Non-genomic glucocorticoid effects to provide the basis for new drug developments

https://doi.org/10.1016/j.mce.2005.11.012Get rights and content

Abstract

Glucocorticoids act via genomic and non-genomic actions. The genomic glucocorticoid actions are well known and new details on processes of transactivation and transrepression have been reported recently. Here we describe the current knowledge on non-genomic glucocorticoid actions and discuss why these actions are considered to be of therapeutic relevance.

It is assumed that rapid non-genomic glucocorticoid effects are mediated by three different mechanisms: (1) physicochemical interactions with cellular membranes (non-specific non-genomic effects); (2) membrane-bound glucocorticoid receptor (mGCR)-mediated non-genomic effects; and (3) cytosolic glucocorticoid receptor (cGCR)-mediated non-genomic effects. With regard to the first mechanism, we discuss here lazaroids and the novel development of drug targeting with liposomes as the carrier system for glucocorticoids. The clinical use of the latter two mechanisms is still speculative, but intriguing ideas are being discussed in this regard.

Introduction

The mechanisms of glucocorticoid actions can be divided into genomic and non-genomic effects. Genomic effects are known to be mediated by transactivation and transrepression processes. These mechanisms have been recently reviewed more in detail (Schacke et al., 2004, Perretti et al., 2003, Song et al., 2005).

Here we discuss the therapeutic relevance of rapid non-genomic glucocorticoid effects which are assumed to be mediated by three mechanisms: (1) physicochemical interactions with cellular membranes (non-specific non-genomic effects) (Buttgereit and Scheffold, 2002, Buttgereit et al., 2004); (2) cytosolic glucocorticoid receptor (cGCR)-mediated non-genomic effects (Croxtall et al., 2000); and (3) membrane-bound glucocorticoid receptor (mGCR)-mediated non-genomic effects (Bartholome et al., 2004). The therapeutic relevance of non-genomic glucocorticoid actions is an issue of ongoing discussion. For example, very recently, beneficial rapid non-genomic glucocorticoid effects have been described for the first time in vivo in the treatment of seasonal allergic rhinitis (Tillmann et al., 2004).

Rapid glucocorticoid actions (which occur within seconds) are being considered as a result of physicochemical interactions with cellular membranes. How can we explain these non-specific non-genomic effects? Currently, the following hypothesis is favoured: Glucocorticoid molecules intercalate at high concentrations in cellular membranes (plasma and mitochondrial membranes) which alter cell functions by influencing cation transport through the plasma membrane and by increasing the proton leak of the mitochondria. The impaired cation cycling across and the compromised ATP production via oxidative phosphorylation are considered to result in immunosuppressive effects and to a reduced activity of inflammatory processes (Buttgereit and Scheffold, 2002, Buttgereit et al., 2004). In the following section, we would like to explain more in detail the experimental results that have led to this theory.

The background to explain the rapid non-specific non-genomic glucocorticoid effects is provided by considerations of the cellular energy metabolism. Every organism and every cell needs metabolic energy whereby ATP is the major source. Also immune cells produce and consume certain amounts of ATP for their housekeeping activities and for specific immune functions. ATP-dependent immune functions include cytokinesis, migration, phagocytosis, antigen processing and antigen presentation, signalling and effector functions such as the synthesis of antibodies, cytotoxicity and regulatory functions (Buttgereit et al., 2000). The major sources for ATP production are glycolysis and oxidative phosphorylation. The most important ATP-consuming pathways are identified to be the transport of cations and synthesis of macromolecules (Buttgereit et al., 2000). In the case of energy deficit, the functions of immune cells are known to be impaired (Meldrum et al., 1994, Karlsson and Nassberger, 1992, Sanchez-Alcazar et al., 1995).

