Abstract
Depression-associated cognitive impairments are among the most prevalent and persistent symptoms during remission from a depressive episode and a major risk factor for relapse. Consequently, development of antidepressant drugs, which also alleviate cognitive impairments, is vital. One such potential antidepressant is vortioxetine that has been postulated to exhibit both antidepressant and pro-cognitive effects. Hence, we tested vortioxetine for combined antidepressant and pro-cognitive effects in male Long-Evans rats exposed to the chronic mild stress (CMS) paradigm. This well-established CMS paradigm evokes cognitive deficits in addition to anhedonia, a core symptom of depression. Learning and memory performance was assessed in the translational touchscreen version of the paired-associates learning task. To identify the mechanistic underpinning of the neurobehavioural results, transcriptional profiling of genes involved in the stress response, neuronal plasticity and genes of broad relevance in neuropsychiatric pathologies were assessed. Vortioxetine substantially relieved the anhedonic-like state in the CMS rats and promoted acquisition of the cognitive test independent of hedonic phenotype, potentially due to an altered cognitive strategy. Minor alterations in gene expression profiling in prefrontal cortex and hippocampus were found. In summary, our findings suggest that vortioxetine exhibits an antidepressant effect as well as behavioural changes in a translational learning task.
Similar content being viewed by others
Introduction
Worldwide, around 264 million people suffer from major depressive disorder (MDD) making this the leading burden of disability worldwide1. The recurrent nature of the disease together with insufficient responses to antidepressant treatment add to the devastating burden of the disease2. Core symptoms of MDD are a depressed mood and an attenuated anticipation or experience of pleasure (anhedonia). Additionally, patients suffer from a variable number of associated symptoms, such as impaired cognitive abilities, which affect primarily attention, executive functions and memory. These cognitive symptoms persist in 30–60% of treated patients after remission from the affective MDD symptoms. Furthermore, cognitive impairments are the most persisting residual symptoms of depression and, hence, continue to decrease daily functioning and quality of life after remission3,4,5,6. Moreover, persistent cognitive impairments augment risk of relapse and are increasingly regarded as a core component rather than an epiphenomenon of depression7,8. Recovery from cognitive symptoms is associated with a rapid remission from depression9, further underlining the importance of restoring cognitive impairments when treating depression.
However, current antidepressant treatment focuses mainly on alleviating the affective symptoms, neglecting cognitive impairments10. Therefore, development of novel, pro-cognitive antidepressants is vital and, hence, a translational drug screening platform for depression-associated cognitive impairments is essential. In a previous study11, it was demonstrated that the chronic mild stress (CMS) paradigm fulfils exactly these criteria. The CMS model exhibits the MDD core symptom anhedonia (face validity) evoked by stress exposure (etiological validity). Additionally, CMS anhedonic-like rats display depression-associated cognitive impairments, indicated by lower performance in a translational touchscreen learning task, which was not found in CMS resilient, hedonic rats11. Hence, cognitive impairments are specific to the depression-like phenotype. In the present study, we follow up by assessing the efficacy of a relatively novel, multimodal antidepressant on affective symptoms and cognitive deficits in the CMS model.
Vortioxetine was approved as an antidepressant in 201312. In addition to an antidepressant action, a pro-cognitive effect was ascribed to vortioxetine due to its multimodal mechanism of action13. In MDD patients, executive functions, attention, speed of processing, verbal learning and memory functions, as well as affective symptoms, have been shown to recover after chronic vortioxetine intervention14. In rodents, vortioxetine improved spatial working memory, visuo-spatial memory and contextual fear memory besides increasing synaptic plasticity and decreasing behavioural despair15,16,17,18,19. Although the CMS model shows high predictive validity for antidepressant actions20,21, unexpectedly, vortioxetine was reported to be ineffective in the CMS model22. Thus, we investigated, in the present study, if vortioxetine can alleviate the anhedonic-like phenotype of CMS exposed rats using a different route of drug administration. Moreover, cognition of these rats was assessed in the different paired-associates learning (dPAL) touchscreen task, a standardized tool in clinical as well as in preclinical research23,24. The rather novel rodent touchscreen platform involves appetitive operant conditioning and was developed based on the human Cambridge Neuropsychological Test Automated Battery (CANTAB); the most frequently applied cognitive assessment tool in depression research4. Finally, hippocampal (HPC) and prefrontal cortex (PFC) gene expression was analysed to link neurobehavioral alterations with underlying molecular changes. Genes that are thought to play a role in psychiatric disorders and/or the stress response, such as the mineralocorticoid receptor (Nr3c2), glucocorticoid receptor (Nr3c1), FK506 binding protein 5 (Fkbp5), glycogen synthase kinase 3 beta (Gsk3b), disrupted in Schizophrenia 1 (Disc1) and brain-derived neurotrophic factor (Bdnf) as well as genes important in cognition and neuronal plasticity, such as neuroregulin 1 (Nrg1), homer scaffolding protein 1–3 (Homer1–3), Shank 1–3, Spinophilin and Cofilin 1, were analysed.
In short, this study aimed to investigate the effect of vortioxetine on the affective state, cognitive performance and cerebral gene expression.
Materials and methods
Animals
Male Long-Evans (LE) rats (Janvier Labs, France; n = 242) were 5–6 weeks of age weighing 100–120 g at arrival. Rats were single-housed during the experiment with free access to food and water (unless otherwise stated) and kept on a 12-h light–dark cycle. All experiments were conducted according to EU Directive 2010/63/EU, in compliance with the ARRIVE guidelines and approved by the Danish National Committee for Ethics in Animal Experimentation (2013-15-2934-00814).
