If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Pharmacology and Center for Biomedical Neuroscience, UT Health San Antonio, San Antonio, TexasAudie L. Murphy Memorial Veterans Hospital, South Texas Veterans Health Care System, San Antonio, Texas
Department of Pharmacology and Center for Biomedical Neuroscience, UT Health San Antonio, San Antonio, TexasAudie L. Murphy Memorial Veterans Hospital, South Texas Veterans Health Care System, San Antonio, Texas
Campbell Family Mental Health Research Institute, Centre for Addiction and Mental Health, Toronto, Ontario, CanadaDepartment of Psychiatry, University of Toronto, Toronto, Ontario, Canada
Department of Pharmacology and Center for Biomedical Neuroscience, UT Health San Antonio, San Antonio, TexasAudie L. Murphy Memorial Veterans Hospital, South Texas Veterans Health Care System, San Antonio, Texas
Campbell Family Mental Health Research Institute, Centre for Addiction and Mental Health, Toronto, Ontario, CanadaDepartment of Psychiatry, University of Toronto, Toronto, Ontario, CanadaDepartment of Pharmacology and Toxicology, University of Toronto, Toronto, Ontario, Canada
Department of Pharmacology and Center for Biomedical Neuroscience, UT Health San Antonio, San Antonio, TexasAudie L. Murphy Memorial Veterans Hospital, South Texas Veterans Health Care System, San Antonio, Texas
Aberrant dopamine neuron activity is attributable to hyperactivity in hippocampal subfields driving a pathological increase in dopamine neuron activity, which is positively correlated with psychosis in humans. Evidence indicates that hippocampal hyperactivity is due to loss of intrinsic GABAergic (gamma-aminobutyric acidergic) inhibition. We have previously demonstrated that hippocampal GABAergic neurotransmission can be modulated by targeting α5-GABAA receptors, which are preferentially expressed in hippocampal regions. Positive and negative allosteric modulators of α5-GABAA receptors (α5-PAMs and α5-NAMs) elicit effects on hippocampal-dependent behaviors. We posited that the selective manipulation of hippocampal inhibition, using α5-PAMs or α5-NAMs, would modulate dopamine activity in control rats. Further, α5-PAMs would reverse aberrant dopamine neuron activity in a rodent model with schizophrenia-related pathophysiologies (methylazoxymethanol acetate [MAM] model).
Methods
We performed in vivo extracellular recordings of ventral tegmental area dopamine neurons in anesthetized rats to compare the effects of two novel, selective α5-PAMs (GL-II-73, MP-III-022), a nonselective α-PAM (midazolam), and two selective α5-NAMs (L-655,708, TB 21007) in control and MAM-treated male Sprague Dawley rats (n = 5–9).
Results
Systemic or intracranial administration of selective α5-GABAA receptor modulators regulated dopamine activity. Specifically, both α5-NAMs increased dopamine neuron activity in control rats, whereas GL-II-73, MP-III-022, and L-655,708 attenuated aberrant dopamine neuron activity in MAM-treated rats, an effect mediated by the ventral hippocampus.
Conclusions
This study demonstrated that α5-GABAA receptor modulation can regulate dopamine neuron activity under control or abnormal activity, providing additional evidence that α5-PAMs and α5-NAMs may have therapeutic applications in psychosis and other psychiatric diseases where aberrant hippocampal activity is present.
The pathophysiology of schizophrenia is complex. It has long been hypothesized that positive symptoms (i.e., hallucinations and delusions) are mediated by hyperactivity of the mesolimbic dopamine system (
); however, no overt histopathology has been identified in dopamine neurons. Thus, aberrant dopamine system function is likely a consequence of disruptions in upstream brain regions. The hippocampus regulates the activity of dopamine neurons, its structure and function are altered in schizophrenia (
), together promoting the hippocampus as a putative region mediating aberrant dopamine activity in schizophrenia.
Rodent studies have repeatedly demonstrated that hippocampal hyperactivity drives aberrant dopamine system function via a multisynaptic pathway consisting of the nucleus accumbens, ventral pallidum, and ventral tegmental area (VTA) (
), neurosurgical, and cell-based approaches in rodent models is successful at restoring dopamine system function as well as behavioral correlates of positive, negative, and cognitive symptoms (
). Together, hippocampal hyperactivity may play a crucial role in the pathophysiology of schizophrenia, such that decreasing hippocampal activity may be a beneficial treatment.
