Stress Potentiation of Glucocorticoid Receptor Transactivity Through

HSF1-Dependent and HSF1-Independent Pathways

Thomas Joseph Jones

Medical College of Ohio

2004 DEDICATION

This work is dedicated to my parents, Ron and Rose Jones, to my brother

Chris and sister, Ann. Their support, guidance, and sacrifices have given me the opportunity to pursue and accomplish any goal that I have set out to achieve.

ii ACKNOWLEDGEMENTS

I would like to thank my major advisor, Dr. Edwin R. Sanchez, for giving

me the opportunity to work in his lab. I greatly appreciate his patience and

generosity along with his help in developing the skills that I need to excel in a

science career.

I would also like to thank my committee members, Dr. Brian Rowan, Dr.

Linda Dokas, Dr. Zijian Xie, and Dr. Daniel Ely. They all played important roles in developing my scientific skills and significantly increasing my interest in science and the people around me. Their guidance and respect has been an invaluable tool in molding my personal and scientific endeavors.

The space allotted here would never be enough to thank the people of the

Pharmacology department for their kindness and support. In particular, I would

like to thank Ruth, Martha, and Debra for their technical support and friendly

conversations, if nothing else just to pass the time. I would also like to thank

Marji, for without her support, guidance, conversations, and wisdom, I would surely not have become the person that I am today. She is truly a great human being! I hope that every student has the opportunity to work with such great people just once in their academic or professional career.

Lastly, I want to thank my parents whose love and devotion to me, my

brother, and sister have made us successful in all of our pursuits. Their love and

sacrifice will never go unnoticed.

iii TABLE OF CONTENTS

DEDICATION ii

ACKNOWLEDGEMENTS iii

TABLE OF CONTENTS iv

INTRODUCTION 1

LITERATURE 13

MANUSCRIPT ONE 48

“Enhancement of Glucocorticoid Receptor Transactivity

by Constitutively-active Heat Shock Factor 1”

MANUSCRIPT TWO 87

“Evidence for a Stress-induced Factor that is Released

from Stressed Cells and Enhances Glucocorticoid

Receptor Responsiveness”

SUMMARY 115

BIBLOGRAPHY 120

ABSTRACT 156

iv INTRODUCTION

The focus of this project is to better understand the roles that stress and

heat shock factor 1 (HSF1) play in the potentiation of transcriptional activity from

the glucocorticoid receptor (GR). The general understanding that both

transcription factors, GR and HSF1, contain heat shock proteins (HSPs) in their

inactive complexes (Pratt et al., 1992; Scherrer et al., 1992; Zou et al., 1998),

and function to protect the cell from stress events (Sciandra and Subjeck, 1984;

Tomasovic and Koval, 1985), introduced the plausibility of a communication

network existing between HSF1, GR, and stress.

Gametchu and Harrison were the first to demonstrate the existence of a

connecting mechanism between stress and GR activity. In their research, they demonstrated a correlation between heat shock (HS) stress and the hormone- binding capacity of the GR. They found that under HS conditions, the hormone- binding capacity of the GR was reduced by as much as 85% (Gametchu and

Harrison, 1984). The thought was that stress had reduced the protein levels of the GR through heat denaturation. However, they observed the reduction in GR hormone-binding capacity, which directly correlates with the increased expression levels seen for the heat shock proteins (Gametchu and Harrison,

1984). Recent work showed that the GR protein levels had not been reduced or degraded, but up to 95% of the GR had been relocated to the nuclear compartment (Sanchez, 1992; Shen et al., 1993). These results were also observed under various other conditions like the chemical shock (CS) of 200µM sodium arsenite.

1 The inactive GR is found in the cytosol associated with a heterocomplex of

heat shock proteins and immunophilins components. In the presence of ligand, the GR dissociates from the heterocomplex and is translocated to the nucleus where it binds chromatin. Stress, in the absence of ligand, also appeared to have the ability to transform the GR to a nuclear complex. The ability of stress to transform the GR from a cytosolic heterocomplex to a DNA bound nuclear receptor was confirmed by immunoflourescence studies (Liu et al., 1995). The nuclear localized GR in these cases lost the ability to bind ligand (Sanchez,

1992; Shen et al., 1993). Explanations for the observed loss of hormone binding capacity lie in an understanding of the role that heat shock protein 90 (Hsp90) plays in steroid receptor function. Hsp90, found in the cytosol, is bound to the inactive GR as a homodimer, forming the backbone of the steroid receptor heterocomplex. Binding of Hsp90 to the GR confirms a high affinity ligand binding state on the receptor (Bohen, 1995; Srinivasan et al., 1997). Once the

GR is dissociated from the heterocomplex, the receptor undergoes a conformation change and localizes to the nuclear compartment (Pratt et al.,

1990). Under stress conditions, nuclear localization of the ligand-free GR was found to have minimal transactivation ability from the complex mammary mouse tumor virus (MMTV) promoter (Shen et al., 1993). A 2-3% transactivation of GR was shown with both HS and CS. This activation, though minimal, demonstrated stress can induce translocation and weak transcription from a nuclear receptor.

Steroid-free activation of the GR had shown a minimal induction of receptor mediated gene transcription. Stress, however, in the presence of

2 ligand, demonstrated the ability to super activate the receptor in a concentration- dependent manner, up to and including saturating levels of hormone (1µM). The hormone-activated GR appeared to harbor a latent activity that stress had uncovered. This unexpected increase in GR transactivity is known as the heat shock potentiation effect (HSPE) (Sanchez et al., 1994). The stress induced potentiation effect was not a product of promoter specificity, as it was seen when using both the complex MMTV (Danielsen et al., 1986) and the synthetic minimal

GRE2E1B promoter (Allgood et al., 1993). Both HS and CS were able to produce a 4- and 20-fold increase in ligand-dependent receptor activity (Sanchez et al., 1994; Hu et al., 1996). However, this stress-induced effect was suppressed in the presence of the GR antagonist, RU486 (Sanchez et al., 1994).

A model presented from this work, as shown in Figure 1, suggests three pathways in which the potentiation of the GR may result from stress-induced changes to receptor transactivity. The first pathway suggests stress acts directly on the GR and affects its intrinsic activity, creating a super-active transcription factor. The second pathway suggests stress induces changes at the level of the transcription machinery producing a more responsive transcription complex.

Finally, the third pathway suggests stress activation of heat shock factor 1

(HSF1), a major stress-activated pathway (Sistonen et al., 1994), produces a heat shock adaptor (HSA) protein. This HSA could then function as a coactivator or could release a GR suppressor molecule and, in turn, increase the receptor- mediated expression. The observation that stress potentiation of the GR begins

3 FIGURE 1. Stress Signaling in GR Enhancement

8 h after the stress event, peaking between 16 h and 24 h, suggests that the potentiation occurred through a stress-inducible factor. This was the basis for the hypothesis that stress had activated or synthesized a factor during the

4 recovery phase that interacted with the GR to enhance transcription (Li et al.,

1999).

This hypothesis led to the investigation of HSF1 as a stress-activated site

of modulation and control over GR transactivity. This research focus was based

on previous work which demonstrated a correlation among the stress-induced

rise in HSP levels, a simultaneous decrease in overall cellular protein levels, and

the time dependent response of the HSPE (Li et al., 1999). To investigate the

role HSF1 might play in the potentiation of the receptor, a pharmacological

approach was initially used as shown in Figure 2.

The drugs quercetin (Q), a flavonoid compound (Katengwa and Polla,

1991; Nagai et al., 1995), and sodium vanadate (SV), a protein phosphatase inhibitor (Sun et al., 1993; Ward et al., 1994; Chu et al., 1996), were used to negatively modulate the function of HSF1 (Li et al., 1999, 2000). In both cases, the negative modulation of HSF1 activity directly correlated with a decrease in the observed potentiation of the GR (Li et al., 1999; Li et al., 2000). The effect the drugs had on stress-free GR were more complicated; quercetin (Q) had no effect on the GR (Li et al., 1999), where as, sodium vanadate (SV) produced an increase in GR activity. However, increased GR activity by SV was negated in the presence of stress (Li et al., 2000) and suppressed HSF1 activation and the observed HSPE. To positively modulate the function of HSF1, the

5 Figure 2. Pharmacological Modulation of HSF1 Activation in GR Activity

phosphatidylinositol-3 kinase inhibitor, wortmannin (W) was used. Wortmannin had been to shown to irreversibly inactivate members of the phosphatidylinositol-

3 kinase related kinases, such as DNA-dependent protein kinase (DNA-PK)

(Sarkaria et al., 1998; Izzard et al., 1999). Research showed that the Ku protein subunit of DNA-PK, or DNA-PK itself, could interact with HSF1 (Yang et al.,

1996; Huang et al., 1997; Nueda et al., 1999). The use of wortmannin increases

HSF1 activity while having no promoter specific effect on the GR at the minimal

GRE2E1B promoter (Li et al., 2000). Wortmannin enhanced HSF1 transactivity and directly correlates with an increase in the observed HSPE (Li et al., 2000).

These pharmacological approaches demonstrate a role for HSF1 in the HSPE and suggest that HSF1 activation was a necessary component of the HSPE.

6 The model presented in Figure 2 shows a mechanism where stress, through HSF1, could produce a product that would enhance the activity of the

GR. HSF1 was shown to be negatively modulated by SV and Q and positively by wortmannin (W), and through these pharmacological means, could modulate the enhancement of the GR. This model presents two possible mechanisms by which HSF1 could potentiate the activity of the GR, although other mechanism may exist. First, the HSF1 product, “X”, could directly interact with the GR functioning as a coactivator or, secondly, “X” it could function to release a GR suppressor molecule. In both cases, the HSF1 product enhances the activity of the GR. Evidence has been shown that a direct correlation exists between HSF1 and the GR, establishing a link between the stress pathway and steroid receptors.

To further evaluate the function of HSF1 in the HSPE, a genetic approach was used in an attempt to find the functional aspects of HSF1, in the absence of stress, to determine how it might modulate the HSPE. However, to pursue this approach, a more profound understanding of HSF1 was required. HSF1 is found in the cytosol as an inactive monomer associated with a heat shock protein heterocomplex as shown in Figure 3 (Sarge et al., 1993). Upon activation, HSF1 dissociates from the heterocomplex, trimerizes and translocates to the nucleus where it binds to a heat shock element (HSE). Activation of HSF1, in part, is due to the recruitment of heat shock proteins (HSPs) away from the inactive heterocomplex of HSF1 leading to an active HSF1 homotrimer. Once the HSPs dissociates from the heterocomplex, the HSF1 monomer trimerizes and moves

7 into the nucleus where it binds chromatin. Trimerization is not a passive partner

in the activation process and can lead to activation of HSF1 in the absence of

stress (Zuo et al., 1994, 1995). A point mutation made in the trimerization

domain of HSF1 at amino acid (aa) 189, was able to produce a constitutively

active mutant of HSF1, hHSF1-E189 (Zuo et al., 1995). The point mutation

abolishes the ability of HSF1 to form an inactive heterocomplex with the HSPs.

Using this mutant of HSF1 (hHSF1-E189), we were able to pursue the genetic

approach to investigate the role of HSF1 on the potentiation of the GR under non-stress conditions. Human HSF1-E189 was placed under the control of a tetracycline-inducible vector and stably transfected into cell lines containing

HSF1 or GR responsive reporters. Immunoblot analysis demonstrates that hHSF1-E189 was doxycycline (DOX) -inducible and was also able to induce the endogenous protein, Hsp70. Evaluating the ability of hHSF1-E189 induction of the HSPE was considerably more involved. Potentiation of the GR was seen after DOX treatment for 24 h in the hHSF1-E189 cells. The hHSF1-E189- induced potentiation of the active receptor was found to be considerably lower than that observed under stress conditions. These results were consistent with the lower levels of endogenous Hsp70 produced from the mutant HSF1 when compared to the stress-activated endogenous HSF1. The combination of stress and DOX-induced hHSF1-E189 produced a super activation of HSF1 transcription, which was seen in the HSF1-responsive promoter. This super activation of HSF1 transcription carried over to greater ligand-induced

8 Figure 3. HSF1 Signaling Pathway

9 potentiation from the minimal GR promoter (GRE2E1B). The combination treatment of stress and DOX showed an increased potentiation of the receptor at saturating levels of hormone (1µM Dex), which is above the levels achieved in

the initial stress-observed HSPE. These observations demonstrated two things.

First, trimerization of HSF1 is not sufficient to fully activate HSF1 transcription

and second, nonstress-induced hHSF1-E189 is sufficient to induce a potentiation

of the GR. These demonstrated that HSF1 activation alone is sufficient to

produce a potentiation of the GR. This potentiation could be seen at all levels of

hormone treatment, including saturating levels.

The results obtained from this work provided the first evidence that stress

through HSF1 could modulate the function of a steroid receptor. Further, it

presented evidence that an HSF1-regulated gene product(s) may be responsible

for the enhancement of the steroid receptor response.

The observations that stress through HSF1 appear to produce the HSPE

and that nonstress-induced active HSF1 produced a reduced HSPE, suggest a

two-hit model. In this model, stress could function in an HSF1-dependent or in

an HSF1-independent manner to modulate the function of the GR. To address the latter model, we looked at the function of stress in the absence of HSF1 activation and investigated its ability to modulate the function of the GR.

We were interested to see if stress, independent of HSF1, could

communicate to the GR through a factor(s) released from stressed cells. To

evaluate this hypothesis, we devised experiments using conditioned media from

stressed cells to monitor the activation of both HSF1 and the GR when placed on

10 adjacent, unstressed cells. Stress-conditioned media was harvested in a time- dependent fashion, 5 min to 20 h post-stress. The results demonstrated an ability of the stress-conditioned media to produce an increase in the activation of the GR from the minimal GRE2E1B promoter. The observed increases in GR activity from the stress-conditioned media ranged from two- to five- fold for HS and 10- to 20-fold for CS compared to control-conditioned media. Differences appeared to depend on the time of media harvest and on the activity of the unstressed (control)-conditioned media. Stress- and control-conditioned media did not have any effect on either endogenous HSF1 activation or on transactivation of the GR in the adjacent identical unstressed cells in the absence of ligand.

Next, we looked at the mechanism of action in the potentiation.

Centrifugation studies were performed to demonstrate that the stress-released factor (SRF) was not part of the stress-induced or normal cellular debris. These results validated that the SRF was not part of the particulate matter. When a

10kDa exclusion filter was incorporated into the centrifugation, the SRF remained in the upper fraction greater than 10kDa. The samples from the fractions less than 10kDa had no effect on GR transactivity. The results obtained from these experiments, however, are difficult to interpret and contain a large margin of error. The error in these experiments is associated with the probability of obstruction or clogging of the size exclusion filter system by the presence of serum in the cell culture media. To determine if the SRF was a protein, we heat- denatured the media at 100˚C and digested the media soluble proteins with the

11 nonspecific protein-digesting enzyme, proteinase K. Experimentally, we find that

SRF was a protein that could be inactivated by a denaturing process. To determine if serum played a role in the stress conditioning of the media or in the production of the SRF, we performed experiments in stress-conditioned media supplemented with bovine calf serum (BCS) in a concentration-dependent manner. Concentrations ranged from 0% to 10% BCS with the initial experiments being performed in media supplemented with 10% BCS. The results confirmed that the serum-free, stress-conditioned media did not contain the ability to increase the induction of ligand-dependent GR transactivity.

This work suggested a model where both stress and the constitutively- active HSF1 were able to produce a HSPE. The effect of stress on GR transactivity exhibited a greater response than that achieved by either hHSF1-

E189 induction or stress-conditioned media alone. The increase in the GR activation observed with the stress-conditioned media was attributed to the release of a stress-dependent factor exerting an effect on GR through an as yet unknown mechanism.

12 LITERATURE

Mechanism of Glucocorticoid Action

Glucocorticoids play an important role in cellular development, cell

proliferation, differentiation, and numerous other metabolic events like

gluconeogenesis, stress tolerance, and cell conditioning (Reichardt and Schutz,

1998; Adcock, 2000).

Glucocorticoids are under the control of the hypothalamic-pituitary-adrenal

axis (HPA) and are eventually released from the adrenal cortex following

stimulation (Reichardt and Schutz, 1998; Adcock, 2000). The multi-step process

of GR agonist release begins in the hypothalamus. Corticotropin releasing

hormone (CRH) is released from the hypothalamus and interacts with a G-protein

linked receptor where it stimulates the release of β-endorphins and

adrenocorticotropic hormone, ACTH, from the pituitary. ACTH, in turn, induces

the release of the adrenal steroids; glucocorticoids, mineralocorticoids, and weak

androgens from the adrenal cortex. Once in circulation, the steroids travel to

target organs and produce their receptor-mediated effects in a tissue-specific

manner (Reichardt and Schutz, 1998). The receptor of interest for this research is the glucocorticoid receptor (GR). The endogenous GR ligand in humans is cortisol and is generated endogenously from specific enzymatic reactions and cleavages of the ubiquitous steroid precursor, cholesterol.

13 Glucocorticoid Receptor

The GR belongs to the nuclear receptor superfamily that includes the estrogen receptor (ER), androgen receptor (AR), and the thyroid hormone receptor (THR) (Beato et al., 1995). Transcriptional activation of the GR, like other members of this family, is a ligand-dependent process (Pratt and Toft,

1997). Activation of the GR begins in the cytosol where the GR is found as an inactive monomer in a multimeric hetero-complex containing heat shock proteins

(HSPs), p23, and an immunophilin component (IP). The IP is composed of one of four known associated immunophilins, CYP40, FKBP51, FKBP52, and PP5

(Figure 4) (Pratt and Toft, 1997; Davies et al., 2002).

The nomenclature of the associated immunophilins is derived from their ability to bind immunosuppressive drugs like cyclosporine A (CYP), or FK506 (FK

Binding Protein). Our laboratory and others have shown that the association of an immunophilin with the GR heterocomplex plays an important role in the ligand-induced activation and transformation of the GR to a DNA-bound, transcriptionally-active state (Bresnick et al., 1988; Pratt and Toft, 1997; Davies et al., 2002).

Immunophilins interact with the GR through binding to the Hsp90 dimer at the tricopeptide repeat (TPR) binding domain (TBD) (Fig. 4). Interactions between Hsp90 and the immunophilin component occur at the TPR region located in each immunophilin (IP) (Ramsey et al., 2000; Cheung-Flynn et al.,

2003). Interactions between the Hsp90 dimer and the steroid receptor confir a

14

FIGURE 4. Glucocorticoid Receptor Heterocomplex

high-affinity binding state on the receptor and allow for ligand-induced activation

(Bresnick et al., 1988; Hollenberg and Evans, 1988). Once the agonist binds to the receptor, the binding induces a “switching” of the immunophilin component

followed by translocation to the nucleus and the subsequent release of Hsp90

(Davies et al., 2002). Once the Hsp90 dimer is released from the heterocomplex,

the GR homo-dimerizes and binds to the GR responsive element (GRE) as seen

15 in Figure 5 (Luisi et al., 1991). The GR-mediated gene products are then produced in a tissue- and promoter-specific fashion (Nordeen et al., 1998;

Lambert and Nordeen, 2003). The promoter- and tissue-specific differences observed under these conditions are thought to be due to such factors as macro environment, coactivator recruitment and histone acetylation (Nordeen et al.,

1998; Lambert and Nordeen, 2003).

Domain Structure of the GR

The major functional aspects of the GR are associated with the three known functional domains of the receptor (Fig. 4), the N-terminal activation domain, DNA-binding domain, and the ligand-binding domain. These domains compose a modular structure that can be interchanged between related nuclear receptors without the loss of functionality (Carson-Jurica et al., 1990). The human GR is 777 aa protein incorporating an N-terminal activation domain, DNA- binding domain (DBD), and a C-terminal ligand-binding domain (LBD). The functional aspects of the GR can be attributed to two unique isoforms of the receptor, alpha and beta. The alpha form of the receptor (hGRα) is the transcriptional form that is generally referred to in this research. The beta isoform (hGRβ) has recently been studied and found to generally function as a dominant negative form of the receptor (Bamberger et al., 1995). A more comprehensive understanding of this concept can be found in the work performed by John Cidlowski and colleagues.

16 FIGURE 5. Glucocorticoid Receptor Activation Pathway

N-Terminal Activation Domain

The N-terminal of the human GR (hGR) consists of aa 1-420 (Yudt and

Cidlowski, 2001). This region is known to contain an alternative translation initiation site generating two isoforms, A and B, of the hGRα protein (Yudt and

Cidlowski, 2001). The A isoform is full length GRα and the B isoform is truncated at the first 27 aa. The hGRβ receptor also has been suggested to contain similar

17 A and B isoforms with a currently unknown function in the cell (Yudt and

Cidlowski, 2001).

The N-terminal portion of the hGR also is required for full activation and

contains the hormone-independent transactivation domain, activation function 1

(AF1). This domain of the receptor has been narrowed to a region between aa

77-262 in the human GR (Hollenberg and Evans, 1988; Yudt and Cidlowski,

2001) and contains most of the known phosphorylation sites mapped within the

receptor. Seven to eight known phosphorylation sites exist in the mouse GR,

depending on the publication, with four of these sites conserved in the human

(Krstic et al., 1997; Bodwell et al., 1998; Pocuca et al., 1998). Most of the phosphorylation sites currently mapped in the human GR are known to be serines, with one additional threonine site found in the mouse (Mason and

Housley, 1993; Webster et al., 1997; Bodwell et al., 1998).

DNA Binding Domain

The DNA-binding domain (DBD) of the steroid receptor is the most

conserved domain within the nuclear receptor family. It encompasses the region

from aa 421-488 (Giguere et al., 1986; Schaaf and Cidlowski, 2002). Chromatin

binding in the nuclear receptor family occurs through the two zinc finger motifs

located in the DNA-binding domain (Luisi et al., 1991). The zinc finger motifs are

required for high-affinity DNA-binding and contain nine cysteines sites that are

conserved across the nuclear receptor family. Four cysteines in each finger are

considered invariable and coordinate tetrahedrally one zinc ion that functions to

18 maintain and stabilize the DNA binding pocket (Luisi et al., 1991). The “P Box,”

located at the base of the first zinc finger, is important in recognition and

discrimination of core DNA motifs specific to each steroid receptor (Mader et al.,

1989). The “D Box,” located at the base of the second zinc finger is an important

region involved in receptor dimerization (Alroy and Freedman, 1992; Freedman

and Luisi, 1993)

Hinge Region

This region is generally located between the DBD and the ligand-binding

domain (LBD) in all nuclear receptors and is not well conserved among different

members of the nuclear receptors family (Aranda and Pascual, 2001). The main

function of the hinge region appears to involve the rotation of the DBD to

accommodate the chromatin into the zinc finger region. The hinge region also

harbors a nuclear localization sequence, NLS1, which is inhibited by the LBD. In the presence of ligand, the NLS is exposed and allows for and contributes to the nuclear localization of the receptor (Picard and Yamamoto, 1987; Baumann et al., 1993; Aranda and Pascual, 2001).

Ligand Binding Domain

The ligand-binding domain (LBD) of the receptor is a multi-functional

domain incorporating ligand binding, receptor dimerization, and regions that

interact with HSPs and co-activators (Moras and Gronemeyer, 1998; Adcock,

2000). The LBD is required for the characteristic ligand-induced activation

19 associated with the nuclear receptor family. The deletion of this domain results in a constitutively active receptor in the mouse (Godowski et al., 1987), however, partial deletions of this region result in reduced ligand binding ability (Hollenberg and Evans, 1988). Tertiary structure of the LBD is generated from twelve α-

helices located within this region. These12 α helices are important for the ability

of the receptor to bind ligand and to recruit co-activators (Darimont et al., 1998).

They fold to generate a hydrophobic pocket within which the ligand can reside.

