Main

Neurosteroids are synthesized in the brain, either de novo from cholesterol or from steroid hormone precursors, and accumulate in the nervous system to levels that are at least in part independent of steroidogenic gland secretion (Baulieu 1981). Neurosteroids, such as progesterone (PROG), pregnenolone (PREG), dehydroepiandrosterone (DHEA), and their respective sulfate ester (PREGS or DHEAS), are involved in regulating the imbalance between excitation and inhibition in the CNS (Wu et al. 1991). The neurosteroids, allopregnanolone, allotetrahydrodeoxycorticosterone, PREGS, and DHEAS have been shown to possess antistress, anxiolytic, and antiamnesic properties in experimental animal models (Bitran et al. 1991; Brot et al. 1997; Maurice et al. 1997, 1999; Reddy and Kulkarni 1998; Reddy et al. 1998; Urani et al. 1998; Wieland et al. 1991). Recent evidence suggests that DHEAS and PREGS also play an important role in depression. Interestingly, decreased levels of DHEA, DHEAS, and PREGS have been associated with depression, cognitive dysfunction, aging, and other neurological conditions (Orentreich et al. 1984; Vallée et al. 1997), and DHEA improves the depression score in patients (Wolkowitz et al. 1997). However, the mechanism underlying the beneficial effects of neurosteroids is not yet known.

Neurosteroids have been shown to affect the activity of neurotransmitter systems, which are involved in a variety of neuropsychiatric illnesses. DHEAS and PREGS are negative allosteric modulators of the γ-aminobutyric acid (GABA)A receptors (Majewska 1992), and positive modulators of N-methyl-D-aspartate (NMDA) receptor-mediated responses (Bowlby 1993; Irwin et al. 1992, 1994). Evidence is mounting that DHEAS and PREGS act as agonists at central sigma receptors, especially sigma1 sites, and exert facilitatory actions on NMDA-mediated glutamatergic and noradrenergic neurotransmission (Maurice and Lockhart 1997; Monnet et al. 1995). Some neurosteroids such as PREGS and PROG have been shown to have an affinity for sigma1 receptors (Su et al. 1988; Walker et al. 1990). Further, Maurice et al. (1996) have reported that exogenous administration of neurosteroids leads to dose-dependent inhibition of in vivo binding of [3H]-(+)-N-allylnormetazocine ((+)-SKF-10,047) to sigma1 receptors. Thus, it is possible that neurosteroids are the main endogenous modulator of sigma1 receptors, and that the pharmacological effects of neurosteroids described above are mediated, at least partly, through the sigma1 receptors.

To date, we have tried to clarify the functional role of sigma1 receptors in a stressful situation by using the conditioned fear stress response defined by Fanselow (1980). Rats and mice exhibit a marked suppression of motility when they are re-placed in an environment in which they have previously received an electric footshock (Kameyama and Nagasaka 1982, 1983; Yamada and Nabeshima 1995). This motor suppression is regarded as a conditioned emotional response to the environment associated with previous footshock (Kameyama and Nagasaka 1982, 1983; Yamada and Nabeshima 1995). We have found that (+)-SKF-10,047 and dextromethorphan, prototype sigma1 receptor agonists, dose dependently attenuate the conditioned fear stress (Kamei et al. 1994, 1996a) and that this effect was antagonized by NE-100, a sigma1 receptor antagonist stress (Kamei et al. 1994, 1996a), indicating that the activation of sigma1 receptors is responsible for the attenuation of conditioned fear stress response. These observations suggest that the sigma1 receptors play an important role in stress responses. However, the functional role of neurosteroids having affinity for sigma1 receptors, in the conditioned fear stress response is unclear.

In the present study, therefore, we examined the effects of neurosteroids on the conditioned fear stress response in mice, in comparison with those of the sigma1 receptor ligand, (+)-SKF-10,047 and discussed the relationship between neurosteroids and sigma1 receptors in the stress response.

