Ketamine and ketamine metabolites as novel estrogen receptor ligands: Induction of cytochrome P450 and AMPA glutamate receptor gene expression

Abstract

Major depressive disorder (MDD) is the most common psychiatric illness worldwide, and it displays a striking sex-dependent difference in incidence, with two thirds of MDD patients being women. Ketamine treatment can produce rapid antidepressant effects in MDD patients, effects that are mediated-at least partially-through glutamatergic neurotransmission. Two active metabolites of ketamine, (2R,6R)-hydroxynorketamine (HNK) and (2S,6S)-HNK, also appear to play a key role in ketamine's rapid antidepressant effects through the activation of α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) glutamate receptors. In the present study, we demonstrated that estrogen plus ketamine or estrogen plus active ketamine metabolites displayed additive effects on the induction of the expression of AMPA receptor subunits. In parallel, the expression of estrogen receptor alpha (ERα) was also significantly upregulated. Even more striking, radioligand binding assays demonstrated that [³H]-ketamine can directly bind to ERα (K_D: 344.5 ± 13 nM). Furthermore, ketamine and its (2R,6R)-HNK and (2S,6S)-HNK metabolites displayed similar affinity for ERα (IC₅₀: 2.31 ± 0.1, 3.40 ± 0.2, and 3.53 ± 0.2 µM, respectively) as determined by [³H]-ketamine displacement assays. Finally, induction of AMPA receptors by either estrogens or ketamine and its metabolites was lost when ERα was knocked down or silenced pharmacologically. These results suggest a positive feedback loop by which estrogens can augment the effects of ketamine and its (2R,6R)-HNK and (2S,6S)-HNK metabolites on the ERα-induced transcription of CYP2A6 and CYP2B6, estrogen inducible enzymes that catalyze ketamine's biotransformation to form the two active metabolites. These observations provide novel insight into ketamine's molecular mechanism(s) of action and have potential implications for the treatment of MDD.

Keywords: AMPA receptors; CYP2A6; CYP2B6; Estrogens; Ketamine; Ketamine metabolites.

Figure 1

Schematic diagram illustrating ketamine mechanism(s) of action. CYP2A6 and CYP2B6 are major enzymes responsible for ketamine metabolism. Ketamine is an NMDA receptor (NMDAR) antagonist and the active ketamine metabolites (2R,6R-HNK and 2S,6S-HNK) also reportedly contribute to antidepressant effects through the activation of AMPA receptors (AMPRs). Ketamine induces glutamate release. Glutamate in turn binds to AMPARs to induce depolarization as well as sodium (Na²⁺) and calcium (Ca²⁺) influx through the AMPA receptors and L-type voltage gated calcium channels (VDCCs), respectively. This results in increased brain-derived neurotropic factor (BDNF) release from synaptic vesicles. BDNF binds to tyrosine kinase receptor B (TrkB) with high affinity, which leads to activation of both the ERK and AKT/mTOR pathways, increasing synaptic protein translation and GRIA1 trafficking to the cell membrane.

Figure 2

Effects on the expression of GRIA1, GRIA2, GRIA3 and GRIA4 in U251-MG cells in response to various treatment conditions, including exposure to increasing concentrations of (A) E2 (0.001-0.1 nM) and (B) ketamine or the active ketamine metabolites (2R,6R)-HNK or (2S,6S)-HNK (0-400 nM) alone or ketamine (400 nM) together with increasing concentration of E2 (0.0001-0.1 nM). ANOVA was performed to compare gene expression, followed by Tukey’s multiple comparisons when significant effects were detected. *p ≤0.05, **p ≤0.005 as compared to vehicle treatment. + p ≤0.05 as compared to the same treatment with or without E2.

Figure 3

Effects on the expression of AMPA receptor subunits GRIA1, GRIA2, and GRIA4 in primary human astrocytes. ANOVA was performed to compare gene expression, followed by Tukey’s multiple comparisons when significant effects were detected. *p ≤0.05, as compared to vehicle treatment. + p ≤0.05 as compared to the same treatment with or without E2.

Figure 4

(A) ERα mRNA expression was determined by qPCR 24 hours after U251-MG cells were exposed to increasing concentrations of ketamine, (2R,6R)-HNK, or (2S,6S)-HNK. E2 plus ketamine or E2 plus the active metabolites resulted in additive effects on the induction of ERα expression. (B), U251-MG cells were transfected with the ERE dual luciferase reporter constructs, and were exposed to increasing concentrations of ketamine, (2R,6R)-HNK, or (2S,6S)-HNK alone or in combination with increasing concentrations of E2. ANOVA was performed to compare gene expression, followed by Tukey’s multiple comparisons. *p ≤0.05, **p ≤0.005 as compared to vehicle treatment. + p ≤0.05 as compared to the same with or without E2.

