SUMMARY

In response to stress, the hypothalamus-pituitary-adrenal (HPA) axis is activated, which ultimately results in the release of corticosterone from the adrenal glands. Corticosterone can pass the blood-brain barrier and bind to intracellular receptors in the brain. The high affinity mineralocorticoid receptor (MR) and the lower affinity glucocorticoid receptor (GR) affect gene transcription via either transrepression, i.e. binding of monomeric corticosteroid receptors to transcription factors, or transactivation, i.e. binding of receptor dimers to glucocorticoid response elements on the DNA.

Both the MR and the GR are abundantly expressed in the hippocampus, a brain region involved in memory function. The hippocampus consists of several subfields, of which the CA1 region and the dentate gyrus (DG) express both corticosteroid receptors. As a result, many properties of principal cells in these regions are affected by corticosteroids and stress. In general, exposure to stress is thought to promote adaptation to the environment, while repeated or chronic stress can lead to maladaptation and increased susceptibility to various diseases, including depression.

Among other things, voltage-dependent calcium currents are affected by differential corticosteroid receptor occupation. Activation of the GR in addition to the MR, e.g. in response to acute stress exposure, results in a dramatic increase of mainly the L-type calcium current in the CA1 area. This effect depends on GR transactivation, but it is presently not clear which genes are responsible for the physiological effect. In the DG, the effect of a high level of corticosterone on voltage-dependent calcium currents had not been studied at the start of this thesis.

A second property affected by differential occupation of corticosteroid receptors in hippocampal neurons is the serotonin (5-HT) 1A receptor-mediated K + current. Again, in the CA1 area, acute stress or corticosterone application was found to increase serotonergic responses. Also in this case, transactivation via the GR is responsible for the altered physiology. However, no changes in mRNA expression of the 5-HT 1A receptor were found that could explain the alterations in serotonergic currents in the CA1 area. Long-term exposure to corticosterone, on the other hand, leads to attenuation of 5-HT 1A receptor-mediated responses. In accordance, a downregulation of 5-HT 1A receptor mRNA expression has been found in some studies, although several others found no difference.

Acute as well as chronic stress or corticosterone application has been shown to affect gene expression and physiological properties of the hippocampus. To date, however, no direct link has been found between changes in cellular excitability and corticosteroid-responsive genes (or vice versa). In this thesis, we tried to answer the question whether and how transcriptional regulation induced by stress and corticosteroids can be linked to functional changes in principal cells of the rodent hippocampus.

First, we searched for possible transcriptional changes that could underlie known physiological alterations in response to corticosterone or stress. In chapter 2 , we tried to explain the increase in 5-HT 1A receptor-mediated responses found in the CA1 area after acute corticosterone application by examining hippocampal mRNA expression of two candidate genes: regulator of G-protein signaling 4 (RGS4), which can inhibit G-protein mediated signaling, and serum- and glucocorticoid regulated kinase 1 (SGK1), which stimulates surface expression of G-protein coupled K + channels. Thus, we hypothesized that either decreased RGS4 or enhanced SGK1 mRNA expression might be found in the CA1 area one hour after a single corticosterone injection, which could account for the increased 5-HT 1A receptor-mediated responses found previously. In situ hybridization revealed that mRNA expression of RGS4 and SGK1 was not affected by an injection with a high level of corticosterone in any of the hippocampal subfields (CA1, CA3, and DG). The corticosterone injection did lead to strong upregulation of SGK1 mRNA expression in the corpus callosum as well as decreased MR expression in the dentate gyrus, indicating that a single corticosterone injection can indeed affect transcript levels in the brain within 1 hour. However, we concluded that the increase in 5-HT 1A receptor-mediated K + currents in the CA1 area after corticosterone application is not due to alterations in mRNA expression of RGS4 or SGK1.

