Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Evaluation of ginsenosides in transactivation and transrepression of human glucocorticoid receptor alpha Hu, Catherine 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


24-ubc_2016_february_hu_catherine.pdf [ 1.17MB ]
JSON: 24-1.0221313.json
JSON-LD: 24-1.0221313-ld.json
RDF/XML (Pretty): 24-1.0221313-rdf.xml
RDF/JSON: 24-1.0221313-rdf.json
Turtle: 24-1.0221313-turtle.txt
N-Triples: 24-1.0221313-rdf-ntriples.txt
Original Record: 24-1.0221313-source.json
Full Text

Full Text

EVALUATION OF GINSENOSIDES IN TRANSACTIVATION AND TRANSREPRESSION OF HUMAN GLUCOCORTICOID RECEPTOR ALPHA  by Catherine Hu  B.A., University of California, Berkeley, 2005  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Pharmaceutical Sciences)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  December 2015  ©  Catherine Hu, 2015 ii  Abstract Ginsenosides are pharmacologically active compounds in ginseng, a medicinal herb that is highly valued and widely consumed. They are reported to have diverse effects, including neuromodulation, anticancer, and immunomodulation. Glucocorticoid receptor (GR) is a nuclear receptor involved in transcriptional regulation of genes in numerous important physiological processes, such as stress-related homeostasis, gluconeogenesis, bone remodeling, and anti-inflammation. Previous research suggested ginsenosides as agonists of rodent GRα. Studies on human GRα (hGRα) mainly focused on a single ginsenoside and its effect on either hGRα-mediated transactivation or transrepression. However, only a few ginsenosides (compound K, Rh1, Rh2, Re, Rg1) were examined and it is not known whether ginsenosides activate hGRα in an analog-selective manner. In this study, seven protopanaxadiol (PPD)-type ginsenosides (Rb1, Rb2, Rc, Rd, compound K, Rh2, PPD) and five protopanaxatriol (PPT)-type ginsenosides (Re, Rf, Rg1, Rh1, PPT) were investigated to determine whether they act as functional ligands of hGRα for both its transactivation and transrepression activity. In vitro time resolved-fluorescence resonance energy transfer (TR-FRET) competitive ligand-binding assay revealed that ginsenosides can weakly bind to the ligand-binding domain of hGRα. Among the selected ginsenosides, monoglycosylated PPD-type ginsenosides compound K and Rh2 exhibited strongest binding to the receptor. Dual-luciferase reporter gene assays employing firefly luciferase reporter vectors carrying either glucocorticoid response element or NF-κB response element were conducted in human colon adenocarcinoma cells (LS180). None of the ginsenosides increased or attenuated hGRα-mediated transactivation or transrepression activity. Furthermore, hGRα target gene (hTAT and hCBG) expression was studied in human hepatocellular carcinoma cells (HepG2) and quantified by real-time PCR. The data indicated that iii  ginsenoside Rh2 did not influence hGRα target gene expression. In summary, among all PPD-type ginsenosides (Rb1, Rb2, Rc, Rd, compound K, Rh2, PPD) and PPT-type ginsenosides (Re, Rf, Rg1, Rh1, PPT) tested, monoglycosylated PPD-type ginsenosides compound K and Rh2 exhibited stronger binding to hGRα-LBD, while others could only bind weakly. Nevertheless, none of the ginsenosides could modulate hGRα activity or affect target gene expression. Therefore, these ginsenosides are not functional ligands of hGRα in LS180 and HepG2 cells.  iv  Preface  This thesis is an independent work by Catherine Hu under the supervision of Dr. Thomas Chang; data presented in this work are unpublished.  v  Table of Contents  Abstract ........................................................................................................................................... ii Preface............................................................................................................................................ iv Table of Contents .............................................................................................................................v List of Tables ................................................................................................................................. ix List of Figures ..................................................................................................................................x List of Abbreviations ..................................................................................................................... xi Acknowledgements ........................................................................................................................xv 1. Introduction ............................................................................................................................. 1 1.1 Glucocorticoid Receptor Overview .......................................................................... 1 1.1.1 GR Structure and Splice Variants ....................................................................... 1 1.1.2 Mechanism of GR-mediated Transactivation and Transrepression .................... 3 GR-mediated Transactivation ......................................................................... 3 GR-mediated Transrepression ........................................................................ 4 Nongenomic Effects........................................................................................ 7 1.1.3 GR Function and Target Genes........................................................................... 8 Negative Feedback Regulation of Hypothalamic-pituitary-adrenal Axis ....... 8 Gluconeogenesis ............................................................................................. 9 Bone Remodeling.......................................................................................... 10 Anti-inflammation......................................................................................... 11 1.1.4 GR Ligands ....................................................................................................... 13 vi  1.2 Ginsenosides ........................................................................................................... 16 1.2.1 Biotransformation and Pharmacokinetics ......................................................... 18 1.2.2 Biological Activities ......................................................................................... 23 Central Nervous System Effects ................................................................... 23 Anticancer Effects ......................................................................................... 24 Anti-inflammatory Effects ............................................................................ 24 1.2.3 Ginsenosides’ Mechanisms of Action .............................................................. 25 1.2.4 Ginsenosides as GRα Agonists ......................................................................... 26 Studies in Rodent Cell Lines......................................................................... 26 Studies in Human Cell Lines ........................................................................ 30 1.3 Rationale of the Study ............................................................................................. 34 1.4 Research and Experimental Hypotheses ................................................................. 34 1.5 Specific Aims .......................................................................................................... 35 2. Materials and Methods .......................................................................................................... 36 2.1 Chemicals and Reagents ......................................................................................... 36 2.2 Cell Lines and Cell Culture..................................................................................... 37 2.3 Lactate Dehydrogenase Cytotoxicity Detection Assay ........................................... 37 2.4 Time-resolved Fluorescence Resonance Energy Transfer Competitive Ligand-binding Assay........................................................................................................................ 38 2.5 Transient Transfection ............................................................................................ 39 2.6 Dual-luciferase Reporter Gene Assay ..................................................................... 40 2.7 HepG2 Cell Culture and Treatment for Gene Expression Study ............................ 41 2.8 Isolation of Total RNA ........................................................................................... 41 vii  2.9 Reverse Transcription and cDNA Quantification ................................................... 42 2.10 Primers and Real-time Polymerase Chain Reaction ............................................... 43 2.11 Statistical Analysis .................................................................................................. 44 3. Results ................................................................................................................................... 45 3.1 Determination of Non-cytotoxic Concentrations of Ginsenosides in Cultured LS180 Cells ........................................................................................................................... 45 3.2 In vitro Binding of Ginsenosides to hGRα Ligand-binding Domain...................... 45 3.3 Ginsenosides in hGRα-mediated Transactivation................................................... 47 3.4 Glycosylated PPD-type Ginsenosides in hGRα-mediated Transrepression ........... 47 3.5 Ginsenoside Rh2 in hGRα-mediated Transactivation and Transrepression Increased by Dexamethasone ................................................................................................................ 48 3.6 Ginsenoside Rh2 in hGRα Target Gene Expression ............................................... 49 4. Discussion ............................................................................................................................. 63 4.1 Ginsenosides Bind Weakly to hGRα-LBD ............................................................. 63 4.2 Ginsenosides Were Unable to Modulate hGRα-mediated Transactivation ............ 65 4.3 Ginsenosides Were Unable to Modulate hGRα-mediated Transrepression ........... 67 4.4 Ginsenoside Rh2 Was Unable to Attenuate hGRα-mediated Transactivation and Transrepression ..................................................................................................................... 68 4.5 Ginsenoside Rh2 Was Unable to Influence hGRα-mediated Transactivation and Transrepression of Target Genes .......................................................................................... 69 4.6 Potential Explanations for Discrepancy among Different Studies ......................... 70 4.6.1 Cell Type Differences in Endogenous Coregulator Level ................................ 70 SRC-1 Level and hGRα-mediated Transactivation ...................................... 71 viii CBP Level and hGRα-mediated Transrepression ......................................... 72 4.6.2 Cell Type Differences in P-glycoprotein Expression ....................................... 73 4.7 Limitations of the Study.......................................................................................... 75 4.8 Future Experiments ................................................................................................. 76 4.8.1 Assessment of Ginsenosides Rb1, Rb2, Rc, and Rd at High Concentrations ... 76 4.8.2 Determination of SRC-1’s Role in Ginsenosides’ Ability to Modulate hGRα-mediated Transactivation .................................................................................................. 76 4.8.3 Determination of CBP’s Role in Ginsenosides’ Ability to Modulate hGRα-mediated Transrepression ................................................................................................. 77 4.8.4 Investigation of Effect of P-gp Drug Efflux ..................................................... 77 4.9 Summary and Conclusion ....................................................................................... 78 References ......................................................................................................................................80 Appendix ........................................................................................................................................88   ix  List of Tables  Table 3.1. IC50 of dexamethasone and glycosylated PPD-type ginsenosides ............................... 46  x  List of Figures  Figure 1.1. Structure of GR............................................................................................................. 3 Figure 1.2. Mechanisms of GR-mediated transcriptional regulation .............................................. 6 Figure 1.3. Structures of GR ligands ............................................................................................ 15 Figure 1.4. Structures of selected ginsenosides ............................................................................ 18 Figure 1.5. Proposed mechanism of PPD-type ginsenoside biotransformation............................ 21 Figure 1.6. Proposed mechanism of PPT-type ginsenoside biotransformation ............................ 22 Figure 3.1. Determination of non-cytotoxic concentrations of ginsenosides in cultured LS180 cells ............................................................................................................................................... 51 Figure 3.2. In vitro binding of ginsenosides to hGRα ligand-binding domain ............................. 53 Figure 3.3. Ginsenosides had no effect on hGRα-mediated transactivation ................................. 55 Figure 3.4. Ginsenosides had no effect on hGRα-mediated transrepression ................................ 57 Figure 3.5. Ginsenoside Rh2 could not attenuate hGRα-mediated transactivation and transrepression increased by dexamethasone ................................................................................ 59 Figure 3.6. Ginsenoside Rh2 did not influence hGRα target gene expression ............................. 61 Figure 4.1. Structures of GR ligands and ginsenosides with atom numbering ............................. 79       xi  List of Abbreviations  Aβ  beta-amyloid AF-1  activation function 1 AF-2  activation function 2 ANOVA analysis of variance AR  androgen receptor ATGL  adipose triglyceride lipase BTM  basal transcription machinery CBP  CREB-binding protein C/EBPβ CCAAT/enhancer binding protein beta CK  ginsenoside compound K CNS  central nervous system COX-2 cyclooxygenase 2 CpdA  compound A CRH  corticotrophin-releasing hormone CS-HI-FBS charcoal-stripped heat-inactivated fetal bovine serum DBD  DNA-binding domain DMSO  dimethyl sulfoxide DUSP1 dual specificity protein phosphatase 1 EC50  half maximal effective concentration ER  estrogen receptor FBS  fetal bovine serum xii  FoxO1  forkhead box protein O1 FP  fluorescence polarization G6P  glucose-6-phosphatase GC  glucocorticoid GR  glucocorticoid receptor GRE  glucocorticoid response element GSS  total ginseng saponins GST  glutathione S-transferase hCBG  human corticosteroid binding globulin hGRα  human glucocorticoid receptor alpha HPA  hypothalamic-pituitary-adrenal HR  hinge region HSP90  heat shock protein 90 hTAT  human tyrosine aminotransferase IC50  half maximal inhibitory concentration IL-1  interleukin 1  IL-2  interleukin 2 LBD  ligand-binding domain LDH  lactate dehydrogenase LPS  lipopolysaccharide MEM-EBSS minimum essential medium-Earle’s balanced salt solution MR  mineralocorticoid receptor NCoR1 nuclear receptor corepressor 1  xiii  NCoR2 nuclear receptor corepressor 2 NF-κB  nuclear factor kappa-light-chain-enhancer of activated B cells nGRE  negative glucocorticoid response element NL1  nuclear localization signal sequence 1  NL2  nuclear localization signal sequence 2 NO  nitric oxide NOS  nitric oxide synthase NR  nuclear receptor NTD  N-terminal domain PBS  phosphate-buffered saline PCR  polymerase chain reaction PEPCK phosphoenolpyruvate carboxykinase P-gp  p-glycoprotein PI3K  phosphatidylinositol 3-kinase POMC  proopiomelanocortin PPD  protopanaxadiol PPT  protopanaxatriol PR  progesterone receptor Rb1  ginsenoside Rb1 Rb2  ginsenoside Rb2 Rc  ginsenoside Rc Rd  ginsenoside Rd RE  response element xiv  Re  ginsenoside Re Rf  ginsenoside Rf Rg1  ginsenoside Rg1 Rg3  ginsenoside Rg3 Rh1  ginsenoside Rh1 Rh2  ginsenoside Rh2 RU486  mifepristone SEGRA selective glucocorticoid receptor agonists SEM  standard error of the mean SGK  serum and glucocorticoid-regulated kinase 1 siRNA  small interfering RNA SRC-1  steroid receptor coactivator 1  SRC-2  steroid receptor coactivator 2 SRC-3  steroid receptor coactivator 3 Tb  terbium TF  transcription factor TLR4  toll-like receptor 4 TNF-α  tumor necrosis factor alpha TR-FRET time-resolved fluorescence resonance energy transfer  xv  Acknowledgements  I would like to express my gratitude to my supervisor, Dr. Thomas Chang, for his guidance, encouragement, and support throughout the entire project. I thank my committee members and external examiner, Dr. Stelvio Bandiera, Dr. Tim Chen, Dr. Adam Frankel, and Dr. Zhaoming Xu, for their valuable advice. I would also like to thank my previous and current lab members, Dr. Aik Jiang Lau, Dr. Jayakumar Surendradoss, Devinder Sharma, and Abdullah Turkistani, for all their assistance, encouragement, and support. I would like to give my sincere gratitude to the Faculty of Pharmaceutical Sciences for awarding me the Kam Li Ma Scholarship in Pharmaceutical Sciences (2012-2014).  I am very grateful for my wonderful family and friends; I thank them for all their love, care, understanding, and prayers throughout the entire program while I was struggling and fighting through numerous trials in life at the same time.  Lastly, but most importantly, I thank my God for shepherding me through the valley of the shadow of death for the past years, and not forsaking me even in my utmost failures. I also praise Him for giving me my knight, who has been patiently staying by my side throughout all the waves and winds, and to whom I am much indebted.  1  1. Introduction  1.1 Glucocorticoid Receptor Overview Glucocorticoid receptor (GR), with the designation NR3C1 (nuclear receptor subfamily 3 group C member 1), is a nuclear receptor belonging to the family of steroid hormone receptors, which also includes mineralocorticoid receptor (MR), progesterone receptor (PR), androgen receptor (AR), and estrogen receptor (ER) (Heitzer et al., 2007). The GR gene is located on chromosome 5q31-32, and consists of nine exons that are subjected to alternative splicing (Oakley et al., 2011). Upon binding to a glucocorticoid (GC), its endogenous ligand, GR is known to regulate the expression of numerous genes involved in important physiological processes such as growth, development, metabolism, anti-inflammation, and stress-related homeostasis.  