Pro-neurogenic, Memory-Enhancing and Anti-stress Effects of DF302, a Novel Fluorine Gamma-Carboline Derivative with Multi-target Mechanism of Action
A comparative study performed in mice investigating the action of DF302, a novel fluoride-containing gamma-carboline derivative, in comparison to the structurally similar neuroprotective drug dimebon. Drug effects on learning and memory, emotionality, hippocampal neurogenesis and mitochondrial functions, as well as AMPA-mediated currents and the 5-HT6 receptor are reported. In the step-down avoidance and fear-conditioning paradigms, bolus administration of drugs at doses of 10 or 40 mg/kg showed that only the higher dose of DF302 improved long-term memory while dimebon was ineffective at either dosage. Short-term memory and fear extinction remained unaltered across treatment groups. During the 5-day predation stress paradigm, oral drug treatment over a period of 2 weeks at the higher dosage regimen decreased anxiety-like behaviour. Both compounds supressed inter-male aggression in CD1 mice, the most eminent being the effects of DF302 in its highest dose. DF302 at the higher dose decreased floating behaviour in a 2-day swim test and after 21-day ultrasound stress. The density of Ki67-positive cells, a marker of adult neurogenesis, was reduced in the dentate gyrus of stressed dimebon-treated and non-treated mice, but not in DF302-treated mice. Non-stressed mice that received DF302 had a higher density of Ki67-positive cells than controls unlike dimebon-treated mice. Similar to dimebon, DF302 effectively potentiated AMPA receptor-mediated currents, bound to the 5-HT6 receptor, inhibited mitochondrial permeability transition and displayed cytoprotective properties in cellular models of neurodegeneration. Thus, DF302 exerts multi-target effects on the key mechanisms of neurodegenerative pathologies and can be considered as an optimized novel analogue of the neuroprotective agent dimebon.
KeywordsAlzheimer’s disease Multi-target mechanisms Hippocampal plasticity AMPA receptor 5-HT6 receptor Stress and depression Aggression
Alzheimer’s disease (AD) is a progressive neurodegenerative disorder with increasing prevalence [1, 2]. Its core feature is dementia which together with behavioural symptoms, such as aggression, anxiety, depression and psychosis, represents a serious burden for patients and care takers. Treating of AD patients is challenging and while enormous efforts and resources have been invested in a search for new remedies during the past decades, new drugs have not yet reached the market [3, 4, 5].
One of the promising approaches for pharmacological management of AD is believed to be the development of multi-target compounds [6, 7, 8]. The use of these types of drugs is based on the understanding of AD as a multifactorial disorder, in which numerous biological processes are impaired [9, 10]. Several multi-functional drugs, in which pharmacophores simultaneously affect multiple targets, such as AChE and M2 receptors , AChE and central MAO-A/B  and other combinations of targets [13, 14], have been proposed. Multi-functional therapies are believed to be advantageous over monotherapies, since a greater number of pathological mechanisms are targeted by these drugs, resulting in the amelioration of a broader scope of symptoms of AD. The lower concentrations required by multi-functional therapies also decreases potential side effects and increases drug compliance .
Dimebon, a gamma-carboline drug originally developed and now used in Russia as one of the standard therapies for AD patients, is one of the current multi-target anti-AD compounds available on the market. It inhibits alpha-adrenergic, histamine and serotonin receptors, including the 5-HT6 receptor [15, 16, 17], potentiates AMPA receptors, blocks NMDA receptors via a low-affinity mechanism [18, 19] and stimulates autophagia [20, 21, 22]. This drug also decreases the mitochondrial permeability transition, exerts cytoprotective properties [23, 24, 25], promotes neurite growth [26, 27, 28] and adult neurogenesis .
These and other properties of dimebon are believed to underlie its beneficial systemic effects. In animal studies, dimebon improves measures of cognition [30, 31, 32, 33] and emotionality [31, 34]. In initial human trials, dimebon has significantly ameliorated a number of AD symptoms and delayed their progression [35, 36, 37]. Subsequent clinical trials with dimebon have reportedly failed to replicate these findings; however, concerns were expressed about the applied study designs and the absence of publication in peer-reviewed scientific journals . The latest studies with dimebon have provided solid evidence, suggesting the efficacy of dimebon in reducing various processes associated with neurodegeneration [20, 21, 22, 39].
Numerous new gamma-carboline chemical analogues of dimebon have recently been synthesized . Among them is DF302, a fluorinated derivative of dimebon (Supplementary Fig. 1) which was found to be water soluble and stable in aqueous solutions, highly bioavailable for the brain and liver, non-toxic in in vitro and in vivo assays, capable of crossing the blood-brain barrier and displaying improved pharmacokinetics (half-life elimination of DF302 is 15.26 vs 2.59 h for dimebon; [41, 42, 43], see Supplementary data). Chronic dosing with DF302 in the range of 5.5–11 mg/kg has ameliorated the degeneration of motor neurons associated with ectopic expression of gamma-synuclein in Thy1mγSN mice , reduced accumulation of tau-protein in TauP301S mice and ameliorated abnormal expression of amyotrophic lateral sclerosis factor FUS . Preliminary data suggests positive effects of DF302 on processes of cellular protection and survival, as well as memory-enhancing action in conventional laboratory animals subjected to long-term memory paradigms and foot shock stress (Markova and Vinogradova, unpublished results).
