Serotonergic systems in the balance: CRHR1 and CRHR2 differentially control stress-induced serotonin synthesis

Abstract

Anxiety and affective disorders are often associated with hypercortisolism and dysfunctional serotonergic systems, including increased expression of TPH2, the gene encoding the rate-limiting enzyme of neuronal serotonin synthesis. We previously reported that chronic glucocorticoid exposure is anxiogenic and increases rat Tph2 mRNA expression, but it was still unclear if this also translates to increased TPH2 protein levels and in vivo activity of the enzyme. Here, we found that adult male rats treated with corticosterone (CORT, 100 μg/ml) via the drinking water for 21 days indeed show increased TPH2 protein expression in the dorsal and ventral part of the dorsal raphe nucleus (DRD, DRV) during the light phase, abolishing the enzyme's diurnal rhythm. In a second study, we systemically blocked the conversion of 5-hydroxytryptophan (5-HTP) to serotonin immediately before rats treated with CORT or vehicle were either exposed to 30 min acoustic startle stress or home cage control conditions. This allowed us to measure 5-HTP accumulation as a direct readout of basal versus stress-induced in vivo TPH2 activity. As expected, basal TPH2 activity was elevated in the DRD, DRV and MnR of CORT-treated rats. In response to stress, a multitude of serotonergic systems reacted with increased TPH2 activity, but the stress-, anxiety-, and learned helplessness-related dorsal and caudal DR (DRD/DRC) displayed stress-induced increases in TPH2 activity only after chronic CORT-treatment. To address the mechanisms underlying this region-specific CORT-dependent sensitization, we stereotaxically implanted CORT-treated rats with cannulae targeting the DR, and pharmacologically blocked either corticotropin-releasing hormone receptor type 1 (CRHR1) or type 2 (CRHR2) 10 min prior to acoustic startle stress. CRHR2 blockade prevented stress-induced increases of TPH2 activity within the DRD/DRC, while blockade of CRHR1 potentiated stress-induced TPH2 activity in the entire DR. Stress-induced TPH2 activity in the DRD/DRC furthermore predicted TPH2 activity in the amygdala and in the caudal pontine reticular nucleus (PnC), while serotonin synthesis in the PnC was strongly correlated with the maximum startle response. Our data demonstrate that chronically elevated glucocorticoids sensitize stress- and anxiety-related serotonergic systems, and for the first time reveal competing roles of CRHR1 and CRHR2 on stress-induced in vivo serotonin synthesis.

Keywords: Corticotropin receptor; Dorsal raphe nucleus; Glucocorticoids; Serotonin; Stress; Tryptophan hydroxylase.

Figure 1. Experimental Design

Timelines of Experiments (A) 1, (B) 2, and (C) 3. (A) Adult, male rats (N = 32) were treated with vehicle (n = 16) or corticosterone (CORT, 100 μg/ml, n = 16) via the drinking water for 21 days to quantify TPH2 protein expression during the light and dark phases of the light cycle. (B) Adult, male rats (N = 32) were treated with vehicle (n = 16) or CORT (n = 16) via the drinking water for 21 days to assess the rats’ emotionality on the elevated plus-maze (EPM) and in the forced swim test (FST), to measure basal and acoustic startle stress (AS)-induced TPH2 activity, and to record treatment effects on plasma CORT concentrations, adrenal weight, and thymus weight. (C) Adult, male rats (N = 60) received stereotaxic implants of a unilateral guide cannula targeting the dorsocaudal parts of the dorsal raphe nucleus (DRD/DRC). After all rats were chronically treated with CORT for 21 days, they received either an intra-DR microinjection of vehicle (n = 20), antalarmin (n = 20) or antisauvagine-30 (A30, n = 20) 10 min prior to 30 min of AS stress or home cage control conditions (3 × 2 design). TPH2 activity was measured in the form of in vivo 5-hydroxytryptophan (5-HTP) accumulation after i.p. injection of 100 mg/kg of the aromatic L-amino acid decarboxylase inhibitor NSD-1015 immediately prior to AS stress.