1.1.1 GR Structure and Splice Variants GR is composed of an N-terminal domain (NTD), a DNA-binding domain (DBD), a hinge region (HR), and a ligand-binding domain (LBD). The NTD of GR contains a ligand-independent Activation Function-1 (AF-1) domain, which interacts with coregulators, chromatin remodeling enzymes, and components of basal transcription machinery (BTM). DBD contains two zinc finger motifs that enable the receptor to bind the glucocorticoid response element (GRE). LBD is responsible for ligand-binding and contains AF-2, a ligand-dependent transactivation domain that recruits coregulators for transcriptional regulation. In addition, regions of DBD, HR and LBD also contain nuclear localization signal sequences (NL1 and NL2) and receptor dimerization motifs (Fig. 1.1) (Nicolaides et al., 2010; Oakley et al., 2011). 2  There are several GR isoforms, and the two most well-studied forms are GRα and GRβ. GRα and GRβ are generated through alternative splicing of exon 9, which causes a 50 amino acid (a.a.) truncation in the LBD of GRβ, and addition of 15 new a.a. in the domain compared to GRα (Yudt et al., 2003). Whereas the LBD of GRα is capable of GC-binding and the receptor is transcriptionally active, GRβ is capable of neither ligand-binding nor transcriptional activation. Furthermore, in the absence of ligand, GRα is mostly sequestered in the cytoplasm, but GRβ predominantly resides in the nucleus (Oakley et al., 1996). It is proposed that GRβ may function as a dominant negative regulator by forming a heterodimer with GRα and annulling its function in transcriptional regulation. Both isoforms are ubiquitously expressed in most organs and tissues; co-expression of both forms renders a tight regulation of GR activity (Oakley et al., 1996; Yudt et al., 2003). Other variants produced from alternative splicing are GRγ, GR-A, and GR-P. These isoforms are expressed in various tissues, and are often associated with GC-resistance in cancer cells. Studies have shown that some of these isoforms may either stimulate or suppress GRα activity (Oakley et al., 2011). GR isoforms can also be generated through alternative translation initiation sites. GRα-A, -B, -C1-3, -D1-3 are all products from differential translation initiation, and differ from GRα in the length of NTD. These variants with truncated NTD are still capable of GC binding and transactivation, yet differ in their cellular localization and target genes. It is proposed that GRβ may have similar translational variants (Oakley et al., 2011). The various GR isoforms, which are distinct in their expression, localization, and sensitivity to GC, show the complexity in GR regulation and function. This study focused on human GRα (hGRα), the most abundant and active form of hGR.  3        Figure 1.1 Structure of GR. AF-1/2: activation function 1 or 2; DBD: DNA-binding domain; HR: hinge region; LBD: ligand-binding domain; NL1/2: nuclear localization signal sequence 1 or 2; NTD: N-terminal domain.  1.1.2 Mechanism of GR-mediated Transactivation and Transrepression GR, like the other nuclear receptors, is a transcription factor (TF) capable of binding ligands. In its unliganded state, GR is complexed with several proteins, such as the HSP90, HSP70 chaperones and immunophilins, which render it inactive, cytoplasm-bound, but with high binding affinity for GC (Pratt, 1993). Upon binding to GC, the receptor’s conformation changes, causing it to dissociate from the repressor complex and exposing its nuclear localization sequences. The released GR is then translocated into the nucleus, and there to carry out either its transactivation or transrepression functions (Fig. 1.2A) (Nicolaides et al., 2010). GR-mediated Transactivation In the nucleus, two monomers of liganded GR dimerize and activate gene transcription in numerous ways. In the simplest case, the receptor dimer recognizes and binds to the GRE on the promoter region of target genes. Upon DNA-binding, GR recruits coactivators and components 4  of BTM to the promoter and activates gene transcription (Fig. 1.2B). The consensus GRE, 5ꞌ RGnACAnnnTGTnCY-3ꞌ, is composed of 2 half sites of 6 base pairs (bp) palindromic repeat separated by a 3-bp spacer. However, actual GRE sequences of GR target genes diverge around the consensus, while these sequences are conserved among different species (Nicolaides et al., 2010; So et al., 2007).  Once bound to GRE, GR recruits numerous coactivators in a promoter-specific and cell type-specific manner. The p160 family coactivators, steroid receptor coactivator (SRC)-1, SRC-2, and SRC-3 contain histone acetyltransferase (HAT) and assist in chromatin modification to ensure the DNA template is accessible to BTM. The CREB-binding protein (CBP) is another important coactivator. Like the p160 coactivators, CBP contains intrinsic HAT activity, but can also act as an adaptor protein that serves to connect other coregulators and components of BTM to GR (Heitzer et al., 2007; McKay et al., 1999). More often, GR works in concert with other TF. GRE-bound GR dimer interacts with nearby TF, which is also bound to its cognate response element (TF RE) (Fig. 1.2C). Together, they recruit proper coactivators and cooperatively activate downstream transcription. In addition, GR can also carry out transactivation in a DNA-binding independent manner. In this method of transactivation, GR dimer tethers to other TF through protein-protein interaction without binding to GRE. The tethering of GR stimulates the expression of the target gene (Fig. 1.2D) (Newton, 2000). GR-mediated Transrepression Similar to transactivation, GR can also act in DNA-binding dependent and independent manner for negative transcriptional regulation. The response element for transrepression, 5  negative GRE (nGRE), varies greatly and has no apparent consensus sequence. While nGRE can act independently within the promoter of its target gene for transrepression, it can also be found within a response element of another TF and competes for access to the binding site, thereby repressing gene transcription (Fig. 1.2E) (Newton, 2000). Corepressors such as nuclear receptor corepressors (NCoR) 1 and NCoR2 may be involved in GR-mediated transrepression through nGRE (Surjit et al., 2011). More frequently, GR represses target genes through protein-protein interaction. Similar to GR-mediated transactivation, the receptor can interact with other TF in a composite or tethering manner (Fig. 1.2F, G). In this case, the interaction inhibits the TF from transcriptional activation and leads to repression of gene expression (Newton, 2000).               6                         B C GR BTM TF GRE TF RE D BTM TF RE GR TF BTM GR GRE GR Cytoplasm Nucleus Nuclear translocation Ligand binding A  GR  0 HSP90 E G F GR BTM TF RE TF GR BTM TF TF RE nGRE BTM GR nGRE 7  Figure 1.2 Mechanisms of GR-mediated transcriptional regulation. BTM: basal transcription machinery; GR: glucocorticoid receptor; GRE: glucocorticoid response element; nGRE: negative glucocorticoid response element; HSP90: heat shock protein 90; TF: transcription factor; TF RE: response element for a transcription factor. Nongenomic Effects The effects exerted by GC via GR are not limited to transcription-dependent effects at genomic level. Many effects mediated by GC are considered “nongenomic,” which is transcription-independent and does not involve new transcript or protein synthesis. Therefore, nongenomic effects are often rapid, occurring within minutes, as opposed to genomic effects, which may take hours or even up to days (Losel et al., 2003). The nongenomic effects can be carried out in a GR-independent or GR-dependent manner. In the GR-independent pathway, GC intercalates into cell membrane and alters membrane permeability or interacts with membrane proteins (Stahn et al., 2008). In GR-dependent pathways, GC binds and activates either cytosolic GR or membrane GR. The activated GR, however, does not translocate into the nucleus to regulate gene transcription. Rather, it interacts with other proteins or secondary messengers to stimulate or inhibit signaling pathways, producing a rapid response (Revollo et al., 2009; Stahn et al., 2008). An example would be the activation of phosphatidylinositol 3-kinase (PI3K) and Akt kinase by GR, which subsequently induces NO production by endothelial nitric oxide synthase (eNOS) for downstream cardiovascular protective effects (Hafezi-Moghadam et al., 2002). GR’s nongenomic activities are of particular interest, as most of the side effects of GC are mediated through genomic actions; developing GC that can selectively activate nongenomic effects may enhance the therapeutic usefulness of GC (Hafezi-Moghadam et al., 2002). 8  1.1.3 GR Function and Target Genes GR is a key transcription regulator of genes involved in important physiological processes. GC binds to GR in different tissues and organs such as brain, liver, bones, and immune cells, where the activated receptor mediates an array of biological responses including stress-related homeostasis, gluconeogenesis, bone remodeling, and anti-inflammation. Negative Feedback Regulation of Hypothalamic-pituitary-adrenal Axis The mechanism by which GR modulates stress response is through a negative feedback regulation of the hypothalamic-pituitary-adrenal (HPA) axis. Stimuli such as infection, injury, or emotional stress trigger the hypothalamus’ secretion of corticotrophin-releasing hormone (CRH), which then acts on anterior lobe of pituitary gland. The pituitary gland synthesizes a peptide hormone precursor, proopiomelanocortin (POMC), and processes it into adrenocorticotropic hormone (ACTH). ACTH in turn stimulates zona fasciculata of the adrenal cortex to produce and secrete cortisol. (Laryea et al., 2015; Newton, 2000). Cortisol binds to GR to regulate gene expression for dealing with stress, and at the same time, represses gene expression of CRH and POMC to downregulate the stress response (de Kloet et al., 2005).  The liganded GR acts on CRH promoter to repress its transcription via nGRE and methylation of the promoter (Malkoski et al., 1997; Sharma et al., 2013). In addition to downregulation of CRH, the receptor also suppresses the transcription of POMC. This is again achieved by binding to the nGRE of the POMC promoter, and also by inhibiting Nur77, a nuclear receptor for POMC transcription, in a composite manner (Drouin et al., 1993; Webster et al., 1999). Negative feedback regulation of HPA-axis is important, as long-term activation of stress response leads to prolonged elevated level of GC, which is detrimental to the body. 9  Therefore, it is necessary to reinstate the homeostasis within the body after stress (de Kloet et al., 2005; Patel et al., 2014). Gluconeogenesis One way the body deals with acute stress is by increasing available glucose supply through gluconeogenesis (Patel et al., 2014). Gluconeogenesis is the process through which the body synthesizes glucose from non-carbohydrate precursors such as pyruvate, amino acids, and glycerol. GC-liganded GR modulates gluconeogenesis through transcriptional upregulation of enzymes such as phosphoenolpyruvate carboxykinase (PEPCK), tyrosine aminotransferase (TAT), and adipose triglyceride lipase (ATGL).  PEPCK is one of the key enzymes catalyzing the rate-limiting step of gluconeogenesis. It is responsible for the conversion of oxaloacetate into phosphoenolpyruvate, which is synthesized into glucose through subsequent steps. GC is known to induce PEPCK expression through GR. Two GRE sites are located on the promoter of PEPCK (Imai et al., 1993). GR works in concert with many other TF such as hepatocytes nuclear factor-4 (HNF-4) and chicken ovalbumin upstream promoter transcription factor (COUP-TF), and recruits coactivators to activate the transcription of PEPCK (Hall et al., 1995). TAT is one of the enzymes responsible for amino acid catabolism, which catalyzes the deamination of the amino acid tyrosine into 4-hydroxyphenylpyruvate. The metabolism of tyrosine can be further carried out to produce carbon skeleton for glucose synthesis (Hagopian et al., 2003). Therefore, aside from amino acid metabolism, TAT is also responsible for converting tyrosine into precursors for gluconeogenesis. In the TAT promoter, proximal and distal GREs can 10  be found. The two sites work in a cooperative manner to upregulate TAT expression when bound by GR (Jantzen et al., 1987).   GR also regulates the expression of enzymes involved in lipid metabolism, which can provide glycerol as another substrate for gluconeogenesis. ATGL is a lipase responsible for lipolysis, and it is indirectly regulated by GR through forkhead box protein O1 (FoxO1) (Wang et al., 2012). GR increases the expression of FoxO1, a TF which in turn upregulates the expression of ATGL. ATGL hydrolyzes triglycerides and leads to increased influx of free fatty acids and glycerol into the liver, thus providing precursors for the synthesis of glucose via gluconeogenesis (Campbell et al., 2011; Patel et al., 2014). By increasing the expression of PEPCK, TAT, ATGL, and other related enzymes, GR activates the process of gluconeogenesis. Over-activation of gluconeogenesis caused by supraphysiological GC concentration can lead to hyperglycemia, an adverse effect associated with long-term GC treatment. Bone Remodeling GR is known to be involved in the process of bone remodeling. It regulates genes involved in proliferation and differentiation of both osteoblasts and osteoclasts, thereby affecting bone formation and resorption. GR exerts an antiproliferative effect on osteoblasts through transactivation of cyclin-dependent kinase (CDK) inhibitors p21 and p27 expression, which leads to G1 cell cycle arrest. The receptor also reduces osteoblast population by repressing the expression of the anti-apoptotic Bcl-2 gene (Rogatsky et al., 1999). In addition, GR can both activate and inhibit osteoblast differentiation (Moutsatsou et al., 2012). GR negatively regulates the expression of osteocalcin, which is responsible for osteoblast differentiation and bone mineralization, by binding to nGRE to suppress its transcription (Newton, 2000). However, it 11  was also demonstrated that GR can increase the transcription of bone morphogenic protein-6 (BMP-6), a growth factor that stimulates osteoblast differentiation (Liu et al., 2004). These studies showed GR’s modulation in the bone formation process.  GR also regulates bone resorption. It promotes the development of osteoclasts by modulating the receptor activator of NF-κB/RANK ligand/osteoprotegerin (RANK/RANKL/ OPG) system. GR increases RANKL expression, a ligand which binds to RANK on pre-osteoclasts, and activates its differentiation into osteoclasts. GR also decreases the expression of OPG, a decoy receptor which intercepts RANKL from interacting with RANK (Kondo et al., 2008). Endogenous GC at physiological concentration contributes in maintaining a balance in the process of bone remodeling via GR. However, high level of GC can disrupt the balance, and lead to decreased bone formation and increased bone resorption, causing GC-induced osteoporosis (Moutsatsou et al., 2012). Anti-inflammation Glucocorticoids have been long employed as anti-inflammatory agents for inflammatory and autoimmune disorders, such as rheumatoid arthritis, asthma, dermatitis, inflammatory bowel disease, etc. (Coutinho et al., 2011; Schäcke et al., 2007). The anti-inflammatory effects of GC are often attributed to its inhibition of other TF via GR. Nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) is a major TF that transcriptionally activates numerous proinflammatory genes including interleukin (IL)-1, IL-2, cyclooxygenase (COX)-2, tumor necrosis factor (TNF)-α, and inducible nitric oxide synthase (iNOS) (McKay et al., 1999). Early studies suggested that GC-liganded GR sequesters NF-κB through protein-protein interaction, and prevents it from binding to its cognate response element for transactivation. Later studies 12  showed that GR can also tether to DNA-bound NF-κB to inhibit transcription of proinflammatory genes (Laryea et al., 2015; McKay et al., 1999). In addition, GR is known to induce the expression of nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor, alpha (IκBα), a repressor which retains NF-κB in the cytoplasm. By elevating the level of IκBα, GR suppresses nuclear translocation of NF-κB for gene transcription (Newton, 2000).  Although GR-mediated transrepression of proinflammatory genes has been considered as the main mechanism through which GR exerts its anti-inflammatory effects, recent discoveries have shown that GR-mediated transactivation also contributes to anti-inflammation. Dual specificity protein phosphatase 1 (DUSP1), Annexin A1 (AnxA1), and glucocorticoid-induced leucine zipper (GILZ) are examples of anti-inflammatory genes upregulated by GR. DUSP1 inactivates mitogen-activated protein (MAP) kinase-regulated inflammatory signaling (Kassel et al., 2001); AnxA1 suppresses synthesis of lipid mediators such as prostaglandins and leukotrienes (Newton 2000); and similar to GR, GILZ inhibits the proinflammatory TF, NF-κB, through physical interaction (Ayroldi et al., 2009).  Furthermore, GR also influences the development of immune cells during the inflammation process. For example, GR can modulate the differentiation of dendritic cells into a hypo-responsive, tolerogenic state instead of immunogenic state. Also, it can induce monocyte differentiation into active phagocytic macrophages for the resolution of inflammation (Coutinho et al., 2011). Because GR modulates the expression of numerous genes involved in the inflammatory response, it remains one of the main therapeutic targets. However, severe side effects can develop from long-term GC administration. Therefore, researchers seek to develop drugs that would dissociate GR’s transrepression function from its transactivation function, in hope of 13  generating therapeutic agents that are equally efficacious as GC, yet with greatly reduced adverse reactions.  1.1.4 GR Ligands GR ligands are referred as “glucocorticoids” (GC). The term “glucocorticoid” is derived from the words “glucose,” “cortex,” and “steroid,” as these compounds are steroidal hormones produced from the adrenal cortex, and they are known to be involved in the regulation of glucose metabolism  (De Bosscher, 2010). The main GC in humans is cortisol, which activates GR to mediate myriads of physiological responses (Fig. 1.3A) (de Kloet et al., 2005). Because of GC’s anti-inflammatory and immunosuppressive activities, various synthetic GC were developed to improve efficacy and potency. Some examples of synthetic GC are prednisone (Fig. 1.3B), which has increased specificity for GR over MR, and dexamethasone, which has improved efficacy due to fluorination at C-9 position (Fig. 1.3C)  (Stahn et al., 2008). These compounds are used to treat chronic inflammatory and autoimmune diseases. Nevertheless, many harmful side effects, namely, hyperglycemia, osteoporosis, skin and muscle atrophy, glaucoma, Cushing’s syndrome, etc., are associated with long-term administration (Stahn et al., 2008).  