In summation, DF302 has in numerous transgenic animal models displayed promising effects that are relevant to the pathophysiology of neurodegeneration and other neuropsychiatric pathologies. However, little is known about the mechanisms of action of these effects, as well as whether or not DF302 exerts similar activities in the absence of the severe pathology resulting from genetic manipulations. In this study, we sought to investigate the potential effects of DF302 under conventional conditions, using previously validated assays of the hippocampal neurogenesis, electrophysiological recordings of AMPA-mediated currents in cerebellar neurons, 5-HT6 receptor binding, mitochondrial functions, cellular neuroprotection and tests for contextual fear conditioning, anxiety, aggression and depressive-like behaviours, in standard and stressful conditions. In most of the in vivo experiments, the effects of DF302 were compared against the effects of dimebon. The selection of doses was based on previously reported data demonstrating the efficacy of both drugs in small rodents and zebrafish used at the concentration range corresponding to the rage of dose 10–40 mg/kg in mice [42, 43], (Lillesaar, unpublished results).
Materials and Methods
Male C57BL/6J mice were used in all behavioural studies except the assay of aggressive behaviour in which CD1 were employed. All mice were 3–3.5 months old and were obtained from RAS, Moscow region via a provider licenced by Charles River (http://www.spf-animals.ru/about/providers/animals). Wister rats were used for the collection of non-synaptosomal mitochondria and neuronal cultures. Animals were housed individually, unless specified, under standard conditions (see Supplementary Data). Experimental conditions conformed to Directive 2010/63/EU, and were approved by the local veterinarian committee. All efforts were undertaken to minimize the potential discomfort of experimental animals.
Behavioural Studies and Stress Experiment
Study of Contextual Learning and Its Extinction
Study of Hippocampal Neurogenesis and Behavioural Changes After Stress
Mice received DF302 or dimebon at doses of 10 or 40 mg/kg/day via drinking water for 2 weeks, or had no treatment; 14–16 mice were used per group. On days 9 and 10, all animals were tested in a 2-day swim test for depressive-like behaviour (Fig. 1b). A subset of mice was subjected to a 5-day rat exposure stress, as described elsewhere [31, 47, 48]; dosing was continued during the stressful period. Six hours after the termination of the stress inducing protocol, a cohort of stressed mice was tested for anxiety-like behaviours in the light dark box and anxiety step-down tests. Mice were sacrificed on average 2 h after behavioural testing and their brains were perfused for a subsequent neurogenesis assay (Fig. 1b). Another subset of mice was not stressed and then used to assess the potential effects of the drug treatments on markers of neurogenesis in naïve mice.
In addition, potential anti-stress like effects of DF302 were investigated in a model of emotional stress induced by ultrasound. A subset of mice was not pharmacologically treated, while others received DF302 or dimebon at a dosage of 40 mg/kg/day via drinking water. Simultaneous to treatment, mice underwent a 21-day long exposure to ultrasound of alternating frequencies between 20 and 45 Hz, where emotionally “negative” lower frequencies mimic natural signals of fear, as described elsewhere . Control mice were housed under standard conditions; 10 to 12 mice were used per group. On day 22, all groups were tested in the forced swim test for depressive-like behaviour (Fig. 1c). For further details on experimental conditions, see below and Supplementary material.
Study of Inter-male Aggression
Individual CD1 mice with pro-aggressive traits were pre-selected using the resident-intruder test and housed in groups of five for 2 weeks. Half of the mice received DF302 or dimebon at doses of 10 or 40 mg/kg via drinking water (Fig. 1d). On the first day after the 2-week period, mice were re-tested in the resident-intruder test, which was carried out as previously described ; (see below).
Step-Down Passive Avoidance Model
Mice were placed on the platform inside a transparent cylinder before relocation to the testing apparatus; after the cylinder’s removal, the time until the animal left the platform was measured and recorded as the baseline latency to step-down (see Supplementary data). Thereafter, a single electric foot shock (0.5 mA, 1 s) was delivered, mice were then returned to their home cages. Both 1 and 24 h later, animals were allowed to step-down from the platform under shock-free conditions, latency to step-down was scored as a measure of memory recall with a maximum recorded latency of 180 s. The percentage of “good learners”, which was defined using previously validated criteria of > 30 s latency to step-down was noted [44, 46, 51].
Acquisition and Extinction of Contextual Freezing
Mice were placed on the grid floor of the apparatus and an electric current (0.5 mA, 1 s) was delivered after a 2 min acclimatization period; animals were returned to their home cages immediately thereafter (see Supplementary data). Twenty-four hours later, freezing behaviour was scored for 180 s in the same apparatus by visual observation and taken as a measure of recall of contextual conditioning. The percentage of time spent freezing was assessed as described elsewhere [19, 45, 46]. At the end of this session, mice were left for another 7 min for memory extinction under shock-free conditions. After a further 24 h, mice were exposed to the chamber again, freezing behaviour was re-scored, a percentage of “good learners” was defined in accordance with the previously established criteria of > 50% time spent in contextual induced freezing [44, 46].
