Isoallopregnanolone reduces tic-like behaviours in the D1CT-7 mouse model of Tourette syndrome

Abstract

Tourette syndrome (TS) is a neuropsychiatric disorder characterised by multiple, persistent tics. These semi-voluntary motor and phonic manifestations are typically aggravated by exposure to acute stress, yet the mechanisms underlying this exacerbation remain unclear. Using a well-characterised animal model of TS, the D1CT-7 mouse, we recently showed that acute stress increases tic-like responses and causes sensorimotor gating deficits, as measured by the prepulse inhibition of the startle. We showed that these effects are promoted by the brain synthesis of the neurosteroid allopregnanolone (AP). In line with this idea, inhibition of AP synthesis by finasteride was found to suppress the tic-exacerbating effects of stress; conversely, AP administration resulted in a marked enhancement of the number of tic-like motor bursts. Given that the primary mechanism of AP is based on the positive allosteric modulation of GABA_A receptors, in the present study, we hypothesised that the enhancement in tic-like behaviours induced by either stress or AP may be countered by isoallopregnanolone (isoAP), the natural 3β-epimer of AP that acts as an antagonist to the AP-binding site within GABA_A receptors. In agreement with our hypothesis, isoAP (5-10 mg kg⁻¹, s.c.) dose-dependently reduced the number of tic-like behaviours induced by stress in D1CT-7 mice. These effects were comparable to those elicited by both the benchmark TS therapy haloperidol (0.3 mg kg⁻¹ , i.p.), as well as finasteride (25 mg kg⁻¹ , i.p.). IsoAP also countered the prepulse inhibition deficits secondary to stress in D1CT-7 mice. Finally, isoAP opposed the enhancement of tic-like behaviours induced by AP (15 mg kg⁻¹, i.p.). Given that isoAP is well-tolerated and has an optimal safety profile, these data suggest that this steroid may have therapeutic properties in TS.

1. INTRODUCTION

Tourette syndrome (TS) is a childhood-onset disorder characterised by multiple motor tics and at least one phonic tic for at least 1 year. Tics are repetitive spasmodic manifestations, typically executed in an abrupt and purposeless fashion. Recent phenomenological studies have revealed that tics are often the epiphenomena of complex behavioural sequences, which typically begin under the guise of premonitory urges and intrusive feelings of fixation on somatic cues.¹ These sensory antecedents are either initiated or exacerbated by stressful environmental triggers^{2, 3} and are associated with a marked sense of discomfort, which can only be alleviated by tic execution. Although tics can be transiently suppressed, their deferment results in an escalation of the psychological burden associated with the corresponding urges, until the individual can no longer refrain from their execution.¹

This background highlights the importance of stress in the modulation and exacerbation of tics. To understand the underlying mechanisms, we have examined the involvement of neurosteroids, a class of brain-borne steroids that are synthesised in response to acute stress to promote coping and resilience.⁴ We first reported that tic severity in TS patients is reduced by inhibitors of the enzyme 5α-reductase (5αR), the enzyme catalysing the rate-limiting step of neurosteroid synthesis.^5-7 5αR converts progesterone into 5α-dihydroprogesterone, which is further converted into the key neurosteroid allopregnanolone (3α,5α, tetrahydroprogesterone; AP) by 3α-hydroxysteroid oxidoreductase.⁸

Building on this evidence, we investigated whether 5αR inhibitors may reduce TS-related phenotypes using animal models. In pharmacological models of TS-related phenotypes, 5αR inhibitors were found to reduce the deficits of sensorimotor gating associated with dopaminergic activation.^9-11 We then studied the impact of stress and neurosteroids in the D1CT-7 mouse, one of the best-validated animal models of TS. This transgenic mutant was generated by the attachment of the D1 dopamine receptor promoter to the gene encoding the neuropotentiating cholera toxin. This mutation results in spontaneous tic-like jerks that are reduced by benchmark therapies for TS, such as the antipsychotic haloperidol.^{12, 13} We reported that D1CT-7 mice display exacerbated tic-like responses and prepulse inhibition (PPI) deficits in response to a naturalistic environmental stressor, consisting of a 20-minute spatial confinement within a 10-cm diameter cylindrical enclosure in their home cages.¹⁴ This stress was associated with an increase in plasma corticosterone, as well as cortical neurosteroids, such as progesterone, 5α-dihydroprogesterone and AP.^{14, 15} Our analyses ascertained that AP, but not other neurosteroids, exacerbated tic-like behaviours in D1CT-7 mice; conversely, the suppression of AP synthesis by the prototypical 5αR inhibitor finasteride reduced tic frequency.¹⁵ Collectively, this evidence strongly suggests that AP mediates stress-induced enhancement of tics.

