Discuss the evidence for non-genomic actions of steroid hormones.
steroid hormones
All natural steroid hormones are made from cholesterol and are modifications of the basic C19 structure. There are 5 distinct classes of steroid hormones: progestogens, glucocorticoids, mineralocorticoids, androgens, oestrogens and the steroid derivative D3. They are all structurally and functionally distinct, and therefore act on their target cells in various ways, with genomic and/or nongenomic actions, as recently discovered. Through these effects, steroids control many physiological processes.
How do steroid hormones act?
Traditionally, it has been thought that steroid hormones enter the cell and bind to classical steroid receptors, located mainly in the cytosol. Ligand binding induces a conformational change in the receptor protein and causes dimerisation, as well as disassociation of heat-shock proteins. These receptors act as transcription factors. In the nucleus, the DNA binding domain (DBD) on the protein binds the hormone response element (HRE), modulating gene transcription. The protein made from the newly synthesised mRNA elicits the genomic response. This process occurs with a time lag of many hours, and the pathway is sensitive to inhibitors such as actinomycin D or cyclohexmide.
However, it has been noted in the past (Hans Selye, 1942) and only recently verified that steroids also have rapid actions through binding steroid receptors located on the plasma membrane (nongenomic actions), much quicker than the above pathway would allow. Association with DNA in the nucleus and the subsequent gene expression is avoided. The receptor types initiate signalling pathways, similar to the actions of peptide hormones. There are 2 main types of peptide hormone receptors: 7TMD receptors linked to the production of a second messenger, or receptors activating a kinase directly. With steroids, the exact structure of these atypical receptors is not yet defined but possibilities of receptor types include GCPRs, ion channels or enzyme-linked receptors. Many of these nongenomic effects appear to involve second messengers such as cAMP and DAG, kinases including PKC and MAPK, and ion fluxes (mainly calcium). This is clearly incompatible with the genomic model described in the past.
Evidence for nongenomic actions
Evidence for their rapid effects is available for steroids of all clones and for a multitude of species and tissues. Nongenomic actions seem to depend on the type of steroid, cells, tissues or species used. Therefore, different pathways and receptors are involved in this speedy process. Tissues traditionally considered as 'non-targets' for genomic actions are vividly found to be regulated by nongenomic mechanisms. Examples of supposed nongenomic actions include rapid oestrogen and aldosterone effects in vascular smooth muscle, and progesterone in the acrosome reaction of sperm.
Hans Seyle provided criteria for defining nongenomic action. These include: the absence of a nucleus, effects that are not obstructed by transcriptional inhibitors, a short time frame, and actions elicited by steroid analogues that do not enter the cell. Seyle's paper was the first to actually detail the nongenomic phenomenon by observing rapid steroid induced effects, such as anaesthesia. This discovery led to development of steroidal anaesthetics used in human medicine and on animals.
Zhu et al were the first to show that their receptor (progestogens) meets all of the criteria for a steroid receptor, including structural plausibility, tissue specificity, membrane localisation, characteristic steroid binding, signalling cascade activation, hormonal regulation and biological relevance.
The pharmacological agonist and antagonist profiles for genomic and nongenomic actions often differ markedly. Antagonists inhibiting genomic effects are inactive towards the nongenomic effects. However, there are only a few selective inhibitors for nongenomic steroid action known at present.
Antibodies have been used to prove their presence, but this approach is sometimes misleading due to crossreactivity and low abundance of antibodies. Various steroids have been shown to bind membrane proteins, but their characterisation has been limited to determination of their molecular weight or antibody identification.
steroid receptors
Three types of receptors have been recorded.
Classical receptors can either lead to changes in gene expression or have nongenomic actions. Nonclassical receptors merely lead to a direct and rapid physiological response. There is a controversy with regards to the existence and identity of nonclassical receptors, and the question being asked is do they exist, and if so, how do they work?
There is evidence that kinases affect gene transcription through binding to CREB, and that transcription factors are linked to kinase cascades. This phenomenon is called cross-talk. (This is true of both steroid and peptide hormones binding to their respective membrane receptors). In thyroid hormones, following initiation of a signal cascade, other events such as phosphorylation of signal transducers, STATs and p53 take place These events lead to a modulation of gene expression, exerted by a nonclassical pathway. Another example of this is the admission of dexamethasone (a synthetic glucocorticoid), which rapidly stabilises lysosomal membranes within 10 minutes, yet a similar effect is detected 24 hours later, indicating a dual action through classical and nonclassical receptors. A multi-step model for steroid action is an extension of the cross-talk model. It explains and inter-relates these 2 mechanisms.
As mentioned above, in some cases, classical nuclear receptors have been shown to drive nongenomic actions. For example, the classical PR has been shown recently to interact with signalling components. The interaction of PR with Src (a SH3-containing protein) activates Ras and/or Raf1 and ERK/MAPK cascades, which then influence the activity of TFs in the nucleus.
Furthermore, there is evidence of receptor-free direct nongenomic action as can be seen with enzymes, for example in the rapid effect of thyroid hormones on mitochondrial respiration.
Steroid groups and evidence for receptor initiated signalling cascades
Conclusion
The importance of nongenomic steroid actions has been underestimated, regardless of early reports. Rapid nongenomic responses are essential when presented with similarly rapid stimuli. These effects have been observed in humans and many other vertebrate and are therapeutically relevant.
However, many studies suffer from technical limitations and quite simply prove the existence of membrane receptors. Further experimentation, such as classical-receptor knockout animals, is required to rule out the possibility of classical receptors being involved and identify nonclassical receptor proteins, if such a thing exists. Clinical trials and animal models are needed. However, with human trials, it is not possible to induce genomic inhibitors. Incorporation of specific 'non-classical' receptors will help identify targets for therapeutic intervention.