Given this background, we have shown a clear hierarchy of energy-consuming pathways in case of reduced energy supply: Pathways of macromolecule biosynthesis (protein synthesis and RNA/DNA synthesis) are most sensitive to energy restriction whereas cation transport ATPases are much less affected. The mitochondrial proton leak is least sensitive to enery supply. These results were obtained by quantitating main ATP-consuming pathways under progressively restricted mitochondrial ATP production by using myxothiazol, a specific inhibitor of the electron transport chain. Obviously, processes not essential for the immediate needs of the cell will be given up before those that are more critical for ionic integrity (Buttgereit and Brand, 1995, Buttgereit et al., 1999).

Why are these experimental results important to understand rapid glucocorticoid effects? If immune cells are treated with high, but clinically relevant concentrations of methylprednisolone (the glucocorticoid most commonly used for high-dose therapy), then the following effects on bioenergetics are observed within seconds (Buttgereit and Scheffold, 2002, Buttgereit et al., 1997, Buttgereit et al., 2004):

  • 1.

    The glucocorticoid instantly inhibits respiration of Con A-stimulated thymocytes and human immune cells (Fig. 1) at concentrations that leave quiescent cells unaffected.

  • 2.

    Methylprednisolone not only reverses but also prevents the Con A-effect on respiration in a dose dependent manner.

  • 3.

    Con A is known to produce a dramatic increase of cytoplasmic calcium concentration. This effect is clearly reduced in the presence of therapeutically relevant drug concentrations or even abolished at suprapharmacological doses.

  • 4.

    Methylprednisolone inhibits calcium and sodium cycling across the plasma membrane but has little effect on protein synthesis. The inhibition of cation cycling in Con A-stimulated thymocytes by the glucocorticoid is caused by direct effects and not by a reduction in ATP production even though methylprednisolone reduces ATP availability to some extent by inhibiting the reactions of substrate oxidation and by increasing mitochondrial proton leak.

The comparison of these methylprednisolone effects with those of myxothiazol we have reported above leads to the following hypothetical explanation: Methylprednisolone dissolves in membranes and affects physicochemical membrane properties and the activities of membrane-associated proteins (Buttgereit and Scheffold, 2002, Buttgereit et al., 2004). The resulting inhibition of calcium and sodium entry across the plasma membrane would explain (i) the decrease in ATP use (and therefore oxygen consumption, see Fig. 1) for plasma membrane ion cycling and (ii) the drop in cytosolic free calcium. A direct effect on the mitochondrial inner membrane would explain the observed increase in proton permeability and the consequent partial uncoupling of oxidative phosphorylation. This is suggested to result in immunosuppression since immune cell function depends on proper functioning of these processes.

Substances which act similarly to high-dose glucocorticoids are lazaroids (21-aminosteroids such as tirilazad). Lazaroids intercalate into biological membranes without binding to glucocorticoid receptors. In this way, they cause a stabilisation of membranes and inhibition of destructive lipid peroxidation which results in the prevention of oxidative cell damage. Tirilazad is used in neurotraumatology (Braughler and Pregenzer, 1989, Dissemond et al., 2003, Bath et al., 2001). To our knowledge, however, these drugs have not been used to investigate their immunosuppressive effects in clinical trials.

Coming back to glucocorticoids, a very interesting approach in the context of rapid non-specific non-genomic glucocorticoid effects is drug targeting with liposomes as the carrier system for glucocorticoids. Newly developed glucocorticoid-containing liposomes accumulate selectively at the site of inflammation where there is an increased permeability of the local vascular endothelium. By reaching very high concentrations (>10−5 M) at the site of inflammation for several hours, 100% utilisation of the genomic actions and, in addition, significant therapeutically beneficial non-genomic actions are achieved (Song et al., 2005). Although the systemic glucocorticoid concentrations (plasma levels) remain relatively high, they give rise to fewer adverse reactions because the steroid is encapsulated in the liposome. Schmidt et al. (2003) recently used prednisolone liposomes to demonstrate the success of this innovative approach in the treatment of experimental autoimmune encephalomyelitis in Lewis rats (EAE).