Step-Down Anxiety Test
In the step-down anxiety test, the latency of stepping-down was measured to assess anxiety-like behaviour in mice, as described previously [31, 51]. Mice were placed on the platform (7 × 7 × 1 cm) and covered with a cylinder (7 cm Ø, 15 cm height) and then moved to a transparent cubical apparatus (30x30x50cm; Open Science, Moscow, Russia). After removal of the cylinder, the latency of stepping-down from the platform with all four paws was measured and recorded as a measure of anxiety.
Light/Dark Box Test
Mice were placed into the black compartment (15 × 20 × 25 cm) from which they could visit the light compartment (30 × 20 × 25 cm, illumination intensity 25 Lux). During a 5-min period, the latency of the first transition, time spent in the light compartment and the number of transitions between compartments were recorded as described elsewhere [19, 34].
In the swim test, mice were placed in a transparent tank filled with water for 6-min and scored for the duration of floating behaviour during the first 2 min period of the test, as described elsewhere [34, 47, 52] (see Supplementary data). A two-session protocol with an interval of 24 h was employed in the rat exposure stress paradigm, and a single testing procedure was applied following chronic ultrasound stress.
The CD1 mice were placed individually in an observation cage (30 × 60 × 30 cm) for 30 min, a C57BL/6J mouse was then introduced to the cage for 4 min during the pre-selection assay and for 8 min during the assessment of effects of group housing and drug treatment. CD1 mice were scored for their latency to and number of attacks, as described elsewhere [50, 53]. In the pre-selection test, CD1 mice that displayed moderate aggression, as defined by a latency of attack of 50–240 s, were selected for further assessment.
Rat Exposure Stress
Mice were introduced into cylindrical transparent containers (15 cm × Ø 8 cm, holes in covers: Ø < 0.5 cm), which were placed into a rat home cage for 15 h (from 18:00 to 9:00 h) for 5 nights, as described elsewhere [31, 47, 48].
Mice were exposed to a 21-day continuous ultrasound stimulation of alternating frequencies between 20 and 45 Hz that was recently shown to induce depressive-like state  (see Supplementary material). The ultrasound device (Weitech, Wavre, Belgium) was hung 2 m above the cages with the alignment of the cages rotated weekly. Ultrasound radiation was measured by a detector (Discovery Channel, Crozet, VA, USA).
Quantitative Immunohistochemistry of Ki67 Binding
Mice from the rat exposure experiment were sacrificed under isoflurane-induced anaesthesia and perfused with 4% paraformaldehyde; brains were extracted, processed and scored for staining with Ki67 antibody, as described elsewhere [31, 47, 54] (see Supplementary material). Immunohistochemistry has been performed on free-floating 50 μm thick frontal hippocampal sections; incubated with mouse anti-Ki67 primary antibodies (Becton Dickinson Biosciences, San Jose, CA, USA), biotinylated goat anti-mouse secondary antibodies (Vector Laboratories, Burlingame, CA, USA) and streptavidin-biotin peroxidase complex. The number of Ki67-positive cells per mm3 of the subgranular zone of the dentate gyrus including the granule cell layer was scored using the Stereo Investigator software (MicroBrightField, Inc., Willsiton, VT, USA) using an Olympus BX51 microscope (Olympus, Hamburg, Germany) and the NeuroLucida software (MBF Bioscience, Williston, VT, USA).
To study the effects of DF302 on AMPA receptors, partial receptor agonist kainic acid (KA) was applied on isolated Purkinje neurons, as described elsewhere (Vignisse et al., 2014; see Supplementary Methods). Following an initial application of KA (20 μΜ), AMPA receptor-mediated transmembrane currents were recorded in the presence of increasing concentrations of the DF302 compound and expressed as a percentage of KA-induced values. Drug concentrations ranged from 0.001 to 30 μM and each dosage was tested on 4–6 Purkinje neurons.
Radioligand 5-HT6 Receptor Binding Competition Assay
In the radioligand binding assay, competition binding was performed using human Serotonin 5-HT6 receptor (FAST-0509B) non-membrane extracts against 0.1–10.000 nm concentrations of DF302 (see Supplementary Material). Prior to each assay, reference compounds were tested, and EC50 and/or IC50 values and a dose-response curve were obtained. The maximum variability accepted was ± 20% around the average of the replicates.
Assessment of Mitochondrial Functions
Non-synaptosomal mitochondria were isolated from the brains of Wistar rats using discontinuous Percoll density gradient, protein concentration was determined using a biuret procedure as previously described [55, 56, 57] (see Supplementary Material). Mitochondrial membrane potential was monitored using potential-dependent dye Safranine O, as described elsewhere .
Absorbance at 530 nm as measured with the Victor3 fluorescence plate reader (Perkin Elmer, Waltham, MA, Germany) was used to assess mitochondrial permeability transition (MPT) in a CaCl2-induced mitochondrial swelling protocol, as described previously  (see Supplementary Material). The maximum CaCl2-induced mitochondrial swelling rate during drug application was represented as the maximal rate of absorbance decline ((dA530/dt)×103) in a probe.