Because it was observed that the anti-inflammatory effects are mainly executed through GR’s transrepression mechanism, whereas many of the side effects, for instance hyperglycemia, are caused by GR’s transactivation mechanism, many research began to focus on developing selective glucocorticoid receptor agonists (SEGRA), which are compounds that dissociate GR-mediated transrepression and transactivation. Some examples of SEGRA are RU24782 (RU782), RU24858 (RU858), AL-438, and compound A (CpdA) (De Bosscher, 2010; Schäcke et al., 2007). Some of these compounds (RU782 and RU858) are steroids resembling GC, while others 14  (AL-438 and CpdA) have non-steroidal structures that may be very different from GC (Fig. 1.3D-G). At present, SEGRA are still in preclinical phase (De Bosscher, 2010). The task of developing a fully dissociated compound with strong anti-inflammatory effect remains challenging, as many of the promising dissociative effects observed in vitro failed to be reproduced in vivo; many of the SEGRA still produce undesirable effects such as decreased bone mass and weight loss (De Bosscher, 2010; Schäcke et al., 2007). Moreover, in recent years it has been discovered that several anti-inflammatory responses are modulated by GR’s transactivation activity, while some of the adverse reactions can be mediated by GR’s transrepression activity. This adds another layer of complexity and difficulty in the development of SEGRA.               15                      Figure 1.3 Structures of GR ligands   A Cortisol B Prednisone C Dexamethasone D RU24782 E RU24858 F AL-438 G Compound A 16  1.2 Ginsenosides Ginseng is a popular medicinal herb that is widely consumed in many cultures. It is considered to be a panacea and an adaptogen, and it is frequently used as a body tonic to boost vitality and promote general well-being. Current estimate of the global market of ginseng roots and related processed products reached $2084 million (Baeg et al., 2013). Because of their popularity and many of the claimed medicinal properties, they also have been extensively investigated for their anticancer, antioxidant, antihyperglycemic, anti-inflammatory, central nervous system (CNS), and cardiovascular effects (Attele et al., 1999; Christensen, 2008).  The complex nature and diverse effects of ginseng are thought to be mediated by a myriad of pharmacologically active compounds known as ginsenosides. Ginsenosides are triterpenoid saponins that structurally resemble steroid hormones. Currently, approximately 150 ginsenosides have been isolated from various parts of the plant (Christensen, 2008). Majority of the ginsenosides have a dammarane steroidal backbone and are classified into three types based on differences in position of the carboxyl groups and the cyclization of the side chain: protopanaxadiol (PPD)-, protopanaxatriol (PPT)-, and ocotillol-type. In the root of Panax ginseng and Panax quinquefolius, the two most commonly consumed species of ginseng, PPD- and PPT-type ginsenosides dominate, and thus are more comprehensively investigated (Fig. 1.4). Some examples of PPD-type ginsenosides are Rb1, Rb2, Rc, Rd, Rh2, compound K (CK), and PPD. These ginsenosides have glycosyl groups such as glucose, rhamnose, arabinose, or simply a hydroxyl group attached to the C-3 and C-20 position. In comparison, PPT-type ginsenosides, such as Re, Rf, Rg1, Rh1, and PPT, have an additional sugar substituent at the C-6 position. PPD- and PPT-type ginsenosides differ not only in the position and types of sugar that are attached to their steroid core, but also in the number of sugar substituents. Ginsenosides such 17  as Rb1, Rb2, Rc, Rd, Re, Rf, and Rg1 are referred to as intact or multi-glycosylated ginsenosides, due to the multiple sugar moieties linked to the backbone. Ginsenosides such as CK, Rh1, and Rh2 are known as metabolite or monoglycosylated ginsenosides because they are the metabolic product of intact ginsenosides, and only a single sugar group remains on the backbone. Ginsenosides PPD and PPT are called aglycone ginsenosides because of the absence of sugar molecules. The number and position of the sugar groups are important in structure-activity relationship of ginsenosides and the biological responses they elicit. For instance, previous studies have shown that the anticancer efficacy of ginsenosides is inversely correlated with the number of sugar attachments (Nag et al., 2012).  In this study, multi-glycosylated ginsenosides Rb1, Rb2, Rc, Rd, Re, Rf, and Rg1, were investigated. These ginsenosides were selected because they are abundant in the roots of P. ginseng and P. quinquefolius, and account for over 90% of the total ginsenoside content (Christensen, 2008). In addition, metabolites of these ginsenosides, CK, Rh1, Rh2, and the aglycone PPD and PPT, were also examined.        18   Figure 1.4 Structures of selected ginsenosides. Glc: glucosyl; Ara(f): arabinofuranosyl; Ara(p): arabinopyranosyl; Rha: rhamnosyl.  1.2.1 Biotransformation and Pharmacokinetics While ginsenosides are considered as the major pharmacologically active compounds in ginseng, recent studies have shown that monoglycosylated and aglycone ginsenosides may be the predominant forms that reach systemic circulation after oral administration. In vitro studies have shown that one or more sugar moieties of ginsenosides may be cleaved off in a low pH environment. When incubated with 0.1 N hydrochloric acid at 37oC, an acidic environment similar to that within the stomach, multi-glycosylated PPT-type ginsenoside Rg1 was found to be deglycosylated at C-20 position into monoglycosylated Rh1 (Han et al., 1982). This implicates similar acid hydrolysis may occur within the stomach. Similar studies also had demonstrated the 19  hydrolysis of sugar moieties of multi-glycosylated PPD-type ginsenosides Rb1 and Rb2 at C-20 to form ginsenoside Rg3 (Karikura et al., 1991b; Yang et al., 2007). However, the biotransformaiton of PPD-type ginsenosides may be more complicated in vivo, as in rodent studies, Rb1 and Rb2 were found to be only slightly decomposed in the stomach, and were subjected to subsequent metabolism in intestines (Karikura et al., 1991a). Nevertheless, these data suggest that upon oral administration of ginseng, majority of the multi-glycosylated ginsenosides are further transformed into active metabolites to mediate downstream biological responses.  Other enzymatic studies demonstrated that intestinal microflora are also responsible for pre-systemic metabolism of ginsenosides into monoglycosylated or aglycone form (Hasegawa, 2004). Hasegawa et al. (1996) isolated multi-glycosylated ginsenosides Rb1, Rb2, Rc, Re, and Rg1 from standardized ginseng extract, and incubated these chemicals with human intestinal bacteria. This study showed multi-glycosylated PPD-type ginsenosides (Rb1, Rb2, and Rc) were metabolized into CK and PPD, whereas the PPT-type (Re and Rg1) were metabolized into Rh1 and PPT (Hasegawa et al., 1996). Bae et al. (200) also presented similar results when incubating Rb1 and Rb2 with human intestinal microflora. In addition, their data inferred that differences in gut microflora influenced the metabolic pathway of Rb1 and Rb2, which were converted to CK via different intermediate metabolites (Bae et al., 2000). Few pharmacokinetic studies had been performed in human, yet the results from these works corroborate the findings in vitro and in rodent models. In a pilot study, human volunteers were orally administered capsules of P. ginseng extract that mainly consisted of multi-glycosylated ginsenosides Rb1, Rb2, Rc, Rd, Re, Rf, and Rg1 (Tawab et al., 2003). Analysis of blood samples revealed that monoglycosylated PPT-type ginsenoside Rh1 was detected in 20  human plasma as early as within the first three hours post dosing. This implies that the multi-glycosylated PPT-type ginsenosides may be quickly absorbed from the upper gastrointestinal tract into the bloodstream soon after acid hydrolysis in the stomach (Tawab et al., 2003). Approximately after 7 to 8 hours of intake, monoglycosylated PPD-type ginsenoside CK was detected in the blood. The delay of metabolite detection in the plasma suggests that multi-glycosylated PPD-type ginsenosides may be metabolized and absorbed in the lower gastrointestinal tract into the bloodstream (Tawab et al., 2003). Contrary to previous studies in rats, ginsenoside Rb1 could be detected in blood and urine sample of one subject, showing that it is possible for multi-glycosylated ginsenosides to reach systemic circulation, and there are inter-individual variations in metabolism of ginsenosides (Tawab et al., 2003).  In another study, a ginsenoside Re tablet (200 mg) was orally administered to human subjects (Liu et al., 2011). In agreement to earlier findings, Rh1 and PPT were detected in both plasma and urine sample. Again, multi-glycosylated Re and Rg1 (as the intermediate metabolite of Re) could also be detected, indicating that ginsenosides are subjected to metabolism in the body. Figure 1.5 and 1.6 summarize proposed pathways for ginsenoside biotransformation.         21                        Figure 1.5 Proposed mechanism of PPD-type ginsenoside biotransformation Acid hydrolysis Stomach Intestines Rb1 Rh2 CK PPD Deglycosylation Deglycosylation Deglycosylation Rg3 22                        Figure 1.6 Proposed mechanism of PPT-type ginsenoside biotransformation Re Rg2 Deglycosylation Deglycosylation Rg1 Rh1 F1 PPT Deglycosylation Acid hydrolysis Stomach Intestines 23  1.2.2 Biological Activities Many effects of ginsenosides are currently under investigation. From cell line and animal model studies, ginsenosides are considered to have biological activities such as CNS, anticancer, anti-inflammatory, antioxidant, cardiovascular, and angiogenic activities. Among these, CNS, anticancer and anti-inflammatory effects are most extensively studied. Central Nervous System Effects Ginsenosides are thought to enhance memory and learning, provide neuroprotection, and modulate neurotransmission by acting on several receptors (Kim et al., 2013; Radad et al., 2011). Ginsenosides can affect central cholinergic system, which plays a role in learning and memory. For instance, Rb1 was shown to increase choline uptake, upregulate the expression of choline acetyltransferase, and thereby augment the release of acetylcholine in rats (Benishin, 1992; Salim et al., 1997), and ginsenosides Rh1, Rh2, and PPT were shown to enhance memory and learning ability of scopolamine-impaired mice (Wang et al., 2009; Yang et al., 2009). Ginsenosides are also implicated in neuroprotection. It was demonstrated that Re ameliorated damage in dopaminergic neurons in 1-methyl-4-phenyl-1,2,3,6-tetra- hydropyridine (MTPT)-treated mice through increasing and decreasing the expression of the antiapoptotic Bcl-2 and proapoptotic BAX proteins, respectively (Xu et al., 2005). Moreover, another study reported that Rh2 was able to increase neurotrophic factors and decrease Aβ-induced apoptosis in rat astrocytes (Shieh et al., 2008). Lastly, ginsenosides were shown to interact with numerous receptors in the CNS, such as the γ-aminobutyric acid subtype A (GABAA), γ-aminobutyric acid subtype B (GABAB), N-methyl-D-aspartate (NMDA) receptors, and voltage-dependent sodium, calcium, and potassium channels (Kim et al., 1998; Kimura et al., 1994; Radad et al., 2011). 24 Anticancer Effects Ginsenosides are considered as potential anticancer agents due to a diverse range of effects, among which, their antiproliferative, cytotoxic, and antimetastatic effects have been extensively examined. One study reported that Rh2 exerted its anti-proliferative effect by decreasing the expression of cell cycle-regulating proteins such as cyclin D1 and cyclin-dependent kinase (CDK) 6, leading to G1 arrest and growth inhibition in human lung cancer cells, A549. Furthermore, Rh2 caused cytotoxicity by increasing the expression of death receptor 4 (DR4), a receptor upstream of the caspase cascade for the induction of apoptosis (Cheng et al., 2005). Similarly, another report demonstrated the anti-estrogenic activity of PPD led to inhibition of cell proliferation in human breast cancer cells, MCF-7. In addition, it was shown that PPD acted synergistically with tamoxifen to inhibit tumor growth in an ER-independent manner (Yu et al., 2007). The antimetastatic effect of PPD and CK were also examined. These ginsenosides were shown to inhibit migration and invasion of human fibrosarcoma and murine colon carcinoma cells by suppressing the expression of key enzymes involved in metastasis, matrix metallopeptidase (MMP)-2 and MMP-9, respectively (Choo et al., 2008; Li et al., 2006). There are also many other effects of ginsenosides that contribute to ginsenosides’ anticancer activity, including induction of cell differentiation, suppression of telomerase reverse transcriptase (TERT) activity, and inhibition of multi-drug resistance (Nag et al., 2012). Anti-inflammatory Effects Many studies have examined ginsenosides’ anti-inflammatory effects in the context of neuroinflammation and inflammation related to cancer. Nevertheless, research has also showed their benefits in other types of inflammatory diseases. Kim et al. (2007) tested Rb1 in both cell 25  culture and rodent model. In the in vitro cell-based assays, peripheral mononucleated blood cells (PMBC), chondrocytes, and fibroblast-like synoviocytes were stimulated with interferon (IFN)-γ, LPS or IL-1β. Rb1 treatment decreased the TNF-α synthesis in stimulated cells. Moreover, when Rb1 was given prophylactically to collagen-induced arthritis (CIA) mice model, it was able to reduce the number of cells expressing TNF-α, alleviate damage of cartilage, and decrease inflammatory cell infiltration into the synovial tissue (Kim et al., 2007). Zheng et al. (2011) showed that treating oxazolone-challenged murine atopic dermatitis model with Rh1 can effectively ameliorate the development of skin lesions and reduce swelling in ears. Furthermore, Rh1 also significantly decrease infiltration of inflammatory cells, serum IgE and IL-6 level, and mast cell degranulation (Zheng et al., 2011). All these reports demonstrate ginsenosides’ potential in being employed as therapeutics to inflammatory diseases.  1.2.3 Ginsenosides’ Mechanisms of Action Ginsenosides are amphipathic compounds composed of a lipophilic steroid backbone and hydrophilic glycosyl substituents. As such, it is proposed that they can intercalate into plasma membrane. Insertion into the membrane is facilitated by the lipophilic steroid core, as well as the polar hydroxyl groups present on ginsenosides, which interact with the β-OH of cholesterol anchored in the phospholipid bilayers, and determine the insertion orientation (Attele et al., 1999). Once intercalated into the membrane, ginsenosides may alter the membrane fluidity or interact with membrane proteins and channels to elicit downstream cell responses. Studies also showed that ginsenosides can be actively transported into cells, and thereby exert its genomic effects inside the cells (Meng et al., 2007; Xie et al., 2005; Xiong et al., 2009). Because of its resemblance to steroid hormone, ginsenosides are thought to be capable of modulating steroid 26  hormone receptors such as ER and GR in the cytoplasm, and affecting their transcriptional regulation activities.  1.2.4 Ginsenosides as GRα Agonists To date, it has been reported that ginsenosides Rg1, Rh1, Rh2, and CK are rodent GR agonists capable of receptor activation in rodent cell lines (Du et al., 2011; Lee et al., 1998; Lee et al., 1997; Lee et al., 1996; Wu et al., 2012; Yang et al., 2008). Studies were also conducted in human cell lines and suggested that ginsenosides such as Re, Rg1, and CK may modulate hGRα (Du et al., 2011; Leung et al., 2006a; Leung et al., 2007; Leung et al., 2006b; Yang et al., 2008). Nevertheless, there are conflicting reports concerning ginsenosides’ ability to activate hGRα. Studies in Rodent Cell Lines Ginsenosides Rh1 and Rh2 Two monoglycosylated ginsenosides Rh1 and Rh2 were studied mouse testicular teratoma cell line (F9) and were shown to induce cell differentiation. At 50 μM, Rh1 and Rh2 were able to induce morphological changes and increased expression of laminin B1, a marker gene for cell differentiation (Lee et al., 1998; Lee et al., 1996). These effects were readily blocked by GR antagonist, RU486. To confirm that GR was involved in Rh1 and Rh2 induced cell differentiation, an electrophoretic mobility shift assay (EMSA) was performed in F9 cells. The data showed that Rh1 and Rh2 were able to induce the complex formation of nuclear proteins and radiolabeled GRE (32P-GRE) oligonucleotides. In addition, a band shift of the detected nuclear protein-32P-GRE complex was observed when mouse GR-specific antibody was 27  added, suggesting that GR was the nuclear protein induced by Rh1 and Rh2 to bind to GRE. Furthermore, the authors also heterologously transfected hGRα in F9 cells, and demonstrated the ability of Rh1 and Rh2 to induce hGRα nuclear translocation, and to cause a moderate increase of GRE promoter activity via GR in a luciferase reporter gene assay. Data from this study indicate that Rh1 and Rh2 are agonists capable of activating both mouse GR and hGRα. However, conflicting result was observed for Rh1 in another mouse cell line. Li et al. (2014) showed that Rh1 was capable of reversing downregulation of GR expression in mouse macrophage cells (RAW264.7) and thereby potentiating dexamethasone’s anti-inflammatory effect via mouse GR. Rh1 treatment (10 μM) alone was unable to activate GR to increase mRNA expression of DUSP1, a target gene for GR-mediated transactivation, nor was Rh1 able to inhibit NF-κB via GR for transrepression. However, when added in combination with dexamethasone, Rh1 enhanced dexamethasone-mediated NF-κB inhibition by GR and also increased the dexamethasone-mediated upregulation of DUSP1 mRNA. The discrepancy of between Lee et al., 1998 and Li et al., 2014 may be explained by differences in concentrations of Rh1 tested or by differences of cell type used in these studies. The notion that ginsenosides’ effect on GR can be cell-type specific may be further supported by the study of Lee et al., 2003b. In this study, the authors found that Rh1 acted as a weak phytoestrogen which could activate ER for promoter-binding and transcriptional activation of target gene. Specifically in one experiment, hER and hGRα were heterologously transfected into a monkey kidney cell line (CV-1) along with luciferase reporter gene vectors carrying corresponding response elements (ERE and GRE, respectively). While 50 μM of Rh1 was capable of activating hER to increase luciferase activity, Rh1 was unable to activate hGRα, in contrast to the results observed in Lee et al., 1998. 