METHODS

Animals

Male mice of the ddY strain (Japan SLC, Inc., Shizuoka, Japan), at 7–8 weeks of age were used. The animals were housed in plastic cages and were kept in a regulated environment (23 ± 1°C, 50 ± 5% humidity), with a 12/12 hr light-dark cycle (lights on at 7:00 A.M.). Food (CE2; Clea Japan Inc., Tokyo, Japan) and tap water were available ad libitum.

All experiments were performed in accordance with the Guidelines for Animal Experiments of the Nagoya University School of Medicine. The procedures involving animals and their care conformed with the international guidelines set out in ‘Principles of Laboratory Animal Care’ (NIH publication no. 85–23; revised 1985).

Drugs

The following drugs were used: (+)-N-allylnormetazocine hydrochloride ((+)-SKF-10,047; Research Biochemicals Inc., Natick, MA, USA), dehydroepiandrosterone sulfate (DHEAS; Sigma, St. Louis MO, USA), pregnenolone sulfate (PREGS; Sigma), allopregnanolone (Sigma), and progesterone (PROG; Sigma). NE-100 (N, N-dipropyl-2-[4-methoxy-3-(2-phenylethoxy)phenyl]-ethylamine monohydrochloride) was kindly supplied by Taisho Pharmaceutical Co. Ltd. (Saitama, Japan). DHEAS radioimmunoassay kit was obtained from Japan DPC Co. (Chiba, Japan).

(+)-SKF-10,047 and NE-100 were dissolved in saline and distilled water, respectively. DHEAS, PREGS, allopregnanolone, and PROG were solubilized in 7% dimethylsulfoxide (Sigma). All compounds were administered in a volume of 0.1 ml/10 g body weight.

Schedule for Conditioned Fear Stress

The experiments were carried out according to the method of Kamei et al. (1994, 1996a,b), with a minor modification. A transparent acrylic rectangular test cage (25.5 × 31.0 × 14.0 cm) equipped with a metal wire floor was used. The test cage was located in a sound-attenuated room and was illuminated with a 20-W bulb.

Each mouse was placed in the test cage and received electric shocks (0.1 Hz, 200 ms, 1 mA) for 6 min through a shock generator-scrambler (Neuroscience Inc., Tokyo, Japan). The test trial was then carried out 24 h after shock treatment; the animals were again placed into the same test cage, but no electric footshocks were given (shocked group). The spontaneous motility of the animal was determined for 6 min in the same test cage by using Scanet SV-10 (Toyo Sangyo, Toyama, Japan) equipped with photosensors. The non-shocked control group was prepared exactly the same way except without electric footshock (non-shocked group).

All test drugs were administered before measuring motility in the test trial; (+)-SKF-10,047, DHEAS, PREGS, allopregnanolone, and PROG were administered subcutaneously (s.c.) 30 min before the test trial. In the antagonistic experiments, NE-100 and PROG were administered intraperitoneally (i.p.) 45 min before the test trial. Groups receiving no drugs were given an appropriate vehicle.

Determination of DHEAS Concentration in the Serum

Immediately after the measurement of motility, each mouse was decapitated and the blood was rapidly collected. The blood was centrifuged at 15,000g for 5 min and the serum was extracted. The DHEAS concentration in the serum was quantified with a radioimmunoassay kit.

Statistical Analysis

Statistical significance was determined by a Dunn-type non-parametric test. Comparisons of two sample means were performed with the Mann-Whitney U-test and Student t-test. P-values less than .05 were taken to indicate statistically significant differences.

RESULTS

Effects of Neurosteroids and (+)-SKF-10,047 on the Conditioned Fear Stress Response in Mice

The shocked mice treated with vehicle (shocked group) exhibited a marked suppression of motility; the shocked group showed 4.3–10.1% of the motility exhibited by the non-shocked group when returned to the apparatus in which they had been given an electric shock (Figure 1 and Table 1A), in agreement with our previous results (Kamei et al. 1994, 1996a,b). The shocked mice mostly froze on the floor and crouched, while the non-shocked mice showed typical exploratory behavior such as ambulation, sniffing and rearing.