Depending on its concentrations, AP acts as either a positive allosteric modulator or agonist of GABAA receptors.^{16, 17} The effects of AP on these receptors are antagonised by its 3β-epimer, isoallopregnanolone (isoAP; 3β,5α, tetrahydroprogesterone).18-20 Similar to other GABAA receptor modulating steroid antagonists, isoAP opposes the positive modulation of this receptor and does not inhibit GABA-evoked currents.^{18, 21} Recently, a randomised clinical trial validated that isoAP is a highly effective therapy for premenstrual dysphoric disorder, with optimal safety and tolerability profiles.²² Building on these findings, the present study aimed to test whether isoAP may reduce the enhancement of tic-like responses in D1CT-7 mice and PPI deficits induced by stress.

2. MATERIALS AND METHODS

2.1. Animals

We used 3-4-month-old, experimentally naïve male Balb/c mice weighing 20-30 g. Animals were purchased from Jackson Laboratories (Bar Harbor, ME, USA) and genotyped as described previously.¹² Because the inheritance pattern of the D1CT-7 mutation is autosomal dominant, we bred wild-type (WT) females with heterozygous D1CT-7 sires. This breeding scheme was selected to standardise maternal behaviour. Animals were housed in group cages with access to food and water available ad lib. The room was maintained under a 12 : 12 hour light/dark cycle (lights on 8.00 am) at 22 °C. Animals were tested during their light cycle between 12.00 and 4.00 pm to minimise any potential circadian effects. All experimental procedures were conducted in accordance with the NIH guidelines and approved by the IACUC of the University of Utah.

2.2. Drugs

The drugs used were: AP, finasteride (Tocris Bioscience, St Louis, MO, USA), haloperidol (Sigma-Aldrich, St Louis, MO, USA) and isoAP (donated by Asarina Pharma AB, Solna, Sweden). IsoAP was suspended in 3% hydroxypropyl β-cyclodextrin and given s.c. Finasteride and AP were suspended in 5% Tween 80, diluted with 0.9% saline, and administered by i.p. injections. Haloperidol was dissolved in 10% acetic acid buffered with NaOH and diluted with saline. The injection volume for all administrations was 10 mL kg⁻¹.

2.3. Spatial confinement

Spatial confinement in the home cage was used as an environmental stressor to evoke tic-like responses and PPI deficits, as described previously.¹⁴ Briefly, mice were isolated for 3 days to allow them to establish territorial behaviour (a key factor to maximise the stressfulness of spatial confinement in the home cage). Animals were then confined within a clear, bottomless Plexiglas cylinder (diameter: 10 cm; height: 30 cm), within their home cages, deeply embedded in bedding to ensure stability. Spatial confinement lasted 30 minutes, and behaviours were video-recorded for the last 20 minutes (to allow neophobia to subside). Tic-like manifestations, as well as digging, rearing and grooming responses, were scored by trained observers who were blinded to the genotype and treatment. Tic-like manifestations were defined as rapid (<1 second) twitches of the head and/or body.

2.4. Bar test

Catalepsy was measured by means of the bar test, performed as described previously.²³ Briefly, the forepaws of the mouse were placed on a bar fixed at the height of 5 cm above the working surface. The latency to descend from the bar was calculated.