28 Ginsenoside Rg1 Rg1 is another ginsenoside that has been extensively studied and was one of the first ginsenosides to be established as an agonist of rodent GR. Lee et al. (1997) tested the compound’s ability to bind to endogenous rat GR by conducting a whole cell ligand-binding assay using a rat hepatoma cell line, FTO2B. They found that while Rg1 was able to inhibit the radiolabeled dexamethasone binding to the rat GR, 100- to 1000-fold higher concentration is required for Rg1 to have the same effect as the unlabeled dexamethasone. In addition, the authors also noted Rg1 could activate endogenous rat GR to stimulate transcription of GRE-containing luciferase reporter gene in a concentration-dependent manner. Similar to the binding assay, the EC50 of Rg1 for the luciferase reporter gene assay was two to three orders of magnitude higher when compared to dexamethasone. Addition of RU486 could block the effects of Rg1 (Lee et al., 1997). Du et al. (2011) studied Rg1’s ability to reduce inflammatory response and discovered that the chemical could decrease the production of proinflammatory cytokines and ameliorate damages caused by inflammatory responses in mice. In addition, they also found that Rg1’s anti-inflammatory effect was unaccompanied by detrimental effects to bones and blood glucose level. Specifically, Rg1 decreased the LPS-induced TNF-α and IL-6 secretion via NF-κB inhibition in a concentration-dependent manner. To show that GR was involved in the anti-inflammatory effect by Rg1, the authors demonstrated that 10 μM of Rg1 treatment was able to induce mouse GR nuclear translocation in RAW264.7 cells similar to dexamethasone (Du et al., 2011).  Wu et al. (2012) investigated Rg1’s neuroprotective effect in primary rat cerebrocortical neurons. The study demonstrated that 20 μM of Rg1 pretreatment was able to reduce β-amyloid (Aβ) peptide-induced neuronal death to a level that was comparable to dexamethasone, and the 29  protective effect was annulled by the GR antagonist, RU486. In addition, they also showed that Rg1 could decrease Aβ-stimulated expression of iNOS and thereby reduce the level of nitric oxide (NO), which has been reported to be involved in neuronal apoptosis. Again, this effect was blocked by RU486 and GR-targeting small interfering RNA (siRNA), implicating that Rg1 exerts its neuroprotective effect through GR (Wu et al., 2012).  To confirm that Rg1 indeed reduced neuronal death via GR, the authors performed several experiments. First, immunofluorescence imaging was performed and showed that 20 μM of Rg1 induced nuclear translocation of GR in primary rat cortical neuron cells. Then, gene expression study was conducted in rat neurons and demonstrated that Rg1 was able to increase the expression of two GR target genes, serum and glucocorticoid-regulated kinase 1 (SGK) and DUSP1, by 4.0- and 3.0-fold, respectively. Finally, an in vitro competitive ligand-binding assay was employed and the data suggested that Rg1 was able to compete with a commercial GR ligand and bind to hGRα (Wu et al., 2012). Ginsenoside CK Ginsenoside CK was also reported to be a mouse GR agonist (Yang et al., 2008). It was demonstrated that CK exerts anti-inflammatory and immunomodulatory effects by inhibition of Toll-like receptor 4 (TLR4) pathway via GR. CK at 10 μg/ml (~16.1 μM) was able to decrease LPS-stimulated TNF-α production by 80% and lower NO production by 60 % in RAW264.7 cells. Pretreatment with RU486 or GR-targeting siRNA reverted these effects. In a concentration-dependent manner, CK treatment of murine bone marrow-derived macrophages (BMBD) cells also reduced LPS-induced phosphorylation and activation of p38 and pERK proteins, which are involved in the TLR4-mediated inflammatory pathway. Whole cell ligand-30  binding assay and chromatin immunoprecipitation (ChIP) assay also demonstrated that CK was able to bind to the mouse GR and it effectively inhibited NF-κB from interacting with interferon regulatory factor (IRF) and its recruitment to interferon stimulated response element (ISRE) in the promoter of the pro-inflammatory genes. Studies in Human Cell Lines Ginsenoside CK In addition to their study of CK and its anti-inflammatory effect via mouse GR, Yang et al. (2008) also investigated CK’s effect via human GR, and conducted experiments using human monocytes and human embryonic kidney cells (HEK293T). In agreement with data from mouse cell lines, CK (10 μg/ml; ~16.1μM) significantly reduced the LPS-induced secretion of proinflammatory cytokines, TNF-α and IL-6, in human monocytes. When NF-κB-responsive luciferase reporter gene was transiently transfected in TLR4-expressing HEK293T cells, CK repressed NF-κB activation of reporter gene in a concentration-dependent manner (1-16 μg/ml; ~1.6-25.7 μM). To establish that GR is involved in the observed suppression of proinflammatory factors, the authors demonstrated that CK was capable of binding to full-length hGRα using an in vitro fluorescence polarization (FP)-based competitive ligand binding assay at high concentrations (1 μM-1 mM). Moreover, CK’s ability to stimulate hGRα transactivation was confirmed through GR-responsive luciferase reporter gene assay conducted in HEK293T cells transfected with hGRα. As expected, CK was shown to increase luciferase activity in a concentration-dependent manner.  31 Ginsenosides Rg1 and Re The aforementioned study by Du et al. (2011) also tested the anti-inflammatory effect of Rg1 in a human lung cancer cell line (A549) in addition to their assays performed in the rodent cell line. A549 cells were transiently transfected with NF-κB-responsive luciferase reporter vector and the effect of Rg1 on hGRα-mediated transrepression was assayed. Both 1 μM of dexamethasone and 10 μM of Rg1 significantly suppressed the luciferase activity by approximately 40-45% compared to the vehicle control. The transrepressive effect was annulled when siRNA was introduced to knock down the expression of endogenous hGRα in A549. Furthermore, at 10 μM, Rg1 promoted translocation of GR from cytoplasm to nucleus. These data suggest that Rg1 could activate hGRα to mediate anti-inflammatory effects. Three studies investigated ginsenosides Re’s and Rg1’s effects on angiogenesis and discovered that Re and Rg1 mediated their proangiogenic effects in human umbilical vein endothelial cells (HUVEC) through nongenomic, GR-dependent PI3K/Akt pathway (Leung et al., 2006a; Leung et al., 2007; Leung et al., 2006b). The PI3K/Akt pathway regulates cell proliferation and is known to activate eNOS through phosphorylation. The activated eNOS then produces NO, which has been reported to be involved in angiogenesis through unknown mechanisms. Treatment with 150 nM of Rg1 increased the expression of phosphorylated Akt, PI3K, and eNOS to a level that is comparable to dexamethasone (500 nM) within 15-30 min, and elevated NO production in HUVEC within 60 min. These effects were abolished when either RU486 or GR-targeting siRNA was introduced. The relatively short response time to the Rg1 treatment suggests these effects were nongenomic via GR, i.e. did not involved GR transcriptional regulation. In addition, Rg1 (150 nM) was shown to upregulate protein expression 32  of a proangiogenic growth factor, vascular endothelial growth factor (VEGF), and the effect was annulled by addition of RU486. Similarly, Re, another PPT-type ginsenoside, was able to elevate the level of phosphorylated eNOS and enhance NO production in HUVEC from a concentration of 250-1000 nM. These effects were also reversed by RU486 and GR-siRNA. Moreover, Re treatment was able to stimulate calcium ion (Ca2+) influx and transiently increase its intracellular concentration, which led to eNOS activation (Leung et al., 2007). To confirm that Rg1 and Re mediated these proangiogenic effects through hGRα, the authors assayed Rg1’s and Re’s capacity to bind to hGRα through a competitive ligand-binding assay using the FP-based technique (Leung et al., 2006a; Leung et al., 2007). They found that Rg1 and Re were able to compete with a fluorescent hGRα ligand and bind to the full-length receptor. Moreover, the data also showed that Rg1 was capable of inducing hGRα nuclear translocation, and Re stimulated GR-mediated transactivation of luciferase reporter gene at low concentrations (150 nM and 240 nM, respectively) (Leung et al., 2006a; Leung et al., 2007). Conflicting Reports Albeit different studies showed the capability of various ginsenosides in hGRα activation, a few studies presented conflicting results (Lee et al., 2003a; Lee et al., 2003b; Lee et al., 2003c; Ling et al., 2005). One study investigated the effects of total ginseng saponins (GSS) in human liver cells (HL7702) (Ling et al., 2005). The GSS were extracted from leaves and stems of P. ginseng, and using high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS), the content of four major ginsenosides were characterized, which contained 20.9% of Rg1, 18.5% of Re, 5.7% of Rb1, and 2.4% Rb2. In this study, the authors found that GSS alone 33  did not have an effect on hGRα activation. However, the extract was able to potentiate dexamethasone-increased hGRα activity. Briefly, HL7702 cells were transiently transfected with GRE-carrying luciferase reporter vector. Treating the transfected cells with various concentrations of GSS (6.25-100 μg/ml) did not increase the activity, whereas 10 nM of dexamethasone treatment lead to an increase of 55-fold. When adding GSS (25 μg/ml) along with dexamethasone, GSS was able to potentiate its activation of GR by 1.5-fold compared to dexamethasone alone. Using reverse transcription PCR and Western blotting, it was discovered that GSS was able to partially reverse the downregulation of GR mRNA and protein expression caused by dexamethasone. Another three studies investigated the effects of Rb1, Rh1, and Rc and Re on endogenous hER in human breast cancer cells (MCF-7), and on transfected hER in a monkey kidney cell lines (CV-1) (Lee et al., 2003a; Lee et al., 2003b; Lee et al., 2003c). In addition, the authors also briefly examined these ginsenosides’ effect on other human steroid hormone receptors such as GR, AR, retinoic acid receptor (RAR), to determine whether the observed estrogenic effect by ginsenosides were solely via hER. To this end, hGRα, hAR, and hRAR were transiently and heterologously expressed in CV-1 cells, along with luciferase reporter vectors carrying the corresponding response elements: GRE, ARE, and β2RARE. It was found that none of the ginsenosides tested were able to activate hGRα in CV-1 cells. This finding is especially interesting, since in Lee et al., 1997 and Leung et al., 2007 Rh1 and Re were shown to activate hGRα in a luciferase reporter gene assay.  34  1.3 Rationale of the Study It has been reported that ginsenosides Rg1 and CK are capable of activating hGRα (Du et al., 2011; Leung et al., 2006a; Leung et al., 2006b; Wu et al., 2012; Yang et al., 2008). In contrast, Rb1, Rc, Rh1, and total ginseng saponins, which contained multiple ginsenosides, had no effect on hGRα activation (Lee et al., 2003a; Lee et al., 2003b; Lee et al., 2003c; Ling et al., 2005). Conflicting results were reported concerning Re’s ability to modulate hGRα (Leung et al., 2007). Many other major ginsenosides, such as Rb2, Rd, Rf, Rh2, PPD, and PPT, are yet to be investigated. It is likely that these ginsenosides activate the receptor in an analog-selective manner. Moreover, ginsenosides that were studied previously were only tested for their effects mainly on the receptor’s transactivation activity. Therefore, this study aimed to investigate the effect and selectivity among different major ginsenosides on hGRα modulation, and both the transactivation and transrepression activities of the receptor were examined.  1.4 Research and Experimental Hypotheses The research hypothesis of this study was that ginsenosides can modulate hGRα-mediated transcriptional regulation in an analog-selective manner. The experimental hypotheses were that 1) in an analog-selective manner, selected ginsenosides are able to compete with an hGRα ligand for LBD-binding in a competitive ligand-binding assay; 2) selected ginsenosides will increase or attenuate hGRα-mediated transactivation and transrepression in LS180 cells in dual-luciferase reporter gene assays, and 3) selected ginsenosides will increase hTAT and decrease hCBG target gene expressions in HepG2 cells as assessed by real-time PCR.  35  1.5 Specific Aims The specific aims of this study are listed as followed: 1. To determine non-cytotoxic concentrations of ginsenosides in cultured LS180 cells using LDH cytotoxicity detection assay for subsequent cell-based assays. 2. To examine whether ginsenosides are able to bind to the LBD of hGRα through TR-FRET competitive ligand-binding assay. 3. To investigate whether ginsenosides are capable of activating or attenuating hGRα-mediated transactivation and transrepression activity in cell-based luciferase reporter gene assay. 4. To assess the effect of ginsenosides on hGRα target gene expression by real-time PCR. 36  2. Materials and Methods  2.1 Chemicals and Reagents Ginsenosides Rb1, Rb2, Rc, Rd, Re, Rf, and Rg1 (purity > 98%) were purchased from INDOFINE Chemical Company, Inc. (Hillsborough, NJ, U.S.A.), whereas ginsenosides Rh1, Rh2, compound K, and 20(S)-protopanaxatriol (purity > 98%) were from ChromaDex, Inc. (Irvine, CA, USA). 20(S)-protopanaxadiol (purity > 98%) was bought from Quality Phytochemicals, LLC (East Brunswick, NJ, U.S.A.). Dexamethasone, dextran, dimethyl sulfoxide (DMSO), mifepristone (RU486), SR12813, testosterone, and Triton X-100 were purchased from Sigma-Aldrich (St. Louis, MO, U.S.A.). Compound A (CpdA) was ordered from Calbiochem, Merck Millipore (Darmstadt, Hesse, Germany). Minimum essential medium, Earle’s balanced salt solution (MEM-EBSS), non-essential amino acids (NEAA), charcoal-stripped heat-inactivated fetal bovine serum (CS-HI-FBS), L-glutamine, penicillin G, and streptomycin were bought from HyClone, Thermo Fisher Scientific (Nepean, ON, Canada). Opti-MEM, Gibco FBS, AmbionTM PureLink RNA Mini Kit, AmbionTM PureLink®  DNase Set, Superscript®  III reverse transcriptase, Quant-iTTM PicoGreen®  dsDNA Assay Kit, diethylpyrocarbonate (DEPC) water, UltraPure™ DNase/RNase-Free Distilled Water, and LanthaScreen TR-FRET GR Competitive Binding Assay Kit were obtained from InvitrogenTM, Life Technologies (Burlington, ON, Canada). FuGENE®  6 transfection reagent and Dual-luciferase Reporter Gene Assay System were purchased from Promega (Madison, WI, U.S.A.). LDH Cytotoxicity Detection Kit was ordered from Roche Diagnostics (Laval, QC, Canada). EvaGreen 2X qPCR MasterMix-ROX was purchased from Applied Biological Materials (Richmond, BC, Canada). MicroAmp®  Optical 96-Well Reaction Plate and MicroAmp®  Optical 37  Adhesive Film were purchased from Applied BiosystemsTM, Life Technologies (Burlington, ON, Canada). Forward and reverse primers for hTAT, hCBG and hHPRT were synthesized at Integrated DNA Technologies (Coralville, IA, U.S.A.). pCMV6-XL5 and pCMV6-XL5-hGR were obtained from OriGene Technologies, Inc. (Rockville, MD, U.S.A.). PathDetect GRE Cis-Reporting System (pGRE-luc) was purchased from Agilent Technologies (Santa Clara, CA, U.S.A.). pGL4.32 [luc2P/NF-κB-RE/Hygro] (NFkB-luc) and pGL4.74[hRluc/TK] (Rluc) were ordered from Promega (Madison, WI, U.S.A.)  2.2 Cell Lines and Cell Culture Human colorectal adenocarcinoma (LS180) and human hepatocellular carcinoma (HepG2) cells were ordered from American Type Culture Collection (Manassas, VA, U.S.A.). Cells were cultured in MEM-EBSS supplemented with 1X NEAA, 2 mM L-glutamine, 100 U/ml penicillin G, 100 μg/ml streptomycin, and 10% FBS in T-75 flasks and grown in a 37oC incubator supplied with 5% CO2. Cell culture media were changed every 3 days and cells were subcultured every 5-6 days. Passages 6-20 were used for all cell-based assays.   2.3 Lactate Dehydrogenase Cytotoxicity Detection Assay LS180 cells were cultured in MEM-EBSS supplemented with abovementioned reagents with 10% CS-HI-FBS in place of FBS, and seeded at a density of 1 × 105 cells per well in a 24-well microplate. Forty-eight hours (h) post-seeding, culture medium was aspirated and cells were treated with fresh culture medium containing a ginsenoside (Rb1, Rb2, Rc, Rd, Re, Rf, Rg1, Rh1, Rh2, CK, PPD, or PPT; 10, 30 or 60 µM), DMSO (0.1% v/v; vehicle control), Triton X-100 (0.1% v/v; positive control), or dextran (1% w/v; negative control) for 24 h. Following the 38  treatment, culture medium was removed from each well, and cells were lysed using lysis buffer (2% v/v Triton X-100 and 20 mM EDTA in phosphate-buffered saline, pH 7.4) and homogenized by shaking at 650 rpm for 2 h on a plate shaker (Jitterbug-4, Boekel Scientific). The activity of lactate dehydrogenase (LDH) in the medium and in the cell lysates were measured at 492 nm using Ascent MultiskanTM absorbance plate reader. Data are expressed as the percentage of LDH released in medium over the total LDH content (medium plus lysate) from each well, and shown as mean  SEM of three independent experiments performed in triplicates.  2.4 Time-resolved Fluorescence Resonance Energy Transfer Competitive Ligand-binding Assay Glutathione S-transferase (GST)-tagged hGRα-LBD (0.5 nM) complexed with terbium-labeled anti-GST antibody (2 nM) provided in the LanthaScreen®  Time-resolved Fluorescence Resonance Energy Transfer (TR-FRET) GR Competitive Binding Kit (Life Technologies) was incubated with FluoromoneTM GS1 Green (GR ligand labeled with fluorescein; 5 nM) and one of the following: a ginsenoside (Rb1, Rb2, Rc, Rd, Re, Rf, Rg1, Rh1, Rh2, CK, PPD, or PPT; 100 μM), DMSO (1% v/v; vehicle control ), SR12813 (5 μM; negative control), or dexamethasone (5 μM; GR agonist) for three hours at room temperature in the dark. The dose-response experiments were performed at concentrations of 0.001, 0.01, 0.1, 1, 10, 100, 300, 1000, and 5000 nM for dexamethasone; 0.1, 1, 3, 10, 30, 60, 100, and 500 μM for ginsenosides Rb1, Rb2, Rc, Rd; and 0.1, 1, 3, 10, 30, 60, and 100 μM for ginsenosides CK and Rh2. The Tb-label conjugated to GR-LBD acts as a donor molecule in fluorescence resonance energy transfer at an excitation wavelength of 340 nm and an emission wavelength of 495 nm. 39  The fluorescein-label on FluormoneTM GS1 acts as an acceptor molecule, which when bound to GR-LBD, is excited by Tb emission and the fluorescein emission is subsequently detected at 520 nm. When a test compound is able to bind to GR-LBD and competes the fluorescent ligand off the receptor, a decrease of fluorescein emission is observed. In my study, the TR-FRET signals were detected at the aforementioned wavelengths using a BMG LABTECH PHERAstar FS plate reader, and delay time and integration time were set to 100 μs and 200 μs, respectively.  TR-FRET ratio was calculated by dividing the fluorescence signal of fluorescein by that of terbium. Subsequently, a net TR-FRET ratio was obtained by subtracting the TR-FRET ratio of each reaction from the background ratio value (vehicle control without human GRα-LBD added). For concentration response curves, One Site Competition Model of Sigma Plot 11.0 was used for curve fitting and generation of IC50 values. Results are expressed as a percentage of net TR-FRET ratio in the vehicle control, and shown as mean  SEM of three independent experiments conducted in triplicate.    