Calcium Retention Capacity of Mitochondria
CaG5N (ex/em = 506/535 nm) levels were evaluated using a Victor3 multi-well fluorescence plate reader (Perkin Elmer, Waltham, MA, Germany), which monitored extramitochondrial Ca2+ during bolus mode calcium applications to mitochondrial suspensions, as described elsewhere  (see Supplementary Material).
Models of Neuronal Death with Primary Cortical Neuron Cultures
Experiments with induction of neurotoxicity were carried out on the 8–10-day-old primary cultures of rat cortical neurons, which were prepared as previously described  (see Supplementary Material). Glutamate toxicity and the tert-butyl hydroperoxide (tBHP)-toxicity assays were performed and neuronal viability was determined as described elsewhere  (see Supplementary Material).
In studies involving intraperitoneal drug administration, drugs were dissolved in sterile saline, with concentrations based on the selected dosage of 10 or 40 mg/kg and a volume of injection of 0.01 ml/kg of body weight. The control group was injected with equivalent volumes of saline. For dosing via drinking water, drugs were dissolved in tap water. Their concentrations were calculated to obtain the desirable dosage of 10 or 40 mg/kg/day based on averaged values of 24 h water consumption, dosage was performed as described elsewhere [34, 61]. Control group mice received pure tap water.
The data was analysed using GraphPad Prism 6.00 software (San Diego, CA, USA) by one-way ANOVA followed by the Dunnett and Tukey multiple comparison tests. Repeated measures ANOVA was used for a comparison of repeated measurements. The trend of DF302 effect in different concentrations was analysed using univariate ANOVA and post hoc test for linear trend. Electrophysiological data were examined by t test using software from HEKA (Lambrecht, Germany). The level of statistical significance was set at p < 0.05. Results are presented as mean ± SEM.
Effects of Dimebon and DF302 on Contextual Learning and Its Extinction
Overall group comparison revealed no significant changes in the percentage of time spent freezing during a recall of fear conditioning 24 h after initial training, but a potential trend towards differences (p = 0.09, one-way ANOVA). A significant increase in freezing was found in mice treated with DF302 at the dose of 40 mg/kg versus the respective control group (p < 0.05, one-way ANOVA and post hoc Dunnett test; Fig. 2b, Table 2B–C). Other drug-treated groups showed no changes in freezing behaviour during this session when compared with the control group (p > 0.05). No significant group differences in the duration of freezing behaviour where observed during the latter recall of memory extinction time point (p > 0.05; Fig. 2b, c) suggesting that neither treatment alters this form of memory. A lack of changes of extinction in DF302-treated mice that received the higher concentration of this drug may have resulted from a heightened initial level of contextual freezing in this group.
Effects of Dimebon and DF302 on Hippocampal Neurogenesis of Naïve and Stressed Mice
Significant differences in cell density were observed between groups following rat exposure stress regardless of dimebon administration status (F = 3.683, p < 0.05, one-way ANOVA). All stressed mice, regardless of treatment status, displayed a significant reduction in the density of Ki67-positive cells in the subgranular zone of dentate gyrus (p < 0.05, one-way ANOVA and Tukey test; Table 3B; Fig. 3b). This finding indicates that this stress paradigm supressed hippocampal neurogenesis and treatment with dimebon at either dose did not interfere with this effect. In the rat exposure experiment with DF302 administration, there were significant group differences (F = 4.572, p < 0.05, one-way ANOVA); stressed mice treated with DF302 showed no decrease in a density of Ki67-positive cells in the subgranular zone of dentate gyrus as compared with non-stressed control mice, a group that received this drug at the dose 40/mg/kg, had significantly higher scores of this measure than non-treated stressed group (p < 0.05). Treatment with DF302 dose-dependently increased the density of proliferating Ki67-positive cells in the SGZ of dentate gyrus (univariate ANOVA for stressed groups, F (2, 21) = 6.197, p = 0.0077; post-test for linear trend, R2 = 0.3687, p = 0.0021, Table 3C, Fig. 3c).
Effects of Dimebon and DF302 on Parameters of Emotionality in the Rat Exposure Stress Study
Mouse groups showed significant differences in time spent in the light compartment (F = 4.812, p < 0.05, respectively, one-way ANOVA) and a trend towards significant differences in the number of exits to the light compartment. The stressed animal group which did not receive any drug treatment had significantly decreased scores of these behavioural parameters, in comparison to the non-stressed control group (p < 0.05; one-way ANOVA and post hoc Dunnett test; Table 4C–D, Fig. 4c, d). Stressed mice that were treated with either dosage of dimebon displayed no significant differences from control non-stressed animals in the time spent in the light compartment or the number of exits (p > 0.05 one-way ANOVA, Table 4C–D). Similarly, stressed mice that were dosed with either concentration of DF302 had no significant differences from control non-stressed animals in these two parameters (p > 0.05 one-way ANOVA, Table 4C–D, one-way ANOVA). Moreover, DF302-treated mice showed a dose-dependency in the number of exits to the light compartment that was highest in the group that was treated with the higher concentration of this compound (p < 0.05, one-way ANOVA and post hoc test for linear trend).