Metselaar et al., 2003, Metselaar et al., 2004 have recently reported a complete remission of experimental arthritis by joint targeting of glucocorticoids with long-circulating liposomes (Song et al., 2005). A single intravenous administration of 10 mg/kg liposomal prednisolone phosphate in rat experimental arthritis and in murine collagen type II-induced arthritis showed a profound antiinflammatory effect which lasted for more than 1 week. In contrast, 10 mg/kg of unencapsulated liposomal prednisolone phosphate was less effective even after repeated daily injections. Moreover, these authors also found the process of cartilage erosion to be profoundly attenuated under this treatment. From the mechanistic point of view, the observation is interesting that the liposomes accumulate in the inflamed joint, i.e. in the proximity of blood vessels mainly in the synovial lining. This strongly suggested that liposomes selectively localise in cells with phagocytotic capacity.

These initial observations suggest targeted delivery using long-circulating liposomal glucocorticoids to be an effective and novel therapeutic option in arthritis (Song et al., 2005). Hopefully clinical trials in humans will yield similar intriguing results as in animal models. Future work will also have to optimise both the lipid shells and the incorporated glucocorticoids.

Apart from the physicochemical interactions with cellular membranes (non-specific non-genomic effects) mentioned above, other mechanisms of rapid glucocorticoid actions are being discussed. So called “mGCR-mediated non-genomic effects” are suggested to be mediated by membrane-bound glucocorticoid receptors (mGCR). The existence of such membrane-bound glucocortioid receptors had already been shown in amphibian neuronal membranes (Orchinik et al., 1991) and in lymphoma cells (Gametchu et al., 1993, Chen et al., 1999, Sackey et al., 1997). Recently, we were able to identify for the first time mGCR in normal human peripheral blood mononuclear cells (Bartholome et al., 2004, Buttgereit et al., 2004, Song et al., 2005). This observation was made using the novel technique of high-sensitivity immunofluorescent staining. The method uses antibody-conjugated magnetofluorescent liposomes, which can increase fluorescence signal intensity up to 1000-fold compared with conventional methods and allows the detection of 50–100 target molecule per cell. Up to 9.2 and 12.3% of the monocytes and B lymphocytes were positive for mGCR, respectively. T lymphocytes never showed any significant mGCR expression (Bartholome et al., 2004). Further experiments demonstrated immunostimulation with lipopolysaccharide to increase significantly the percentage of mGCR-positive monocytes. In the presence of brefeldin A, immunostimulation with lipopolysaccharide did not increase the number of mGCR-positive monocytes demonstrating the ability of brefeldin A to abrogate the induced up-regulation of mGCR by inhibiting the secretory pathway. We concluded that immunostimulation induces cellular events whereby mGCR are actively upregulated and transported through the cell.

On the basis of this data, we hypothesized that there is a clinically relevant correlation between the number of mGCR-positive cells and disease-related activity of the immune system. We indeed found a strong positive correlation between the frequency of mGCR-positive monocytes and various parameters of disease activity in patients suffering from rheumatoid arthritis (Bartholome et al., 2004). One way to interpret these data is that mGCR may play a role in the aetiopathogenesis of disease. However, we suggest being more likely that immunostimulation or high disease activity stimulate mGCR expression in immune cells such as monocytes which in turn leads to a significantly higher percentage of cells undergoing mGCR-mediated glucocorticoid-induced apoptosis (Sackey et al., 1997). This putative process would diminish the activity of the immune system and could be therefore considered as a mechanism of negative feed-back regulation. Further experiments will have to be made in order to prove this hypothesis. A new approach of optimising glucocorticoid therapy could be to develop drugs selectively binding to the mGCR, but this is clearly speculative. First the function of mGCR needs to be investigated in detail. Though there is no evidence for specific signalling pathways associated with mGCR, this mechanism might explain the rapid dexamethasone-induced adenylate cyclase/protein kinase A-dependent inhibition of chloride ion secretion in bronchial epithelium (Goulding, 2004, Urbach et al., 2002).