Hippocampal hyperactivity is observed in individuals with schizophrenia as well as in rodent models and is thought to result from deficits in GABAergic (gamma-aminobutyric acidergic) inhibition (
). Prenatal exposure to methylazoxymethanol acetate (MAM) is frequently used in rodents to induce anatomical, physiological, and behavioral deficits that model those observed in schizophrenia (
), which can be reversed by transplanting into the hippocampus parvalbumin and somatostatin interneurons, derived from embryonic stem cells or fetal interneuron precursor cells (
). Thus, augmenting hippocampal interneuron function may represent a novel therapeutic modality for the symptoms of schizophrenia.
GABAergic control of hippocampal pyramidal neurons occurs via GABAA receptors (GABAARs) and GABAB receptors (GABABRs). GABAARs are pentameric, ionotropic chloride channels typically containing a combination of two α, two β, and one γ subunit (
). The subunit composition of GABAARs lends to unique expression patterns, with those containing the α5 subunit being highly expressed in the hippocampus, expressed to a much lesser extent in the cortex, and very minimally expressed in subcortical regions (
). The enriched distribution of α5-GABAARs in the hippocampus makes them an ideal target for the selective manipulation of hippocampal activity. Previous studies have demonstrated that knockdown of the α5-GABAARs in the hippocampus can produce deficits in behaviors associated with positive symptoms of schizophrenia, including latent inhibition and prepulse inhibition of startle (
). Conversely, overexpression of the α5 subunit of the GABAA receptor in the ventral hippocampus (vHipp) not only restores dopamine system function in MAM-treated rats, but also alleviates deficits in cognitive flexibility (
Modulation of extrasynaptic GABAA alpha 5 receptors in the ventral hippocampus normalizes physiological and behavioral deficits in a circuit specific manner.
). Further, systemic injection of a positive allosteric modulator selective for the α5 subunit of the GABAA receptor (α5-PAM) (SH-053-2′F-R-CH3) was effective at normalizing dopamine neuron activity and improved dopamine-dependent behaviors in MAM-treated rats (
). Here, we tested two novel selective α5-PAMs, GL-II-73 and MP-III-022. GL-II-73 has recently been developed and displays pro-cognitive, anxiolytic, and antidepressant-like effects in stressed and old mice (
Positive modulation of alpha5GABAA receptors leads to dichotomous effects in rats on memory pattern and GABRA5 expression in prefrontal cortex and hippocampus.
). Thus, GL-II-73 and MP-III-022 may be beneficial in treating the negative and cognitive symptom domains of schizophrenia, although their effects on dopamine system function have yet to be determined.
Negative allosteric modulators of α5-GABAARs (α5-NAMs) were originally developed to serve as cognitive enhancers; however, recent studies have demonstrated multiple uses for these compounds. Specifically, α5-NAMs can improve performance in hippocampal-dependent cognitive tasks (
An inverse agonist selective for alpha5 subunit-containing GABAA receptors improves encoding and recall but not consolidation in the Morris water maze.
Negative modulation of GABAA alpha5 receptors by RO4938581 attenuates discrete sub-chronic and early postnatal phencyclidine (PCP)-induced cognitive deficits in rats.
Selective pharmacological augmentation of hippocampal activity produces a sustained antidepressant-like response without abuse-related or psychotomimetic effects.
Rapid antidepressant action and restoration of excitatory synaptic strength after chronic stress by negative modulators of alpha5-containing GABAA receptors.
). This idea is supported by the observation that the rapid-acting antidepressant, ketamine, can acutely restore dopamine neuron activity and synaptic plasticity in the hippocampus in a rodent model of helplessness (
). Therefore, it stands that increasing hippocampal activity with an α5-NAM may normalize dopamine system function in models used to study depression.
We posited that α5-PAMs and α5-NAMs modulate dopamine neuron activity through their activity on α5-GABAARs in the hippocampus (Figure 1). Here, we performed in vivo extracellular recordings of dopamine neurons in the VTA to investigate the effects of both systemic administration and direct intrahippocampal microinjection of two novel selective α5-PAMs (GL-II-73 and MP-III-022), a nonselective α-PAM (midazolam), and two selective α5-NAMs (L-655,708 and TB 21007) in saline-treated control rats and MAM-treated rats. Gaining a better understanding of how α5-GABAA receptor modulation can regulate dopamine neuron activity is warranted, as such compounds have multiple potential therapeutic applications, including in schizophrenia and other psychiatric diseases where aberrant hippocampal activity is present.