The 12th helix also performs three additional functions: 1) it serves as a cover over the ligand-binding pocket, securing the ligand to the receptor (Darimont et al., 1998; Aranda and Pascual, 2001), 2) it is the site where most co-activators bind in steroid receptors (Darimont et al., 1998), and 3) it is the core site of the activation function 2 (AF2) (Darimont et al., 1998). The AF2 domain is located between aa 526-556 (Onate et al., 1998) and is crucial for complete activation of the receptor, in a ligand-dependent manner. The AF2 domain function is inverse to the ligand-independent activation associated with the AF1 domain located in the N-terminal of the receptor.

Ligand binding in steroid receptors is a selective process and plays an important role in the activation and downstream production of receptor-specific gene products. Recent crystal structure resolutions of the LBD in the GR have revealed an additional side pocket that is formed by helices six and seven

(Bledsoe et al., 2002). The function of this side pocket may explain the selective binding of glucocorticoids and some mineralocorticoids as opposed to other steroids within the pocket (Bledsoe et al., 2002). 20 Glucocorticoid Response Elements

Upon activation, the GR translocates to the nucleus where it binds to a GR

response element (GRE). The GRE is a 15-mer palindrome sequence of 5’-

GGTACAnnnTGTTCT-3’ (Beato, 1989), separated by three base pairs between

each element (Green et al., 1988). Specific binding of the GR to the GRE occurs

through the P-box located in the first zinc finger of the DBD (Luisi et al., 1991).

Binding of the GR to the response elements occurs one monomer at a time.

Monomer on of the GR binds to the first half-site and enhances binding of the

second monomer (Dahlman-Wright et al., 1990; Dahlman-Wright et al., 1991),

generating a DNA-bound homodimer.

Regulation of the GR gene products occurs through modulation of the

DNA-bound receptors. GR has the ability to suppress specific genes by binding

to negative GREs (nGREs). Negative regulation occurs through the

displacement of transcription factors that would induce the intended gene transcription (Drouin et al., 1987), in a ligand-dependent manner (Drouin et al.,

1989). Negative GRE sequences are not conserved within or across species

and a more variable sequence is defined as ATYACnnTnTGATCn (Truss and

Beato, 1993).

21 MODULATION OF GR ACTIVITY

Receptor Phosphorylation

The GR, like other steroid receptors, is basally phosphorylated and

undergoes post-translational modifications as a means of control of activity.

Hyperphosphorylation occurs in response to hormone-induced activation (Orti et

al., 1989) and depending on the publication, seven to eight phosphorylation sites

have been identified within the mouse GR (mGR). Four of these sites

correspond to the rat, human, and yeast (Hollenberg and Evans, 1988; Bodwell

et al., 1991, 1998; Krstic et al., 1997; Pocuca et al., 1998), and are clustered in the N-terminal AF1 region of the receptor (Bodwell et al., 1991).

All known phosphorylation sites located in the GR appear to be either serine or threonine. The serine phosphorylation sites are located in proline- directed and conserved sequences; with the only exception found in the mGR, the threonine site (Mason and Housley, 1993; Bodwell et al., 1998). The serine sites in the GR are phosphorylated by cell cycle-associated kinases like the cyclin-dependent and mitogen-activated protein (MAP) kinases (Bodwell et al.,

1998). An example of this type of phosphorylation can be seen in the mGR where serine 122 is phosphorylated by the kinase, casein kinase II (Bodwell et al., 1998). Garabedian et al. also supported this observation by showing that serine 203 and 211 in human GR (hGR), which correspond to serine 224 and serine 232 in mGR, have a higher incidence of hormone-induced phosphorylation in a cell cycle-dependent fashion (Wang et al., 2002). These results demonstrate

22 the necessity of cyclin-dependent kinases in the phosphorylation-induced post-

translational modifications of GR-mediated gene expression (Wang et al., 2002).

Control of GR-mediated gene expression occurs in a promoter- and

phosphorylation-specific manner regardless of cell type (Webster et al., 1997;

Bodwell et al., 1998; Wang et al., 2002). The phosphorylation state of the GR is

important in receptor activation and the ability to bind hormone. Glucocorticoid

receptor found in G2, M and G1 phases of cell cycle exhibit a high basal

phosphorylation state and an associated decrease in the ability to bind hormone.

High basal phosphorylation generates a high negative charge in the AF1 domain

(Danielsen et al., 1987), which suppresses the hormone-induced

hyperphosphorylation of the receptor and defines the ligand or hormone

resistance phase (Danielsen et al., 1987). When the cell enters S phase or the

ligand sensitivity phase, the basal phosphorylation of the receptor is decreased

and allows for increased ligand-induced hyperphosphorylation. This increase in

ligand-induced hyperphosphorylation increases GR-specific DNA binding

(Bodwell et al., 1998).

Phosphorylation also plays a role in the down-regulation of the GR

response at the level of transcription (Burnstein et al., 1994). Phosphorylation

decreases the GR response through changes in the mRNA levels of the receptor.

Dexamethasone (Dex) treatment in mouse cells reduces the GR mRNA levels by

more than 50% and directly corresponds to an increase in the phosphorylation of

the receptor (Bodwell et al., 1998). This was also seen in phosphorylation

mutants of the GR, where decreased phosphorylation increased GR protein

23 levels (Bodwell et al., 1998). These observations demonstrate a correlation

between the receptor phosphorylation and stabilization of the protein (Bodwell et al., 1998).

Destabilization in GR protein levels can occur through changes in the receptor half-life. Half-life of the GR decreases as the number of phosphorylation sites increases in the receptor (Bodwell et al., 1998). Research by Munck et al. supports this observation by demonstrating that ligand-induced hyperphosphorylation of the GR is a function of accelerated phosphorylation over time (Orti et al., 1993). This suggests that the longer GR is activated the more likely it was to be hyperphosphorylated and shunted to destabilization.

General Transcription Factors, Co-Activators and Co-Repressors

Nuclear receptors are sequence-specific transcription factors that interact with RNA polymerase II and the transcription initiation complex transcribing receptor-dependent genes (Aranda and Pascual, 2001). Once a receptor is activated, it binds to chromatin and initiates transcription through the recruitment of RNA polymerase II to the TATA box found in the promoter region of all steroid receptors. The recruitment of RNA polymerase II to this region also initiates the association of other factors to generate the transcription initiation complex. The factors that are usually associated with the complex include: TFIID, TFIIB, TFIIA,

TFIIF, TFIIE, TFIIH. The formation and complete function of the transcription initiation complex from RNA polymerase to the association of transcription factors can be reviewed in the work by Wilson et al. (1996).

24 Steroid receptors, like the GR, can interact with the transcription initiation

complex through direct or indirect approaches, influencing the rate at which

receptor-specific genes are transcribed (Collingwood et al., 1999). Indirect

methods of interaction generally occur through the recruitment of cofactors to

assist in the release of chromatin-induced suppression, through chromatin

remodeling (Collingwood et al., 1999). Examples of chromatin remodeling can

be seen with complexes like BRG1 and BRM in humans, or the Swi/Snf complex

found in yeast (Wang et al., 1996a, b; Deroo and Archer, 2001). The remodeling

activity of the cofactors is achieved through the process of histone acetylation

and deacetylation that induces changes to the nucleosome structure. The

cofactor-induced changes in the nucleosome allows the receptor/cofactor

complex to recruit the basal transcription machinery to the pre-initiation complex

(Kurihara et al., 2002). The complete complex then initiates the ligand-

dependent receptor-specific transcription of genes.

The ligand-dependent activation of the receptor can also affect the

function of the AF2 domain (Webster et al., 1988). Once the ligand is bound, the

receptor dissociates from its inactive heterocomplex and promotes the protruding

twelfth helix to fold back over the hydrophobic pocket and secure the ligand. The orientation of helix twelve, with respect to the ligand-binding pocket, is crucial to the ability of the ligand-bound receptor to interact with cofactors and the general transcription machinery (Tsai and O'Malley, 1994; McKenna et al., 1999).

The cofactor/GR interaction occurs through the LXXLL motif located in

each cofactor. Each motif contains three highly conserved LXXLL regions

25 composed of leucine (L) and any other amino acid (X). These motifs are

required for receptor interaction and are necessary and sufficient to support

interactions with the ligand-bound receptor (Heery et al., 1997; Ding et al., 1998).

The AF1 region, located in the N-terminus of the receptor, can also

interact with cofactors. However, this AF1 cofactor interaction is poorly

understood, but appears to involve the cooperation of the AF2 domain (Onate et

al., 1998). Binding of cofactors to the AF1 region of the receptor does not involve

LXXLL motifs but glutamine-rich sequences (Bevan et al., 1999). In any case, the result is an interaction of cofactor/GR through the cooperation of both AF1

and AF2 regions inducing transcription. Examples of this type of interaction can

be seen in the binding of steroid receptor coactivator 1 (SRC1) to both the AF1

and AF2 domains in the estrogen and other nuclear receptors (McInerney et al.,

1996).

COACTIVATORS

Coactivators are a diverse family of proteins that contain the ability to

induce nucleosome remodeling. The first coactivator to be discovered was the

steroid receptor coactivator 1 (SRC1), recovered from cloning experiments

performed in B-lymphocytes (Onate et al., 1995). This cofactor was found to be

a member of a much larger family of coactivators referred to as the P160 family

of proteins. The P160 family was divided into three different groups, identified

from alternative splice variants in different species (Aranda and Pascual, 2001).

26 SRC-1 was the first P160 coactivator found to interact with the nuclear receptor family, through the N-terminal, AF2 domain in a ligand-dependent manner (Onate et al., 1995). Other members of this family were later identified based on their molecular size, including SRC-2 (Voegel et al., 1996), and SRC-3

(Torchia et al., 1997). The SRC coactivators are all different in size but show a

40% sequence homology. The main function of this family of proteins is histone acetylation through histone acetyltransferace activity (McKenna et al., 1998; Xu et al., 1998; Gehin et al., 2002). Histone acetyltransferace (HAT) activity has been mapped to the C-terminal region of these proteins (Chen et al., 1997;

Spencer et al., 1997).

Although the SRC cofactors show homology in their sequence and action, they have functional differences in the effects they produce on the ligand- activated receptor. Research suggests that this difference occurs as a result of ligand-induced preferential recruitment of specific coactivators to the active receptor (Bramlett and Burris, 2002).

ACETYLATION and DEACETYLATION

The functional aspect of cofactors can be attributed to their ability to acetylate and deacetylate histones. To understand the properties of acetylation, a brief synopsis of the eukaryotic chromosome is necessary. Eukaryotic chromosomes are organized into regularly spaced protein-wrapped DNA units termed nucleosomes. The basic subunit of the nucleosome is the histone, a small globular protein (Tse et al., 1998). Nucleosome are arranged into

27 octomers formed from an 8-histone heterocomplex. Each heterocomplex is made up of two copies of each of the histones; H2A, H2B, H3, and H4. The octomers are connected together by the linker histone, H1. The complete complex contains the histone octomer and H1 wrapped 1.7 turns forming the

DNA super helix of compact chromatin. This spatial arrangement creates a transcription initiation barrier against unwanted transcription and prevents access of transcription factors to their DNA substrate (Ura et al., 1997). Transcription factors gain access to the DNA through histone acetylation (HAT) (Tse et al.,

1998). In each case, the permissive binding of the coactivator to the ligand- bound receptor leads to acetylation of the histones. Once acetylated, derepression of the chromatin occurs through unwinding the chromatin structure and facilitates transcription initiation (Lee et al., 1993; Ura et al., 1997).

Histone acetyltransferase activity (HAT) is the process whereby the histone tails associated with each of the histones are acetylated. The acetylation process decreases the interaction between the histones and the negative charged DNA (Tse et al., 1998). This “uncoupling” of their interactions, along with disruption of the higher order chromatin structure (Rhodes, 1997; Tse et al.,

1998), grants access of the activated receptor to the chromatin. The inverse of this process is histone deacetylation (HDAC). HDAC activity involves the deacetylation of the histones and the reassociation of the histones and the DNA.

This reassociation sequesters and tightly winds the DNA creating compact chromatin and suppression of receptor-mediated transcription.

28 COREPRESSORS

While coactivators function to increase the activity of nuclear receptor in the presence of ligand, corepressors induce transcriptional repression, or silencing of the receptor. Repression is an antagonist-specific or ligand-free process (Smith et al., 1997), whereby competition for chromatin binding occurs through a direct interaction between the repressor and receptor. The interaction suppresses or hinders receptor dimerization quenching transcription (Hudson et al., 1990). The interactions between the corepressor and transcription initiation complex induce changes in the histone associated tertiary structure of the DNA and obscure the chromatin. This process, transrepression (Baniahmad et al.,

1992, 1995), occurs through the deacetylation of the core histones leading to nucleosome condensation. As the nucleosome condenses and tightens the DNA super helix, receptor access is limited (Laherty et al., 1997). Corepressors lack deacetylation activity and achieve this function through the recruitment of factors such as Sin3 and histone deacetylase (HDAC) (Heinzel et al., 1997; Nagy et al.,

1997).

The first corepressors discovered were the silencing mediator for retinoid and thyroid hormone receptor, SMART, and nuclear receptor corepressor, N-

CoR (Chen and Evans, 1995; Horlein et al., 1995). These two corepressors have been shown to function in the repression of the thyroid receptor (TR) and the retinoic acid receptor (RAR) through an indirect process of corepressor- mediated receptor-specific deacetylation (Seol et al., 1996). Current literature suggests that N-CoR, and possibly the sequence similar partner SMART (Chen

29 and Evans, 1995), have little or no interaction with the GR (Seol et al., 1996).

While other researchers provide evidence for such an interaction (Stevens et al.,

2003), the exact mechanism of corepressor-mediated suppression is still to be resolved.

Repression by Protein Antagonism

One of the most researched methods of GR suppression is through the protein antagonist nuclear factor kappa B (NF-kB). Antagonism between the GR and NF-kB is unique, in that, the functions of the two proteins are diametrically opposed to each other. The GR functions as a suppressor of the cells immune and inflammatory pathways through the inhibition of the same cytokines and cytokine–induced genes that are activated by the NF-kB. This interaction and

NF-kB function can be reviewed in work by McKay and Cidlowski (McKay and

Cidlowski, 1999).

The GR and NF-kB antagonize one another at the level of the promoter, through a direct physical interaction (Ray and Prefontaine, 1994; Caldenhoven et al., 1995; Scheinman et al., 1995; McKay and Cidlowski, 1998). The antagonism occurs in a ligand-specific manner and is associated with the activation of steroid receptors. The NF-kB is a heteromeric protein composed of p65 and p50 subunits (Baeuerle and Baltimore, 1989). The inactive form of the protein, found in the cytosol, is bound to IkB and functions as the protein inhibitor of NF-kB activation (Baeuerle and Baltimore, 1988). Cells stimulated by stress induce the phosphorylation of IkB (Brown et al., 1995) targeting the protein for degradation

30 (Krappmann et al., 1996). As IkB degrades, IkB is released of IkB released from

the NF-kB/IkB inactive complex and activates NF-kB (Palombella et al., 1994).

Active NF-kB interacts with the DBD of the GR and suppresses transcription

(McKay and Cidlowski, 1998). The interaction of NF-kB and the GR occurs through a member of the Rel family of transcription factors (Mercurio et al.,

1992), at the Rel homology domain of the p65 subunit of NF-kB (Ray and

Prefontaine, 1994; Caldenhoven et al., 1995; Scheinman et al., 1995; McKay and

Cidlowski, 1998).

HEAT SHOCK FACTOR 1 (HSF1)

Significance

The first observed example of stress-activated genes was seen in the

chromosomal puffing that appeared in Drosophila following sublethal doses of increased body temperature (Ritossa, 1962). This observation was unique because high temperature stress had caused the inactivation of normally active genes. The proteins expressed under the stress conditions were eventually termed heat shock proteins (HSPs). Subsequently, HSPs have been found to function in the stress response of cells in many tissues and species. Induction of

HSPs is not limited to heat shock, but can be induced by a variety of stressors, including environmental and physiological conditions as illustrated in Figure 6.

Non-stress conditions, like cell growth and differentiation, also have been shown to induce the production of heat shock proteins. Regardless of the mechanism of stimulation, the major shared characteristic between these conditions is the

31 generation of unfolded, denatured, or aggregated proteins. The accumulation of unfolded or abnormally folded proteins is the initiating step in the induction of

HSPs and the cell stress response (Hightower, 1980; Anathan et al., 1986).

In cells, HSPs are under the control of a family of transcription factors which consists of four members, HSF1-4 (Wu, 1995), three of which are found in human, mouse, and chicken. HSF3, the exception, has only been found in

FIGURE 6. Environmental Regulation of Heat Shock Factor 1 (HSF1)

32 chickens, where it is involved in cellular development (Rabindran et al., 1991;

Sarge et al., 1991; Schuetz et al., 1991; Nakai and Morimoto, 1993; Nakai et al.,

1997). The HSF family of proteins, although diverse, shows 40% conservation of

sequence homology within a species. Most of this conservation is confined to the DNA-binding (Harrison et al., 1994) and oligomerization (Peteranderl and

Nelson, 1992) domains within the family.

HSFs 1 and 3 are the stress responsive family members in humans and

chickens, respectfully (Nakai and Morimoto, 1993; Nakai et al., 1995). Their

activation is required to achieve the maximal stress response (Tanabe et al.,

1998). The stress-induced activation of the HSF1 has been mapped to the C-

terminal end of the protein (Green et al., 1995). Stress-activated transcription

factor HSF2, appears to be the factor that is activated during the developmental

and differentiation stages associated with cell cycle progression (Schuetz et al.,

1991). Not only do HSF1 and 2 show a difference in the stage of activation, but

also show a relevant difference in the time frame and duration of action. HSF1

undergoes a rapid activation following a stress event, while HSF2 is activated

over a period of time, from sixteen to twenty four hours, and remains active for up to seventy-two hours (Sistonen et al., 1994; Fiorenza et al., 1995; Tanabe et al.,

1997). The fourth family member, HSF4, is also unique in that it lacks a transactivation domain and appears to function as a repressor of the heat shock response and stress-induced gene expression (Nakai et al., 1997).

33 HSF1 Structure

The HSFs are organized into multi-domain structures that incorporate four functional aspects. The structure of HSF1 is representative of most of the members of this family. A model of the functional domains can be seen in Figure

7. HSF1 is divided into a DNA binding domain, trimerization domain, regulatory domain, and a C terminal activation domain. These domains were characterized through the use of GAL4 DNA-binding domains fused to various segments of mammalian HSF1 (Green et al., 1995; Shi et al., 1995; Zuo et al., 1995). Fusion proteins help to determine the functional and structural aspects of each of the domains found in the family. The DNA-binding domain of HSF1 is composed of a classical helix-turn-helix motif, whose structure is found in members of the helix-turn-helix class of proteins (Harrison et al., 1994). The third helix, located in this region, is the most important for the protein to bind DNA. Ten of the fifteen

amino acids comprising the third helix were found to be invariant, and are the

most conserved sequences in the structure of HSF1 (Hubl et al., 1994; Kim et al.,

1994). The trimerization domain of HSF1 is composed of a classical helical coiled-coil structure, which is essential to trimerization and conversion into the active protein and is involved in the high affinity binding to chromatin in higher mammals (Peteranderl and Nelson, 1992). The C terminal region of HSF1 contains the activation domain of which deletions and mutation within this domain have resulted in constitutive trimerization and activity in the absence of stress

(Rabindran et al., 1993).

34

Figure 7. Model of HSF1 Domain Structure

The ability of mutations in the C terminal region to activate HSF1, suggest that this region is involved in HSF1 suppression. Suppression of HSF1 occurs through intramolecular coiled-coil interactions between the C terminal and the amino trimerization region of the protein (Zuo et al., 1994). Stress-induced activation of HSF1 involves the stress-dependent change of the monomeric protein from an intramolecular to a trimeric intermolecular interaction between multiple HSF1 monomers (Zuo et al., 1994). The activation process itself is associated with the collapse of the intramolecular coiled-coil structure of inactive

HSF1 and the formation of an active HSF1 trimer.

35 HSF1 Response Element

Upon activation HSF1 translocates to the nucleus and binds to the HSF1

responsive element, heat shock element (HSE), (Pelham, 1982). This HSE is

found upstream of stress-responsive genes and is composed of multiple

elements containing pentameric sequences of 5’-nGAAn-3’ (Fernandes et al.,

1994; Kroeger and Morimoto, 1994; Fernandes et al., 1995). These sequences are arranged in groups of four to five repeats for each binding site (Kroeger and

Morimoto, 1994).

Heat Shock Protein Family

The heat shock protein family is produced through the activation of HSFs

and is composed of highly conserved stress-inducible proteins and a

constitutively synthesized set of proteins. The constitutively synthesized HSPs

function under normal growth conditions and are expressed at a basal level. The

stress-induced forms of the HSPs are almost completely absent under non-stress

conditions and significantly induced following stress (Becker and Craig, 1994).

All HSPs are classified based on their molecular size, regardless of the

stress or mechanism of induction. One of the major stress-induced proteins is

the 70kDa protein, Hsp70 (Moran et al., 1983). The multigene family of Hsp70 is

composed of a constitutively expressed, Hsc70, and an inducible Hsp72 form of

the protein. The stress-induced Hsp72 protein is the one that is commonly

referred to as Hsp70 (Voellmy et al., 1985; Wu and Morimoto, 1985; Mues et al.,

1986; Sorger and Pelham, 1987; Kiang and Tsokos, 1998). Sequence homology

36 within the Hsp70 family varies between 60-78% among eukaryotic organisms

(Kiang and Tsokos, 1998). Hsp70 is functionally important to the stress response and is the major component responsible for down regulation of the active HSF1 trimer. The role Hsp70 plays in the down regulation of HSF1 will be discussed further in the following section.

One of the most abundant HSPs in the cell is Hsp90, which has a variety of functions including, the chaperoning of most, if not all, of the steroid receptors and the binding to microtubules and actin microfilaments (Koyasu et al., 1986).

Hsp90 is an essential backbone component of the GR heterocomplex (Pratt,

1993); lack of Hsp90 in the GR heterocomplex leaves the receptor unable to bind ligand and in an inactive state.

Regulation of HSF1

The stress response is a major factor in the ability of a cell to survive a stressful event. Stress is characterized by the synthesis and accumulation of unfolded or mis-folded proteins (Kelley and Schlesinger, 1978; Hightower, 1980;

Goff and Goldberg, 1985; Anathan et al., 1986). The presence of unfolded or mis-folded proteins recruits endogenous HSPs, like Hsp70, away from their chaperoning functions to contribute to re-folding and chaperoning of denatured proteins. The recruitment of HSPs away from the HSF1 heterocomplex induces the activation of HSF1. Once active, HSF1 increases the expression of the inducible HSPs and helps to regulate the cells response to stress (Rabindran et al., 1991).

37 HSPs also function in the regulation of HSF1 activity. Controversy

however, does exist over the exact role that HSPs may play in regulation.

Currently, two camps exist to explain HSF1 regulation. The first camp suggests

that HSF1 is regulated by the abundance of free Hsp70. As the presence of

unfolded or misfolded proteins increases in stressed cells, the equilibrium of

Hsp70 shifts away from the HSF1 heterocomplex and toward the denatured

proteins. This shift releases Hsp70 from HSF1 and allows it to function in the

mediation of the stress. At the same time, unchaperoned HSF1 trimerizes to an

active protein and begins to increase HSP expression. This model is suggested

in the work by Richard Morimoto et al. (Abravaya et al., 1992; Morimoto et al.,

1992). His hypothesis, however, has been refuted in the work of C. Wu et al.,

which suggests that the suppression of HSF1 does not occur through an Hsp70-

dependent regulatory mechanism (Rabindran et al., 1994). Recent work

presents evidence that Hsp90 may interact with HSF1 and potentially function as

the HSF1-inactivating protein (Bharadwaj et al., 1999). Both cases provide strong evidence for their arguments, with no clear picture emerging to completely explain the regulatory mechanism of HSF1.

