Distinct histone deacetylases repress expression of LH and FSH beta genes in the immature gonadotrope alphat3 1 cells and the repression is reversed by GNRH

Chromatin immunoprecipitation ChIP assays have shown that in the immature pituitary gonadotrope αT3-1 cells, distinct sets of histone deacetylases HDACs, along with SMRT and Sin3A, assoc

DISTINCT HISTONE DEACETYLASES REPRESS EXPRESSION OF LH AND FSH β GENES IN THE

REPRESSION IS REVERSED BY GNRH

THE NATIONAL UNIVERSITY OF SINGAPORE

2007

D ISTINCT HISTONE DEACETYLASES REPRESS EXPRESSION OF LH AND FSH β GENES IN THE

REPRESSION IS REVERSED BY GNRH

A T HESIS S UBMITTED

F OR THE D EGREE OF M ASTERS OF S CIENCE

D EPARTMENT OF B IOLOGICAL S CIENCES

T HE N ATIONAL U NIVERSITY OF S INGAPORE

2007

This is perhaps the easiest and hardest chapter that I have to write It will be simple to name all the people that helped to get this done, but it will be tough to thank them enough I will nonetheless try…

First of all, I must send thanks to all the HDAC team members Without their support and striving development the project would have never reached the great result it did

Dr Philippa Melamed, my supervisor, she always encourages me to bring forth my own ideas and to test them independently Those warm discussions, just as warm as the weather in Singapore, are unforgettable I am very happy that at the early stage of my research life, I could establish my faith on science and learn to figure out the problems, which is common in bio-research, with great enthusiasm

My wonderful lab mates, Luo Min, Stefan, Jia Jun, Siew Hoon, Fai, who’ve made our lab

a lively and enjoyable place to work in And those previous colleagues, their preliminary tests and hypothesis are the source of my inspiration Dr Martin Lee, his suggestions really saved me for some of the protein tests

My lovely friends, church or campus which are too many to mention, always stood by

my side asking over and over again “When will you get it done? Next week? Next Month? When?”

My final words go to my family In this type of work the relatives are always mistreated

A great thanks to all

The gonadotropins, luteinizing hormone (LH) and follicle-stimulating hormone (FSH) are synthesized in and secreted by the pituitary gland, and play crucial roles in regulating reproduction The synthesis of both LH and FSH is repressed soon after birth until puberty, when the repression is reversed by gonadotropin-releasing hormone (GnRH) Chromatin immunoprecipitation (ChIP) assays have shown that in the immature pituitary gonadotrope αT3-1 cells, distinct sets of histone deacetylases (HDACs), along with SMRT and Sin3A, associate with LH and/or FSH β-subunit gene promoters in a repressive complex In order to understand the de-repression of LH and FSH β genes in αT3-1 cells, the effect of GnRH treatment on this repressive complex must be elucidated ChIP assays have shown that GnRH is able to remove several HDACs and Sin3A from

LH and/or FSH β gene promoters, which results in the disruption of the repressive complex De-repression of the LH and FSH β genes after GnRH stimulation might be caused by class IIa HDAC modifications which lead to the nuclear export of those HDACs SENP1, a nuclear protease that appears to deconjugate sumoylated proteins, reverses the repression of both the LH and FSH β genes in the αT3-1 cells It is possible that GnRH stimulation recruits the de-sumoylation pathway to export the class IIa HDACs from the nucleus resulting in the disruption of the repressive complex

TABLE OF CONTENTS

LIST OF FIGURES 1

LIST OF TABLES 3

LIST OF ABBREVIATIONS 4

CHAPTER 1 1 INTRODUCTION N 6

1.1The gonadotropins: lutenizing hormone (LH) and follicle-stimulating hormone (FSH) 6