Figure 2. Effects of chronic CORT on behavior and TPH2 protein expression

(A) Light phase and dark phase TPH2 protein expression in the dorsal (DRD) and ventral part (DRV) of the dorsal raphe nucleus, and in the median raphe nucleus (MnR) following treatment with chronic corticosterone (CORT) or vehicle. Half of each treatment group was sacrificed +10 h into the (active) dark phase at 1600 h, the other half +10 h into the (inactive) light phase at 0400 h. The photomicrographs depict representative, color-inverted images of enhanced chemoluminescence detection of TPH2 protein content in bands from region-specific western blots in comparison to β-actin protein, the loading control. #p < 0.05, ##p < 0.01 light/dark effect within the same treatment group; *p < 0.05 CORT-treatment effect versus vehicle-light group, ⁽*⁾p = 0.063 versus vehicle-dark group, after post hoc analysis with Fisher’s Protected LSD test (Vehicle-light, n = 8; Vehicle-dark, n = 8; CORT-light, n = 8; CORT-dark, n = 8). (B) Diameter and location of microdissection samples at a representative neuroanatomical section adapted from Paxinos and Watson (1998) at bregma level −8.10 mm. (C) Anxiety-related behavioral parameters of rats from Experiment 1 and (D) Experiment 2, as assessed on the elevated plus-maze (EPM) during the rats’ active (dark) phase on day 16 of treatment with either vehicle or CORT. Panels on the left side depict the percent time the rats spent either in the closed arms, the open arms, the neutral area, or grooming. The number of entries into either arm type, and the number of rears at the walls are indicated in the panels on the right side. Active (climbing and swimming) versus passive (floating) stress coping behavior was assessed on day 18 of treatment in the forced swim test (FST), again in both (E) Experiment 1 and (F) Experiment 2. On the right side are the percentages of time the rats spent climbing, swimming, or immobile. The panels on the right side display the difference in rectal temperature (delta temperature, °C) from before and after the FST. Further abbreviations: Aq, cerebral aqueduct; CIC, central nucleus of the inferior colliculus; mlf, medial longitudinal fasciculus. Data are displayed as means + SEM. *p < 0.05, **p < 0.01, versus Vehicle (Vehicle, n = 16; CORT, n = 16).

Figure 3. Effects of chronic CORT and acute stress on TPH2 activity

The bar graphs display the region-specific amount of 5-hydroxytryptophan (5-HTP) in pg/μg total protein, as a measure of the enzymatic activity of TPH2, following a 2 × 2 design of treatment (Vehicle or corticosterone (CORT), via the drinking water for 21 days) and home cage control conditions (‘C’) versus acoustic startle stress (‘S’) on day 21. Displayed are TPH2 activity in (A) each functional subdivision of the dorsal raphe nucleus (DR), including the entire DR, (B) the median raphe nucleus (MnR), (D) the lateral orbital cortex (LO), (E) the infralimbic cortex (IL), (F) the CA1 region of the dorsal hippocampus (dCA1), and (G) the caudal pontine reticular nucleus (PnC). Panels H, I and J display significant correlations of TPH2 activity in the DRD/DRC or the DRD by itself with TPH2 activity in the basolateral amygdala (BL), the central amygdala (CE) and the PnC. Panel C maps the neuroanatomical location and diameter of microdissections within the serotonergic DR and MnR. Further abbreviations: 2, second cerebellar lobule; 4V, fourth ventricle; aca, anterior commissure, anterior part; Aq, cerebral aqueduct; CIC, central nucleus of the inferior colliculus; DRC, dorsal raphe nucleus, caudal part; DRD dorsal raphe nucleus, dorsal part; DRI, dorsal raphe nucleus, interfascicular part; DRV, dorsal raphe nucleus, ventral part; DRVL/VLPAG, dorsal raphe nucleus, ventrolateral part/ventrolateral periaqueductal gray; fmi, forceps minor of the corpus callosum; mlf, medial longitudinal fasciculus; PrL, prelimbic cortex; ts, tectospinal tract; VO, ventral orbital cortex. Data are displayed as means + SEM. #p < 0.05, ##p < 0.01 stress effect within the same treatment group; *p < 0.05, **p < 0.01 treatment effect versus respective vehicle group, after post hoc analysis with Fisher’s Protected LSD test (Vehicle-Control, n = 8; Vehicle-Stress, n = 8; CORT-Control, n = 8; CORT-Stress, n = 8).

Figure 4. Effects of intra-DR microinjections of CRHR1 and CRHR2 antagonists on stress-induced TPH2 activity

Graphs illustrating the amount of 5-hydroxytryptophan (5-HTP) measured in pg/μg total protein in (A) each functional subdivision of the dorsal raphe nucleus (DR) and in (B) the median raphe nucleus (MnR), following a 2 × 3 design of stress (home cage, HC control; versus acoustic startle, AS stress) and intra-DR antagonist microinjection (vehicle, Veh; antalarmin, Ant; and antisauvagine-30, A30). All rats in Experiment 3 were exposed to chronic corticosterone (CORT) via the drinking water for 21 days. Data are displayed as means + SEM. #p < 0.05, ##p < 0.01 stress effect versus the HC control group of the same antagonist microinjection; *p < 0.05, **p < 0.01 antagonist effect versus the AS stress-group that received microinjection of vehicle, after post hoc analysis with Fisher’s Protected LSD test (Veh-HC, n = 8; Veh-AS, n = 7; Ant-HC, n = 8; Ant-AS, n = 7; A30-HC, n = 8; A30-AS, n = 9). Panel C summarizes the verification and mapping of cannulae placements. Unilateral stereotaxic placements of guide and injection cannulae were verified via histochemistry and comparison with a rat brain atlas (Paxinos and Watson, 1998). Depicted are those injection sites between −8.10 mm and −8.70 mm from bregma that were considered within target. (D) Representative photomicrograph of guide and injection cannula tracks and of an injection site itself at −8.40 mm from bregma in a 25 μm-thick brain section stained histochemically with cresyl violet. The vertical scales in panels C and D indicate the dorsoventral coordinates in millimeters. Panels E and F display significant correlations of TPH2 activity in the dorsocaudal DR (DRD/DRC) and TPH2 activity in the basolateral (BL) and central amygdala (CE) under home cage control conditions, and representative BL and CE microdissection sites and diameters at bregma level -3.00 mm. Further abbreviations: 2, second cerebellar lobule; Aq, cerebral aqueduct; DRD, dorsal raphe nucleus, dorsal part; DRV, dorsal raphe nucleus, ventral part; DRVL/VLPAG, dorsal raphe nucleus, ventrolateral part/ventrolateral periaqueductal gray; DRC, dorsal raphe nucleus, caudal part; DRI, dorsal raphe nucleus, interfascicular part; DTgP, dorsal tegmental nucleus, pericentral part; mlf, medial longitudinal fasciculus; scp, superior cerebellar peduncle; xscp, decussation of the superior cerebellar peduncle (Veh-HC, n = 8; Veh-AS, n = 7; Ant-HC, n = 8; Ant-AS, n = 7; A30-HC, n = 8; A30-AS, n = 9).