We have elected to follow a model presented by Moroimoto et al., as presented in Figure 3. In this model, the activation of HSF1 begins in the cytosol where the 75kDa monomer protein of HSF1 is constitutively synthesized and associates with a heterocomplex of heat shock proteins, like Hsp90 and Hsp70

(Baler et al., 1993; Zou et al., 1998; Guo et al., 2001). When the cell experiences stress, the repressive effects of the heterocomplex are released by the

38 recruitment of the associated HSPs away from HSF1. The release of the HSPs

then function to alleviate the accumulation of denatured proteins. Upon the

release of the repressive effects, HSF1 obtains a trimerized state and is

translocated to the nucleus where it binds to its response element, the heat

shock element (HSE). Once bound, HSF1 induces the production of HSPs to

reduce and remove stress. This, in turn, inactivates HSF1 in an autoregulatory

fashion (DiDomenico et al., 1982; Craig and Gross, 1991). Hsp70 performs a

pivotal role in this process by releasing HSF1 from the chromatin and sequestering it back to an inactive monomeric complex (Mosser et al., 1988,

1990; Abravaya et al., 1992; Baler et al., 1996; Shi et al., 1998).

The regulation of HSF1 and the induction of heat shock proteins are

important in the cells ability to deal with and survive stressful events. However,

HSF1 is not required for a cell’s survival in the absence of stress. Work in HSF1-

null mice suggests that they develop and survive normally. Research in HSF1-

deficient fibroblast cells found them unable to produce HSPs, yet were viable

under normal conditions (McMillan et al., 1998; Xiao et al., 1999). Further, heat

shock proteins are not the only mechanism involved in the regulation of HSF1.

The stress activation of HSF1 is considerably more complex than this model

suggests. For example, post-translational modifications, like phosphorylation,

are required for complete activation (Guo et al., 2001).

Phosphorylation in HSF1

39 The initial observation that phosphorylation functions in the activation of

HSF1 came from electromobility shift assays done in the presence of

phosphatases (Sorger et al., 1987; Larson et al., 1988; Sorger and Pelham,

1988; Sarge et al., 1993). Results show HSF1 is a basally- or constitutively-

phosphorylated protein that under stress conditions is inducible phosphorylated

(Cotto et al., 1996; Kline and Morimoto, 1997). The view is, the inactive form of

HSF1 is constitutively-phosphorylated and, once activated, becomes

hyperphosphorylated (Baler et al., 1993; Sarge et al., 1993; Cotto et al., 1996).

Phosphorylation work has revealed negative regulatory sites within HSF1

at Serine -303, -307, and -363 (Chu et al., 1996, 1998; Knauf et al., 1996; Kline

and Morimoto, 1997). These constitutively-phosphorylated serines negatively

modulate the function of HSF1 through multiple kinases (Kline and Morimoto,

1997; Xia et al., 1998). The extracellular signal-regulated kinase (ERK) members of the mitogen-activated protein kinase (MAPK) family phosphorylate

HSF1 at serine 303. Phosphorylation by ERK subsequently primes the system

for a second phosphorylation at serine 307 by glycogen synthase kinase 3

(GSK3) and the negative modulation of HSF1 function (Chu et al., 1996, 1998).

A negative regulatory site was also found at serine 363, which can be

phosphorylated by both the MAPK Jun-N-terminal kinase (JNK) and protein kinase C (Chu et al., 1998; Dai et al., 2000).

Currently, only two positive regulatory sites have been located within

HSF1. One located at serine 230 (Holmberg et al., 2001) and the other at

threonine 142 (Soncin et al., 2003). Serine 230 is a site of basal phosphorylation

40 and is enhanced under stress conditions (Holmberg et al., 2001). This site is phosphorylated by the calcium calmodulin-dependent protein kinase II (CaMKII)

(Holmberg et al., 2001). The recently discovered threonine 142 site is phosphorylated under stress conditions by the protein kinase CK2 (Gerber et al.,

2000; Soncin et al., 2003). An understanding of phosphorylation in the modulation of HSF1 is still very elusive. The general consensus is that hyperphosphorylation leads to an increase in HSF1 transactivity over that seen in the basal state of phosphorylation (Xia and Voellmy, 1997).

Potential Role of Glucocorticoid and Heat Shock Protein Cross Talk in

Stress Physiology

Cells have developed many ways to deal with the diverse conditions that are encountered throughout a normal life span. These conditions range from normal cellular growth conditions to pathophysiological states like fever and inflammation, and may include environmental stresses such as heavy metals and heat shock (Morimoto et al., 1992). In all cases, environment is important in the differential induction of the various stress-responsive genes, including heat shock proteins. Stress-response genes allow the cell to endure the stress and survive

(Morimoto et al., 1992). Induction of the stress-response genes alone, however, does not ensure a cell’s survival. The ability to survive lies in the capacity of a cell to regulate and control the stress-response itself. If a cell loses control of the stress response system, a case of chronic inflammation may arise and generate conditions like those observed in rheumatoid arthritis and asthma. Cellular

41 responses to stress need to be modulated in order to prevent damaging

situations. The regulation of the stress response can, in part, be controlled by

the actions of the glucocorticoid receptor (GR). In mammals, the GR is activated following stress through the release of cortisol from the HPA axis. Cortisol serves as the endogenous ligand activating GR-regulated gene transcription, in both a direct and indirect manner. Once activated, the GR induces multiple DNA and non-DNA binding-dependent functions that suppress stress-activated genes like cytokines and chemokines (Karin, 1998; Reichardt and Schutz, 1998;

Reichardt et al., 1998, 2001).

The physiological function of the GR, in mammals, is the mediation of the

stress response. This observation evolved from four distinct mechanisms of GR

action in the work by Munck et al.. These four mechanisms, as described below,

work together to regulate the cell’s normal stress response and prevent its over-

activation. Regulation of the stress response occurs through a receptor- mediated mechanism that contributes to a cell’s ability to recover, as opposed to protecting against a stress event (Munck et al., 1984; Munck and Naray-Fejes-

Toth, 1992).

The four mechanisms of action suggested by Munck et al. can be

separated into permissive, suppressive, stimulatory, and, finally, preparative.

The first mechanism of action, permissive, occurs before the stress event. It is

associated with the basal activity of the receptor and occurs regardless of the

presence or absence of stress. The permissive actions of the GR can be seen

as part of the cell’s normal defense and provides a mechanism that primes the

42 cell to respond to a stimulus. Examples of this response can be seen in the GR-

mediated enhancement or full activation of pathways like the vasoconstrictive

pathway associated with catecholamine action, (Davies and Lefkowitz, 1984),

enhancement of gluconeogenesis following stress (Exton et al., 1972), and the

stimulation of the mammalian immune response by an increase in the cytokine

receptors for IL-1 (Akahoshi et al., 1988).

The second mechanism of GR action is the suppressive effect, which is

attributed to the enhancement of glucocorticoid release and is associated with

normal activity. Suppressive effects protect the body, and cell, from the damage

that is associated with over activation of the defense mechanisms (Munck et al.,

1984; Munck and Naray-Fejes-Toth, 1992). Examples of this activity can be

seen in the glucocorticoid-mediated inhibition of insulin (Philippe and Missotten,

1990) and IL-1 (Philippe and Missotten, 1990). The suppressive and permissive

effects of the GR demonstrate a paradox in the receptor activity under normal and stress conditions. Consider this a duality in receptor function, the permissive effects occur in the early stages of a stress event and allow the cell to respond to the stress. The suppressive effects follow the permissive effects. The suppressive effects function as a reply to the permissive resulting in a normalization of the cells reaction to stress and re-establishes normal growth situations (Munck and Naray-Fejes-Toth, 1992).

The third mechanism of GR activation is the stimulatory response, which

occurs through stress-induced increases in GR activity and enhances the cells

normal defense mechanisms. This response is slow to occur and happens only

43 after the stress, responding within an hour. Once the stress is removed, the

response is attenuated and gradually returns to a basal state. The stimulatory

response can be seen in the enhancement of the permissive effects observed

prior to the stress. An example of the stimulatory response can be seen in

muscle tissue treated with excess glucocorticoids and the increase in glucose

production (Cleasby et al., 2003).

The final mechanism is the preparative effect, which does not influence

the immediate response to a stress but modulates the response to subsequent

stressors, creating a stress tolerant cell.

In reality, the effects observed in GR signaling work in conjunction with each other to achieve the desired result. The realization that an organism is multicellular with different tissues and cell types makes this a complex physiological system to unravel. Furthermore, the observation that the endogenous and synthetic ligands, like dexamethasone, bind to the GR and to the mineralocorticoid receptors with different affinities (Schmidt and Meyer, 1994;

Lim-Tio and Fuller, 1998; Farman and Rafestin-Oblin, 2001) and can induce non- physiological results make the mechanism even more complex.

This in mind, we suggest one way cells may communicate under stress conditions is through cell culture media. A stress release factor (SRF) might function in these cases to increase the activity of the GR through two of the above-mentioned mechanisms. First, the SRF could function in a manner consistent with a permissive effect and occur prior to and in the absence of stress. Second, the actions of the SRF, in the presence of hormone, may be

44 interpreted as a preconditioning effect. Preconditioning would produce an

increase in GR function and induce GR-mediated stress-response genes in unstressed cells. This multifunctional approach is consistent with the ideas

suggested by Allan Munck.

The mechanism of cells reaction to stress have been a continuous subject

of research dating back to the original observations of a stress response in the

work by Ritossa in 1962 (Ritossa, 1962). The observed stress-responsive genes

and HSPs have since been understood to function in the protection and

resistance to stress (Schlesinger, 1990; Parsell and Lindquist, 1993). Under

these circumstances, the HSPs function as chaperones to refold and stabilize the

accumulating denatured proteins. However, under stress conditions HSPs can

function in the stabilization of the cellular membrane (Multhoff and Hightower,

1996; Sapozhnikov et al., 2002) and have the ability to function as cell surface signaling molecules (Guzhova et al., 1998; Asea et al., 2000, 2002). An exact

mechanism for their cell surface signaling is still unresolved, though potential

functions can be observed in autoimmune diseases and cell stress tolerance

(Multhoff and Hightower, 1996; Guzhova et al., 2001). Signaling observed from

HSPs is not found to be part of the debris associated with cell death or lysis

(Tytell et al., 1986; Hightower and Guidon, 1989), but is released from

mammalian cells under normal conditions (Hightower and Guidon, 1989;

Guzhova et al., 2001). An example of this can be seen in the HSPs expressed in

the autoimmune diseases resulting in high levels of Hsp70 and Hsp60 outside

45 the cell, which are necessary for induction of the associated disease effects

(Minota et al., 1988; Engman et al., 1990).

The observation of extracellular HSPs was found to induce a

preconditioning or stress tolerance in neuronal cells (Guzhova et al., 2001).

Preconditioning in the neuronal cells occurred through the direct uptake of Hsp70

and Hsc70 into the cytoplasmic and nuclear compartments (Guzhova et al.,

1998, 2001; Fujihara and Nadler, 1999). The exact method of this HSP uptake is

not entirely understood. However, HSPs can interact with the plasma membrane

and function to form an ion pore (Alder et al., 1990; Arispe and De, 2000; Arispe

et al., 2002), suggesting the HSP access is through a membrane pore system.

Exogenous HSPs can also interact with the plasma membrane through a

receptor-based system. Work by Calderwood et al. demonstrated HSPs, like

Hsp70, Hsp90, and Hsp60, can interact with the membrane-based Toll and Toll

like receptors (TLRs) (Asea et al., 2002). The Toll receptors are in a family of

proteins characterized by extracellular leucine-rich repeats, consistent with those

found in the IL-1 receptor family (Medzhitov et al., 1997). The extracellular

leucine-rich domain of the TLRs is the site responsible for ligand and HSP

binding to the receptor (Vabulas et al., 2002; Direskeneli and Saruhan-

Direskeneli, 2003; Janssens and Beyaert, 2003). The intracellular, cytoplasmic

domain of the TLRs is responsible for signal transduction.

The concept of an extracellular signal functioning as a stress sensor is

relatively new. R.J. Rowbury suggests that extracellular sensors function as

“alarmones” or warning signals of an impending stress, inducing a tolerance for

46 stress in bacteria (Rowbury and Goodson, 2001; Rowbury, 2001a, b, 2003). His

work suggests that extracellular signals do more than just function in an

autocrine fashion, they have the ability to travel to adjacent unstressed cells or

cells failing to produce the extracellular signals and function to prepare them for a

stress event (Rowbury and Goodson, 2001). This concept of a stress-released,

extracellular signal, functioning to protect unstressed cells is seen in neuronal

cells. Work by Cunningham et al. presented evidence for a media-soluble

peptide that has the ability to promote survival from a stress event in

retinoblastoma and mouse hippocampal cells (Cunningham et al., 1998). The

peptide did not only function on the same cell in an autocrine fashion, but had the

ability to work like a “pheromone” warning the adjacent unstressed cells of an

impending stress. Though no direct evidence was shown demonstrating a role

for the GR or HSPs in the survival of these cells, it does demonstrate the ability

of a stressed cell to release a signaling product with the ability to promote cell survival.

47 Enhancement of Glucocorticoid Receptor Transactivity

by Constitutively-active Heat Shock Factor 1

Thomas J. Jones

Dapei Li

Irene M. Wolf

Subhagya A. Wadekar

Sumudra Periyasamy

Edwin R. Sánchez*

Department of Pharmacology

3035 Arlington Avenue

Medical College of Ohio

Toledo, Ohio 43614

Phone: (419) 383-4182

FAX: (419) 383-2871

Email: [email protected]

*To whom correspondence should be addressed

Keywords: steroids, glucocorticoid, nuclear receptor, heat shock, heat shock

transcription factor, Hsp70, Hsp70 promoter, PCR

48 ABSTRACT

To further define the role of heat shock factor 1 (HSF1) in the stress potentiation

of glucocorticoid receptor (GR) activity, we placed a constitutively-active mutant

of human HSF1 (hHSF1-E189) under the control of a doxycycline-inducible

vector. In mouse L929 cells, doxycycline (DOX)-induced expression of hHSF1-

E189 correlated with in vivo occupancy of the hHsp70 promoter (chromatin-

immunoprecipitation assay) and with increased activity under non-stress

conditions at the human Hsp70 promoter controlling expression of CAT (p2500-

CAT). Comparison of hHSF1-E189 against stress-activated, endogenous HSF1

for DNA-binding, p2500-CAT and Hsp70 protein expression activities showed the

mutant factor to have lower, but clearly detectable, activities as compared to wild-

type factor. Thus, the hHSF1-E189 mutant is capable of replicating these key

functions of endogenous HSF1, albeit at reduced levels. To assess the

involvement of hHSF1-E189 in GR activity, DOX-induced expression of hHSF1-

E189 was performed in L929 cells expressing the minimal pGRE2E1B-CAT reporter. hHSF1-E189 protein expression in these cells was maximal at 24 h of

DOX and remained constant up to 72 h. hHSF1-E189 expressed under these conditions was found both in the cytosolic and nuclear compartments, in a state capable of binding DNA. More importantly, GR activity at the pGRE2E1B-CAT promoter was found to increase following DOX-induced expression of hHSF1-

E189. The potentiation of GR by hHSF1-E189 occurred at low and saturating concentrations of hormone and was dependent on at least 48 h of hHSF1-E189 up-regulation – suggesting that time was needed for an HSF1-induced factor to 49 accumulate to a threshold level. Initial efforts to characterize how hHSF1-E189 controls GR signaling showed that it does not occur through alterations of GR protein levels or changes in GR hormone binding capacity. In summary, our observations provide the first molecular evidence for the existence of HSF1- regulated genes that serve to elevate the response of steroid receptors under stress conditions.

50 INTRODUCTION

The glucocorticoid receptor (GR) is a ligand-activated transcription factor that serves as the principal target of steroids produced by the adrenal cortex (Evans,

1988; Tsai and O'Malley, 1994). In the absence of hormone, the GR is known to exist in a complex with several members of the heat shock protein family (Pratt and Toft, 1997) – proteins that are integral to the “heat shock” stress response found in almost all cells (Morimoto, 1993). At an organismal level, glucocorticoid hormones are known to play a variety of roles that serve to maintain homeostasis in response to stress events (Munck et al., 1984; Sapolsky et al., 2000). One of the best understood of these roles is the ability of glucocorticoids to protect against over-activity by immune and inflammatory pathways. In this respect, our recent finding (Wadekar et al., 2001) (Wadekar et al., 2004) that glucocorticoids can suppress the heat shock response in cells by inhibiting the actions of heat shock factor 1 (HSF1) serves to underscore the central role of GR in modulating stress responses.

In addition to suppression of HSF1 activity by GR, our laboratory and those of others have found evidence that reciprocal control of steroid receptor responses by stress can also occur (Edwards et al., 1992); (Sanchez et al.,

1994); (Nordeen et al., 1994); (Sivo et al., 1996). In this case, however, most studies report that heat shock and other forms of cellular stress will cause an increase in steroid receptor transcriptional activity. Keys features of this response include: 1) heat shock potentiation of GR activity at all concentrations of hormone, up to and including saturating levels of hormone, 2) stress potentiation

51 does not occur in cells devoid of GR, or containing hormone- or DNA-binding defective GR, 3) stress potentiation does not occur in response to classical GR antagonists, such as RU486, and 4) stress potentiation does not occur by cooperative binding to GR-regulated promoters by GR and any other DNA- binding transcription factor, including HSF1 (Sanchez et al., 1994). Moreover, no obvious change in amount of GR protein, or in hormone-induced translocation of

GR to the nucleus has been observed under the conditions of stress potentiation.

Although identification of the precise stage of GR signaling affected by stress has not yet been achieved, we have recently made progress by providing evidence that HSF1 activity is centrally involved in this mechanism. Through use of drugs that selectively modulate HSF1 activity under stress conditions, we have shown a corresponding modulation of GR under the same conditions (Li et al.,

1999; Li et al., 2000). For example, a flavonoid compound, quercetin, was used to prevent HSF1 activation in response to stress, while having no effect on HSF1 after activation. Under these conditions, it was found that quercetin blocked heat shock potentiation of the GR, but only when administered before the stress event. Similarly, increasing HSF1 activity under stress conditions by treating cells with a PI-3 kinase inhibitor (wortmannin) caused a concomitant increase in

GR transactivity.

Even though these pharmacological approaches provide strong evidence for the involvement of HSF1, we decided that a strictly molecular approach to this question was needed. There were several reasons for this decision. First, use of drugs to inhibit HSF1 had to be performed under conditions of stress. Thus, it

52 could not be confidently concluded that the drugs were targeting HSF1 alone, as opposed to other stress-induced signal pathways. Moreover, pharmacological approaches could not rule out the possible involvement of other members in the

HSF family, such as HSF2, which are known to be expressed in the mouse

(Mathew et al., 2000). Lastly, if HSF1 is indeed responsible for the stress potentiation of GR, then discovery of the HSF1-regulated genes involved would be much easier using a stress-free molecular approach rather than under combined conditions of stress and drug treatment. With this in mind, we report here that expression of a constitutively-active mutant of HSF1 in cells can indeed up-regulate GR transcriptional enhancement activity under stress-free conditions.

Thus, the mechanism by which heat shock and other forms of stress cause elevation of GR function most likely requires expression of HSF1-regulated genes during the post-stress recovery period.

53 RESULTS

Non-stress Expression of hHSF1-E189 in Mouse L929 Cells Mimics the Function of Endogenous Stress-activated Factor.

To further define the role of HSF1 in the stress potentiation of GR, we set out to separate intrinsic HSF1 activity from all other stress-induced mechanisms. We achieved this through use of a constitutively-active mutant of human HSF1

(hHSF1-E189) originally developed by Voellmy and coworkers (Zuo et al., 1995). hHSF1-E189 (which we also refer to as E189) contains a single amino acid substitution at residue 189 residing in one of three hydrophobic LZ domains (Fig.

1A). The LZ domains of HSF1 are thought to interact with heat shock protein chaperones, serving to maintain HSF1 in an inactive state. The E189 mutant, therefore, has stress-free activity because it cannot be properly chaperoned, leading to active HSF1 trimers under non-stress conditions (Zuo et al., 1995;

Wagstaff et al., 1998). As further diagramed in Fig. 1, the cDNA for hHSF1-E189 was placed under the control of a doxycycline-inducible vector (Gossen et al.,

1995) in cells that had previously been stably-transfected with a CAT reporters driven by the hHsp70 promoter (LHSE-CAT cells) or by the minimal GR- responsive GRE2E1B construct (LGRE-CAT cells). After selection, the stably- transfected LHSE-E189 and LGRE-E189 cells were thus established.

As an initial test, LHSE-E189 cells were exposed to 10 µg/ml of doxycycline (DOX) followed by assay of hHSF1-E189 expression by Western- blotting using an antibody specific to the human HSF1 (Fig. 2A). Here the results show appearance of E189 protein in response to DOX treatment. As a further

54 test, the ability of this protein to bind the HSP promoters in vivo was determined

by use of the chromatin immunoprecipitation assay (ChIP) using primers specific

to the human Hsp70 promoter (Fig. 2B). The results show occupancy of the

hHsp70 promoter by DOX-induced E189. To demonstrate that promoter-bound

E189 can indeed stimulate transcription in the absence of stress, a time course

of exposure to DOX was performed in LHSE-E189 cells, followed by assay for

E189 by Western blotting and CAT activity assay (Fig. 3). The results show

detectable levels of hHSF1-E189 protein as early as 4 h following DOX

treatment, with levels of protein appearing to plateau at about 20 h of DOX exposure. However, CAT expression from the hHsp70 promoter was not appreciably increased until 20 h of DOX and was still increasing at 40 h of treatment. As would be expected, this suggests that expression of E189- regulated genes lag behind expression of hHSF1-E189 protein itself.

To guard against the possibility that the CAT activity observed in these

cells was actually due to activation of endogenous mouse HSF1 by DOX, we

treated the parental LHSE-CAT cells (no E189 vector) with this compound. DOX

treatment up to 48 h had no effect on CAT expression from the hHsp70 promoter

(data not shown). Because the pBI vector used for hHSF1-E189 expression also

controls expression of green fluorescent protein (GFP), we also tested the

possibility that GFP could be activating mHSF1, perhaps by causing recruitment

of Hsp70 and other chaperones away from the inactive mouse factor. In this test

(Fig. 4A), activation of E189 and mHSF1 was assayed by Western blotting using

an antibody that detects both species of this factor. The results show presence

55 of activated mHSF1 in the nuclear fraction of heat shocked cells and the

presence of E189 in both the cytosolic and nuclear fractions. However,

endogenous, activated mHSF1 has an apparent Mr larger than that of E189 and

this band is not detected in cells exposed to DOX alone. Thus, it is unlikely that

E189 or GFP expression is leading to simultaneous activation of the endogenous factor.

Although these data show that DOX-expressed E189 is active in the absence of stress at the exogenous hHsp70 promoter, we wanted to determine whether E189 could act at endogenous promoters within these cells and the extent of this activity. We chose to analyze the endogenous Hsp70 promoters by use of Western blotting with an antibody that can detect both the “constitutive”

(Hsp70c) and “inducible” (Hsp70i) forms of Hsp70 (Fig. 4B). The data show that

DOX treatment of LHSE-E189 cells can indeed cause increased expression of both Hsp70c and Hsp70i. However, the levels of induction by DOX for each of these proteins, although clearly elevated, were low compared to levels obtained in response to sodium arsenite (a potent inducer of HSF1 activity). Because of this, we reasoned that E189 activity at the exogenous hHsp70 promoter (p2500-

CAT) may also be weak compared to arsenite and other stressors. The results of Fig. 5A show this to be the case, as DOX-induced CAT activity in the LHSE-

E189 cells was about 50% of the activity obtained in response to heat shock and only about 15% of the activity seen after arsenite. To help determine whether reduced activity by E189 was due to relative lack of DNA-binding or to a deficiency of transcription activation function by this factor, we compared

56 activation of E189 and endogenous mouse HSF1 by EMSA (Fig. 5B). Here we

found that binding to DNA by DOX-expressed E189 was reduced compared to

heat shock-activated mHSF1. Thus, it is likely that low E189 activity at

endogenous promoters may be due to reduced promoter binding compared to

that seen for stress-activated endogenous factor. Taken as a whole, however,

our data clearly show that the hHSF1-E189 mutant is capable of replicating

several key functions of endogenous HSF1 without the need for stress, albeit at

reduced levels.