Figure 1
figure 1

Effects of neurosteroids on the conditioned fear stress in mice. Dehydroepiandrosterone sulfate (DHEAS), pregnenolone sulfate (PREGS), progesterone (PROG), and allopregnanolone were administered s.c. 30 min before motility was measured. The results are expressed as the mean ± S.E.M. for the number of animals shown in each column. *p < .05 and **p < .01 compared to the corresponding vehicle-treated, shocked group (Dunn-type non-parametric test)

Full size image
Table 1 Effect of (+)-SKF-10,047 on the Conditioned Fear Stress in Mice+
Full size table

Figure 1 and Table 1A show the effects of neurosteroids and (+)-SKF-10,047, respectively, on motor suppression in the shocked group. DHEAS (25 and 50 mg/kg), PREGS (10–50 mg/kg) and (+)-SKF-10,047 (3 and 6 mg/kg) partially attenuated the motor suppression of the shocked group in a dose-dependent manner, without changing motility in the non-shocked mice (Figure 1 and Table 1A); significant effects of DHEAS, PREGS, and (+)-SKF-10,047 were observed at a dose of 50, 50, and 6 mg/kg, respectively.

Allopregnanolone (5 and 20 mg/kg) and PROG (10–50 mg/kg) had no effects on motility in the shocked and non-shocked groups (Figure 1).

Antagonistic Effects of Sigma1 Receptor Antagonist and PROG on the Attenuation of Conditioned Fear Stress Induced by DHEAS, PREGS, and (+)-SKF-10,047

To clarify whether the attenuating effects of DHEAS and PREGS on motor suppression in the shocked mice were mediated by sigma1 receptors, we investigated the antagonistic effects of a selective sigma1 receptor antagonist, NE-100 and a sigma1 receptor antagonistic neurosteroid, PROG, on the effects of DHEAS and PREGS.

NE-100 and PROG were used at doses that could block the attenuating effect of sigma1 receptor ligands on the conditioned fear stress (Kamei et al. 1996a) and on scopolamine-induced learning impairment (Urani et al. 1998), respectively. As shown in Figures 2 and 3,, NE-100 (5 mg/kg) and PROG (5 mg/kg) significantly inhibited the attenuation of the motor suppression induced by DHEAS (50 mg/kg) and PREGS (50 mg/kg). Similarly, the attenuating effect of (+)-SKF-10,047 (6 mg/kg) on motor suppression was significantly antagonized by NE-100 (5 mg/kg) and PROG (10 mg/kg) (Table 1B and Figure 3).

Figure 2
figure 2

Effect of NE-100 on dehydroepiandrosterone sulfate (DHEAS)- and pregnenolone sulfate (PREGS)-induced attenuation of conditioned fear stress in mice. DHEAS and PREGS were administered s.c. 30 min before motility was measured. NE-100 was administered i.p. 45 min before motility was measured. The results are expressed as the mean ± S.E.M. for the number of animals shown in each column. **p < .01 compared to the corresponding vehicle-treated, shocked group (Mann-Whitney U-test). #p <, .05 compared to the DHEAS- or PREGS-treated, shocked group (Mann-Whitney U-test)

Full size image
Figure 3
figure 3

Effect of progesterone (PROG) on dehydroepiandrosterone sulfate (DHEAS)-, pregnenolone sulfate (PREGS)-, and (+)-SKF-10,047 (SKF)-induced attenuation of conditioned fear stress in mice. DHEAS, PREGS, and SKF were administered s.c. 30 min before motility was measured. PROG was administered i.p. 45 min before motility was measured. The results are expressed as the mean ± S.E.M. for the number of animals shown in each column. *p < .05 and **p < .01 compared to the corresponding vehicle-treated, shocked group (Mann-Whitney U-test). #p < .05 and ##p < .01 compared to the DHEAS-, PREGS-, or SKF-treated, shocked group (Dunn-type non-parametric test)