1.2 Gonadotropin-releasing hormone (GnRH) regulates LH and FSH β-subunit synthesis 10

1.2.1 Basal expression of LH and FSH β-subunit genes 10

1.2.2 GnRH-mediated expression of LH and FSH β-subunit genes 12

1.3 Regulation of LH and FSH β genes through their transcriptional repression 15

1.3.1 Histone deacetylases (HDACs) repress gene expression 15

1.3.2 Transcription regulators: class IIa HDACs, N-CoR/SMRT and mSin3A 20

1.3.3 Modification of Class IIa HDACs by phosphorylation or SUMOylation affects their repressive functions .24

1.4 Mouse gonadotrope cell lines: αT3-1 and LβT2 28

1.5 Hypothesis and aims 30

CHAPTER 2 2 MATERIALS AND METHODS S 31

2.1 Tissue culture 31

2.1.1 Growth condition 31

2.1.2 Transient transfection 31

2.2 Preparation of plasmid DNA 32

2.2.1 Expression vectors 32

2.2.2 siRNA constructs to target N-CoR, SMRT and Sin3A 32

2.2.2.1 Design of oligonucleotides 32

2.2.2.2 Annealing of oligonucleotides 33

2.2.2.3 Restriction digestion of vectors 34

2.2.2.4 DNA purification 34

2.2.2.5 Ligation of annealed oligos and linearized pSUPER vector 34

2.2.3 Isolation, verification and large scale preparation 35

2.2.3.1 Transformation of plasmids into Escherichia coli (E.coli) 35

2.2.3.2 Plasmid isolation and verification 36

2.3 RT-PCR analysis 37

2.3.1 RNA isolation 37

2.3.2 First strand cDNA synthesis 38

2.3.3 PCR and gel electrophoresis analysis 38

2.4 Western blotting 40

2.4.1 Whole cell extraction 40

2.4.2 Nuclear and cytoplasmic extraction 40

2.4.3 SDS page and blotting 40

2.5 Co-immunorecipitation (Co-IP) 43

2.6 Chromatin Immunoprecipitation (CHIP) 44

2.6.1 Cross-linking of protein and DNA 44

2.6.2 Immunoprecipitation of protein-DNA complex and DNA extraction 45

2.6.3 PCR detection 46

CHAPTER 3 3 RESULTS S 48

3.1 Gonadotropin β-subunit genes are repressed by HDACs in the immature gonadotrope αT3-1 cell line and GnRH is able to overcome this repression 48

3.2 Co-repressors are identified that repress the FSH β gene in both αT3-1 and LβT2 cells 50

3.3 Distinct sets of HDACs and co-repressors are associated with the LH and FSH β gene promoters, and the association is affected by GnRH treatment 53

3.4 Co-immunoprecipitation indicates that the repressive factors associated with the LH and FSH β-subunit gene, are contained in more than one complex at each gene promoter 60

3.5 GnRH-mediated modification of HDAC4 and HDAC5 facilitates their nuclear export 63

CHAPTER 4 4 DISCUSSION N 67

LIST OF FIGURES

Figure 1 Functional connection between the hypothalamus and pituitary gland 7 Figure 2 Regulation of gonadotropin gene expression 8 Figure 3 A model of basal gonadotropin subunit gene expression 11 Figure 4 The gonadotropin-releasing hormone (GnRH) receptor-modulated signaling

network

13 Figure 5 Histone acetylation-deacetylation cycle 16 Figure 6 Schematic depiction of the different isoforms of various HDACs 18 Figure 7 Repressive complexes are associated with the promoter to repress gene

anterior pituitary

29

Figure 12 LH and FSH β-subunit gene expression is repressed in immature αT3-1

cells and this is overcome by GnRH

49 Figure 13 SMRT represses expression of the FSHβ gene in both cell types 50 Figure 14 In αT3-1 cells, both transcript and protein levels of N-CoR and SMRT are

decreased following the respective siRNA-mediated knock down

52

Figure 15 In αT3-1 cells, several HDACs are associated with the LH β promoter and

the association is differentially affected by GnRH treatment

54

Figure 16 In αT3-1 cells, several HDACs are associated with the FSH β promoter

and this is differentially affected by GnRH treatment

55

Figure 17 In αT3-1 cells, neither N-CoR nor SMRT is associated with the LH β

promoter but SMRT is recruited following GnRH treatment

57

Figure 18 In αT3-1 cells, SMRT is associated with the FSH β promoter and this is

not affected by GnRH treatment

58

Figure 19 In LβT2 cells, several HDACs along with co-repressors are associated with

the FSH β promoter and the association is differentially affected by GnRH treatment

59

Figure 20 In αT3-1 cells, class I HDACs precipitate with class IIa HDACs and

co-repressors

61

Figure 21 In αT3-1 cells, HDAC4 co-precipitated with the co-repressor Sin3A but

not with either SMRT or HDAC5

62

Figure 22 In αT3-1 cells, the repression of LH and FSH β-subunit genes is overcome

by over-expression of SENP1

64

Figure 23 In αT3-1 cells, HDAC5 is present in two forms in both cytoplasmic and

nuclear extraction, one appears 20 kD bigger, and this is not affected by NEM

treatment

64

Figure 24 In αT3-1 cells, localization of wild-type HDAC5 in both nucleus and

cytoplasm is SUMO-dependant, and this is affected by GnRH treatment

66 Figure 25 SUMOylation and nuclear import 74 Figure 26 Repressive complexes containing distinct HDACs repress expression of

