**3. Mechanism of action of GC**

A better understanding of the mechanisms of GC action is crucial for understanding how to utilise these drugs more effectively in the clinical setting while minimising their adverse effects. In general terms the mechanisms of action can be divided into genomic and nongenomic. The genomic actions of GC are medicated through gene transcription and take hours to days to occur while the non-genomic actions are more rapid (Fig 5).

#### **3.1 Genomic mechanisms**

8 Rheumatoid Arthritis – Treatment

Indeed a cohort study examining the interaction between GC therapy and cardiovascular risk in RA showed GC therapy to be associated with an increased risk only if patients were rheumatoid factor (RF) positive (Davis et al., 2007). In fact in RF negative patients GC were not associated with increased risk regardless of the cumulative dose and indeed showed a

**percentiles)** 

**AEs per 100 patient years**

15 (3-28)

9 (2-236)

7 (3-34)

**Type of Adverse Event Median:(25th-75th**

Infectious (viral, bacterial, skin infections) 15 (3-15) Gastrointestinal (peptic ulcer, pancreatitis) 10 (4-20)

Dermatological (cutaneous atrophy, acne, hirsutism, alopecia) 5 (2-80) Musculoskeletal (osteoporosis, osteonecrosis, myopathy) 4 (3-9) Ophthalmological (glaucoma, cataract) 4 (0-5)

Fig. 4. Reported adverse events in GC treated patients with rheumatological diseases.

In summary, GC are widely used in the management of RA and rheumatologists have over 60 years experience in their use. At low doses they act as to reduce the symptoms of RA in the first 6 to 12 months but in addition, their use early in the disease process substantially slows the progression of joint destruction and results in less disability in the long term. Remarkably this joint protective effect seems to be sustained years after GC are discontinued and for this reason GC can both be considered to be true "disease modifying" antirheumatic drugs (Bijlsma et al., 2010) and to have some kind of effect on the underlying long term disease process. At higher doses they are effective in treating severe and life threatening flares of disease. Adverse effects remain a significant problem but in the balance of risk versus benefit, GC (especially at lower doses) can be considered relatively safe. The summary of the EULAR recommendations in GC use are reproduced below and are a useful

A better understanding of the mechanisms of GC action is crucial for understanding how to utilise these drugs more effectively in the clinical setting while minimising their adverse effects. In general terms the mechanisms of action can be divided into genomic and nongenomic. The genomic actions of GC are medicated through gene transcription and take

hours to days to occur while the non-genomic actions are more rapid (Fig 5).

trend towards being protective.

Psychological and behavioural (minor mood disturbance, psychosis)

Endocrine and metabolic

(dyslipidemia, oedema, hypertension, heart failure)

(glucose intolerance, diabetes, fat redistribution)

tool for clinicians to refer to in their daily practice.

**3. Mechanism of action of GC** 

Cardovascular

GC have a lipophilic structure and low molecular mass. They therefore pass easily through the cell membrane and exert their effects mainly through binding with the glucocorticoid receptor (GCR) in the cytoplasm (Rhen & Cidlowski, 2005). There are two isoforms of the GCR, α and β. GCR-α is the biologically active form of the receptor and mediates the intracellular effects of GC. GCR-β is an alternativley spliced form which may act as dominant neagtive inhibitor of GC action (Lewis-Tuffin, 2006). Over expression of GCR-β may be implictaed in GC resistance as will be discussed later in this chapter.

The GCR-α in the cytoplasm is associated with various heat shock proteins (HSPs) including HSP40, HSP56, HSP 70 and HSP90 (McLaughlin & Jackson, 2002) which are released when the receptor binds to GC After binding, the complex translocates to the neucleus where it exerts its effects on gene transcription (Davies et al., 2002). At the neucleus GCRs homodimerise and bind to GC response elements (GREs) in the promoter region of the target genes and lead to activation or inhbition of gene transcription. In addition the DNA bound GCR can also directly bind transcription co-activator molecules and exert further actions this way (Barnes, 2006).

