Address correspondence and reprint requests to Dr. Esther L. Sabban, Department of Biochemistry and Molecular Biology, New York Medical College, Valhalla, NY 10595, USA. E-mail: firstname.lastname@example.org
The corticotropin-releasing hormone (CRH) family regulates the endocrine stress response. Here, we examined the effect of immobilization stress (IMO) on gene expression of adrenomedullary CRH family members. Urocortin 2 (Ucn2) has the highest basal gene expression and is increased by > 30-fold in response to single IMO and about 10-fold after six daily repeated IMO. IMO also induced a smaller rise in CRH (six-fold) and CRH receptor type 1 (CRHR1; two-fold with single IMO). The influence of glucocorticoids was examined. Dexamethasone (DEX) or corticosterone greatly increased Ucn2 mRNA levels in PC12 cells in a dose-dependent manner. The DEX elicited rise in Ucn2 was abolished by actinomycin D pre-treatment, indicating a transcriptionally mediated response. DEX also triggered a rise in CRHR1 and lowered CRH mRNA levels. In CRH-knockout mice, where the IMO-induced rise in corticosterone was attenuated, the response of IMO on Ucn2, as well as CRHR2 mRNAs was absent. Overall, the results suggest that the stress-triggered rise in glucocorticoids is involved in the large induction of Ucn2 mRNA levels by IMO, which may allow Ucn2 to act in an autocrine/paracrine fashion to modulate adrenomedullary function, or act as an endocrine hormone.
Urocortin 2 (Ucn2), also known as stresscopin-related peptide, was discovered in 2001 (Hsu and Hsueh 2001; Reyes et al. 2001) and is the most recent addition to the corticotropin-releasing hormone (CRH) family of peptides. Ucn2 is a 38 amino acid active peptide, whose mRNA is expressed in brain areas such as the hypothalamus, cerebellum, and locus coeruleus, as well as in a number of peripheral tissues with high levels in skin, skeletal muscles, lung, stomach, ovary, and adrenal gland and lower levels in heart, spleen, thymus, and kidney (Reyes et al. 2001; Chen et al. 2004; Yamauchi et al. 2005). Other CRH family members include CRH, urocortin 1 (Ucn1), and urocortin 3 (Ucn3). All CRH peptides share moderate amino acid sequence identity, but Ucn1 is most closely related to CRH. On the other hand, Ucn2 and Ucn3 resemble each other more than CRH. These structural differences are reflected in distinct binding affinities to respective receptors and also in different physiological effects [reviewed in (Fekete and Zorrilla 2007)].
The biological effects of CRH and urocortins are mediated by two distinct receptors, CRH receptor type 1 (CRHR1) and type 2 (CRHR2), which belong to the G protein–coupled receptor superfamily (Chen et al. 1993; Perrin et al. 1993; Vita et al. 1993; Dautzenberg and Hauger 2002). These receptors are encoded by two separate genes and significantly differ in pharmacological profile, tissue distribution and function, although they display 70% homology in amino acid sequence. CRH has a very high affinity for CRHR1. Ucn1 binds with the same affinity to both receptors, and Ucn2 and Ucn3 have high affinity to CRHR2 (Tu et al. 2007). Although Ucn3 is essentially a selective CRHR2 agonist, Ucn2 at high concentrations may also bind to CRHR1 (Dautzenberg and Hauger 2002; Fekete and Zorrilla 2007).
The role of CRH in the hypothalamic–pituitary–adrenal (HPA) axis stress response is well documented (Dallman 1993; Aguilera 1998; Bale and Vale 2004; McEwen 2007). Hypothalamic CRH triggers the release of pituitary ACTH to stimulate a rise in circulating glucocorticoids (corticosterone in rodents and cortisol in humans). Activation of the HPA axis, together with increased levels of plasma norepinephrine (NE) and epinephrine (EPI) as a result of activation of the sympatho-adrenomedullary system, are the key regulators of the organism's stress response (Kvetnansky et al. 2009). However, these systems are interconnected at a number of levels. For example, synthesis of EPI in the adrenal medulla is primarily regulated by the HPA axis, through glucocorticoids (Jeong et al. 2000; Tai et al. 2002; Kvetnansky et al. 2006).