Chronic mild stress paradigm
A 1-h sucrose consumption test (SCT, 1.5%) was carried out weekly to assess the hedonic state of each rat throughout the experiment (Supplementary Methods). Following three baseline SCTs, rats were exposed to a number of variable, unpredictable mild stressors in a two-week repeated protocol (Table S1) to provoke a depressive-like phenotype.
After five weeks of CMS, stress exposed rats with a SCT index ≤ 0.7 (average of SCTs in week 4–5 normalised to baseline) were categorized anhedonic-like according to an a priori cutoff25,26 and remained in the study.
Following nine weeks of CMS, which included an initial four weeks of drug treatment, a modified CMS protocol was used (Fig. 1A). Stressors were only applied during the nights reserving daytime for touchscreen assessment. Every Friday, the SCT was carried out followed by 4 h of grouping and light stressors. Thus, touchscreen testing was discontinued on Fridays. The modified CMS schedule (Table S2) was changed every second week to prevent habituation to the milder stress protocol.
Drug administration
After five weeks of CMS, 45 and 12 anhedonic-like animals were randomly assigned to treatment with vortioxetine or vehicle (Fig. 1A). Group means and standard deviations of the last SCT index before treatment start were comparable for treatment and vehicle group. Standard rat chow (Altromin 1324, Brogaarden, Denmark) was supplemented with vortioxetine (Carbosynth Ltd., UK) at a concentration of 1.8 g/kg rat chow in order to reach a therapeutic dose range with a SERT occupancy above 90%27. Following four weeks of treatment combined with CMS, rats were subdivided into high responders (10 rats with highest recovery according to SCT index) and low responders (10 rats with lowest recovery according to SCT index) and subjected to touchscreen testing.
Touchscreen operant platform
Food reduction and touchscreen pre-training
After nine weeks of CMS and four weeks of treatment, 40 rats (control, anhedonic-like, responder, low-responder; n = 10/group) were used for touchscreen testing. First, rats were gradually food restricted to 75% of their individual ad libitum consumption (Table S3)11. Body weights were monitored daily to ensure rats maintain at least 90% of their body weight during food restriction. Additionally, rats were introduced to peanut butter (Bilka, Denmark) and bacon pellets (45 mg dustless precision pellets, Bio Serv, Flemington, NJ, USA) used for operant conditioning during touchscreen testing. Pre-training was conducted after eight days of food restriction. In four steps, rats were conditioned to operate the touchscreen chamber (Fig. 1B). For further details on pre-training and the Bussey-Saksida touchscreen operant chambers (Campden Instruments Ltd., Loughborough, UK) see Supplementary Method section. Experimenters carrying out behavioural testing were blinded to group identity.
Paired-associates learning touchscreen task
Cognitive performance was assessed in the dPAL task, in which a specific symbol-location association needs to be learned. In each trial, only two of the three symbols (spider, flower, plane) would be displayed, one in its correct location (S +) and the other symbol in an incorrect location (S-) on the touchscreen. The third window was left blank (Supplementary Fig. S1). A touch to S + resulted in reward pellet delivery followed by a 20 s inter-trial interval (ITI). Poking S- was followed by a 5 s time out with house light on, the ITI and a correction trial (repetition of the incorrect trial until correct). The six trial types resulting from the stimulus-location association pairs were balanced over the course of a session. dPAL criterion was achieved by completing 75 trials (excluding correction trials) with at least 60 correct trials (≥ 80% accuracy) within 45 min on two consecutive days. Rats that did not acquire the task within 46 session were marked as failing the task by an a priori criterion from a previous study11.
Retention of the dPAL task
Passing the dPAL task was ensued by a 10-day hiatus without touchscreen testing and an increase in food availability. Rats were then re-tested on the dPAL task for two days to assess long-term memory.
Cerebral gene expression
A circadian rhythm of BDNF has been reported in certain brain regions28,29. Therefore, the rats were sacrificed under similar standardized time conditions from 2 to 4 pm, 1–3 days (Mean = 1.3 days) after completing the dPAL retention testing. To diminish a possible effect of the testing, the rats were distributed across the four groups at day 1 to 3. The brain was removed and PFC, dorsal and ventral HPC were dissected and snap frozen on dry-ice. RNA was extracted using the PARIS RNA isolation kit (Ambion, TX, USA). The samples were processed as previously described30 and real-time qPCR was performed. A detailed description of RNA extraction and qPCR can be found in Supplementary Methods.
Statistical analysis
SCT data were analysed by a two-way ANOVA (time x group), followed by group-wise post-hoc comparisons. SCT data are displayed and included in the analysis until the time point when the first animal was terminated after completing the dPAL task.
Summary statistics of the dPAL task (see results section ‘Acquisition of the dPAL task’) were analysed by applying two-way ANOVA (hedonic state x treatment) or by rank aligned two-way ANOVA (indicated with Frank) if assumptions of normality (assessed with QQ-plots) or homogeneity of variance (assessed with Bartlett’s test) were violated. Furthermore, one outlier in the control group for median response latency and two outliers (control and low-responder) for number redundant screen touches were determined by Grubbs (α = 0.05) or ROUT (Q = 1%) test (Prism 7, GraphPad Software Inc., CA, USA) and excluded.
Repeated measurement data analysing learning behaviour across the task (results section ‘Learning phase of the dPAL task’) and learning behaviour within a session (‘Learning behaviour within a single dPAL session’ & Supplementary Results) included all animals (acquiring and failing dPAL acquisition), whereas retention data (‘Long-term memory of dPAL task’ & Supplementary Results) only included animals passing the dPAL task. The data were analysed with repeated measures ANOVA of type III if significant interaction effect was present, otherwise with type II. Mauchly’s sphericity test, if significant, led to Greenhouse–Geisser (GG) (ε < 0.75) or Huynh–Feldt (HF)-corrected repeated measures ANOVA (indicated with FGG or FHF). Post-hoc comparisons were Bonferroni-corrected. In a separate analysis of memory and relearning performance (Supplementary Results), data were analysed by two-way ANOVA as described in summary statistics.