Full size image

Effects of (+)-SKF-10,047 in Combination with DHEAS on Conditioned Fear Stress

Figure 4 shows the effects of (+)-SKF-10,047 in combination with DHEAS on the motor suppression in the shocked group. Low doses of DHEAS (25 mg/kg) and (+)-SKF-10,047 (3 mg/kg) themselves had no effects on motility in the shocked group. However, when DHEAS was co-administered with (+)-SKF-10,047, motor suppression was significantly attenuated, but not the motility in the non-shocked groups. The shocked mice given (+)-SKF-10,047 in combination with DHEAS exhibited exploratory behavior.

Figure 4
figure 4

Effect of (+)-SKF-10,047 (SKF) in combination with dehydroepiandrosterone sulfate (DHEAS) on conditioned fear stress in mice. SKF and DHEAS were administered s.c. 30 min before motility was measured. The results are expressed as the mean ± S.E.M. for the number of animals shown in each column. *p < .05 compared to the (+)-SKF-10,047 or DHEAS alone-treated, shocked group (Mann-Whitney U-test)

Full size image

The Serum Concentration of DHEAS in Mice Showing the Conditioned Fear Stress Response

The serum concentrations of DHEAS in the non-shocked and shocked groups are shown in Figure 5. In mice showing the conditioned fear stress response, the serum concentration of DHEAS was significantly decreased compared with that in the non-shocked mice. However, the serum DHEAS concentration in mice, which did not place into the same test cage 24 h after shock treatment, was unchanged (data not shown).

Figure 5
figure 5

Dehydroepiandrosterone sulfate (DHEAS) levels in the serum of mice showing conditioned fear stress response. The results are expressed as the mean ± S.E.M. for the number of animals shown in each column. *p < .05 compared to the vehicle-treated, non-shocked group (Student's t-test)

Full size image

DISCUSSION

Sigma1 receptors have been demonstrated to play an important role in conditioned fear stress response (Kamei et al. 1994, 1996a,b, 1997). For example, (+)-SKF-10,047 and dextromethorphan, putative sigma1 receptor ligands attenuate the conditioned fear stress-induced motor suppression in mice, the effects being antagonized by NE-100, a selective sigma1 receptor antagonist (Kamei et al. 1994, 1996a). In the present study, (+)-SKF-10,047 (6 mg/kg), neurosteroids, DHEAS (50 mg/kg) and PREGS (50 mg/kg) could partially attenuate the conditioned fear stress-induced motor suppression in mice. The effects of both neurosteroids and (+)-SKF-10,047 were antagonized by NE-100, a selective sigma1 receptor antagonist. Several steroids including PROG, PREG, PREGS, testosterone, and 17β-estradiol have been shown to inhibit the in vitro binding of the sigma1 receptor radioligands [3H]-SKF-10,047, [3H]-dextromethorphan, [3H]-3-PPP, and [3H]-haloperidol to rat brain, splenocytes, plasma membranes, and liver microsomes (Klein and Musacchio 1994; Mc-Cann and Su 1991; Ross 1991; Schwartz et al. 1989; Su et al. 1988; Yamada et al. 1994). DHEAS and PREGS dose-dependently inhibited the in vivo haloperidol-sensitive [3H]-(+)-SKF-10,047 binding to sigma1 receptor in the mouse hippocampus and cortex (Maurice et al. 1996). Taken together, these findings suggest that the attenuating effects of neurosteroids on the conditioned fear stress are mediated at least partly through the sigma1 receptors. That there is interaction between DHEAS and sigma1 receptors is further supported by the finding that the combination of DHEAS with (+)-SKF-10,047 had an additive effect to attenuate the motor suppression in the conditioned fear stress.