Besides the central role of CRH in the HPA axis, there are also local (tissue specific) CRH systems. Expression of CRH, Ucn1, Ucn2, and their receptors CRHR1 and CRHR2 was detected in rat and human adrenal medulla and cortex (Dermitzaki et al. 2007; Tsatsanis et al. 2007). There is evidence that Ucn2 can be involved in regulation of adrenomedullary catecholamine synthesis and release (Nemoto et al. 2005; Dermitzaki et al. 2007; Gu et al. 2010). Our microarray data from the adrenal medulla of rats exposed to immobilization stress revealed that Ucn2 is one of the most markedly increased gene transcripts (Liu et al. 2008). However, the mechanism that leads to its induction as well as the role of Ucn2 in the stress response remains to be elucidated.
In this study, we examined the effect of immobilization stress on gene expression of rat adrenomedullary Ucn2 and other CRH family members. To investigate a requirement for an intact HPA axis and possible involvement of the stress-triggered rise in glucocorticoids, we used CRH-knockout (KO) animals, which display an attenuated rise in plasma corticosterone in response to stress. Moreover, the response to glucocorticoids of the CRH family members was examined in PC12 cells. Our data show that stress induces a robust increase in Ucn2 gene expression in the adrenal medulla, which may be mediated by a hypothalamic CRH-elicited rise in glucocorticoid levels.
Materials and methods
Male, Sprague-Dawley rats (250–320 g) were obtained from Taconic Farms (Germantown, NY, USA). Male, corticotropin-releasing hormone knockout (CRH-KO) mice (C57B1/129SV) and wild-type (WT) mice (20–25 g) were bred at the Institute of Experimental Endocrinology Slovak Academy of Sciences, Bratislava, Slovakia. The CRH-KO mouse line was originally a generous gift from Dr. Joseph A. Majzoub (Harvard Medical School, Department of Endocrinology, Boston, MA, USA) (Muglia et al. 1995). Animals were maintained under controlled conditions (23 ± 2°C, 12-h light/dark cycle, lights on from 6 AM) with food and water ad libitum.
All animal experiments were performed in accordance with the Guide for the Care and Use of Laboratory Animals and approved by the NYMC Institutional Animal Care and Use Committee (rat experiments) or by the Ethical Committee of the Institute of Experimental Endocrinology SAS, Bratislava, Slovakia (mouse experiments).
Stress procedure and tissue collection
Immobilization stress (IMO) was performed as in previous experiments (Sabban et al. 2006a; Tillinger et al. 2010) and as originally described by Kvetnansky and Mikulaj (Kvetnansky and Mikulaj 1970). For single IMO (1xIMO), rats were immobilized once for 2 h. For repeated IMO, they were immobilized 2 h daily for six consecutive days (6xIMO). The IMO was performed at the same time of the day (between 8 AM and noon). Rats were killed immediately or 3 h after termination of IMO. Immobilization of WT and CRH-KO mice was performed the same way as in rats, except that the size of immobilization board was adjusted for use with mice. Mice were killed immediately after 1xIMO. Control animals were not exposed to stress and were killed immediately after removal from their home cage.
The left and right adrenals were dissected from the animals. Subsequently, any cortex tissue adhering to the adrenal medulla was carefully removed. The left and right adrenal medullae from each individual animal were frozen separately in liquid nitrogen, and kept at −80°C. Trunk blood from mice was collected into heparinized tubes and centrifuged at 10 000 g for 20 min. Plasma was removed and stored at −80°C.
PC12 cell cultures
Rat adrenomedullary-derived PC12 cells were maintained in Dulbecco's modified Eagle's medium (DMEM) (Invitrogen, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (Gemini Bio-Products, West Sacramento, CA, USA), 5% horse serum (Gemini Bio-Products), and 50 μg/mL of streptomycin with 50 IU/mL penicillin (0.5%; Invitrogen) in a 37°C humidified incubator with 5% CO2, as described previously (Serova et al. 1997).
