While clinically used general anesthetics are considered relatively safe, there is an ongoing interest in the development of new, potentially safer general anesthetics, especially for specific populations such as children. Neuroactive steroids are utilized as anesthetic agents due to their ability to modulate the activity of γ-aminobutyric acid type A (GABA_A) receptors in the brain, leading to a sedative and anesthetic effect. Currently, neuroactive steroids are primarily used for the induction and maintenance of general anesthesia in veterinary medicine (Alphaxalone), but there is ongoing interest in exploring their potential for human use as well. Here, we are comparing the spectral characteristics and sex differences in the hypnotic effect of two neuroactive steroids, Allopregnanolone and its synthetic analogue, Alphaxalone. Our data demonstrate that it is important to consider the use of neuroactive steroids for clinical anesthesia, particularly for their potential to be safer than traditional anesthetics.

Neuroactive steroids are endogenous or exogenous steroids with direct effects on neuronal excitability. It has been shown that the spatial arrangement of the hydroxyl group at the C3 position of the steroid skeleton dramatically affects the biological properties of the neuroactive steroids [1, 2]. Previous reports demonstrated that 3α-hydroxy neuroactive steroids, like endogenous Allopregnanolone (AlloP, 3α-hydroxy-5α-pregnan-20-one) or its synthetic analog Alphaxalone (Alpx, 3α-hydroxy-5α-pregnane-11,20-dione), act like positive allosteric modulators of γ-aminobutyric acid type A (GABA_A) receptors in nanomolar concentrations [3–6]. This effect on γ-aminobutyric acid (GABA) currents is consistent with known behavioral effects of neuroactive steroids including anxiolysis, analgesia, anticonvulsant activity, sedation, hypnosis and anesthesia [7].

Furthermore, neuroactive steroids have been used in human and veterinary clinics as potentially safer anesthetic agents [8]. In comparison to traditional general anesthetics, many neuroactive steroids have favorable clinical properties: minimal cardiorespiratory depression, rapid onset and recovery, less neurotoxicity in young populations [2, 8–10]. It has been shown that AlloP is a potent hypnotic drug not just in wild-type (WT) animals but also in animals that lack progesterone receptors, suggesting that progesterone receptors are not involved in the hypnotic response to AlloP [11, 12]. Similarly, a synthetic drug, Alpx, has potent hypnotic/anesthetic and sedative properties [2, 8]. Although it was abandoned for clinical use due to unwanted side effects caused by its vehicle, there has been recent interest in the therapeutic value of Alpx: a new aqueous formulation (Phaxan^®) has been developed for use as an intravenous sedative and anesthetic in humans [9, 13]. While the sex-differences in various behavioral effects of neuroactive steroids are well documented in the literature, their neurophysiological signature is poorly understood [14].

Central medial nucleus of thalamus (CMT), as a part of intralaminar thalamic complex is implicated in arousal and is a key hub through which general anesthesia and sleep are initiated [15]. Here, we investigated sex-differences and spectral characteristics in the hypnotic effect of AlloP and Alpx in WT mice using behavioral testing (loss of righting reflex - LORR) and in vivo electrophysiology by recording local field potential (LFP) from CMT and electroencephalogram (EEG) from barrel cortex. Since there is an ongoing interest in developing new hypnotics and general anesthetics, we hope that our study may rekindle interest in clinical use of these agents as safer alternative to other common sedatives and anesthetics.

All experimental procedures with mice were performed according to a protocol approved by the Institutional Animal Care and Use Committee of the University of Colorado Anschutz Medical Campus, Aurora, CO, USA. Treatments of animals adhered to guidelines set forth in the NIH Guide for the Care and Use of Laboratory Animals. All efforts were made to minimize animal suffering and to use only the necessary number of animals to produce reliable scientific data. The authors confirm that the study complies with the ARRIVE guidelines.

Male and female adult C57BL/6J WT 3–4 months old mice (Jackson Laboratory, Bar Harbor, ME, USA) were used for behavioral and in vivo electrophysiology studies. All mice were maintained on a 14/10h light-dark cycle with food and water ad libitum. All tests were done in a blinded fashion. The estrus cycle of the female animals was not monitored during the experiments and can be a limitation of the study.

Although the introduction of Alpx and related 3α-OH analogues with hypnotic/anesthetic properties occurred in the 1970s, there is an increased interest recently in the development of new neuroactive steroids as therapeutic agents. For example, AlloP was formulated and FDA-approved as a medicine for postpartum depression (Brexanolone^®), and a 3β-methylated analogue of AlloP (Ganaxolone^®) was FDA-approved for treatment of seizure associated with the rare disease cyclin-dependent kinase-like 5 deficiencies [10]. Furthermore, a new aqueous formulation of Alpx (Phaxan^®) has been developed for use as an intravenous sedative and anesthetic in humans [9, 13].