2.5. Startle testing

Testing was performed as described previously.¹⁴ Briefly, the apparatus used for detection of startle reflexes (SR-LAB; San Diego Instruments, San Diego, CA, USA) consisted of five Plexiglas cages (diameter 5 cm) in sound-attenuated chambers with fan ventilation. Each cage was mounted on a piezoelectric accelerometric platform connected to an analogue-to-digital converter. The response to each stimulus was recorded as 65 consecutive 1-ms readings. A dynamic calibration system was used to ensure comparable sensitivities across chambers. The startle testing protocol featured a 70-dB background white noise, and consisted of a 5-minute acclimatisation period, followed by three consecutive blocks of pulse, prepulse + pulse and ‘no stimulus’ trials. During the first and the third block, mice received only five pulse-alone trials of 115 dB. Conversely, in the second block, mice were exposed to a pseudorandom sequence of 50 trials, consisting of 12 pulse-alone trials, 30 trials of pulse preceded by 73, 76 or 82 dB pre-pulses intensities (10 for each level of prepulse loudness), and eight ‘no stimulus’ trials, where only the background noise was delivered. Intertrial intervals were selected randomly with a duration of between 10 and 15 seconds. Sound levels were assessed using an A-scale setting. Percent PPI (%PPI) was calculated with the formula:

$$100-\frac{\text{mean startle amplitude for prepulse pulse trials}}{\text{mean startle amplitude for pulse alone trials}}\times 100$$

The first five pulse-alone bursts were excluded from the calculation. Because no interactions between prepulse levels and treatment were found in the statistical analysis, %PPI values were collapsed across prepulse intensity to represent average %PPI.

2.6. Experimental procedures

The study included four experiments.

In the first experiment (n = 8 per group), we tested the effects of isoAP (5-10 mg kg⁻¹, s.c.) in our model of tic-like behaviour exacerbation by spatial confinement stress. D1CT-7 and WT mice were treated with isoAP (5-10 mg kg⁻¹, s.c.) or its vehicle (cyclodextrin) immediately before being subjected to stress, and tic-like responses and other behaviours were measured.

In the second experiment (n = 7-8 per group), we tested whether isoAP (10 mg kg⁻¹, s.c., immediately before spatial confinement) reduced tic-like behaviours of spatially confined WT and D1CT-7 mice, in a fashion comparable with that of the 5αR inhibitor finasteride (25 mg kg⁻¹, i.p., injected 35 minutes before spatial confinement) and the benchmark treatment haloperidol (0.3 mg kg⁻¹, i.p., injected 40 minutes before spatial confinement). These drugs were compared with their respective vehicles (finasteride: 5% Tween 80 in saline; haloperidol: 10% acetic acid diluted in saline and buffered with NaOH). Immediately after this experiment, animals were tested for catalepsy.

In the third experiment, (n = 8 per group), we tested whether isoAP could also reverse the PPI deficits induced by spatial confinement. Accordingly, D1CT-7 and WT littermates were placed in the cylinder (as described above). After 20 minutes, mice were briefly removed from the cylinder and treated with either isoAP (10 mg kg⁻¹, s.c.) or its vehicle (cyclodextrin). Following treatment, animals were left in the cylinder for an additional 10 minutes, after which they were tested for startle and PPI.

Given that we had previously shown that the mechanism by which spatial confinement promotes tic-like behaviours via AP, in the fourth experiment (n = 7-8/group), we tested whether isoAP (10 mg kg⁻¹, s.c.) could block the enhancement in tic-like responses induced by AP (15 mg kg⁻¹, i.p.). Accordingly, each animal was treated with either isoAP or its vehicle (cyclodextrin), immediately followed by AP (15 mg kg⁻¹, i.p.). Behavioural testing lasted 30 minutes. Throughout the study, all analyses were performed in counterbalanced order, so as to minimise potential differences in testing time. Treatments were randomly assigned to each experimental group.

2.7. Statistical analysis

Normality and homoscedasticity of data distribution were verified by using Kolmogorov-Smirnov and Bartlett's tests. Statistical analyses of parametric data were performed using two-way ANOVA (with treatment and genotype as factors), followed by Newman-Keuls’ test for post-hoc comparisons. P < 0.05 was considered statistically significant.

