The endogenous neurosteroid allopregnanolone (ALLO; 5α-pregnan-3α-ol-20-one) is synthesized inside and outside the central nervous system.1 Unlike its precursors, progesterone and α5-dihydroxyprogesterone, it acts as a positive allosteric modulator on γ-aminobutyric acid type A (GABAA) receptors.2 Pharmacologic blockade of ALLO de novo synthesis in the rodent neocortex shortened the duration of GABAA receptor-mediated synaptic events.3 Conversely, increasing the neurosteroid synthesis enhanced the synaptic inhibition in neocortical brain slices.4 These observations suggest that ALLO impacts GABAA receptor function under physiologic conditions. A recent study investigated the interactions between the neurosteroid pregnanolone (5β-pregnan-3α-ol-20-one), another precursor of progesterone, and etomidate, an anesthetic that is mainly acting at GABAA receptors.5 Pregnanolone substantially amplified the potentiating action of etomidate on α1β2γ2L-GABAA receptors expressed in HEK 293 cells.5 The effects of etomidate and pregnanolone on GABA-evoked currents were supra-additive, indicating a synergistic interaction between these 2 compounds. This finding prompted us to hypothesize that synergistic interactions between neuroactive steroids and anesthetic agents at GABAA receptors could be a universal principle. To test this hypothesis, we explored how GABAA receptor-mediated inhibition is modified by ALLO, propofol, and a combination of both agents in cultured neocortical slices. Our results indicate a supra-additive action of ALLO and propofol. We further asked how this molecular mechanism translates into neuronal firing patterns. ALLO and propofol reduced the discharge rates of cortical neurons, but these effects were additive rather than synergistic. These findings suggest that combining propofol and neuroactive steroids is a novel and potentially beneficial approach to reduce anesthetic dose requirements and unwanted side effects in clinical anesthesia.6
Organotypic Slice Cultures
All procedures were approved by the animal care committee (Eberhard Karls University, Tübingen, Germany) and were in accordance with the German Animal Welfare Act (TierSchG). At any time, all efforts were made to minimize both the suffering and the number of animals. Wild-type C57BL6 mice of both sexes were used for this study. Neocortical organotypic slice cultures were prepared from neonatal mice of both sexes as described previously.7,8 The age of the cultures at the time of recording was 10 to 19 days in vitro for intracellular recording and 14 to 35 days in vitro for extracellular recording.
Extracellular multiunit recordings were performed at 34°C to detect action potential activity, as described in detail before.7 Slices were perfused with artificial cerebrospinal fluid (ACSF), consisting of (in millimolar) 120 NaCl, 3.5 KCl, 1.13 NaH2PO4, 1.0 MgCl2, 26 NaHCO3, 1.2 CaCl2, and 11 glucose (Merck, Darmstadt, Germany, and AppliChem, Darmstadt, Germany) bubbled with 95% oxygen and 5% carbon dioxide. ACSF-filled glass electrodes with a resistance of 3 to 5 MΩ were advanced into the tissue until a spike activity exceeding 100 µV in amplitude was detectable. Data were acquired using the DigiData 1200 AD/DA interface and AxoScope 9 software (Axon Instruments, Foster City, CA).
Voltage-clamp recordings were performed, as previously reported, to investigate the actions of ALLO and propofol on visually identified pyramidal neurons at room temperature.9 Pipettes were filled with a solution containing (in millimolar) 121 CsCL, 24 CsOH, 10 HEPES, 5 EGTA, 1 MgCl2, and 4 ATP adjusted to pH 7.2 with 1N HCl (Merck). Cells were voltage-clamped at −70 mV. To suppress glutamatergic synaptic transmission, we added 60 µM D-L-2-amino-5-phosphonopentanoic acid and 50 µM 6-cyano-7-nitroquinoxaline-2.3-dione (both from Tocris, Ellisville, MS).
The actions of ALLO and propofol on the neuronal network were explored by extracellular recordings, whereas we investigated the effects on the cellular level by intracellular recordings. Besides the single application of ALLO or propofol as well as the combined application of both substances, we conducted sham experiments (application of the solvent dimethyl sulfoxide [DMSO]) to test whether action potential firing rates of extracellular recordings and inhibitory postsynaptic currents (IPSCs) or baseline currents of intracellular recordings changed spontaneously during the timespan of the experiment. We observed a significant decrease of the action potential firing rate over time (residual median activity, 79% of control recordings; interquartile range [IQR], 40%; n = 72), as indicated (eg, in Figure 1). As a consequence of our finding, we statistically compared the concentration-dependent effects of ALLO and propofol with the sham application. However, in the case of IPSC kinetics and baseline currents, we did not detect any significant changes over time.