Expression of hHSF1-E189 Causes Non-stress Potentiation of Glucocorticoid

Receptor Transactivity

To test the effect of intrinsic HSF1 activity on GR function, we generated the

LGRE-E189 cells (Fig. 1) in which DOX-regulated expression of hHSF1-E189

occurs in cells containing the pGRE2E1B-CAT reporter. A time course of DOX exposure was performed in these cells to establish the kinetics of E189 expression (Fig. 6A). The results show an unusual but reproducible pattern in which E189 protein expression is not appreciably detected until 24 h of DOX exposure, with E189 levels remaining constant thereafter (up to 72 h of treatment). Because we could not directly measure E189 transactivity in these cells, we tested the ability of E189 to localize to the nucleus (Fig. 6B) and to bind

DNA (Fig. 6C). Like in the LHSE-E189 cells (Figs. 4 and 5), E189 in the LGRE-

E189 cells was found both in the cytosolic and nuclear compartments and was capable of binding DNA.

57 As we have previously shown (Li et al., 2000), the response to hormone in

LGRE-CAT cells is relatively low due to the intrinsic limitations of the minimal

pGRE2E1B-CAT reporter. However, in these same cells, the response at this promoter can be dramatically increased when heat shock or arsenite treatment is combined with hormone. It can be seen in Fig. 7A that the LGRE-E189 cells show a similar pattern of responses to hormone and stress conditions, with arsenite typically giving a much higher potentiation of GR activity than heat shock. In Fig. 7B, we measured GR activity in the same cells under conditions of

E189 up-regulation (DOX). An increase in GR activity at the pGRE2E1B-CAT

reporter was seen in response to DOX treatment for 48 h and the response was

even greater at 72 h of treatment. DOX alone had no effect on this activity.

Moreover, DOX plus hormone had no effect on the GR response in the parental

LGRE-CAT cells containing no E189 vector (data not shown). Thus, it appears

that intrinsic HSF1 activity can indeed control ligand-induced GR responses in

these cells. It should be noted that the magnitude of the effect seen at 72 of

DOX (Fig. 7B) is starting to approach the level of response seen following heat

shock treatment in these same cells (Fig. 7A).

Because HSF1 is known to be the major regulator of Hsp70 and Hsp90

levels in cells (Morimoto, 1993) and because these HSPs are known to associate

with unliganded GR heterocomplexes (Pratt and Toft, 1997), we reasoned that

E189 could be causing potentiation of the GR by altering GR heterocomplexes in

a way that leads either to more GR or to GR with increased hormone-binding

capacity. Interestingly enough, both of these possible effects of E189 up-

58 regulation were not observed (Fig. 8). Thus, it is likely that HSF1 is targeting a site of action downstream of the hormone-free GR heterocomplex.

59 DISCUSSION

We have shown that the E189 mutant under non-stress conditions can effectively

replicate most of the key properties of HSF1, including the ability to activate HSP gene expression at both heterologous (p2500-CAT) and endogenous (Hsp70) genes. In so doing, we have been able to reconcile a long-standing issue with respect to the mechanism by which heat shock and other forms of stress cause enhancement of glucocorticoid receptor activity – namely, whether HSF1 signaling itself (as opposed to other stress-activated events) was the principal mechanism responsible for GR up-regulation. Although in prior publications we have shown evidence for involvement of HSF1 in the GR potentiation (Li et al.,

1999; Li et al., 2000), the approaches taken in those studies involved the use of pharmacological agents applied to cells experiencing stress. Thus, a complete separation of all possible stress-activated signal mechanisms from the HSF1 pathway was not possible until the present study. It is now clear that intrinsic

HSF1 activity can indeed lead to a potentiation of GR transactivation.

Based on these findings, the next major unanswered question is how

HSF1 achieves this effect on GR. Because HSF1 is a transcription factor best

known for its regulation of Hsp90 and Hsp70 and because both of these HSPs

are known to regulate assembly of GR heterocomplexes (Morishima et al., 2000)

and the ability of receptor to bind hormone (Dittmar and Pratt, 1997), we

reasoned that HSF1-induced changes to GR cytoplasmic heterocomplexes could

explain increased activity by the receptor. However, this mechanism now seems

less likely, as E189 potentiation of GR transactivity occurred without any changes

60 to GR expression levels or overall hormone-binding function (Fig. 8). Yet it

remains a possibility that GR complexes are altered in a way that can still affect

transactivation without changing hormone-binding capacity. One such

mechanism is through up-regulation of immunophilins, such as FKBP52 and

Cyp40 – both of which show increased expression following stress (Sanchez,

1990; Mark et al., 2001). In the case of FKBP52, this protein appears to play a role in the targeting of GR to the nucleus following the hormone-binding event

(Czar et al., 1995; Davies et al., 2002). Of course, it is also possible that an

HSF1-regulated gene may control GR at any number of other steps in the GR signal pathway, including the transactivation stage. Our demonstration here that

E189 activity under non-stress conditions can essentially replicate the stress potentiation of GR will now make it easier to identify this HSF1-induced product(s) through use of genomic or proteomic approaches, for example.

If HSF1 can indeed stimulate GR activity, what is the function of this event? Although we do not yet have an answer to this question, one possible explanation is that GR activity under stress serves to promote cell survival and that heat shock signaling stimulates this activity. In this sense, GR may serve a similar function to HSF1, whose role in protecting against stress-induced cellular lethality is well documented (Jolly and Morimoto, 2000). Although studies showing a protective role of glucocorticoids are numerous [see reviews by Munck and colleagues (Munck et al., 1984; Sapolsky et al., 2000)], an interesting example is the ability of glucocorticoid agonists (in the absence of stress) to induce a state of “thermotolerance” similar to that seen when cells are subjected

61 to a conditioning, sub-lethal heat stress (Fisher et al., 1986); (Anderson et al.,

1991). In our laboratory, we have observed that combined stress and

glucocorticoid treatment leads to a rate of cell survival that is dramatically higher

than that seen following stress treatment alone (unpublished observations).

Thus, by the measure of cellular viability, a synergistic relationship between the

heat shock response and GR activity does appear to occur.

However, the simplicity of the above model must be tempered by our

concurrent observation that glucocorticoid agonists appear to have an inhibitory

effect on the heat shock response itself, principally by inhibiting the ability of

HSF1 to act as a transcription factor (Wadekar et al., 2004). How, therefore, can

these seemingly contradictory phenomena be reconciled? One explanation is

that HSF1 potentiation of GR is simply a mechanism by which to insure its own

down-regulation (potentiated negative-feedback). Yet, the rapid nature of GR actions on HSF1 makes this unlikely (Wadekar et al., 2004). Instead, we believe

(as further explained below) that the principal role for HSF1 potentiation of GR is to maximize production of genes involved in survival. Moreover, the feedback model does not explain how HSF1 can cause potentiation of GR when it is being

“simultaneously” inhibited. A potential, albeit quantitative, solution to this

problem is that glucocorticoid inhibition of HSF1 activity is not 100% effective.

Typically, about 30% HSF1 activity remains when glucocorticoid treatment

occurs before the stress event, even at a concentration of 1 µM dexamethasone

(Wadekar et al., 2001). Thus, under the most stringent conditions, enough HSF1

activity may remain to cause the actions on GR documented in this work. In

62 most experiments of this kind, however, we typically add the hormone after the stress event, as this appears to yield a higher potentiation effect on the receptor

– although a rigorous investigation of this has proven difficult to do in a way that maintains both equal exposure time to hormone and equal recovery time following stress.

The above issues aside, we believe that a more relevant model involves intertwined actions of HSF1 and GR that are not simultaneous (Fig. 9). In most experiments designed to inhibit HSF1, we have added hormone to cells at or before the time of stress. It is likely that such treatment is an artificial condition that most cells do not experience in a physiological context. Instead, the stress event is likely to occur first in an environment of relatively low glucocorticoid hormone concentration. In this case, the heat shock response in affected tissues would proceed uninhibited until the stress event triggers a rise in glucocorticoid secretion as controlled by the hypothalamus-pituitary-adrenal axis – a result that has indeed been observed in rats exposed to restraint stress (Bhatnagar and

Vining, 2003). Elevated hormone levels would then secondarily lead to a rapid attenuation of the heat shock response, presumably to prevent over-stimulation by this response, or to provide an alternative mechanism of cell survival, or both.

Yet at this point in the course of events, HSF1-controlled genes responsible for potentiation of GR activity would have already been expressed, producing an elevated response to hormone that most likely serves to restore normal cellular functioning through gene products that cannot be produced by the heat shock pathway itself. Thus, one way to look at this model is that the heat shock

63 response is the cell’s initial survival mechanism that, in addition to producing

protective heat shock proteins, also serves to prime optimal response for a later-

acting survival mechanism mediated by the GR – a mechanism that involves

rapid moderation of the heat shock response itself and increased production of

gene products that likely serve to re-establish cellular homeostasis.

Although many aspects of the above model remain to be confirmed, we do

know that the peak time for either heat or chemical shock potentiation of GR

occurs approximately 16 h into the recovery period (Hu et al., 1996) – kinetics

that are consistent with a temporal pattern in which the protective role of

hormone follows the initial stress event. Lastly, we believe that our elucidation of

this complex relationship between GR and HSF1 has important implications for

the treatment of disorders arising from pathophysiological stress, especially if

novel GR-regulated genes can be identified with primary roles in the restoration

of cellular health.

MATERIALS AND METHODS

Materials

[3H]Dexamethasone (NET467; 42.8 Ci/mmol), [3H]acetate (10.3

µCi/mmol), and [125I]conjugates of goat anti-mouse IgG (NET159; 11.8 µCi/µg)

and goat anti-rabbit IgG (NET155; 9.0 µCi/µg) were obtained from New England

Nuclear. Doxycycline, ATP, DMSO, sodium arsenite, dexamethasone, G418

(Geneticin) antibiotic, hygromycin, acetyl CoA synthetase, acetyl Co-enzyme A,

Tris, Hepes, EDTA, protein A-Sepharose, protein G-Sepharose, DMEM

64 powdered medium were from Sigma Chemical Co. Horseradish peroxidase-

conjugates of goat anti-mouse and goat anti-rabbit IgG were from Calbiochem.

Iron-supplemented newborn calf serum was from Hyclone. Immobilon® PVDF membranes were obtained from Millipore Corp. GenePorter transfection reagent was obtained from Gene Therapy Systems, Inc. The FiGR mouse monoclonal antibody against GR (Bodwell et al., 1991) was a gift from Jack Bodwell

(Dartmouth Medical School) and was expressed and affinity-purified by Biocon.

The Stressgen SPA-812 antibody was used to detect the inducible and constitutive forms of Hsp70. To analyze HSF1, several antibodies were employed. Neomarker’s HSF1-AB4 antibody was used to detect both mouse and human HSF1, while the PA3-017 (Affinity Bioreagents) and the SPA-901

(Stressgen) antibodies showed selectivity for human HSF1. Rat monoclonal antibody against human HSF1 (HSF1-AB4) was purchased from Neomarkers.

The PA3-017 antibody against mouse HSF1 was from Affinity BioReagents, while the SPA-901 antibody recognizing mouse and human HSF1 was from

Stressgen. Technical grade rat IgG and mouse IgG2a were bought from Sigma.

In the p2500-CAT reporter used in this study, expression of

chloramphenicol acetyltransferase (CAT) is controlled by the human Hsp70

promoter. This promoter contains consensus heat shock elements (HSEs) that

are activated by binding of heat shock factor 1 (Schiller et al., 1988). The

pGRE2E1B-CAT minimal reporter is composed of two synthetic GREs derived from the tyrosine aminotransferase (TAT) promoter linked to the adenovirus E1B

TATA sequence (Allgood et al., 1993). The pBI-EGFP vector was obtained from

65 Clontech. In this vector, expression is controlled by a tetracycline-response

element and two minimal cytomegalovirus promoters arranged in opposite

orientations. Expression of Enhanced Green Fluorescent Protein (EGFP) was

used to isolate positive cell colonies. The pUHD172-1hygro vector (Gossen et

al., 1995), expressing the “reverse tet” trans-activator and hygromycin resistance

genes, was obtained from Hermann Bujard (Universitat Heidelberg). The cDNA

for the E189 mutant of human HSF1 (Zou et al., 1995) was the generous gift of

Richard Voellmy.

Transfection of Cell Lines

The LHSE-CAT and LGRE-CAT cell lines were established as previously

described (Sanchez et al., 1994; Li et al., 1999). Briefly, mouse L929 cells were

co-transfected with pSV2neo and a two-fold excess of p2500-CAT (to yield

LHSE-CAT cells) or pGRE2E1B-CAT (to yield LGRE-CAT cells) using

GenePorter as carrier. This was followed by selection for stably-transfected,

cloned cell lines using G418 (Geneticin) antibiotic at 0.4 mg/ml. Once

established, each cell line was grown in an atmosphere of 5% CO2 at 37°C in

DMEM containing 0.2 mg/ml of G418 and 10% iron-supplemented newborn calf

serum. The tetracycline-inducible LHSE-E189 and LGRE-E189 cells were made

by stably-transfecting LHSE-CAT or LGRE-CAT cells with the pUHD172-1hygro

plasmid and a seven-fold excess of pBI-E189 plasmid, followed by selection and

cloning using 0.4 µg/ml of hygromycin. The pBI-E189 construct was made by

excising the cDNA for the constitutively-active hHSF1-E189 mutant from the 66 pGEM-E189 vector originally developed by Voellmy and coworkers (Zou et al.,

1995). This cDNA was then inserted into the multiple cloning site of the pBI-

EGFP vector (Clontech).

Stress Treatment of Cell Lines

For all experiments, the newborn calf serum was stripped of endogenous

steroids by extraction with dextran-coated charcoal. Most stress experiments were performed on cells that were at or near confluence; although similar results were obtained with sub-confluent cultures. Heat shock treatment was achieved by shifting replicate flasks to a second 5% CO2 incubator set at 43°C. Duration of heat shock treatment was 2 h, or as indicated. Cells were also subjected to chemical shock by addition of 200 µM sodium arsenite to the medium. In the

chemical shock experiments, the arsenite-treated and non-treated cells were

incubated at 37°C for 2 h and were then washed with DMEM medium and

allowed to recover, or were harvested immediately after stress.

Chromatin Immunoprecipitation Assay

To detect binding of HSF1 to the hHsp70 promoter in vivo, chromatin

immunoprecipitation assay was performed according to the method of Nissen

(Nissen and Yamamoto, 2000) with some modifications. Briefly, replicate flasks

of LHSE-E189 cells were treated as described in the legend to Fig. 2, followed by

cross-linking with formaldehyde and preparation of nuclear extracts. After

sonication, crude fragments of protein-linked chromatin were further subjected to

67 immunoprecipitation using an antibody specific to human HSF1 or an equivalent amount of non-immune serum as control, followed by immobilization on protein G

Sepharose. The samples were washed, digested with proteinase K and cross links were reversed by heating. DNA was extracted and purified and subjected to 25 cycles of PCR. The 20 bp forward primer 5’-GGA AGG TGC GGG AAG

GTT CG-3’ was designed to bind at –75 of upper strand of the human Hsp70 promoter used in the p2500-CAT reporter. The backward primer 5’-TTC TTG

TCG GAT GCT GGA-3’ was chosen to bind at +110 of the lower strand. The size of product obtained was 185 bp. PCR products were run on 2% agarose gels containing ethidium bromide and photographed.

Fractionation, Immune Purification and Western-blotting

In the experiments of Figs. 4B and 7B, cells were fractionated into cytosolic and nuclear portions by Dounce A homogenization in hypotonic buffer, followed by centrifugation at 1,000 x g. The cytosolic fractions were saved and the nuclear pellets were washed 2 times by resuspension and pelleting in hypotonic buffer.

Hypotonic buffer containing 0.5 M NaCl was added to the pellet fractions and incubated on ice with occasional vortexing for 1 h. After salt extraction, the nuclear pellets were centrifuged at 14,000 x g and the supernatants saved.

Cytosolic and nuclear fractions were adsorbed in-batch to protein A-Sepharose, using the HSF1-AB4 antibody recognizing both the human and mouse forms of

HSF1. Sepharose pellets were washed with TEG buffer (10 mM TES, 1 mM

68 EDTA, 10% glycerol, 50 mM NaCl, 10 mM sodium molybdate, pH 7.6) and eluted

with 2X SDS sample buffer.

In the experiments of Figs. 2, 3, 5, 7 and 10, whole cell extracts were

prepared by freezing of cells at –80°C and resuspension in WCE buffer (20 mM

Hepes, 25% glycerol, 0.42 M NaCl, 1.5 mM MgCl2, 0.2 mM EDTA, 0.5 mM

PMSF, 0.5 mM DTT, pH 7.9) followed by centrifugation at 100,000 x g for 10 min.

All samples were resolved by electrophoresis in 10% polyacrylamide SDS

gels, followed by transfer to Immobilon® PVDF membranes. The relative

amounts of hHSF1-E189, endogenous mouse HSF1 or GR were determined via

a Western-blotting technique previously described (Tienrungroj et al., 1987),

employing primary antibody and both peroxidase- and 125I-conjugated counter antibodies. After color development, the blots were exposed to Kodak XAR-5 film with an intensifying screen at –80°C. Relative amounts of proteins were determined by densitometric scanning of films using a Molecular Dynamics scanner and software.

CAT Assay

Measurement of CAT enzyme activity was performed according to the method of

Nordeen et al (Nordeen et al., 1987) with minor modifications. In this assay, a

reaction mixture containing acetyl CoA synthetase, [3H]sodium acetate,

coenzyme A (CoA) and ATP is briefly preincubated to enzymatically generate

labeled acetyl coenzyme A from CoA and labeled acetate. Acetylation of

chloramphenicol was then initiated by addition of cell lysate containing CAT

69 enzyme. The reaction was stopped by extraction with cold benzene and 75% of

the organic phase was counted. Cell lysates were prepared by sequential

freezing and thawing in 0.25 M Tris, 5 mM EDTA (pH 7.5) and centrifugation at

14,000 X g. Aliquots of lysate containing equal protein content were added to the

enzymatic reaction mixtures. As the HSE- and GRE-containing promoters

employed in this study have distinct basal and inducible activities, all data is

represented as percent of control, maximum or the equivalent. In this way, the

relative effects of each treatment can be readily seen.

Electrophoretic Mobility Shift Assay (EMSA)

EMSA assays for HSF1 were performed according to the protocol of Mosser et al

(Mosser et al., 1993), with minor modifications. Briefly, cells were harvested, centrifuged and rapidly frozen at –80°C. The frozen pellets were resuspended in

WCE buffer and centrifuged at 100,000 x g for 10 min. The supernatants were either stored at –80°C or used immediately. EMSA was performed by mixing 10

µg of whole cell extract with 0.1 ng (50,000 cpm) of [32P]-labeled HSE

oligonucleotide (5’-GAT CTC GGC TGG AAT ATT CCC GAC CTG GCA GCC

GA-3’) and 1.0 µg of poly (dI-dC) in 10 mM Tris (pH 7.8), 50 mM NaCl, 1 mM

EDTA, 0.5 mM DTT, 5% glycerol in a final volume 25 µl. For competition

experiments, the binding reactions contained 0.1 ng of the [32P]HSE and a 100- fold molar excess of unlabeled HSE. Reactions were incubated at 25°C for 30 min and loaded onto 4% polyacrylamide gels in 0.5X TBE. The gels were run at room temperature for 1.5 h at 150 V and were exposed to Kodak XAR-5 film with

70 an intensifying screen at –80°C. The relative amounts of probe-bound HSF1 were measured by densitometric scanning of the film using the Bio-Rad

Molecular Analyst system.

ACKNOWLEDGEMENTS

The authors wish to thank Dr. Richard Voellmy for his generous gift of cDNAs encoding the human HSF1-E189 mutant and the p2500-CAT reporter. We are also grateful to Dr. Hermann Bujard for the pUHD172-1 vector, Dr. Jack Bodwell for the FiGR antibody and Dr. John Cidlowski for the pGRE2E1B-CAT reporter.

This work was supported by National Institutes of Health grant (DK43867) to

E.R.S.

71 FIGURE LEGENDS

Fig. 1. Transfection of hHSF1-E189, a Constitutively-active Mutant of HSF1, into

GR- and HSF1-responsive Backgrounds. A, HSF1 is a 529 aa protein

incorporating multiple leucine zipper regions (LZ), a DNA-binding domain (DBD)

and a C-terminal transactivation domain (CTR). A point mutation at residue 189

introduced a collapse of LZ2 to produce a constitutively-active form of HSF1

(E189). B, Mouse L929 cells stably selected for the p2500-CAT (Hsp70) or the

minimal pGRE2E1B-CAT promoters were used as parent cells for the

construction of stable cell lines expressing E189. C, E189 was placed in the

doxycycline-inducible bi-directional vector PBI-EGFP. Parent cells were co-

transfected with pBI-EGFP-E189 and the transactivator vector pUHD172-1hygro,

followed by hygromycin selection and cloning of GFP-positive cells to generate

the LHSE-E189 and LGRE-E189 cell lines.

Fig. 2. Dox-Induction and In Vivo Promoter Binding of E189 in LHSE-E189-CAT cells. A, LHSE-E189-CAT cells were treated with DOX (10 µg/ml) for 24 h, followed by lysate preparation and analysis by Western-blotting using an antibody against human HSF1. B, Chromatin Immunoprecipitation Assay (ChIP) was performed, as described in Materials and Methods. Briefly, equal amounts of cross-linked lysates were sonicated and immunoabsorbed (IP) with antibody

against HSF1 (åHSF1) or non-immune control (NI). Following washing and

reversal of cross-linking, PCR was performed using primers against the hHsp70i

promoter (p2500-CAT), as indicated. Conditions: C, no treatment; Dex, 1 µM

72 dexamethasone for 24 h; DOX, 10 g/ml of doxycycline for 24 h.

Fig. 3. Transcription Enhancement Activity of DOX-induced E189 in LHSE-

E189-CAT Cells. A, LHSE-E189-CAT cells were subjected to the indicated time-

course of DOX (10 µg/ml) treatment and analyzed for E189 protein expression by

Western-blotting with antibody against human HSF1. B, Same time-course as in

Panel A, except that lysates were analyzed for CAT gene expression. Results

represent the mean +/– S.E.M. of three independent experiments.

Fig 4. DOX-induced Expression of HSP70 Genes by Constitutively Active E189.

A, DOX treatment does not activate endogenous HSF1. LHSE-CAT cells were subjected to DOX for 72h (DOX) or heat shock at 43˚C for 2 h with no recovery

(HS). After Dounce homogenization, the cells were analyzed for subcellular

localization of HSF1 by immunoabsorption (IP) of cytosolic (C) and nuclear (N)

fractions with nonimmune antibody (åNI) or antibody that detects both mouse and human HSF1 (åHSF1). B, DOX increases expression of inducible and constitutive Hsp70 (Hsp70i and Hsp70c, respectfully). LHSE-E189-CAT cells were treated with 10 µg/ml of DOX for the indicated time or were subjected to chemical shock (CS) with 200µM sodium arsenite for 2 h and allowed to recover for 24 h. After treatment, Western blot analysis was performed using an antibody against the constitutive and inducible forms of Hsp70.

Fig. 5. Comparison of Transactivity and DNA-binding Properties of E189 and

Endogenous mHSF1. A, Promoter activities by E189 and mHSF1 in LHSE-

73 E189-CAT cells were measured in response to no treatment (Con), 10 µg/ml of

doxycycline for 24 h (DOX), heat shock at 43 ºC for 2 h (HS), or chemical shock using 200 µM sodium arsenite for 2 h (CS). Stressed cells were allowed to grow for an additional 24 h under normal conditions prior to harvesting. The results represent the mean +/– S.E.M. of 3 to 6 independent experiments. B, DNA- binding activities of E189 and mHSF1 in LHSE-E189-CAT cells were measured by EMSA assay and subsequent quantitation by densitometric scanning. Cells were subjected to the following conditions. Lane 1, no treatment (Con); lane 2,

10 µg/ml doxycycline for 24h (DOX); lane 3, lysates of DOX-treated cells incubated with antibody against HSF1; lane 4, lysates of DOX-treated cells incubated with unlabeled oligonucleotide; lane 5, heat shock at 43 ºC for 2 h with no recovery (HS); lane 6, lysates of HS-treated cells incubated with antibody against HSF1; lane 7, DOX treatment for 24 h followed by HS (DOX/HS); lane 8, lysates of DOX/HS-treated cells incubated with antibody against HSF1. Results represent the mean +/– S.E.M. of nine independent experiments.