In chapter 3 , we studied possible transcriptional changes underlying the attenuated 5-HT 1A receptor-mediated responses found after chronic application of high corticosterone levels. Using in situ hybridization, we studied RGS4 and SGK1 mRNA expression, as well as the transcript level of neural cell adhesion molecule (NCAM) in the hippocampus of rats injected with corticosterone for 21 consecutive days. NCAM reduces cell surface expression of the 5-HT 1A receptor-coupled K + channel, indicating that enhanced NCAM expression might be responsible for the attenuated serotonergic responses found in the CA1 area after long-term corticosterone treatment. Although the 21-day injection paradigm clearly resulted in enhanced plasma corticosterone levels - even one day after the last injection -, hippocampal mRNA expression of RGS4, SGK1, nor NCAM was affected. Thus, these genes are not likely to be involved in the physiological changes found in the hippocampal CA1 area in response to chronic corticosterone injections.

Using the reverse approach, we next studied the physiological effects of earlier observed alterations in gene expression. In chapter 4 , we tested whether the previously observed alterations in voltage-dependent calcium channel (VDCC) subunit expression after chronic stress in the DG result in similar physiological changes in voltage-dependent calcium currents. We thus subjected rats to a 21-day chronic unpredictable stress paradigm, and recorded whole cell calcium currents in hippocampal slices after acute in vitro application of corticosterone or vehicle solution. In addition, half of the control and half of the chronically stressed animals were treated with the GR antagonist mifepristone during the last 4 days of the stress protocol, to test whether this treatment would reverse the putative stress-induced effects. We found that chronic stress in itself did not result in altered voltage-dependent calcium currents recorded in the DG one day after the last stressor. However, calcium current amplitude was significantly enhanced after chronic stress (compared to control) when slices had been treated with a high dose of corticosterone 1-4 hours earlier. In animals treated with mifepristone for the last 4 days of the stress paradigm, this effect was no longer seen. Interestingly, our findings on voltage-dependent calcium currents were in line with the previous results on VDCC a 1C mRNA expression after chronic stress and acute corticosterone incubation.

Finally, in chapter 5 , we studied corticosterone-induced alterations in mRNA levels, protein levels, and physiological responses in parallel in the hippocampal subfields CA1 and DG. We first studied voltage-dependent calcium currents in both subfields 1-4 hours after in vitro corticosterone application. The previously reported corticosterone-induced increase in calcium current amplitude in CA1 pyramidal cells was replicated. In the DG, however, incubation with 100 nM corticosterone did not affect voltage-dependent calcium currents. We then used in situ hybridization to study mRNA expression of VDCC subunits in both regions after a single corticosterone injection, and found that mRNA expression of the b 4 subunit was enhanced in corticosterone-injected versus naïve animals, while a 1D mRNA expression was only upregulated in corticosterone- versus vehicle-injected animals. Importantly, the observed corticosterone-induced differences were found in all hippocampal areas and could thus not explain the differences in physiological effects between the CA1 area and the DG. Finally, we studied VDCC subunit b 4 and a 1C protein expression in the hippocampus 2-3 hours after corticosterone incubation using the Western blot technique. We found that corticosterone incubation led to enhanced b 4 as well as a 1C protein expression in the CA1 area, but not in the DG, thus providing a possible explanation for the differential corticosterone-induced effect on calcium currents. The results from this study thus suggest that corticosterone-induced changes at the posttranscriptional level are essential for the differential effect on voltage-dependent calcium currents between both hippocampal subfields.

In conclusion, the research described in this thesis shows that corticosterone-induced changes in physiological parameters can often not be directly linked to changes in mRNA expression of obvious candidate genes. At least in some cases, protein levels seem to correlate better with the physiological outcome than the level of mRNA expression. We thus propose the possibility that corticosterone influences the expression of proteins involved in postranscriptional processes as well, thereby influencing many physiological parameters simultaneously. Possible mechanisms underlying these putative posttranscriptional changes, as well as the experimental design of the research described in this thesis are discussed extensively in chapter 6 .