In activated inflammatory cells there is an additional route for GC action. This is because inflammatory stimuli ultimately lead to the activation of neuclear factor κB (NFκB) which binds to specific κB recognition sites on promoter regions of inflammatory genes in addition to coactivators such as cyclic AMP response element binding protein (CBP). The coactivators cause acetylation of core histones which leads to their unravelling and opens up the genes for transcription. Activated GCRs and the HSPs that are released when the GC binds to the receptor inhibit this effect directly by binding the coactivators and recruiting histone deacetylase (HDAC) 2 which inhibits acetylation (Rhen & Cidlowski, 2005). GC also switch on the transcription of certain genes including mitogen-activated protein (MAP) kinase phosphatase 1 (MKP1) hence inhibiting the MAP kinase pathway which is involved in proinflammatory gene transcription (Clark, 2003). The existence of this pathway has led to a search for ways of enhancing this GC effect, which would apply only in activated inflammatory cells and would therefore not be relevant to other body tissues, and hence would not contribute to adverse effects.

#### **3.2 Non-genomic mechanisms**

Some of the effects of GC occur within minutes of their administration especially at high intravenous doses (Croxtall et al., 2000). The mechanisms involved in mediating this rapid action are non-genomic as they are transcription independent. So far three such nongenomic actions of GC have been described. The first involved the observation of the rapid reversal by dexamethasone of epidermal growth factor-stimulated activation of phospholipase A2. It is thought that this effect is medicated by chaperone molecules such as Src which are rapidly released from the GCR-GC complex on ligation (Croxtall et al., 2000).

Non-specific non-genomic effects are seen at very high doses of GC therapy and are thought to be due the saturation of all available GCRs in the cells at doses above 100mg perdnisolone or equivalent (Tyrrell & Baxter, 1995). At these doses it is thought that GC molecules dissolve into the membranes and alter proton leak hence influencing membrane transport (Buttgereit & Scheffold, 2002).

The Clinical Role of Glucocorticoids in the Management of Rheumatoid Arthritis 11

Skin atrophy Type I collagen (X) Type II collagen (X) (X)

Wound healing Pro-inflmmatory genes X

 OPG (X) (X) Osteocalcin X Type I collagen (X)

Psychiatric 5HT1A receptor X

PEPCK X

 sgk X Fig. 6. Adverse effect associated proteins: regulation by GC and mechanisms. Reproduced

POMC/ACTH X

TAT X

**Molecule DNA Dependent** 

Tenscin C (X) (X)

(X)

CRH X

**Adverse Effect Primary Targeted** 

Osteoporosis Osteoblast/osteocyte apoptosis

 Ubiquitin-proteasome pathway

Glaucoma TIGR/MYOC gene product

HPA suppression

Diabetes Mellitus

glycosaminoglycans

OPG-L X

Muscle atrophy Glutamine synthetase (X)

 Fibronectin (X) Type IV collagen (X) Type I collagen (X)

 AAT X G6Pase X

Hypertension αENAC X

from Schäcke et al, Pharmacology and therapeutics 96 (2002) 23-43

Sulfated

**Mechanism [X= Confirmed, (X)= Possible]**

X

X

**Activation Repressio**

**DNA Independen t** 

**<sup>n</sup> Repression** 

(X) (X)

Fig. 5. Cellular action of glucocorticoids

It is now thought that GC also have specific non-genomic effects that are mediated through membrane bound GCRs which are found in small numbers on human peripheral blood mononuclear cells (Bartholome et al., 2004). Moreover, stimulation of these cells in vitro by lipopolyscaharide (LPS) increases the percentage of membrane GCR expressing monocytes indicating an active upregulation of this process (Bartholome et al., 2004). Interestingly in patients with RA who have an activated immune system, the percentage memebrane GCR expressing monocytes is increased, in keeping with the in-vitro observations. These membrane expressed receptors are thought to be variants of the classical cytoplasmic GCRs (Löwenberg et al., 2007) and have recently been shown to also interact with the MAP kinase pathway (Strehl et al., 2011) Moreover, the engagement of these receptors is thought to inhibit T cell signalling by acting through downstream TCR associated signalling proteins lymphocytespecific tyrosine kinase (LCK) and FYN oncogene (Lowenberg et al., 2006). It is possible that memberane glucocorticoid receptors will prove to have therapeutic implications.