Fig. 6. DOX-Induced E189 Nuclear Expression and DNA-binding in LGRE-E189 cells. A, LGRE-E189 cells were subjected to a time-course of 10 µg/ml doxycycline (DOX) followed by Western-blot analysis of E189 protein. Results are representative of four independent experiments. B, LGRE-E189 cells were treated with doxycycline for 72 h (DOX) or vehicle control, followed by Dounce homogenization and analysis for subcellular localization of HSF1. Aliquots of cytosolic (C) and nuclear (N) fractions were immunoabsorbed with antibody

74 against mouse and human HSF1 followed by Western blotting. C, LGRE-E189

cells were treated with 10 µg/ml doxycycline (DOX) for 0, 24, 48 and 72 h, or were subjected to heat shock (HS) or chemical shock (CS). Cells were

harvested immediately after the stress or DOX treatment and analyzed by EMSA.

Results are representative of four independent experiments.

Fig. 7. DOX-induced E189 Expression Increases GR Transactivity. A, LGRE-

E189 cells were subjected to no treatment (Con), 1 µM dexamethasone for 24 h

(Dex), heat shock for 2 h followed by 1 µM dexamethasone for 24 h (HS/Dex), or

chemical shock for 2 h followed by 1 µM dexamethasone for 24 h (CS/Dex). B,

LGRE-E189 cells were subjected to no treatment (Con), 10 µg/ml doxycycline for

72 h (DOX), 1 µM dexamethasone for 24 h (Dex), or 10 µg/ml doxycycline for 48

and 72 h with 1 µM dexamethasone present during the last 24 h of treatment

(DOX + Dex). Each panel shows relative CAT activity from the GR-responsive

promoter. Results represent the mean +/– S.E.M. of three (Panel A) and 12 to

18 (Panel B) independent experiments.

Fig. 8. DOX Induction of E189 Expression Has No Effect on GR Protein Levels

or Hormone Binding Capacity. A, LGRE-E189 cells were subjected to a time-

course of 10 µg/ml doxycycline, as indicated, followed by Western blot analysis

of GR and quantitation by densitometric scanning. Results are the mean +/–

S.E.M. of four independent experiments. B, LGRE-E189 cells were subjected to

0 or 72 h of 10 µg/ml doxycycline (DOX), followed by measurement of specific

75 hormone-binding capacity using [3H]dexamethasone. Results are the mean +/–

S.E.M. of six independent experiments.

Fig. 9. Model for Reciprocal Regulation of GR and HSF1 Signaling. In this work and in our companion paper (Wadekar et al., 2004), we have provided evidence for a complex functional relationship between GR and HSF1 with the following overall properties. In stressed cells, activation of GR signaling causes rapid inactivation of HSF1 by blocking its actions at the promoter of Hsp70 and probably other HSP genes. Meanwhile, the stress event also leads to a stimulation of GR transcription enhancement activity that requires, at least in part, functional HSF1. As detailed in the Discussion, we believe that the best model by which to understand these seemingly opposing processes is one in which full recovery of cells following a stress event occurs through precise, ordered activities on the part of HSF1 and the GR. According to this model, the initial stress event is likely to occur under conditions of low glucocorticoid hormone concentration. Under these conditions, HSF1 action would be un- impeded and would provide “First-Order Protection”, principally through up-

regulation of molecular chaperones (HSPs). At some later point, the stress event

is likely to lead to elevated secretion of glucocorticoids as controlled by the

hypothalamus-pituitary-adrenal (HPA) axis. Glucocorticoid action at the stressed

cell at this stage of recovery is likely to be two-fold: rapid inactivation of HSF1 (to

prevent over-stimulation) and production of gene products that likely serve to

complete the re-establishment of cellular homeostasis. We call the latter effect

76 “Second-Order Protection” and it should be noted that one or more gene products produced by HSF1 earlier in the stress response are likely to serve as facilitators of this GR activity.

77 FIGURE 1.

78 FIGURE 2.

79 FIGURE 3.

80 FIGURE 4.

81 FIGURE 5.

82 FIGURE 6.

83 FIGURE 7.

84 FIGURE 8.

85 FIGURE 9.

86

Evidence for a factor that is released from stressed cells and enhances

glucocorticoid receptor responsiveness

Thomas J Jones

Edwin R Sanchez*

Department of Pharmacology

3035 Arlington Ave. Toledo

The Medical College of Ohio

Toledo, Ohio 43614

Phone: (419) 383-4128

FAX: (419) 383-2871

Email: [email protected]

* To whom correspondence should be addressed

87 ABSTRACT

We present evidence of a novel stress-induced factor, which is released from

stressed cells and can enhance steroid receptor activity in unstressed, quiescent

cells. Stress experiments utilizing heat shock (HS) at 43˚C or chemical shock

(CS) of 200µM sodium arsenite for 2 hours (h) caused the release of the factor in

a time-dependent fashion from L929 mouse fibroblasts. This stress-conditioned

media induced a ligand-dependent increase in glucocorticoid receptor (GR)

transactivity at the minimal GRE2E1B promoter. The increase observed under these conditions was greater than that seen with control-conditioned media or under normal growth conditions. Media collected from the induction of the tet- inducible constitutively active mutant of HSF1, hHSF1-E189, were unable to enhance GR activity and suggest, that HSF1 activation, in the absence of stress, was not sufficient to produce the stress released factor (SRF). Cells were also subjected to a combined treatment of DOX and stress. These results produced a decrease in GR enhancement and were symptomatic of a stress-tolerant cell with suppressed SRF release. The stress-tolerance in these cells could occur through a preconditioning effect from hHSF1-E189 transactivation. The stress- induced factor was not incorporated in normal stress-induced cellular debris and was found to be media-soluble. The existence of a stress-released, media- soluble factor, which may act in paracrine signaling, is intriguing and presents a novel avenue for drug intervention in inflammatory and other GR-dependent pathways.

88 INTRODUCTION

We have shown that cellular stress, heat shock (HS) and chemical shock (CS),

cause a dramatic increase in glucocorticoid receptor (GR) activity (Sanchez,

1992; Sanchez et al., 1994; Hu et al., 1996; Jones et al., 2004). Pharmacological

and molecular approaches have demonstrated that heat shock factor 1 (HSF1) is

responsible, at least in part, for the stress effect on the GR (Li et al., 1999; Li et al., 2000). However, it is unclear if HSF1 activity is the only factor necessary for the enhancement of GR transactivity in response to stress. To address this question we consider whether stress, in the absence of HSF1 activation, can enhance the GR by releasing a factor into the cell culture media.

Work by Munck et al. has provided the framework for this concept. His

findings suggest multiple physiological mechanisms through which the GR can

regulate a cell’s response to stress (Munck et al., 1984). He suggests regulation

of a cell’s stress response occurs through receptor-mediated mechanisms that

limit the size of the stress and contribute to the cell’s overall ability to recover,

rather than protect against a stress (Munck et al., 1984; Munck and Naray-Fejes-

Toth, 1992). Traditionally, GR regulation and the stress-responsive genes are

the accepted mechanisms that ensure a cell’s survival to stress (Morimoto et al.,

1992)

The role stress-responsive genes play in cell survival and cellular

signaling is now an emerging field of research. Some examples are seen with

stress-responsive proteins, like heat shock proteins (HSPs), maintaining the

plasma membrane during stress (Multhoff and Hightower, 1996; Sapozhnikov et

89 al., 2002). Work has also implicated HSPs as cell surface signaling molecules involved in immunoregulatory effects and neuronal survival (Guzhova et al.,

1998; Guzhova et al., 2001; Asea et al., 2002). Cunningham et al. have demonstrated, in hippocampal and retinoblastoma cells, an extracellular signaling peptide, which can protect unstressed cells and assist in cell survival

(Cunningham et al., 1998). Recent research has provided evidence that extracellular signaling, at least in the case of HSPs, can occur through Toll-like receptors (TLRs) located on the cell’s surface (Asea et al., 2002).

R.J. Rowbury suggests the extracellular sensors function as “alarmones” or warning signals of stress events and can induce stress tolerance in bacteria cells (Rowbury and Goodson, 2001; Rowbury, 2001a; Rowbury, 2001b;

Rowbury, 2003). He suggests extracellular signals function not only in an autocrine fashion, but also in a paracrine fashion on adjacent unstressed cells in advance of an impending stress (Rowbury and Goodson, 2001).

The potential for extracellular stress signaling to be involved in GR regulation is intriguing. Experimentally, we present a new model of stress signaling which includes an extracellular stress released factor (SRF), enhancing the activity of the glucocorticoid receptor (GR) independent of heat shock factor 1

(HSF1) activation in the target cell.

RESULTS AND DISCUSSION

90 In our laboratory stress has been observed to potentiate GR activity (Fig.

1). To test the hypothesis that media from stressed cells can enhance GR

transactivity, we developed a protocol, which allows for the harvest and pooling

of stress- and control-conditioned media (Fig. 2). Harvested stress- and control-

conditioned media was placed on GR responsive cells and evaluated. GR

activity was measured in mouse L929 fibroblast cells stably selected for a

minimal GR responsive promoter. HSF1 activation was evaluated through HSF1

responsive cell lines stably expressing the human heat shock protein 70 (Hsp70)

promoter-linked CAT reporter. The production of these cells and others relevant

to this study can be seen in Figure 3. To separate the effects of stress and HSF1

activation, we employed additional stable L929 cell lines containing a tet-

inducible constitutively-active mutant of HSF1, hHSF1-E189 (Wadekar et al.,

2001). In these cells, the presence of doxycycline (DOX) alone induced the

constitutively-active mutant of HSF1and allowed for the collection of non-stress

HSF1-conditioned media.

Our results demonstrated that both heat shock (HS) and chemical shock

(CS) can release a factor into the stress-conditioned cell culture media (Fig. 4).

The stress-conditioned media has the ability to enhance the activity of the GR

from the minimal GR-responsive CAT reporter, GRE2E1B-CAT. Enhancement is considerable when compared to the activity observed from control-conditioned media. Moreover, this enhancement is seen in a time-dependent fashion beginning as early as 4 hours after the stress treatment and lasts up to 20 hours after the stress was relieved. During this time, the control-conditioned media

91 showed a time-dependent decrease in the ability to enhance GR transactivity in

the presence of saturating levels of dexamethasone (1µM Dex). The time- dependent decrease in GR activity from the control-conditioned media is

consistent with that observed under normal cell growth conditions and with time-

dependent depletion of media nutrients. Stress-conditioned media appears to

contain a factor that is able to overcome the time-dependent nutrient depletion

and enhance transactivity of the GR. GR enhancement is not seen in

conditioned media after five minutes of recovery and demonstrates the GR

enhancement is not the result of residual stress effects. Four hours of recovery

is required before stress-conditioned media can induce an enhancement of GR transactivity. The apparent delayed response of the GR suggests time is required for the stress factor to be made and subsequently released into the cell culture media.

To determine if stress-conditioned media can activate endogenous HSF1, conditioned media was evaluated from the HSF1-responsive cell line containing the hHsp70-linked CAT reporter (Fig 5.). These observations are essential in determining if we had uncovered a new mechanism of crosstalk between the GR and stress or an additional HSF1-dependent mechanism of GR potentiation.

Results in Figure 5 demonstrate that stress-conditioned media has no effect on

HSF1 activation. The activity from control- and stress-conditioned media is identical to that achieved under normal growth conditions, but less than achieved with HS and CS conditions. Although a potential exists for endogenous heat shock proteins (HSPs) to be involved, we suggest SRF functions through the cell

92 culture media and is a novel, HSF1-independent pathway, with the ability to

enhance GR transactivity in a direct, or indirect manner, in target cells.

Stress-conditioned media is able to enhance the activity of the GR in a concentration- and time-dependent manner (Fig 6A). A response was seen at all concentrations of ligand with a robust response first observed at 10-7 M Dex. We propose that the response is occurring at all concentrations of hormone and is limited by the activity of the minimal GR promoter and assay. In Figure 6B we test the ability of the GR antagonist, RU486, to inhibit the conditioned media response. Treatment with 1µM Dex and 1µM RU486 for twenty-four hours suppressed the ability of SRF to potentiation the receptor, returning it to near normal Dex response levels. The results demonstrate a ligand-dependent, SRF- induced, potentiation of the GR.

To determine if HSF1, in the absence of stress, can produce a conditioned-media induced enhancement of the GR, we evaluated the constitutively-active mutant of HSF1, hHSF1-E189 (Wadekar et al., 2001). In

Figure 7A cells were pretreated with doxycycline (DOX) in a time-dependent fashion and DOX-conditioned media was collected and placed on GR responsive cells in the presence of 1µM Dex. DOX-conditioned media does not contain the ability to enhance the activity of the GR. The stress-conditioned media appears to require a stress event for the release of SRF into the media. However, hHSF1-E189 is a weak transactivator (Wadekar et al., 2001; Jones et al., 2004) and may simply produce too little SRF to elicit a GR response in the target cells.

To conclusively determine a role for HSF1 in the production and release of SRF 93 further experiments are needed to modulate the function of HSF1 in the absence

of stress. This can be addressed with the use of dominant-negative mutants of

HSF1, or through the use of HSF1-deficient cell lines and animals (Xiao et al.,

1999). A pharmacological approach may be used with drugs such as sodium

vanadate, quercetin, and wortmannin to modulate the activity of HSF1 (Li et al.,

1999; Li et al., 2000), or the stress can be increased in the cells following DOX

pretreatment.

Combined treatments of stress and DOX showed a greater ability to

enhance GR and HSF1 transactivity over stress conditions alone (Jones et al.,

2004). This in mind, we predicted the combined treatment of stress and DOX

would also increase the production and release of SRF into the conditioned

media. DOX pretreatment followed by stress exhibited an unexpected suppression of GR enhancement (Fig. 7B). However, this observation is consistent with HSP-induced preconditioned cells (Xia et al., 1999). We suggest

DOX treatment and the induction of hHSF1-E189 increased the amount of HSPs

(Jones et al., 2004) and generated a stress tolerant cell. We speculate stress tolerance decreased the release of the SRF, suppressing the conditioned-media induced enhancement of the GR.

Initial evaluations of SRF characteristics were consistent with a media soluble protein. In Figure 8A, centrifugation was performed on control- and stress-conditioned media to eliminate any particulate matter associated with cellular stress. SRF is media-soluble and is not part of the cellular debris. To determine the molecular mass (kDa) of the SRF, mass exclusion filtration

94 methods involving centrifugation were performed on both control- and stress-

conditioned media. These results suggested that SRF has a molecular mass

greater than 10kDa (data not shown), but were error prone due to obstruction or

clogging of the size exclusion filter system by the presence of serum in the cell

culture media. To determine the exact molecular mass of the SRF, column

chromatography will need to be performed in future experiments. Finally to

determine if SRF was a protein, we denatured the conditioned media with

proteinase K, a serine proteinase, and boiling (Fig. 8B). SRF can be denatured

by heat or combined treatment of proteinase K and heat. The combined

treatments did inactivate the factor in both control- and stress-conditioned media.

The decrease seen in the control-conditioned media can be attributed to the non- specific nature of proteinase K, digesting any proteins and growth factors present

in the culture media.

To this point, results have been hindered by the presence of serum in the

culture media. To determine if serum is required for the conditioned media-

induced GR enhancements, experiments were repeated under serum-free

conditions (Fig 9). Cells were cultivated and stressed in normal cell culture

media containing 10% bovine calf serum (BCS). Following treatment, cells were

allowed to recover in culture media with and without serum for 4 hours. After the

recovery period, the media was transferred to GR responsive unstressed cells

and allowed to recover under normal conditions. Results demonstrated that the

SRF was not present or active under serum-free conditions. Aliquots of

conditioned serum-free media (0%) were also supplemented with 1% BCS. The

95 addition of a small amount (1%) of serum to both the control- and stress-

conditioned media produced an increase in GR transactivity from the stress-

conditioned media but not the control-conditioned media. These results suggested SRF was present in the serum-free media but required serum for its activity. Follow-up experiments are necessary to support these results and need to include concentrating the serum-free conditioned media to detect activity and a proteomics approach for identification.

Based on these observations, we propose two mechanisms by which stress can enhance GR transactivity (Fig. 10). The classical GR model involves ligand-activated dissociation of the receptor heterocomplex and binding to the

GR responsive element (GRE). Previous work in our lab has demonstrated that stress, through HSF1, enhanced the activity of the GR inducing a robust increase in activity (Hu et al., 1996; Li et al., 1999; Li et al., 2000). Use of the constitutively-active mutant of HSF1 (Zuo et al., 1995), hHSF1-E189 (E189), also demonstrated an ability to increase GR activity, in the absence of stress (Jones et al., 2004) although, less than that achieved under stress conditions. This work presents evidence for a novel stress-sensitive factor, SRF, which, through cell culture media, enhances transactivity of the GR under stress conditions. We speculate that this enhancement may occur through a novel membrane-based receptor independent of HSF1 activation in the target cell.

96

MATERIALS AND METHODS

Cell Culture:

All cell lines were grown in an atmosphere of 5% CO2 at 37˚C in Dulbecco’s modified Eagles medium (DMEM) (Sigma Chemical Company, St. Louis MO) containing 10% iron-supplemented new born calf serum (BCS) (Hyclone). All experiments were performed on cells at or near confluence.

Conditioned Media:

At confluence, cells were divided into 2 groups: Donor and Host cells. Donor cells were treated with either heat shock (HS) of 43˚C for 2 hours (h), chemical shock (CS) of sodium arsenite (Sigma) for 2h followed by washing, or were subjected to 10µg/ml of doxycycline (DOX) (Sigma) for the desired time. Cells were not washed prior to stress treatment. Following treatment, all cells were washed 3 times, including the control cells with DMEM without serum and then allowed to recover for the desired time in DMEM supplemented with the appropriate amount of bovine calf serum (BCS). Following recovery, the conditioned media was harvested and pooled for distribution onto the host cells.

Donor-conditioned media was placed onto host cells following washing 3 times with nonsupplemented DMEM. Host cells bathed in donor media were then treated with desired concentrations of dexamethasone and allowed to recover under normal growth conditions for 24h prior to harvest.

97 CAT Assay:

Measurement of CAT enzyme activity was performed according to the modified method of Nordeen et al (Nordeen et al., 1987) with minor modifications. Briefly, cell lysates are prepared by sequential freezing and thawing in a hypotonic buffer, 25M Tris, 5mM EDTA (pH 7.5), and centrifuged at 8,000xg. Aliquots of lysate containing equal amount of protein are added to an enzyme reaction mixture containing acetyl-CoA synthase (Sigma), [3H]-sodium acetate (ICN), coenzyme A (CoA) (Sigma) and ATP (Sigma). Radioactively-labeled acetyl coenzyme A is first generated enzymatically from CoA and labeled acetate through preincubation. Acetylation of chloramphenicol is then initiated by adding cell lysate containing CAT enzyme. The reaction is stopped with cold benzene and 75% of the organic phase is counted.

98 FIGURE LEGENDS

Fig 1 - Stress-Induced Potentiation of the Glucocorticoid Receptor. LGRE-E189 cells were subject to control (Con), 1µM Dex (d), heat shock (HS) at 43˚C for 2

hours (h), or chemical shock (CS) using sodium arsenite for 2h followed by

washing. Cells were the allowed to grow under non-stress conditions in the

presence of 1µM Dex (d) for 24 hours. Lysates were prepared and analyzed for

CAT activity. The results represent the mean average +/- of 6 individual

experiments.

Fig 2 - Generation of Stress and Control Conditioned Media. At confluence cells

were divided into 2 groups, Donor and Host cells. Donor cells were treated with

either, heat shock (HS) of 43˚C for 2 hours (h), Chemical shock (CS) of sodium

arsenite for 2h followed by washing, or were subjected to 10µg/ml DOX for the

desired time. Following treatment, media was harvested from these flasks

generating donor-conditioned media. Donor conditioned media was then placed

on the host cells and left to recover in the presence and absence of Dex for 16h.

Lysates were prepared and analyzed.

Fig 3 - Transfection of hHSF1-E189, a Constitutively-active Mutant of HSF1, into

GR- and HSF1-responsive backgrounds. A, HSF1 is a 529 aa protein

incorporating multiple leucine zipper regions (LZ), DNA-binding domain (DBD)

and a C-terminal transactivation domain (CTR). A point mutation at residue 189 99 introduced a collapse of LZ2 to produce a constitutively-active form of HSF1

(E189). B, Mouse L929 cells stably selected for the p2500-CAT (Hsp70) or the minimal pGRE2E1B-CAT promoters were used as parent cells for the construction of stable cell lines expressing E189. C, E189 was placed in the doxycycline-inducible bi-directional vector PBI-EGFP. Parent cells were co- transfected with pBI-EGFP-E189 and the transactivator vector pUHD172-1hygro, followed by hygromycin selection and cloning of GFP-positive cells to generate the LHSE-E189 and LGRE-E189 cell lines.

Fig 4 - Time-Dependent Release of a Stress-Induced Steroid Receptor

Transactivation Enhancing Factor. LGRE-E189 Cells were subjected to control conditions or stress treatments of heat shock (HS) of 43˚C (Panel A) or chemical shock (CS) using 200µM sodium arsenite for 2 hours (h) (Panel B). Media conditioned for the time intervals indicated, from HS, CS, or control cells, was transferred to identical untreated LGRE cells. The cells were allowed to grow under non-stress conditions in the presence of 1µM Dex for 24h. Lysates were prepared and analyzed for relative CAT activity. The results represent the mean average SEM +/- of nine individual experiments.

Fig 5 - Effect of Stress- and Time-Dependent Conditioned Media on HSF1

Activation. LGRE-E189 cells were subjected to control (Con), heat shock (HS) of

43˚C, or chemical shock (CS) of 200µM sodium arsenite for 2 hours (h).

Following stress treatment, cells were recovered for the indicated time and the 100 conditioned media was transferred to identical LHSE cells. Experimental controls of LHSE cells were evaluated under identical conditions. Following media transfer or stress treatment, cells were allowed to grow under non-stress conditions for 24h. Lysates were prepared and analyzed for CAT activity. The results represent the mean average SEM +/- of three individual experiments.

Fig 6 - Effect of SRF on Ligand-Dependent GR Transactivity. In panel A, LGRE cells were subjected to Control (Con), or chemical shock (CS) of 200µM sodium arsenite for 2h followed by washing. The CS cells were allowed to recover for 4 and 20 hours respectfully. Media from the 3 groups was directly transferred to adjacent identical untreated LGRE cells and treated with Dex at indicated concentration. Lysates were prepared and analyzed for CAT activity. The results represent the mean +/- SEM of four individual experiments. In panel B,

LGRE cells were subjected to Control (Con), or chemical shock (CS) of 200µM sodium arsenite for 2h followed by washing. The CS cells were allowed to recover for 4. Media from the control- and CS-conditioned cells was directly transferred to adjacent identical untreated LGRE cells and left untreated (Con), treated with the GR antagonist RU486 (1µM), or treated with a 1:1 ratio (1µM) of

Dex and RU486 for 16 hours. Lysates were prepared and analyzed for CAT activity. The results represent the mean +/- SEM of 3 individual experiments.

Fig. 7 - Effect of Constitutively-active hHSF1-E189 Conditioned Media and stress on GR Transactivity. In panel A., LGRE-E189 cells were subjected to 10µg/ml 101 DOX for the indicated time. Following DOX treatment the media was transferred

to identical, untreated, LGRE-E189 cells and treated with 1µM Dex.