2.5 Transient Transfection LS180 cells were cultured in MEM-EBSS supplemented with 1X NEAA, 2 mM L-glutamine, 100 U/ml penicillin G, 100 μg/ml streptomycin, 10% CS-HI-FBS, and seeded at a density of 1X105 cells per well in a 24-well microplate. Twenty-four h post-seeding, cells were subjected to transient transfection using FuGENE6® transfection reagent. In the hGRα-mediated transactivation assay, cells were transfected with a firefly luciferase reporter vector carrying glucocorticoid response element (pGRE-luc; 100 ng), a Renilla reniformis luciferase reporter vector, which is an internal control for normalization (Rluc; 1 ng), and either pCMV-XL5 (empty vector; 100 ng) or pCMV6-XL5-hGR (hGRα-containing vector; 100 ng). FuGENE6®  40  transfection reagent and the plasmid DNA were diluted in Opti-MEM in a 1 μg DNA: 3 μl reagent ratio, the mixture was then added into each well. Twenty-four h following transfection, the culture medium was removed and fresh culture medium containing either a ginsenoside (Rb1, Rb2, Rc, Rd, Re, Rf, Rg1, Rh1, Rh2, CK, PPD, or PPT; 30 or 60 µM, as indicated in each figure legend), DMSO (0.1% v/v; vehicle control), SR12813 (10 µM negative control), RU486 (100 nM; hGRα antagonist), dexamethasone (100 nM; hGRα agonist), or both dexamethasone (100 nM) and RU486 (100 nM). At the end of the 24 h treatment period, cells were lysed and the luciferase assay was performed. In the hGRα-mediated transrepression assay, cells were transfected with a firefly luciferase reporter vector carrying NF-κB response element (NFkB-luc; 50 ng), Rluc for internal control (10 ng), and either pCMV-XL5 or pCMV6-XL5-hGR (100 ng) using FuGENE6®  transfection reagent (1 μg DNA: 3 μl reagent ratio). Twenty-four h later, transfection medium was aspirated, and cells were treated with 5 ng/ml of TNF-α and one of the following: a ginsenoside (Rb1, Rb2, Rc, Rd, CK, or Rh2; 30 or 60 µM, as indicated), DMSO (0.1% v/v; vehicle control), dexamethasone (500 nM; hGRα agonist), RU486 (10 µM; hGRα antagonist) or both dexamethasone (500 nM) and RU486 (10 µM). At the end of 5 h treatment period, cells were lysed for luciferase activity measurement. Cells were treated for 5 h before cell lysis for subsequent assay.  2.6 Dual-luciferase Reporter Gene Assay Treatment medium was removed at the end of indicated treatment period, and cells were lysed using Passive Lysis Buffer supplied in the Dual-luciferase Reporter Gene System. Cell lysates were transferred to a 96-well microplate and firefly luciferase activity and Renilla 41  luciferase activity were measured in a GlomaxTM 96 Microplate Luminometer. Results are expressed as ratio of firefly luciferase activity normalized to Renilla luciferase activity and shown as mean  SEM of three independent experiments conducted in triplicate.     2.7 HepG2 Cell Culture and Treatment for Gene Expression Study HepG2 cells were cultured in were cultured in MEM-EBSS supplemented with 1X NEAA, 2 mM L-glutamine, 100 U/ml penicillin G, 100 μg/ml streptomycin, 10% CS-HI-FBS, and plated in 24-well microplates at a density of 1x105 cells per well. The culture medium was aspirated 24 h post-seeding, and replaced with fresh culture medium containing drug treatment.  For hTAT gene expression study, cells were treated with DMSO (0.1% v/v; vehicle control), ginsenoside Rh2 (30 μM), RU486 (100 nM; hGRα antagonist), dexamethasone (10 nM; hGRα agonist), or a combination of either dexamethasone and ginsenoside Rh2 (10 nM and 30 μM, respectively) or dexamethasone and RU486 (100 nM each). For hCBG gene expression study, cells were treated with DMSO (0.1% v/v; vehicle control), ginsenoside Rh2 (30 μM), testosterone (10 μM), CpdA (1 μM; hGRα agonist), or a combination treatment of either CpdA and ginsenoside Rh2 (1 μM and 30 μM, respectively) or CpdA and testosterone (1 μM and 10 μM, respectively). Cells were treated for 24 h before total RNA isolation.  2.8 Isolation of Total RNA At the end of the treatment period, cells were lysed using the lysis buffer provided in the AmbionTM PureLink®  RNA Mini Kit. The lysate was passed through a 21G needle attached to a syringe for 5-10 times to homogenize the sample. RNA isolation was performed according to the manufacturer’s protocol using a column-based method. During column purification, the isolated 42  RNA samples were subjected to on-column DNase treatment using the AmbionTM PureLink®  DNase Set to remove any genomic DNA contamination following the provided instructions. The DNase-treated and purified RNA was eluted with 50 μl of RNase-free water (provided in the kit).  The concentration of total RNA was quantified using the Thermo Scientific NanoDrop 2000 UV-Vis Spectrophotometer by measuring absorbance at 260 nm. Absorbance at 280 nm and 230 nm were also measured; the ratio of absorbance at 260 nm and 280 nm (260/280), and at 260 nm and 230 nm (260/230) were taken to assess RNA sample purity.  2.9 Reverse Transcription and cDNA Quantification RNA templates (2.5 μg) were reversely transcribed using SuperScript®  III reverse transcriptase (RT). First, the template, 0.5 μg of oligo(dT)12-18 primer, and 1 μl of 10mM dNTP were incubated in the presence of DEPC water (added to make a reaction volume of 12 μl) at 65oC for 5 min. The mixture was then chilled on ice, and 8 μl of SuperScript® III reverse transcriptase (RT) master mix (containing 200 unit of RT in the presence of 1X reaction buffer and 1 μl of 0.1 M DTT) was added for each reaction. The reaction was incubated for 60 min at 50oC, followed by 15 min at 70oC for enzyme inactivation. cDNA quantification was performed using Quant-iTTM PicoGreen®  dsDNA Assay Kit according to the manufacturer’s protocol. Briefly, DNA standards of 0, 50, 100, 200, 400, 800 ng/ml were made by diluting the provided λ-phage DNA in TE buffer. To quantify cDNA concentration, cDNA samples were diluted 50X in TE buffer. A 200X diluted Picogreen®  dye solution, a fluorescent dye which is capable of binding to double-stranded nucleic acids, was freshly prepared with TE buffer, and added to standards and cDNA samples at a 1:1 volume ratio in a black 96-well microplate. The reactions were incubated at room temperature for 5 min 43  protected from light. The Picogreen®  fluorescence was then detected (excitation wavelength 485 nm; emission wavelength 527 nm) in a BMG PHERAstar FS plate reader.  2.10 Primers and Real-time Polymerase Chain Reaction hTAT mRNA was amplified using forward primer 5ꞌ-CTG-AAG-TTA-CCC-AGG-CAA-TGA-AAG-3ꞌ and reverse primer 5ꞌ-TAA-TAA-GAA-GCA-ATC-TCC-TCC-CGA-C-3ꞌ were used. hCBG was amplified using forward primer 5ꞌ-CAC-CAA-CCA-GGC-AAA-TTT-CT-3ꞌ and reverse primer 5ꞌ-GGA-CGT-CAG-GTT-TAG-GGT-GA-3ꞌ were used. Human hypoxanthine-guanine phosphoribosyltransferase (hHPRT) was employed as the endogenous control and was amplified using forward primer 5ꞌ-GAA-GAG-CTA-TTG-TAA-TGA-CC-3ꞌ, and reverse primer 5ꞌ-GCG-ACC-TTG-ACC-ATC-TTT-G-3ꞌ.  Each PCR reaction contained either 10 ng (hTAT mRNA determination) or 5 ng (hCBG mRNA determination) of cDNA, 300 nM each of forward and reverse primers, 1X EvaGreen qPCR MasterMix, and UltraPureTM DNase/RNase-free distilled water (PCR water) added to make a reaction volume of 20 μl. The PCR reaction mixtures were loaded onto MicroAmp®  optical 96-well reaction plate and sealed with MicroAmp®  optical adhesive film. Real-time PCR analysis was conducted in Applied BiosystemsTM StepOnePlusTM real-time PCR system. The reaction plate was first incubated at 95oC for 10 min (enzyme activation for the HotStart Taq in EvaGreen qPCR mastermix), then 40 cycles of 95oC denaturation for 15 s, followed by 60oC annealing/extension for 1 min. Relative mRNA expression level was calculated using the comparative Ct (ΔΔCt) method in which the relative abundance of the transcript is expressed as 2-ΔΔCt, where ΔΔCt = (Ct Target – Ct Endogenous )Treatment – (Ct Target – Ct 44  Endogenous )DMSO (Livak et al., 2001). Data are expressed as mean  SEM of three independent experiments performed in triplicate.  2.11 Statistical Analysis Three independent experiments were conducted for each assay.  Data are expressed as mean  SEM. For cell-based assays, cells from a specific passage were used for each independent experiment. SigmaPlot 11.0 was employed for statistical analyses; either one-way or two-way analysis of variance (ANOVA) was performed followed by Student-Newman-Keuls post-hoc test. Statistical significance was set at 0.05; a p-value of less than 0.05 was considered as statistically significant.  45  3. Results  3.1 Determination of Non-cytotoxic Concentrations of Ginsenosides in Cultured LS180 Cells  LDH cytotoxicity detection assay was conducted to determine a non-cytotoxic concentration range of ginsenosides in cultured LS180 cells. At 10 μM and 30 μM, all tested PPD- and PPT-type ginsenosides were not cytotoxic. The percentage of LDH release from these ginsenosides were comparable to culture medium and DMSO (vehicle controls), and dextran (negative control). In contrast, Triton X-100 (positive control) yielded 99.8  0.1% of LDH release (Fig. 3.1A and B). At 60 μM, monoglycosylated ginsenosides CK and Rh2 the LDH release was 95.5  1.6% and 87.2  12.0%, respectively. By comparison, the LDH release was 21.5  1.5% in LS180 cells treated with 60 μM of the aglycone ginsenoside PPD (Fig. 3.1C). Therefore, all ginsenosides were tested at either 30 or 60 μM in the subsequent cell-based assays.  3.2 In vitro Binding of Ginsenosides to hGRα Ligand-binding Domain To investigate whether the ginsenosides can bind to the ligand-binding domain (LBD) of hGRα, an in vitro TR-FRET competitive ligand-binding assay was performed. In the first part of this experiment, all ginsenosides were screened at a single high concentration of 100 μM. Such a high concentration was chosen because in a preliminary study of a selected ginsenoside, no significant effect in receptor activation was observed when it was tested at lower concentrations.  (Appendix A). At 100 μM, all ginsenosides were able to compete with FluormoneGS1TM, a fluorescein-labeled hGRα ligand. Among the ginsenosides, glycosylated PPD-type ginsenosides (Rb1, Rb2, Rc, Rd, CK and Rh2) had greater effect in displacing the fluorescent ligand from the 46  LBD, and caused a decrease in net TR-FRET ratio by 35 to 66%. In comparison, the aglycone PPD and PPT-type ginsenosides (Re, Rf, Rg1, Rh1, and PPT) only decreased the net TR-FRET ratio by less than 30%, suggesting weaker binding to the LBD of the receptor. As expected, SR12813 (negative control) was unable to compete with the fluorescent ligand and no change in net TR-FRET ratio was observed, whereas dexamethasone (hGRα agonist) exhibited strong binding to the receptor, and caused a decrease of more than 98% in the net TR-FRET ratio (Fig. 3.2A).   In the second part of the experiment, dexamethasone and glycosylated PPD-type ginsenosides were selected, and concentration-response curves were characterized for these compounds (Fig. 3.2B-H). Table 3.1 summarizes the IC50 values obtained from the concentration-response curves. By comparison, multi-glycosylated ginsenosides Rb1, Rb2, Rc, and Rd having a much larger IC50  values (60-100 μM) than monoglycosylated ginsenosides CK and Rh2 (39.8  9.7 μM and 15.0  1 μM, respectively). An IC50 of 24.1  5.5 nM was obtained for dexamethasone (positive control).  Table 3.1. IC50 of dexamethasone and glycosylated PPD-type ginsenosides 47  3.3 Ginsenosides in hGRα-mediated Transactivation To determine whether ginsenosides activate hGRα-mediated transactivation, a dual-luciferase reporter gene assay was conducted in LS180 cells, and a firefly luciferase reporter vector containing four copies of GRE (pGRE-luc) was employed. Compared to the DMSO-treated group, 30 μM of ginsenosides did not increase luciferase activity whereas dexamethasone (hGRα agonist) at 100 nM activated hGRα and increased luciferase activity by 7.4  2.4-fold. Both SR12813 (negative control) and RU486 (hGRα antagonist) behaved as expected. Furthermore, RU486 decreased the luciferase activity to control level. The empty vector did not increase luciferase activity, indicating that the effect was receptor-specific (Fig. 3.3A).  The multi-glycosylated PPD-type ginsenosides Rb1, Rb2, Rc, and Rd were tested at a greater concentration (60 μM). However, none of them affected hGRα transactivation activity, whereas dexamethasone increased the luciferase activity by 5.4  0.2-fold. All the other controls yielded expected results (Fig. 3.3 B).  3.4 Glycosylated PPD-type Ginsenosides in hGRα-mediated Transrepression To investigate whether glycosylated PPD-type ginsenosides can increase hGRα transrepression activity, a firefly luciferase reporter vector containing NF-κB response element (NFkB-luc) was employed in a dual-luciferase reporter gene assay. First, factors such as concentration of dexamethasone and TNF-α (an NF-κB activator) treatment, and amount of pCMV6-XL5-hGR transfection were characterized in order to optimize the assay conditions. Dexamethasone at 500 nM was sufficient to decrease the luciferase activity by 55.1  4.5% compared to vehicle control, whereas increasing the concentration to 1000 nM did not augment dexamethasone’s effect on hGRα transrepression activity (Fig. 3.4A).  48  When TNF-α was tested from 0-10 ng/ml, the result suggested that at least 5 ng/ml of TNF-α treatment was required to show  statistically significant effect on hGRα transrepression mediated by dexamethasone (56.2  8.9% decrease in luciferase activity). Increasing TNF-α treatment to 10 ng/ml did not further affect transrepression (52.7  9.3% decrease; Fig 3.4B).  Lastly, different amounts of pCMV6-XL5-hGR vector (50-200 ng) were transfected in cultured LS180 cells. Each of transfection groups decreased the luciferase activity by approximately 55% (Fig. 3.4C). Based on these results, 500 nM of dexamethasone, 5 ng/ml TNF-α, and 100 ng of pCMV6-XL5-hGR transfection were used in the subsequent hGRα transrepression assay. Using the established parameters, multi-glycosylated PPD-type ginsenosides Rb1, Rb2, Rc, and Rd were tested at 60 μM, and monoglycosylated PPD-type ginsenosides CK and Rh2 were tested at 30 μM. Compared to DMSO, glycosylated PPD-type ginsenosides did not affect hGRα transrepression activity, whereas dexamethasone decreased the luciferase activity by 54.3  2.2%. All the other controls gave the expected results (Fig. 3.4D).  3.5 Ginsenoside Rh2 in hGRα-mediated Transactivation and Transrepression Increased by Dexamethasone Based on previous findings, ginsenosides were unable to act as agonists to activate hGRα for neither its transactivation nor transrepression activity. Nevertheless, TR-FRET competitive ligand-binding assay had shown their ability to bind to the hGRα-LBD. Thus, an experiment was performed to determine whether a ginsenoside can act as an antagonist to attenuate hGRα transactivation or transrepression activity increased by dexamethasone. Ginsenoside Rh2 was selected due to its strong binding profile to the receptor (Fig. 3.2H). 49  Concentration-response experiments were performed for dexamethasone. For hGRα transactivation, dexamethasone concentrations from 0.001-0.03 nM did not activate hGRα and had no effect on luciferase activity. At 0.06 nM, dexamethasone increased luciferase activity by 8.8  0.3-fold and reached maximal response (18.7  4.3-fold) at 1 nM for this assay (Fig. 3.5A, upper panel). For hGRα transrepression, dexamethasone concentrations from 0.01-0.3 nM had no effect on luciferase activity. The luciferase activity was decreased by 37.8  8.6% and 44.8  2.3% at 0.6 nM and 1 nM, respectively. Transrepression response peaked at 10 nM with a 55.5  2.1% decrease in luciferase activity (Fig. 3.5B, upper panel). For the subsequent assays, dexamethasone concentrations of 0.06 nM and 1 nM were selected for transactivation and transrepression assay, respectively.  Rh2 (30 μM) alone did not increase luciferase activity in the hGRα transactivation assay, whereas dexamethasone at 0.06 nM increased the luciferase activity by 7.9  0.3-fold over the DMSO-treated control group. Furthermore, Rh2 (30 μM) did not affect the increase in luciferase activity by dexamethasone. In contrast, RU486 attenuated the increase in luciferase activity by dexamethasone (Fig 3.5A, lower panel).  For hGRα transrepression, dexamethasone at 1 nM decreased the luciferase activity by 53.3  3.5%. Rh2 at 30 μM did not attenuate the effect of dexamethasone on hGRα transrepression activity. Again, RU486 was able antagonize and reverse dexamethasone’s effect on hGRα-mediated transrepression (Fig 3.5B, lower panel) . 3.6 Ginsenoside Rh2 in hGRα Target Gene Expression hGRα target genes were studied in HepG2 cells, a human hepatoma cell line which expresses endogenous hGRα. Two target genes, human tyrosine aminotransferase (hTAT) and 50  human corticosteroid binding globulin (hCBG), were selected for investigation. hTAT encodes for an enzyme involved in tyrosine metabolism and is transcriptionally activated by hGRα. Ginsenoside Rh2 at 30 μM did not increase hTAT mRNA expression. In contrast, dexamethasone at 10 nM increased hTAT mRNA expression by 2.5  0.3-fold over the vehicle-treated control group. Rh2 did not attenuate the effect of dexamethasone on hTAT mRNA expression. As expected, RU486 decreased hTAT mRNA expression to a level comparable to that in the vehicle-treated control group (Fig. 3.6A).  hCBG encodes for a glucocorticoid transport protein and is a target gene of hGRα-mediated transrepression. Twenty-four h treatment of Rh2 at 30 μM did not decrease hCBG mRNA expression in HepG2 cells. Compound A (CpdA; 1 μM), a naturally-derived compound that was reported to modulate hGRα transrepression activity, decreased hCBG mRNA expression by 36.4  3.6%. Addition of ginsenoside Rh2 did not attenuate the mRNA downregulation caused by CpdA. So far, there has not been an antagonist reported for hGRα-mediated transrepression of hCBG. Therefore, testosterone was also tested to determine whether it can antagonize hGRα. At 10 μM, testosterone alone did not affect hCBG mRNA expression. Adding testosterone to CpdA treatment also did not attenuate transrepression by CpdA. Therefore, both ginsenoside Rh2 and testosterone were not antagonists of hGRα-mediated transrepression.   51   Figure 3.1. Determination of non-cytotoxic concentrations of ginsenosides in cultured LS180 cells. LS180 cells were treated with a ginsenoside at 10 μM (A), 30μM (B), or 60 μM (C), along with control treatments including culture medium (vehicle control for dextran and Triton X-100), DMSO (0.1% v/v; vehicle control for ginsenosides), dextran (1% w/v; negative control), or Triton X-100 (0.1% v/v; positive control) for 24 h before LDH detection. Data are presented as mean ± SEM for three independent experiments performed in triplicate. For statistical analysis, MediumDMSO Rb1Rb2 RcRdCKRh2sPPD Re RfRg1Rh1sPPTDextranTriton X-100% LDH Release02080100120* ***60 MMediumDMSO Rb1Rb2 RcRdCKRh2sPPD Re RfRg1Rh1sPPTDextranTriton X-10002080100120*30 MMediumDMSO Rb1Rb2 RcRdCKRh2sPPD Re RfRg1Rh1sPPTDextranTriton X-100% LDH Release02080100120*10 MBCA52  one-way ANOVA was performed followed by Student-Newman-Keuls post-hoc test; *, significantly different from the vehicle control group; p˂ 0.05.  