3. RESULTS

3.1. IsoAP dose-dependently reduces tic-like behaviours induced by spatial confinement

In the first experiment, we tested the effects of isoAP (5-10 mg kg⁻¹, s.c.) in our model of stress-exacerbated tic-like behaviours. Our results showed that D1CT-7 mice exhibited a significantly higher number of tic-like behaviours than those observed in WT counterparts (main effect of genotype: F_1,42 = 42.01, P < 0.0001; partial η² = 0.50); furthermore, isoAP produced a significant reduction in tics across both genotypes (F_2,42 = 41.52, P < 0.0001; partial η² = 0.66). Finally, a significant genotype × treatment interaction was found (F_2,42 = 10.31, P = 0.0002; partial η² = 0.33). Specifically, post-hoc analyses revealed that both doses of isoAP reduced tic-like behaviours in D1CT-7 (isoAP 5 mg kg⁻¹ vs vehicle: P < 0.0001; isoAP 10 mg kg⁻¹ vs vehicle: P < 0.00001), although not significantly so in WT mice (Figure 1A). The analysis of digging behaviour (Figure 1B) showed a significant effect of treatment (F_2,42 = 5.16, P = 0.01; partial η² = 0.20), which was found to reflect a reduction of this behaviour in animals treated with the highest isoAP dose (P < 0.001). However, neither a main effect of genotype, nor a significant genotype × treatment was detected. Finally, no significant effects were found for rearing (Figure 1C) or grooming (Figure 1D) behaviours.

Figure 1:

Effects of isoallopregnanolone (isoAP) (5-10 mg kg⁻¹, s.c.) on tic-like responses evoked by spatial confinement stress in D1CT-7 and wild-type (WT) mice. Comparisons are based on post-hoc testing following two-way ANOVA analysis and are indicated by dotted lines. ***P < 0.001. NS, not significant. The numbers of each experimental group are indicated. For further details, see text.

3.2. The tic-suppressing effects of isoAP are comparable with those of haloperidol and finasteride

In the second experiment, we then tested whether isoAP (10 mg kg⁻¹, s.c.) reduced tic-like behaviours in a fashion comparable with that of the 5αR inhibitor finasteride (25 mg kg⁻¹, i.p.) and the benchmark treatment haloperidol (0.3 mg kg⁻¹, i.p.). As expected, main effects were found for genotype (F_1,84 = 179.6, P < 0.0001; partial η² = 0.68) and treatment (F_5,84 = 19.33, P < 0.0001; partial η² = 0.58). A significant genotype × treatment interaction was also found (F_5,48 = 25.14, P < 0.0001; partial η² = 0.49), which was shown to reflect the effects of all three compounds in D1CT-7 mice (P < 0.00001 vs vehicle), although not in WT mice (Figure 2). Notably, no significant differences were found between drugs, indicating that the efficacy of isoAP was comparable with the other treatments. By contrast to haloperidol, however, neither isoAP, nor finasteride produced any cataleptic side-effects (Table 1).

Figure 2:

Effects of isoallopregnanolone (isoAP) (10 mg kg⁻¹, s.c.) on tic-like responses evoked by spatial confinement stress in D1CT-7 and wild-type (WT) mice compared to haloperidol and finasteride. Every drug was compared with its vehicle. Comparisons are based on post-hoc testing following two-way ANOVA analysis and are indicated by dotted lines. ***P < 0.001. NS, not significant. The numbers of each experimental group are indicated. For further details, see text.

Table 1.

Effects of isoallopregnanolone (isoAP) (10 mg kg⁻¹, s.c.) on tic-like responses evoked by spatial confinement stress in D1CT-7 and wild-type (WT) mice compared to haloperidol and finasteride. Every drug was compared with its vehicle. The numbers of each experimental group are indicated. For further details, see text. ^a P < 0.001 for main effect of haloperidol vs vehicle.

Genotype	Drugs	Catalepsy (s)	N
WT	Vehicle of isoAP (s.c.)	0.13 ± 0.05	8
	IsoAP (10 mg kg−1, s.c.)	0.14 ± 0.06	8
	Vehicle of haloperidol (i.p.)	0.3 ± 0.05	8
	Haloperidol (0.3 mg kg−1, i.p.)	11.38 ± 0.86a	8
	Vehicle of finasteride (i.p.)	0.13 ± 0.05	8
	Finasteride (25 mg kg−1, i.p.)	0.16 ± 0.04	8
D1CT-7	Vehicle of isoAP (s.c.)	0.28 ± 0.03	7
	IsoAP (10 mg kg−1, s.c.)	0.21 ± 0.05	8
	Vehicle of haloperidol (i.p.)	0.17 ± 0.06	8
	Haloperidol (0.3 mg kg−1, i.p.)	11.25 ± 0.98a	8
	Vehicle of finasteride (i.p.)	0.29 ± 0.05	8
	Finasteride (25 mg kg−1, i.p.)	0.25 ± 0.07	8