Experimental control LGRE-E189 cells were subjected to control (Con), 1µM

Dex, heat shock (HS) at 43˚C for 2h, chemical shock (CS) of 200µM sodium arsenite for 2h. Following media transfer or stress treatment the Dex treated cells were allowed to grow under non-stress conditions for 24h in the presence of

1µM Dex. Lysates were prepared and analyzed for CAT activity. The results represent the mean SEM +/- of six individual experiments. In panel B., LGRE-

E189 cells were subject to control media (CM), 10µg/ml DOX for the indicated

time, a combined treatment of Dox followed by chemical shock (CS) of 200µM

sodium arsenite for 2h, or CS media alone (CSM). Following stress treatment

the cells were washed and allowed to recover for 4h. Following recovery the

media from the Con, DOX treated, and DOX/CSM cells was transferred to

adjacent untreated LGRE cells and allowed to grow under non-stress conditions

in the presence of 1µM Dex for 24h. Lysates were made and analyzed for CAT

activity. The results represent the mean average SEM +/- of three individual

experiments.

Fig 8 - Stress-Released Factor (SRF) is Media soluble and can be Heat

Denatured. In panel A., LGRE-E189 cells were subjected to control (C) or chemical shock (CS) of 200µM sodium arsenite for 2h followed by washing.

Media (M) from control and CS cells was then transferred to adjacent untreated

102 LGRE cells by direct transfer or following 100Kxg centrifugation (100Kxg). All cells were then treated with 1µM Dex and allowed to grow under non-stress conditions for 24h. Lysates were made and analyzed for CAT activity. The results represent the mean average SEM +/- of three individual experiments. In panel B., LGRE cells were subjected to control or chemical shock (CS) of 200µM sodium arsenite for 2h followed by washing. The cells were allowed to recover for 4h generating conditioned media. The media from the cells was then left as a control, heat inactivated at 100˚C for 15 min (100˚C), or treated with 250 µg/ml proteinase K (K) for 30 minutes followed by heat inactivation at 100˚C for 15 min

(100˚C). All media was then centrifuged at 100KxG for 30 min. The media was then transferred to adjacent identical untreated LGRE cells and treated with 1µM

Dex for 24h. Lysates were prepared and analyzed for CAT. The results represent the mean +/- SEM of six individual experiments.

Fig 9 - Serum Free Enhancement of GR Transactivity by the SRF. LGRE cells grown in normal cell culture media supplemented with 10% BCS and subjected to control or chemical shock (CS) of 200µM sodium arsenite for 2h followed by washing. The cells were allowed to recover for 4h in serum concentrations of 0% or 10% bovine calf serum (BCS). Following recovery the media was directly transferred to adjacent identical untreated LGRE cells in the 0 and 10% treatment groups. Aliquots of media from the 0% group were supplemented with

1% BCS to generate the 1% serum and again transferred to adjacent identical untreated LGRE cells. Cells were allowed to recover in the presence of 1µM Dex 103 for 24h. Lysate were prepared and analyzed for CAT activity. The results represent the mean +/- SEM of 12 individual experiments.

Fig 10 - Model for SRF and HSF1 Regulation of GR Transactivity. The classical model of glucocorticoid receptor (GR) activation involves the diffusion of the ligand across the plasma membrane followed by binding to the receptor. Binding of the ligand eventually leads to a dissociation of the GR associated heterocomplex and translocation to the nucleus followed by binding of the GR to the responsive element (GRE) producing GR regulated genes. Stress in the presence of ligand has the ability to enhance the activity of the GR through the heat shock factor (HSF1) signaling pathway, inducing a robust increase in GR activity we term the heat shock potentiation effect (HSPE). Genetic manipulation of HSF1 provided a constitutively active mutant of HSF1, E189. E189 produced, in the absence of stress, an HSPE albeit to a lesser extent than that achieved under stress conditions. Following stress, a factor (SRF) is released into the media, and we suggest, through interaction with a novel membrane based receptor enhances the GR.

104 FIGURE 1.

105 FIGURE 2.

106 FIGURE 3.

107 FIGURE 4.

108 FIGURE 5.

109 FIGURE 6.

110 FIGURE 7.

111 FIGURE 8.

112 FIGURE 9.

113 FIGURE 10.

114 SUMMARY

The work in our lab has demonstrated that stress events can enhance the

transcriptional activity of the glucocorticoid receptor. We call this phenomenon

the heat shock potentiation effect (HSPE). HSF1 activity has been shown to

function in the HSPE and positively affect transactivity of the GR while the GR

was shown to feedback and suppress HSF1 by releasing it from the chromatin

(Sanchez, 1992; Sanchez et al., 1994; Hu et al., 1996; Li et al., 1999; Li et al.,

2000; Wadekar et al., 2001; Wadekar et al., 2004). Work presented here

expands the knowledge of stress signaling to include the nonstress-induced activation of HSF1 as a mechanism of GR potentiation and also presents a novel stress released factor (SRF), found in cell culture media that enhances GR transactivity.

Work by Dapei Li presented evidence for HSF1 involvement in the HSPE

through pharmacological modulation of HSF1. He positively modulated the

function of HSF1 with the phosphatidylinositol-3 kinase inhibitor, wortmannin, (Li et al., 2000) and negatively modulated the function with sodium vanadate and quercetin (Li et al., 2000; Li et al., 1999). In all cases, the modulation of HSF1 activity directly correlated with the potentiation of the GR.

Suppression of HSF1 activity by the GR was presented in the work of

Subhagya Wadekar (Wadekar et al., 2001; Wadekar et al., 2004). She

demonstrated negative modulation of both the endogenous HSF1 and a

constitutively-active mutant of human HSF1, hHSF1-E189 by the ligand-activated

GR. The GR suppression of HSF1 activity was lost in the presence of the GR

115 antagonist, RU486. RU486 results were supported with ligand- and DNA-binding deficient mutants of the GR (Wadekar et al., 2001) and demonstrate the GR- specific suppression of HSF1 activity is through the inhibition of HSF1 chromatin binding (Wadekar et al., 2004).

Results however, did not resolve the exact mechanism of the GR potentiation. Observations demonstrate HSF1 could crosstalk with the GR, but did not provide evidence if HSF1 per se was sufficient to produce a potentiation of the GR in the absence of stress. Work here demonstrates the constitutively- active mutant of HSF1, hHSF1-E189, is sufficient to induce a potentiation of the

GR, in the absence of stress. Results from immunoblots, electromobility shift assays (EMSA), and chromatin immunoprecipitation (ChIP) analysis demonstrated that hHSF1-E189 functioned in a manner consistent with that observed from endogenous HSF1. Further, induction of hHSF1-E189, in the presence of saturating levels of hormone, produced a potentiation of the GR.

The potentiation occurred in the absence of stress and in a time frame consistent with stress treatment. These results suggest a downstream product(s) of HSF1 activation is required for the increase in GR transactivation to occur.

This suggests HSF1 product(s) have the ability to function outside the normal realm associated with traditional stress induced HSPs and provide the potential for HSF1-specific genes to interact with the GR through four possible mechanisms. These mechanisms include, but are not limited to HSF1 product(s)

1) functioning like a coactivator or a protein kinase, 2) releasing a GR repressor molecule, 3) interacting directly with the receptor, and finally, 4) interacting with

116 the transcription machinery. In any case, the activation of HSF1, in the absence

of stress, induced a potentiation of the GR. Work still did not elucidate what the

HSF1-induced gene product may be. One approach to answer this may be to

evaluate the HSF1-induced gene products through gene or protein chip arrays.

Research had demonstrated the ability of HSF1 to communicate to the GR

and produce the HSPE. A question that had been discussed in this laboratory

was, can stress communicate to adjacent unstressed cells through cell culture

media? To look at this possible mechanism of potentiation, we generated a media exchange protocol. This protocol provided evidence of a stress-released factor (SRF) that enhanced GR transactivity in non-stressed cells, through the cell culture media. The enhancement did not occur through the activation of endogenous HSF1 in the target cells, thus presents a novel stress signaling mechanism.

Stressed-conditioned media was able to enhance the GR in a ligand- and

concentration-dependent manner. Non-stress-induced human HSF1-E189-

conditioned media was unable to elicit the same response. Human HSF1-E189

however, is a weak transactivator (Jones et al., 2004) and may produce

insufficient SRF to induce a GR response in the target cells. SRF was not part of

the normal- or stress-induced cellular debris and was inactivated by heat and

proteinase K cleavage. Analysis of SRF activity in depleted media indicated the

factor was not able to function under serum-free conditions. These results reflect

that SRF is a novel stress-release protein that is media soluble and functions to

117 enhance the activity of the GR in the target cells through an HSF1-independent

pathway.

Our work suggests that stress-induced potentiation of the GR may be the

result of two stress-induced events. First, stress activates HSF1 and produces

an enhancement of the GR approximately sixteen to twenty hours after the stress

treatment. Second, as an early response to stress, SRF is released into the cell

culture media and signals to adjacent cells to increase GR transactivation. The

SRF-induced potentiation does not occur immediately following the stress, but

occurs after four hours of recovery and can be maintained for up to 20 hours.

The two-step model presented evidence as to how stress and HSF1 modulate their own suppression in a GR-specific manner. First, stress activates

HSF1 through the recruitment of HSPs away from the inactive HSF1 monomer.

Active HSF1 induces HSP production, alleviating the stress, and through protein production, potentiates the GR. Secondly, stress induces the release of SRF into the cell culture media, which has the ability to modulate and enhance GR transactivity. Active GR can also modulates and suppress the activation of

HSF1. The increase in GR transactivity, HSP production and the activation of

HSF1 regulate the overall stress response. This is done through three mechanisms: 1) through GR-mediated gene products, 2) through the suppression of HSF1 activity, and 3), through SRF-induced GR enhancement.

The major stress activated pathways, HSF1 and the GR, appear to function in close communication with each other through multiple crosstalk mechanisms. HSF1 protects the cells from the perceived stress event through

118 HSP production while the GR protects the cell from an overshoot of the stress response and suppresses HSF1. Stress also initiates the release of SRF into the cell culture media. We suggest the function of SRF is to warn the adjacent cells of an impending stress and induce the formation of a stress containment system through the ligand-dependent activation of the GR. The result is the containment of the stress to a limited area and the formation of a “perimeter” by the GR signal transduction pathways.

119 BIBLIOGRAPHY

Abravaya, K.; Myers, M. P.; Murphy, S. P. and Morimoto, R. I. 1992 The

human heat shock protein hsp70 interacts with HSF, the transcription factor that

regulates heat shock gene expression. Genes Dev., 6:1153-1164.

Adcock, I. M. 2000 Molecular mechanisms of glucocorticosteroid actions.

Pulm.Pharmacol.Ther., 13:115-126.

Akahoshi, T.; Oppenheim, J. J. and Matsushima, K. 1988 Induction of high-affinity interleukin 1 receptor on human peripheral blood lymphocytes by glucocorticoid hormones. J. Exp. Med., 167:924-936.

Alder, G. M.; Austen, B. M.; Bashford, C. L.; Mehlert, A. and Pasternak, C.

A. 1990 Heat shock proteins induce pores in membranes. Biosci. Rep., 10:509-

518.

Allgood, V. E.; Oakley, R. H. and Cidlowski, J. A. 1993 Modulation by vitamin B6 of glucocorticoid receptor-mediated gene expression requires transcription factors in addition to the glucocortcoid receptor J. Biol. Chem.,

268:20870-20876.

Alroy, I. and Freedman, L. P. 1992 DNA binding analysis of glucocorticoid

receptor specificity mutants. Nucleic Acids Res., 20:1045-1052.

Anathan, J. A.; Goldberg, A. L. and Voellmy, R. 1986 Abnormal proteins

serve as eukaryotic stress signals and trigger the activation of heat shock genes

Science, 232:252-254.

Anderson, R. L.; Kraft, P. E.; Bensaude, O. and Hahn, G. M. 1991 Binding

activity of glucocorticoid receptors after heat shock. Exp. Cell Res., 197:100-106.

120 Aranda, A. and Pascual, A. 2001 Nuclear hormone receptors and gene

expression. Physiol. Rev., 81:1269-1304.

Arispe, N. and De, M. A. 2000 ATP and ADP modulate a cation channel formed by Hsc70 in acidic phospholipid membranes. J. Biol. Chem., 275:30839-

30843.

Arispe, N.; Doh, M. and De, M. A. 2002 Lipid interaction differentiates the constitutive and stress-induced heat shock proteins Hsc70 and Hsp70. Cell

Stress Chaperones, 7:330-338.

Asea, A.; Kraeft, S. K.; Kurt-Jones, E. A.; Stevenson, M. A.; Chen, L. B.;

Finberg, R. W.; Koo, G. C. and Calderwood, S. K. 2000 HSP70 stimulates cytokine production through a CD14-dependant pathway, demonstrating its dual role as a chaperone and cytokine Nat. Med., 6:435-442.

Asea, A.; Rehli, M.; Kabingu, E.; Boch, J. A.; Bare, O.; Auron, P. E.;

Stevenson, M. A. and Calderwood, S. K. 2002 Novel signal transduction pathway utilized by extracellular HSP70: role of toll-like receptor (TLR) 2 and TLR4. J.

Biol. Chem., 277:15028-15034.

Baeuerle, P. A. and Baltimore, D. 1988 I kappa B: a specific inhibitor of

the NF-kappa B transcription factor. Science, 242:540-546.

Baeuerle, P. A. and Baltimore, D. 1989 A 65-kappaD subunit of active NF-

kappaB is required for inhibition of NF-kappaB by I kappaB. Genes Dev., 3:1689-

1698.

121 Baler, R.; Dahl, G. and Voellmy, R. 1993 Activation of human heat shock

genes is accompanied by oligomerization, modification, and rapid translocation of

heat shock transcription factor HSF1. Mol. Cell. Biol., 13:2486-2496.

Baler, R.; Zou, J. and Voellmy, R. 1996 Evidence for a role of Hsp70 in the regulation of the heat shock response in mammalian cells Cell Stress

Chaperones, 1:33-39.

Bamberger, C. M.; Bamberger, A. M.; de, C. M. and Chrousos, G. P. 1995

Glucocorticoid receptor beta, a potential endogenous inhibitor of glucocorticoid

action in humans. J. Clin. Invest., 95:2435-2441.

Baniahmad, A.; Kohne, A. C. and Renkawitz, R. 1992 A transferable silencing domain is present in the thyroid hormone receptor, in the v-erbA oncogene product and in the retinoic acid receptor. EMBO J., 11:1015-1023.

Baniahmad, A.; Leng, X.; Burris, T. P.; Tsai, S. Y.; Tsai, M. J. and

O'Malley, B. W. 1995 The tau 4 activation domain of the thyroid hormone receptor is required for release of a putative corepressor(s) necessary for transcriptional silencing. Mol. Cell. Biol., 15:76-86.

Baumann, H.; Paulsen, K.; Kovacs, H.; Berglund, H.; Wright, A. P.;

Gustafsson, J. A. and Hard, T. 1993 Refined solution structure of the

glucocorticoid receptor DNA-binding domain. Biochemistry, 32:13463-13471.

Beato, M. 1989 Gene regulation by steroid hormones Cell 56, 335-344.

Beato, M.; Herrlich, P. and Schutz, G. 1995 Steroid hormone receptors: many actors in search of a plot. Cell, 83:851-857.

122 Becker, J. and Craig, E. A. 1994 Heat-shock proteins as molecular

chaperones. Eur. J. Biochem., 219:11-23.

Bevan, C. L.; Hoare, S.; Claessens, F.; Heery, D. M. and Parker, M. G.

1999 The AF1 and AF2 domains of the androgen receptor interact with distinct

regions of SRC1. Mol. Cell. Biol., 19:8383-8392.

Bharadwaj, S.; Ali, A. and Ovsenek, N. 1999 Multiple components of the

HSP90 chaperone complex function in regulation of heat shock factor 1 In vivo

Mol. Cell. Biol., 19:8033-8041.

Bhatnagar, S. and Vining, C. 2003 Facilitation of hypothalamic-pituitary- adrenal responses to novel stress following repeated social stress using the resident/intruder paradigm. Horm. Behav., 43:58-165.

Bledsoe, R. K.; Montana, V. G.; Stanley, T. B.; Delves, C. J.; Apolito, C. J.;

McKee, D. D.; Consler, T. G.; Parks, D. J.; Stewart, E. L.; Willson, T. M.;

Lambert, M. H.; Moore, J. T.; Pearce, K. H. and Xu, H. E. 2002 Crystal structure of the glucocorticoid receptor ligand binding domain reveals a novel mode of receptor dimerization and coactivator recognition. Cell, 110:93-105.

Bodwell, J. E.; Orti, E.; Coull, J. M.; Pappin, D. J.; Smith, L. I. and Swift, F.

1991 Identification of phosphorylated sites in the mouse glucocorticoid receptor.

J. Biol. Chem., 266:7549-7555.

Bodwell, J. E.; Webster, J. C.; Jewell, C. M.; Cidlowski, J. A.; Hu, J. M. and Munck, A. 1998 Glucocorticoid receptor phosphorylation: overview, function and cell cycle-dependence. J. Steroid Biochem. Mol. Biol., 65:91-9.

123 Bohen, S. P. 1995 Hsp90 mutants disrupt glucocorticoid receptor ligand

binding and destabilize aporeceptor complexes J. Biol. Chem., 270:29433-

29438.

Bramlett, K. S. and Burris, T. P. 2002 Effects of selective estrogen receptor modulators (SERMs) on coactivator nuclear receptor (NR) box binding to estrogen receptors. Mol. Genet. Metab., 76:225-233.

Bresnick, E. H.; Sanchez, E. R. and Pratt, W. B. 1988 Relationship between glucocorticoid receptor steroid-binding capacity and association of the

Mr 90,000 heat shock protein with the unliganded receptor. J. Steroid Biochem.,

30:267-269.

Brown, K.; Gerstberger, S.; Carlson, L.; Franzoso, G. and Siebenlist, U.

1995 Control of I kappa B-alpha proteolysis by site-specific, signal-induced

phosphorylation. Science, 267:1485-1488.

Burnstein, K. L.; Jewell, C. M.; Sar, M. and Cidlowski, J. A. 1994

Intragenic sequences of the human glucocorticoid receptor complementary DNA mediate hormone-inducible receptor messenger RNA down-regulation through multiple mechanisms. Mol. Endocrinol., 8:1764-1773.

Caldenhoven, E.; Liden, J.; Wissink, S.; Van de Stolpe, A.; Raaijimakers,

J.; Koenderman, L.; Okret, S.; Gustafsson, J.-A. and Van der Saag, P. T. 1995

Negative cross-talk between Rel A and the glucocorticoid receptor: a possible mechanism for the anti-inflammatory action of glucocorticoids Molecular

Endocrinology, 9:401-412.

124 Carson-Jurica, M. A.; Schrader, W. T. and O'Malley, B. W. 1990 Steroid

receptor family: structure and functions. Endocr. Rev., 11:201-220.

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, 90:569-80.

Chen, J. D. and Evans, R. M. 1995 A transcriptional co-repressor that interacts with nuclear hormone receptors Nature, 377:454-47.

Cheung-Flynn, J.; Roberts, P. J.; Riggs, D. L. and Smith, D. F. 2003 C- terminal sequences outside the tetratricopeptide repeat domain of FKBP51 and

FKBP52 cause differential binding to Hsp90. J. Biol. Chem. 278:17388-17394.

Chu, B.; Soncin, F.; Price, B. D.; Stevenson, M. A. and Calderwood, S. K.

1996 Sequential phosphorylation by mitogen-activated protein kinase and glycogen synthase kinase 3 represses transcriptional activation by heat shock factor-1 J. Biol. Chem., 271:30847-30857.

Chu, B.; Zhong, R.; Soncin, F.; Stevenson, M. A. and Calderwood, S. K.

1998 Transcriptional activity of heat shock factor 1 at 37 degrees C is repressed through phosphorylation on two distinct serine residues by glycogen synthase kinase 3 and protein kinases Calpha and Czeta. J. Biol. Chem. 273:18640-

18646.

Cleasby, M. E.; Kelly, P. A.; Walker, B. R. and Seckl, J. R. 2003

Programming of rat muscle and fat metabolism by in utero overexposure to

glucocorticoids. Endocrinology, 144:999-1007.

125 Collingwood, T. N.; Urnov, F. D. and Wolffe, A. P. 1999 Nuclear receptors:

coactivators, corepressors and chromatin remodeling in the control of

transcription. J. Mol. Endocrinol., 23:255-275.

Cotto, J. J.; Kline, M. and Morimoto, R. I. 1996 Activation of heat shock

factor 1 DNA binding precedes stress-induced serine phosphorylation. Evidence

for a multistep pathway of regulation. J. Biol. Chem., 271:3355-3358.

Craig, E. A. and Gross, C. A. 1991 Is hsp70 the cellular thermometer?

Trends Biochem. Sci., 16:135-140.

Cunningham, T. J.; Hodge, L.; Speicher, D.; Reim, D.; Tyler-Polsz, C.;

Levitt, P.; Eagleson, K.; Kennedy, S. and Wang, Y. 1998 Identification of a

survival-promoting peptide in medium conditioned by oxidatively stressed cell

lines of nervous system origin. J. Neurosci., 18:7047-7060.

Czar, M. J.; Lyons, R. H.; Welsh, M. J.; Renoir, J. M. and Pratt, W. B.

1995 Evidence that the FK506-binding immunophilin heat shock protein 56 is

required for trafficking of the glucocorticoid receptor from the cytoplasm to the

nucleus. Mol. Endocrinol., 9:1549-1560.

Dahlman-Wright, K.; Siltala-Roos, H.; Carlstedt-Duke, J. and Gustafsson,

J. A. 1990 Protein-protein interactions facilitate DNA binding by the glucocorticoid receptor DNA-binding domain. J. Biol. Chem., 265:14030-14035.

Dahlman-Wright, K.; Wright, A.; Gustafsson, J. A. and Carlstedt-Duke, J.

1991 Interaction of the glucocorticoid receptor DNA-binding domain with DNA as

a dimer is mediated by a short segment of five amino acids. J. Biol. Chem.,

266:3107-3112.

126 Dai, R.; Frejtag, W.; He, B.; Zhang, Y. and Mivechi, N. F. 2000 c-Jun NH2-

terminal kinase targeting and phosphorylation of heat shock factor-1 suppress its

transcriptional activity. J. Biol. Chem., 275:18210-1828.

Danielsen, M.; Northrop, J. P.; Jonklaas, J. and Ringold, G. M. 1987

Domains of the glucocorticoid receptor involved in specific and nonspecific

deoxyribonucleic acid binding, hormone activation, and transcriptional

enhancement. Mol. Endocrin., 1:816-822.

Danielsen, M.; Northrop, J. P. and Ringold, G. M. 1986 The mouse glucocorticoid receptor: mapping of functional domains by cloning, sequencing and expression of wild-type and mutant receptor proteins. Embo. J., 5:2513-

2522.

Darimont, B. D.; Wagner, R. L.; Apriletti, J. W.; Stallcup, M. R.; Kushner,

P. J.; Baxter, J. D.; Fletterick, R. J. and Yamamoto, K. R. 1998 Structure and

specificity of nuclear receptor-coactivator interactions. Genes Dev., 12:3343-

3356.

Davies, A. O. and Lefkowitz, R. J. 1984 Regulation of beta-adrenergic

receptors by steroid hormones. Annu. Rev. Physiol., 46:119-130.

Davies, T. H.; Ning, Y. M. and Sanchez, E. R. 2002 A new first step in activation of steroid receptors: hormone-induced switching of FKBP51 and

FKBP52 immunophilins. J. Biol. Chem., 277:4597-4600.

Deroo, B. J. and Archer, T. K. 2001 Glucocorticoid receptor-mediated

chromatin remodeling in vivo. Oncogene, 20:3039-3046.

127 DiDomenico, B. J.; Bugaisky, G. E. and Lindquist, S. 1982 The heat shock response is self-regulated at both the transcriptional and posttranscriptional levels. Cell, 31:593-603.