For experiments with dexamethasone or corticosterone, one day prior to treatment, the media was replaced with stripped media containing DMEM supplemented with 10% charcoal stripped FBS (Sigma-Aldrich, St. Louis, MO, USA), dialyzed horse serum (Gemini Bio-Products), and antibiotics. Cells were treated with 1–1000 nM dexamethasone (Sigma-Aldrich) or corticosterone (Sigma-Aldrich) in 0.01% ethanol or vehicle for 3–24 h and subsequently harvested for RNA isolation. To determine the role of transcription, 4 μM actinomycin D (transcription inhibitor; Enzo Life Sciences, Farmingdale, NY, USA) dissolved in 0.04% DMSO, or vehicle was added to the cells immediately prior to 8 h treatment with 1 μM dexamethasone.
RNA isolation and Real-Time PCR
Total RNA was isolated using the RNeasy Mini kit (Qiagen, Valencia, CA, USA) and concentrations were quantified using the NanoDrop 2000 (Thermo Fisher Scientific, Waltham, MA, USA). Reverse transcription of RNA (1 μg from rat adrenal medulla or PC12 cells and 600 ng from mouse adrenal medulla) was performed with the RevertAid First Strand cDNA Synthesis kit (Thermo Fisher Scientific) according to the manufacturer's protocol, using an oligo dT primer. For quantitative Real-Time PCR, 30 ng of cDNA product was mixed with 12.5 μL of FastStart Universal SYBR Green Master Rox (Roche Diagnostics, Indianapolis, IN, USA) and 1 μL of the following primer pair sets: rat or mouse Ucn1, Ucn2, Ucn3, CRH, CRHR1, CRHR2, and GAPDH (RT2 qPCR Primer Assay from Qiagen), according to the manufacturer's protocol and analyzed on an ABI7900HT Real-Time PCR instrument (Applied Biosystems, Carlsbad, CA, USA). Data are normalized to GAPDH levels and expressed as the relative fold change, calculated using the ΔΔCt method (Livak and Schmittgen 2001).
Measurement of plasma corticosterone
Concentration of plasma corticosterone in WT and CRH-KO mice was measured by RIA kit as previously described (Newsome et al. 1972; Etches 1976).
All data are presented as mean ± SEM, unless otherwise noted, with n = 4–8 per group for animal experiments and n = 4 per group for cell culture experiments. Differences were analyzed by Student's t-test (if only two groups) or analysis of variance (anova) followed by Bonferroni's post hoc analysis (if more than two groups), with F ratios given where applicable, using GraphPad Prism 4 software (GraphPad Software, Inc., La Jolla, CA, USA). A value of p ≤ 0.05 was considered significant.
Effect of single and repeated immobilization stress on Ucn2, CRH, CRHR1, and CRHR2 gene expression in adrenal medulla
First, we estimated basal mRNA levels of CRH peptides and their receptors in rat adrenal medulla by quantitative RT-PCR. There were several orders of magnitude differences in basal mRNA levels for the various CRH family peptides (Table 1). Ucn2 mRNA levels are about 200 times higher than that of CRH, which is the second most abundant CRH family member. Basal Ucn1 mRNA was at the border of detection and thus its quantification is less precise. We were unable to detect gene expression of Ucn3 in rat adrenal medulla. Basal mRNA levels of CRHR2 were about three-fold higher than CRHR1 and in the range of CRH.
Table 1. Relative mRNA levels of CRH family peptides and their receptors in rat adrenal medulla
In rats exposed to single IMO there was an enormous increase in Ucn2 gene expression. As compared to unstressed control animals, Ucn2 mRNA levels were 18.4-fold higher immediately after a 2 h IMO (not shown) and more than 30-fold higher three hours later (Fig. 1), a time when the stress-elicited rise in adrenal catecholamine biosynthetic enzymes mRNA levels peak (Nankova et al. 1994; Sabban and Kvetnansky 2001). CRH and CRHR1 mRNA levels were also significantly increased, although to a lesser extent than Ucn2. No significant changes were detected in CRHR2 mRNA levels.
With repeated daily IMO for six consecutive days the mRNA for Ucn2 was up-regulated, but considerably less than compared to 1xIMO. CRH mRNA levels were increased following 6xIMO to a similar extent as with 1xIMO. There were no changes in CRHR1 and CRHR2 gene expression with repeated IMO.