Normalised target genes were displayed as percent of control group mean (PFC data) or percent of dorsal HPC control mean (dorsal and ventral HPC data) and analysed by two-way ANOVA as described in summary statistics. Differences between dorsal and ventral HPC gene expression were analysed with Student’s t-test. Supplementary Table S5 displays n-number for each gene and group, thus, the number of outliers removed.
Statistical significance was accepted at p < 0.05, two-tailed. Effect size is reported as eta squared (η2; summary statistics) or generalised eta squared (η2G; repeated measures) for cognitive results31. All post-hoc comparisons were Bonferroni-corrected. Statistical analyses were performed with RStudio (Version 0.99.892, Boston, USA) and data were displayed with GraphPad Prism 7.
Results
Hedonic-like status in response to CMS and vortioxetine treatment
Following a significant interaction effect of group x time (F(45,540) = 5.52, p < 0.0001; two-way ANOVA) and main effects of time (F(15,540) = 12.82, p < 0.0001) and group (F(3,36) = 32.24, p < 0.0001), Bonferroni-corrected post-hoc analysis revealed that anhedonic-like rats consumed significantly less sucrose during all SCTs compared to non-stressed control rats (p < 0.0001). Sixty-five percent of treated rats responded well to vortioxetine and their sucrose intake was not statistically significant different from non-stressed controls, but significantly increased compared to untreated, anhedonic-like rats (p < 0.0001). Rats that responded poorly to vortioxetine, thus low-responders, consumed significantly less sucrose than responders (p < 0.0001) or non-stressed controls (p < 0.0001), but were not statistically significantly different to anhedonic-like rats (Fig. 2).
Paired-associates learning touchscreen task
Acquisition of the dPAL task
Acquisition of the dPAL task, indicated by the accumulated number of trials over all sessions to reach criterion for passing, did not differ significantly between groups (Fig. 3A).
Two-way ANOVA revealed that drug treatment increased the number of redundant screen touches compared to untreated animals (main effect of treatment: F(1,28) = 9.98, p = 0.004, η2 = 0.23). This treatment effect is possibly driven by a trend in hedonic state x treatment interaction effect (F(1,28) = 1.12, p = 0.063, η2 = 0.08), i.e. responders diverging (Fig. 3B).
Median response latency was altered due to a hedonic state x treatment interaction effect (F(1,29) = 9.03, p = 0.005, η2 = 0.15; Fig. 3C). Specifically, anhedonic-like rats (p = 0.013), responders (p = 0.0001) and low-responders (p = 0.001) responded faster to touchscreen stimuli than non-stressed control rats. Furthermore, treatment alone reduced median response latency (F(1,29) = 17.58, p = 0.0002, η2 = 0.30; Fig. 3C).
There was no difference in reward collection latency (Fig. 3D) or number of correction trials between groups.
Six animals (one non-stressed control, three anhedonic-like rats, one responder and one low-responder) did not pass dPAL and, thus, were excluded from this analysis.
Learning phase of the dPAL task
To compare learning curves with repeated measures ANOVA, the rats’ variable number of sessions and trials per session was normalised32. Thus, for each rat, the total number of trials (trials + correction trials) to learn the dPAL task was split into ten equal bins11.
The percentage of correct trials (accuracy) increased significantly over time, thus, with increasing number of bins (FGG(3.00,107.98) = 30.08, p < 0.0001, η2G = 0.08), indicating task learning. No effect of group on accuracy was observed (Fig. 4A).
The number of trials performed increased significantly over time with growing bin number (FGG(3.08,110.85) = 47.90, p < 0.0001, η2G = 0.10), whereas the number of correction trials decreased significantly by bin number (FGG(3.08,110.73) = 48.37, p < 0.0001, η2G = 0.17; Fig. 4B). This also indicates learning of the task, however, no statistically significant differences between groups were observed.
Control rats responded slower to stimuli compared to anhedonic-like rats (p = 0.002), low-responders (p < 0.0001) or responders (p < 0.0001). However, vortioxetine responders showed the shortest median response latency compared to controls, anhedonic-like rats (p = 0.002) and low-responders (p = 0.029; main effect of group: F(3,36) = 3.24, p = 0.033, η2G = 0.15). The median response latency decreased significantly during dPAL acquisition (main effect of time: FGG(4.32,155.40) = 9.14, p < 0.0001, η2G = 0.08; Fig. 4C).
Vortioxetine responders executed the highest number of redundant screen touches per trial compared to control rats (p < 0.0001) and anhedonic-like rats (p < 0.0001). Vortioxetine low-responders also performed more redundant screen touches than control (p = 0.022) and anhedonic-like rats (p = 0.005; main effect of group: F(3,36) = 3.10, p = 0.039, η2G = 0.13). The number of redundant screen touches decreased during dPAL acquisition (main effect of time: FGG(2.46,88.45) = 5.67, p < 0.0001, η2G = 0.06; Fig. 4D).
Collection latency was not significantly different between groups or over time, suggesting equal motivation for reward collection and for engaging in the dPAL task.
Learning behaviour within a single dPAL session
Every single session of an animal was divided into six blocks by reference to the total number of trials (trials + correction trials). Average performance per block was determined for each animal. This allowed analysis of learning behaviour within the time course of a session.