Preparation and Application of Test Solutions
We dissolved ALLO (Sigma-Aldrich, Taufkirchen, Germany) in DMSO (Sigma-Aldrich) to a 1 mM stock solution. Before the experiment, we diluted ALLO and propofol (Fresenius Kabi, Bad Homburg, Germany) in ACSF to reach the target concentration. Then we applied the drug containing ACSF through a bath perfusion using roller and syringe pumps (Ismatec, Cole-Parmer, Wertheim, Germany, and ZAK, Marktheidenfeld, Germany) and recorded γ-aminobutyric acid (GABA)ergic IPSCs from voltage-clamped neurons. After establishing the whole-cell configuration, we monitored the current traces for at least 7 minutes in the presence of D-L-2-amino-5-phosphonopentanoic acid and 6-cyano-7-nitroquinoxaline-2.3-dione (control condition). Thereafter, we added ALLO and/or propofol to the bathing solution and continuously recorded the synaptic activity for at least 9 minutes. To ensure that drug actions approached a steady state, we only used the last 3 minutes of each drug condition for IPSC analysis.
We used self-written MATLAB routines (MathWorks, Natick, MA) to filter and offline count extracellular recorded signals. When analyzing these signals, we detected action potentials and computed the average firing rate with a threshold set approximately 2 times higher than the baseline noise. Intracellularly recorded IPSCs and baseline changes (tonic current) were analyzed using self-written MATLAB routines. IPSC decay time was fitted with a double exponential fit. Absolute parameters are displayed for decay time and baseline changes as mean values and 95% confidence interval (CI). For IPSC analysis, cumulative probability diagrams are displayed, demonstrating the changes of the decay time. For the majority of our experiments, we had to reject the hypothesis that data samples derive from a normally distributed population based on the results obtained by q-q plots and the Lilliefors test. Data were normalized to the activity immediately recorded before applying ALLO or propofol (control condition). Results are presented by boxplots (line: median, box: lower quartile = 25th percentile and upper quartile = 75th percentile, lateral notch of the box: 95% CI; whisker: 1.5 × IQR; IQR: difference between the upper and the lower quartiles). To check for significant differences between 2 groups, we used the Mann-Whitney U test, eg, drug versus sham application. For statistical analysis of multiple groups, eg, propofol versus propofol and ALLO versus sham, we performed nonparametric analysis of variance (Kruskal-Wallis test) combined with multiple comparison post hoc correction (Bonferroni correction) using the MATLAB function “multcompare.” When working with this function, the number of comparisons equals the number of all combinations of pairs of groups, which can be created. Furthermore, 2-tailed tests are performed per pair of groups. For this reason k × (k − 1), tests are performed in total (k is the number of groups). Accordingly, the significance level in Figure 1 was divided by 42 (7 groups) and in Figure 5 by 12 (4 groups).
Actions of Allopregnanolone on the Discharge Rates of Neocortical Neurons
In this study, 115 slice cultures were used for extracellular recordings and 68 cultures for voltage-clamp recordings. The median action potential firing rate under control conditions was 10.6 ± 1.4 Hz (IQR, 13.5) and lies in the range of previously reported values.7,10,11 The concentration-dependent effects of ALLO (50 nM to 5 µM) on the firing rates are displayed in Figure 1. The results are normalized to the discharge rate under drug-free conditions. At ALLO concentrations >250 nM, neuronal activity was <50% of controls, and at 5 µM, action potential firing was almost completely depressed (residual median activity, 3%; IQR, 17%, n = 31). The lowest concentration of ALLO, which significantly altered firing rates (approximately 21% inhibition, n = 79, Mann-Whitney U test, P = .0017, Bonferroni correction) was 100 nM.
In further experiments, we quantified the actions of 100 nM ALLO on GABAA receptor-mediated synaptic inhibition by performing whole-cell voltage-clamp recordings. In our preparation, ALLO (100 nM) induced a tonic current, which was reversed by the GABAA-antagonist bicuculline (Figure 2A). Data from all recorded cells are summarized in Figure 2B. At 100 nM, ALLO (n = 6) shifted the baseline toward more negative values (mean and 95% CI, −24.6 ± 13.6 pA). However, we did not observe this effect in sham-treated preparations (DMSO, +6.6 ± 14.5 pA; n = 8). The difference between drug-treated and sham-treated cells was statistically significant (P < .002, 2-sample t tests).