Another hypothesis to explain rapid glucocorticoid effects is that glucocorticoids bind to its cytosolic glucocorticoid receptor where they not only cause classical genomic, but also rapid non-genomic effects resulting in interactions with signalling processes. The underlying theory is the following: Arachidonic acid (AA) is an essential mediator which is important for cell growth and several metabolic reactions (e.g. production of inflammatory cytokines in the setting of the inflammatory cascade). AA release from cell membrane phospholipids is controlled by mediators which among others involve growth factors (like insulin or PDGF), adaptor proteins such as SOS, Grb2, the GTPase p21ras, erk2/ MAPK (mitogen activated protein kinases), cPLA2 (cytosolic phospholipaseA2) and especially lipocortin 1 (LC1) (Ahn et al., 1991, De Vries-Smits et al., 1992, Lin et al., 1993, Croxtall et al., 1995, Croxtall et al., 1996, Croxtall et al., 1998).

With regard to cGCR-mediated glucocorticoid effects, Croxtall et al. (2000). have recently reported that epidermal growth factor (EGF) stimulated cPLA2 (cytosolic PLA2) activation with subsequent arachidonic acid release can be inhibited by dexamethasone. This effect is considered to be a glucocorticoid receptor-dependent (RU486-sensitive), but transcription-independent (actinomycin-insensitive) mechanism.

What could be the explanation for this observed effect? To answer this question, it should be noted that the unactivated (unligated) cGCR is retained in the cytoplasm as a multi-protein complex consisting of heat shock proteins and several kinases of the MAPK signalling system (chaperones and co-chaperones), including Src. Following glucocorticoid binding, the cGCR is released from this complex, thus revealing domains of the receptor that are able to bind specifically to glucocorticoid responsive DNA elements. This mechanism is known to account for genomic glucocorticoid effects. However, the role of the remaining signalling components of the multi-protein complex is less clearly defined. In their study, Croxtall et al. (2000) showed that there is a rapid release of not only glucocorticoid receptor protein, but also of Src from the multi-protein complex which follows the binding of dexamethasone to the cGCR. However, the release of glucocorticoid receptor and Src appear to be distinct events, since the nuclear translocation of glucocorticoid receptor is not required for subsequent activation of lipocortin 1 and rapid inhibition of arachidonic acid release. Rather, it appeared that it is the rapid effect of Scrd that fulfils this role.

In summary, it is suspected that the binding of glucocorticoids to the cGCR not only induces classical genomic effects, but also rapidly controls intracellular signalling through other components of the multi-protein complex. It may be possible in the future to make therapeutical use of these glucocorticoid effects. The clinical background for this assumption is that glucocorticoids have been shown recently to activate endothelial nitric oxide synthase in a non-genomic manner and mediated by PI3K and Akt phosphorylation. This ultimately leads to vasorelaxation, which might explain some of the cardioprotective effects of glucocorticoids (Lösel and Wehling, 2003, Hafezi-Moghadam et al., 2002). There are other examples in the field of oncology where blocking of small molecules represent very promising therapeutic approaches. In this regard, we would like to mention the inhibition of IGFBP2 (insulin-like growth factor binding protein) which enhances the invasion capacity of ovarian cancer cells (Lee et al., 2005). Other examples include the EWS-FlI1 fusion protein (Ewing sarcoma family of tumors), translocation fusion protein ETV6-NTRK3 (congenital fibrosarcoma or cellular mesoblastic nephroma) and BCR-ABL kinase inhibitor imatinib mesylate (chronic myeloid leukaemia). This is only a selection to demonstrate that advances in molecular biology have already led to the identification of quite a lot new protein targets along with an increased understanding of the biologic role of these proteins (Uren and Toretsky, 2005). This research is ultimately designed to improve therapeutic option as, e.g. demonstrated with tyrosine kinase inhibitors such as gefitinib which is currently tested in phase II trials in breast cancer. Monoclonal antibodies targeting epidermal growth factor receptor (EGRF) represent another approach to interfere with EGRF-driven signal transduction. For example, the chimerized monoclonal antibody cetuximab is designed to treat advanced colorectal cancer (Caponigro et al., 2005). We have discussed these examples to illustrate that the cGCR-mediated non-genomic glucocorticoid effects could provide the basis for targeted therapies in the future.