As expected, we found that both AlloP and Alpx can reliably produce LORR in male and female mice. We further found that there are notable sex-differences in the effects of both neuroactive steroids in tested behavior. Specifically, the female animals are more sensitive to the hypnotic effect. Previous studies reported sex specific effect with both endogenous neuroactive steroids and synthetic analogues [21, 22]. The longer duration of LORR in females was previously described with Alpx in both mice and rats [22–24]. It was postulated that the interactions of neuroactive steroids with synaptic membranes may be more sex-specific [22, 23] and can potentially explain why the female mice in our study had a longer duration of hypnosis with the same dose of AlloP and Alpx. The authors found that the sex differences with Alpx is age-dependent, can be abolished by administering estrogen to the males, does not depend on sexual differentiation of the brain, and cannot be attributed to a sex difference in the metabolic clearance rate of neuroactive steroids [22].

Similarly, it was reported that neuroactive steroids such as AlloP exhibit sex differences in their anticonvulsant activity indicating reduced potency in the males relative to females [25]. This effect can be likely due to a greater abundance of extra-synaptic δ subunit of GABA_A receptors that mediates neuroactive steroid-sensitive, tonic GABA currents and seizure protection [26] and can at least partially explain sex specific effect in our experiments.

Interestingly, recent studies demonstrated that the female brain in mice and humans is less responsive to the hypnotic effects of volatile anesthetics largely due to acute effects of sex hormones [27]. Despite clear behavioral sex differences in anesthetic sensitivity, sex-differences were not visible in clinically used EEG but were detected in subcortical recordings [27]. This hidden resistance to volatile anesthetics may explain the higher incidence of awareness under anesthesia in females and requires investigation of sex-differences using both EEG (cortical) and LFP recordings [27].

Neuronal inhibition in the brain is predominantly provided by activation of GABA_A receptors - heteropentameric receptors formed from an array of subunits (α1-6, β1-3, γ1-3, δ, ε, π, θ and ρ1-3) [28, 29]. Synaptic receptors typically comprise α1-3, β and γ subunits, while extra-synaptic receptors contain α4/6, β and δ subunits [30]. It has been shown that AlloP has higher selectivity for tonic δ-containing GABA_A receptors, and that by being more selective to extra-synaptic GABA_A receptors, AlloP plays an essential neurophysiological role in the fine-tuning of neuronal inhibition mediated by GABA_A receptors [31, 32]. However, when injected in high doses, as in our case (100 mg/kg, i.p.), AlloP produced drastic suppression of total power in both male and female mice after LORR probably by acting on both synaptic and extra-synaptic GABA_A receptors. The clinically relevant dose range for AlloP in rodents varies depending on the desired effect and route of administration, and for anesthetic effect is higher (15–20 mg/kg i.v) than for anxiolytic/sedative effect in animals [28]. Additionally, the typical dose range for Alpx with hypnotic effect is between 50 and 60 mg/kg i.p. in females but higher doses are needed for male rodents [23]. To compare hypnotic effect of AlloP and Alpx, based on our preliminary dose-response experiments, we used the lowest dose of neuroactive steroids that induces LORR in all tested female and male mice.

Since the introduction of EEG, neural oscillations have been systematically categorized across behavioral states into canonical frequency bands: delta, theta, alpha, beta, gamma [33]. Both natural sleep and general anesthesia are associated with synchronized, high-amplitude slow oscillations, whereas conscious wakefulness is dominated by faster, desynchronized activity patterns [34]. Low doses of sedative-hypnotic agents, such as positive allosteric modulators of GABA_A receptors (e.g., propofol, barbiturates, etomidate), can paradoxically induce cortical excitation, typically reflected as enhanced β-band activity [35]. As anesthetic depth increases, EEG signatures progress through characteristic stages: Phase 1 (light sedation) is marked by reduced beta and elevated alpha power; Phase 2 (intermediate depth) displays prominent alpha and delta oscillations, closely resembling non-rapid eye movement slow-wave sleep; Phase 3 exhibits burst-suppression patterns; and Phase 4 is characterized by a near-isoelectric EEG, indicative of profound cortical silencing [35]. Although full anesthesia can be assessed by investigating withdrawal reflex to a painful stimulus, like tail or toe clamping and pinching, here we investigated just the hypnotic effect and were unable to assess if AlloP or Alpx induce surgical levels of anesthesia in high dose. However, strong EEG/LFP suppression that we observed with AlloP resembles burst suppression seen with potent volatile general anesthetics [36]. Similarly to sex-differences observed in behavioral test, the EEG/LFP suppression seen with AlloP was more profound in female animals in comparison to males. On the contrary, with the same dose of Alpx we did not observe LFP/EEG suppression, confirming previous results that Alpx alone can induce hypnosis or light anesthesia but even in higher doses probably cannot induce deeper surgical levels of general anesthesia [23, 24]. The observed changes between AlloP and Alpx-induced hypnosis could be at least partially explained by the differences in modulation of synaptic α₁β₂γ₂ GABA_A receptors; AlloP is a more potent positive allosteric modulator of these receptors than Alpx (ED₅₀ for α₁ containing GABA_A receptors for AlloP 0.18 ± 0.002 μM (mean ± SEM) and for Alpx 2.2 ± 0.04 μM) [37]. The same rationale can explain longer LORR durations induced by AlloP in comparison to Alpx in both female and male mice. Somewhat surprisingly, sex-differences in total EEG/LFP power in our experiments were not seen with the Alpx.