Figure 5

Luciferase reporter assays in U251-MG cells expressing either CYP2A6 or CYP2B6. (A) Schematic representations of the promoters for CYP2A6 and CYP2B6 with the locations of ERE motifs indicated. Tss represents the site of transcription initiation. Cells were transfected with the dual luciferase reporter constructs and treated with increasing concentrations of E2 for 24 hours. (B) Transcriptional activities for both the CYP2A6 and CYP2B6 promoter constructs could be induced by E2 in a concentration-dependent fashion. Results shown are average values for three independent experiments (*p<0.0001 when compared with the vehicle control). (C) U251-MG Cells and (D) HepaRG cells were transfected with the dual luciferase reporter constructs and were treated with increasing concentrations of ketamine, (2R,6R)-HNK or (2S,6S)-HNK and the three compounds plus increasing concentrations of E2. All three compounds significantly induced transcriptional activities for CYP2A6 or CYP2B6 reporter constructs in a dose-dependent fashion (*p<0.005, **p<0.0005 vs vehicle treatment). + p ≤0.05 as compared to the same with or without E2. Results are presented as fold change in relative luciferase units compared with the pGL3 basic vector. ANOVA was performed to compare luciferase activities, followed by Tukey’s multiple comparison tests for individual comparisons when significant effects were detected.

Figure 6

Effects on the expression of CYP2A6 and CYP2B6 in primary human astrocytes (A) or HepaRG cells (B) in response to various treatment conditions. ANOVA was performed to compare gene expression, followed by Tukey’s multiple comparisons when significant effects were detected. *p <0.05, as compared to vehicle treatment. + p ≤0.05 as compared to the same treatment with or without E2.

Figure 7

(A) Induction of both CYP2A6 and CYP2B6 expression by ketamine and the ketamine metabolites (2R,6R)-HNK, or (2S,6S)-HNK was lost when ERα was silenced by the use of siRNA. *p ≤0.005 as compared to control siRNA. (B) Induction of both CYP2A6 and CYP2B6 expression by ketamine and the ketamine metabolites (2R,6R)-HNK, or (2S,6S)-HNK was lost when ERα was blocked using ICI, an estrogen receptor blocker.

Figure 8

Induction of GRIA1, GRIA2, GRIA3 and GRIA4 expression by ketamine and the ketamine was lost when ERα was silenced by the use of siRNA in both U138-MG and U251-MG cells.

Figure 9

Induction of the expression of GRIA1, GRIA2 and GRIA4 by ketamine and ketamine metabolites was lost after ERα blockade by fulvestant (ICI). Specifically, the expression of AMPA receptors is known to be enhanced in response to ketamine therapy. However, we observed that GRIA1, GRIA2 and GRIA4 were not induced by increasing concentrations of ketamine or its metabolites (100, 200, 400 nM) after ER blockade. Values shown are mean +/− SEM of three independent determinations.

Figure 10

(A) SPR sensorgrams showing the dose-dependent binding of ketamine (52 – 823 nM) to ERα, as indicated by relative response units (RUs) after background subtraction. (B) Plot of RU_max versus ketamine concentration. Inset, Scatchard plot of ketamine binding.

Figure 11

(A) Saturation curves for the binding of increasing concentrations of [³H]-ketamine to ERα (K_D=344.5 ± 13 nM). (B) [³H]-Ketamine competition binding curves. Data are expressed as percentages of specific binding of [³H]-ketamine vs. log of the competitor concentration (non-radioactive drug concentrations ranged from 0.4 nM to 400 μM). Each point represents the mean±SEM of three independent determinations.

Figure 12

(A) Saturation curves for the binding of increasing concentrations of [³H]-estradiol to ERα (K_D=3.67 ± 0.56 nM). (B) [³H]-estradiol competition binding curves. Data are expressed as percentages of specific binding of [³H]-estradiol vs. log of the competitor concentration (non-radioactive E2 concentrations ranged from 0.0002 nM to 20 μM, concentrations of ketamine and 4nM to 40mM for the two metabolites. Each point represents mean±SEM of three independent determinations.

Figure 13

Immunofluorescence staining of U251-MG cells showing ERα nuclear translocation after the cells were treated with E2 (0.1 nM), or with 400 nM ketamine or its active metabolites.