the gonadotropin β-subunit genes in αT3-1 cells, and this is overcome by GnRH

treatment

78

LIST OF TABLES

Table 1 Optimized amount of plasmids transfected into the cell lines 32

Table 2 Oligonucleotides designed for synthesis of siRNA 33

Table 3 Conditions for annealing of oligonucleotides 33

Table 4 Components of reaction for T4 ligatio 34

Table 5 Mix preparation for the restriction digestion 36

Table 6 Mix for the sequencing reaction 37

Table 7 Cycling parameters for sequencing reaction 37

Table 8 Mix for the first strand cDNA synthesis 38

Table 9 Mix of PCR to test expression level of the LHβ, FSHβ and β-actin 39 Table 10 PCR cycling parameters to analyze LHβ, FSHβ and β-actin gene expression 39

Table 11 Primers used to amplify LHβ, FSHβ and β-actin 39

Table 12 Composition of buffers used in western blot 42

Table 13 Antibodies used in western blotting; Regarding anti-SMRT, one is

particularly used for western blotting following co-precipitation 42

Table 14 Antibodies used in immuno-precipitation 43

Table 15 Composition of buffers used in ChIP experiments 46

Table 16 Antibodies used in ChIP 47

Table 17 PCR cycling parameters to amply specific regions of the FSHβ and LHβ

Table 18 Primers used to amply LH and FSH β promoter region 47

LIST OF ABBREVIATIONS

CaM Ca 2+ sensor calmodulin

CaMKs Ca 2+ /CaM-dependent protein kinases

ChIP Chromatin immunoprecipitation

DAD Deacetylase activating domain

DAG Diacylglycerol

EGF Epidermal growth factor

ERK Extracellular-signal-regulated kinase

FSH Follicle-stimulating hormone

FSHβ Follicle-stimulating hormone β-subunit

GnRH Gonadotropin-releasing hormone

GnRHR Gonadotropin-releasing hormone receptor

HAT Histone acetyltransferase

Hda1 Histone deacetylase 1

HDAC Histone deacetylase

HDACi Histone deacetylase inhibitors

IP3 Inositol 1,4,5 triphosphate

JNK Jun N-terminal kinase

LH Luteinizing hormone

LHR Luteinizing hormone receptor

LHβ Luteinizing hormone β-subunit

MAPK Mitogen-activated protein kinase

MEK Mitogen-activated protein kinase kinase

N-CoR Nuclear receptor corepressor

NES Nuclear export signal

NLS Nuclear localization signal

NPC Nuclear pore complex

NURD Nucleosome remodeling and histone deacetylation

PIP Plasmid immunoprecipitation

Pitx1 Pituitary homeobox 1

PKA Protein kinase A

PKC Protein kinase C

PLC Phospholipase C

SENP Sentrin/SUMO-specific protease

Sf-1 Steroidogenic factor 1

Sir2 Silent information regulator 2

siRNA Short interfering ribonucleic acids

SMRT Silencing mediator of retinoic and thyroid hormone receptors

SUMO Small ubiquitin-related modifier

TSA Trichostatin A

Ubc Ubiquitin conjugating

or rear lobe (Figure 1) The anterior pituitary is composed of a number of different cell

types, including five endocrine cells (Jacobson et al 1979) One of these are gonadotrope

cells which secrets two gonadotropins: luteinizing hormone (LH) and follicle-stimulating hormone (FSH) The synthesis and secretion of gonadotropins are subject to the complex

control of many factors (Figure 2) including GnRH (Papavasiliou et al 1986; Kato et al

1989; Ruf and Sealfon 2004), steroid hormones (testosterone, estrogen and progesterone)

and gonadal peptides (activin, inhibin and follistatin) (Gharib et al 1990; Joshi et al

1993)

Anterior pituitary Posterior pituitary

Figure 1 Functional connection between the hypothalamus and pituitary gland

The anterior pituitary gland is functioning connected with the hypothalamus; nerve cells in the hypothalamus secrete neurohormones that act on the endocrine cells of the anterior lobe to stimulate or inhibit their synthesis and secretion Abbreviations: AL, anterior lobe; PL, posterior lobe; MB, mammillary body (Nussey and Whitehead 1999)