Recent evidence has suggested that sigma receptors are divided into at least two subtypes, sigma1 and sigma2 receptors (Walker et al. 1990; Quirion et al. 1992). Additionally, binding studies with an anticonvulsant, phenytoin, have shown that sigma1 receptors can be differentiated into two different binding sites, namely phenytoin-sensitive and phenytoin-insensitive sites (DeHaven-Hudkins et al. 1993). The binding affinity of (+)-SKF-10,047 and dextromethorphan to sigma1 sites is markedly increased by phenytoin while that of (+)-pentazocine (a sigma1 receptor ligand) and DTG (a sigma1/sigma2 receptor ligand) is unaffected by phenytoin (DeHaven-Hudkins et al. 1993). The activation of phenytoin-regulated type sigma1 receptors is involved in the attenuation of (+)-SKF-10,047 and dextromethorphan on conditioned fear stress, since (+)-pentazocine and DTG failed to attenuate its effect.

In the present study, DHEAS and PREGS, as well as (+)-SKF-10,047, attenuated the conditioned fear stress in mice. Further, the attenuating effects of both neurosteroids were antagonized by a selective sigma1 receptor antagonist. Thus, we speculate that the attenuating effects of DHEAS and PREGS may be mediated via sigma1 receptors, particularly, phenytoin-regulating sigma1 site, but not sigma2 receptors. Further research, however, is needed to confirm this hypothesis. In addition, allopregnanolone at the dose of 50 mg/kg did not attenuate the conditioned fear stress response. It is unlikely that the doses of allopregnanolone are insufficient, since allopregnanolone less than 20 mg/kg, as well as diazepam, has anxiolytic effects in numerous anxiety paradigms, including the Vogel conflict, open field light/dark transition and elevated plus maze tasks in rodents (Bitran et al. 1991; Brot et al. 1997; Wieland et al. 1991). The anxiolytic actions of this drug have been proposed to be related to an augmentation of GABAA receptor function, although the mechanisms underlying these effects are yet unclear (Bitran et al. 1991; Brot et al. 1997; Wieland et al. 1991). It has been demonstrated that anxiolytics such as diazepam or chlordiazepoxide do not attenuate the motor suppression in conditioned fear stress (Kameyama and Nagasaka 1982; Nagasaka and Kameyama 1983). Thus, it is likely that the conditioned fear stress is resistant to GABAA mimic neurosteroids and agents such as allopregnanolone and benzodiazepines.

A crossed pharmacology between the effects of sigma1 receptor ligands and neurosteroids was recently described for neuronal responses, DHEAS and PREGS behaving as agonists and PROG as an antagonist (Bergeron et al. 1996; Monnet et al. 1995). In the present study, PROG at 10–50 mg/kg, which inhibit the in vivo haloperidol-sensitive [3H]-(+)-SKF-10,047 binding to sigma1 receptor in the mouse hippocampus and cortex (Maurice et al. 1996), had no effect on the conditioned fear stress response in mice. Like NE-100, PROG behaved as a potent sigma1 receptor antagonist in the conditioned fear stress response, since it antagonized the attenuating effects of (+)-SKF-10,047, DHEAS, and PREGS on the response. Such an effect of PROG was observed in other experiments, that is the attenuating effects of DHEAS and sigma1 receptor agonists on immobility in the Porsolt forced swim test and on memory impairment were antagonized by PROG (Maurice et al. 1998; Reddy et al. 1998; Urani et al. 1998). Thus, these results suggest that PROG acts as a potent sigma1 receptor antagonist, whereas DHEAS and PREGS act as sigma1 receptor agonists.

PROG antagonized the attenuating effects of neurosteroids (DHEAS and PREGS) at smaller doses compared to those of (+)-SKF-10,047. One possible explanation is that this may be due to difference of affinities for sigma1 receptors between (+)-SKF10,047 and neurosteroids used since the affinity of (+)-SKF-10,047 for sigma1 receptors is higher than that of DHEAS, PREGS, and PROG. However, this point must be considered with caution, as other neuropharmacology of sigma ligands and neurosteroids remains to be clarified.