Figure 1Under normal conditions, the VP exerts tonic inhibitory control of VTA dopamine neurons. Our hypothesis suggested that in the MAM model, increased hippocampal activity drives the NAc, which in turn inhibits the tonic activity of the VP. This results in a loss of GABAergic transmission from the VP to the VTA, which causes an increase in dopamine neuron population activity. NAMs of the α5 receptor cause an increase in hippocampal activity and mimic what is observed in models that display psychosis-related pathologies. In contrast, PAMs of the α5-GABAA receptor decrease aberrant vHipp activity, thus restoring dopamine system function in models with hippocampal hyperactivity and dopamine system dysfunction. GABAergic, gamma-aminobutyric acidergic; MAM, methylazoxymethanol acetate; NAc, nucleus accumbens; NAM, negative allosteric modulator; PAM, positive allosteric modulator; vHipp, ventral hippocampal; VP, ventral pallidum; VTA, ventral tegmental area.
All experiments were performed in accordance with the guidelines outlined in the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and the Use Committee of UT Health San Antonio.
Animals
To generate rodents displaying circuit-level alterations relevant to psychosis, we administered MAM as previously described (
). In brief, timed pregnant female Sprague Dawley rats were obtained from Envigo on gestational day 16 and injected with MAM (22 mg/kg diluted in saline via intraperitoneal route) or saline (1 mL/kg via intraperitoneal route) on gestational day 17. Male pups were weaned on postnatal day 21 and housed with littermates in groups of two or three. All experiments were performed on multiple litters of MAM- and saline-treated rats during adulthood (>8 weeks old; approximately 250–400 g). Male Sprague Dawley rats (>12 weeks old; 350–450 g) were obtained from Envigo and used to perform studies with pharmacological hyperactivation of the vHipp and respective control rats. The doses of PAMs and NAMs used for systemic administration were 1 mg/kg for midazolam , 10 mg/kg for GL-II-73 and MP-III-022, 3 mg/kg for L-655,708, and 0.3 mg/kg for TB 21007 and were administered approximately 20 minutes before in vivo electrophysiology. Doses were chosen based on previously published literature using these compounds (
Ester to amide substitution improves selectivity, efficacy and kinetic behavior of a benzodiazepine positive modulator of GABAA receptors containing the alpha5 subunit.
Rats were anesthetized with chloral hydrate via intraperitoneal injection before placement in a stereotaxic apparatus (Kopf). A core body temperature of 37 °C was maintained. This anesthetic (chloral hydrate) was used for examination of dopamine neuron physiology, as it does not significantly alter dopamine activity when compared to conscious rats (
). Supplemental anesthesia was administered as required to maintain suppression of the limb withdrawal reflex. Before dopamine neuron electrophysiology, untreated male Sprague Dawley rats were unilaterally injected into the vHipp (anteroposterior +5.3 mm and mediolateral ±5.3 mm from bregma; dorsoventral −7.0 mm ventral of the brain surface) with vehicle (0.5 μL Dulbecco’s phosphate-buffered saline) or NMDA (0.75 μg/0.5 μL), followed by an α5 modulator (GL-II-73, MP-III-022, midazolam, L-655,708, TB 21007; 0.75 μL of 100 ng/μL) or vehicle (0.5 μL; 50% dimethyl sulfoxide in distilled water or 1% Tween 80 [Sigma-Aldrich], 14% propylene glycol, and 85% distilled water) at a rate of approximately 0.5 μL/min. A subset of MAM- and saline-treated rats received an intra-vHipp injection of an α5 modulator or vehicle. Intra-vHipp injections specifically target the ventral CA1/subiculum region of the vHipp, as this region strongly regulates the mesolimbic dopamine system (
Organization of the projections from the subiculum to the ventral striatum in the rat. A study using anterograde transport of Phaseolus vulgaris leucoagglutinin.
). Dopamine neuron electrophysiology was performed starting approximately 20 minutes after microinjection. Hippocampal microinjections and dopamine recordings were performed bilaterally to minimize the number of experimental animals used.
Dopamine Neuron Electrophysiology
Extracellular glass microelectrodes (impedance 6–10 MΩ) were lowered into the VTA (anteroposterior −5.3 mm and mediolateral ±0.6 mm from bregma; dorsoventral −6.5 to −9.0 mm ventral of the brain surface). Six to 9 vertical passes were made throughout the cell body region of the VTA. Spontaneously active dopamine neurons were identified using the following previously established criteria (
): 1) action potential duration >2 ms and 2) frequency between 0.5 and 15 Hz. The following three parameters of dopamine activity were measured: 1) population activity (the number of spontaneously active dopamine neurons encountered per track); 2) basal firing rate; 3) the proportion of action potentials occurring in bursts (defined as the incidence of spikes with <80 ms between them; termination of the burst is defined by >180 ms between spikes). The same vehicle control rats were used in the PAM and NAM analyses.
