An increasing body of evidence

An increasing body of evidence demonstrates that MXC can cause neurobehavioral and neuropathological alterations [7]. In the brain, there are several steroidogenic enzymes, which are responsible for neurosteroid production. Neurosteroids are subsets of steroids that rapidly change neuronal excitability via binding the ligand-gated ion channels, such as GABA-A receptor [8]. Two neurosteroids, allopregnanolone and 3α,5α-androstanediol (DIOL), are potent GABA-A receptor agonists [9]. These neurosteroids may regulate various functions in the nerve system. For example, increased allopregnanolone level significantly induces analgesia for neuropathic pain [10]. Allopregnanolone has potent anticonvulsant activity in the adult mouse bimatoprost [8]. DIOL is also a positive modulator of GABA-A receptors to protect against the seizures induced by hippocampus kindling [11]. The levels of allopregnanolone and DIOL are regulated by two separate enzymes that have oppositely catalytic directions: 3α-hydroxysteroid dehydrogenase (AKR1C14, encoded by Akr1c14) and retinol dehydrogenase 2 (RDH2, encoded by Rdh2) in rats. AKR1C14 is a cytosolic NADPH-dependent enzyme [12], which primarily catalyzes the formation of allopregnanolone or DIOL from dihydroprogesterone and dihydrotestosterone (DHT), respectively (Supplementary Fig. S1). AKR1C14 belongs to the aldo-keto reductase family and adds a hydrogen to the 3α-position of many steroids, including neurosteroids [13]. RDH2 is a microsomal NAD+-dependent enzyme, which catalyzes the opposite direction. RDH2 belongs to the short-chain dehydrogenase/reductase family and removes a hydrogen from 3α-position of many steroids [14]. The brain contains both AKR1C14 and RDH2 [[15], [16], [17]]. Since MXC and HPTE directly suppress many steroidogenic enzyme activities [[4], [5], [6],18], in the present study, we investigated their direct effects on AKR1C14 and RDH2 enzyme activities.
Methods
Results
Discussion MXC can be metabolically activated into HPTE, which exerts many toxic effects [28]. Indeed, HPTE is more potent than MXC to inhibit many steroidogenic enzymes, including human testicular 3β-hydroxysteroid dehydrogenase 2, rat 11β-hydroxysteroid dehydrogenase 1, human and rat 11β-hydroxysteroid dehydrogenase 2, and human and rat 17β-hydroxysteroid dehydrogenase 3 [5,18]. Interestingly, HPTE was also more potent to bind to estrogen receptor α than MXC [29]. The homeostasis of neurosteroids such as allopregnanolone and DIOL depends on the catalysis of AKR1C14 and RDH2. Interestingly, these two enzymes are present in the different subcellular regions, with AKR1C14 in the cytoplasm and RDH2 in the smooth endoplasmic reticulum (microsome), and they use different cofactors, with AKR1C14 of NADPH and RDH2 of NAD+ [19]. Therefore, the cofactor availability determines the catalytic direction. The present study found that HPTE was more potent to inhibit AKR1C14 (IC50 = 2.602 μM) than RDH2 (IC50 = 20.473 μM). This indicates that HPTE suppresses the accumulation of neurosteroids. The IC50 were measured at 1000 nM steroid (which was within the Km range) in the present study. However, the brain neurosteroid concentrations could be much lower, as in the case of rat brain allopregnanolone concentration was around 1 ng/g (3 nM). In this regard, this possibly leads to nanomolar range of IC50 value of HPTE to inhibit AKR1C14, since HPTE is a competitive inhibitor. HPTE inhibited AKR1C14 and RDH2 in a competitive manner when a steroid was used. Interestingly, HPTE inhibited other steroidogenic enzymes such human and rat 11β-hydroxysteroid dehydrogenase 1 and 17β-hydroxysteroid dehydrogenase 3 competitively, when the steroid substrate was used [5]. These results indicate that hydroxyl group of HPTE binds to the active sites of these steroidogenic enzymes. Both AKR1C14 and RDH2 have two substrate binding sites, the steroid binding and the cofactor binding. Both MXC and HPTE inhibited rat AKR1C14 in a competitive mode against DHT. Indeed, using a docking algorithm based on the docking software EADock DSS [26], we found that HPTE bound the steroid binding site with high affinity (−7.43 Kcal) because its energy was lower than MXC (−7.1 Kcal). This indicates that HPTE competes with DHT more potently than MXC and the inhibition is reversible.