Ding, X. F.; Anderson, C. M.; Ma, H.; Hong, H.; Uht, R. M.; Kushner, P. J. and Stallcup, M. R. 1998 Nuclear receptor-binding sites of coactivators glucocorticoid receptor interacting protein 1 (GRIP1) and steroid receptor coactivator 1 (SRC-1): multiple motifs with different binding specificities. Mol.

Endocrinol., 12:302-313.

Direskeneli, H. and Saruhan-Direskeneli, G. 2003 The role of heat shock proteins in Behcet's disease. Clin. Exp. Rheumatol., 21:S44-8.

Dittmar, K. D. and Pratt, W. B. 1997 Folding of the glucocorticoid receptor by the reconstituted Hsp90-based chaperone machinery. The initial hsp90.p60.hsp70-dependent step is sufficient for creating the steroid binding conformation. J. Biol. Chem., 272:13047-13054.

Drouin, J.; Charron, J.; Gagner, J. P.; Jeannotte, L.; Nemer, M.; Plante, R.

K. and Wrange, O. 1987 Pro-opiomelanocortin gene: a model for negative regulation of transcription by glucocorticoids. J. Cell Biochem., 35:293-304.

Drouin, J.; Trifiro, M. A.; Plante, R. K.; Nemer, M.; Eriksson, P. and

Wrange, O. 1989 Glucocorticoid receptor binding to a specific DNA sequence is required for hormone-dependent repression of pro-opiomelanocortin gene transcription. Mol. Cell. Biol., 9:5305-5314.

Edwards, D. P.; Estes, P. A.; Fadok, V. A.; Bona, B. J.; Onate, S.;

Nordeen, S. K. and Welch, W. J. 1992 Heat shock alters the compostion of

128 heteromeric steroid receptor complexes and enhances receptor activity in vivo.

Biochemistry, 31:2482-2491.

Engman, D. M.; Dragon, E. A. and Donelson, J. E. 1990 Human humoral

immunity to hsp70 during Trypanosoma cruzi infection. J. Immunol., 144:3987-

3991.

Evans, R. M. 1988 The steroid and thyroid hormone receptor superfamily.

Science, 240:889-895.

Exton, J. H.; Friedmann, N.; Wong, E. H.; Brineaux, J. P.; Corbin, J. D.

and Park, C. R. 1972 Interaction of glucocorticoids with glucagon and

epinephrine in the control of gluconeogenesis and glycogenolysis in liver and of lipolysis in adipose tissue. J. Biol. Chem., 247:3579-3588.

Farman, N. and Rafestin-Oblin, M. E. 2001 Multiple aspects of mineralocorticoid selectivity. Am. J. Physiol. Renal Physiol., 280:F181-92.

Fernandes, M.; Xiao, H. and Lis, J. T. 1994 Fine structure analyses of the

Drosophila and Saccharomyces heat shock factor--heat shock element

interactions. Nucleic Acids Res., 22:167-173.

Fernandes, M.; Xiao, H. and Lis, J. T. 1995 Binding of heat shock factor to and transcriptional activation of heat shock genes in Drosophila. Nucleic Acids

Res., 23:4799-4804.

Fiorenza, M. T.; Farkas, T.; Dissing, M.; Kolding, D. and Zimarino, V. 1995

Complex expression of murine heat shock transcription factors. Nucleic Acids

Res., 23:467-474.

129 Fisher, G. A.; Anderson, R. L. and Hahn, G. M. 1986 Glucocorticoid-

induced heat resistance in mammalian cells J. Cell. Phys., 128:127-132.

Freedman, L. P. and Luisi, B. F. 1993 On the mechanism of DNA binding by nuclear hormone receptors: A strutural and functional perspective. J. Cell.

Biochem., 51:140-150.

Fujihara, S. M. and Nadler, S. G. 1999 Intranuclear targeted delivery of functional NF-kappaB by 70 kDa heat shock protein. EMBO J., 18:411-419.

Gametchu, B. and Harrison, R. W. 1984 Characterization of a monoclonal antibody to the rat liver glucocorticoid receptor. Endocrinology, 114:274-279.

Gehin, M.; Mark, M.; Dennefeld, C.; Dierich, A.; Gronemeyer, H. and

Chambon, P. 2002 The function of TIF2/GRIP1 in mouse reproduction is distinct

from those of SRC-1 and p/CIP. Mol. Cell. Biol., 22:5923-5937.

Gerber, D. A.; Souquere-Besse, S.; Puvion, F.; Dubois, M. F.; Bensaude,

O. and Cochet, C. 2000 Heat-induced relocalization of protein kinase CK2.

Implication of CK2 in the context of cellular stress. J. Biol. Chem., 275:23919-

23926.

Giguere, V.; Hollenberg, S. M.; Rosenfeld, M. G. and Evans, R. M. 1986

Functional domains of the human glucocorticoid receptor. Cell, 46:645-652.

Godowski, P. J.; Rusconi, S.; Miesfeld, R. and Yamamoto, K. R. 1987

Glucocorticoid receptor mutants that are constitutive activators of transcriptional enhancement. Nature, 325:365-368.

Goff, S. A. and Goldberg, A. L. 1985 Production of abnormal proteins in E. coli stimulates transcription of lon and other heat shock genes. Cell, 41:587-95.

130 Gossen, M.; Freundlieb, S.; Bender, G.; Muller, G.; Hillen, W. and Bujard,

H. 1995 Transcriptional activation by tetracyclines in mammalian cells. Science,

268:1766-1769.

Green, M.; Schuetz, T. J.; Sullivan, E. K. and Kingston, R. E. 1995 A heat

shock-responsive domain of human HSF1 that regulates transcription activation

domain function. Mol. Cell. Biol., 15:3354-3362.

Green, S.; Kumar, V.; Theulaz, I.; Wahli, W. and Chambon, P. 1988 The

N-terminal DNA-binding 'zinc finger' of the oestrogen and glucocorticoid receptors determines target gene specificity. EMBO J., 7:3037-3044.

Guo, Y.; Guettouche, T.; Fenna, M.; Boellmann, F.; Pratt, W. B.; Toft, D.

O.; Smith, D. F. and Voellmy, R. 2001 Evidence for a mechanism of repression of heat shock factor 1 transcriptional activity by a multichaperone complex. J. Biol.

Chem., 276:45791-45799.

Guzhova, I. V.; Arnholdt, A. C.; Darieva, Z. A.; Kinev, A. V.; Lasunskaia, E.

B.; Nilsson, K.; Bozhkov, V. M.; Voronin, A. P. and Margulis, B. A. 1998 Effects of

exogenous stress protein 70 on the functional properties of human promonocytes

through binding to cell surface and internalization. Cell Stress Chaperones, 3:67-

77.

Guzhova, I.; Kislyakova, K.; Moskaliova, O.; Fridlanskaya, I.; Tytell, M.;

Cheetham, M. and Margulis, B. 2001 In vitro studies show that Hsp70 can be

released by glia and that exogenous Hsp70 can enhance neuronal stress

tolerance. Brain Res., 914:66-73.

131 Harrison, C. J.; Bohm, A. A. and Nelson, H. C. 1994 Crystal structure of

the DNA binding domain of the heat shock transcription factor. Science, 263:224-

227.

Heery, D. M.; Kalkhoven, E.; Hoare, S. and Parker, M. G. 1997 A signature motif in transcriptional co-activators mediates binding to nuclear receptors. Nature, 387:733-76.

Heinzel, T.; Lavinsky, R. M.; Mullen, T. M.; Soderstrom, M.; Laherty, C. D.;

Torchia, J.; Yang, W. M.; Brard, G.; Ngo, S. D.; Davie, J. R.; Seto, E.; Eisenman,

R. N.; Rose, D. W.; Glass, C. K. and Rosenfeld, M. G. 1997 A complex containing N-CoR, mSin3 and histone deacetylase mediates transcriptional repression. Nature, 387:43-48.

Hightower, L. 1980 Cultured animal cells exposed to amino acid analogues or puromycin rapidly synthesize several polypeptides. J. Cell. Phys.,

102:407-427.

Hightower, L. E. and Guidon, P. T. J. 1989 Selective release from cultured

mammalian cells of heat-shock (stress) proteins that resemble glia-axon transfer

proteins. J. Cell. Physiol., 138:257-266.

Hollenberg, S. M. and Evans, R. M. 1988 Multiple and cooperative trans- activation domains of the human glucocorticoid receptor. Cell, 55:899-906.

Holmberg, C. I.; Hietakangas, V.; Mikhailov, A.; Rantanen, J. O.; Kallio,

M.; Meinander, A.; Hellman, J.; Morrice, N.; MacKintosh, C.; Morimoto, R. I.;

Eriksson, J. E. and Sistonen, L. 2001 Phosphorylation of serine 230 promotes inducible transcriptional activity of heat shock factor 1. EMBO J., 20:3800-3810.

132 Horlein, A. J.; Naar, A. M.; Heinzel, T.; Torchia, J.; Gloss, B.; Kurokawa,

R.; Ryan, A.; Kamei, Y.; Soderstrom, M.; Glass, C. K.; et al. 1995 Ligand-

independent repression by the thyroid hormone receptor mediated by a nuclear receptor co-repressor. Nature, 377:397-404.

Hu, J. L.; Guan, X. J. and Sanchez, E. R. 1996 Enhancement of

glucocorticoid receptor-mediated gene expression by cellular stress: evidence for

the involvement of a heat shock-initiated factor or process during recovery from

stress. Cell Stress Chaperones, 1:197-205.

Huang, J.; Nueda, A.; Yoo, S. and Dynan, W. S. 1997 Heat shock

transcription factor 1 binds selectively in vitro to Ku protein and the catalytic

subunit of the DNA-dependent protein kinase. J. Biol. Chem., 272:26009-26016.

Hubl, S. T.; Owens, J. C. and Nelson, H. C. 1994 Mutational analysis of the DNA-binding domain of yeast heat shock transcription factor. Nat. Struct.

Biol., 1:615-620.

Hudson, L. G.; Santon, J. B.; Glass, C. K. and Gill, G. N. 1990 Ligand-

activated thyroid hormone and retinoic acid receptors inhibit growth factor

receptor promoter expression. Cell, 62:1165-1175.

Izzard, R. A.; Jackson, S. P. and Smith, G. C. 1999 Competitive and noncompetitive inhibition of DNA-dependent protein kinase. Cancer Res.,

59:2581-2586.

Janssens, S. and Beyaert, R. 2003 Role of Toll-like receptors in pathogen

recognition. Clin. Microbiol. Rev., 16:637-646.

133 Jolly, C. and Morimoto, R. I. 2000 Role of the heat shock response and

molecular chaperones in oncogenesis and cell death. J. Natl. Cancer Inst.,

92:1564-1572.

Jones, T. J.; Li, D.; Wolf, I. M.; Wadekar, S. A.; Periyasamy, S. and

Sanchez, E. R. 2004 Enhancement of Glucocorticoid Receptor-mediated Gene

Expression by Constitutively-active Heat Shock Factor 1. Mol. Endocrinol.,

18:509-520.

Karin, M. 1998 New twists in gene regulation by glucocorticoid receptor: is

DNA binding dispensable? Cell, 93:487-490.

Katengwa, S. and Polla, B. S. 1991 Flavonoids, but not protein kinase C

inhibitors, prevent stress protein synthesis during erythrophagocytosis.

Biochemical Biophysical Research Communications, 180:308-314.

Kelley, P. M. and Schlesinger, M. J. 1978 The effect of amino acid analogues and heat shock on gene expression in chicken embryo fibroblasts.

Cell, 15:1277-1286.

Kiang, J. G. and Tsokos, G. C. 1998 Heat shock protein 70 kDa:

molecular biology, biochemistry, and physiology. Pharmacol. Ther., 80:183-201.

Kim, S. J.; Tsukiyama, T.; Lewis, M. S. and Wu, C. 1994 Interaction of the

DNA-binding domain of Drosophila heat shock factor with its cognate DNA site: a

thermodynamic analysis using analytical ultracentrifugation. Protein Sci., 3:1040-

1051.

134 Kline, M. P. and Morimoto, R. I. 1997 Repression of the heat shock factor

1 transcriptional activation domain is modulated by constitutive phosphorylation.

Mol. Cell. Biol., 17:2107-2115.

Knauf, U.; Newton, E. M.; Kyriakis, J. and Kingston, R. E. 1996

Repression of human heat shock factor 1 activity at the control tempreture by phosphorylation. Genes Dev., 10:2782-2793.

Koyasu, S.; Nishida, E.; Kadowaki, T.; Matsuzaki, F.; Iida, K.; Harada, F.;

Kasuga, M.; Sakai, H. and Yahara, I. 1986 Two mammalian heat shock proteins,

HSP90 and HSP100, are actin-binding proteins. Proc. Natl. Acad. Sci. USA,

83:8054-8058.

Krappmann, D.; Wulczyn, F. G. and Scheidereit, C. 1996 Different

mechanisms control signal-induced degradation and basal turnover of the NF-

kappaB inhibitor IkappaB alpha in vivo. EMBO J., 15:6716-6726.

Kroeger, P. E. and Morimoto, R. I. 1994 Selection of new HSF1 and HSF2

DNA-binding sites reveals difference in trimer cooperativity. Mol. Cell. Biol.,

14:7592-7603.

Krstic, M.; Rogatsky, I.; Yamamoto, K. and Garabedian, M. J. 1997

Mitogen-activated and cyclin-dependent protein kinases selectively and

differentially modulate transcriptional enhancement by the glucocorticoid receptor

Mol. Cell. Biol., 17:3947-3954.

Kurihara, I.; Shibata, H.; Suzuki, T.; Ando, T.; Kobayashi, S.; Hayashi, M.;

Saito, I. and Saruta, T. 2002 Expression and regulation of nuclear receptor coactivators in glucocorticoid action. Mol. Cell Endocrinol., 189:181-189.

135 Laherty, C. D.; Yang, W. M.; Sun, J. M.; Davie, J. R.; Seto, E. and

Eisenman, R. N. 1997 Histone deacetylases associated with the mSin3

corepressor mediate mad transcriptional repression. Cell, 89:349-356.

Lambert, J. R. and Nordeen, S. K. 2003 CBP Recruitment and Histone

Acetylation in Differential Gene Induction by Glucocorticoids and Progestins.

Mol. Endocrinol., 17:1085-1094.

Larson, J. S.; Schuetz, T. J. and Kingston, R. E. 1988 Activation in vitro of sequence-specific DNA binding by a human regulatory factor. Nature, 335:372-

375.

Lee, D. Y.; Hayes, J. J.; Pruss, D. and Wolffe, A. P. 1993 A positive role for histone acetylation in transcription factor access to nucleosomal DNA. Cell,

72:73-84.

Li, D. P.; Li Calzi, S. and Sanchez, E. R. 1999 Inhibition of heat shock factor activity prevents heat shock potentiation of glucocorticoid receptor- mediated gene expression. Cell Stress Chaperones, 4:223-234.

Li, D. P.; Periyasamy, S.; Jones, T. J. and Sanchez, E. R. 2000 Heat and chemical shock potentiation of glucocorticoid receptor transactivation requires heat shock factor (HSF) activity. Modulation of HSF by vanadate and wortmannin. J. Biol. Chem., 275:26058-26065.

Lim-Tio, S. S. and Fuller, P. J. 1998 Intracellular signaling pathways confer specificity of transactivation by mineralocorticoid and glucocorticoid receptors. Endocrinology, 139:1653-1661.

136 Liu, W.; Hillmann, A. G. and Harmon, J. M. 1995 Hormone-independent

repression of the AP-1-inducible collagenase promoter activity by glucocorticoid

receptors. Mol. Cell. Biol., 15:1005-1013.

Luisi, B. F.; Xu, W. X.; Otwinowski, Z.; Freedman, L. P.; Yamamoto, K. R. and Sigler, P. B. 1991 Crystallographic analysis of the interaction of the glucocorticoid receptor with DNA [see comments]. Nature, 352:497-505.

Mader, S.; Kumar, V.; de, V. H. and Chambon, P. 1989 Three amino acids

of the oestrogen receptor are essential to its ability to distinguish an oestrogen

from a glucocorticoid-responsive element. Nature, 338:271-274.

Mark, P. J.; Ward, B. K.; Kumar, P.; Lahooti, H.; Minchin, R. F. and

Ratajczak, T. 2001 Human cyclophilin 40 is a heat shock protein that exhibits altered intracellular localization following heat shock. Cell Stress Chaperones,

6:59-70.

Mason, S. A. and Housley, P. R. 1993 Site-directed mutagenesis of the

phosphorylation sites in the mouse glucocorticoid receptor. J. Biol. Chem.,

268:21501-2154.

Mathew, A.; Shi, Y.; Jolly, C. and Morimoto, R. I. 2000 Analysis of the

mammalian heat-shock response. Inducible gene expression and heat-shock

factor activity. Methods Mol. Biol., 99:217-255.

McInerney, E. M.; Tsai, M. J.; O'Malley, B. W. and Katzenellenbogen, B.

S. 1996 Analysis of estrogen receptor transcriptional enhancement by a nuclear

hormone receptor coactivator. Proc. Natl. Acad. Sci. USA, 93:10069-10073.

137 McKay, L. I. and Cidlowski, J. A. 1998 Cross-talk between nuclear factor

kB and the steroid hormone receptors: mechanisms of mutual antagonism. Mol.

Endocrinol., 12:45-56.

McKay, L. I. and Cidlowski, J. A. 1999 Molecular control of

immune/inflammatory responses: interactions between nuclear factor-kappa B

and steroid receptor-signaling pathways. Endocr. Rev., 20:435-459.

McKenna, N. J.; Lanz, R. B. and O'Malley, B. W. 1999 Nuclear receptor coregulators: cellular and molecular biology. Endocr. Rev., 20:321-44.

McKenna, N. J.; Nawaz, Z.; Tsai, S. Y.; Tsai, M. J. and O'Malley, B. W.

1998 Distinct steady-state nuclear receptor coregulator complexes exist in vivo.

Proc. Natl. Acad. Sci. USA, 95:11697-11702.

McMillan, D. R.; Xiao, X.; Shao, L.; Graves, K. and Benjamin, I. J. 1998

Targeted disruption of heat shock transcription factor 1 abolishes thermotolerance and protection against heat-inducible apoptosis. J. Biol. Chem.,

273:7523-7528.

Medzhitov, R.; Preston-Hurlburt, P. and Janeway, C. A. J. 1997 A human

homologue of the Drosophila Toll protein signals activation of adaptive immunity.

Nature, 388:394-397.

Mercurio, F.; Didonato, J.; Rosette, C. and Karin, M. 1992 Molecular cloning and characterization of a novel Rel/NF-kappa B family member displaying structural and functional homology to NF-kappa B p50/p105. DNA Cell

Biol., 11:523-537.

138 Minota, S.; Cameron, B.; Welch, W. J. and Winfield, J. B. 1988

Autoantibodies to the constitutive 73-kD member of the hsp70 family of heat

shock proteins in systemic lupus erythematosus. J. Exp. Med., 168:1475-1480.

Moran, L. A.; Chauvin, M.; Kennedy, M. E.; Korri, M.; Lowe, D. G.;

Nicholson, R. C. and Perry, M. D. 1983 The major heat-shock protein (hsp70)

gene family: related sequences in mouse, Drosophila, and yeast. Can. J.

Biochem. Cell. Biol., 61:488-499.

Moras, D. and Gronemeyer, H. 1998 The nuclear receptor ligand-binding domain: structure and function. Curr. Opin. Cell. Biol., 10:384-391.

Morimoto, R. I. 1993 Cells in stress: transcriptional activation of heat shock genes. Science, 259:1409-1410.

Morimoto, R. I.; Sarge, K. D. and Abravaya, K. 1992 Transcriptional

regulation of heat shock genes J. Biol. Chem., 267:21987-21990.

Morishima, Y.; Murphy, P. J.; Li, D. P.; Sanchez, E. R. and Pratt, W. B.

2000 Stepwise assembly of a glucocorticoid receptor.hsp90 heterocomplex resolves two sequential ATP-dependent events involving first hsp70 and then hsp90 in opening of the steroid binding pocket. J. Biol. Chem., 275:18054-

18060.

Mosser, D. D.; Duchaine, J. and Massie, B. 1993 The DNA-binding activity

of the human heat shock transcription factor is regulated in vivo by hsp70. Mol.

Cell. Biol., 13:5427-5438.

Mosser, D. D.; Kotzbauer, P. T.; Sarge, K. D. and Morimoto, R. I. 1990 In vitro activation of heat shock transcription factor DNA-binding by calcium and

139 biochemical conditions that affect protein conformation. Proc. Natl. Acad. Sci.

USA, 87:3748-3752.

Mosser, D. D.; Theodorakis, N. G. and Morimoto, R. I. 1988 Coordinate changes in heat shock element-binding activity and HSP70 gene transcription rates in human cells. Mol. Cell. Biol., 8:4736-4744.

Mues, G. I.; Munn, T. Z. and Raese, J. D. 1986 A human gene family with sequence homology to Drosophila melanogaster Hsp70 heat shock genes. J.

Biol. Chem., 261:874-877.

Multhoff, G. and Hightower, L. E. 1996 Cell surface expression of heat

shock proteins and the immune response. Cell Stress Chaperones, 1:167-176.

Munck, A.; Guyre, P. M. and Holbrook, N. J. 1984 Physiological functions

of glucocorticoids in stress and their relation to pharmacological actions. Endocr.

Rev., 5:25-44.

Munck, A. and Naray-Fejes-Toth, A. 1992 The ups and downs of

glucocorticoid physiology. Permissive and suppressive effects revisited. Mol.

Cell. Endocrinol., 90:C1-4.

Nagai, N.; Nakai, A. and Nagata, K. 1995 Quercetin suppresses heat

shock response by down regulation of HSF1 Biochemical Biophysical Research

Communications, 208:1099-1105.

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.

140 Nakai, A.; Kawazoe, Y.; Tanabe, M.; Nagata, K. and Morimoto, R. I. 1995

The DNA-binding properties of two heat shock factors, HSF1 and HSF3, are induced in the avian erythroblast cell line HD6. Mol. Cell. Biol., 15:5268-5278.

Nakai, A. and Morimoto, R. I. 1993 Characterization of a novel chicken heat shock transcription factor, heat shock factor 3, suggests a new regulatory pathway. Mol. Cell. Biol., 13:1983-1997.

Nakai, A.; Tanabe, M.; Kawazoe, Y.; Inazawa, J.; Morimoto, R. I. and

Nagata, K. 1997 HSF4, a new member of the human heat shock factor family which lacks properties of a transcriptional activator. Mol. Cell. Biol., 17:469-481.

Nissen, R. M. and Yamamoto, K. R. 2000 The glucocorticoid receptor inhibits NFkappaB by interfering with serine-2 phosphorylation of the RNA polymerase II carboxy-terminal domain. Genes Dev., 14:2314-229.

Nordeen, S. K.; Green, P. P. I. I. I. and Fowlkes, D. M. 1987 A rapid, sensitive, and inexpensive assay for chloramphenicol acetyltransferase. DNA,

6:173-178.

Nordeen, S. K.; Moyer, M. L. and Bona, B. J. 1994 The coupling of multiple signal transduction pathways with steroid response mechanisms

Endocrinology, 134:1723-1732.

Nordeen, S. K.; Ogden, C. A.; Taraseviciene, L. and Lieberman, B. A.

1998 Extreme position dependence of a canonical hormone response element.

Mol. Endocrinol., 12:891-898.

141 Nueda, A.; Hudson, F.; Mivechi, N. F. and Dynan, W. S. 1999 DNA-

dependent protein kinase protects against heat-induced apoptosis. J. Biol.

Chem., 274:14988-14996.

Onate, S. A.; Boonyaratanakornkit, V.; Spencer, T. E.; Tsai, S. Y.; Tsai, M.

J.; Edwards, D. P. and O'Malley, B. W. 1998 The steroid receptor coactivator-1

contains multiple receptor interacting and activation domains that cooperatively

enhance the activation function 1 (AF1) and AF2 domains of steroid receptors. J.