During a session, accuracy did not change significantly over time, nor between groups. The number of trials executed during a session changed depending on session block (main effect of session block: F(5,180) = 3.38, p = 0.006, η2G = 0.02; Supplementary Fig. S2A). For further details see Supplementary Result Section.
Long-term memory of dPAL task
Long-term memory performance was assessed by re-testing rats in dPAL following a 10-day hiatus after dPAL acquisition. Included in the analysis was accuracy of the last session of dPAL acquisition before the break as well as the two dPAL retention sessions after the break. A trend of an interaction effect of group x session (FGG(4.28,42.75) = 2.10, p = 0.066, η2G = 0.05) and a main effect of time (FGG(1.43,42.75) = 8.91, p = 0.0004, η2G = 0.36) on accuracy was observed. Bonferroni post-hoc comparisons revealed that all groups decreased accuracy of task retention in session one, vortioxetine responders significantly increased their accuracy on retention session two and all groups continued to show a lower accuracy on session two compared to passing criterion (Supplementary Fig. S3A). For further details see Supplementary Result Section.
Cerebral gene expression
Alterations in gene expression levels were analysed in response to vortioxetine treatment and hedonic state. Furthermore, differences between dorsal and ventral HPC gene expression were examined. Regulated genes are presented in Fig. 5. Supplementary Table S5 contains all gene expression levels for the four groups and all tissues.
Prefrontal cortex gene expression
In the PFC, the expression level of Cofilin 1 was increased in the anhedonic-like group (p = 0.022) compared to controls (interaction effect of hedonic state x treatment: Frank(1,28) = 5.51, p = 0.026). A trend of treatment reducing expression of Nr3c1 mRNA was observed (F(1,27) = 4.07, p = 0.054). The mRNA expression levels of Nr3c2, Fkbp5, Disc1, Gsk3b, Bdnf, Shank 1–3, Homer1–3, Nrg1, and Spinophilin were not affected.
Hippocampal gene expression
The gene expression of Gsk3b, Disc1, Shank1, Shank2, and Nrg1 was higher in the ventral compared to dorsal HPC (t(35) = − 3.13, p = 0.004; t(34) = − 4.72, p < 0.0001; t(34) = − 3.99, p = 0.0003; t(32) = − 3.58, p = 0.001; and t(32) = − 5.84, p < 0.0001, respectively). For Homer1 the expression was decreased in the ventral compared to the dorsal HPC (t(35) = 3.01, p = 0.005).
In the dorsal HPC, Homer2 gene expression was decreased in groups with anhedonic-like phenotype (main effect of hedonic state: F(1,33) = 5.63, p = 0.024; Fig. 5).
Close to significant trends due to treatment and/or hedonic state were observed for Nr3c2, Disc1, Gsk3b, Bdnf and Homer3 mRNA levels (Fig. 5; statistics in Table S5); with no notable observations on Fkbp5, Nr3c1, Shank3, Spinophilin, or Cofilin 1 gene expression across tissues, hedonic state or treatment.
Discussion
In the present CMS study, non-stressed controls, anhedonic-like rats and vortioxetine treated rats were assessed for hedonic state, cognitive performance and cerebral gene expression profiling.
Vortioxetine recovers the hedonic state
CMS exposed rats decreased sucrose intake over time, indicating a reduced reward sensitivity and, hence, mirroring the MDD core symptom anhedonia. Administration of the antidepressant vortioxetine recovered the hedonic state in a major fraction of anhedonic-like rats (65%), whereas the remaining rats responded poorly and remained in an anhedonic-like state. Previously vortioxetine was reported to be ineffective when tested in the CMS model22. However, vortioxetine was administered by intraperitoneal injections once daily (Mariusz Papp, personal communication), and the relatively short half-life of vortioxetine in rodents22 may explain for the ineffective treatment outcomes in this study. In the present study vortioxetine was mixed into the diet and, hence, this route of drug administration ensured a more even and continuous diurnal drug exposure. In a parallel study using the same dose and route of administration, we confirmed comparable vortioxetine serum levels (unpublished data) as shown to be therapeutically relevant27,33. Food restriction necessary for touchscreen training likely resulted in a slightly reduced dose of vortioxetine. However, this reduction was comparable across animals and groups (Suplementary Fig. S5). Furthermore, monitoring of the hedonic state with SCTs throughout the study (Fig. 2) showed that vortioxetine responders remained comparable to controls and above the criterion for anhedonia even during food reduction.
Vortioxetine affects cognition
In the present study we also investigated whether vortioxetine-induced alleviation of the hedonic state is associated with alterations in cognitive performance. Vortioxetine has been reported to augment cognitive functions22 and is believed to be a directly mediated effect rather than caused through remission from affective symptoms14. In the present study, vortioxetine did not alter primary touchscreen parameters (accuracy, number of trials) compared to non-stressed controls or anhedonic-like rats. However, we noticed that three out of ten anhedonic-like rats did not pass the dPAL task within 46 sessions whereas only one animal failed to pass in any of the other groups. This observation might be attributed to normal biological variation considering the small group size (n = 10). Alternatively, the inability to acquire the dPAL task might suggest cognitive impairment in the anhedonic-like group and, consequently, a potential pro-cognitive effect of vortioxetine treatment. Future studies are needed to validate this interpretation.
Importantly, the latency for collecting reward pellets did not differ between groups. This suggests equal incentive to consume the reward and presumably to participate in the touchscreen task. Likely, this behaviour is driven by hunger due to the food restriction accompanying touchscreen testing11.