Role of glucocorticoids in stress-mediated changes in Ucn2, CRH, CRHR1, and CRHR2 gene expression
Since Ucn2 gene expression was enormously induced by IMO, we first determined whether the stress-triggered rise in glucocorticoids may mediate this response. As shown in Fig. 2a, treatment of rat adrenomedullary-derived PC12 cells with corticosterone elicited a concentration-dependent increase in Ucn2 mRNA levels (F = 122.9, p < 0.0001). The synthetic glucocorticoid dexamethasone (DEX) also elicited a significant concentration-dependent rise in Ucn2 mRNA levels (F = 101.45, p ≤ 0.001) (Fig. 2b). The changes with DEX were more pronounced than with corticosterone and reached values at the highest DEX concentration resembling the magnitude of the effect of IMO on the rat adrenal medulla in vivo. The effects of 1 μM DEX on Ucn2 gene expression are time dependent (F = 11.16, p ≤ 0.001) (Fig. 2c). A greater than 20-fold increase in Ucn2 gene expression was observed as early as 3 h (earliest time point tested) and as long as 24 h (longest time point tested) after treatment.
To examine if the effect of DEX is transcriptionally mediated, PC12 cells were treated with 1 μM DEX in the presence of 4 μM of the transcriptional inhibitor actinomycin D or vehicle. DEX did not alter Ucn2 mRNA levels in the actinomycin D pre-treated cells (Fig. 2d), indicating a transcriptional mechanism for the response to DEX.
DEX treatment also significantly increased CRHR1 gene expression in PC12 cells after 10 and 24 h (Fig. 3), but the increase was much lower than for Ucn2 (Fig. 2c). In contrast, CRH mRNA levels were significantly decreased by DEX treatment in PC12 cells at all times examined, whereas CRHR2 gene expression was decreased after 10 h treatment but unchanged at the other time points (Fig. 3).
To explore the involvement of glucocorticoids in the stress-elicited induction of adrenomedullary Ucn2 gene expression in vivo, we used CRH deficient (CRH-KO) mice. These mice are unable to produce CRH, have lower basal plasma corticosterone levels, and have a greatly attenuated increase in corticosterone levels in response to IMO (Fig. 4a) (Muglia et al. 1995, 2000; Kvetnansky et al. 2006).
Basal mRNA levels of Ucn2 were similar between the wild-type (WT) and CRH-KO mice. Single exposure to IMO led to an increase in Ucn2 mRNA in WT mice as in rats. However, this induction of Ucn2 in the adrenal medulla of mouse was not as robust as observed in rat, most probably because of interspecies differences. In contrast to WT mice, Ucn2 mRNA levels were not altered in CRH-KO mice (Fig. 4b).
The basal and stress-triggered expression of CRHR2, the major receptor for Ucn2, was also examined in CRH-KO animals. The CRH-KO mice displayed significantly higher basal levels of CRHR2 mRNA than the WT animals (Fig. 4c). Exposure to a single episode of IMO reduced CRHR2 mRNA in the adrenal medulla of the wild type, but not in the CRH-KO animals.
The results of this study demonstrate the effect of stress on a number of components of the CRH family in rat adrenal medulla. Basal levels of their mRNAs were very distinct, with expression of Ucn2 several orders of magnitude higher than CRH, Ucn1, and CRH receptors. Ucn3 mRNA was undetectable in the rat adrenal medulla. Furthermore, we found that a single episode of immobilization stress triggers a dramatic increase in Ucn2 gene expression. With repeated stress Ucn2 mRNA levels were also elevated. Single and repeated IMO also led to an increase in CRH mRNA levels. CRHR1 gene expression was up-regulated only after single IMO. Stress had no significant effect on CRHR2 mRNA levels in rats 3 h after either 1x or 6xIMO, however a modest, but significant reduction, was observed immediately following 1xIMO in WT mice. The results from glucocorticoid-treated PC12 cells and CRH-KO mice point to the likely importance of the HPA axis and the stress-induced rise in glucocorticoids for increased Ucn2 gene expression, and perhaps also for other components of the CRH system.