Pituitary Hypothalamus

Steroids

Testosterone, Estrogen, Progesterone

Figure 2 Regulation of gonadotropin gene expression GnRH, synthesized in and

released from the hypothalamus, binds to GnRH receptors on the surface of the gonadotrope This leads to the synthesis and secretion of LH and FSH, which stimulate the production of steroid hormones Testosterone, estrogen and progesterone negatively or positively regulate the synthesis of the gonadotropins directly at the pituitary or indirectly by modulating GnRH secretion from the hypothalamus The gonadal peptides, inhibin, activin and follistatin, also have roles

in the regulation of gonadotropin gene expression by exerting positive or negative feedback (Brown and Mcneilly 1999)

Both LH and FSH are glycoproteins and are composed of one α-subunit which is identical, and one β-subunit which is unique and endows each hormone with the ability

to bind to its own receptor These hormones stimulate the activation of the gonads: in females, the ovaries; and in males, the testes FSH in females initiates follicular growth, specifically through the actions on granulosa cells, whereas in males, FSH stimulates the maturation of germ cells The name LH is derived from its effect of inducing luteinization of ovarian follicles LH receptors are expressed on the maturing follicle that produces an increasing amount of estradiol with the rise in estrogens With maturation of the follicle, the estrogen rise leads a surge in LH levels over a 24-48 hour period This

LH surge triggers ovulation and initiates the conversion of the residual follicle into a

corpus luteum that, in turn, produces progesterone to prepare the endometrium for a

possible implantation Progesterone is necessary for maintenance of pregnancy, and in

humans, LH is required for continued development and function of corpora lutea In

males, LH acts upon the Leydig cell of the testis and stimulates testosterone production that promotes spermatogenesis and is responsible for the male secondary sexual characteristics

1.2 Gonadotropin-releasing hormone (GnRH) regulates LH and FSH β-subunit

synthesis

1.2.1 Basal expression of LH and FSH β-subunit genes

During embryogenesis, a controlled cascade involving multiple signaling pathways determines the transcription factor expression which initiates the basal expression of

gonadotropin genes (Treier et al 1998) The initiation of the gonadotrope cell lineage is

characterized by expression of the α-subunit followed by the expression of LH and FSH

β subunit transcripts after a further 5 - 6 days Transcription factors such as pituitary homeobox 1 (Ptx1) and steroidogenic factor 1 (SF-1) have been reported to activate

gonadotropin α and β subunit gene transcription at the basal level (Figure 3) Although

both LH and FSH are produced during fetal development, their synthesis is repressed after birth until re-activation at puberty, when the GnRH pulse generator is activated, resulting in increased GnRH release (Grumbach 2003) As such, the appropriate GnRH delivery to the gonadotrope is the only endogenous block to the reawakening of the GnRH-gonadoptropin axis at puberty which has been shown by the experiment that precocious puberty leading to ovulation can be stimulated in monkeys merely by the

appropriate administration of GnRH (Wildt et al 1980; Plant et al 1989)

Figure 3 A model of basal gonadotropin subunit gene expression Transcription

factors involved in basal gonadotropin subunit gene expression are activated during anterior pituitary development and are shown bound to their cognate DNA elements

at the gonadotropin subunit gene promoters (Brown and Mcneilly 1999)

1.2.2 GnRH-mediated expression of LH and FSH β-subunit genes

The decapeptide GnRH (pGlu-His-Trp-Ser-Tyr-Gly-Leu-Arg-Pro-Gly-NH2), released in

a pulsatile pattern by GnRH-producing neurons of the hypothalamus, is the chief regulator of the reproductive system in mammals The GnRH receptor, a heptahelical membrane protein on the surface of the anterior pituitary gonadotrope is activated following GnRH binding, resulting in a signal transduction network After receptor activation, the signal transduction network of the gonadotrope reliably decodes the instructions received to generate the appropriate rates of gonadotropin bio-systhesis and

secretion (Figure 4)

When GnRH interacts with its receptor, it stabilizes a conformational change in the receptor that promotes the activation of heterotrimeric G proteins The principle G protein activated by GnRH belongs to the Gq/11 subclass, while G proteins of the Gi/o

and Gs subclasses have also been reported to be activated by GnRH (Ruf et al 2003) In

αT3-1 cells, the endogenous mouse receptor was found to be consistent with activation solely of Gq/11 subtype G-proteins, while, in the related LβT2 gonadotrope cell line, the endogenous mouse receptor was found to activate both Gq/11 and Gs sub-type G-

proteins (Grosse et al 2000; Liu et al 2002b)

Figure 4 The gonadotropin-releasing hormone (GnRH) receptor-modulated signaling network Activation of the