Effects of Dimebon and DF302 on Depressive-Like Behaviour in the Ultrasound Stress
In the ultrasound stress paradigm experiments, there were significant differences between the groups in the latency and duration of floating behaviour (F = 9.147, p = 0.001 and F = 4.431, p < 0.05 one-way ANOVA). In comparison to the control non-stressed group, stressed non-pharmacologically treated mice showed significantly lowered latency of floating (p < 0.05, one-way ANOVA and post hoc Dunnett test; Table 4E–F, Fig. 4e) and increased duration of floating (p < 0.05, one-way ANOVA and post hoc Dunnett test; Fig. 4f). DF302 treatment of stressed mice resulted in the absence of significant differences in these parameters between these groups and non-stressed non-pharmacologically treated control mice (p > 0.05; Table 4E–F, Fig. 4e, f).
Effects of Dimebon and DF302 on Aggressive Behaviour
There were significant group differences between groups in the latency to attack and total number of attacks in a study of inter-male aggression during group housing (in (F = 8.414, p = 0.0001 and F = 10.50, p < 0.0001, respectively, one-way ANOVA). As compared to CD1 mice that were not treated pharmacologically, mice of the same strain which were dosed with dimebon and DF302 at concentrations of 10 or 40 mg/kg showed prolonged latency of attack (p < 0.05, one-way ANOVA and post hoc Dunnett test; Table 4G). Also, in comparison to pharmacologically naïve group housed controls, mice treated with dimebon and DF302 at either dose had a diminished total number of attacks (p < 0.05; Table 4H, Fig. 4h). The effect of DF302 was on average greater than dimebon and displayed a significant dose-dependency on both aggression parameters (one-way ANOVA and post hoc test for linear trend). Thus, both dimebon and DF302 evoked anti-aggressive effects, with the behavioural effects of DF302 being more profound.
Effects of DF302 on AMPA Receptor-Mediated Currents
Effects of DF302 on the Binding of 5-HT6 Receptor
DF302 bound to the 5-HT6 receptor with high affinity; IC50 of 8.4 ± 0.4 nM (Fig. 5c), while the reference substance mianserin, showed a much lower affinity in this assay (IC50 = 135.4 ± 8.5 nM, see Supplementary data).
Effects of DF302 and Dimebon on Mitochondrial Functions and on Neuronal Viability
Calcium retention capacity in control samples was 148 ± 13 nmol/mg protein, while it was significantly higher in dimebon-treated samples 180 ± 11 nmol/mg protein (p < 0.05 vs. control, t test), and DF302-treated samples 224 ± 21 nmol/mg protein (p < 0.05 vs. control, t test, Fig. 6b). There was significantly greater effects of DF302 than dimebon on this parameter (p < 0.05). Thus, both dimebon and DF302 significantly improved calcium retention capacity of mitochondria with the effects of DF302 being more pronounced.
Finally in two models of neurotoxicity induced by applications of glutamate or tBHP, dimebon and DF302 increased the viability of neurons in comparison to controls (p < 0.05, t test, for all comparisons, Fig. 6c). These results provide evidence that both compounds display similar abilities to protect from cellular death.
Our current study revealed generally similar in vivo and in vitro activities of DF302 to those previously described for dimebon, while in a number of measured parameters DF302 displayed a greater potency. The activities investigated concern a number of the key pathophysiological processes associated with AD and neurodegeneration and as such could potentially underlie the mechanisms of action of anti-AD therapies. DF302 improved learning in paradigms concerning contextual memory, sharply enhanced hippocampal neurogenesis under normal and stressful conditions, decreased aggression and reduced stress-related behaviours in mice. These beneficial properties of DF302 could be due to its ameliorative effects on mitochondrial functions and cell survival, as well as its AMPA receptor stimulation ability and high binding affinity to the 5-HT6 receptor.
Present results suggest superior memory-enhancing effects of acute treatment with DF302 over those of dimebon in the weak protocols of contextual training which were employed here. Weak training protocols were selected for the current study in order to mimic a clinical situation of poor memory abilities that is typical for AD patients. While the administration of dimebon has previously been shown to exert memory-enhancing effects at doses of 0.5–1 mg/kg under standard protocols of learning in the step-down and fear-conditioning paradigms , we found no such effects in our weak learning conditions. In contrast to dimebon, DF302 at the higher 40 mg/kg dose significantly improved long-term memory both in the step-down and fear-conditioning tasks. Neither treatment altered short-term memory in the step-down avoidance paradigm, nor extinction of fear conditioning, suggesting that these forms of memory are not sensitive to the mechanisms of action of DF302 or dimebon. A lack of effects of DF302 on memory extinction may also be due to the greater initial scores of contextual memory prior to the extinguishing trial. Generally, the memory-enhancing effects of DF302 are in line with previously published results regarding gamma-carbolines use in rodent models of hippocampus-dependent memory [19, 31].