In a next step, we explored the actions of ALLO on the time course of GABAA receptor-mediated IPSCs. Figure 3 displays representative recordings of spontaneous IPSC activity. We observed a broadening of IPSCs in the presence of ALLO (100 nM, highlighted in blue on the left), but not during sham experiments (highlighted in red on the right).
Figure 4A shows the effects of ALLO (100 nM, n = 23) on the IPSC decay time for all individual experiments. The application of ALLO clearly prolonged the decay time. We found that this effect was absent in cells treated only with DMSO (n = 22). The mean prolongation of the decay time was 12.9 ± 2.2 milliseconds and differed significantly (P = .01, Mann-Whitney U test) from the sham group (Figure 4B). Taken together, 100 nM ALLO inhibited the cortical action potential firing, which likely involved a tonic conductance and, in addition, a prolongation of GABAergic IPSCs.
Actions of the Combined Application of Allopregnanolone and Propofol on Neocortical Neurons
In further experiments, we characterized the interactions between ALLO and propofol by using almost equieffective concentrations of both agents. On the basis of findings from previous studies,9,11 we estimated that 250 nM propofol should reduce the discharge rates of cortical neurons by approximately 20%. Figure 5 shows the effects of 250 nM propofol and 100 nM ALLO on the discharge rates of neocortical neurons. As expected, both propofol (250 nM, n = 42) and ALLO (100 nM, n = 79) were similarly effective in reducing action potential firing (approximately 24% and 21% inhibition, respectively).
The application of ALLO together with propofol significantly reduced the firing rates by approximately 42% (n = 40, Mann-Whitney U test, P = .01, Bonferroni correction). Thus, the depression induced by ALLO and propofol is roughly the sum of the depression induced by the single application of 1 agent. In a next step, we analyzed drug actions on spontaneous GABAergic IPSCs to elucidate interactions between ALLO and propofol at the synaptic level. The cumulative probability plots in Figure 6A show distributions of IPSC decay times taken from a typical experiment under drug-free conditions in the presence of ALLO alone (100 nM, n = 168) and in a combination of ALLO (100 nM) and propofol (250 nM, n = 87). In this cell, ALLO increased the decay time by approximately 12 milliseconds. Furthermore, administering propofol together with ALLO increased the decay time by about the same amount.
However, propofol was much less effective without ALLO. Figure 6B shows a neuron pretreated with the solvent DMSO (number of recorded IPSCs = 321) before administering propofol (number of recorded IPSCs = 315). Here, the decay time was prolonged by only 4 milliseconds.
The IPSC decay times obtained from individual experiments under drug-free conditions, during ALLO administration, or a combination of both agents are presented in Figure 6C. Furthermore, Figure 6D shows the effects of DMSO or propofol. We used these data sets to calculate the effects of propofol on the decay time of synaptic events monitored either in the absence or in the presence of ALLO (Figure 7). In the presence of ALLO, propofol prolonged the decay time, on average, by approximately 10 milliseconds (mean change of IPSC decay time, 10.4 ± 6.1 milliseconds, n = 9); in the absence of ALLO, however, propofol prolonged the decay time only by approximately 4 milliseconds (mean change of decay time, 3.8 ± 2 milliseconds, n = 13). This difference is statistically significant (P = .0056, Mann-Whitney U test), indicating that ALLO increases the sensitivity of GABAA receptors for propofol.
We found that ALLO (100 nM) increases the decay time of IPSCs in neocortical neurons by approximately 50%. A quantitatively similar effect was reported for miniature IPSCs, recorded from CA1 pyramidal neurons.12 Furthermore, we observed that ALLO provoked a bicuculline-sensitive tonic current (Figure 2). However, the magnitude of this current was, albeit statistically significant, much more variable than the changes in IPSC decay times, confirming the findings of Harney et al.12 Because of this limitation, we only evaluated the interaction between ALLO and propofol on GABAA receptor-mediated synaptic events. The main finding of this study is that propofol enhanced the IPSC decay time double the amount in the presence of ALLO (Figure 7), indicating that ALLO increases the sensitivity of GABAA receptors for allosteric modulation by propofol. Our results are in line with the observation that the neurosteroid pregnanolone enhances the modulatory action of etomidate (0.1–10 μM), pentobarbital (100 μM), and propofol (5 μM) on GABA-evoked currents in HEK 293 cells.5 On the molecular level, the synergistic interaction between neurosteroids and anesthetic agents may involve a simultaneous binding on GABAA receptors. Previous studies found that the neurosteroid binding site on GABAA receptors differs from the binding sites for anesthetics6 and that neurosteroids enhance the affinity of muscimol,13 flunitrazepam,14 and etomidate5 for GABAA receptors.
