A Critical Role of Mitochondria in BDNF-Associated Synaptic Plasticity After One-Week Vortioxetine Treatment

A Critical Role of Mitochondria in BDNF-Associated Synaptic Plasticity After One-Week... Background: Preclinical studies have indicated that antidepressant effect of vortioxetine involves increased synaptic plasticity and promotion of spine maturation. Mitochondria dysfunction may contribute to the pathophysiological basis of major depressive disorder. Taking into consideration that vortioxetine increases spine number and dendritic branching in hippocampus CA1 faster than fluoxetine, we hypothesize that new spines induced by vortioxetine can rapidly form functional synapses by mitochondrial support, accompanied by increased brain-derived neurotrophic factor signaling. Methods: Rats were treated for 1 week with vortioxetine or fluoxetine at pharmacologically relevant doses. Number of synapses and mitochondria in hippocampus CA1 were quantified by electron microscopy. Brain-derived neurotrophic factor protein levels were visualized with immunohistochemistry. Gene and protein expression of synapse and mitochondria- related markers were investigated with real-time quantitative polymerase chain reaction and immunoblotting. Results: Vortioxetine increased number of synapses and mitochondria significantly, whereas fluoxetine had no effect after 1-week dosing. BDNF levels in hippocampus DG and CA1 were significantly higher after vortioxetine treatment. Gene expression levels of Rac1 after vortioxetine treatment were significantly increased. There was a tendency towards increased gene expression levels of Drp1 and protein levels of Rac1. However, both gene and protein levels of c-Fos were significantly decreased. Furthermore, there was a significant positive correlation between BDNF levels and mitochondria and synapse numbers. Conclusion: Our results imply that mitochondria play a critical role in synaptic plasticity accompanied by increased BDNF levels. Rapid changes in BDNF levels and synaptic/mitochondria plasticity of hippocampus following vortioxetine compared with fluoxetine may be ascribed to vortioxetine’s modulation of serotonin receptors. Keywords: synapse, mitochondria, BDNF, depression, vortioxetine Received: January 30, 2018; Revised: February 26, 2018; Accepted: March 2, 2018 © The Author(s) 2018. Published by Oxford University Press on behalf of CINP. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any 603 medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 604 | International Journal of Neuropsychopharmacology, 2018 Significance Statement Preclinical studies have indicated that the antidepressant effect of vortioxetine involves increased synaptic plasticity and pro- motion of spine maturation. Mitochondria dysfunction may contribute to the pathophysiological basis of major depressive dis- order. The interplay between mitochondria and synapses has been less well studied. In the present study, our results indicate that mitochondria play a critical role in synaptic plasticity accompanied by increasing BDNF levels. In particular, the coincidence of rapid changes in the BDNF levels and synaptic/mitochondria plasticity of hippocampus following vortioxetine compared with fluoxetine may be ascribed to vortioxetine’s modulation of one or more serotonin receptors. Moreover, gene and protein expression related to the link between the mitochondria and synapses may possibly offer a molecular mechanism explanation in response to vortioxetine treatment. changes to regulate synaptic activity. Importantly, the activity Introduction of synapses differentially influences the mitochondrial morph- The multimodal antidepressant vortioxetine is an antago- ology and distribution between axons and dendrites (Chang nist at 5-HT, 5-HT , and 5-HT receptors, a partial agonist at 3 7 1D et al., 2006). Therefore, mitochondria not only provide dynamic 5-HT receptors, an agonist at 5-HT receptors, and a seroto- 1B 1A energy support for normal synaptic functioning but also directly nin transporter (SERT) inhibitor (Sanchez et al., 2015). The pro- modulate synaptic structural and functional plasticity. cognitive effects of vortioxetine in preclinical rodent models The interplay between mitochondria and synapses has been and also improvement of certain aspects of cognitive function less well studied. Based on our previous observation that vorti- in major depressive disorder (MDD) patients have been docu- oxetine increased spine number and dendritic branching in the mented (Frampton, 2016; Li et  al., 2017). Preclinical studies in hippocampus CA1 faster than the SSRI fluoxetine, we hypoth- rodents indicate that vortioxetine’s effects on synaptic trans- esize that the concomitant function of mitochondria and brain- mission, neurogenesis, dendritic branching, and dendritic derived neurotrophic factor (BDNF)-signaling is necessary for spine maturation differ significantly from a selective serotonin rapid adaptive synaptic plasticity by vortioxetine. Identification reuptake inhibitor (SSRI) (Dale et  al., 2014W ; aller et  al., 2016). of molecular mechanisms related to the link between the mito- Furthermore, several lines of evidence reveal that vortioxetine chondria and synapses in response to vortioxetine treatment can promote expression of various genes that play a role in syn- may aid in a better understanding of their therapeutic capacity. aptic plasticity (du Jardin et al., 2016 W; aller et al., 2016). We have previously reported that 1-week vortioxetine treatment induced Materials and Methods changes in spine number and dendritic morphology, whereas an equivalent dose of fluoxetine had no effects (Chen et al., 2016). Animals In the central nervous system, most excitatory postsynaptic ter - minals reside in dendritic spines. Spine remodeling is usually Adult male Sprague-Dawley rats (180–200  g) (n = 42) w ere kept associated with changes in synaptic connectivity. However, we on a normal light:dark cycle with free access to food and water. do not know whether the rapid induction of new spines by vorti- There were 6/8 rats (6 for morphological and 8 for molecular oxetine form functional synapses and react to presynaptic stim- studies) in each group (vortioxetine, fluoxetine, and vehicle). All ulation. Therefore, we conducted further studies to demonstrate rats were handled for 7 d before any treatment was initiated. changes of excitatory synapses located on dendritic spines after vortioxetine treatment using electron microscopy. Antidepressant Treatment Moreover, several lines of preclinical and clinical evidence indicate that vortioxetine can enhance cognitive function via Vortioxetine was given in chow (1.6  g/kg food, provided by modulation of a wide range of neurotransmitters (Sanchez et al., H. Lundbeck A/S) for 7 d. This dosage has previously been shown 2015; Pehrson et al., 2016; Li et al., 2017; Pan et al., 2017; Smith to produce >80% SERT occupancy (Chen et al., 2016). Fluoxetine, et  al., 2017). Pathophysiological processes in MDD are intim- also provided by H. Lundbeck A/S, was delivered in the drinking ately linked to the biological underpinnings of cognitive loss water (160 mg/L) for 7 d. This dose was chosen to achieve >80% (Anderson, 2018). The various pathophysiological changes, such SERT occupancy with fluoxetine (Chen et al., 2016). Regular food as increased oxidative stress, occurring over the course of neu- pellets with the same composition were given to vehicle rats for roprogression in MDD may lead to suboptimal mitochondrial the same period. Drug delivery via the chow ensures a stable functioning (Anderson, 2018; Czarny et  al., 2018). Suboptimal target engagement and removed any stressors associated with mitochondrial functioning may be the primary driver of the dosing. Body weight, food intake, and fluid intake were meas- pathophysiological changes and emerging cognitive deficits in ured regularly. All procedures are approved by the Danish ani- MDD (Anderson, 2018). Substantial data support that MDD was mal ethics committee (2012-15-2934-00254; C –sheet 1). accompanied by oxidative stress dysregulation, which could indicate a dysfunction of mitochondria, and that antidepressant Tissue Preparation and Sampling treatment may reduce oxidative stress (Alcocer-Gomez et  al., Rats were perfused transcardially 24 h after treatment with fixa- 2014; Klinedinst and Regenold, 2015). Increased production of reactive oxygen species (ROS) associated with age- and disease- tives (4% paraformaldehyde and 2% glutaraldehyde in 0.1 M PBS). Brains were removed and postfixed in the same fixative, and dependent loss of mitochondrial function, altered metal homeo- stasis, and reduced antioxidant defense directly affect synaptic kept at 4°C until further processing. Hippocampi were isolated, and left or right hippocampus was selected randomly, manually activity and neurotransmission in neurons leading to cognitive dysfunction (Vavakova et al., 2015Czarn ; y et al., 2018). Synapses straightened along the septotemporal axis to diminish its natu- ral curvature, embedded in 5% agar, and cut into 65-µm-thick are dynamic, and mitochondria constantly move along axons and dendrites by dividing and fusing in response to synaptic sections perpendicularly to its longest axis on a Vibratome 3000. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Chen et al. | 605 Three sets of sections were chosen based on a systematic ran- from spines by a less densely stained cytoplas containing micro- dom sampling principle and a section sampling fraction of 1/15. tubules and mitochondria. The spine synapses may be subdivided One set was stained with thionin for estimating the volume of into perforated and nonperforated synapses. Perforated synapses hippocampus with light microscopy and immunohistochemi- displayed discontinuous or perforated PSD profiles, whereas non- cally staining for BDNF was performed on a second set. The last perforated synapses exhibited continuous PSD files in all con- set of tissue sections for electron microscopy was embedded in secutive sections (Figure 1) (Geinisman et al., 2001). TAAB 812 Epon (TAAB) for cutting 20 consecutive serial ultrathin The synapse number density was estimated using the PSD sections. The actual mean ultrathin section thickness (around as a counting unit. At least 2 neuropil fields were randomly pho- 67–72 nm) was determined according to Small’s method of mini- tographed on each ultrathin section. From each section series mal folds (Small JV, 1968). (16–20 sections), 10 consecutive sections were selected from sec- tion 2 (section 1 was look-up section) and used as the reference section of the disectors. The last 5 sections of each series were Ultrastructural Study of Synapses and Mitochondria used as look-up sections to ensure that all counted PSDs were The ultrathin sections were observed in a FEI Morgagni included in their entirety in the section series. Axo-spinous per - transmission electron microscope. Electron micrographs forated synapses and shaft synapses were counted with ~120 were taken with a digital camera at an initial magnification disectors and axo-spinous nonperforated synapses with ~48 of 10 500×, and digitally enlarged to a final magnification of disectors in each animal. The total synapse number was esti- 23 850×. The synapses and mitochondria were sampled by mated as the product of the synapse number density and vol- the physical disector and analyzed using the digitized elec- ume of the CA1 stratum radiatum. Detailed information can be tron micrographs via iTEM software (Olympus Soft Imaging found in our previous paper (Chen et al., 2010). Solutions GmbH) (Chen et al., 2010). Mitochondria were counted throughout the neuropil and The synapses were identified primarily on the basis of the specifically in the axon terminals and dendrites. The criteria presence of a postsynaptic density (PSD) with vesicles in close for identifying mitochondria were the presence of distinctive proximity to the presynaptic zone. Only spine and shaft syn- cristae and a double membrane (Figure  1). Neuropil structures apses of asymmetric synapses were analyzed in this study. In the were identified as axon terminals (presence of 3 or more synap- spine synapses, PSD located on the spines that were small spher - tic vesicles), dendrites (postsynaptic to a synapse or having an ical postsynaptic protrusions of the dendrite filled with a clear attached spine), or astroglial processes (presence of fibrils and cytoplasmic matrix and a distinct spine apparatus but without watery cytoplasm). The total number of mitochondria in neu- mitochondria and microtubules. The shaft synapses terminated ropil and the number of mitochondria in axon terminals and directly on the dendritic shaft. The dendrites were differentiated dendrites were determined. Figure 1. Count of synapses and mitochondria in serial sections. Electron micrographs of consecutive ultrathin sections (a–d) showed nonperforated synapses (black arrow), a perforated synapse (large white arrow), and shaft synapse (large black arrow). The postsynaptic spine exhibited PSD discontinuities (black stars). Mitochondria: Isolated particle (I), A branch dividing (white stars), or branches connection (B). Scale bar, 0.5 μm. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 606 | International Journal of Neuropsychopharmacology, 2018 Combining the disector principle with the object’s 0.3% H O in TBS for 7  min and washed by TBS 3 times for 2 2 3D Euler number estimates the number of mitochondria 10  min. Sections were then mounted on the gelatin-coated (Kroustrup and Gundersen, 2001). In terms of disector section slides and dehydrated with alcohol gradient and cleared with pairs, mitochondria profiles in one section plane are com- xylene. pared with mitochondria profiles on the next section plane. Images of whole immunostained sections were taken by the Mitochondria are identified in each section plane, and a Virtual Slides System VS120 (Olympus) including the software change between planes is deduced as being 1 of 2 significant VS ASW OIL 2.7 (Olympus Soft Imaging Solutions GmbH). ImageJ possibilities: a new isolated part, a so-called “Island, , or a ” I software was used to analyze the images of immunostained sec- new connection between isolated mitochondria, a “Bridge,” tions and calculated the mean optical density of BDNF-positive B, in an existing mitochondrion. The total Euler number, areas in subregions of hippocampus (Figure 2). Σx, contribution from all disectors is obtained as the signed sum of Islands, and Bridges (Figure  1). The total mitochon- Measurement of mRNA Levels with Real-time dria number was estimated as the product of the mitochon- Quantitative Polymerase Chain Reaction dria number density and volume of CA1 stratum radiatum. (Real-Time qPCR) Detailed information can be found in our previous paper The rats were decapitated and the brain collected. Briefly, (Chen et al., 2013). coronal brain sections (200  µm) were cut using a cryostat at -20ºC. Using a scalpel, the CA1 area was dissected and stored Immunohistochemistry at -80°C until extraction of RNA with the Qiagen Rneasy Free-floating 8–9 vibratome sections from each animal were Mini kit (Qiagen) as previously described (Silva Pereira et al., washed 3 times for 10  min in Tris-buffered saline (TBS) (pH 2017). The dissected CA1 tissues were homogenized in Lysis 7.4), immersed in endogenous peroxidase blocking solution buffer (Qiagen) with a mixer-mill (Retsch; twice for 40 s at 30 for 30 min, and were incubated in preheated Target Retrieval Hz/s). Total RNA was isolated following the manufacturer’s solution at 85°C for 40  min (Dako, EnVision System HRP). instructions (Qiagen). The RNA concentration, A260/280, and Tissue sections were then incubated at 4°C overnight in a A260/230 ratios were determined with a NanoDrop spec- solution containing the rabbit anti-BDNF polyclonal antibody trometer (Thermo Fisher Scientific). Afterwards, RNA was (diluted 1: 500) (AB1779, Merck Millipore). The next day, sec- reversely transcribed using random primers and Superscript tions were washed 3 times for 10  min with buffer (1% BSA IV Reverse Transcriptase (Invitrogen) following the manu- and 0.3% Triton-X in TBS) and incubated in buffer (1% BSA in facturer’s instructions. The input RNA concentration was TBS) added goat anti-rabbit IgG (1:200) for 2  h at room tem- adjusted to 29  ng/µL in the individual samples. The cDNA perature. Sections were washed 3 times for 10 min in TBS and samples were stored undiluted at -80°C until real-time qPCR then visualized with 0.1% 3, 3’-diaminobenzidine containing analysis. Figure  2. (A) BDNF expression levels were examined by immunohistochemistry in the subregions of hippocampus. (B–D) Immunohistochemistry-examined BDNF expression levels in each group. Mean optical density (MOD) was calculated with the following formula: = lo OD g10(max pixel intensity/mean pixel intensity), where max pixel intensity = 255. MOD in the vortioxetine-treated group significantly increased compared with the vehicle and fluoxetine groups in DG (A) and CA1 (C) sub- regions of hippocampus. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Chen et al. | 607 Signaling #2250; 1:500), and rabbit anti-β-actin (Licor 926–42212; Real-Time qPCR 1:3000) overnight at 4°C followed by incubation with the appro- The samples were diluted 1:10 with DEPC water before being priate IRDye conjugated secondary antibody for 1 h at RT: IRDye used as a qPCR template. The real-time qPCR reactions were 800CW goat anti-mouse IgG, IRDye 680RD goat anti-rabbit, or carried out in 96-well PCR-plates using an Mx3005P (Stratagene) IRDye 800CW goat-anti rabbit IgG, at 1:15 000 dilution (Licor). and SYBR Green. The gene expression of synapse-related genes Infrared signals were detected using the Odyssey CLx infrared (Arc, Cdc42, c-fos, Cofilin1, Homer1, Psd95, Rac1, RhoA, Spinophilin, imaging system, and bands were quantified using Image Studio Synapsin1), mitochondria-related genes (Drp1F , is1, Mfn1, Mfn2, software (LI-COR Biosciences). Band intensities were normalized Opa1), as well as 8 different reference genes (18s rRNA, ActB, to β-actin levels within the same lane. CycA, Gapdh, Hmbs, Hprt1, Rpl13A, Ywhaz) was investigated. The reference genes were selected as previously described Statistical Procedures (Bonefeld et  al., 2008). The primers were designed and tested as in our previous work (Elfving et  al., 2008). Essential gene- All data were subjected to 1-way ANOVA to compare treatment specific data about primer sequence and amplicon sizes is responses followed by posthoc tests (Tukey and least signifi- given in Supplement Table  1. The primers were obtained from cant difference) and Dunnett’s test for multiple comparisons Sigma-Aldrich. Each SYBR Green reaction (10  µL total volume) (Immunoblotting). P < .05 was considered statistically significant. contained 1x SYBR Green master mix (Sigma-Aldrich), 0.5  µM Statistical analyses and graphical representations of the find- primer pairs, and 3 µL of diluted cDNA and were carried out as ings were carried out using SPSS11 (SPSS Corp) and Sigmaplot described previously (Elfving et al., 2008 2010 , ). The mixture was 10 (SYSTAT Inc.) software. heated initially to 95°C for 3 min to activate hot-start iTaq DNA polymerase and then 40 cycles with denaturation at 95°C for Results 10  s, annealing at 60°C for 30  s, and extension at 72°C for 60  s were applied. To verify that only one PCR product was detected, BDNF Expression Levels by Immunohistochemistry the samples were subjected to a heat dissociation protocol; after the final cycle of the PCR, the reactions were heat-denatured by BDNF immunoreactivity was mainly localized in the intra-cyto- increasing the temperature from 60°C to 95°C. All samples and plasm of neurons. Mean optical density of BDNF in DG and CA1 the standard curve were run in duplicates. A standard curve was (P < .05) of hippocampus was significantly higher in rats treated generated on each plate. with vortioxetine compared with vehicle rats (Figure 2). There was Initially, the mRNA levels were determined for the 8 refer - no difference between the fluoxetine-treated rats and vehicle rats. ence genes. Stability comparison of the expression of the refer - ence genes was then conducted with the Normfinder software. The Volume of Hippocampus and Hippocampal CA1 Values of the target genes were subsequently normalized with Stratum Radiatum the geometric mean of the 2 optimal reference genes (Ywhaz and Hmbs), based on the NormFinder mathematical algorithm The volume of whole hippocampus and hippocampal CA1-SR (Andersen et al., 2004). in the vortioxetine group was significantly increased compared with the vehicle group (P < .05) (Figure 3; Table ).1 There was no difference in fluoxetine-treated rats compared with vehicle rats. Immunoblotting Aliquots (20 µg total protein) of hippocampal CA1 tissue homog- The Number of Synapses in Serial Sections enized in Cell Lysis Buffer (Bio-Rad) containing 1x protease inhibitor cocktail (Roche) were separated on 10% criterion TGX The total number of synapses (P < .001), nonperforated spine gels (Bio-Rad), transferred to nitrocellulose membranes, blocked synapses (P < .01), and perforated spine synapses (P < .01) was in odyssey blocking buffer (Licor), and probed with the primary significantly increased in the vortioxetine group compared with antibodies: mouse anti-Rac1 (Abcam ab33186; 1:500), mouse the fluoxetine and vehicle groups. No changes in the number of anti-Drp1 (Cell Signaling #14647; 1:1000), rabbit anti-c-Fos (Cell shaft synapses were observed (Figure 4; Table ). 1 Figure 3. The volume of hippocampus and CA1 stratum radium. The volume of hippocampus and CA1-SR in vortioxetine group is significantly increased compared with the vehicle and fluoxetine groups (*P < .05). Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 608 | International Journal of Neuropsychopharmacology, 2018 Figure 4. The number of synapses including subtypes of synapse in CA1 (*P < .05; ** P < .01; ***P < .001) (∆: vehicle group; ▲: vortioxetine; ▽: fluoxetine). (A) The total num- ber of synapses was significantly higher in the vortioxetine group compared with the vehicle and fluoxetine groups. (B) The number of nonperforated spine synapses was significantly higher in vortioxetine group compared with the vehicle and fluoxetine groups. (C) The number of perforated spine synapses was significantly higher in the vortioxetine group compared with the vehicle and fluoxetine groups. (D) The number of shaft synapses was not any different in the vortioxetine group compared with the vehicle and fluoxetine groups. treatment relative to vehicle. There was a tendency towards The Number of Mitochondria in Serial Sections and increase of protein expression levels of Rac1 in the vortioxetine Mitochondria Volume treatment group compared with vehicle group. No changes of The mitochondria number in total neuropil (P < .05) and axon the protein levels of Rac1 in the fluoxetine treatment group terminal (P < .05) was significantly increased in the vortioxetine were observed. There were no significant differences in protein group compared with fluoxetine and vehicle groups. There were expression levels of Drp1 in the vortioxetine and fluoxetine no significant differences in mitochondria volume and the den- treatment groups compared with vehicle group (Figure 7). drites’ mitochondria number (Figure 5; Table). 1 Correlations between Synapses, Mitochondria, Synapse- and Mitochondria-Related Gene Volume, and BDNF Level in the Hippocampus Expression and Protein Levels following Antidepressant Treatment Real-time qPCR analysis revealed that Rac1 in the vortioxetine Correlations of morphological data and BDNF values are shown treatment group was significantly increased compared with the in Figure 8. BDNF levels were positively correlated with the vol- vehicle group (P < .05) (Figure 6). There was no significant differ - ume of hippocampal CA1-SR (r = 0.80; P < .001), synapse number ence in mRNA levels of Rac1 in the fluoxetine treatment group (r = 0.78; P < .001), and mitochondrial number (r = 0.76; P < .001). compared with the vehicle group. There was a tendency towards Moreover, the volume of hippocampal CA1-SR was positively increased mRNA levels of Drp1 in the vortioxetine treatment correlated with synapse number (r = 0.78; P < .001) and mitochon- group compared with vehicle group. The mRNA expression lev- drial number (r = 0.61; P < .001). Furthermore, there was a positive els of the immediate early gene (IEG), c-fos, were significantly correlation between total mitochondria number and total num- reduced following vortioxetine (P < .01) and fluoxetine (P < .05) ber of synapses (r = 0.81; P < .001). treatment relative to vehicle; however, neither vortioxetine nor fluoxetine had any effect on Arc mRNA expression. The other Discussion selected synapse-related and mitochondria-related genes given in Supplement Table 1 were not regulated. Here, we have shown that vortioxetine significantly increased The protein expression levels of c-Fos were significantly the number of synapses and mitochondria accompanied by reduced following vortioxetine (P < .05) and fluoxetine (P < .01) BDNF level elevation, whereas fluoxetine had no effect after 7 Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Chen et al. | 609 Figure 5. The number of mitochondria in the various structures (neuropil, axons, and dendrites) and the mean volume of mitochondria in CA1 (*, < .05) (∆: v P ehicle group; ▲: vortioxetine; ▽ : fluoxetine). (A) The total number of mitochondria in neuropil displayed a significant increase in the vortioxetine group compared with the vehicle and fluoxetine groups. (B) The number of mitochondria in the axon terminal also showed a significant increase in the vortioxetine group compared with the vehicle and fluox- etine groups. (C) The number of mitochondria in dendrites was not significantly different in the vortioxetine group compared with the vehicle and fluoxetine groups. (D) The mean volume of mitochondria in CA1 stratum radiatum was not significantly different in the vortioxetine group compared with the vehicle and fluoxetine groups. Figure  6. mRNA expression investigated with real-time qPCR after 1-week treatment with fluoxetine and vortioxetine. Plotted data show normalized mean group values± SEM and are expressed as percent of control rats. Asterisks represent significant differences from control rats (*P < .05). = 8 in eac n h group. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 610 | International Journal of Neuropsychopharmacology, 2018 Figure  7. Protein expression levels of Rac1 and Drp1 in hippocampal total lysate. (A) Representative immunoblots of Rac1, Drp1, and c-Fos protein expression in response to vortioxetine and fluoxetine treatment. (B–D) Bar graphs representing the β-actin normalized density of Rac1, Drp1, and c-Fos (n = 8 rats/gr oup). Data are expressed as mean percentage± SEM of control mean values (*P < .05, **P < .01, 1-way ANOVA followed by Dunnett’s test for multiple comparisons). Figure 8. The correlations between BDNF, synaptic plasticity, and mitochondria plasticity of hippocampus (*** < .001). P The volume of hippocampal CA1-SR correlated positively with synapse number (A) and mitochondrial number (B). The levels of BDNF expression correlated positively with the volume of hippocampal CA1-SR (D), synapse number (E), and mitochondrial number (F). Furthermore, there was a strong significant positive correlation between the total mitochondria number density and total number of synapses (C). d treatment. The gene expression level of Rac1 was significantly correlation between BDNF level and mitochondria/synapse increased and Drp1 showed a tendency towards an increase in number. Based on our previous observation that vortioxetine the vortioxetine treatment group. Moreover, there was a ten- increases spine number and dendritic branching in the hippo- dency towards increase of protein expression levels of Rac1 campus CA1 faster than the SSRI fluoxetine, the present study after vortioxetine treatment by immunoblotting. However, supports the hypothesis that the new spines induced by vorti- both gene and protein expression levels of c-Fos were signifi- oxetine can rapidly form functional synapses by mitochondrial cantly decreased. Furthermore, there was a significant positive support, accompanied by increased BDNF-signaling. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Chen et al. | 611 Table 1. Summary of the Overall Measurement Results of Ultrastructure Veh vs Vor Veh vs Flu Vehicle Vortioxetine Fluoxetine P P Total volume (hippocampus) 34.870 (6.907) 42.373 (4.162) 35.553 (6.301) <.05 .86 V(CA1) 3.961 (0.748) 4.736 (0.403) 4.233 (0.393) <.05 .44 Total N(syn/CA1) x10 (9) 8.684 (1.490) 12.4511 (0.472) 9.498 (1.488) <.001 .36 N(np-syn) x10 (9) 6.296 (1.135) 8.9710 (1.055) 6.978 (0.907) <.01 .28 N(p-syn) x10 (9) 1.982 (0.361) 3.027 (0.779) 2.242 (0.907) <.05 .45 N(sh-syn) x10 (9) 0.406 (0.109) 0.513 (0.1511) 0.277 (0.153) .11 .12 Total N(mit/CA1) x10 (9) 3.438 (1.483) 5.657 (1.105) 3.802 (1.307) <.05 .66 N(a-mit)x10 (9) 2.366 (0.957) 3.895 (0.656) 2.752 (1.052) <.01 .52 N(d-mit)x10 (9) 1.065 (0.521) 1.745 (0.556) 1.048 (0.349) .05 .95 VN(mit/CA1) μm 0.080(0.018) 0.062(0.015) 0.069(0.013) .11 .29 Total V(mit/CA1) mm 0.257(0.071) 0.255(0.012) 0.252(0.043) <.05 .880 Abbreviations: N(a-mit), number of mitochondria in axons; N(d-mit), number of mitochondria in dendrites; N(mit/CA1), number of mitochon- dria in CA1; N(np-syn), number of nonperforated synapses; N(p-syn), the number of perforated synapses; N(sh-syn), the number of shaft syn- apses; N(syn/CA1), total number of synapses in CA1; V(CA1), volume of CA1 stratum radiatum; V(mit/CA1), total volume of mitochondria in CA1; V (mit/CA1), mean volume of mitochondria. documented a role for Rac-regulators in hippocampal neurogen- Effect of Vortioxetine on Hippocampal Synaptic esis and synaptogenesis and shown that dysregulation leads to Plasticity and Synapse-Related Gene Expression cognitive impairment. Thus, Rho GTPases and their downstream Several lines of evidence suggest that, in addition to neurogen- effectors may represent important therapeutic targets for dis- esis, more rapid synaptic plasticity may play an important role orders associated with cognitive dysfunction. Interestingly, the in the neurobiology of depression and effects of antidepressant role of Rac1 in modulation of synapses is not limited to hippo- therapy (Popoli et  al., 2002Duman, ; 2004). The current results campus and reported in other brain regions including nucleus showed that the number of synapses (including both nonper - accumbens (Golden et  al., 2013). A  recent study indicated the forated and perforated synapses) significantly increased in the impact of vortioxetine on synaptic integration in prefrontal-sub- vortioxetine group compared with the vehicle group. However, cortical circuits by showing diminished mPFC-msNAc afferent there were no changes in the fluoxetine group compared with drive (Chakroborty et al., 2017). the vehicle group at this time point. Previously, we have reported Emerging findings have suggested that BDNF plays a critical that 1-week vortioxetine treatment induced changes in spine role as a regulator of neuronal development and function, through number and density and also dendritic morphology, whereas an Rac1-mediated remodeling of the actin cytoskeleton (Zhou et al., equivalent dose of fluoxetine had no effects (Chen et al., 2016). 2007) and the defect in BDNF-induced spine morphogenesis is a Our recent findings implied that induction of synapses was asso- result of impaired activation of Rac1 (Lai et al., 2012). Consistent ciated with the creation of new spines at adjacent sites in the with these findings, our results indicate that the morphological dendrite. Therefore, the increased number of synapses is corre- changes of synapses to some extent may rely on Rac1-dependent lated with changes of spine number and dendritic morphology. signaling cascade accompanied by BDNF elevation. Previous experiments indicated that vortioxetine enhanced LTP The IEGs, such as c-fos, has been widely used as a molecular in the CA1 subregion of the hippocampus (Dale et al., 2014). marker tightly associated with synaptic plasticity (Minatohara Furthermore, these morphological changes were associated et al., 2015). The transcription factor IEGs c-fos is considered as with the improvement of depression-like behavior in the forced a good candidate for the initial steps of learning inducing long- swim test (FST) according to our previous behavioral results. term synaptic plasticity (Abraham et  al., 1991). This indicates Li et  al. (Li et  al., 2013) used a progesterone withdrawal model that the threshold of synaptic activation inducing the expres- in female Long-Evans rats to investigate the effects of several sion of transcription factor genes in particular brain regions classes of antidepressants (including vortioxetine and fluoxet- can be closely linked to synaptic plasticity. In the present study, ine) in the FST as a measure of antidepressant activity. Acute c-fos was downregulated in response to vortioxetine treatment. (3 injections over 2  days) treatment results showed that vorti- Consistent with these results, 4 weeks chronic vortioxetine can oxetine significantly reduced immobility in the FST, whereas promote a decrease in c-fos expression in the hippocampus fluoxetine was ineffective. Therefore, our findings support that (Waller et  al., 2016). Although several lines of evidence reveal synaptic plasticity may play an important role in the effects of that vortioxetine can promote expression of various genes that antidepressant therapy. play a role in synaptic plasticity (Li et al., 2015 du J ; ardin et al., Evidence suggests that Rac1, a well-known Rho GTPase, con- 2016; Waller et al., 2017), these studies measured mRNA expres- tributes to the regulation of dendritic spine formation, excitatory sion and were not focused on a particular layer of hippocampus, synapses, and synaptic function (Penzes et al., 2003Cala ; brese as well as by species differences in the response to vortioxetine. et al., 2006; Oh et al., 2010). The longer dendritic spines in hip- pocampal CA1 pyramidal neurons of Srgap 3-/- mice were found Effect of Vortioxetine on Hippocampal Mitochondrial through increasing Rac1 activity by negatively regulating Srgap3 Plasticity and Mitochondria-Related Gene Expression (Waltereit et al., 2012). The definitive evidence showed compel- ling links between Rho GTPases signaling and cognitive function Synaptic transmission requires mitochondrial ATP generation (De et al., 2014). Zamboni et al. (Zamboni et al., 2016) have also for neurotransmitter exocytosis, vesicle recruitment, activation Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 612 | International Journal of Neuropsychopharmacology, 2018 of ion conductance, signaling at metabotropic receptors, potenti- Interestingly, several studies have demonstrated that fluox- ation of neurotransmitter release, and synaptic plasticity (Li etine induces mitochondria triggering apoptosis and/or inter - et  al., 2010; Jiao and Li, 2011). Mitochondrial dynamics, a cor - feres with mitochondrial function by modulating the activity porate term for the processes of