Treatment with DF302 has resulted in a dose-dependent increase of density of Ki67-positive neurons, a marker of proliferating cells, in the dentate gyrus of the hippocampus of both non-stressed and stressed mice, suggesting for elevated adult neurogenesis in these groups. While previous studies have demonstrated pro-neurogenic effects of treatment with dimebon that occurred at substantially lower doses than those employed in present study , no significant effects of dimebon on the density of Ki67-positive neurons were found. This might be due to the U-shape character of its systemic effects . Our results are in line with previously obtained data that suggest a lack of effect of administration of dimebon on parameters of hippocampal neurogenesis in naïve animals (Trofimov and Strekalova, unpublished results). In summary, unlike dimebon, DF302 has exerted pronounced pro-neurogenic activity under normal and stressful conditions in our study. Given that increased hippocampal neurogenesis is a well-established correlate of general anti-stress and pro-cognitive effects [62, 63], it could be suggested that the pro-neurogenic actions of DF302 may be a source of the ameliorative action of DF302 on the above-described behaviours.
Furthermore, the administration of DF302 but not of dimebon induced anti-depressive-like effects in stressed mice which underwent the forced swim test, a test of depression like behaviour . This reduction in depressive-like behaviour was found to parallel the changes in hippocampal neurogenesis . Notably, dosing with dimebon did not ameliorate reduced hippocampal neurogenesis in stressed mice and was ineffective in changing depressive-like behaviour. Furthermore, our studies revealed anti-aggressive and anti-anxiety effects of treatments with both DF302 and dimebon. The anti-anxiety and anti-aggressive effects of the administration of DF302 were found to be more pronounced that those of dimebon. Previous studies revealed modest ameliorative effects of dimebon in tasks examining anxiogenic and cognitive behaviours in the 5xFAD transgenic model of AD  and in similar tasks carried out on conventional models. According to the available literature, we believe our findings concerning anti-aggressive effects of dimebon and DF302 are the first demonstration of such effects in rodents. AD patients that have received dimebon were previously reported to show improvements in a broad spectrum of behavioural symptoms [35, 66].
Our results show that DF302, like dimebon , potentiates AMPA receptor-mediated currents, a mechanism associated with increased neurogenesis , pro-cognitive action  and anti-depressant and anti-anxiety-like effects [67, 68]. This potentiation is more pronounced in DF302 than in dimebon and could potentially explain DF302’s increased efficacy in the assays for plasticity and emotionality investigated here.
We found that DF302 binds to the 5-HT6 receptor with a high affinity that exceeds that of dimebon. While no data is currently available regarding the effects of DF302 binding on this subtype of serotonin receptor, previously obtained data with other gamma-carbolines strongly suggests an inhibitory effect of DF302 binding to the 5-HT6 receptor . Inhibition of the 5-HT6 receptor has previously be shown to result in anti-depressive and memory-enhancing effects, an increase in brain plasticity and anti-stress effects in the CNS [15, 69, 70]. These binding effects may aid the pro-neurogenic, memory-enhancing and anti-stress effects of administration of DF302.
Finally, we demonstrated the ameliorative effects of DF302 on mitochondrial permeability transition and on neuronal cell viability in two models of neurotoxicity. Recent experiments of Pieper and co. (2010) have showed that the pro-neurogenic activity of dimebon and other gamma-carbolines and carbazole derivatives correlated with their inhibitory action on mitochondrial permeability transition. These effects may be due to the well-established pivotal role of mitochondria in cell survival, cell differentiation and programmed cell death of neurons [71, 72]. The inhibition of mitochondrial permeability transition, a key step of programmed cell death, has been shown to stimulate hippocampal neurogenesis by increasing the survival of newly differentiated cells [29, 73]. Positive action of DF302 on mitochondrial permeability transition could also underlie protective effects of this drug under conditions of neurotoxicity and oxidative stress as seen in the two models of neuronal culture toxicity utilized here. Increased neuronal survival during application of DF302 could be particularly important to the stress-protective effects of this drug that were found in our ultrasound and predation stress models. While DF302 and dimebon displayed similar behavioural affects overall, the greater cellular effects of DF302 on mitochondrial functions and neuronal cell survival, along with the other above-described molecular effects, may explain the superior activity of DF302 on brain plasticity.
It is likely that DF302 impacts multiple pathways other than the targets reported here. Experiments that address other potential targets are in progress. The greater pharmacokinetic properties of DF302 in comparison to those of dimebon may also contribute to the improved efficacy of this drug over dimebon. The higher stability in concentration and longer lasting presence of DF302 in blood and CNS during chronic and acute treatments may result in a more pronounced biological response.
In summary, the data presented here demonstrates that DF302 affects multiple key mechanisms that are associated with the clinical efficacy of dimebon and other anti-AD therapies. The in vivo and in vitro effects of DF302 are generally greater in number and magnitude than those of dimebon, suggesting a strong potential for DF302 in therapeutic treatment of patients suffering from neurodegenerative conditions, including AD. In a broader sense, these results support the need for development of multi-target drugs which are capable of modulating the pathophysiology of neurodegenerative disorders. Our study provides further evidence for the immense clinical potential of multi-target drugs to ameliorate cognitive and behavioural symptoms.