GnRH receptor leads to the activation of at least two G-protein subtypes, Gs and Gq Signaling downstream of protein kinase C (PKC) leads to transactivation of the epidermal growth factor (EGF) receptor and activation of mitogen-activated protein kinases (MAPKs), including extracellular-signal-regulated kinase (ERK), Jun N-terminal kinase (JNK) and p38 MAPK Active MAPKs translocate to the nucleus, resulting in activation of transcription factors and rapid induction of early genes This figure illustrates the distributed and interconnected movement of information from the receptor to the

Phospholipase Cβ is activated by Gq/11 proteins (Hsieh and Martin 1992), leading to the hydrolysis of phosphatidylinositol 4, 5-bisphosphate to 1,4,5 - inositol trisphosphate (IP3) and diacylglycerol (DAG) IP3 mobilizes intracellular calcium which activates conventional protein kinase C (PKC) isoforms such as β and βII, which have been

identified in gonadotrope cell lines (Junoy et al 2002; Liu et al 2002a) Phospholipase D

is also activated by GnRH-receptor signaling (Shacham et al 2001), and subsequently releases DAG (Zheng et al 1994), which might cause the activation of Ca2+-independent

PKC isoforms such as PKCδ or PKCε (Shacham et al 1999) Signaling downstream of

PKC leads to transactivation of the epidermal growth factor (EGF) receptor and activation of mitogen-activated protein kinases (MAPK), including extracellular-signal-regulated kinase (ERK), Jun N-terminal kinase (JNK) and p38 MAPK Active MAPKs translocate to the nucleus, resulting in activation of transcription factors and rapid induction of early genes

GnRH has also been reported to stimulate an increase in cAMP (Bourne 1988; Garrel et

al 1997) Burger and coworkers performed in vitro studies in rat pituitary cells to detect

the stimulatory effect of cAMP on gonadotropin subunit mRNA and found that a diffusible cAMP analog stimulated a rise in α, but not LH or FSH β mRNA Interestingly, more recent studies suggest that the cAMP/PKA patyway plays a role in cross-talk between specific intracellular messenger systems like PKC and ERK in response to

GnRH stimulation (Burger et al 2004)

1.3 Regulation of LH and FSH β genes through their transcriptional repression

1.3.1 Histone deacetylases (HDACs) repress gene expression

Distinct sets of HDACs have been found to be associated with the LH or FSH β gene promoter, repressing the expression of both gonadotropin β-subunit genes, and GnRH

was shown to overcome this repression (Lim et al 2007) Reversible suppression of gene

expression is achieved through the actions of DNA-associated repressors, which block the binding sites of activators and/or compact the chromatin making it less accessible to the activator, through recruitment of co-repressors and chromatin modifying enzymes Those repressors One of these well-studied chromatin modifications is histone

deacetylation (Figure 5a) Histone deacetylation is caused by HDACs, which act to

deacetylate histone tails in the nucleosomes that bind to the TATA box and other

regulatory regions of the genes they repress In vitro studies have shown that when

promoter DNA is assembled onto a nucleosome with deacetylated histones, the general transcription complex is not able to bind to the TATA box and initiation region, because the deacetylated N-terminal lysines of the hitones are positively charged and interact

strongly with DNA phosphates (Figure 5b) The deacetylated histone tails also interact

with neighboring histone octamers, favoring the folding of chromatin into condensed, higher-ordered structure, although its precise conformation is not well understood

(Lodish et al 2004)

a

b

Figure 5 Histone acetylation-deacetylation cycle a Equilibrium of steady-state

histone acetylation is maintained by opposing activities of histone acetyltransferases and deacetylases Acetyl coenzyme A is the high-energy acetyl moiety donor for histone acetylation Histone acetyltransferases (HATs) transfer the acetyl moiety to the ε-NH3 group of internal lysine residues of histone N-terminal domains Reversal of

this reaction is catalyzed by histone deacetylases (HDACs) (Kuo and Allis 1998) b

Histone tails are modified to alter transcriptional competence, change interactions with the DNA and to serve as a code for interacting proteins which has been entitled as the

“histone code”

Mammalian HDACs are usually classified into two classes based on their sequence similarity to yeast HDACs Class I includes HDAC 1, 2, 3, 8 and 11; class II includes

HDAC 4, 5, 6, 7, 9 and 10 (Figure 6) Members of class I contain a well-conserved

catalytic domain that in HDAC1, 2 and 3 encompasses almost two thirds of the protein (Khochbin and Wolffe 1997) HDAC1 and HDAC2 were identified as components of two multi-protein complexes known as Sin3/HDAC and NuRD/Mi2/NRD (Knoepfler

and Eisenman 1999) (Figure 7) HDAC3, which was not found in either Sin3/HDAC or

NURD/Mi2/NRD core complexes, appears to be a nuclear receptor co-repressor (Ahringer 2000) Increasing evidence points to HDAC3 being a member of the stable core of the Silencing Mediator for Retinoid and Thyroid receptors (SMRT) and/or