mitochondrial fission, fusion, of respiratory chain components and enzymes of the Krebs and transport, controls mitochondrial function and localization cycle. Furthermore, fluoxetine alters mitochondria-related within the cell (Detmer and Chan, 2007). Mitochondria con- redox aspects in different experimental models (de Oliveira, stantly move along axons and dendrites, dividing and fusing 2016). However, there was no significant effect of fluoxetine in response to synaptic changes (Mattson et  al., 2008 P; almer on hippocampal mitochondrial plasticity and mitochondria- et  al., 2011), and help to regulate synaptic activity and conse- related genes expression in this study. This discrepancy may quently learning and memory (Li et al., 2010 Jiao and Li, ; 2011). be explained by the fact that the dosage and time of fluoxetine Mitochondria, therefore, not only provide dynamic energy sup- administration in the present study was different compared port for normal synaptic functioning but also directly modulate with the previous reports (Adzic et al., 2013). synaptic structural and functional plasticity (Cheng et al., 2010). Our results suggest that the increased mitochondrial number 5-HT Receptor Activities of Vortioxetine provided ATP to support features of synaptogenesis. Therefore, mitochondrial biogenesis may play an important physiological Recent studies demonstrated that vortioxetine prevented the 5-HT-induced increase in inhibitory post-synaptic potentials role in synaptic plasticity in the hippocampus. Mitochondria are dynamically transported in and out of recorded from CA1 pyramidal cells, most likely by 5-HT recep- tor antagonism (Dale et  al., 2014). Furthermore, 5-HT receptor axons and dendrites to maintain neuronal and synaptic func- tion. It is shown that the movement of mitochondria in axons antagonists exhibited memory-enhancing properties in a num- ber of preclinical cognition models (Arnsten et al., 1997 Pitsikas ; and dendrites are different due to the movement of mitochon- dria in axons with a consistently rapid velocity than are those and Borsini, 1996). A  selective 5-HT R agonist (flesinoxan) and 1A a selective 5-HT receptor antagonist (ondansetron) reversed in dendrites (Ligon and Steward, 2000). Therefore, the difference in motility and metabolic properties of mitochondria in axons memory impairments in 5-HT depleted rats (Jensen et al., 2014). Thus, accumulating evidence suggests that both 5-HT and and dendrites reflects alterations in energy metabolism during 1A synaptic plasticity. Since most metabolic activity takes place in 5-HT receptors play an important role in the effect of vortiox- etine on cognitive function. axon terminals (Zinsmaier et al., 2009), an increased number of mitochondria in axon terminals after vortioxetine treatment in A selective 5-HT receptor antagonist, Tropisetron, protected PC12 cells against high glucose-induced apoptosis and oxida- our study implies that the increased metabolism may support high synaptic activity by the generation of action potentials and tive stress by preventing mitochondrial pathways (Aminzadeh, 2017). Oxidative stress results from an imbalance between the trafficking of synaptic vesicles (neurotransmitter exocytosis and vesicle recruitment). production of ROS and antioxidant capacity (Uttara et al., 2009). ROS could increase the gating potential of mitochondrial pores Regulation of mitochondrial dynamics is partly accomplished through posttranslational modification of mitochondrial fission and contribute to cytochrome c release (Sharifi et al., 2007). The and fusion enzymes, in turn influencing mitochondrial bioen- 5-HT receptor agonist promoted axonal transport of mitochon- 1A ergetics and transport (Cheng et  al., 2010). The importance of dria in hippocampus neurons by activating the Akt- GSK3β path- posttranslational regulation is highlighted in numerous neuro- way (Chen et  al., 2007). GSK3β is a common therapeutic target degenerative disorders associated with posttranslational modi- of many antidepressant drugs (Jope and Roh, 2006). Thus, it is fication of the mitochondrial fission enzyme Drp1 (Li et al., 2004). suggested that aberrant 5-HT-induced mitochondrial movement One study showed that mitochondrial abnormality including may be linked to depression, and effects of 5-HT on mitochon- morphological impairment is due to the imbalance between dria functionality may be mediated by both 5-HT and 5-HT 1A 3 mitochondrial fusion and fission via a glycogen synthase kinase receptors. Furthermore, we suggest that vortioxetine’s 5-HT 1A 3β (GSK3β)/dynamin-related protein-1 (Drp1)-dependent mech- receptor agonism and 5-HT receptor antagonism may con- anism (Huang et al., 2015). tribute to cognitive function through mitochondrial biogenesis Furthermore, increasing mitochondrial fragmentation by based on our present findings. overexpression of Drp1 enhances synapse formation, whereas dominant-negative inhibition of Drp1 has the opposite effect The Relationship between BDNF, Synaptic Plasticity, in cultured hippocampal neurons, indicating a prominent and Mitochondria Plasticity of Hippocampus in role for mitochondrial dynamics in synaptogenesis (Dickey Depression and Strack, 2011). Our finding showing a tendency towards increased levels of Drp1 gene is consistent with their findings. Emerging findings suggest roles for mitochondria as media- But, there was no significant change of protein levels of Drp1. tors of at least some of the effects of glutamate and BDNF on However, it was not certain whether a vortioxetine-induced synaptic plasticity (Shuttleworth et  al., 2003 Lu et  ; al., 2009). increase in Drp1 gene expression was associated with a simi- Several neurotrophic factors including BDNF have been shown lar enhancement on protein level of Drp1 and the functional to promote neuronal differentiation, survival, and modifying activity of the protein. Moreover, correlation analysis in large- synaptic plasticity (Duman and Monteggia, 2006 K; ermani and scale data sets reported approximately 50% correspondence Hempstead, 2007). BDNF promotes synaptic transmission and between messenger RNA and protein levels (Tian et al., 2004). plasticity, in part by increasing mitochondrial energy produc- Thus, the gene expression of Drp1 in the present study may tion (Markham et  al., 2014). Consequences of abnormalities in possibly offer a mechanistic explanation for increased mito- the BDNF and glucocorticoid receptor signaling are impairment chondria number induced by vortioxetine. Furthermore, a link of mitochondrial respiration efficiency and synaptic plastic- between Drp1 and mitochondria plasticity may indicate that ity (Jeanneteau and Arango-Lievano, 2016). A  study of BNDF- there is an increase of mitochondrial fission after vortioxetine induced mitochondrial motility arrest and presynaptic docking treatment. suggests that mitochondrial transport and distribution play Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Chen et al. | 613 an essential role in BDNF-mediated synaptic transmission (Su Stochastic Geometry and Advanced Bioimaging. Dr. Wegener et al., 2014). The neuroprotective effect of BDNF shares molecu- reported having received lecture/consultancy fees from lar signaling pathways (MEK-Bcl-2 pathway) with mitochondrial H.  Lundbeck A/S, Servier SA, Astra Zeneca AB, Eli Lilly A/S, respiratory coupling (Markham et al., 2012). Sun Pharma Pty Ltd, Pfizer Inc, Shire A/S, HB Pharma A/S, Arla More importantly, the ability of mitochondria in regulation Foods A.m.b.A., Alkermes Inc, and Mundipharma International 2+ of Ca clearance in connection with BDNF signaling pathways Ltd., and research funding from the Danish Medical Research is a key point of modulating the synaptic plasticity, especially Council, Aarhus University Research Foundation (AU-IDEAS ini- synaptogenesis (Markham et  al., 2014). Consistent with these tiative (eMOOD)), the Novo Nordisk Foundation, the Lundbeck findings, our results also indicated that BDNF has a significant Foundation, and EU Horizon 2020 (ExEDE). Drs Ardalan, Danladi, positive correlation with synapse and mitochondrial number. Elfving and Müller reported no biomedical financial interests or The mitochondria are essential components in synaptic trans- potential conflicts of interest. mission, since they are a major source of energy required for main- tenance and restoration of ion gradients (Todorova and Blokland, References 2017). High levels of monocarboxylate (Gerhart et  al., 1998) and glucose (Gerhart et al., 1991) transporters have been observed in Abraham WC, Dragunow M, Tate WP (1991) The role of immedi- CA1, suggesting elevated metabolic and synaptic activity. Several ate early genes in the stabilization of long-term potentiation. studies suggest that physical proximity between mitochondria Mol Neurobiol 5:297–314. and synapses is