This work was primarily supported by the Russian Science Foundation grant #14-23-00160P. The DF302 compound and model agent dimebon were synthesized in frames of the IPAC Research Program Framework. Behavioural and immunohistochemical assays have been conducted with the support of the European Community EC: AGGRESSOTYPE FP7/No.602805, and the 5-100 Russian Academic Excellence Project.
- 1.Harrison TM, Burggren AC, Small GW, Bookheimer SY (2016) Altered memory-related functional connectivity of the anterior and posterior hippocampus in older adults at increased genetic risk for Alzheimer’s disease. Hum Brain Mapp 37(1):366–380. https://doi.org/10.1002/hbm.23036 CrossRefPubMedGoogle Scholar
- 3.Schneider LS (2013) Alzheimer disease pharmacologic treatment and treatment research. Continuum (Minneap Minn) 19(2 Dementia):339–357. https://doi.org/10.1212/01.CON.0000429180.60095.d0 Google Scholar
- 5.Bachurin SO, Bovina EV, Ustyugov AA (2017) Drugs in clinical trials for Alzheimer’s disease. The major trends. Med Res Rev 37(5):1186–1225Google Scholar
- 7.Combarros O, Cortina-Borja M, Smith AD, Lehmann DJ (2009) Epistasis in sporadic Alzheimer’s disease. Neurobiol Aging 30(9):1333–1349. https://doi.org/10.1016/j.neurobiolaging.2007.11.027 CrossRefPubMedGoogle Scholar
- 9.Calzà L, Baldassarro VA, Giuliani A, Lorenzini L, Fernandez M, Mangano C, Sivilia S, Alessandri M et al (2013) From the multifactorial nature of Alzheimer’s disease to multitarget therapy: the contribution of the translational approach. Curr Top Med Chem 13(15):1843–1845Google Scholar
- 12.Weinstock M, Gorodetsky E, Poltyrev T, Gross A, Sagi Y, Youdim M (2003) A novel cholinesterase and brain-selective monoamine oxidase inhibitor for the treatment of dementia comorbid with depression and Parkinson’s disease. Prog Neuro-Psychoph 27(4):555–561. https://doi.org/10.1016/S0278-5846(03)00053-8 CrossRefGoogle Scholar
- 17.Giorgetti M, Gibbons JA, Bernales S, Alfaro IE, Drieu La Rochelle C, Cremers T, Altar CA, Wronski R et al (2010) Cognition-enhancing properties of Dimebon in a rat novel object recognition task are unlikely to be associated with acetylcholinesterase inhibition or N-methyl-D-aspartate receptor antagonism. J Pharmacol Exp Ther 333(3):748–757. https://doi.org/10.1124/jpet.109.164491
- 19.Vignisse J, Steinbusch HW, Grigoriev V, Bolkunov A, Proshin A, Bettendorff L, Bachurin S, Strekalova T (2014) Concomitant manipulation of murine NMDA- and AMPA-receptors to produce pro-cognitive drug effects in mice. Eur Neuropsychopharmacol 24(2):309–320. https://doi.org/10.1016/j.euroneuro.2013.06.010 CrossRefPubMedGoogle Scholar
- 26.Protter A, Vartiainen V, Yrjanheikki J, Bernales S (2009) Neurite outgrowth and mitochondrial function in dimebon treated rat cortical cultures. Neurodegener Dis 6:1536Google Scholar
- 27.Bernales S, Alarcon R, Guerrero J, Higaki JN, Protter AA (2009) Dimebon induces neurite outgrowth from hippocampal, spinal, and cortical neurons. Neurology 72:A385Google Scholar
- 29.Pieper AA, Xie S, Capota E, Estill SJ, Zhong J, Long JM et al (2010) Discovery of a proneurogenic, neuroprotective chemical. Cell 142:39–51. https://doi.org/10.1016/j.cell.2010.06.018
- 30.Bachurin S, Bukatina E, Lermontova N, Tkachenko S, Afanasiev A, Grigoriev V, Grigorieva I, Ivanov Y et al (2001) Antihistamine agent Dimebon as a novel neuroprotector and a cognition enhancer. Ann N Y Acad Sci 939:425–435Google Scholar
- 31.Vignisse J, Steinbusch HW, Bolkunov A, Nunes J, Santos AI, Grandfils C, Bachurin S, Strekalova T (2011) Dimebon enhances hippocampus-dependent learning in both appetitive and inhibitory memory tasks in mice. Prog Neuro-Psychoph 35(2):510–522. https://doi.org/10.1016/j.pnpbp.2010.12.007 CrossRefGoogle Scholar
- 32.Webster SJ, Wilson CA, Lee CH, Mohler EG, Terry AV Jr, Buccafusco JJ (2011) The acute effects of dimebolin, a potential Alzheimer’s disease treatment, on working memory in rhesus monkeys. Br J Pharmacol 164(3):970–978. https://doi.org/10.1111/j.1476-5381.2011.01432.x CrossRefPubMedPubMedCentralGoogle Scholar