Figure 5. Effects of chronic CORT and CRHR1 antagonist on the startle response

Acoustic startle (AS) responses as measured during the 30 min exposure to AS stress on day 21 in (A) Experiment 2 and (C) Experiment 3, and correlation of 5-hydroxytryptophan (5-HTP in pg/μg total protein) measured in the caudal pontine reticular nucleus (PnC), and the maximum startle response in Experiments 2 and 3 (panels B and D, respectively). The maximum acoustic startle response is displayed in Newton, N. Further abbreviations: A30, antisauvagine-30; Ant, antalarmin, CORT, corticosterone. Data are displayed as means + SEM. *p < 0.05 versus Veh (Veh, n = 7; Ant, n = 7; A30, n = 9); **p < 0.01 versus Vehicle (Vehicle, n = 8; CORT, n = 8).

Figure 6. Mechanistic model

Hypothetical model of corticotropin-releasing hormone (CRH)-mediated serotonergic control executed from the dorsocaudal subdivisions of the dorsal raphe nucleus (DRD/DRC) towards the anxiety-related basolateral amygdala (BL) and the acoustic startle-controlling caudal pontine reticular nucleus (PnC). Chronic glucocorticoid exposure increases CRH expression in the central amygdala (CE) and in the bed nucleus of the stria terminalis (BNST) (Makino et al., 1994; Shepard et al., 2000), while sustained input from CE-derived CRH neurons drives CRH expression in the BNST, particularly in the (dorso)lateral part (BNST_L) (Shepard et al., 2006; Walker et al., 2009). BNST_L-derived CRH is then released into the DRD/DRC (Sink et al., 2012), acting onto CRH receptor type 2 (CRHR2) to increase, and on CRH receptor type 1 (CRHR1) to restrain, TPH2 activity and thus serotonin synthesis in DRD/DRC serotonergic neurons during an acute stressor. Both CRHR1 and CRHR2 are G protein-coupled receptors that mainly activate Gs. Most CRHR1 is not expressed in serotonergic neurons, but is located in γ-aminobutyric acid (GABA)ergic interneurons (Valentino et al., 2010), likely acting to increase GABAergic inhibition of TPH2 activity. In contrast, CRHR2 has been detected within the soma of serotonergic neurons (Waselus et al., 2009; Lukkes et al., 2011). The DRD/DRC innervates, amongst other forebrain target regions, the BL. In the BL, activation of the excitatory serotonin receptor 5-HT_2C has anxiogenic effects and mediates the exaggerated anxiety in a model of learned helplessness (Christianson et al., 2010). Within the PnC, immunohistochemistry has identified excitatory 5-HT_2C receptors on giant PnC neurons (Weber et al., 2008). Electrophysiological studies furthermore suggest that serotonin increases the excitability of giant PnC neurons via two-pore K⁺ channels, namely the TWIK-related acid-sensitive K⁺ channel type 3 (Weber et al., 2008), indicating how serotonergic projections from the DR to the PnC (Hobson et al., 1986) may potentiate the AS response. During an acute stressor, enhanced translocation of CRHR2 from the cytoplasm to the cell membrane (Waselus et al., 2009) may further potentiate stress-induced serotonin synthesis within DRD/DRC serotonergic neurons projecting to the BL and PnC, activating conflict anxiety-controlling output neurons and enhancing the startle response. Within the DRD/DRC, chronic corticosterone (CORT) exposure increases basal TPH2 expression, and facilitates stress-induced serotonin synthesis by increasing the sensitivity of local, CRHR2-expressing serotonergic neurons in response to an acute stressor, resulting in passive stress-coping behavior, increased conflict anxiety, and learned helplessness (Christianson et al., 2010). Further abbreviations: 4V, fourth ventricle; Aq, cerebral aqueduct; f, fornix; LV, lateral ventricle; mlf, medial longitudinal fasciculus; scp, superior cerebellar peduncle; ts, tectospinal tract.