Histological Verification
To verify electrode and cannula placement, rats were rapidly decapitated at completion of all experiments. Brains were extracted, fixed for at least 24 hours (4% formaldehyde in saline), and cryoprotected (10% w/v sucrose in phosphate-buffered saline) until saturated. Brains were coronally sectioned (25 μm) using a cryostat (Leica BioSystems). Sections containing electrode or cannula tracks were mounted onto gelatin-chrome alum-coated slides, stained with neutral red (0.1%) and thionin acetate (0.01%) and cover slipped with DPX Mountant (Sigma-Aldrich) for histochemical confirmation within the VTA (electrode) (Figure 2A) or vHipp (cannula) (Figure 2B) with reference to a stereotaxic atlas (
Figure 2(A) Representative brain slice with an electrode track (arrows) in the ventral tegmental area (left) and corresponding schematic of the brain section (right). Gray boxes indicate the area where cannula and electrode tracks were observed. (B) Representative brain slice with a cannula track (arrow) in the ventral hippocampus (left) and corresponding schematic of the brain section (right).
Data are represented as the mean ± SEM with n values representing the number of animals per experimental group unless otherwise stated. Statistical analyses were performed using SigmaPlot (Systat Software Inc.). Electrophysiological data were analyzed by two-way analysis of variance (strain × treatment) followed by the Holm-Sidak post hoc test. Significance was determined at p < .05.
Materials
MAM was purchased from MRIGlobal. Proprietary compounds, MP-III-022 and GL-II-73, were generated by the University of Wisconsin in Milwaukee and supplied by the Campbell Family Mental Health Research Institute, Centre for Addiction and Mental Health. L-655,708 and TB 21007 were purchased from Tocris. Midazolam hydrochloride was obtained from Akorn, Inc. Chloral hydrate, NMDA, propylene glycol, and dimethyl sulfoxide were obtained from Sigma-Aldrich.
Results
Systemic Administration of α5-PAMs and α5-NAMs Exerts Differential Effects on Dopamine Neuron Population Activity in MAM-Treated Rats
Rodent models used to study schizophrenia-related pathologies consistently display aberrant dopamine system function (
). As expected, we observed a significant increase, about a doubling, in dopamine neuron population activity (strain: F1,20 = 101.511, p < .001; treatment: F3,50 = 4.123, p = .012; interaction: F3,50 = 5.538, p = .003) in MAM-treated vehicle rats (n = 7, 1.97 ± 0.14) compared with saline-treated vehicle rats (n = 9, 0.94 ± 0.05; t = 7.53, p < .001) (Figure 3A). In saline-treated rats, systemic administration with α5-PAMs GL-II-73 (n = 6, 1.05 ± 0.06) and MP-III-022 (n = 6, 1.06 ± 0.08) and with the nonselective α-PAM midazolam (n = 6, 1.00 ± 0.04) failed to alter population activity (Figure 3A), firing rate (Figure S1A), or bursting (Figure S1B). Aberrant population activity in MAM-treated rats was significantly decreased by systemic injection of MP-III-022 (n = 6, 1.29 ± 0.08; t = 4.67, p < .001); however, GL-II-73 (n = 6, 1.96 ± 0.18) and midazolam (n = 6, 1.87 ± 0.13) had no effect (Figure 3A). No differences were observed in average firing rate and burst firing pattern between any of the groups treated with α5-PAMs.
Figure 3Dopamine neuron population activity can be modulated by systemic administration of positive and negative allosteric modulators of α5-GABAA receptors. (A) Population activity (average number of spontaneously active dopamine neurons per electrode track) is significantly higher in MAM-treated rats, which is reversed by systemic administration of the selective α5-GABAA positive allosteric modulator, MP-III-022. (B) Systemic administration of the α5-GABAA negative allosteric modulator, L-655,708, restored normal dopamine system function in MAM-treated rats and increased dopamine activity in saline-treated rats. In contrast, the α5-GABAA negative allosteric modulator, TB 21007, had no effect in saline-treated rats and only attenuated population activity in MAM-treated rats. ∗p < .05 denotes significance from respective saline-treated control; ∗∗∗p < .05 denotes significance from MAM-treated vehicle group; ∗∗p < .05 denotes significance from saline-treated vehicle group. GABAA, gamma-aminobutyric acid A; MAM, methylazoxymethanol acetate; MZ, midazolam.