Interestingly, we found that the serum DHEAS concentration in mice showing the conditioned fear stress response was lower than that in the non-shocked mice, but the serum DHEAS concentration in mice which did not place into the same test cage 24 h after shock treatment, was unchanged (data not shown). This finding suggests that changes in the serum level of DHEAS are involved in the stress-induced physiopathological changes such as anxiety and depression. A recent clinical study suggests that DHEA is involved in depression: when DHEA was openly administered to patients with major depression and low plasma levels of DHEA and DHEAS for four weeks, depression ratings were significantly improved with the increase in DHEA and DHEAS plasma levels (Wolkowitz et al. 1997). Further, when DHEA was administered to patients with treatment-resistant depression for six months, a marked improvement in depression ratings was observed (Wolkowitz et al. 1997).

The conditioned fear stress has been regarded as an animal model of treatment-resistant psychoses, since anxiolytics and/or antidepressants did not attenuate the motor suppression in conditioned fear stress (Kameyama and Nagasaka 1982; Nagasaka and Kameyama 1983). Taken together, these findings suggest that the decrease in the serum levels of DHEAS is related to the development of conditioned fear stress response and neurosteroids might be useful anti-psychotics for treatment-resistant psychoses.

Conditioned fear stress has been shown to affect central neurotransmitters such as the dopamine (Herman et al. 1982; Inoue et al. 1994; Katoh et al. 1996; Nabeshima et al. 1986), norepinephrine (Tsuda et al. 1986), and serotonin (Inoue et al. 1993, 1994) neuronal systems, which are well known to be involved in stress responses. In this regard, our previous behavioral and neurochemical experiments suggest that the reduction of dopamine release from dopaminergic terminals in the nucleus accumbens is responsible for the development of the motor suppression in conditioned fear stress, and that (+)-SKF-10,047 attenuates the conditioned fear stress-induced motor suppression by reversing the conditioned fear stress-induced dysfunction in the mesolimbic dopaminergic systems (Kamei et al. 1996b, 1997). A preliminary experiment showed that DHEAS, like (+)-SKF-10,047, restored the decreased dopamine turnover in the nucleus accumbens of shocked mice to control levels, as indicated by the increase in DOPAC/dopamine and HVA/dopamine (data not shown).

There is evidence that sigma1 receptors modulate the activity of dopaminergic neurons: the existence of sigma1 receptors on the dopaminergic neurons has been demonstrated (Gundlach et al. 1986) and electrophysiological studies have shown that these receptors are involved in the regulation of neuronal activity of the midbrain dopaminergic neurons (Freeman and Bunney 1984; Steinfels and Tam 1989). Further, the in vivo brain microdialysis technique has demonstrated that sigma ligands as (+)-SKF-10,047 (Volonte et al. 1995) and (+)-pentazocine (Patrick et al. 1993) increase the endogenous dopamine release in rat brain. In conclusion, we speculate that DHEAS and PREGS, as well as sigma1 receptor agonists, reverse the reduction of dopamine release in the nucleus accumbens via activation of presynaptic sigma1 receptors of the mesolimbic dopaminergic system and, as a result, conditioned fear stress response is attenuated. In the present study, neurosteroids and (+)-SKF-10,047 did not totally attenuate the conditioned fear stress response in mice. Previously, we have found that opioid μ-agonists also attenuate the conditioned fear stress response via dopaminergic systems. Thus, the partial attenuating effects of DHEAS, PREGS, and (+)-SKF-10,047 on conditioned fear stress response in mice, may be due to involvement in non-sigma-mediated mechanisms. However, further studies are needed to verify this point.

At present, much evidence suggests that there is a common mechanism of action between neurosteroids and sigma1 receptor agonists in stress-induced disorders such as anxiety and depression (Reddy et al. 1998; Wolkowitz et al. 1997). However, this mechanism has still to be examined, particularly as concerns the different neurotransmission systems involved in the pathological state. We believe that the use of neurosteroids is a novel therapeutic approach for at least some mood disorders including treatment-resistant depression.