regulated by neuronal activity (Courchet et  al., Adzic M, Lukic I, Mitic M, Djordjevic J, Elaković I, Djordjevic A, 2013; Sheng, 2014). Our results showed that increased total syn- Krstic-Demonacos M, Matić G, Radojcic M (2013) Brain region- apse numbers are strongly positively correlated with total mito- and sex-specific modulation of mitochondrial glucocorticoid chondria number. The mitochondrial changes may be crucially receptor phosphorylation in fluoxetine treated stressed rats: linked to changed energy metabolism and, therefore, may have effects on energy metabolism. Psychoneuroendocrinology consequences for cell plasticity, resilience, and survival in patients 38:2914–2924. with MDD. Antidepressants might ultimately enhance energy Alcocer-Gómez E, de Miguel M, Casas-Barquero N, Núñez-Vasco metabolism and reduce the damage of oxidative stress. However, J, Sánchez-Alcazar JA, Fernández-Rodríguez A, Cordero MD because our experimental model is in vivo, it is difficult to show (2014) NLRP3 inflammasome is activated in mononuclear directly the role of mitochondria in synaptic plasticity. Further in blood cells from patients with major depressive disorder. vitro studies would be necessary to elucidate how mitochondria Brain Behav Immun 36:111–117. modulate synaptic plasticity, e.g., changes of dendritic mitochon- Aminzadeh A (2017) Protective effect of tropisetron on high glu- dria content or mitochondrial activity. cose induced apoptosis and oxidative stress in PC12 cells: In conclusion, our study suggests that mitochondria play a roles of JNK, P38 mapks, and mitochondria pathway. Metab critical role in synaptic plasticity accompanied by increasing Brain Dis 32:819–826. BDNF levels. In particular, the coincidence of rapid changes in the Andersen CL, Jensen JL, Ørntoft TF (2004) Normalization of real- BDNF levels and synaptic/mitochondria plasticity of hippocam- time quantitative reverse transcription-PCR data: a model- pus following vortioxetine compared with fluoxetine may be based variance estimation approach to identify genes suited ascribed to vortioxetine’s modulation of one or more serotonin for normalization, applied to bladder and colon cancer data receptors. A newly published patent (US 9820984 B1) documents sets. Cancer Res 64:5245–5250. that a treatment regimen of a single i.v. dose of vortioxetine fol- Anderson G (2018) Linking the biological underpinnings of lowed by oral vortioxetine achieves a faster onset of therapeutic depression: role of mitochondria interactions with melatonin, action in MDD patients (Sanchez et al., 2017). Further studies are inflammation, sirtuins, tryptophan catabolites, DNA repair necessary to clarify how BDNF levels trigger changes in mito- and oxidative and nitrosative stress, with consequences for chondria and synapses. These findings might provide insights in classification and cognition. Prog Neuropsychopharmacol to how to target specific 5-HT receptors to induce mitochondrial Biol Psychiatry 80:255–266. biogenesis and may provide new avenues for development of Arnsten AF, Lin CH, Van Dyck CH, Stanhope KJ (1997) The effects novel therapeutics targeting the biogenesis response. of 5-HT3 receptor antagonists on cognitive performance in aged monkeys. Neurobiol Aging 18:21–28. Bonefeld BE, Elfving B, Wegener G (2008) Reference genes for nor - Supplementary Material malization: a study of rat brain tissue. Synapse 62:302–309. Supplementary data are available at International Journal of Calabrese B, Wilson MS, Halpain S (2006) Development and regu- lation of dendritic spine synapses. Physiology (Bethesda) Neuropsychopharmacology online. 21:38–47. Chakroborty S, Geisbush TR, Dale E, Pehrson AL, Sánchez C, West Acknowledgments AR (2017) Impact of vortioxetine on synaptic integration in prefrontal-subcortical circuits: comparisons with escitalo- Herdis Krunderup and Lone Lysgaard are gratefully acknowl- pram. Front Pharmacol 8:764. edged for their skillful EM technical assistance. Chang DT, Honick AS, Reynolds IJ (2006) Mitochondrial traf- ficking to synapses in cultured primary cortical neurons. J Statement of Interest Neurosci 26:7035–7045. Dr. Chen reports having received salary support from Chen F, Madsen TM, Wegener G, Nyengaard JR (2010) Imipramine H.  LundbeckA/S. Dr. Ardalan reported having salary support treatment increases the number of hippocampal synapses from Lundbeck Foundation. Connie Sanchez was full-time and neurons in a genetic animal model of depression. employee at H.  Lundbeck A/S when the study was conducted. Hippocampus 20:1376–1384. Dr. Nyengaard reports having received research funding from Chen F, Wegener G, Madsen TM, Nyengaard JR (2013) Sino-Danish Center and the Villum Foundation via Centre for Mitochondrial plasticity of the hippocampus in a genetic rat Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 614 | International Journal of Neuropsychopharmacology, 2018 model of depression after antidepressant treatment. Synapse AJ, Gonzalez-Maeso J, Neve RL, Turecki G, Ghose S, Tamminga 67:127–134. CA, Russo SJ (2013) Epigenetic regulation of RAC1 induces Chen F, du Jardin KG, Waller JA, Sanchez C, Nyengaard JR, synaptic remodeling in stress disorders and depression. Nat Wegener G (2016) Vortioxetine promotes early changes in Med 19:337–344. dendritic morphology compared with fluoxetine in rat hippo- Huang S, Wang Y, Gan X, Fang D, Zhong C, Wu L, Hu G, Sosunov campus. Eur Neuropsychopharmacol 26:234–245. AA, McKhann GM, Yu H, Yan SS (2015) Drp1-mediated mito- Chen S, Owens GC, Crossin KL, Edelman DB (2007) Serotonin chondrial abnormalities link to synaptic injury in diabetes stimulates mitochondrial transport in hippocampal neurons. model. Diabetes 64:1728–1742. Mol Cell Neurosci 36:472–483. Jeanneteau F, Arango-Lievano M (2016) Linking mitochondria to Cheng A, Hou Y, Mattson MP (2010) Mitochondria and neuroplas- synapses: new insights for stress-related neuropsychiatric ticity. ASN Neuro 2:e00045. disorders. Neural Plast 2016:3985063. Courchet J, Lewis TL Jr, Lee S, Courchet V, Liou DY, Aizawa S, Jensen JB, du Jardin KG, Song D, Budac D, Smagin G, Sanchez Polleux F (2013) Terminal axon branching is regulated by the C, Pehrson AL (2014) Vortioxetine, but not escitalopram or LKB1-NUAK1 kinase pathway via presynaptic mitochondrial duloxetine, reverses memory impairment induced by cen- capture. Cell 153:1510–1525. tral 5-HT depletion in rats: evidence for direct 5-HT receptor Czarny P, Wigner P, Galecki P, Sliwinski T (2018) The interplay modulation. Eur Neuropsychopharmacol 24:148–159. between inflammation, oxidative stress, DNA damage, DNA Jiao S, Li Z (2011) Nonapoptotic function of BAD and BAX in repair and mitochondrial dysfunction in depression. Prog long-term depression of synaptic transmission. Neuron Neuropsychopharmacol Biol Psychiatry 80:309–321. 70:758–772. Dale E, Zhang H, Leiser SC, Xiao Y, Lu D, Yang CR, Plath N, Jope RS, Roh MS (2006) Glycogen synthase kinase-3 (GSK3) in Sanchez C (2014) Vortioxetine disinhibits pyramidal cell func- psychiatric diseases and therapeutic interventions. Curr tion and enhances synaptic plasticity in the rat hippocam- Drug Targets 7:1421–1434. pus. J Psychopharmacol 28:891–902. Kermani P, Hempstead B (2007) Brain-derived neurotrophic De Filippis B, Romano E, Laviola G (2014) Aberrant rho gtpases factor: a newly described mediator of angiogenesis. Trends signaling and cognitive dysfunction: in vivo evidence for a Cardiovasc Med 17:140–143. compelling molecular relationship. Neurosci Biobehav Rev Klinedinst NJ, Regenold WT (2015) A mitochondrial bioenergetic 46:285–301. basis of depression. J Bioenerg Biomembr 47:155–171. de Oliveira MR (2016) Fluoxetine and the mitochondria: a review Kroustrup JP, Gundersen HJ (2001) Estimating the number of of the toxicological aspects. Toxicol Lett 258:185–191. complex particles using the conneulor principle. J Microsc Detmer SA, Chan DC (2007) Functions and dysfunctions of mito- 203:314–320. chondrial dynamics. Nat Rev Mol Cell Biol 8:870–879. Lai KO, Wong AS, Cheung MC, Xu P, Liang Z, Lok KC, Xie H, Palko Dickey AS, Strack S (2011) PKA/AKAP1 and PP2A/bβ2 regulate ME, Yung WH, Tessarollo L, Cheung ZH, Ip NY (2012) Trkb neuronal morphogenesis via drp1 phosphorylation and phosphorylation by cdk5 is required for activity-depend- mitochondrial bioenergetics. J Neurosci 31:15716–15726. ent structural plasticity and spatial memory. Nat Neurosci du Jardin KG, Müller HK, Sanchez C, Wegener G, Elfving B (2016) 15:1506–1515. A single dose of vortioxetine, but not ketamine or fluoxetine, Li Y, Raaby KF, Sánchez C, Gulinello M (2013) Serotonergic recep- increases plasticity-related gene expression in the rat frontal tor mechanisms underlying antidepressant-like action in cortex. Eur J Pharmacol 786:29–35. the progesterone withdrawal model of hormonally induced Duman RS (2004) Neural plasticity: consequences of