53   54  Figure 3.2. In vitro binding of ginsenosides to hGRα ligand-binding domain. DMSO (0.1% v/v; vehicle control), a ginsenoside (100 μM), SR12813 (5 μM ; negative control), and dexamethasone (Dex) (5 μM ; hGRα agonist) were incubated with glutathione-S-transferase (GST)-conjugated hGRα-LBD (0.5 nM) labeled with terbium anti-GST antibody (2 nM) in the presence of FluromoneTM GS1 (5 nM; hGRα ligand), and analyzed using a TR-FRET based method (A). Concentration-response curves were characterized for dexamethasone (0.001 nM-5 μM), multiply-glycosylated PPD-type ginsenosides (Rb1, Rb2, Rc, and Rd; 0.1-500 μM), and monoglycosylated PPD-type ginsenosides (CK and Rh2; 0.1-100 μM) using the same TR-FRET based analysis (B-H). Data are presented as mean ± SEM for three independent experiments performed in triplicate. For statistical analysis, one-way ANOVA was performed followed by Student-Newman-Keuls post-hoc test; *, significantly different from the vehicle control group; p˂ 0.05 (A).            55   Figure 3.3. Ginsenosides had no effect on hGRα-mediated transactivation. LS180 cells were transiently transfected for 24 h with either hGRα-expressing (pCMV6-XL5-hGR) or corresponding empty vector (pCMV6-XL5), along with GRE-containing firefly luciferase reporter vector (pGRE-luc) and Renilla luciferase internal control vector (pGL4.74[hRluc/TK]). Transfected cells were subsequently treated with DMSO (0.1% v/v; vehicle control), a ginsenoside (30 μM), SR12813 (10 μM; negative control), RU486 (100 nM; hGRα antagonist), dexamethasone (100 nM; hGRα agonist), or combination treatment of dexamethasone (100 nM) DMSO Rb1Rb2 RcRdSR12318RU486DexDex+RU486Normalized Luciferase Activity0. pCMV6-XL5-hGR *60 MDMSO Rb1Rb2 RcRdCKRh2sPPD Re RfRg1Rh1sPTTSR12813RU486DexDex+RU486Normalized Luciferase Activity0. pCMV6-XL5-hGR *30 MAB56  and RU486 (100 nM) for 24 h before cell lysis for luminescence detection (A). Multi-glycosylated PPD-type ginsenosides were further examined at higher concentration (60 μM) using the same method (B). Data are expressed as normalized ratio of firefly luciferase and Renilla luciferase luminescence signal, and shown as mean ± SEM for three independent experiments performed in triplicate. For statistical analysis, two-way ANOVA was performed followed by Student-Newman-Keuls post-hoc test; *, significantly different from the vehicle control group; p˂ 0.05.               57   Figure 3.4. Ginsenosides had no effect on hGRα-mediated transrepression. To establish proper assay condition, LS180 cells were transiently transfected with either pCMV6-XL5-hGR or pCMV6-XL5, along with NF-κB RE-containing firefly luciferase reporter vector (NFkB-luc) and pGL4.74[hRluc/TK] for 24 h. Then the cells were treated with TNF-α (5ng/ml; NF-κB activator) and either DMSO (0.1% v/v; vehicle control) or dexamethasone (500 nM or 1000nM; hGRα agonist) for 5 h before luminescence detection (A). TNF-α concentration-response (0-10 ng/ml) and pCMV6-XL5-hGR dose-response (50-200 ng) relations were also assayed with DMSO Dex 500 nM Dex 1000 nMNormalized Luciferase Activity0. pCMV6-XL5 pCMV6-XL5-hGR * *ATNF- Concentration0 ng/ml 1 ng/ml 5 ng/ml 10 ng/ml012345DMSO Dex **BCpCMV6-XL5-hGR Transfected50 ng 100 ng 200 ngNormalized Luciferase Activity0. Dex DDMSO Rb1Rb2 RcRdCKRh2RU486DexDex+RU4860. pCMV6-XL5-hGR *58  similar approach (B and C). Using the established condition, the effect of ginsenosides Rb1, Rb2, Rc, Rd (60 μM), and CK and Rh2 (30 μM) on hGR-mediated transrepression was tested with DMSO (0.1% v/v; vehicle control), RU486 (10 μM; hGRα antagonist), dexamethasone (500 nM; hGRα agonist), or both dexamethasone (500 nM) and RU486 (10 μM) as controls. Cells were co-treated with TNF-α (5ng/ml) for 5 h before luminescence detection (D). Data are shown as mean ± SEM for three independent experiments performed in triplicate. For statistical analysis, two-way ANOVA was performed followed by Student-Newman-Keuls post-hoc test; *, significantly different from the vehicle control group; p˂ 0.05                59   Figure 3.5. Ginsenoside Rh2 could not attenuate hGRα-mediated transactivation and transrepression increased by dexamethasone. To establish dexamethasone concentration-response relation in hGRα transactivation and transrepression luciferase reporter gene assay, LS180 cells were transiently transfected for 24 h with pCMV6-XL5-hGR, pGL4.74[hRluc/TK], and either pGRE-luc or NFkB-luc. The cells were then treated with DMSO (0.1% v/v; vehicle for dexamethasone, labeled as 0 nM), and indicated concentrations of dexamethasone (hGRα Dexamethasone Concentration0 nM0.001 nM0.01 nM0.03 nM0.06 nM0.1 nM1 nM10 nM100 nMNormalized Luciferase Activity0.*****Dexamethasone Concentration0 nM0.01 nM0.1 nM0.3 nM0.6 nM1 nM10 nM100 nM500 nM0.*** * *DMSO Rh2RU486DexDex+Rh2Dex+RU486Normalized Luciferase Activity0. pCMV6-XL5-hGR * *DMSO Rh2RU486DexDex+Rh2Dex+RU4860. pCMV6-XL5-hGR * *A BTransactivation Transrepression60  agonist) for 24 h before luminescence detection (A and B, upper panel). Effect of ginsenoside Rh2 on dexamethasone-induced hGRα transactivation (A, lower panel) and transrepression (B, lower panel) was assayed. Transiently transfected cells were treated with DMSO (0.1% v/v; vehicle control), ginsenoside Rh2 (30 μM), dexamethasone, (0.06 nM for transactivation assay and 1 nM for transrepression assay; concentrations selected based on the concentration-response relation), RU486 (100 nM or 10 μM for transactivation or transrepression assay, respectively; hGRα antagonist), or combination treatments of dexamethasone and ginsenoside Rh2 or dexamethasone and RU486 at aforementioned concentrations. Data are presented as mean ± SEM for three independent experiments performed in triplicate. For statistical analysis, either One-way ANOVA (A and B, upper panel) or two-way ANOVA (A and B, lower panel) was performed followed by Student-Newman-Keuls post-hoc test; *, significantly different from the vehicle control group; p˂ 0.05.            61    Figure 3.6. Ginsenoside Rh2 did not influence hGRα target gene expression. HepG2 cells were treated for 24 h with DMSO (0.1% v/v; vehicle control), ginsenoside Rh2 (30 μM), RU486 (100 nM; hGRα antagonist), dexamethasone (10 nM; hGRα agonist), or combination treatment of either dexamethasone (10 nM) and ginsenoside Rh2 (30 μM) or dexamethasone (10 nM) and RU486 (100 nM) before RNA isolation and reverse transcription (A). HepG2 cells were treated with DMSO (0.1% v/v; vehicle control), ginsenoside Rh2 (30 μM), testosterone (TT) (10 μM; hGRα antagonist), CpdA (1 μM; hGRα agonist), or combination treatment of either CpdA (1 μM) and ginsenoside Rh2 (30 μM) or CpdA (1 μM) and testosterone (10 μM) for 24 h (B). Relative expression level of hTAT mRNA, a target gene of hGRα-mediated transactivation, and hCBG mRNA, a target gene of hGRα-mediated transrepression, were quantified using real-time PCR using ΔΔCt method after normalizing to a housekeeping gene, human hypoxanthine-guanine phosphoribosyltransferase (hHPRT). Data are expressed as mean ± SEM for three independent DMSO Rh2RU486DexDex+Rh2Dex+RU486Relative mRNA Expression01234**hTAT mRNAADMSO Rh2 TTCpdACpdA+Rh2CpdA+TT0.*hCBG mRNAB62  experiments performed in triplicate. For statistical analysis, one-way ANOVA was performed followed by Student-Newman-Keuls post-hoc test; *, significantly different from the vehicle control group; p˂ 0.05. 63  4. Discussion  4.1 Ginsenosides Bind Weakly to hGRα-LBD From the TR-FRET competitive ligand-binding assay, the data showed that both PPD- and PPT-type ginsenosides bind weakly to the LBD of hGRα at a high concentration (100 μM). Comparing these two types of ginsenosides, PPD-type ginsenoside were able to decrease the net TR-FRET ratio, a marker of competitive binding, by 35-66%, and exhibited stronger binding than PPT-type ginsenosides, which were able to decrease the net TR-FRET ration by only less than 30%. PPD- and PPT-type ginsenosides differ by the substituent at C-6 position (Fig. 4.1C and D). While the aglycone PPD and PPT, in which sugar substituents are absent, displayed similar binding capacity, all glycosylated PPD-type ginsenosides showed stronger binding than glycosylated PPT-type ginsenosides, suggesting that the sugar group at C-6 position may hinder binding to hGRα. Bledsoe et al. (2002) had resolved the LBD crystal structure in the presence of dexamethasone. Analysis of the crystal structure showed that C-6 of dexamethasone, which contains no substituents, may interact with A605 of the LBD through hydrophobic interaction (Bledsoe et al., 2002). In general, all natural-occurring corticosteroids such as cortisol, corticosterone, aldosterone, as well as synthetic GC, such as prednisone, dexamethasone, lack a substituent at C-6 with the exception of fluticasone and flunisolide, which have a single fluorine atom attached, and methylprednisolone, with a small methyl group attached. Therefore, sugar substituents at C-6 may lead to increased steric hindrance which weakens the receptor binding of PPT-type ginsenosides whereas PPD-type ginsenosides lack the sugar moiety at C-6 display stronger receptor binding. 64   Among PPD-type ginsenosides, multi-glycosylated ginsenosides exhibited weaker binding when compared to monoglycosylated ginsenosides, as assessed by IC50 values generated from concentration-response curves. Table 3.1 shows that as the number of sugar substituents decreases, the IC50 also decreases, suggesting that sugar moieties may hinder the compound from receptor binding. Furthermore, between the two monoglycosylated ginsenosides, CK displayed weaker binding compared to Rh2. Both CK and Rh2 possess a single glucosyl group, but differ in the position of attachment. The glucose on CK is attached to C-20 of the D ring, while on Rh2, it is linked to C-3 of the A ring (Fig. 1.4; Fig. 4.1C). In general, GC is bound to the GR-LBD with an orientation that its D-ring is positioned toward AF-2 helix of the LBD (Bledsoe et al., 2002). Upon ligand binding, the AF-2 helix serves to stabilize the ligand-bound LBD structure. Thus, the bulky sugar substituent at C-20 in CK may reduce the stability of the LBD structure when compared to the glucose-free Rh2. When compared to Rh2, the aglycone PPD lacked the bulky sugar moiety at C-3, yet did not exhibit stronger binding. Conventionally, GR, MR, PR, and AR are classified into a subfamily called 3-ketosteroid receptor and differentiate themselves from ER because their ligands generally contain a ketone functional group at C-3 instead of a phenolic hydroxyl group like ER ligands (Ekena et al., 1998). This is due to the difference in contact points in the binding pockets between the 3-ketosteroid receptors and ER. It is probable that the hydroxyl group on C-3 of PPD may not interact properly with the key residues in the GR ligand binding pocket. Nevertheless, it is intriguing that Rh2 with a bulky glucosyl substituent at C-3 can bind more effectively than the aglycone. One explanation may be that the functional groups on the glucose may serve to stabilize the ligand-receptor complex through interactions with other residues in the LBD domain, resulting in stronger binding compared to PPD. 65  Four studies had conducted FP-based competitive ligand-binding assays using full-length recombinant hGRα and a fluorescein-labeled GR ligand (Leung et al., 2006a; Leung et al., 2007; Wu et al., 2012; Yang et al., 2008). In agreement with this study, Yang et al. (2008) reported that CK was able to compete with the fluorescein-labeled ligand and bind to the full-length receptor. Unfortunately, the IC50 value of CK for this assay was not determined. Two other studies examined dexamethasone and Rg1. Based on the concentration-response curve, IC50 values for both compounds were approximated (Leung et al., 2006a; Wu et al., 2012). Intriguingly, while both Leung et al. (2006a) and Wu et al. (2012) reported very similar IC50 values for dexamethasone (10.6 nM and 9.0 nM, respectively), the IC50 values generated for Rg1 differed by two orders of magnitude (128 nM and 12.8 μM, respectively). Lastly, Re was reported to compete with a GR fluorescent ligand with an IC50 value at nanomolar range (156.6 nM) (Leung et al., 2007). In this study, although Re and Rg1 could bind to hGRα-LBD, the binding was very weak even when these compounds were tested at high concentrations. At 100 μM, Re and Rg1 could only decrease the net TR-FRET ratio by less than 15% (Fig. 3.2A). The discrepancy of results may be due to differences in the experimental set up. In particular, the amount of fluorescein-labeled ligand used can greatly influence the concentration of the ginsenosides required to compete for binding, and this may be reflected by the differences in IC50 values of dexamethasone from their studies and this study (24.1 nM).  4.2 Ginsenosides Were Unable to Modulate hGRα-mediated Transactivation To assess whether ginsenosides can affect hGRα-mediated transactivation, a dual-luciferase reporter gene assay was employed. To detect ligand activation that is specific to hGRα, 66  human colon adenocarcinoma LS180 cells, a cell line which does not express detectable endogenous GR, was employed (Maier et al., 2007). By transfecting hGRα-expression vector or corresponding empty vector to into LS180 cells, hGRα-specific activity can be discriminated from nonspecific activity. In this study, all ginsenosides (Rb1, Rb2, Rc, Rd, CK, Rh2, PPD; Re, Rf, Rg1, Rh1, and PPT) at 30 μM were unable to increase the luciferase activity. On the other hand, dexamethasone at 100 nM increased the luciferase activity by 7.4-fold when compared to vehicle control. Similar results were obtained even when cells were treated with 60 μM of multi-glycosylated PPD-type ginsenosides. This suggests that dexamethasone but not ginsenosides was able to activate hGRα and modulate its transactivation activity.  These data do not seem to be in agreement with the current findings. Leung et al. (2007) employed a secreted embryonic alkaline phosphatase (SEAP) reporter gene assay system to test Re’s ability in hGRα modulation. Using HUVEC, which expresses endogenous hGRα, the authors showed that at nanomolar concentration (240 nM), Re was able to cause a small increase (2.4-fold) in SEAP activity. Dexamethasone was also tested at 50 nM and showed a similar fold increase (2.5-fold). In another study, Yang et al. (2008) tested CK at various concentrations (0.1-16 μg/ml; ~0.16-25.7 μM) in HEK293T cells transfected with a hGRα expression vector and a GRE-containing firefly luciferase reporter gene vector. Starting from 2 μg/ml (~3.2 μM), a small increase in luciferase activity (~6 fold) can be observed. At 16 μg/ml (~25.7 μM), CK caused roughly a 20-fold increase in luciferase activity. It is worthy to note that the cells were pre-treated with LPS (1 μg/ml) for 4 h to induce a state of inflammation.  In contrast, three studies from Lee et al. (2003a, 2003b, 2003c), in agreement with this study, reported that Rb1 (50 μM), Rh1 (50 μM), Rc (45 μg/ml; ~41.7 μM), and Re (45 μg/ml; 67  ~47.5 μM) were all incapable of modulating hGRα-mediated transactivation of luciferase reporter gene in CV-1 cells. The discrepancy of the results from this study, Lee et al. (2003a, 2003b, 2003c), and the previous published works may be due to differences in cell type, which will be discussed further in section 4.6.  4.3 Ginsenosides Were Unable to Modulate hGRα-mediated Transrepression GR plays a very important role in transcriptional regulation in that it is capable of mediating both transactivation and transrepression of numerous target genes. GR’s function in gene transrepression is essential to the receptors’ anti-inflammatory effect. By inhibiting TF such as NF-κB, GR represses the expression of various proinflammatory genes (McKay et al., 1999). Compounds like AL-438 and CpdA are SEGRA that can modulate hGRα for transrepression but not transactivation. Because ginsenosides were unable to stimulate hGRα-mediated transactivation, the next question asked was whether these compounds act as SEGRA and modulate hGRα-mediated transrepression. For this, a dual-luciferase reporter gene assay was employed, and glycosylated PPD-type ginsenosides (Rb1, Rb2, Rc, Rd, CK, and Rh2) were tested in LS180 cells transfected with hGRα expression vector and NF-κB RE-containing luciferase reporter vector. Dexamethasone at 500 nM was able to decrease luciferase activity by 54.3% (Fig. 3.4D). This indicates that dexamethasone was able to promote the tethering of hGRα to NF-κB, which leads to inhibition of NF-κB-mediated transcription. However, the ginsenosides were unable to suppress luciferase activity, suggesting that these compounds were unable to activate hGRα and therefore failed to transrepress NF-κB-mediated transcriptional regulation. 68  Even though in this study, glycosylated PPD-type ginsenosides had no effect on hGRα’s inhibition of NF-κB, a previous study showed that Rg1 could modulate hGRα-mediated transrepression. In Du et al., 2011, A549 cells, which expresses endogenous hGRα, were transfected with luciferase reporter vector carrying NF-κB RE. Dexamethasone (1 μM) or Rg1 (10 μM) was given as a pretreatment for 2 h, followed by a co-treatment of LPS to induce inflammatory state for another 6 h. Both dexamethasone and Rg1 were able to decrease the luciferase activity by approximately 45%.  CBP had been implicated in enhancing the tethering between hGRα and p65 subunit of NF-κB (McKay et al., 2000). Discrepancy between the two results may be explained by differences in CBP level in different cell types. Alternatively, differences in drug efflux may also be involved. These will be elaborated in section 4.6.  4.4 Ginsenoside Rh2 Was Unable to Attenuate hGRα-mediated Transactivation and Transrepression Albeit the competitive ligand-binding assay indicated that ginsenosides were capable of binding hGRα in vitro, these chemicals were unable to activate hGRα for transactivation or transrepression in cell-based assays. Therefore, the next experiment was to determine whether ginsenosides act as a GR antagonist which abolishes the receptor’s transcriptional regulation activities when bound to hGRα. For this, Rh2, which had exhibited strongest receptor binding among all tested ginsenosides, was selected and assayed for its ability to attenuate dexamethasone’s activation of hGRα. Nevertheless, Rh2 treatment (30 μM) did not alter the hGRα transactivation or transrepression activity caused by dexamethasone (Fig. 3.4A and B). In contrast, RU486, a known hGRα antagonist, was able to annul the dexamethasone-increased 69  transactivation and transrepression activity by hGRα. The lack of attenuation suggests inability of Rh2 to antagonize the receptor. Combined with the accumulated data thus far, ginsenosides act as neither an agonist nor an antagonist to hGRα in LS180 cells . 