Consistently, median response latency was reduced in all CMS-exposed groups compared to controls. During task acquisition, vortioxetine responders displayed the shortest median response latency and controls the longest latency. Prolonged median response latency in the control group is consistent with a previous study25, suggesting increased cognitive appraisal, before executing a choice in control animals. Consequently, reduced response latency in the anhedonic and mainly in the vortioxetine treated groups can be considered as impulsive behaviour, executing a less evaluated, spontaneous choice. Reduced response latency may indicate impaired HPC functioning since inactivation of the dorsal HPC with lidocaine and scopolamine significantly shortened reaction time in the rat dPAL task as well34 and is in line with the important role of HPC in visuospatial learning tasks32,35. An alternative explanation might include a frontostriatal reorganization causing a shift from effortful, goal-directed to habitual behaviour. Such changes have been observed after stress exposure36 and might explain the reduced response latency observed in the present study. Noticeably, responders to vortioxetine treatment displayed the shortest response latency of all groups suggesting an association between treatment response and decreased appraisal.
A shift to habit-like or impulsive behaviour is further supported by the number of redundant screen touches per trial. Consistently, vortioxetine treated rats executed more redundant touches than any other group. Thus, vortioxetine seems to increase impulsive or compulsive behaviour. This lack of inhibitory control may suggest impairments in executive function associated with the PFC37.
In order to address long-term memory, accuracy was re-tested after a 10-day hiatus subsequent to passing dPAL. Vortioxetine responders decreased most in accuracy after the 10-day hiatus and performed significantly worse than low-responders. Hence, a high response to vortioxetine treatment was associated with reduced memory performance. Interestingly, only the control group restored performance to the dPAL passing criterion level (≥ 80% accuracy) on the second day of retention. All other groups still performed below 80% accuracy and the anhedonic-like group even decreased in accuracy on the second day of retention.
Altered cerebral gene expression associated with vortioxetine treatment and hedonic state
Expression levels of genes regulated in neuropsychiatric diseases or associated with neuronal plasticity were measured in the PFC, dorsal and ventral HPC. Cofilin 1 is a key regulator in growth cone dynamics and, thus, in neuronal plasticity important for learning and memory38,39. In the PFC, Cofilin 1 expression was upregulated in anhedonic-like rats compared to controls. Excessive up- or down-regulation of Cofilin 1 was associated with impaired synaptic plasticity and learning deficits39. Thus, altered Cofilin 1 gene expression might suggest subthreshold cognitive impairments associated with anhedonia, especially in untreated rats.
DISC1 is a scaffolding protein involved in neurodevelopmental signalling and suggested as candidate gene in neuropsychiatric disorder40,41. In the present study, Disc1 gene expression levels were higher in the ventral compared to the dorsal HPC. In the ventral HPC, an interaction trend may indicate a regulatory association of the hedonic state and vortioxetine treatment on Disc1 gene expression. These changes support the literature that DISC1 dysregulation is involved in the pathology of mental illnesses including cognitive deficits and dendritic arborisation42,43.
DISC1 regulates downstream Gsk3b expression and in the present study, Gsk3b expression was upregulated in the ventral compared to the dorsal HPC, which might be linked to an increased Disc1 gene expression. Gsk3b expression is known to be inhibited by most antidepressant treatments, e.g. SSRIs, and a dysregulation of Gsk3b expression is suggested to be implicated in depression44,45,46,47. Gsk3b upregulation is associated with impairments in spatial memory, attention and long-term potentiation, which are all important elements in acquisition of the dPAL task48,49,50,51,52. Consequently, borderline increased Gsk3b gene expression levels in the dorsal HPC in the present study may underlie the observed memory impairments during dPAL retention in the vortioxetine responder group compared to low-responders.
Homer proteins, which are scaffolding proteins facilitating post-synaptic signalling, are vital for learning and memory functions53. Moreover, decreased Homer1 expression is associated with an enhanced stress response and susceptibility to psychiatric diseases such as MDD54,55. In the present study, Homer1 was higher expressed in the dorsal than in the ventral HPC, possibly in response to spatial learning required for dPAL acquisition56. In the dorsal HPC, Homer2 mRNA expression was decreased in rats with anhedonic phenotype (treated and untreated). Homer2 is required for alcohol-seeking57 and, thus, reduced seeking of reward in anhedonic-like rats may be reflected by decreased Homer2 levels. Although Homer3 was upregulated in rat frontal cortex in response to vortioxetine treatment (not correcting of multiple comparisons)58, only a trend of vortioxetine downregulating Homer3 expression in the ventral hippocampus was observed in the present study.
Bdnf is involved in neuronal plasticity59, a mechanism which might be upregulated by vortioxetine treatment58. Moreover, Bdnf expression levels are reduced following stress exposure as well as in PFC and HPC post-mortem tissue of MDD suicide victims60,61. Furthermore, antidepressant treatment elevates Bdnf levels and, in turn, treatment efficacy appears dependent on Bdnf levels62,63,64. Consequently, the trend of higher Bdnf levels in the dorsal HPC of vortioxetine treated animals is in accordance with the literature.
NR3C2 expression is an important player in the stress response, HPA axis activity and MDD. Increased NR3C2 function is associated with resilience, whereas decreased NR3C2 levels suggest stress-susceptibility for developing depression65. Hence, the anhedonic phenotype, i.e. susceptibility to CMS including a low treatment response to vortioxetine, might be linked to a reduced Nr3c2 expression in the HPC.
In future studies, it would be interesting to include gene expression profiling before start of behavioural testing as well as after or, alternatively, a behaviourally naïve, vortioxetine-treated group can be added to disentangle the effects of the learning paradigm from treatment effects.
Touchscreen testing
To our knowledge, this was the first touchscreen study to show that not only sweet rewards, such as sugar pellets or milkshakes, generate successful operant conditioning. This might become crucial in addiction, diabetes or reward studies and expands the applicability of touchscreen testing. Furthermore, continuous SCTs throughout the experiment revealed the impact of food reduction, treatment and appetitive touchscreen testing on rodents.