stress and depression in rats. Behav Brain Res 256:520–528. actions of antidepressant treatment. Dialogues Clin Neurosci Li Y, Abdourahman A, Tamm JA, Pehrson AL, Sánchez C, Gulinello 6:157–169. M (2015) Reversal of age-associated cognitive deficits is Duman RS, Monteggia LM (2006) A neurotrophic model for accompanied by increased plasticity-related gene expression stress-related mood disorders. Biol Psychiatry 59:1116–1127. after chronic antidepressant administration in middle-aged Elfving B, Bonefeld BE, Rosenberg R, Wegener G (2008) Differential mice. Pharmacol Biochem Behav 135:70–82. expression of synaptic vesicle proteins after repeated elec- Li Y, Sanchez C, Gulinello M (2017) Distinct antidepressant- troconvulsive seizures in rat frontal cortex and hippocam- like and cognitive effects of antidepressants with different pus. Synapse 62:662–670. mechanisms of action in middle-aged female mice. Int J Elfving B, Plougmann PH, Wegener G (2010) Differential brain, Neuropsychopharmacol 20:510–515. but not serum VEGF levels in a genetic rat model of depres- Li Z, Okamoto K, Hayashi Y, Sheng M (2004) The importance of sion. Neurosci Lett 474:13–16. dendritic mitochondria in the morphogenesis and plasticity Frampton JE (2016) Vortioxetine: a review in cognitive dysfunc- of spines and synapses. Cell 119:873–887. tion in depression. Drugs 76:1675–1682. Li Z, Jo J, Jia JM, Lo SC, Whitcomb DJ, Jiao S, Cho K, Sheng M (2010) Geinisman Y, Berry RW, Disterhoft JF, Power JM, Van der Zee EA Caspase-3 activation via mitochondria is required for long- (2001) Associative learning elicits the formation of multiple- term depression and AMPA receptor internalization. Cell synapse boutons. J Neurosci 21:5568–5573. 141:859–871. Gerhart DZ, Djuricic B, Drewes LR (1991) Quantitative immuno- Ligon LA, Steward O (2000) Movement of mitochondria in the cytochemistry (image analysis) of glucose transporters in the axons and dendrites of cultured hippocampal neurons. J normal and postischemic rodent hippocampus. J Cereb Blood Comp Neurol 427:340–350. Flow Metab 11:440–448. Lu CW, Lin TY, Chiang HS, Wang SJ (2009) Facilitation of glutam- Gerhart DZ, Enerson BE, Zhdankina OY, Leino RL, Drewes LR ate release from rat cerebral cortex nerve terminal by suban- (1998) Expression of the monocarboxylate transporter MCT2 esthetic concentration propofol. Synapse 63:773–781. by rat brain glia. Glia 22:272–281. Markham A, Cameron I, Bains R, Franklin P, Kiss JP, Schwendimann Golden SA, Christoffel DJ, Heshmati M, Hodes GE, Magida J, Davis L, Gressens P, Spedding M (2012) Brain-derived neurotrophic K, Cahill ME, Dias C, Ribeiro E, Ables JL, Kennedy PJ, Robison factor-mediated effects on mitochondrial respiratory Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Chen et al. | 615 coupling and neuroprotection share the same molecular sig- expression in the prefrontal cortex of adult male rats. Eur J nalling pathways. Eur J Neurosci 35:366–374. Pharmacol 815:304–311. Markham A, Bains R, Franklin P, Spedding M (2014) Changes in Small JV (1968) Measurement of section thickness. [Fourth mitochondrial function are pivotal in neurodegenerative and European Conference on Electron Microscopy]:609–610. psychiatric disorders: how important is BDNF? Br J Pharmacol Smith J, Browning M, Conen S, Smallman R, Buchbjerg J, Larsen 171:2206–2229. KG, Olsen CK, Christensen SR, Dawson GR, Deakin JF, Hawkins Mattson MP, Gleichmann M, Cheng A (2008) Mitochondria in neu- P, Morris R, Goodwin G, Harmer CJ (2017) Vortioxetine reduces roplasticity and neurological disorders. Neuron 60:748–766. BOLD signal during performance of the N-back working Minatohara K, Akiyoshi M, Okuno H (2015) Role of immediate- memory task: a randomised neuroimaging trial in remitted early genes in synaptic plasticity and neuronal ensembles depressed patients and healthy controls. Mol Psychiatry. underlying the memory trace. Front Mol Neurosci 8:78. Su B, Ji YS, Sun XL, Liu XH, Chen ZY (2014) Brain-derived neuro- Oh D, et  al. (2010) Regulation of synaptic rac1 activity, long-term trophic factor (BDNF)-induced mitochondrial motility arrest potentiation maintenance, and learning and memory by BCR and and presynaptic docking contribute to BDNF-enhanced syn- ABR rac gtpase-activating proteins. J Neurosci 30:14134–14144. aptic transmission. J Biol Chem 289:1213–1226. Palmer CS, Osellame LD, Stojanovski D, Ryan MT (2011) The regu- Tian Q, Stepaniants SB, Mao M, Weng L, Feetham MC, Doyle MJ, lation of mitochondrial morphology: intricate mechanisms Yi EC, Dai H, Thorsson V, Eng J, Goodlett D, Berger JP, Gunter B, and dynamic machinery. Cell Signal 23:1534–1545. Linseley PS, Stoughton RB, Aebersold R, Collins SJ, Hanlon WA, Pan Z, Grovu RC, Cha DS, Carmona NE, Subramaniapillai Hood LE (2004) Integrated genomic and proteomic analyses M, Shekotikhina M, Rong C, Lee Y, McIntyre RS (2017) of gene expression in mammalian cells. Mol Cell Proteomics Pharmacological treatment of cognitive symptoms in 3:960–969. major depressive disorder. CNS Neurol Disord Drug Targets Todorova V, Blokland A (2017) Mitochondria and synaptic 16:891–899. plasticity in the mature and aging nervous system. Curr Pehrson AL, Hillhouse TM, Haddjeri N, Rovera R, Porter JH, Mørk Neuropharmacol 15:166–173. A, Smagin G, Song D, Budac D, Cajina M, Sanchez C (2016) Uttara B, Singh AV, Zamboni P, Mahajan RT (2009) Oxidative Task- and treatment length-dependent effects of vortioxe- stress and neurodegenerative diseases: a review of upstream tine on scopolamine-induced cognitive dysfunction and hip- and downstream antioxidant therapeutic options. Curr pocampal extracellular acetylcholine in rats. J Pharmacol Exp Neuropharmacol 7:65–74. Ther 358:472–482. Vavakova M, Durackova Z, Trebaticka J (2015) Markers of oxida- Penzes P, Beeser A, Chernoff J, Schiller MR, Eipper BA, Mains RE, tive stress and neuroprogression in depression disorder. Oxid Huganir RL (2003) Rapid induction of dendritic spine morpho- Med Cell Longev 2015:898393. genesis by trans-synaptic ephrinb-ephb receptor activation Waller JA, Chen F, Sánchez C (2016) Vortioxetine promotes mat- of the rho-GEF kalirin. Neuron 37:263–274. uration of dendritic spines in vitro: a comparative study in Pitsikas N, Borsini F (1996) Itasetron (DAU 6215)  prevents age- hippocampal cultures. Neuropharmacology 103:143–154. related memory deficits in the rat in a multiple choice avoid- Waller JA, Tamm JA, Abdourahman A, Pehrson AL, Li Y, Cajina M, ance task. Eur J Pharmacol 311:115–119. Sánchez C (2017) Chronic vortioxetine treatment in rodents Popoli M, Gennarelli M, Racagni G (2002) Modulation of synap- modulates gene expression of neurodevelopmental and plas- tic plasticity by stress and antidepressants. Bipolar Disord ticity markers. Eur Neuropsychopharmacol 27:192–203. 4:166–182. Waltereit R, et al (2012) Srgap3⁻/⁻ mice present a neurodevelop- Sanchez C, Asin KE, Artigas F (2015) Vortioxetine, a novel anti- mental disorder with schizophrenia-related intermediate depressant with multimodal activity: review of preclinical phenotypes. Faseb J 26:4418–4428. and clinical data. Pharmacol Ther 145:43–57. Zamboni V, Armentano M, Sarò G, Ciraolo E, Ghigo A, Germena Sanchez C, Søby KK, Bang-Andersen B (2017) Dosing regimens G, Umbach A, Valnegri P, Passafaro M, Carabelli V, Gavello for fast onset of antidepressant effect. United States Patent US D, Bianchi V, D’Adamo P, de Curtis I, El-Assawi N, Mauro A, 9820984 B1 Nov. 21, 2017. Priano L, Ferri N, Hirsch E, Merlo GR (2016) Disruption of arh- Sharifi K, Mostaghni K, Maleki M, Badiei K (2007) Ischaemia/rep- gap15 results in hyperactive rac1, affects the architecture and erfusion injury in experimentally induced abomasal volvulus function of hippocampal inhibitory neurons and causes cog- in sheep. Vet Res Commun 31:575–590. nitive deficits. Sci Rep 6:34877. Sheng ZH (2014) Mitochondrial trafficking and anchoring in neu- Zhou P, Porcionatto M, Pilapil M, Chen Y, Choi Y, Tolias KF, Bikoff rons: new insight and implications. J Cell Biol 204:1087–1098. JB, Hong EJ, Greenberg ME, Segal RA (2007) Polarized signaling Shuttleworth CW, Brennan AM, Connor JA (2003) NAD(P)H fluor - endosomes coordinate BDNF-induced chemotaxis of cerebel- escence imaging of postsynaptic neuronal activation in m-ur lar precursors. Neuron 55:53–68. ine hippocampal slices. J Neurosci 23:3196–3208. Zinsmaier KE, Babic M, Russo GJ (2009) Mitochondrial transport Silva Pereira V, Elfving B, Joca SRL, Wegener G (2017) Ketamine dynamics in axons and dendrites. Results Probl Cell Differ and aminoguanidine differentially affect Bdnf and Mtor gene 48:107–139. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png International Journal of Neuropsychopharmacology Oxford University Press

A Critical Role of Mitochondria in BDNF-Associated Synaptic Plasticity After One-Week Vortioxetine Treatment

Free
13 pages

Loading next page...