We observed that ALLO and propofol applied together have a cumulative effect on the action potential firing, however, not a supra-additive effect. Thus, the synergistic action observed on the synaptic level did not manifest in network recordings. These data raise the following question: What is the reason for this discrepancy is? To provide an explanation, it is important to note that in this study, we obtained extracellular recordings of action potential activity and intracellular recordings of IPSCs under different experimental conditions. To isolate GABAA receptor-mediated IPSCs, we pharmacologically blocked the glutamatergic synaptic transmission. In neocortical slice cultures, the majority of IPSCs is generated by GABAergic interneurons without a glutamatergic excitation. Neither ALLO (100 nM) nor propofol (250 nM) decreased the frequency of IPSCs, indicating that the action potential firing of GABAergic interneurons remained unaffected. However, action potential activity was monitored during glutamatergic transmission. Under these conditions, GABAergic interneurons are excited by glutamatergic pyramidal cells. We hypothesize that enhancing GABAA receptor function decreases excitation of GABAergic interneurons by glutamatergic pyramidal cells and thereby reduces GABAA receptor-mediated inhibition. Thus, the supra-additive effect of ALLO and propofol on GABAA receptor-mediated IPSCs is masked by a decreasing excitatory drive onto GABAergic interneurons. It seems also possible that combined propofol and ALLO enhancement of tonic GABA currents caused a shunting of IPSCs, thus reducing their contribution to depressed pyramidal cell discharge frequency while simultaneously contributing to this depression.
Concentrations of propofol causing anesthesia in vivo are considered to be below 370 nM.15 In a previous study, we showed that 1 μM propofol reduces spontaneous action potential firing in organotypic neocortical slice cultures by 50% to 60%.9,11 Although a synergistic effect of ALLO and propofol did not manifest on the firing rates, coapplying these agents reduced the need for propofol to depress neuronal activity to “anesthetic levels” (ie, 50% in our test system) more than 2-fold. These findings suggest that a coapplication of neurosteroids and propofol may open new ways to substantially reduce anesthetic dose requirements, thereby reducing unwanted side effects, as previously suggested.6
Some compounds, based on their chemical structure of neurosteroids, were especially developed for anesthesia and sedation, eg, althesin16 or eltanolone.17 However, this approach was not successful because of a hypersensitivity caused by the solvent. Interestingly, there is growing interest in reintroducing neurosteroids into clinical anesthesia.18,19
On the behavioral level, ALLO signaling through GABAA receptors causes anxiolysis,20 sedation, loss of consciousness, and neuroprotection.1,18,21 Furthermore, there is evidence that neurosteroids protect immature neurons from damage.22
Thus, the combined use of neurosteroids and anesthetic agents could be beneficial for 2 reasons. First, this approach should reduce drug dosage and, thereby, minimize undesired side effects of anesthetic agents.5 Second, the neuroprotective and anti-inflammatory properties of neurosteroids may improve patient recovery.23
Name: Berthold Drexler, MD.
Contribution: This author helped design the study and write the manuscript.
Name: Monika Balk, MD.
Contribution: This author helped design the study, conduct the study, collect and analyze the data, and prepare the manuscript.
Name: Bernd Antkowiak, PhD.
Contribution: This author helped design the study, analyze the data, and prepare the manuscript.
This manuscript was handled by: Markus W. Hollmann, MD, PhD, DEAA.
1. Schumacher M, Mattern C, Ghoumari A, et al. Revisiting the roles of progesterone and allopregnanolone in the nervous system: resurgence of the progesterone receptors. Prog Neurobiol. 2014;113:6–39.
2. Seljeset S, Laverty D, Smart TG. Inhibitory neurosteroids and the GABAA receptor. Adv Pharmacol. 2015;72:165–187.
3. Puia G, Mienville JM, Matsumoto K, et al. On the putative physiological role of allopregnanolone on GABA(A) receptor function. Neuropharmacology. 2003;44:49–55.