Nuclear Receptor Co-Repressor (N-CoR) complexes (Huang et al 2000; Li et al 2000; Urnov et al 2000; Wen et al 2000)

Figure 6 Schematic depiction of the different isoforms of various HDACs Bars

depict the length of the protein The catalytic domain is shown in blue; black depicts

a NLS N, N-terminus, C, C-terminus (De Ruijter et al 2003)

Figure 7 Repressive complexes are associated with the promoter to repress gene expression

Co-repressor complexes include the Sin3/HDAC complex, which has been proposed to be recruited via the NR

co-repressors N-CoR or SMRT This complex possesses histone deacetylase activity and is thought to

reverse actions of histone acetyltransferase-containing complexes Adapted and modified from Glass and

Rosenfeld, 2000

1.3.2 Transcription regulators: class IIa HDACs, N-CoR/SMRT and mSin3A

Class II HDACs, which are almost twice the size of class I HDACs, are subdivided into two subclasses: IIa includes HDAC4, 5, 7 and 9 and its splice variant MITR, and IIb includes HDAC6 and HDAC10 Besides the catalytic domain which is located in the carboxy-terminal, class IIa HDACs contain several other domains for interacting with diverse proteins

HDAC4, 5 and 7 have been reported to function in a matrix-associated deacetylase body (MADB) which contains certain members of NuRD and Sin3 complexes, as well as the class I HDACs, HDAC1, 2 and 3 The nuclear receptor co-repressors, SMRT and N-CoR

have also been found in these bodies (Downes et al 2000; Kao et al 2000) Although the

combination of the class I and II HDACs are found in nuclear complexes, the involvement of each member is dependent on an additional level of regulation controlling

their intracellular localization (Khochbin et al 2001) HDAC4, 5 and 7 shuttle between

nucleus and the cytoplasm after binding to 14-3-3 proteins, a family of highly conserved acidic proteins (Muslin and Xing 2000) This binding is believed to be dependant on the phosphorylation of two or three conserved N-terminal serine residues in class II HDACs

and mediates their cytoplasmic sequestration (Figure 8)

Figure 8 Translocation of HDAC7 causes de-repression of Nur77 gene in developing thymic T cells HDAC7 represses Nur77 gene through either preventing

the binding of co-activators or direct associated enzymatic activities Activation of the T-cell receptor causes activation of Ca2+/CaM-dependent protein kinases (CaMKs), and elevation of intracellular Ca2+ levels, which in turn activates the Ca2+sensor calmodulin (CaM) Phosphorylation of HDAC7 by CaMKs facilitates its binding to 14-3-3 protein, mediating its CRM1-dependent nuclear export Activated CaM binds to Cabin1 and removes it from repression complex Either one of the two

pathways is able to restore the expression of Nur77 gene (Verdin et al 2003)

SMRT and N-CoR are encoded by two distinct loci but share a common molecular architecture and approximately 45% amino acid identity, while additional forms of SMRT and N-CoR are generated by alternative mRNA splicing (Privalsky 2004) Both SMRT and N-CoR can be conceptually divided into an N-terminal portion having three

or four distinct transcriptional repression domains (RDs), and a C-terminal portion

composed of two or three nuclear receptor interaction domains (NDs) (Figure 9)

Therefore, SMRT and N-CoR can be viewed as protein platforms, which recruit other repressors like HDACs, TBL-1, mSin3 and associated proteins through their RDs, while also tethering to the nuclear receptors through their NDs (Privalsky 2004)

co-SMRT and N-CoR interact with both HDAC3 and class IIa HDACs, which explains why HDAC3 co-immunoprecipitates with class IIa HDACs Class IIa HDACs are enzymatically inactive unless they bind to the SMRT/N-CoR-HDAC3 complex HDAC3

is also catalytically inactive alone as purified protein but becomes enzymatically active when bound to SMRT/N-CoR, even in the absence of class IIa HDACs In contrast, class IIa HDACs alone are still enzymatically inactive after binding to the SMRT/N-CoR

proteins in vitro (Verdin et al 2003)

Another co-repressor, Sin3 is also a large multi-domain protein, with four imperfect repeats of a paired amphipathic helix (PAH) motif that facilitates its interaction with

other proteins (Shiio et al 2006) Sin3A has been reported to interact directly with