3.3. IsoAP reduces PPI deficits induced by spatial confinement

In the third experiment, we tested whether isoAP (10 mg kg⁻¹, s.c.) could oppose the effects of spatial confinement stress on PPI deficits.¹⁴ Our results showed no significant differences in startle amplitude between genotypes and treatment (Figure 3A). As expected, the analysis of PPI revealed that stress impaired PPI in D1CT-7 mice, and this deficit was significantly reversed by iso-AP (F_1,27 = 18.73, P = 0.0002; partial η² = 0.41; P < 0.01 for comparison of vehicle-treated WT and vehicle-treated D1CT-7 mice; P < 0.001 for comparison of isoAP-and vehicle-treated D1CT-7 mice; Tukey's) (Figure 3B).

Figure 3:

Effects of isoallopregnanolone (isoAP) (10 mg kg⁻¹, s.c.) or its vehicle on startle response (A) and prepulse inhibition (PPI) (B) following spatial confinement stress in D1CT-7 and wild-type (WT) mice. Comparisons are based on post-hoc testing following two-way ANOVA analysis and are indicated by dotted lines. **P < 0.01, ***P < 0.001. The numbers of each experimental group are indicated. For further details, see text.

3.4. IsoAP counters the enhancement in tic-like behaviours induced by AP

Finally, we tested whether isoAP (10 mg kg⁻¹, s.c.) could reverse the increase in tic-like behaviours induced by AP (15 mg kg⁻¹, i.p.) (n = 7-8 per group). In addition to main effects for genotype (F_1,26 = 48.63, P < 0.0001; partial η² = 0.63) and isoAP (F_1,26 = 22.14, P = 0.0004; partial η² = 0.43), a significant interaction was detected (F_1,26 = 6.82, P = 0.01; partial η² = 0.19). Post-hoc comparisons revealed that iso-AP significantly reduced tic-like behaviours in AP-treated D1CT-7 mice (P = 0.001) (Figure 4A). The analysis of digging (Figure 4B) and rearing (Figure 4C) behaviours showed no significant effects of isoAP and genotype and their interaction. Finally, isoAP significantly reduced grooming responses (main effect: F_1,24 = 18.35, P = 0.0003; partial η² = 0.22) (Figure 4D).

Figure 4:

Effects of isoallopregnanolone (isoAP) (10 mg kg⁻¹, s.c.) or its vehicle on tic-like responses caused by allopregnanolone (AP) in D1CT-7 and wild-type (WT) mice. Comparisons are based on post-hoc testing following two-way ANOVA analysis and are indicated by dotted lines. ***P < 0.001. NS, not significant. The numbers of each experimental group are indicated. For further details, see text.

4. DISCUSSION

The results of the present study show that isoAP dose-dependently ameliorates tic-like responses induced by either spatial confinement stress or AP administration in the D1CT-7 mouse model of TS. Such amelioration was not consistently observed for other spontaneous behaviours, emphasising the specificity of this effect. We previously reported that spatial-confinement stress increases the levels of AP and its precursors in the prefrontal cortex (PFC) of D1CT-7 mice, without modifying GABAA receptor subunit composition.¹⁵ The findings of the present study complement those data by indicating that GABAA receptors contribute to the tic-exacerbating effects of AP.

Of clinical relevance, our findings also show that isoAP produced effects akin to those of haloperidol and finasteride; however, unlike haloperidol, isoAP did not elicit cataleptic reactions. Furthermore, in contrast to finasteride, isoAP has not been linked to any demasculinising effects, and may therefore be employed during childhood. In particular, finasteride recently has been shown to be associated with depression, sometimes irreversible sexual dysfunction and even suicide attempts,^24-26 raising concerns about its potential application as a therapy for neuropsychiatric disorders. Clinical trials have recently shown that isoAP is effective in the treatment of premenstrual dysphoric disorder - a condition characterized by mood changes associated with the menstrual cycle -, and has an optimal safety and tolerability profile.^{22, 27} These results suggest that, in contrast to haloperidol and finasteride, isoAP may be a promising strategy for TS therapy, with potential applicability in children and adolescents.