4. Rupprecht R, Rammes G, Eser D, et al. Translocator protein (18 kD) as target for anxiolytics without benzodiazepine-like side effects. Science. 2009;325:490–493.
5. Li P, Bracamontes JR, Manion BD, et al. The neurosteroid 5β-pregnan-3α-ol-20-one enhances actions of etomidate as a positive allosteric modulator of α1β2γ2L GABAA receptors. Br J Pharmacol. 2014;171:5446–5457.
6. Li GD, Chiara DC, Cohen JB, Olsen RW. Neurosteroids allosterically modulate binding of the anesthetic etomidate to gamma-aminobutyric acid type A receptors. J Biol Chem. 2009;284:11771–11775.
7. Drexler B, Zinser S, Hentschke H, Antkowiak B. Diazepam decreases action potential firing of neocortical neurons via two distinct mechanisms. Anesth Analg. 2010;111:1394–1399.
8. Gähwiler BH. Organotypic monolayer cultures of nervous tissue. J Neurosci Methods. 1981;4:329–342.
9. Drexler B, Jurd R, Rudolph U, Antkowiak B. Distinct actions of etomidate and propofol at beta3-containing gamma-aminobutyric acid type A receptors. Neuropharmacology. 2009;57:446–455.
10. Hentschke H, Schwarz C, Antkowiak B. Neocortex is the major target of sedative concentrations of volatile anaesthetics: strong depression of firing rates and increase of GABAA receptor-mediated inhibition. Eur J Neurosci. 2005;21:93–102.
11. Antkowiak B. Different actions of general anesthetics on the firing patterns of neocortical neurons mediated by the GABA(A) receptor. Anesthesiology. 1999;91:500–511.
12. Harney SC, Frenguelli BG, Lambert JJ. Phosphorylation influences neurosteroid modulation of synaptic GABAA receptors in rat CA1 and dentate gyrus neurones. Neuropharmacology. 2003;45:873–883.
13. Harrison NL, Simmonds MA. Modulation of the GABA receptor complex by a steroid anaesthetic. Brain Res. 1984;323:287–292.
14. Harris BD, Wong G, Moody EJ, Skolnick P. Different subunit requirements for volatile and nonvolatile anesthetics at gamma-aminobutyric acid type A receptors. Mol Pharmacol. 1995;47:363–367.
15. Franks NP. General anaesthesia: from molecular targets to neuronal pathways of sleep and arousal. Nat Rev Neurosci. 2008;9:370–386.
16. Gyermek L, Soyka LF. Steroid anesthetics. Anesthesiology. 1975;42:331–344.
17. Tang J, Qi J, White PF, Wang B, Wender RH. Eltanolone as an alternative to propofol for ambulatory anesthesia. Anesth Analg. 1997;85:801–807.
18. Qian M, Krishnan K, Kudova E, et al. Neurosteroid analogues. 18. Structure-activity studies of ent-steroid potentiators of γ-aminobutyric acid type A receptors and comparison of their activities with those of alphaxalone and allopregnanolone. J Med Chem. 2014;57:171–190.
19. Monagle J, Siu L, Worrell J, Goodchild CS, Serrao JM. A phase 1c trial comparing the efficacy and safety of a new aqueous formulation of alphaxalone with propofol. Anesth Analg. 2015;121:914–924.
20. Rupprecht R. Neuroactive steroids: mechanisms of action and neuropsychopharmacological properties. Psychoneuroendocrinology. 2003;28:139–168.
21. Bruun DA, Cao Z, Inceoglu B, et al. Combined treatment with diazepam and allopregnanolone reverses tetramethylenedisulfotetramine (TETS)-induced calcium dysregulation in cultured neurons and protects TETS-intoxicated mice against lethal seizures. Neuropharmacology. 2015;95:332–342.
22. Peper JS, van den Heuvel MP, Mandl RC, Hulshoff Pol HE, van Honk J. Sex steroids and connectivity in the human brain: a review of neuroimaging studies. Psychoneuroendocrinology. 2011;36:1101–1113.
© 2016 International Anesthesia Research Society
23. Rupprecht R, Papadopoulos V, Rammes G, et al. Translocator protein (18 kDa) (TSPO) as a therapeutic target for neurological and psychiatric disorders. Nat Rev Drug Discov. 2010;9:971–988.