Relative gene expression of CRH family members in rat adrenal medulla
Although other studies (Fukuda et al. 2005; Dermitzaki et al. 2007) have also observed the expression of various CRH family members (Ucn2, CRH, Ucn1, CRHR1, and CHRR2) in the adrenal medulla, consistent with a local CRH system, this is the first study, to our knowledge, to indicate that gene expression of Ucn2 is the most abundant of all CRH members. We also showed the ratio of mRNA abundance of individual CRH family members. Basal Ucn2 mRNA levels are about 200-fold higher than that of CRH, while Ucn1 mRNA is on the border of detection and Ucn3 mRNA is essentially absent in rat adrenal medulla. An absence of Ucn3 mRNA has also been demonstrated by in situ hybridization in human adrenal medulla (Fukuda et al. 2005). In this regard, studies in knockout animals showed a limited effect of Ucn3 deficiency, as compared to Ucn2 deficiency, on adrenal function and growth (Riester et al. 2012).
In contrast to our study, Dermitzaki et al. (2007), using immunohistochemical methods, reported similar levels of CRH, Ucn1, Ucn2, and their receptors CRHR1 and CRHR2 throughout the rat adrenal medulla. However, this is based solely on the specificity of the antibodies used. Unfortunately, to our knowledge, there are no suitable commercially available antibodies for rat Ucn2 for western blot analysis.
Stress-induced changes in gene expression of adrenomedullary CRH family members and potential functional implications
Not only is the mRNA for Ucn2 the most abundant of the CRH family members under basal conditions but it also displayed the greatest increase in mRNA levels (greater than 30-fold) with a single exposure to IMO, highlighting its likely functional significance perhaps on the local catecholaminergic system. We hypothesize that the robust IMO-induced increase in Ucn2 gene expression in rat adrenal medulla may have an autocrine/paracrine regulatory role on catecholamine secretion and biosynthesis with stress conditions. It was previously shown that Ucn2 can modulate catecholamine biosynthesis and secretion. Treatment of PC12 cells with Ucn2 stimulates NE secretion, induces phosphorylation of tyrosine hydroxylase (TH) (Nemoto et al. 2005), and elevates TH mRNA and protein levels (Dermitzaki et al. 2007). Furthermore, in Ucn2 KO mice mRNA levels of adrenomedullary TH and phenylethanolamine N-methyltransferase are significantly decreased compared to WT mice (Riester et al. 2012). However, depending on the dose and duration of treatment, Ucn2 can also act as an inhibitory factor on the catecholaminergic system (Dermitzaki et al. 2007). In addition to an autorcrine/paracrine role in the adrenal medulla, secretion of adrenomedullary-derived Ucn2 into the bloodstream may also have an endocrine function. In this regard, peripheral administration of Ucn2 has a beneficial effect on cardiac function in animal experiments and pilot human studies [reviewed in (Emeto et al. 2011)]. Since Ucn2 has a vasodilatory effect (Kageyama et al. 2003) it may reverse the vasoconstrictive effect of catecholamines released during stress and help the organism restore homeostasis.
In addition to the large IMO-triggered rise in Ucn2, our study demonstrated a significant (about six-fold) rise in adrenomedullary CRH mRNA levels with single and repeated IMO though the significance of these changes may be more modest given the low basal CRH mRNA levels and less profound induction by stress compared to Ucn2. The rise in adrenal CRH gene expression in the adrenal medulla could also influence catecholamine synthesis and release. Activation of CRHR1 in rat chromaffin cells triggered catecholamine secretion and elevated TH mRNA and protein levels (Nanmoku et al. 2005).
Role of glucocorticoids in transcriptional regulation of Ucn2 under stress conditions
HPA axis activation with subsequent elevation of plasma levels of ACTH and corticosterone is a key response of the organism to stress. Our study revealed that treatment of PC12 cells with the synthetic glucocorticoid receptor agonist, DEX, or the endogenous rodent adrenal glucocorticoid, corticosterone, induces a dose-dependent elevation of Ucn2 mRNA. This differs from the negative feedback effects of glucocorticoids on the HPA axis, where DEX lowered Ucn2 mRNA expression levels in cells of the anterior pituitary (Nemoto et al. 2007). The DEX triggered elevation of Ucn2 mRNA levels in PC12 cells is transcriptionally mediated as it was inhibited by actinomycin D. Previously, mouse Ucn2 gene promoter activity was found to be stimulated by DEX and to contain 14 putative glucocorticoid responsive elements (GRE) within the 1.2 kb 5′ flanking region (Chen et al. 2003). Using TESS (Transcription Element Search Software; http://www.cbil.upenn.edu/tess), we found five putative GRE sites within the proximal 1.2 kb 5′ flanking region of the rat Ucn2 gene.