Conclusion
Our study expands on the relatively new drug treatment approach of antidepressants targeting depression-associated cognitive impairments. Hence, the effect of vortioxetine on the hedonic state, on cognition and selected gene expression was assessed. In contrast to a previous report (reviewed in Sanchez et al.22), we have shown that vortioxetine recovers the hedonic state in anhedonic-like rats and, hence, demonstrated its efficacy in a well-validated preclinical model of depression26,66,67. Moreover, cognitive performance was assessed with the touchscreen operant platform, which was developed with focus on its translational value. In the present study, the primary readouts did not reveal beneficial cognitive effects of vortioxetine treatment although it was observed that a higher number of treated rats managed to pass the dPAL task. Furthermore, effects on behavioural strategy was evident from secondary read-outs. The potential pro-cognitive effect of vortioxetine requires more detailed evaluation since the observed effects, such as shortened reaction time and a shift to habitual behaviour might be beneficial in a different context than what the dPAL touchscreen task is actually designed for addressing. Finally, the most pronounced alterations in the selected genes were in the dorsal versus the ventral HPC. However, it cannot be excluded that the learning paradigm has affected the gene expression profiles.
References
World Health Organisation. Depression-Fact sheet. World Health Organisation (2020). Available at: https://www.who.int/en/news-room/fact-sheets/detail/depression. (Accessed 26 April 2020).
Warden, D., Rush, J., Trivedi, M. H., Fava, M. & Wisniewski, S. R. The STAR*D project results: a comprehensive review of findings. Curr. Psychiatry Rep. 9, 449–459 (2007).
Conradi, H. J., Ormel, J. & de Jonge, P. Presence of individual (residual) symptoms during depressive episodes and periods of remission: a 3-year prospective study. Psychol. Med. 41, 1165–1174 (2011).
Darcet, F., Gardier, A. M., Gaillard, R., David, D. J. & Guilloux, J. P. Cognitive dysfunction in major depressive disorder. A translational review in animal models of the disease. Pharmaceuticals 9, 9 (2016).
Reppermund, S., Ising, M., Lucae, S. & Zihl, J. Cognitive impairment in unipolar depression is persistent and non-specific: further evidence for the final common pathway disorder hypothesis. Psychol. Med. 39, 603–614 (2009).
Rock, P. L., Roiser, J. P., Riedel, W. J. & Blackwell, A. D. Cognitive impairment in depression : a systematic review and meta-analysis. 44, 2029–2040 (2014).
Lee, R. S. C., Hermens, D. F., Porter, M. A. & Redoblado-Hodge, M. A. A meta-analysis of cognitive deficits in first-episode Major Depressive Disorder. J. Affect. Disord. 140, 113–124 (2012).
McIntyre, R. S. et al. Cognitive deficits and functional outcomes in major depressive disorder: determinants, substrates, and treatment interventions. Depress. Anxiety 30, 515–527 (2013).
Gudayol-Ferré, E., Guàrdia-Olmos, J. & Peró-Cebollero, M. Effects of remission speed and improvement of cognitive functions of depressed patients. Psychiatry Res. 226, 103–112 (2015).
Gonda, X. et al. The role of cognitive dysfunction in the symptoms and remission from depression. Ann. Gen. Psychiatry 14, 1–7 (2015).
Martis, L.-S., Brision, C., Holmes, M. C. & Wiborg, O. Resilient and depressive-like rats show distinct cognitive impairments in the touchscreen paired-associates learning (PAL) task. Neurobiol. Learn. Mem. 155, 287–296 (2018).
Gibb, A. & Deeks, E. D. Vortioxetine: first global approval. Drugs 74, 135–145 (2014).
McIntyre, R. S., Lophaven, S. & Olsen, C. K. A randomized, double-blind, placebo-controlled study of vortioxetine on cognitive function in depressed adults. Int. J. Neuropsychopharmacol. 17, 1557–1567 (2014).
Katona, C., Hansen, T. & Olsen, C. K. A randomized, double-blind, placebo-controlled, duloxetine-referenced, fixed-dose study comparing the efficacy and safety of Lu AA21004 in elderly patients with major depressive disorder. Int. Clin. Psychopharmacol. 27, 215–223 (2012).
Jensen, J. B. et al. Vortioxetine, but not escitalopram or duloxetine, reverses memory impairment induced by central 5-HT depletion in rats: Evidence for direct 5-HT receptor modulation. Eur. Neuropsychopharmacol. 24, 148–159 (2014).
Li, Y., Sanchez, C. & Gulinello, M. Memory impairment in old mice is differentially sensitive to different classes of antidepressants. Eur. Neuropsychopharmacol. 23, S282 (2013).
Li, Y. et al. Reversal of age-associated cognitive deficits is accompanied by increased plasticity-related gene expression after chronic antidepressant administration in middle-aged mice. Pharmacol. Biochem. Behav. 135, 70–82 (2015).
Mørk, A. et al. Pharmacological effects of Lu AA21004: a novel multimodal compound for the treatment of major depressive disorder. J. Pharmacol. Exp. Ther. 340, 666–675 (2012).
Mørk, A. et al. Vortioxetine (Lu AA21004), a novel multimodal antidepressant, enhances memory in rats. Pharmacol. Biochem. Behav. 105, 41–50 (2013).
Bondi, C. O., Rodriguez, G., Gould, G. G., Frazer, A. & Morilak, D. A. Chronic unpredictable stress induces a cognitive deficit and anxiety-like behavior in rats that is prevented by chronic antidepressant drug treatment. Neuropsychopharmacology 33, 320–331 (2008).