- 34.Malatynska E, Steinbusch HW, Redkozubova O, Bolkunov A, Kubatiev A, Yeritsyan NB, Vignisse J, Bachurin S et al (2012) Anhedonic-like traits and lack of affective deficits in 18-month-old C57BL/6 mice: implications for modeling elderly depression. Exp Gerontol 47(8):552–564. https://doi.org/10.1016/j.exger.2012.04.010
- 35.Doody RS, Gavrilova SI, Sano M, Thomas RG, Aisen PS, Bachurin SO, Seely L, Hung D et al (2008) Effect of dimebon on cognition, activities of daily living, behaviour, and global function in patients with mild-to-moderate Alzheimer’s disease: a randomised, double-blind, placebo-controlled study. Lancet 372(9634):207–215. https://doi.org/10.1016/S0140-6736(08)61074-0
- 38.Bharadwaj PR, Bates KA, Porter T, Teimouri E, Perry G, Steele JW, Gandy S, Groth D et al (2013) Latrepirdine: molecular mechanisms underlying potential therapeutic roles in Alzheimer’s and other neurodegenerative diseases. Transl Psychiatry 3(3):e332Google Scholar
- 41.Bachurin SO, Shelkovnikova TA, Ustyugov AA, Peters O, Khritankova I, Afanasieva MA, Tarasova TV, Alentov II et al (2012) Dimebon slows progression of proteinopathy in γ-synuclein transgenic mice. Neurotox Res 22(1):33–42. https://doi.org/10.1007/s12640-011-9299-y
- 43.Peters OM, Shelkovnikova T, Tarasova T, Springe S, Kukharsky MS, Smith GA, Brooks S, Kozin SA et al (2013) Chronic administration of Dimebon does not ameliorate amyloid-β pathology in 5xFAD transgenic mice. J Alzheimers Dis 36(3):589–596. https://doi.org/10.3233/JAD-130071
- 46.Veniaminova E, Cespuglio R, Cheung CW, Umriukhin A., Markova N, Shevtsova E, Lesch K-P, Anthony DC, Strekalova T (2017) Autism-like behaviours and memory deficits result from a Western diet in mice. Neural Plasticity, in pressGoogle Scholar
- 47.Strekalova T, Evans M, Chernopiatko A, Couch Y, Costa-Nunes J, Cespuglio R, Chesson L, Vignisse J et al (2015) Deuterium content of water increases depression susceptibility: the potential role of a serotonin-related mechanism. Behav Brain Res 277:237–244. https://doi.org/10.1016/j.bbr.2014.07.039
- 48.Markova N, Bazhenova N, Anthony DC, Vignisse J, Svistunov A, Lesch KP, Bettendorff L, Strekalova T (2016) Thiamine and benfotiamine improve cognition and ameliorate GSK-3β-associated stress-induced behaviours in mice. Prog Neuro-Psychopharmacol Biol Psychiatry 75:148–156. https://doi.org/10.1016/j.pnpbp.2016.11.001 CrossRefGoogle Scholar
- 49.Morozova A, Zubkov E, Strekalova T, Kekelidze Z, Storozeva Z, Schroeter CA, Bazhenova N, Lesch KP et al (2016) Ultrasound of alternating frequencies and variable emotional impact evokes depressive syndrome in mice and rats. Prog Neuro-Psychopharmacol Biol Psychiatry 68:52–63. https://doi.org/10.1016/j.pnpbp.2016.03.003
- 50.Couch Y, Trofimov A, Markova N, Nikolenko V, Steinbusch HW, Chekhonin V, Schroeter C, Lesch KP et al (2016) Low-dose lipopolysaccharide (LPS) inhibits aggressive and augments depressive behaviours in a chronic mild stress model in mice. J Neuroinflammation 13(1):108. https://doi.org/10.1186/s12974-016-0572-0
- 54.Sun P, Knezovic A, Parlak M, Cuber J, Karabeg MM, Deckert J, Riederer P, Hua Q et al (2015) Long-term effects of intracerebroventricular streptozotocin treatment on adult neurogenesis in the rat hippocampus. Curr Alzheimer Res 12(8):772–784Google Scholar
- 59.Rathinam ML, Watts LT, Narasimhan M, Riar AK, Mahimainathan L, Henderson GI (2012) Astrocyte mediated protection of fetal cerebral cortical neurons from rotenone and paraquat. Environ Toxicol Pharmacol 33(2):353–360. https://doi.org/10.1016/j.etap.2011.12.027 CrossRefPubMedPubMedCentralGoogle Scholar
- 68.Harro J, Kanarik M, Matrov D, Panksepp J (2011) Mapping patterns of depression-related brain regions with cytochrome oxidase histochemistry: relevance of animal affective systems to human disorders, with a focus on resilience to adverse events. Neurosci Biobehav Rev 35(9):1876–1889. https://doi.org/10.1016/j.neubiorev.2011.02.016 CrossRefPubMedGoogle Scholar
- 70.Mitchell ES, Hoplight BJ, Lear SP, Neumaier JF (2006) BGC20-761, a novel tryptamine analog, enhances memory consolidation and reverses scopolamine-induced memory deficit in social and visuospatial memory tasks through a 5-HT6 receptor-mediated mechanism. Neuropharmacology 50(4):412–420CrossRefPubMedGoogle Scholar