Glucocorticoids are clinically very important drugs; however, their use is limited by side effects. This drives interesting research in order to understand better the underlying mechanisms of action. The non-genomic glucocorticoid effects we describe here are currently in the focus of research. The ultimate aim is to convert results of basic research in this regard into novel therapeutic options to improve the benefit-risk-ratio of glucocorticoid therapy.

References (38)

  • B. Bartholome et al.

    Membrane glucocorticoid receptors (mGCR) are expressed in normal human peripheral blood mononuclear cells and up-regulated after in nitro stimulation and in patients with rheumatoid arthritis

    FASEB J.

    (2004)
  • Bath, P.M., Iddenden, R., Bath, F.J., Orgogozo, J.M., 2001. Tirilazad for acute ischaemic stroke. Cochrane Database...
  • F. Buttgereit et al.

    Bioenergetics of immune functions: fundamental and therapeutic aspects

    Immunol. Today

    (2000)
  • F. Buttgereit et al.

    Glucocorticoids in the treatment of rheumatic diseases

    Arthritis Rheum.

    (2004)
  • F. Buttgereit et al.

    A hierarchy of ATP-consuming processes in mammalian cells

    Biochem. J.

    (1995)
  • F. Buttgereit et al.

    Methylprednisolone inhibits uptake of Ca2+ and Na+ ions into concanavalin A-stimulated thymocytes

    Biochem. J.

    (1997)
  • F. Caponigro et al.

    New drugs in cancer therapy, national tumor institute, Naples 17–18 June 2004

    Anticancer Drugs

    (2005)
  • F. Chen et al.

    Association of the glucocorticoid receptor alternatively-spliced transcript 1A with the presence of the high molecular weight membrane glucocorticoid receptor in mouse lymphoma cells

    J. Cell Biochem.

    (1999)
  • J.D. Croxtall et al.

    Glucocorticoids act within minutes to inhibit recruitment of signalling actors to activated EGF receptors through a receptor-dependent, transcription-independent mechanism

    Br. J. Pharmacol.

    (2000)
  • Cited by (171)

    • The Use of Albuterol/Budesonide as Reliever Therapy to Reduce Asthma Exacerbations

      2024, Journal of Allergy and Clinical Immunology: In Practice
    • Comparison of mechanisms of angiostasis caused by the anti-inflammatory steroid 5α-tetrahydrocorticosterone versus conventional glucocorticoids

      2022, European Journal of Pharmacology
      Citation Excerpt :

      Indeed membrane glucocorticoid receptors have different ligand binding specificity from cytosolic glucocorticoid receptors (Mitre-Aguilar et al., 2015) and are present in immune cells (Buttgereit et al., 2004; Stahn et al., 2007). Non-genomic cardioprotective effects of glucocorticoids have been demonstrated (Haller et al., 2008; Tasker et al., 2006) having rapid therapeutic effects after stroke or myocardial infarction (Song and Buttgereit, 2006). In fact fewer adverse cardiovascular effects could be achieved through making use of non-genomic mechanisms of glucocorticoids; for example to decrease/eliminate genomic effects on wound repair and cardiac cell remodelling (Lee et al., 2012).

    • The physiology of stress and the human body’s response to stress

      2022, Epigenetics of Stress and Stress Disorders
    View all citing articles on Scopus
    View full text