Jayatissa, M. N., Bisgaard, C., Tingström, A., Papp, M. & Wiborg, O. Hippocampal cytogenesis correlates to escitalopram-mediated recovery in a chronic mild stress rat model of depression. Neuropsychopharmacology 31, 2395–2404 (2006).
Sanchez, C., Asin, K. E. & Artigas, F. Vortioxetine, a novel antidepressant with multimodal activity: review of preclinical and clinical data. Pharmacol. Ther. 145, 43–47 (2015).
Bussey, T. J. et al. New translational assays for preclinical modelling of cognition in schizophrenia: the touchscreen testing method for mice and rats. Neuropharmacology 62, 1191–1203 (2012).
Hvoslef-Eide, M. et al. The NEWMEDS rodent touchscreen test battery for cognition relevant to schizophrenia. Psychopharmacology 232, 3853–3872 (2015).
Martis, L.-S. et al. The effect of rat strain and stress exposure on performance in touchscreen tasks. Physiol. Behav. 184, 83–90 (2018).
Wiborg, O. Chronic mild stress for modeling anhedonia. Cell Tissue Res. 354, 155–169 (2013).
Wallace, A., Pehrson, A. L., Sánchez, C. & Morilak, D. A. Vortioxetine restores reversal learning impaired by 5-HT depletion or chronic intermittent cold stress in rats. Int. J. Neuropsychopharmacol. 17, 1695–1706 (2014).
Allen, G. C., Qu, X. & Earnest, D. J. TrkB-deficient mice show diminished phase shifts of the circadian activity rhythm in response to light. Neurosci. Lett. 378, 150–155 (2005).
Schaaf, M. J. M., Duurland, R., De Kloet, E. R. & Vreugdenhil, E. Circadian variation in BDNF mRNA expression in the rat hippocampus. Mol. Brain Res. 75, 342–344 (2000).
Müller, H. K., Wegener, G., Popoli, M. & Elfving, B. Differential expression of synaptic proteins after chronic restraint stress in rat prefrontal cortex and hippocampus. Brain Res. 1385, 26–37 (2011).
Olejnik, S. & Algina, J. Generalized eta and omega squared statistics: measures of effect size for some common research designs. Psychol. Methods 8, 434–447 (2003).
Kim, C. H., Heath, C. J., Kent, B. A., Bussey, T. J. & Saksida, L. M. The role of the dorsal hippocampus in two versions of the touchscreen automated paired associates learning (PAL) task for mice. Psychopharmacology 232, 3899–3910 (2015).
Pehrson, A. L. et al. Task-and treatment length-dependent effects of vortioxetine on scopolamine-induced cognitive dysfunction and hippocampal extracellular acetylcholine in rats. J. Pharmacol. Exp. Ther. 358, 472–482 (2016).
Talpos, J. C., Winters, B. D., Dias, R., Saksida, L. M. & Bussey, T. J. A novel touchscreen-automated paired-associate learning (PAL) task sensitive to pharmacological manipulation of the hippocampus: a translational rodent model of cognitive impairments in neurodegenerative disease. Psychopharmacology 205, 157–168 (2009).
De Rover, M. et al. Hippocampal dysfunction in patients with mild cognitive impairment: a functional neuroimaging study of a visuospatial paired associates learning task. Neuropsychologia 49, 2060–2070 (2011).
Dias-Ferreira, E. et al. Chronic stress causes frontostriatal reorganization and affects decision-making. Science 325, 621–625 (2009).
Graybeal, C., Kiselycznyk, C. & Holmes, A. Stress-induced impairments in prefrontal-mediated behaviors and the role of the N-methyl-d-aspartate receptor. Neuroscience 211, 28–38 (2012).
Kuhn, T. B. et al. Regulating actin dynamics in neuronal growth cones by ADF/cofilin and rho family GTPases. J Neurobiol 44, 126–144 (2000).
Barone, E., Mosser, S. & Fraering, P. C. Inactivation of brain Cofilin-1 by age, Alzheimer’s disease and γ-secretase. Biochim. Biophys. Acta Mol. Basis Dis. 1842, 2500–2509 (2014).
Dahoun, T., Trossbach, S. V., Brandon, N. J., Korth, C. & Howes, O. D. The impact of Disrupted-in-Schizophrenia 1 (DISC1) on the dopaminergic system: a systematic review. Transl. Psychiatry 7, e1015 (2017).
Ishizuka, K., Paek, M., Kamiyaand, A. & Sawa, A. A review of disrupted-in-schizophrenia-1 (disc1): neurodevelopment, cognition, and mental conditions. Biol. Psychiatry 59, 1189–1197 (2006).
Austin, C. P., Ky, B., Ma, L., Morris, J. A. & Shughrue, P. J. Expression of disrupted-in-schizophrenia-1, a schizophrenia-associated gene, is prominent in the mouse hippocampus throughout brain development. Neuroscience 124, 3–10 (2004).
Gill, M., Donohoe, G. & Corvin, A. What have the genomics ever done for the psychoses?. Psychol. Med. 40, 529–540 (2010).
Miller, B. R. & Hen, R. The current state of the neurogenic theory of depression and anxiety. Curr. Opin. Neurobiol. 30, 51–58 (2015).
Eldar-Finkelman, H. & Martinez, A. GSK-3 inhibitors: preclinical and clinical focus on CNS. Front. Mol. Neurosci. 4, 1–18 (2011).
Ochs, S. M. et al. Loss of neuronal GSK3β reduces dendritic spine stability and attenuates excitatory synaptic transmission via β-catenin. Mol. Psychiatry 20, 482–489 (2015).