HDAC1 and HDAC2, and is recruited by many DNA-binding transcriptional repressors SMRT and N-CoR are also found to interact directly with mSin3A, and SMRT is

Figure 9 Domains of the N-CoR and SMRT co-repressors The primary structure of the human N-CoR and murine SMRT

α Codon numbering is indicated on top The locations of the repression domains (RD1 to RD4), the deacetylase activating domain (DAD), the conserved SANT motifs that include sites of histone interaction, and of the CoRNR box/nuclear receptor interaction sites (N1, N2, and N3 in N-CoR verses S1 and S2 in SMRT) are indicated for each co-repressor Interaction sites for transcription factors that utilize SMRT and/or N-CoR for repression are indicated in yellow, whereas interaction sites for additional components of the co-repressor complex or the general transcriptional machinery are shown in red Not all

reported to form a repressor complex together with mSin3A and HDAC1 (Nagy et al

1997) mSin3A is also reported to be involved, possibly through a mSin3-binding

protein-SAP 30, in N-CoR-mediated repression (Laherty et al 1998)

1.3.3 Modification of Class IIa HDACs by phosphorylation or SUMOylation affects their repressive functions

The class IIa HDAC phosphorylation sites recognized by the 14-3-3 proteins are closely related to consensus phosphorylation sites for Ca2+/CaM-dependent protein kinases

(CaMKs) (Mckinsey et al 2000a) Overexpression of constitutively active CaMKs or

signal-dependent activation of CaMKs induces the localization of class IIa HDACs to the

cytoplasm and suppresses their repressive activity (Kerckaert et al 1993) Conversely,

mutation of the serine phosphorylation sites abolishes the HDAC nuclear export and enhances their repressive effects during muscle differentiation and T-cell apoptosis

(Mckinsey et al 2000a; Miska et al 2001) Cytoplasmic localization of class II HDACs

removes these enzymes from the chromatin and dissociates them from the

SMRT/N-CoR–HDAC3 complex, leading to the disruption of the repression complex (Fischle et al 2001; Fischle et al 2002)

Several HDACs have been reported to be targeted by SUMO, a small ubiquitin-like modifier, which is able to enhance their repressive abilities (Gill 2005) Unlike

ubiquitylation, which generally but not always, causes protein degradation through the proteasomal pathway, SUMOylation affects the function of the target protein by either altering its sub-cellular localization or by antagonizing other modifications The covalent attachment of SUMO to its target involves four enzymatic reactions, which are mediated

by SUMO protease, E1, E2 and E3 enzymes, to form an isopeptide bond between the carboxy-terminal glycine of SUMO and the ε-amino group of a lysine residue in the target protein (Seeler and Dejean 2003)

Modification of HDAC1 by SUMO contributes to the repression of AR - mediated transcription It has been reported that over-expression of SENP1, one of the Sentrin/SUMO-specific protease, enhances AR-dependent transcription, which is not mediated though de-sumoylation of AR, but rather through de-SUMOylation of HDAC1,

causing its de-conjugation and reducing its deacetylase ability (Cheng et al 2004)

HDAC4 is also modified by SUMO Mutation of the SUMO-acceptor lysine in HDAC4 correlated with a reduction in both deacetylase activity and transcriptional repressor

activity (Kirsh et al 2002) Thus, the functions of HDACs as either enzymatic histone

deacetylases or transcription co-repressors can be regulated by SUMO modification

HDACs have also been found to regulate the efficiency of SUMOylation of some substrates Because SUMOylation competes for the same lysines with other post-translational modifications like ubiquitination and acetylation, deacetylation of those lysine residues by HDACs could increase accessibility of substrate lysines to SUMO

HDAC4 has been shown to stimulate SUMOylation of the transcription factors MEF2C and MEF2D (Gregoire and Yang 2005)

An increasing amount of work also supports a role for SUMOylation in the control of chromosome dynamics In fact, all SUMO-pathway components, E1, 2, 3, and SUMO proteases, have shown to be associated with the regulation of chromosome condensation, cohesion or mitotic chromosome separation (Seeler and Dejean 2003) One important consideration regarding SUMOylation of histones is that its target lysine residue is a putative substrate for multiple modifications, not only for SUMO, but also for acetylation

and methylation (Shiio and Eisenman 2003) (Figure 10) Histone acetylation by histone

acetyltransferases recruited through co-activator complexees, correlates with gene activation Once the gene has been transcribed, its activity must be attenuated and then finally repressed The signal for recruitment of SUMOylating enzymes may be acetylation itself, as suggested by the observation that H4 SUMOylation increases with increasing H4 acetylation (Shiio and Eisenman 2003) The HDAC-mediated removal of acetyl groups occurs as a result of their recruitment by DNA-bound repressors Repression is then caused by histone methyltransferase (HMT) - mediated methylation, which is required for binding of HP1, in turn providing the structural element for

chromatin condensation (Nathan et al 2003)