The involvement of glucocorticoids in the IMO-elicited induction of Ucn2 gene expression in the adrenal medulla is supported by results with CRH-KO animals. Mice lacking the CRH gene have low basal corticosterone levels and an impaired glucocorticoid response to various stressors (Muglia et al. 1995, 2000). CRH-KO mice exposed to IMO have an attenuated rise in plasma corticosterone and epinephrine (Kvetnansky et al. 2006), as also shown here for corticosterone. Basal adrenomedullary Ucn2 mRNA levels were similar in both CRH-KO and WT mice. However, the IMO-induced increase in Ucn2 gene expression observed in WT mice was absent in CRH-KO mice, suggesting the involvement of glucocorticoids in induction of Ucn2 gene expression with stress. However, we cannot rule out other CRH requiring mechanisms. In addition to the hypothalamic CRH responsible for the release of ACTH from the pituitary, there are widely dispersed CRH systems in other areas of the brain, such as the brainstem, septum, and amygdala which can mediate physiological responses to stress (Heinrichs and Koob 2004; Orozco-Cabal et al. 2006).
With repeated IMO the rise in glucocorticoids is less sustained after termination of the stress than with the first exposure (Tai et al. 2007), and may explain our findings of smaller rise in Ucn2 mRNA 3 h after 6xIMO compared to 1xIMO (Fig. 1).
In contrast to Ucn2, CRH mRNA levels were significantly decreased in PC12 cells after DEX treatment, although DEX led to a delayed elevation of immunoreactive CRH in a human pheochromocytoma cell line (Venihaki et al. 1998). In rat hypothalamic paraventricular nucleus (PVN) neurons in organotypic culture DEX or corticosterone also markedly reduced CRH mRNA levels (Bali et al. 2008). Nevertheless, we found that CRH mRNA was significantly elevated by IMO. Together this suggests that the IMO-induced elevation in CRH gene expression occurs by a non-glucocorticoid-mediated pathway. This may occur perhaps by way of IMO-triggered activation of CREB in the adrenal medulla (Sabban et al. 2006b), a pathway which mediates stress-triggered induction of CRH gene expression in the PVN (Liu et al. 2010; Aguilera and Liu 2012).
IMO not only elevated CRH and Ucn2 mRNA levels in rat adrenal medulla but also CRHR1 mRNA levels, albeit to a lesser extent of about two-fold. A similar increase was also observed in DEX treated PC12 cells. On the other hand, there was no significant change in CRHR2 gene expression in rat adrenal medulla 3 h after 1x or 6xIMO. However, immediately following 1xIMO, CRHR2 mRNA levels were modestly down-regulated in WT mice, but not in CRH-KO mice. Accordingly, treatment of PC12 cells with DEX decreased CRHR2 gene expression at one of the time points and basal CRHR2 mRNA levels were elevated in unstressed, control CRH-KO mice relative to WT mice. This suggests that IMO triggers a small, transient, glucocorticoid-mediated decrease in CRHR2 gene expression and that furthermore, glucocorticoids may also be important for low basal expression of the CRHR2 gene. In this regard, glucocorticoids can down-regulate CRHR2 in rat heart and aortic smooth muscle (Kageyama et al. 2000).
Overall, our results show that Ucn2 is the most abundant CRH family member mRNA transcript in rat adrenal medulla and that its gene expression increases dramatically in response to stress. The elevated Ucn2, by mechanism which appears to involve stress-triggered rise in glucocorticoids, may be a key mediator of not only adrenomedullary function in response to stress, but may also have an endocrine function.
We gratefully acknowledge support of Grant 10GRNT442001 from American Heart Association (ELS) and Grant APVV-0088-10 and 0148-06 and VEGA (2/0188/09, 2/0036/11) from Slovak Research and Development Agency (RK).