Conversely, the α5-NAM, L-655,708, increased population activity (strain: F1,39 = 4.352, p = .045; interaction: F2,39 = 27.244, p < .001) in saline-treated rats (n = 6, 1.71 ± 0.16; t = 4.61, p < .001) and restored normal dopamine system function in MAM-treated rats (n = 6, 1.00 ± 0.06; t = 5.78, p < .001) (Figure 3B). Systemic administration of the α5-NAM, TB 21007, had no effect in saline-treated rats (n = 6, 1.24 ± 0.15), but attenuated activity in MAM-treated rats (n = 6, 1.56 ± 0.13; t = 2.48, p = .018). α5-NAMs did not alter firing rate or bursting in saline-treated rats; however, burst firing was significantly increased in MAM-treated rats (Figure S1C, D).
Intra-vHipp Microinjections of α5-PAMs and α5-NAMs Exert Effects Downstream on Dopamine Neuron Population in Saline- and MAM-Treated Rats
Saline-treated rats receiving intra-vHipp vehicle displayed a population activity of 1.01 ± 0.05 (n = 6), whereas MAM-treated rats receiving vehicle had a significantly higher population activity of 1.98 ± 0.12 (n = 6; t = 7.76, p < .001; strain: F1,42 = 48.211, p < .001; treatment: F3,42 = 6.379, p = .001; interaction: F3,42 = 10.446, p < .001) (Figure 4A). Direct intra-vHipp microinjections of α5-PAMs had no effect in saline-treated rats (GL-II-73: n = 5, 1.14 ± 0.06; MP-III-022: n = 6, 1.06 ± 0.06; midazolam: n = 5, 1.02 ± 0.13), consistent with systemic administration; however, GL-II-73 completely reversed the elevated population activity in MAM-treated rats (GL-II-73: n = 5, 1.10 ± 0.12; t = 6.681; p < .001) (Figure 4A). To a smaller extent, MP-III-022 and midazolam also significantly attenuated the population activity in MAM-treated rats (MP-III-022: n = 5, 1.64 ± 0.12; t = 2.597, p = .04; midazolam: n = 5, 1.37 ± 0.13; t = 4.664, p < .001). Only relatively small effects were observed on firing rate and bursting (Figure S2A, C).
Figure 4Direct intra–ventral hippocampal administration of positive and negative allosteric modulators of α5-GABAA receptors can attenuate dopamine neuron population activity in MAM-treated rats. (A) The α5-GABAA positive allosteric modulator, GL-II-73, completely restored dopamine system function, while MP-III-022 and MZ only attenuated activity in MAM-treated rats, with no effects in saline-treated rats. (B) Saline-treated rats displayed a significant increase in population activity following intra–ventral hippocampal administration of both α5-GABAA negative allosteric modulators, L-655,708 and TB 21007. Further, dopamine system function was restored in MAM-treated rats following L-655,708, and population activity was attenuated following TB 21007. ∗p < .05 denotes significance from respective saline-treated control; ∗∗∗p < .05 denotes significance from MAM-treated vehicle group; ∗∗p < .05 denotes significance from saline-treated vehicle group. GABAA, gamma-aminobutyric acid A; MAM, methylazoxymethanol acetate; MZ, midazolam.
Intra-vHipp microinjection of α5-NAMs (treatment: F1,31 = 3.424, p = .048; interaction: F5,74 = 31.362; p < .001) (Figure 4B) elicited a significant increase in population activity in saline-treated rats (L-655,708: n = 5, 1.66 ± 0.14; t = 4.514, p < .001; TB 21007: n = 5, 1.61 ± 0.09; t = 4.193, p < .001). Both α5-NAMs produced a significant reduction, albeit to a different extent, in population activity of MAM-treated rats (L-655,708: n = 5, 1.03 ± 0.12; t = 6.553, p < .001; TB 21007: n = 5, 1.64 ± 0.09; t = 2.368, p = .026). Only relatively small effects were observed on firing rate and bursting (Figure S2B, D).
Intra-vHipp Administration of α5-PAMs and α5-NAMs Alters Dopamine System Function After Pharmacological Activation of vHipp
Using a model of pharmacological hyperactivation of the vHipp, via infusion of NMDA, we observed a significant increase in dopamine neuron activity (n = 6, 2.03 ± 0.18; t = 5.55, p < .001) (Figure 5A) compared with vehicle (n = 6, 1.04 ± 0.10; strain: F1,47 = 3.869, p = .056; treatment: F3,47 = 4.967, p = .005; interaction: F3,47 = 18.594, p < .001). GL-II-73 caused a significant increase in population activity in vehicle rats (n = 6, 1.78 ± 0.12; t = 4.12, p < .001); that was not observed with MP-III-022 (n = 6, 1.08 ± 0.07) or midazolam (n = 6, 0.94 ± 0.06). Further, both GL-II-73 (n = 6, 0.97 ± 0.03; t = 5.94, p < .001) and MP-III-022 (n = 6, 1.08 ± 0.08; t = 5.35, p < .001) completely reversed the increase in population activity observed following NMDA activation of the vHipp, whereas midazolam had a modest effect (n = 6, 1.47 ± 0.24; t = 3.17, p = .012).