Biol. Chem., 273:12101-12108.

Onate, S. A.; Tsai, S. Y.; Tsai, M. J. and O'Malley, B. W. 1995 Sequence

and characterization of a coactivator for the steroid hormone receptor

superfamily. Science, 270:1354-137.

Orti, E.; Hu, L. M. and Munck, A. 1993 Kinetics of glucocorticoid receptor phosphorylation in intact cells. Evidence for hormone-induced hyperphosphorylation after activation and recycling of hyperphosphorylated receptors. J. Biol. Chem., 268:7779-7784.

Orti, E.; Mendel, D. B.; Smith, L. I. and Munck, A. 1989 Agonist-dependent

phosphorylation and nuclear dephosphorylation of glucocorticoid receptors in

intact cells. J. Biol. Chem., 264:9728-9731.

Palombella, V. J.; Rando, O. J.; Goldberg, A. L. and Maniatis, T. 1994 The

ubiquitin-proteasome pathway is required for processing the NF-kappa B1

precursor protein and the activation of NF-kappa B. Cell, 78:773-785.

142 Parsell, D. A. and Lindquist, S. 1993 The function of heat-shock proteins

in stress tolerance: degradation and reactivation of damaged proteins. Annu.

Rev. Genet., 27:437-496.

Pelham, H. R. 1982 A regulatory upstream promoter element in the

Drosophila hsp 70 heat-shock gene. Cell, 30:517-528.

Peteranderl, R. and Nelson, H. C. 1992 Trimerization of the heat shock

transcription factor by a triple-stranded alpha-helical coiled-coil. Biochemistry,

31:12272-12276.

Philippe, J. and Missotten, M. 1990 Dexamethasone inhibits insulin

biosynthesis by destabilizing insulin messenger ribonucleic acid in hamster

insulinoma cells. Endocrinology, 127:1640-1645.

Picard, D. and Yamamoto, K. R. 1987 Two signals mediate hormone- dependent nuclear localization of the glucocorticoid receptor. European

Molecular Biology Organization Journal, 6:3333-3340.

Pocuca, N.; Ruzdijic, S.; Demonacos, C.; Kanazir, D. and Krstic-

Demonacos, M. 1998 Using yeast to study glucocorticoid receptor

phosphorylation. J. Steroid Biochem. Mol. Biol., 66:303-318.

Pratt, W. B. 1993 The role of heat shock proteins in regulating the

function, folding, and trafficking of the glucocorticoid receptor J. Biol. Chem.,

268:21455-21458.

Pratt, W. B.; Dalman, F. C.; Meshinchi, S. and Scherrer, L. C. 1990 The

relationship between glucocorticoid receptor binding to Hsp90 and receptor

function. Nippon Naibunpi Gakkai Zasshi, 66:1185-1197.

143 Pratt, W. B.; Scherrer, L. C.; Hutchison, K. A. and Dalman, F. C. 1992 A

model of glucocorticoid receptor unfolding and stabilization by a heat shock

protein complex. J. Steroid Biochem. Mol. Biol., 41:223-229.

Pratt, W. B. and Toft, D. O. 1997 Steroid receptor interactions with heat

shock protein and immunophilin chaperones. Endocr. Rev. 18:306-360.

Rabindran, S. K.; Giorgi, G.; Clos, J. and Wu, C. 1991 Molecular cloning and expression of a human heat shock factor, HSF1. Proc. Natl. Acad. Sci.,

USA, 88:6906-6910.

Rabindran, S. K.; Haroun, R. I.; Clos, J.; Wisniewski, J. and Wu, C. 1993

Regulation of heat shock factor trimer formation: role of conserved leucine zipper

Science, 259:230-234.

Rabindran, S. K.; Wisniewski, J.; Li, L.; Li, G. C. and Wu, C. 1994

Interaction between heat shock factor and hsp70 is insufficient to suppress

induction of DNA-binding activity in vivo. Mol. Cell. Biol., 14:6552-6560.

Ramsey, A. J.; Russell, L. C.; Whitt, S. R. and Chinkers, M. 2000

Overlapping sites of tetratricopeptide repeat protein binding and chaperone activity in heat shock protein 90. J. Biol. Chem., 275, 17857-17862.

Ray, A. and Prefontaine, K. E. 1994 Physical association and functional

antagonism between the p65 subunit of transcription factor NF-kB and the

glucocorticoid receptor. Proc. Natl. Acad. Sci. USA, 91:752-756.

Reichardt, H. M.; Kaestner, K. H.; Tuckermann, J.; Kretz, O.; Wessely, O.;

Bock, R.; Gass, P.; Schmid, W.; Herrlich, P.; Angel, P. and Schutz, G. 1998 DNA

144 binding of the glucocorticoid receptor is not essential for survival. Cell, 93:531-

541.

Reichardt, H. M. and Schutz, G. 1998 Glucocorticoid signalling--multiple variations of a common theme. Mol. Cell. Endocrinol., 146:1-6.

Reichardt, H. M.; Tuckermann, J. P.; Gottlicher, M.; Vujic, M.; Weih, F.;

Angel, P.; Herrlich, P. and Schutz, G. 2001 Repression of inflammatory

responses in the absence of DNA binding by the glucocorticoid receptor. EMBO

J., 20:7168-7173.

Rhodes, D. 1997 Chromatin structure. The nucleosome core all wrapped up. Nature, 389:231, 233.

Ritossa, F. 1962 A new puffing pattern induced by temperature shock and

DNP in Drosophila. Experientia. 18:571-573.

Rowbury, R. J. 2001a Extracellular sensing components and

extracellular induction component alarmones give early warning against stress in

Escherichia coli. Adv. Microb. Physiol., 44:215-257.

Rowbury, R. J. 2001b Cross-talk involving extracellular sensors and extracellular alarmones gives early warning to unstressed Escherichia coli of impending lethal chemical stress and leads to induction of tolerance responses.

J. Appl. Microbiol., 90:677-695.

Rowbury, R. J. 2003 Extracellular proteins as enterobacterial

thermometers. Sci. Prog., 86:139-155.

145 Rowbury, R. J. and Goodson, M. 2001 Extracellular sensing and

signalling pheromones switch-on thermotolerance and other stress responses in

Escherichia coli. Sci. Prog., 84:205-233.

Sanchez, E. R. 1990 Hsp56: a novel heat shock protein associated with untransformed steroid receptor complexes. J. Biol. Chem., 265:22067-22070.

Sanchez, E. R. 1992 Heat shock induces translocation to the nucleus of

the unliganded glucocorticoid receptor. J. Biol. Chem., 267:17-20.

Sanchez, E. R.; Hu, J. L.; Zhong, S. J.; Shen, P.; Green, M. J. and

Housley, P. R. 1994 Potentiation of glucocorticoid receptor mediated gene expression by heat and chemical shock Molecular. Endocrinology, 8:408-421.

Sapolsky, R. M.; Romero, L. M. and Munck, A. U. 2000 How do glucocorticoids influence stress responses? Integrating permissive, suppressive, stimulatory, and preparative actions. Endocr. Rev., 21:55-89.

Sapozhnikov, A. M.; Gusarova, G. A.; Ponomarev, E. D. and Telford, W.

G. 2002 Translocation of cytoplasmic HSP70 onto the surface of EL-4 cells

during apoptosis. Cell Prolif., 35:193-206.

Sarge, K. D.; Murphy, S. P. and Morimoto, R. I. 1993 Activation of heat shock gene transcription by heat shock factor 1 involves oligomerization, acquisition of DNA-binding activity, and nuclear localization and can occur in the absence of stress. Mol. Cell. Biol., 13:1392-1407.

Sarge, K. D.; Zimarino, V.; Holm, K.; Wu, C. and Morimoto, R. I. 1991

Cloning and characterization of two mouse heat shock factors with distinct inducible and constitutive DNA-binding ability. Genes Dev., 5:1902-1911.

146 Sarkaria, J. N.; Tibbetts, R. S.; Busby, E. C.; Kennedy, A. P.; Hill, D. E.

and Abraham, R. T. 1998 Inhibition of phosphoinositide 3-kinase related kinases

by the radiosensitizing agent wortmannin Cancer Res., 58:4375-4382.

Schaaf, M. J. and Cidlowski, J. A. 2002 Molecular mechanisms of

glucocorticoid action and resistance. J. Steroid Biochem. Mol. Biol., 83:37-48.

Scheinman, R. I.; Gualberto, A.; Jewell, C. M.; Cidlowski, J. A. and

Baldwin Jr., A. S. 1995 Characterization of mechanisms involved in

transrepression of NF-kB by activated glucocorticoid receptors. Mol. Cell. Biol.,

15:943-953.

Scherrer, L. C.; Hutchison, K. A.; Sanchez, E. R.; Randall, S. K. and Pratt,

W. B. 1992 A heat shock protein complex isolated from rabbit reticulocyte lysate

can reconstitute a functional glucocorticoid receptor-Hsp90 complex.

Biochemistry, 31:7325-7329.

Schiller, P.; Amin, J.; Ananthan, J.; Brown, M. E.; Scott, W. A. and

Voellmy, R. 1988 Cis-acting elements involved in the regulated expression of a

human HSP70 gene. J. Mol. Biol., 203:97-105.

Schlesinger, M. J. 1990 Heat shock proteins. J. Biol. Chem., 265:12111-

12114.

Schmidt, T. J. and Meyer, A. S. 1994 Autoregulation of corticosteroid receptors. How, when, where, and why? Receptor, 4:229-257.

Schuetz, T. J.; Gallo, G. J.; Sheldon, L.; Tempst, P. and Kingston, R. E.

1991 Isolation of a cDNA for HSF2: evidence for two heat shock factor genes in

humans. Proc. Natl. Acad. Sci. USA, 88:6911-6915.

147 Sciandra, J. J. and Subjeck, J. R. 1984 Heat shock proteins and protection

of proliferation and translation in mammalian cells. Cancer Res., 44:5188-5194.

Seol, W.; Mahon, M. J.; Lee, Y. K. and Moore, D. D. 1996 Two receptor

interacting domains in the nuclear hormone receptor corepressor RIP13/N-CoR.

Mol. Endocrinol., 10:1646-1655.

Shen, P.; Xie, Z.-J.; Li, H. and Sanchez, E. R. 1993 Glucocorticoid receptor conversion to high affinity nuclear binding and transcription enhancement activity in Chinese hamster ovary cells subjected to heat and chemical stress J. Steroid Biochem. Mol. Biol., 47:55-64.

Shi, Y.; Kroeger, P. E. and Morimoto, R. I. 1995 The caoboxyl-terminal

transactivation domain of heat shock factor 1 is negatively regulated and stress

responsive. Mol. Cell. Biol., 15:4309-4318.

Shi, Y.; Mosser, D. D. and Morimoto, R. I. 1998 Molecular chaperones as

HSF1-specific transcriptional repressors. Genes Dev., 12:654-666.

Sistonen, L.; Sarge, K. D. and Morimoto, R. I. 1994 Human heat shock

factors 1 and 2 are differentially activated and can synergistically induce hsp70

gene transcription. Mol. Cell. Biol., 14:2087-2099.

Sivo, J.; Harmon, J. M. and Vogel, S. N. 1996 Heat shock mimics glucocorticoid effects on IFN-gamma-induced Fc gamma RI and Ia messenger

RNA expression in mouse peritoneal macrophages. J. Immunol., 156:3450-

3454.

148 Smith, C. L.; Nawaz, Z. and O'Malley, B. W. 1997 Coactivator and

corepressor regulation of the agonist/antagonist activity of the mixed

antiestrogen, 4-hydroxytamoxifen. Mol. Endocrinol., 11:657-666.

Soncin, F.; Zhang, X.; Chu, B.; Wang, X.; Asea, A.; Ann, S. M.; Sacks, D.

B. and Calderwood, S. K. 2003 Transcriptional activity and DNA binding of heat shock factor-1 involve phosphorylation on threonine 142 by CK2. Biochem.

Biophys. Res. Commun., 303:700-706.

Sorger, P. K.; Lewis, M. J. and Pelham, H. R. 1987 Heat shock factor is

regulated differently in yeast and HeLa cells. Nature, 329:81-84.

Sorger, P. K. and Pelham, H. R. 1987 Cloning and expression of a gene encoding hsc73, the major hsp70-like protein in unstressed rat cells. EMBO J.

6:993-998.

Sorger, P. K. and Pelham, H. R. 1988 Yeast heat shock factor is an

essential DNA-binding protein that exhibits temperature-dependent

phosphorylation. Cell, 54:855-864.

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. and O'Malley, B. W.

1997 Steroid receptor coactivator-1 is a histone acetyltransferase. Nature,

389:194-18.

Srinivasan, G.; Post, J. F. and Thompson, E. B. 1997 Optimal ligand

binding by the recombinant human glucocorticoid receptor and assembly of the

receptor complex with heat shock protein 90 correlate with high intracellular ATP

levels in Spodoptera frugiperda cells. J. Steroid Biochem. Mol. Biol., 60:1-9.

149 Stevens, A.; Garside, H.; Berry, A.; Waters, C.; White, A. and Ray, D.

2003 Dissociation of steroid receptor coactivator 1 and nuclear receptor

corepressor recruitment to the human glucocorticoid receptor by modification of

the ligand-receptor interface: the role of tyrosine 735. Mol. Endocrinol., 17:845-

859.

Sun, H.; Charles, C. H.; Lau, L. F. and Tonks, N. K. 1993 MKP1, an immediate early gene product, is a dual specificity phosphatase that dephosphorylates MAP kinase in vivo. Cell, 75:487-493.

Tanabe, M.; Kawazoe, Y.; Takeda, S.; Morimoto, R. I.; Nagata, K. and

Nakai, A. 1998 Disruption of the HSF3 gene results in the severe reduction of heat shock gene expression and loss of thermotolerance. EMBO J., 17:1750-

1758.

Tanabe, M.; Nakai, A.; Kawazoe, Y. and Nagata, K. 1997 Different

thresholds in the responses of two heat shock transcription factors, HSF1 and

HSF3. J. Biol. Chem., 272:15389-15395.

Tienrungroj, W.; Sanchez, E. R.; Housley, P. R.; Harrison, R. W. and

Pratt, W. B. 1987 Glucocorticoid receptor phosphorylation, transformation, and

DNA-binding. J. Biol. Chem., 262:17342-17349.

Tomasovic, S. P. and Koval, T. M. 1985 Relationship between cell survival and heat-stress protein synthesis in a Drosophila cell line. Int. J. Radiat. Biol.

Relat. Stud. Phys. Chem. Med., 48:635-650.

150 Torchia, J.; Rose, D. W.; Inostroza, J.; Kamei, Y.; Westin, S.; Glass, C. K. and Rosenfeld, M. G. 1997 The transcriptional co-activator p/CIP binds CBP and mediates nuclear-receptor function. Nature, 387:677-684.

Truss, M. and Beato, M. 1993 Steroid hormone receptors: interaction with deoxyribonucleic acid and transcription factors. Endocr. Rev., 14:459-479.

Tsai, M.-J. and O'Malley, B. W. 1994 Molecular mechanisms of action of steroid/thyroid receptor superfamily members Annual Review of Biochemistry,

63:451-486.

Tse, C.; Sera, T.; Wolffe, A. P. and Hansen, J. C. 1998 Disruption of

higher-order folding by core histone acetylation dramatically enhances

transcription of nucleosomal arrays by RNA polymerase III. Mol. Cell. Biol.,

18:4629-4638.

Tytell, M.; Greenberg, S. G. and Lasek, R. J. 1986 Heat shock-like protein

is transferred from glia to axon. Brain Res., 363:161-164.

Ura, K.; Kurumizaka, H.; Dimitrov, S.; Almouzni, G. and Wolffe, A. P. 1997

Histone acetylation: influence on transcription, nucleosome mobility and positioning, and linker histone-dependent transcriptional repression. EMBO J.,

16:2096-2107.

Vabulas, R. M.; Wagner, H. and Schild, H. 2002 Heat shock proteins as

ligands of toll-like receptors. Curr. Top. Microbiol. Immunol., 270:169-184.

Voegel, J. J.; Heine, M. J.; Zechel, C.; Chambon, P. and Gronemeyer, H.

1996 TIF2, a 160 kDa transcriptional mediator for the ligand-dependent activation function AF-2 of nuclear receptors. EMBO J., 15:3667-3675.

151 Voellmy, R.; Ahmed, A.; Schiller, P.; Bromley, P. and Rungger, D. 1985

Isolation and functional analysis of a human 70,000-dalton heat shock protein gene segment. Proc. Natl. Acad. Sci. USA, 82:4949-4953.

Wadekar, S. A.; Li, D.; Periyasamy, S. and Sanchez, E. R. 2001 Inhibition

of heat shock transcription factor by glucocorticoid receptor. Mol. Endocrinol.,

15:1396-1410.

Wadekar, S. A.; Li, D. and Sanchez, E. R. 2004 Agonist-activated

Glucocorticoid Receptor Inhibits Binding of Heat Shock Factor-1 to the Hsp70

Promoter In Vivo. Mol. Endocrinol., 18:500-508.

Wagstaff, M. J.; Smith, J.; Collaco-Moraes, Y.; de Belleroche, J. S.;

Voellmy, R.; Coffin, R. S. and Latchman, D. S. 1998 Delivery of a constitutively active form of the heat shock factor using a virus vector protects neuronal cells from thermal or ischaemic stress but not from apoptosis. Eur. J. Neurosci.,

10:3343-3350.

Wang, W.; Cote, J.; Xue, Y.; Zhou, S.; Khavari, P. A.; Biggar, S. R.;

Muchardt, C.; Kalpana, G. V.; Goff, S. P.; Yaniv, M.; Workman, J. L. and

Crabtree, G. R. 1996a Purification and biochemical heterogeneity of the

mammalian SWI-SNF complex. EMBO J., 15:5370-5382.

Wang, W.; Xue, Y.; Zhou, S.; Kuo, A.; Cairns, B. R. and Crabtree, G. R.

1996b Diversity and specialization of mammalian SWI/SNF complexes. Genes

Dev., 10:2117-2130.

152 Wang, Z.; Frederick, J. and Garabedian, M. J. 2002 Deciphering the

phosphorylation "code" of the glucocorticoid receptor in vivo. J. Biol. Chem.,

277:26573-26580.

Ward, Y.; Gupta, S.; Jensen, P.; Wartmann, M.; Davis, R. J. and Kelly, K.

1994 Control of MAP kinase activation by the mitogen induced threonine/tyrosine

phosphatase PAC1. Nature, 367:651-654.

Webster, J. C.; Jewell, C. M.; Bodwell, J. E.; Munck, A.; Sar, M. and

Cidlowski, J. A. 1997 Mouse glucocorticoid receptor phosphorylation status influences multiple functions of the receptor protein. J. Biol. Chem., 272:9287-

993.

Webster, N. J.; Green, S.; Jin, J. R. and Chambon, P. 1988 The hormone- binding domains of the estrogen and glucocorticoid receptors contain an inducible transcription activation function. Cell, 54:199-207.

Wilson, C. J.; Chao, D. M.; Imbalzano, A. N.; Schnitzler, G. R.; Kingston,

R. E. and Young, R. A. 1996 RNA polymerase II holoenzyme contains SWI/SNF

regulators involved in chromatin remodeling. Cell, 84:235-244.

Wu, B. J. and Morimoto, R. I. 1985 Transcription of the human hsp70

gene is induced by serum stimulation. Proc. Natl. Acad. Sci. USA, 82:6070-6074.

Wu, C. 1995 Heat shock transcription factors: structure and regulation.

Annu. Rev. Cell. Dev. Biol., 11:441-469.

Xia, W.; Guo, Y.; Vilaboa, N.; Zuo, J. and Voellmy, R. 1998 Transcriptional activation of heat shock factor HSF1 probed by phosphopeptide analysis of factor

32P-labeled in vivo. J. Biol. Chem., 273:8749-8755.

153 Xia, W.; Vilaboa, N.; Martin, J. L.; Mestril, R.; Guo, Y. and Voellmy, R.

1999 Modulation of tolerance by mutant heat shock transcription factors. Cell

Stress Chaperones, 4:8-18.

Xia, W. and Voellmy, R. 1997 Hyperphosphorylation of heat shock

transcription factor 1 is correlated with transcriptional competence and slow

dissociation of active factor trimers. J. Biol. Chem., 272:4094-102.

Xiao, X.; Zuo, X.; Davis, A. A.; McMillan, D. R.; Curry, B. B.; Richardson,

J. A. and Benjamin, I. J. 1999 HSF1 is required for extra-embryonic

development, postnatal growth and protection during inflammatory responses in

mice. Embo J., 18:5943-552.

Xu, J.; Qiu, Y.; DeMayo, F. J.; Tsai, S. Y.; Tsai, M. J. and O'Malley, B. W.

1998 Partial hormone resistance in mice with disruption of the steroid receptor coactivator-1 (SRC-1) gene. Science, 279:1922-1925.

Yang, S. H.; Nussenzweig, A.; Li, L.; Kim, D.; Ouyang, H.; Burgman, P. and Li, G. C. 1996 Modulation of thermal induction of hsp70 expression by Ku autoantigen or its individual subunits. Mol. Cell. Biol., 16:3799-3806.

Yudt, M. R. and Cidlowski, J. A. 2001 Molecular identification and

characterization of a and b forms of the glucocorticoid receptor. Mol. Endocrinol.,

15:1093-1103.

Zuo, J.; Dahl, G. and Voellmy, R. 1994 Activation of the DNA-binding

ability of human heat shock transcription factor 1 may involve the transition from

an intramolecular to an intermolecular triple-stranded coiled-coil structure. Mol.

Cell. Biol., 14:7557-7568.

154 Zou, J.; Guo, Y.; Guettouche, T.; Smith, D. F. and Voellmy, R. 1998

Repression of heat shock transcription factor HSF1 activation by hsp90 (hsp90 complex) that forms a stress-sensitive complex with HSF1. Cell, 94:471-480.

Zuo, J.; Rungger, D. and Voellmy, R. 1995 Multiple layers of regulation of human heat shock transcription factor 1. Mol. Cell. Biol., 15:4319-4330.

155 ABSTRACT

Stress potentiation of glucocorticoid receptor (GR) transcription enhancement activity (transactivity) has been seen in response to heat shock

(HS) at 43˚C or chemical shock (CS) with 200 µM sodium arsenite. Under these conditions, a robust increase in GR transactivity is observed using GR- responsive promoters. Early work has suggested a role for heat shock factor 1

(HSF1) in the stress potentiation effect. To provide conclusive evidence for this role, a constitutively-active mutant of HSF1 (E189) was used. Results show that expression of E189, in the absence of stress, causes both an increase in heat shock protein expression and in GR transactivity. Further, this potentiation approached the levels observed under HS conditions, providing evidence that stress potentiation of GR activity occurs, at least in part, through HSF1-controlled gene products. To expand the roles of stress signaling in GR activity, we investigated the ability of unstressed cells to respond to stress-conditioned cell culture media. Control- and stress-conditioned media were transferred from

L929 cells to adjacent unstressed cells in the presence of dexamethasone.

Conditioned media was harvested 4 and 20h post-stress treatment and transferred to L929 cells stably selected for the minimal GRE2E1B promoter controlling expression of CAT. Results provide evidence for stress-induced release of a factor (SRF) that has the ability to enhance GR transactivity. The

SRF exhibited characteristics that were consistent with media-soluble proteins, but was unable to function in serum-free media to increase GR transactivation in unstressed cells. In summary, this work provides evidence for two mechanisms

156 through which stress can regulate GR activity. One mechanism is stress- activated HSF1-dependent gene product(s) can directly or indirectly affect the ligand-dependent transactivity of the GR. The second mechanism is stress- induced release of SRF into the cell culture media can cause an increase in transactivity of the GR in identical unstressed cells, or potentially in both stressed and unstressed cells. The latter mechanism has implications for physiological responses to stress, suggesting an “early warning” mechanism in which stressed cells initiate preemptive GR protective pathways through release of SRF.

157