Figure 10 Model for SUMOylation function in regulating transcription A gene with TATA box-containing promoter

and ORF is shown together with histone octamers/nucleosomes represented by ovals An activator, through the help of activators, can recruit a histone acetyltransferase (HAT) which acetylates histones and promotes chromatin structure amenable to transcription This acetylation can potentially recruit SUMO-conjugating enzyems (E2/E3) capable of modifying either histone or activators to achieve attenuation The repressors are able to bind DNA, probably facilitated by SUMO modification, and recruit co-repressors and histone deacetylase (HDAC) to deacetylate histones, making way for the addition

co-of repression-specific methylation marks, like H3 K9-methyl, by an HMT Finally, methylated histones (and possibly SUMO) would recruit HP1, contributing to chromatin structure in a static repressed state Adapted from Nathan et al., 2003

1.4 Mouse gonadotrope cell lines: αT3-1 and LβT2

αT3-1 and LβT2 gonadotrope cell lines were generated by targeted oncogenesis in transgenic mice The αT3-1 cell line was isolated from the carcinoma developed in the embryos following expression of the oncogene driven by the gonadotropin α subunit

gene promoter (Windle et al 1990), and the LβT2 cell line was derived by a similar method, using the LH β subunit gene promoter (Alarid et al 1996)

The αT3-1 cell line, which represents an immature gonadotrope, expresses the α-subunit,

GnRHr and Sf-1, but neither LH nor FSH β genes (Windle et al 1990) The LβT2 cell

line represents a fully differentiated gonadotrope, expressing GnRH receptor, both α- and

LH β-subunits; it can also be induced to express the FSHβ gene with exposure to activin

(Graham et al 1999) (Figure 11)

Figure 11 The gonadotrope cell lines along the developmental cell lineages of the anterior pituitary The distinct

stages of differentiation are represented by the immortalized pituitary cell lines created by target oncogenesis in transgenic mice In this study, the αT3-1 (immature gonadotrope) and LβT2 (fully differentiated gonadotrope) cell lines were used as comparative model systems Adapted from Alarid et al., 1996

1.5 Hypothesis and aims

The hypothesis of this study was that HDACs along with co-repressors N-CoR, SMRT and/or Sin3A comprise distinct repressive complexes at the LH and FSH β gene promoters, and that GnRH is able to disrupt the complex(es) through the removal of some of these proteins, resulting in the de-repression of both genes This removal might

be regulated by GnRH-mediated modifications of the HDACs, possibly including SUMOylation

The aims of this study were to:

1 determine whether the presence of distinct HDACs at both LH and FSH β-subunit gene promoters is affected following GnRH treatment;

2 investigate the roles of the co-repressors N-CoR and SMRT in regulating the gonadotropin β-subunit genes;

3 characterize the repressive complex(es) associated with each gene promoter;

4 investigate a possible role for SUMOylation in GnRH-mediated de-repression of these genes

CHAPTER 2 MATERIALS AND METHODS

2.1 Tissue culture

2.1.1 Growth condition

αT3-1 cells (a gift from Dr P Mellon) were cultured in minimal essential medium (MEM), containing 10% qualified fetal bovine serum (FBS), non essential amino acid solution, sodium pyruvate solution (all from GIBCO™), 100 µg/ml antimycotic solution and 10 mM HEPES (pH 7.4, Sigma Aldrich) The mature pituitary gonadotrope cell line LβT2 cells (from Dr P.Mellon, San Diego) were cultured in Dulbecco’s modified eagle medium (DMEM), certified FBS (all from Gibco™), 100 µg/ml antimycotic solution and

10 mM HEPES (pH 7.4, Sigma Aldrich) These cells were maintained under 5 % CO2 at

The amount of the plasmid for transfection (µg) Cell Type

pSUPER-N-CoR pSUPER-SMRT

(A+B) pSUPER-mSin3A

Table 1: Optimized amount of plasmids transfected into the cell lines

2.2 Preparation of plasmid DNA

Table 2: Oligonucleotides designed for synthesis of siRNA

Step wise cooling to 4 °C 20

Table 3: Conditions for annealing of oligonucleotides