Figure 5(A) Rats with NMDA-induced hippocampal hyperactivity displayed a significant increase in dopamine neuron population activity, which was reversed by intra–ventral hippocampal administration of the specific α5-GABAA positive allosteric modulators, GL-II-73 and MP-III-022, and attenuated by MZ. Intra–ventral hippocampal GL-II-73 administration caused a significant increase in population activity in vehicle-treated control rats. (B) The α5-GABAA negative allosteric modulator, L-655,708, restored dopamine system function in rats with NMDA-induced hippocampal hyperactivity. ∗p < .05 denotes significance from respective saline-treated control; ∗∗∗p < .05 denotes significance from MAM-treated vehicle group; ∗∗p < .05 denotes significance from saline-treated vehicle group. GABAA, gamma-aminobutyric acid A; MAM, methylazoxymethanol acetate; MZ, midazolam.
No significant differences were observed following either α5-NAM in vehicle rats (L-655,708: n = 6, 1.38 ± 0.11; TB 21007: n = 6, 1.51 ± 0.14; strain: F1,35 = 5.733, p = .023; treatment: F2,35 = 4.247, p = .024; interaction F2,35 = 8.474, p = .001) (Figure 5B). L-655,708 caused a significant decrease in population activity (n = 6, 1.03 ± 0.09; t = 4.357, p < .001) following NMDA-induced activation of the vHipp, but no change was observed after administration of TB 21007 (n = 6, 1.82 ± 0.28) (Figure 5B). No differences in firing rate or burst firing were observed (Figure S3).
Discussion
There is increasing evidence that aberrant dopamine system function observed in individuals with schizophrenia may be secondary to hippocampal hyperactivity (
). Therefore, exploring targets within the hippocampus may represent innovative and efficacious approaches for the treatment of schizophrenia and provide an effective treatment without the side effects observed with conventional antipsychotics (
). Here, we provide evidence that selective pharmacological modulation of vHipp activity, using novel selective α5-GABAA receptor modulators, can reduce aberrant dopamine neuron activity observed in a model used to study schizophrenia-related pathologies.
Previous studies have demonstrated schizophrenia-like deficits following knockdown of α5-GABAARs, specifically in the hippocampus (
), whereas overexpression of the GABAA receptor α5 subunit can reverse deficits in dopamine system function commonly observed in rodent models used to study schizophrenia-related pathologies (
Modulation of extrasynaptic GABAA alpha 5 receptors in the ventral hippocampus normalizes physiological and behavioral deficits in a circuit specific manner.
). Previous studies showed that systemic injection of the selective α5-PAM, SH-053-2′F-R-CH3, normalized dopamine system function and improved dopamine-dependent behaviors; however, it was ineffective at resolving behaviors associated with negative symptoms and cognitive deficits (
). In contrast, the selective α5-PAM used in the present study, GL-II-73, has been established as an anxiolytic, antidepressant, and pro-cognitive compound, thus demonstrating therapeutic potential for the treatment of negative symptoms and cognitive decline associated with schizophrenia (
). As shown in this study, when administered directly into the vHipp, GL-II-73 can reverse aberrant VTA dopamine neuron activity implicated in schizophrenia (
MAM-treated rats exhibit anatomical, physiological, and behavioral deficits consistent with those observed in schizophrenia. In these rats, systemic administration of the selective α5-PAM, GL-II-73, did not change dopamine neuron population activity, although MP-III-022 successfully restored dopamine system function. By contrast, after intra-vHipp administration, both GL-II-73 and MP-III-022 were able to modulate dopamine neuron population activity. The difference in effect of GL-II-73 following these two routes of administration is likely associated with its pharmacokinetic profile. GL-II-73 undergoes significant first-pass metabolism in rats (data not shown), resulting in insufficient brain concentrations to produce a detectable effect on population activity. This is specific for rats, as previously published literature demonstrated effects of GL-II-73 in mice when administered systemically (
). Intra-vHipp administration of both GL-II-73 and MP-III-022 reduced the elevated dopamine neuron population activity in MAM-treated rats; however, only GL-II-73 restored dopamine neuron population activity back to control levels, whereas MP-III-022 attenuated it. We suspect the difference in the magnitude of the effect between the α5-selective GABAA receptor PAMs stems from their differential affinities for the α5 subunit. MP-III-022 has a greater binding affinity and functional potentiation efficacy at α5-GABAA receptor compared with GL-II-73 but displays an inverted U–shaped dose response relationship (
Positive modulation of alpha5GABAA receptors leads to dichotomous effects in rats on memory pattern and GABRA5 expression in prefrontal cortex and hippocampus.
