28, 1018–1028. Summers, A.R., Fischer, M.A., Stengel, K.R., Zhao, Y., Kaiser, J.F., Wells, C.E., Hunt, A., Bhaskara, S., Luzwick, J.W., Sampathi, S., et al. (2013). HDAC3 is essential for DNA replication in hematopoietic progenitor cells. J. Clin. Invest. 123, 3112–3123. Sun, Z., Singh, N., Mullican, S.E., Everett, L.J., Li, L., Yuan, L., Liu, X., Epstein, J.A., and Lazar, M.A. (2011). Diet-induced lethality due to deletion of the Hdac3 gene in heart and skeletal muscle. J. Biol. Chem. Sun, Z., Miller, R.A., Patel, R.T., Chen, J., Dhir, R., Wang, H., Zhang, D., Graham, M.J., Unterman, T.G., Shulman, G.I., et al. (2012). Hepatic Hdac3

144 promotes gluconeogenesis by repressing lipid synthesis and sequestration. Nat. Med. 18, 934–942. Sun, Z., Feng, D., Fang, B., Mullican, S.E., You, S.H., Lim, H.W., Everett, L.J., Nabel, C.S., Li, Y., Selvakumaran, V., et al. (2013). Deacetylase-Independent function of HDAC3 in transcription and metabolism requires nuclear receptor corepressor. Mol. Cell 52, 769–782. Tamkun, J.W., Deuring, R., Scott, M.P., Kissinger, M., Pattatucci, A.M., Kaufman, T.C., and Kennison, J.A. (1992). brahma: A regulator of Drosophila homeotic genes structurally related to the yeast transcriptional activator SNF2 SWI2. Cell 68, 561–572. Taunton, J., Hassig, C.A., and Schreiber, S.L. (1996). A mammalian histone deacetylase related to the yeast transcriptional regulator Rpd3p. Science (80-. ). 272, 408–411. Thorvaldsdóttir, H., Robinson, J.T., and Mesirov, J.P. (2013). Integrative Genomics Viewer (IGV): High-performance genomics data visualization and exploration. Brief. Bioinform. 14, 178–192. Tilg, H., Moschen, A.R., and Roden, M. (2016). NAFLD and diabetes mellitus. Nat. Rev. Gastroenterol. Hepatol. 14, 32–42. Ukropec, J., Anunciado, R.P., Ravussin, Y., Hulver, M.W., and Kozak, L.P. (2006). UCP1-independent thermogenesis in white adipose tissue of cold- acclimated Ucp1-/- mice. J. Biol. Chem. 281, 31894–31908. Uldry, M., Yang, W., St-Pierre, J., Lin, J., Seale, P., and Spiegelman, B.M. (2006). Complementary action of the PGC-1 coactivators in mitochondrial biogenesis and brown fat differentiation. Cell Metab. 3, 333–341. Villena, J.A., and Kralli, A. (2008). ERRα: a metabolic function for the oldest orphan. Trends Endocrinol. Metab. 19, 269–276. Villena, J. a, Hock, M.B., Chang, W.Y., Barcas, J.E., Giguère, V., and Kralli, A. (2007). Orphan nuclear receptor estrogen-related receptor alpha is essential for adaptive thermogenesis. Proc. Natl. Acad. Sci. U. S. A. 104, 1418–1423. Vincent, S.D., and Buckingham, M.E. (2010). How to make a heart. The origin and regulation of cardiac progenitor cells. Curr. Top. Dev. Biol. 90, 1–41. Virtanen, K.A., Lidell, M.E., Orava, J., Heglind, M., Westergren, R., Niemi, T., Taittonen, M., Laine, J., Savisto, N.J., Enerback, S., et al. (2009). Functional brown adipose tissue in healthy adults. N Engl J Med 360, 1518–1525. Wang, D., Garcia-Bassets, I., Benner, C., Li, W., Su, X., Zhou, Y., Qiu, J., Liu, W., Kaikkonen, M.U., Ohgi, K. a, et al. (2011). Reprogramming transcription by 145 distinct classes of enhancers functionally defined by eRNA. Nature 474, 390– 394. Wang, H., Guan, Y., Karamercan, M.A., Ye, L., Bhatti, T., Becker, L.B., Baur, J.A., and Sims, C.A. (2015a). Resveratrol Rescues Kidney Mitochondrial Function Following Hemorrhagic Shock. Shock 44, 173–180. Wang, L., Liu, Y., Han, R., Beier, U.H., Bhatti, T.R., Akimova, T., Greene, M.I., Hiebert, S.W., and Hancock, W.W. (2015b). FOXP3 + regulatory T cell development and function require histone / protein deacetylase 3. J. Clin. Invest. 