4.5 Ginsenoside Rh2 Was Unable to Influence hGRα-mediated Transactivation and Transrepression of Target Genes To study hGRα target gene expression, human hepatocellular carcinoma cells (HepG2) were used as experimental model. HepG2 is reported to express endogenous hGRα and GR target genes hTAT and hCBG (Novotna et al., 2014; Verhoog et al., 2014). hTAT is a target gene for hGRα-mediated transactivation. Two GRE can be found in the hTAT promoter, which are recognized and bound by liganded GR. Subsequently, coactivators and components of BTM are recruited to activate transcription of the gene (Jantzen et al., 1987). hCBG is a target gene for hGRα-mediated transrepression. Transcription of hCBG is activated by CCAAT/enhancer binding protein beta (C/EBPβ). GR inhibits C/EBPβ via protein-protein interaction, and transrepresses the expression of hCBG (Verhoog et al., 2014). While 10 nM of dexamethasone caused a 2.5-fold increase in hTAT transcript, 30 μM of Rh2 did not elevate nor attenuate the hTAT mRNA expression increased by dexamethasone (Fig. 3.6A). Likewise, 1 μM of CpdA lowered the hCBG transcript by 36.4%, yet 30 μM of Rh2 was unable to decrease the hCBG mRNA expression by itself, or reverse the downregulation caused by CpdA (Fig. 3.6B). These results implicated that Rh2 is not a modulator of the endogenous hGRα in HepG2, and therefore it was unable to activate the receptor to regulate the transcription of target genes. 70  Two studies had tested ginsenosides’ effect on GR target gene expression in rodent cell lines. Rg1 was reported to increase the mRNA expression of SGK and DUSP1 in primary rat cortical neurons to a level that was comparable to dexamethasone, and the effect was antagonized by RU486 (Wu et al., 2012). In another report, Rh1 alone had no effect on DUSP1 transcript expression but potentiated dexamethasone’s upregulation of DUSP1 by restoring the level of GR in RAW264.7 (Li et al., 2014). Interestingly, in the same study, Rh1 did not significantly potentiate dexamethasone-increased DUSP1 mRNA expression in primary mouse hepatocytes. These conflicting results within the same study further suggest that the effect of ginsenosides may be cell type-specific. Furthermore, the expression of two other GR target genes, PEPCK and glucose-6-phosphatase (G6P), were also studied. Similar to DUSP1, both genes are regulated by GR-mediated transactivation. Whereas Rh1 was able to potentiate DUSP1 expression by dexamethasone, the compound attenuated PEPCK and G6P upregulation by dexamethasone. These data implicated the ginsenosides’ effect may be promoter-specific, and complex mechanisms of transcriptional regulation through different coregulators are involved.  4.6 Potential Explanations for Discrepancy among Different Studies   4.6.1 Cell Type Differences in Endogenous Coregulator Level One explanation for the discrepancy between the conflicting results concerning whether ginsenosides are able to modulate hGRα activities may be that different cell lines were used as experimental model. A few studies have suggested the effects of GC via GR are cell type-specific. Cellular context, such as endogenous level of GR, coregulators, and even the ratio of 71  coactivators and corepressors influence the cell’s sensitivity to GC, and how GC would modulate GR activity (De Bosscher, 2010; Dezitter et al., 2014; Wang et al., 2004). SRC-1 Level and hGRα-mediated Transactivation Dezitter et al. (2014) showed that cellular level of SRC-1 played an important role in determining the ability of RU782 and RU858 to transactivate hGRα, while dexamethasone’s ability to modulate hGRα-mediated transcription remained little affected. RU782 and RU858 are known as dissociated GC, and capable of modulating hGRα for its transrepression of target gene, but not transactivation. To elucidate the mechanism for these compounds’ ability in dissociating the transactivation and transrepression of GR, molecular docking analysis, and luciferase reporter gene assays using wild type GR and GR containing point mutations in different cell types were performed. Comparing the structures between dexamethasone and dissociated GC, one of the main distinctions lies on the α-hydroxyl group at C-17 (17α-OH), which is present in GC, but missing in dissociated GC (Fig. 4.1A and B). The 17α-OH group is crucial in making contact with a glutamine residue (Q642) in the hGRα ligand-binding pocket, which helps in SRC-1 recruitment (Dezitter et al., 2014). Lack of 17α-OH group in dissociated GC impaired their ability to alter GR-LBD into a proper conformation that could efficiently recruit SRC-1, thereby resulting in very little GR transactivation activity. This phenomenon, however, can be compensated when cellular level of SRC-1 is high. Three different cell lines, HeLa, COS-1, and HEK293T were tested in the study and it was found that while RU782 and RU858 induced little GR transactivation activity in HeLa, their ability to increase GR transactivation activity was greatly enhanced in HEK293T cells. Intriguingly, levels of transactivation induced by RU782 and RU858 were similar between the two compounds in 72  HeLa and HEK293T, yet there was a marked difference between the two compounds in COS-1 cells, with RU858 having moderate transactivation activity and RU782 having little transactivation activity via hGRα. This further confirmed that the effect of GR ligands on the receptor is highly dependent on cellular context. Moreover, when SRC-1 was over-expressed in HeLa cells, transactivation activity mediated by RU782 and RU858 increased. Similar to the structure of RU782 and RU858, ginsenosides also lacked the 17α-OH group, which suggests that these chemicals may have lower efficiency in SRC-1 recruitment and high cellular level of SRC-1 are required for these compounds to modulate hGRα transactivation. In this study, LS180 and HepG2 cells were used, and both cell lines have been reported to express low levels of SRC-1 (Martínez-Jiménez et al., 2006; Zhou et al., 2004). Therefore, SRC-1 level may be an explanation for conflicting data from this study and other published studies. CBP Level and hGRα-mediated Transrepression  While the dissociated GC are not able to modulate transactivation activity of hGRα, they are able to effect transrepression by hGRα. However, in this study, ginsenosides had no effect on hGRα-mediated transrepression. This may also be explained by the differences in level of coregulators. Molecular docking analysis in a previous study had shown that RU782 interacted with the I747 residue in GR ligand-binding pocket through its methyl sulfide group at C-21 (Fig. 4.1B) (Dezitter et al., 2014). Compared to dexamethasone (Fig. 4.1A), the contact distance between I747 and methyl sulfide group of RU782 was very short, and led to crowding of the residue. Because I747 is located on the loop connecting helices 11 and 12, the short contact distance may result in displacement of AF-2 domain of hGRα-LBD. This displacement may further compromise RU782’s ability to modulate GR’s recruitment of various coregulators. 73  CBP was shown to be involved in mediating GR’s tethering of NF-κB to transrepress NF-κB regulated transcription (McKay et al., 2000). CBP serves as an adaptor protein which strengthens and stabilizes the interaction between the GR and NF-κB. It had been shown that GR-LBD is required for the recruitment of CBP to GR (Kamei et al., 1996). Having AF-2 displaced may result in an ineffective recruitment of CBP, therefore impaired transrepression. Ginsenosides contain a long aliphatic chain on D ring at C-22 (Fig. 4.1C and D). This may also result in similar crowding near the loop region and displacement of AF-2, and  consequently impairing the recruitment of CBP and compromising ginsenosides’ ability in modulating hGRα-mediated transrepression when CBP level is low. One study had examined the ability of ginsenoside Rg1 to modulate hGRα for transrepression in A549 (Du et al., 2011). There may be differences in CBP level in A549, and the cell lines used in this study leading to the difference in the observed effects of ginsenosides on hGRα. Unfortunately, little is known concerning the levels of CBP in HepG2 and LS180; Western blotting can be performed to examine the expression level of CBP in these cell lines.  4.6.2 Cell Type Differences in P-glycoprotein Expression Another cell type-specific factor may be the difference in efflux rate of the compounds. Ginsenosides Rh2 and Rg1 have been reported to be substrates of P-glycoprotein (P-gp) (Meng et al., 2007; Xie et al., 2005). Xie et al. (2005) investigated the uptake of Rh2 and PPD in Caco-2 cells, and found that both compounds are substrates of P-gp and are actively exported out of Caco-2 cells. Using P-gp inhibitors, verapamil and cyclosporine, the Rh2 uptake rate was increased by 1.3-fold and 2.6-fold, respectively. The efflux by P-gp is an energy-dependent process, and it has been reported metabolic inhibitors, such as 2,4-dinitrophenol and sodium 74  azide, disrupt ATP production and effectively inhibit the extrusion of drugs by P-gp. In accordance to the data from P-gp inhibitors, treating the cells with 2,4-dinitrophenol and sodium azide significantly increase the rate of Rh2 uptake by 2.9-fold and 2.3-fold, respectively.  In another study, Rg1 was also found to be extruded by P-gp. Meng et al. (2007) treated primary rat pulmonary epithelial cells with high concentration of Rg1 (100 μg/ml; ~125 μM) and reported that Rg1 had a low penetration ratio, i.e. the intracellular concentration of Rg1 remained low in comparison to Rg1 concentration in the treatment medium. When the cells were treated with a P-gp inhibitor and a metabolic inhibitor, verapamil and sodium cyanide, the intracellular concentration of Rg1 increased. Like ginsenosides Rh2 and Rg1, dexamethasone was also reported to be a P-gp substrate. However, because of its high affinity to hGRα and high potency in comparison to the ginsenosides, the active extrusion of P-gp may have a greater impact on compromising ginsenosides’ activation of hGRα than on dexamethasone. Previous reports that demonstrated modulation of hGRα by ginsenosides used HEK293T, A549 and HUVEC as experimental models. It had been shown in different studies that these cell lines had either no endogenous expression (A549 and HUVEC) or low expression of P-gp (HEK293T) (Dauchy et al., 2009; Dessilly et al., 2014; Kavallaris et al., 1997). Nevertheless, in this study, where ginsenosides were found incapable of activating hGRα, P-gp is demonstrated to be constitutively expressed in the cell lines used (Chan et al., 2000; Maier et al., 2007). Therefore, cell type-specific difference in P-gp expression may be an alternative explanation for the discrepancy between the findings in this study and the previously published works.  75  4.7 Limitations of the Study  There are several limitations in this study. The first two are related to the LDH cytotoxicity assay. First, this study employed LDH cytotoxicity detection assay, which is an end-point assay that mainly detects dead cells with disrupted membrane. It is unknown whether ginsenosides at tested concentrations could cause stress to the cells, and if so, whether such stress would influence the functionality of cells. However, the expression of Renilla luciferase in the reporter gene assays can be an indirect indicator of the functionality of cellular machinery, showing that the cells are healthy enough to execute the cellular processes of transcription and translation at tested concentrations. Second, the ginsenosides were tested up to a concentration of 60 μM in the LDH cytotoxicity detection assay. However, in the competitive ligand-binding assay, ginsenosides Rb1, Rb2, Rc, and Rd showed IC50 values ranging from 60-120 μM. Because of the high IC50 values, these ginsenosides should be tested at concentrations higher than 60 μM for cytotoxicity. If high concentrations of these ginsenosides remain non-toxic, then the cell-based assays should be repeated at those concentrations. In addition, the cell lines utilized in this study, although originated from different tissues, are similar in that both exhibit low levels of key coregulators, such as SRC-1, and both constitutively express P-gp. Because the effects of GR ligands are highly cell type-specific, and are very much affected by levels of coregulators and drug efflux, therefore a larger variety of cell lines should be employed. Alternatively, methods such as introduction of transiently expressed coregulators, and inhibition of P-gp can be considered.  76  4.8 Future Experiments  4.8.1 Assessment of Ginsenosides Rb1, Rb2, Rc, and Rd at High Concentrations Because ginsenosides Rb1, Rb2, Rc, and Rd showed very weak binding with IC50 values much greater than the previously tested concentrations (30 μM and 60 μM), these compounds should be re-assayed at concentrations higher than their IC50 values. To this end, an additional LDH cytotoxicity detection assay should first be performed to establish the highest non-toxic concentration of each compound, and the ginsenosides should be tested again at those concentrations using luciferase reporter gene assays.  4.8.2 Determination of SRC-1’s Role in Ginsenosides’ Ability to Modulate hGRα-mediated Transactivation In Dezitter et al. (2014), the dissociated GC were unable to modulate GR-mediated transactivation in HeLa cells because of the compounds’ inability to efficiently recruit SRC-1. When the level of SRC-1 was increased, the insensibility to dissociated GC was reversed in HeLa cells. To investigate whether ginsenosides’ lack of effect in modulating GR-mediated transactivation in LS180 and HepG2 is due to the low endogenous level of SRC-1, expression vectors carrying SRC-1 gene can be transiently expressed in these cells. Western blotting can be performed to confirm the increase in SRC-1 level, then luciferase reporter gene assays and TAT gene expression study should follow to re-evaluate ginsenosides’ effect on hGRα-mediated transactivation.  77  4.8.3 Determination of CBP’s Role in Ginsenosides’ Ability to Modulate hGRα-mediated Transrepression CBP was reported to be involved in inhibition of NF-κB by hGRα. It serves as a platform that enhances the inhibitory tethering of the two proteins by interacting with NF-κB’s p65 unit and LBD of GR (McKay et al., 2000). However, the CBP’s interaction with GR-LBD may be weakened when ligands such as ginsenosides are bound to the binding pocket, due to the bulky side chain attached to its D ring. Therefore, analogous to SRC-1 and its effect on dissociated GC in GR-mediated transactivation, expression of CBP in LS180 and HepG2 cells may be low and compromise the ability of ginsenosides’ modulation of hGRα-mediated transrepression. Again, to validate the role of CBP in lack of response from these cell lines, CBP levels can be elevated via transient expression. Then, luciferase reporter gene assays and hCBG gene expression study can be employed to re-assess the effect of ginsenosides.  4.8.4 Investigation of Effect of P-gp Drug Efflux Ginsenosides were reported as P-gp substrates, and inhibition of P-gp efflux via chemical inhibitors led to elevated uptake rate and higher intracellular concentration (Meng et al., 2007; Xie et al., 2005). To elucidate whether the lack of effect of ginsenosides is due to low intracellular concentration caused by drug extrusion, P-gp anti-sense RNA and chemical inhibitors can be employed to knock down P-gp expression or directly inhibit P-gp efflux. It was reported that HepG2 does not respond to P-gp chemical inhibitors such as verapamil and cyclosporine. Therefore, an anti-sense RNA approach should be employed when working with HepG2 (Chan et al., 2000).  After P-pg knock down or inhibition, luciferase assays and gene expression studies can be conducted to re-evaluate ginsenosides’ ability to modulate hGRα. 78  4.9 Summary and Conclusion  In this study, all ginsenosides tested (Rb1, Rb2, Rc, Rd, CK, Rh2, PPD; Re, Rf, Rg1, Rh1, PPT) were capable of weak binding to hGRα-LBD. Among these, monoglycosylated PPD-type ginsenosides (CK and Rh2) exhibited strongest binding to the receptor in an in-vitro TR-FRET competitive ligand-binding assay. Despite their ability to bind the receptor, selected ginsenosides were unable to increase or attenuate hGRα-mediated transactivation and transrepression, as assessed by luciferase reporter gene assays in LS180 cells. Furthermore, ginsenoside Rh2 had no effect on the expression level of hGRα target genes, hTAT and hCBG in HepG2 cells. Taken together, PPD- and PPT-type ginsenosides examined in this study are not functional hGRα ligands in LS180 and HepG2 cells. Whether this lack of effect is attributed to cell type-specific traits, such as low level of endogenous coregulators or high drug efflux rate, remains to be elucidated. Further experiments are required to evaluate whether ginsenosides are hGRα modulators in a cell type-specific manner.        79    Figure 4.1. Structures of GR ligands and ginsenosides with atom numbering.     A B C D Dexamethasone RU24782 PPD-type Ginsenosides PPT-type Ginsenosides 80  References Attele AS, Wu JA, & Yuan C-S (1999). Ginseng pharmacology: multiple constituents and multiple actions. Biochemical Pharmacology 58: 1685-1693.  Ayroldi E, & Riccardi C (2009). Glucocorticoid-induced leucine zipper (GILZ): a new important mediator of glucocorticoid action. The FASEB Journal 23: 3649-3658.  Bae E-A, Park S-Y, & Kim D-H (2000). Constitutive beta-gluccosidases hydrolyzing ginsenoside Rb1 and Rb2 from human intestinal bacteria. Biological & Pharmaceutical Bulletin 23: 1481-1485.  Baeg I-H, & So S-H (2013). The world ginseng market and the ginseng (Korea). Journal of Ginseng Research 37: 1-7.  Benishin CG (1992). Actions of ginsenoside Rb1 on choline uptake in central cholinergic nerve endings. Neurochemistry International 21: 1-5.  Bledsoe RK, Montana VG, Stanley TB, Delves CJ, Apolito CJ, McKee DD, et al. (2002). Crystal structure of the glucocorticoid receptor ligand binding domain reveals a novel mode of receptor dimerization and coactivator recognition. Cell 110: 93-105.  Campbell JE, Peckett AJ, D'souza AM, Hawke TJ, & Riddell MC (2011). Adipogenic and lipolytic effects of chronic glucocorticoid exposure. American Journal of Physiology - Cell Physiology 300: C198-C209.  Chan JY-W, Chu AC-Y, & Fung K-P (2000). Inhibition of P-glycoprotein expression and reversal of drug resistance of human hepatoma HepG2 cells by multidrug resistance gene (mdr1) antisense RNA. Life Sciences 67: 2117-2124.  Cheng C-C, Yang S-M, Huang C-Y, Chen J-C, Chang W-M, & Hsu S-L (2005). Molecular mechanisms of ginsenoside Rh2-mediated G1 growth arrest and apoptosis in human lung adenocarcinoma A549 cells. Cancer Chemother Pharmacol 55: 531-540.  