The effects of these steroids namely

The effects of these steroids, namely the decrease in the peak and acceleration of desensitization could be observed in IGly activated by 100 μM glycine (concentration close to EC50). However, with an increase in glycine to 500 μM (a concentration close to saturating), the effect of steroids on the IGly peak completely disappeared, while the effect on desensitization remained with the same efficiency (Fig. 4). These results suggest the existence of two different mechanisms of action, which are apparently implemented via two different sites on GlyR. According to the literature, the effects of steroids on the GlyR have been described in various models: cultures of chick spinal cord I-BET151 mg (Wu et al., 1990, 1997), cultures of rat spinal dorsal horn neurons (Jiang et al., 2006), recombinant GlyR expressed in Xenopus oocytes (Maksay et al., 2001; Jin et al., 2009; Weir et al., 2004), and cultured rat hippocampal neurons (Jiang et al., 2009). All these studies (except Weir et al., 2004) describe a decrease in the peak amplitude of IGly, with IC50 values of 1–16 μM. It has been shown that the mechanism of inhibition of the IGly peak by steroids can be competitive (Jiang et al., 2006; Maksay et al., 2001) or non-competitive (Jiang et al., 2009). We have not found any literature on the effect of steroids on the desensitization of IGly. Our work is the first to describe the effect of steroids on the desensitization of the IGly of rat hippocampal pyramidal neurons and to show that this effect is stronger than on the peak amplitude of IGly. The mechanism of the complex effect of steroids on GlyRs remains to be elucidated. Our results suggest that steroids interact with at least two independent sites on GlyR, one of which regulates the peak current amplitude, and the second regulates desensitization. The site responsible for the decrease in peak amplitude is presumably located near the binding site of the agonist, as the effect is sensitive to the concentration of agonist. However, the effect on desensitization is insensitive to agonist concentration. In our opinion, the binding of steroids with a desensitization gate in a pore of the chloride channel (Gielen et al., 2015; Keramidas and Lynch, 2013) is the simplest explanation of the acceleration of desensitization. Interestingly, a similar complex profile of action on the GlyR of some other substances, namely, beta amyloid peptide, cyclic nucleotides and lithium ions was described in our previous work (Bukanova et al., 2014a, 2014b; 2016; Solntseva et al., 2016). It might be possible that neurosteroids and some other substances can interact with the same nonspecific sites on GlyR. A comparison of our results with the literature shows that the effect of pregnane steroids on GlyRs and NMDARs differs. In contrast to IGly, pregnane derivatives barely affected the INMDA peak amplitude. However, they significantly accelerated the desensitization of the INMDA (Borovska et al., 2012; Vyklicky et al., 2015). The authors explain this effect by slow steroid diffusion to the site of action at the extracellular vestibule of the NMDAR. The physiological significance of this finding may lie in a more potent suppression of tonically activated extrasynaptic NMDARs than phasically activated synaptic NMDARs. A comparison of our results with literature data on the action of pregnane and androstane steroids on the functional activity of NMDAR, GABAAR and GlyR enables us to draw the following conclusions. First, androstane steroids (with no substituents at C-17) at a concentration up to 50 μM do not affect the IGABA and IGly of rat hippocampal pyramidal neurons. At the same time, electrophysiological experiments on NMDAR expressed in HEK293 cells show an ability of these steroids to inhibit NMDAR-mediated currents at low micromolar concentrations (Adla et al., 2017). Therefore, androstane steroids could be considered to be selective NMDAR antagonists. Second, such pregnane steroids as PA-Glu, PA-hSuc, and PA-hPim affected the IGABA and IGly of rat hippocampal neurons with IC50 values in the range of 3–12 μM, and these values are close to previously published steroid effects on the NMDAR current (Vyklicky et al., 2016). Thus, these new pregnane steroids cannot be considered to be selective NMDAR antagonists.