), which could underlie the results detailed here. In addition, midazolam, a nonspecific α-PAM, was able to attenuate elevated population activity only when given directly into the vHipp and had no effect when administered systemically. Given the widespread effects of benzodiazepines, it is possible that activation of non-α5 receptors opposes the beneficial effects of activation of α5-GABAARs and could reflect differences between synaptic and extrasynaptic GABAARs (
). Indeed, the dichotomy between extrasynaptic and synaptic GABAARs has been highlighted in previous studies, in which overexpression of extrasynaptic (α5) receptors in the vHipp restored dopamine system function in MAM-treated rats, whereas overexpression of synaptic (α1) receptors did not (
Modulation of extrasynaptic GABAA alpha 5 receptors in the ventral hippocampus normalizes physiological and behavioral deficits in a circuit specific manner.
), and recent work has demonstrated that the localization of α5-GABAARs is dynamic, such that changes in neuronal activity can cause receptors to shift between the synaptic and extrasynaptic spaces, altering their contribution to tonic and phasic inhibition (
Interestingly, α5-NAMs have been shown to produce similar behavioral effects (i.e., pro-cognitive and antidepressant-like) to those observed with α5-PAMs under different experimental conditions (
Selective pharmacological augmentation of hippocampal activity produces a sustained antidepressant-like response without abuse-related or psychotomimetic effects.
). Specifically, α5-PAMs produce antidepressant-like effects acutely, whereas the effect of α5-NAMs appear at 24 hours following administration. The reason for this is that α5 modulators restore the excitatory signal-to-noise ratio, albeit by different mechanisms. The α5-PAMs are thought to decrease noise, while the disinhibition produced by α5-NAMs leads to a glutamatergic surge and an augmented signal. Both can support appropriate processing of information, leading to improvements in behavior and reduced symptoms (
). It has been suggested that a delicate balance of excitation and inhibition is necessary for optimal signal transduction and that alterations to α5-GABAA receptor signaling in either direction may negatively impact cognition (
). Therefore, depending on the pathology, either an NAM or a PAM may be the most appropriate therapeutic intervention to influence dopamine neuron activity. Indeed, in conditions of hippocampal hyperfunction such as schizophrenia (
)], an NAM may be a more effective therapeutic to exert vHipp control over dopamine neuron activity in the VTA. However, depression is also associated with reduced cortical inhibition, in which case α5-PAMs seem to provide more beneficial effects than α5-NAMs, in particular on cognition (
), we reasoned that α5-NAMs should increase dopamine neuron population activity in control rats. Indeed, when given either systemically or intra-vHipp, L-655,708 increased dopamine neuron population activity in saline-treated control rats, providing further evidence that activation of the vHipp can drive the increases in dopamine neuron population activity. Further, L-655,708, but not TB 21007, decreased population activity in MAM-treated rats. It is unclear why L-655,708 would reverse aberrant dopamine neuron activity in MAM-treated rats, while increasing it in saline-treated rats.
Given the potential for other disruptions associated with prenatal MAM treatment [for review, see (
)], we employed a simplified model where vHipp activity was enhanced pharmacologically by direct microinjection of NMDA into the vHipp. Consistent with previous reports (
). As expected, intra-vHipp administration of α5-NAMs increased dopamine neuron population activity in control rats receiving intra-vHipp vehicle, albeit to a lesser extent than seen in the control rats for the MAM experiments. This is likely associated with the multiple injections into the vHipp, i.e., both NMDA and the α5-NAMs. In NMDA-treated rats, L-655,708 decreased population activity. No significant effects were present with TB 21007 administration. This mirrored our results in MAM-treated rats, suggesting that in conditions of vHipp hyperactivity, L-655,708, but not TB 21007, is able to restore dopamine system function. The ability of L-655,708 to reverse aberrant dopamine neuron activity was unexpected, but may be due to inherent differences between L-655,708 and TB 21007. L-655,708 has much higher selectivity for α5-GABAA receptor over other GABAARs in terms of affinity compared with TB 21007 (50–100 times and 10 times, respectively). Despite this difference in affinity, TB 21007 is still functionally selective for α5-GABAARs based on efficacy (
). It is possible that in the context of high hippocampal activity, as seen in MAM-treated or intra-vHipp NMDA-treated rats, these otherwise subtle differences are accentuated, leading to the unanticipated effects of L-655,708 in these groups. The important point then is that as a class of drugs, individual α5-NAMs may not all act the same way.