Choo M-K, Sakurai H, Kim D-H, & Saiki I (2008). A ginseng saponin metabolite suppresses tumor necrosis factor-alpha-promoted metastasis by suppressing nuclear factor-kappaB signaling in murine colon cancer cells. Oncology reports 19: 595.  Christensen LP (2008). Ginsenosides: Chemistry, biosynthesis, analysis, and potential health effects. In Advances in Food and Nutrition Research. ed Steve L.T. Academic Press, pp 1-99.  Coutinho AE, & Chapman KE (2011). The anti-inflammatory and immunosuppressive effects of glucocorticoids, recent developments and mechanistic insights. Molecular and Cellular Endocrinology 335: 2-13.  81  Dauchy S, Miller F, Couraud P-O, Weaver RJ, Weksler B, Romero I-A, et al. (2009). Expression and transcriptional regulation of ABC transporters and cytochromes P450 in hCMEC/D3 human cerebral microvascular endothelial cells. Biochemical Pharmacology 77: 897-909.  De Bosscher K (2010). Selective glucocorticoid receptor modulators. The Journal of Steroid Biochemistry and Molecular Biology 120: 96-104.  de Kloet ER, Joels M, & Holsboer F (2005). Stress and the brain: From adaptation to disease. Nat Rev Neurosci 6: 463-475.  Dessilly G, Elens L, Panin N, Capron A, Decottignies A, Demoulin J-B, et al. (2014). ABCB1 1199G>A genetic polymorphism (Rs2229109) influences the intracellular accumulation of tacrolimus in HEK293 and K562 recombinant cell lines. PLoS ONE 9: e91555.  Dezitter X, Fagart J, Taront S, Fay M, Masselot B, Hétuin D, et al. (2014). A structural explanation of the effects of dissociated glucocorticoids on glucocorticoid receptor transactivation. Molecular Pharmacology 85: 226-236.  Drouin J, Sun YL, Chamberland M, Gauthier Y, De Léan A, Nemer M, et al. (1993). Novel glucocorticoid receptor complex with DNA element of the hormone-repressed POMC gene. The EMBO Journal 12: 145-156.  Du J, Cheng B, Zhu X, & Ling C (2011). Ginsenoside Rg1, a novel glucocorticoid receptor agonist of plant origin, maintains glucocorticoid efficacy with reduced side effects. The Journal of Immunology 187: 942-950.  Ekena K, Katzenellenbogen JA, & Katzenellenbogen BS (1998). Determinants of ligand specificity of estrogen receptor-α: estrogen versus androgen discrimination. Journal of Biological Chemistry 273: 693-699.  Hafezi-Moghadam A, Simoncini T, Yang Z, Limbourg FP, Plumier J-C, Rebsamen MC, et al. (2002). Acute cardiovascular protective effects of corticosteroids are mediated by non-transcriptional activation of endothelial nitric oxide synthase. Nat Med 8: 473-479.  Hagopian K, Ramsey JJ, & Weindruch R (2003). Caloric restriction increases gluconeogenic and transaminase enzyme activities in mouse liver. Experimental Gerontology 38: 267-278.  Hall RK, Sladek FM, & Granner DK (1995). The orphan receptors COUP-TF and HNF-4 serve as accessory factors required for induction of phosphoenolpyruvate carboxykinase gene transcription by glucocorticoids. Proceedings of the National Academy of Sciences of the United States of America 92: 412-416.  Han BH, Park MH, Han YN, Woo LK, Sankawa U, Yahara S, et al. (1982). Degradation of ginseng saponins under mild acidic conditions. Planta Med 44: 146-149.  82  Hasegawa H (2004). Proof of the mysterious efficacy of ginseng: Basic and clinical trials: Metabolic activation of ginsenoside: Deglycosylation by intestinal bacteria and esterification with fatty acid. Journal of Pharmacological Sciences 95: 153-157.  Hasegawa H, Sung J-H, Matsumiya S, & Uchiyama M (1996). Main ginseng saponin metabolites formed by intestinal bacteria. Planta Med 62: 453-457.  Heitzer M, Wolf I, Sanchez E, Witchel S, & DeFranco D (2007). Glucocorticoid receptor physiology. Rev Endocr Metab Disord 8: 321-330.  Imai E, Miner JN, Mitchell JA, Yamamoto KR, & Granner DK (1993). Glucocorticoid receptor-cAMP response element-binding protein interaction and the response of the phosphoenolpyruvate carboxykinase gene to glucocorticoids. Journal of Biological Chemistry 268: 5353-5356.  Jantzen H-M, Strähle U, Gloss B, Stewart F, Schmid W, Boshart M, et al. (1987). Cooperativity of glucocorticoid response elements located far upstream of the tyrosine aminotransferase gene. Cell 49: 29-38.  Kamei Y, Xu L, Heinzel T, Torchia J, Kurokawa R, Gloss B, et al. (1996). A CBP integrator complex mediates transcriptional activation and AP-1 inhibition by nuclear receptors. Cell 85: 403-414.  Karikura M, Miyase T, Tanizawa H, Taniyama T, & Takino Y (1991a). Studies on absorption, distribution, excretion and metabolism of ginseng saponins. VI. The decomposition products of ginsenoside Rb2 in the stomach of rats. Chem Pharm Bull (Tokyo) 39: 400-404.  Karikura M, Miyase T, Tanizawa H, Taniyama T, & Takino Y (1991b). Studies on absorption, distribution, excretion and metabolism of ginseng saponins. VII. Comparison of the decomposition modes of ginsenoside-Rb1 and -Rb2 in the digestive tract of rats. Chem Pharm Bull (Tokyo) 39: 2357-2361.  Kassel O, Sancono A, Krätzschmar J, Kreft B, Stassen M, & Cato ACB (2001) Glucocorticoids inhibit MAP kinase via increased expression and decreased degradation of MKP‐1. vol. 20.  Kavallaris M, Kuo DY, Burkhart CA, Regl DL, Norris MD, Haber M, et al. (1997). Taxol-resistant epithelial ovarian tumors are associated with altered expression of specific beta-tubulin isotypes. Journal of Clinical Investigation 100: 1282-1293.  Kim HA, Kim S, Chang SH, Hwang HJ, & Choi Y-n (2007). Anti-arthritic effect of ginsenoside Rb1 on collagen induced arthritis in mice. International Immunopharmacology 7: 1286-1291.  Kim HJ, Kim P, & Shin CY (2013). A comprehensive review of the therapeutic and pharmacological effects of ginseng and ginsenosides in central nervous system. Journal of Ginseng Research 37: 8-29.  83  Kim YC, Kim SR, Markelonis GJ, & Oh TH (1998). Ginsenosides Rb1 and Rg3 protect cultured rat cortical cells from glutamate-induced neurodegeneration. Journal of Neuroscience Research 53: 426-432.  Kimura T, Saunders PA, Kim HS, Rheu HM, Oh KW, & Ho IK (1994). Interactions of ginsenosides with ligand-bindings of GABA alpha and GABA beta receptors. General Pharmacology: The Vascular System 25: 193-199.  Kondo T, Kitazawa R, Yamaguchi A, & Kitazawa S (2008). Dexamethasone promotes osteoclastogenesis by inhibiting osteoprotegerin through multiple levels. Journal of Cellular Biochemistry 103: 335-345.  Laryea G, Muglia L, Arnett M, & Muglia LJ (2015). Dissection of glucocorticoid receptor-mediated inhibition of the hypothalamic–pituitary–adrenal axis by gene targeting in mice. Frontiers in Neuroendocrinology 36: 150-164.  Lee Y-N, Lee H-Y, Lee YM, Chung H-Y, Shin-Il K, Lee S-K, et al. (1998). Involvement of glucocorticoid receptor in the induction of differentiation by ginsenosides in F9 teratocarcinoma cells. The Journal of Steroid Biochemistry and Molecular Biology 67: 105-111.  Lee Y, Jin Y, Lim W, Ji S, Cho J, Ban J, et al. (2003a). Ginsenoside Rc and Re stimulate c-Fos expression in MCF-7 human breast carcinoma cells. Arch Pharm Res 26: 53-57.  Lee Y, Jin Y, Lim W, Ji S, Choi S, Jang S, et al. (2003b). A ginsenoside-Rh1, a component of ginseng saponin, activates estrogen receptor in human breast carcinoma MCF-7 cells. The Journal of Steroid Biochemistry and Molecular Biology 84: 463-468.  Lee Y, Jin Y, Lim W, Park W, Cho J, Jang S, et al. (2003c). Ginsenoside-Rb1 acts as a weak phytoestrogen in MCF-7 human breast cancer cells. Arch Pharm Res 26: 58-63.  Lee YJ, Chung E, Youl Lee K, Hee Lee Y, Huh B, & Lee SK (1997). Ginsenoside-Rg1, one of the major active molecules from Panax ginseng, is a functional ligand of glucocorticoid receptor. Molecular and Cellular Endocrinology 133: 135-140.  Lee YN, Lee HY, Chung HY, Kim SI, Lee SK, Park BC, et al. (1996). In vitro induction of differentiation by ginsenosides in F9 teratocarcinoma cells. European Journal of Cancer 32: 1420-1428.  Leung KW, Cheng Y-K, Mak NK, Chan KKC, David Fan TP, & Wong RNS (2006a). Signaling pathway of ginsenoside-Rg1 leading to nitric oxide production in endothelial cells. FEBS Letters 580: 3211-3216.  Leung KW, Leung FP, Huang Y, Mak NK, & Wong RNS (2007). Non-genomic effects of ginsenoside-Re in endothelial cells via glucocorticoid receptor. FEBS Letters 581: 2423-2428.  84  Leung KW, Pon YL, Wong RNS, & Wong AST (2006b). Ginsenoside-Rg1 induces vascular endothelial growth factor expression through the glucocorticoid receptor-related phosphatidylinositol 3-kinase/Akt and β-catenin/T-cell factor-dependent pathway in human endothelial cells. Journal of Biological Chemistry 281: 36280-36288.  Li G, Wang Z, Sun Y, Liu K, & Wang Z (2006). Ginsenoside 20(S)-protopanaxadiol inhibits the proliferation and invasion of human fibrosarcoma HT1080 cells. Basic & Clinical Pharmacology & Toxicology 98: 588-592.  Li J, Du J, Liu D, Cheng B, Fang F, Weng L, et al. (2014). Ginsenoside Rh1 potentiates dexamethasone’s anti-inflammatory effects for chronic inflammatory disease by reversing dexamethasone-induced resistance. Arthritis Research & Therapy 16: R106-R106.  Ling C, Li Y, Zhu X, Zhang C, & Li M (2005). Ginsenosides may reverse the dexamethasone-induced down-regulation of glucocorticoid receptor. General and Comparative Endocrinology 140: 203-209.  Liu L, Huang J, Hu X, Li K, & Sun C (2011). Simultaneous determination of ginsenoside (G-Re, G-Rg1, G-Rg2, G-F1, G-Rh1) and protopanaxatriol in human plasma and urine by LC–MS/MS and its application in a pharmacokinetics study of G-Re in volunteers. Journal of Chromatography B 879: 2011-2017.  Liu Y, Titus L, Barghouthi M, Viggeswarapu M, Hair G, & Boden SD (2004). Glucocorticoid regulation of human BMP-6 transcription. Bone 35: 673-681.  Livak KJ, & Schmittgen TD (2001). Analysis of relative gene expression data using real-time quantitative PCR and the 2−ΔΔCT method. Methods 25: 402-408.  Losel R, & Wehling M (2003). Nongenomic actions of steroid hormones. Nat Rev Mol Cell Biol 4: 46-55.  Maier A, Zimmermann C, Beglinger C, Drewe J, & Gutmann H (2007). Effects of budesonide on P-glycoprotein expression in intestinal cell lines. British Journal of Pharmacology 150: 361-368.  Malkoski SP, Handanos CM, & Dorin RI (1997). Localization of a negative glucocorticoid response element of the human corticotropin releasing hormone gene. Molecular and Cellular Endocrinology 127: 189-199.  Martínez-Jiménez CP, Gómez-Lechón MJ, Castell JV, & Jover R (2006). Underexpressed coactivators PGC1α and SRC1 impair hepatocyte nuclear factor 4α function and promote dedifferentiation in human hepatoma cells. Journal of Biological Chemistry 281: 29840-29849.  McKay LI, & Cidlowski JA (1999). Molecular control of immune/inflammatory responses: Interactions between nuclear factor-κB and steroid receptor-signaling pathways. Endocrine Reviews 20: 435-459.  85  McKay LI, & Cidlowski JA (2000). CBP (CREB binding protein) integrates NF-κB (nuclear factor-κB) and glucocorticoid receptor physical interactions and antagonism. Molecular Endocrinology 14: 1222-1234.  Meng Z, Zhang H, Zhao Y, Lan J, & Du L (2007). Transport behavior and efflux of Rg1 in rat pulmonary epithelial cells. Biomedical Chromatography 21: 635-641.  Moutsatsou P, Kassi E, & Papavassiliou AG (2012). Glucocorticoid receptor signaling in bone cells. Trends in Molecular Medicine 18: 348-359.  Nag SA, Qin J, Wang W, Wang M-H, Wang H, & Zhang R (2012). Ginsenosides as anticancer agents: in vitro and in vivo activities, structure-activity relationships, and molecular mechanisms of action. Frontiers in Pharmacology 3.  Newton R (2000). Molecular mechanisms of glucocorticoid action: what is important? Thorax 55: 603-613.  Nicolaides NC, Galata Z, Kino T, Chrousos GP, & Charmandari E (2010). The human glucocorticoid receptor: Molecular basis of biologic function. Steroids 75: 1-12.  Novotna A, Korhonova M, Bartonkova I, Soshilov AA, Denison MS, Bogdanova K, et al. (2014). Enantiospecific effects of ketoconazole on aryl hydrocarbon receptor. PLoS ONE 9: e101832.  Oakley RH, & Cidlowski JA (2011). Cellular processing of the glucocorticoid receptor gene and protein: New mechanisms for generating tissue-specific actions of glucocorticoids. Journal of Biological Chemistry 286: 3177-3184.  Oakley RH, Sar M, & Cidlowski JA (1996). The human glucocorticoid receptor beta isoform: Expression, biochemical properties, and putative function. Journal of Biological Chemistry 271: 9550-9559.  Patel R, Williams-Dautovich J, & Cummins CL (2014). Minireview: New molecular mediators of glucocorticoid receptor activity in metabolic tissues. Molecular Endocrinology 28: 999-1011.  Pratt WB (1993). The role of heat shock proteins in regulating the function, folding, and trafficking of the glucocorticoid receptor. Journal of Biological Chemistry 268: 21455-21458.  Radad K, Moldzio R, & Rausch W-D (2011). Ginsenosides and their CNS targets. CNS Neuroscience & Therapeutics 17: 761-768.  Revollo JR, & Cidlowski JA (2009). Mechanisms generating diversity in glucocorticoid receptor signaling. Annals of the New York Academy of Sciences 1179: 167-178.  Rogatsky I, Hittelman AB, Pearce D, & Garabedian MJ (1999). Distinct glucocorticoid receptor transcriptional regulatory surfaces mediate the cytotoxic and cytostatic effects of glucocorticoids. Molecular and Cellular Biology 19: 5036-5049. 86   Salim KN, McEwen BS, & Chao HM (1997). Ginsenoside Rb1 regulates ChAT, NGF and trkA mRNA expression in the rat brain. Molecular Brain Research 47: 177-182.  Schäcke H, Berger M, Rehwinkel H, & Asadullah K (2007). Selective glucocorticoid receptor agonists (SEGRAs): Novel ligands with an improved therapeutic index. Molecular and Cellular Endocrinology 275: 109-117.  Sharma D, Bhave S, Gregg E, & Uht R (2013). Dexamethasone induces a putative repressor complex and chromatin modifications in the CRH promoter. Molecular Endocrinology 27: 1142-1152.  Shieh P-C, Tsao C-W, Li J-S, Wu H-T, Wen Y-J, Kou D-H, et al. (2008). Role of pituitary adenylate cyclase-activating polypeptide (PACAP) in the action of ginsenoside Rh2 against beta-amyloid-induced inhibition of rat brain astrocytes. Neuroscience Letters 434: 1-5.  So AY-L, Chaivorapol C, Bolton EC, Li H, & Yamamoto KR (2007). Determinants of cell- and gene-specific transcriptional regulation by the glucocorticoid receptor. PLoS Genetics 3: e94.  Stahn C, & Buttgereit F (2008). Genomic and nongenomic effects of glucocorticoids. Nat Clin Pract Rheum 4: 525-533.  Surjit M, Ganti Krishna P, Mukherji A, Ye T, Hua G, Metzger D, et al. (2011). Widespread negative response elements mediate direct repression by agonist-liganded glucocorticoid receptor. Cell 145: 224-241.  Tawab MA, Bahr U, Karas M, Wurglics M, & Schubert-Zsilavecz M (2003). Degradation of ginsenosides in humans after oral administration. Drug Metabolism and Disposition 31: 1065-1071.  Verhoog N, Allie-Reid F, Vanden Berghe W, Smith C, Haegeman G, Hapgood J, et al. (2014). Inhibition of corticosteroid-binding globulin gene expression by glucocorticoids involves C/EBP beta. PLoS ONE 9: e110702.  Wang J-C, Gray NE, Kuo T, & Harris CA (2012). Regulation of triglyceride metabolism by glucocorticoid receptor. Cell & Bioscience 2: 1-9.  Wang Q, Blackford JAJ, Song L-N, Huang Y, Cho S, & Simons SSJ (2004). Equilibrium interactions of corepressors and coactivators with agonist and antagonist complexes of glucocorticoid receptors. Molecular Endocrinology 18: 1376-1395.  Wang Y-Z, Chen J, Chu S-F, Wang Y-S, Wang X-Y, Chen N-H, et al. (2009). Improvement of memory in mice and increase of hippocampal excitability in rats by ginsenoside Rg1's metabolites ginsenoside Rh1 and protopanaxatriol. Journal of Pharmacological Sciences 109: 504-510.  87  Webster JC, & Cidlowski JA (1999). Mechanisms of glucocorticoid receptor-mediated repression of gene expression. Trends in Endocrinology & Metabolism 10: 396-402.  Wu J, Pan Z, Wang Z, Zhu W, Shen Y, Cui R, et al. (2012). Ginsenoside Rg1 protection against β-amyloid peptide-induced neuronal apoptosis via estrogen receptor α and glucocorticoid receptor-dependent anti-protein nitration pathway. Neuropharmacology 63: 349-361.  Xie H-T, Wang G-J, Chen M, Jiang X-l, Li H, Lv H, et al. (2005). Uptake and metabolism of ginsenoside Rh2 and its aglycon protopanaxadiol by Caco-2 cells. Biological and Pharmaceutical Bulletin 28: 383-386.  Xiong J, Sun M, Guo J, Huang L, Wang S, Meng B, et al. (2009). Active absorption of ginsenoside Rg1 in vitro and in vivo: the role of sodium-dependent glucose co-transporter 1. Journal of Pharmacy and Pharmacology 61: 381-386.  Xu B-B, Liu C-Q, Gao X, Zhang W-Q, Wang S-W, & Cao Y-L (2005). Possible mechanisms of the protection of ginsenoside Re against MPTP-induced apoptosis in substantia nigra neurons of Parkinson's disease mouse model. Journal Of Asian Natural Products Research 7: 215-224.  Yang C-S, Ko S-R, Cho B-G, Shin D-M, Yuk J-M, Li S, et al. (2008). The ginsenoside metabolite compound K, a novel agonist of glucocorticoid receptor, induces tolerance to endotoxin-induced lethal shock. Journal of Cellular and Molecular Medicine 12: 1739-1753.  Yang J-H, Han S-J, Ryu JH, Jang I-S, & Kim D-H (2009). Ginsenoside Rh2 ameliorates scopolamine-induced learning deficit in mice. Biological and Pharmaceutical Bulletin 32: 1710-1715.  Yang L, Deng Y, Xu S, & Zeng X (2007). In vivo pharmacokinetic and metabolism studies of ginsenoside Rd. Journal of Chromatography B 854: 77-84.  Yu Y, Zhou Q, Hang Y, Bu X, & Jia W (2007). Antiestrogenic effect of 20S-protopanaxadiol and its synergy with tamoxifen on breast cancer cells. Cancer 109: 2374-2382.  Yudt MR, Jewell CM, Bienstock RJ, & Cidlowski JA (2003). Molecular origins for the dominant negative function of human glucocorticoid receptor beta. Molecular and Cellular Biology 23: 4319-4330.  Zheng H, Jeong Y, Song J, & Ji GE (2011). Oral administration of ginsenoside Rh1 inhibits the development of atopic dermatitis-like skin lesions induced by oxazolone in hairless mice. International Immunopharmacology 11: 511-518.  Zhou C, Tabb MM, Sadatrafiei A, Grün F, & Blumberg B (2004). Tocotrienols activate the steroid and xenobiotic receptor, SXR, and selectively regulate expression of its target genes. Drug Metabolism and Disposition 32: 1075-1082.   88  Appendix    Appendix A. Preliminary Assay of Ginsenoside Rg1 in hGRα-mediated Transactivation. LS180 cells were transiently transfected for 24 h with hGRα-expressing vector (pCMV6-XL5-hGR), GRE-containing firefly luciferase reporter vector (pGRE-luc), and Renilla luciferase internal control vector (pGL4.74[hRluc/TK]). Transfected cells were subsequently treated with DMSO (0.1% v/v; vehicle control), ginsenoside Rg1 (30 or 50 μM), RU486 (100 nM; hGRα antagonist), dexamethasone (100 nM; hGRα agonist), or combination treatment of dexamethasone (100 nM) and RU486 (100 nM) for 24 h before cell lysis for luminescence detection. Data are expressed as normalized ratio of firefly luciferase and Renilla luciferase luminescence signal, and shown as mean ± SD for a single experiment performed in triplicate.  Normalized Luciferase Activity0. 30 MRg1 50 MRU486DexDex+RU486


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items