125, 1–13. Wang, X., Wang, Y., Snitow, M.E., Stewart, K.M., Li, S., Lu, M.M., and Morrisey, E.E. (2016a). Expression of histone deacetylase 3 instructs alveolar type I cell differentiation by regulating a Wnt signaling niche in the lung. Dev. Biol. 414, 161–169. Wang, Y., Frank, D.B., Morley, M.P., Zhou, S., Wang, X., Lu, M.M., Lazar, M.A., and Morrisey, E.E. (2016b). HDAC3-Dependent Epigenetic Pathway Controls Lung Alveolar Epithelial Cell Remodeling and Spreading via miR-17-92 and TGF- β Signaling Regulation. Dev. Cell 36, 303–315. Watson, P.J., Fairall, L., Santos, G.M., and Schwabe, J.W.R. (2012a). Structure of HDAC3 bound to co-repressor and inositol tetraphosphate. Nature 481, 335– 340. Watson, P.J., Fairall, L., and Schwabe, J.W.R. (2012b). Nuclear hormone receptor co-repressors: Structure and function. Mol. Cell. Endocrinol. 348, 440– 449. Yam, A.Y., Xia, Y., Lin, H.-T.J., Burlingame, A., Gerstein, M., and Frydman, J. (2008). Defining the TRiC/CCT interactome links chaperonin function to stabilization of newly made proteins with complex topologies. Nat. Struct. Mol. Biol. 15, 1255–1262. Yang, X.-J., and Seto, E. (2008). The Rpd3/Hda1 family of lysine deacetylases: from bacteria and yeast to mice and men. Nat. Rev. Mol. Cell Biol. 9, 206–218. Yang, XJ; Ogryzko, VV; Nishikawa, J; Howard, BH; Nakatani, Y. (1996). A p300/CBP-associated factor that competes with the adenoviral oncoprotein E1A. Nature 382, 319–324. Yoon, H.G., Chan, D.W., Huang, Z.Q., Li, J., Fondell, J.D., Qin, J., and Wong, J. (2003). Purification and functional characterization of the human N-CoR complex: The roles of HDAC3, TBL1 and TBLR1. EMBO J. 22, 1336–1346. You, S.-H., Lim, H.-W., Sun, Z., Broache, M., Won, K.-J., and Lazar, M.A. (2013). Nuclear receptor co-repressors are required for the histone-deacetylase activity 146 of HDAC3 in vivo. Nat. Struct. Mol. Biol. 20, 182–187. Zhang, J., Kalkum, M., Chait, B.T., and Roeder, R.G. (2002). The N-CoR-HDAC3 nuclear receptor corepressor complex inhibits the JNK pathway through the integral subunit GPS2. Mol. Cell 9, 611–623. Zhang, L., He, X., Liu, L., Jiang, M., Zhao, C., Wang, H., He, D., Zheng, T., Zhou, X., Hassan, A., et al. (2016). Hdac3 Interaction with p300 Histone Acetyltransferase Regulates the Oligodendrocyte and Astrocyte Lineage Fate Switch. Dev. Cell 36, 316–330. Zhang, Y., Huypens, P., Adamson, A.W., Chang, J.S., Henagan, T.M., Boudreau, A., Lenard, N.R., Burk, D., Klein, J., Perwitz, N., et al. (2009). Alternative mRNA splicing produces a novel biologically active short isoform of PGC-1α. J. Biol. Chem. 284, 32813–32826. Zhang, Y., Fang, B., Emmett, M.J., Damle, M., Sun, Z., Feng, D., Armour, S.M., Remsberg, J.R., Jager, J., Soccio, R.E., et al. (2015). GENE REGULATION. Discrete functions of nuclear receptor Rev-erbα couple metabolism to the clock. Science 348, 1488–1492. (2011). Using the SRA Toolkit to convert .sra files into other formats. In: SRA Knowledge Base.

147