The potential beneficial effects of α5-PAMs seen in MAM-treated rats were also examined in the NMDA-induced model of hippocampal hyperactivity. These data replicated data observed in MAM-treated rats where intra-vHipp GL-II-73 and MP-III-022 completely reversed aberrant dopamine neuron population activity and midazolam partially restored dopamine system function. These results reinforce the idea that selectively augmenting α5-GABAA receptor activity in the vHipp can reverse aberrant dopamine neuron activity associated with hippocampal activation. Surprisingly, GL-II-73 caused a significant increase in population activity in vehicle-treated rats. While this effect was robust, we believe it to be spurious, as intra-vHipp injection of GL-II-73 did not increase population activity in control rats in our earlier experiment.
In summary, we have demonstrated that pharmacological manipulation of activity in the vHipp with α5-PAMs and α5-NAMs differentially modulates dopamine neuron population activity in the VTA. Specifically, under conditions of hippocampal hyperactivity, aberrant dopamine system function can be reversed by the administration of α5-PAMs. By demonstrating that multiple α5-PAMs can reverse aberrant dopamine neuron activity, this study provides evidence for potential antipsychotic properties of this class of compounds. Furthermore, it adds schizophrenia as well as other conditions in which hippocampal hyperactivity is present to the growing list of psychiatric disorders for which these compounds may have therapeutic utility. Moreover, α5-PAMs offer added benefit over traditional anxiolytic medications (nonspecific GABAAR PAMs), as their preference for α5 minimizes the undesired side effect of sedation, typically associated with the α1 subunit (
). Specifically, the α5-PAM GL-II-73 can reverse pathological increases in dopamine neuron population activity in the MAM model as well as stress-associated cognitive deficits, anxiety, and some depressive-like behaviors (
), suggesting that this compound may address affective and cognitive deficits observed in schizophrenia in addition to dopamine dysfunction. Indeed, future studies will test the utility of α5-PAMs in other rodent models that display psychosis-related pathologies. Furthermore, we have demonstrated that α5-NAMs can significantly increase baseline dopamine neuron activity in control rats suggesting that these might be a therapeutic approach for conditions associated with dopamine hypofunction. Taken together, we have demonstrated robust hippocampal modulation of dopamine system function by α5-GABAA receptor allosteric modulators, which may provide a beneficial approach for the treatment of several psychiatric disorders.
Acknowledgments and Disclosures
This work was supported by the Campbell Family Mental Health Research Institute of the Centre for Addiction and Mental Health (to ES), U.S. Department of Veterans Affairs Biomedical Laboratory Research and Development Service Merit Awards (Grant Nos. BX004693 and BX004646 [to DJL]), and the National Institutes of Health (Grant No. RO1-MH090067 [to DJL]).
ES, JMC, and TDP are coinventors or listed on U.S. patent applications that cover GABAergic ligands and their use in brain disorders. ES is cofounder of DAMONA Pharmaceuticals, a biopharmaceutical company dedicated to treatment of cognitive deficits in brain disorders. DJL and AF are coinventors of a patent application covering novel analgesics, and DJL has an active collaboration with Sosei-Heptares; these are both unrelated to the work described in the current article. All other authors report no biomedical financial interests or potential conflicts of interest.
Modulation of extrasynaptic GABAA alpha 5 receptors in the ventral hippocampus normalizes physiological and behavioral deficits in a circuit specific manner.
Positive modulation of alpha5GABAA receptors leads to dichotomous effects in rats on memory pattern and GABRA5 expression in prefrontal cortex and hippocampus.
An inverse agonist selective for alpha5 subunit-containing GABAA receptors improves encoding and recall but not consolidation in the Morris water maze.
Negative modulation of GABAA alpha5 receptors by RO4938581 attenuates discrete sub-chronic and early postnatal phencyclidine (PCP)-induced cognitive deficits in rats.
Selective pharmacological augmentation of hippocampal activity produces a sustained antidepressant-like response without abuse-related or psychotomimetic effects.
Rapid antidepressant action and restoration of excitatory synaptic strength after chronic stress by negative modulators of alpha5-containing GABAA receptors.
Ester to amide substitution improves selectivity, efficacy and kinetic behavior of a benzodiazepine positive modulator of GABAA receptors containing the alpha5 subunit.
Organization of the projections from the subiculum to the ventral striatum in the rat. A study using anterograde transport of Phaseolus vulgaris leucoagglutinin.
00003-9&id=gr1.jpg){kind=link}
00003-9&id=gr2.jpg){kind=link}
00003-9&id=gr3.jpg){kind=link}
00003-9&id=gr4.jpg){kind=link}
00003-9&id=gr5.jpg){kind=link}