In the cortex and thalamus, power density analysis showed a shift in the power density peak from theta during wakefulness to delta frequencies after both neuroactive steroids, accompanied with a drastic suppression of EEG/LFP power density only in the AlloP group. On the contrary, thalamic spectral analysis showed increase in power density in slower frequencies under Alpx in comparison to the wake state. Similarly to other general anesthetics that act as GABA_A positive allosteric modulators, analysis of relative power showed increase in thalamic relative delta power in AlloP and Alpx animals while increase in EEG delta relative power was observed just with Alpx [38, 39]. Although both alpha and slow-delta oscillations are dominant EEG signature for propofol-induced hypnosis and hypnosis with other GABA_A positive allosteric modulators, with neuroactive steroids we observed a decrease and not an increase in thalamic and cortical alpha relative power [38, 39]. The difference in effect between neuroactive steroids and other GABA_A positive allosteric modulators was previously reported and might be related to effects on different GABA_A receptor isoforms [2, 10, 40, 41]. Additionally, neuroactive steroids have higher affinity for extra-synaptic GABA_A receptors responsible for persistent non-desensitizing inhibitory conductance [42]. Interestingly, we observed a substantial rise in relative cortical beta/low gamma powers after AlloP-induced LORR. A similar rise in low gamma range was seen in rodents after injections of ketamine, an N-methyl-D-aspartate (NMDA) antagonist [43, 44]. By contrast, we found that Alpx decreased thalamic beta and slightly increased cortical beta relative power and reduced low gamma relative power in both the thalamus and cortex. Overall, our data showed that in AlloP-treated mice, relative powers of slower (delta and theta) frequencies are lower but relative powers of higher (beta and low gamma) frequencies are higher than in the Alpx group.

Additionally, we used the Hilbert transform based phase synchronization analysis [19], to investigate the phase coupling during neuroactive steroid-induced hypnosis. Systematic phase synchronization changes were detected after injection of both neuroactive steroids. Generally, we observed reduction in corticocortical phase synchronization in AlloP animals that was more profound in slower than in higher frequency ranges. Similarly, Alpx induced drop in corticocortical synchronization in all except delta frequencies. We observed existence of higher corticocortical phase locking values under Alpx in comparison to AlloP in delta, theta and alpha frequencies. Previous models predicted that anesthesia would induce a decrease in slow wave EEG phase coherence as the cortex transitions away from the wakeful state [45]. In line with these findings, some experiments have confirmed that phase coherence during induction with propofol produces a significant drop in slow-wave bands synchronization between frontal, occipital, and frontal-occipital electrode pairs in humans [45, 46]. With AlloP we observed a similar decrease in cortical delta phase locking values. Previous findings reported that traditional general anesthetics can induce a large and localized increase in synchronization in alpha band as well as smaller and widespread synchrony increases in gamma frequency bands [47]. Studies using propofol in humans have reported that alpha band synchronization between cortical sites increases during induction and decreases during recovery [46]. By contrast, with neuroactive steroids we observed a profound decrease in alpha frequency synchronization in both corticocortical and thalamocortical leads in all treatment groups. Interestingly, we found increased synchronization in thalamocortical low gamma oscillation in both AlloP and Alpx treatment groups. This finding further denotes that neuroactive steroids induce very distinctive spectral changes that were not described with other similar GABA-mimetic general anesthetics.

Furthermore, in contrast to commonly used general anesthetics, both AlloP and Alpx activate progesterone receptors, and have neuroprotective effects that is independent of their GABA_A actions [48]. Recently, it has been shown that Alpx treated subjects scored better than propofol and sevoflurane anesthetized patients in the cognition tests, and that the higher cognition scores were accompanied by higher serum m-Brain-Derived Neurotrophic Factor levels in the Alpx-anesthetized patients [13]. Having this in mind, we posit that use of neuroactive steroids as general anesthetics may have beneficial effects in comparison to commonly clinically used drugs.

In conclusion, our data from behavioral assessments and LFP/EEG changes support the idea that female mice are more sensitive to neuroactive steroid-induced hypnosis. Our data also showed that neuroactive steroid-induced hypnosis has similar spectral characteristics as common general anesthetics with some unique LFP/EEG properties. While most used general anesthetics act as positive allosteric modulators of GABA_A receptors, they have different binding sites in comparison to neuroactive steroids and have markedly different side effects suggesting major differences in off-target binding and clinical effect [10]. Thus, we posit that future clinical use of neuroactive steroids for clinical anesthesia warrants consideration.