The molecular mechanisms by whivh GABAA receptor activation by AP may contribute to the exacerbation of tic-like behaviours in D1CT-7 mice remain elusive; however, we previously showed that spatial confinement enhances AP levels in the PFC, possibly indicating the specific involvement of this brain region. The involvement of the PFC in the action of steroids, especially of progesterone/AP, has also been shown in humans.²⁸

Given that the PFC elicits an inhibitory function on subcortical regions, the action of AP in the PFC may lead to a greater inhibition of pyramidal cells, which in turn may lead to the disinhibition of striatum (via corticostriatal connections) and other subcortical areas. From this perspective, isoAP may also exert its tic-ameliorating properties by preventing an excessive inhibition of pyramidal neurones in the PFC. At the same time, it is unlikely that the effects of isoAP may be based on the striatum because previous functional studies have shown that local dorsal striatal infusions of the GABAA receptor antagonist bicuculline produce tic-like behaviours in rodent models.²⁹

Allopregnanolone is produced by several brain regions in response to acute stress³⁰ as a way of increasing stress coping and, ultimately, curbing the untoward emotional outcomes of stress^{31, 32} and promoting psychological resilience (defined as the adaptive ability to overcome adversity and support healthy behaviour). From this perspective, the idea that enhancements in AP levels may exacerbate tics may appear to be counterintuitive; however, as described above, tics are often enacted by TS patients as a maladaptive strategy to cope with stress and to achieve relief from the burden associated with premonitory urges and sensory phenomena. This interpretation reveals the possibility that the same neurobiological factors that drive healthy behaviour in most situations may be paradoxically problematic when applied to situations characterised by corticolimbic miswiring such as in TS. Future research should be devoted to reducing the neurobiological factors that cause or facilitate maladaptive resilience in TS patients. In addition, it should be noted that, given the relevance of D1CT-7 mice to obsessive-compulsive disorder, the observed effects of isoAP may also signify its therapeutic potential in this condition and other related disorders, such as trichotillomania. In agreement with this possibility, we found that isoAP reduced the duration of repetitive behaviours, such as digging stereotypies.

Several limitations of the present study should be recognised. First, the translational value of these findings is partially limited by the evaluation of the effects of isoAP only in one model of TS. Further studies on alternative models of TS should be performed to fully assess the potential translational value of these findings. Nevertheless, it is worth noting that, among the currently available animal models of TS, D1CT-7 mice feature unique characteristics with respect to face and predictive validity,^33-35 including neuropotentiation of the sensorimotor cortex, which is a characteristic that may overlap with the mechanisms of premonitory urges. Second, our studies were only conducted in adult male D1CT-7 mice because the effects of spatial confinement stress are best evident at this stage; nevertheless, future studies are needed to evaluate the effects of isoAP on TS-relevant phenotypes in juvenile and female mice. Third, we did not measure whether confinement stress may induce changes in the brain content of isoAP; thus, we cannot exclude that the effects of this steroid may be moderated by variations in its endogenous levels. Fourth, although our results have tested the specific hypothesis that isoAP counters the effects of AP, other neurosteroids and AP-sensitive receptors may also participate in the behavioural effects of stress. Further investigations will need to focus on other neurosteroids increased by stress, such as tetrahydrodeoxycorticosterone, or androgenic neuroactive steroids, such as androstanediol, as well as the role of additional AP targets, including membrane progesterone receptors or pregnane X receptors. These analyses may prove essential with respect to clarifying the male predominance of TS, as well as potential mechanisms of comorbidity with other neuropsychiatric problems, including attention deficit hyperactivity disorder and obsessive-compulsive disorder.

Despite these limitations and caveats, the findings of the present study are the first to suggest the potential therapeutic value of isoAP in the treatment of stress-induced exacerbation of tic-like behaviours, highlighting the involvement of positive GABA_A receptor modulating steroids such as AP in the stress-induced exacerbation of TS. Additional clinical studies will be essential to validate these constructs in TS patients and to explore the therapeutic potential of isoAP and other neurosteroid-targeting therapies in TS.

DATA AVAILABILITY

The data that support the findings of this study are available from the corresponding author upon request.