Jope, R. S. & Roh, M.-S. Glycogen synthase kinase-3 (GSK3) in psychiatric diseases and therapeutic interventions. Curr. Drug Targets 7, 1421–1434 (2006).
Beaulieu, J.-M., Gainetdinov, R. R. & Caron, M. G. Akt/GSK3 signaling in the action of psychotropic drugs. Annu. Rev. Pharmacol. Toxicol. 49, 327–347 (2009).
Engel, T. Full reversal of Alzheimer’s disease-like phenotype in a mouse model with conditional overexpression of glycogen synthase kinase-3. J. Neurosci. 26, 5083–5090 (2006).
Hernández, F., Borrell, J., Guaza, C., Avila, J. & Lucas, J. J. Spatial learning deficit in transgenic mice that conditionally over-express GSK-3β in the brain but do not form tau filaments. J. Neurochem. 83, 1529–1533 (2002).
Hooper, C. et al. Glycogen synthase kinase-3 inhibition is integral to long-term potentiation. Eur. J. Neurosci. 25, 81–86 (2007).
Li, X. et al. In vivo regulation of glycogen synthase kinase-3beta (GSK3beta) by serotonergic activity in mouse brain. Neuropsychopharmacology 29, 1426–1431 (2004).
Foa, L. & Gasperini, R. Developmental roles for Homer: more than just a pretty scaffold. J. Neurochem. 108, 1–10 (2009).
Leber, S. L. et al. Homer1a protein expression in schizophrenia, bipolar disorder, and major depression. J. Neural Transm. 124, 1261–1273 (2017).
Szumlinski, K. K., Kalivas, P. W. & Worley, P. F. Homer proteins: implications for neuropsychiatric disorders. Curr. Opin. Neurobiol. 16, 251–257 (2006).
Ménard, C. & Quirion, R. Successful cognitive aging in rats: A role for mGluR5 glutamate receptors, homer 1 proteins and downstream signaling pathways. PLoS ONE 7, e28666 (2012).
Szumlinski, K. K. Homer2 Is necessary for EtOH-induced neuroplasticity. J. Neurosci. 25, 7054–7061 (2005).
Waller, J. A. et al. Neuroplasticity pathways and protein-interaction networks are modulated by vortioxetine in rodents. BMC Neurosci. 18, 1–15 (2017).
Park, H. & Poo, M. M. Neurotrophin regulation of neural circuit development and function. Nat Rev Neurosci 14, 7–23 (2013).
Smith, M. A., Makino, S., Kvetnansky, R. & Post, R. M. Stress and glucocorticoids affect the expression of brain-derived neurotrophic factor and neurotrophin-3 mRNAs in the hippocampus. J. Neurosci. 15, 1768–1777 (1995).
Dwivedi, Y. et al. Altered gene expression of brain-derived neurotrophic factor and receptor tyrosine kinase B in postmortem brain of suicide subjects. Arch. Gen. Psychiatry 60, 804–815 (2003).
Chen, B., Dowlatshahi, D., MacQueen, G. M., Wang, J. F. & Young, L. T. Increased hippocampal BDNF immunoreactivity in subjects treated with antidepressant medication. Biol. Psychiatry 50, 260–265 (2001).
Castrén, E. & Rantamäki, T. The role of BDNF and its receptors in depression and antidepressant drug action: reactivation of developmental plasticity. Dev. Neurobiol. 70, 289–297 (2010).
Sen, S., Duman, R. & Sanacora, G. Serum brain-derived neurotrophic factor, depression, and antidepressant medications: meta-analyses and implications. Biol. Psychiatry 64, 527–532 (2008).
Ter Heegde, F., De Rijk, R. H. & Vinkers, C. H. The brain mineralocorticoid receptor and stress resilience. Psychoneuroendocrinology 52, 92–110 (2015).
Czéh, B., Fuchs, E., Wiborg, O. & Simon, M. Animal models of major depression and their clinical implications. Prog. Neuro-Psychopharmacology Biol. Psychiatry 64, 293–310 (2016).
Willner, P. The chronic mild stress (CMS) model of depression: history, evaluation and usage. Neurobiol. Stress 6, 78–93 (2017).
Acknowledgements
We gratefully acknowledge Stine Dhiin and Karina Lassen Holm for their help with the CMS model as well as Kim Henningsen for helping out with the CMS and the touchscreen testing. Special thanks to Bo Martin Bibby for his statistical advice. Furthermore, we are grateful for Per Fuglsang Mikkelsen assisting with the tissue collection and to Birgitte Hviid Mumm and Sanne Nordestgaard Andersen for their molecular work. L.-S.M. was supported by a scholarship from the Aarhus-Edinburgh Excellence in European Doctoral Education Project (ExEDE). M.C.H is a member of The University of Edinburgh Centre for Cognitive Ageing and Cognitive Epidemiology, part of the cross council Lifelong Health and Wellbeing Initiative (MR/K026992/1). Funding from the BBSRC and MRC is gratefully acknowledged.
Author information
Authors and Affiliations
Contributions
L.-S.M., M.C.H, B.E. & O.W. designed the study. L.-S.M. & K.H. performed the in vivo experiments and tissue collection. B.E. conducted the real-time qPCR experiments. L.-S.M. performed data analyses. L.-S.M wrote the first draft of the manuscript and all authors contributed to the final version of the manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Martis, LS., Højgaard, K., Holmes, M.C. et al. Vortioxetine ameliorates anhedonic-like behaviour and promotes strategic cognitive performance in a rodent touchscreen task. Sci Rep 11, 9113 (2021). https://doi.org/10.1038/s41598-021-88462-7
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-021-88462-7
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.