 
/lp/ou_press/a-critical-role-of-mitochondria-in-bdnf-associated-synaptic-plasticity-iu1wb745k7
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of CINP.
ISSN
1461-1457
eISSN
1469-5111
D.O.I.
10.1093/ijnp/pyy022
Publisher site
See Article on Publisher Site

Abstract

Background: Preclinical studies have indicated that antidepressant effect of vortioxetine involves increased synaptic plasticity and promotion of spine maturation. Mitochondria dysfunction may contribute to the pathophysiological basis of major depressive disorder. Taking into consideration that vortioxetine increases spine number and dendritic branching in hippocampus CA1 faster than fluoxetine, we hypothesize that new spines induced by vortioxetine can rapidly form functional synapses by mitochondrial support, accompanied by increased brain-derived neurotrophic factor signaling. Methods: Rats were treated for 1 week with vortioxetine or fluoxetine at pharmacologically relevant doses. Number of synapses and mitochondria in hippocampus CA1 were quantified by electron microscopy. Brain-derived neurotrophic factor protein levels were visualized with immunohistochemistry. Gene and protein expression of synapse and mitochondria- related markers were investigated with real-time quantitative polymerase chain reaction and immunoblotting. Results: Vortioxetine increased number of synapses and mitochondria significantly, whereas fluoxetine had no effect after 1-week dosing. BDNF levels in hippocampus DG and CA1 were significantly higher after vortioxetine treatment. Gene expression levels of Rac1 after vortioxetine treatment were significantly increased. There was a tendency towards increased gene expression levels of Drp1 and protein levels of Rac1. However, both gene and protein levels of c-Fos were significantly decreased. Furthermore, there was a significant positive correlation between BDNF levels and mitochondria and synapse numbers. Conclusion: Our results imply that mitochondria play a critical role in synaptic plasticity accompanied by increased BDNF levels. Rapid changes in BDNF levels and synaptic/mitochondria plasticity of hippocampus following vortioxetine compared with fluoxetine may be ascribed to vortioxetine’s modulation of serotonin receptors. Keywords: synapse, mitochondria, BDNF, depression, vortioxetine Received: January 30, 2018; Revised: February 26, 2018; Accepted: March 2, 2018 © The Author(s) 2018. Published by Oxford University Press on behalf of CINP. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any 603 medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 604 | International Journal of Neuropsychopharmacology, 2018 Significance Statement Preclinical studies have indicated that the antidepressant effect of vortioxetine involves increased synaptic plasticity and pro- motion of spine maturation. Mitochondria dysfunction may contribute to the pathophysiological basis of major depressive dis- order. The interplay between mitochondria and synapses has been less well studied. In the present study, our results indicate that mitochondria play a critical role in synaptic plasticity accompanied by increasing BDNF levels. In particular, the coincidence of rapid changes in the BDNF levels and synaptic/mitochondria plasticity of hippocampus following vortioxetine compared with fluoxetine may be ascribed to vortioxetine’s modulation of one or more serotonin receptors. Moreover, gene and protein expression related to the link between the mitochondria and synapses may possibly offer a molecular mechanism explanation in response to vortioxetine treatment. changes to regulate synaptic activity. Importantly, the activity Introduction of synapses differentially influences the mitochondrial morph- The multimodal antidepressant vortioxetine is an antago- ology and distribution between axons and dendrites (Chang nist at 5-HT, 5-HT , and 5-HT receptors, a partial agonist at 3 7 1D et al., 2006). Therefore, mitochondria not only provide dynamic 5-HT receptors, an agonist at 5-HT receptors, and a seroto- 1B 1A energy support for normal synaptic functioning but also directly nin transporter (SERT) inhibitor (Sanchez et al., 2015). The pro- modulate synaptic structural and functional plasticity. cognitive effects of vortioxetine in preclinical rodent models The interplay between mitochondria and synapses has been and also improvement of certain aspects of cognitive function less well studied. Based on our previous observation that vorti- in major depressive disorder (MDD) patients have been docu- oxetine increased spine number and dendritic branching in the mented (Frampton, 2016; Li et  al., 2017). Preclinical studies in hippocampus CA1 faster than the SSRI fluoxetine, we hypoth- rodents indicate that vortioxetine’s effects on synaptic trans- esize that the concomitant function of mitochondria and brain- mission, neurogenesis, dendritic branching, and dendritic derived neurotrophic factor (BDNF)-signaling is necessary for spine maturation differ significantly from a selective serotonin rapid adaptive synaptic plasticity by vortioxetine. Identification reuptake inhibitor (SSRI) (Dale et  al., 2014W ; aller et  al., 2016). of molecular mechanisms related to the link between the mito- Furthermore, several lines of evidence reveal that vortioxetine chondria and synapses in response to vortioxetine treatment can promote expression of various genes that play a role in syn- may aid in a better understanding of their therapeutic capacity. aptic plasticity (du Jardin et al., 2016 W; aller et al., 2016). We have previously reported that 1-week vortioxetine treatment induced Materials and Methods changes in spine number and dendritic morphology, whereas an equivalent dose of fluoxetine had no effects (Chen et al., 2016). Animals In the central nervous system, most excitatory postsynaptic ter - minals reside in dendritic spines. Spine remodeling is usually Adult male Sprague-Dawley rats (180–200  g) (n = 42) w ere kept associated with changes in synaptic connectivity. However, we on a normal light:dark cycle with free access to food and water. do not know whether the rapid induction of new spines by vorti- There were 6/8 rats (6 for morphological and 8 for molecular oxetine form functional synapses and react to presynaptic stim- studies) in each group (vortioxetine, fluoxetine, and vehicle). All ulation. Therefore, we conducted further studies to demonstrate rats were handled for 7 d before any treatment was initiated. changes of excitatory synapses located on dendritic spines after vortioxetine treatment using electron microscopy. Antidepressant Treatment Moreover, several lines of preclinical and clinical evidence indicate that vortioxetine can enhance cognitive function via Vortioxetine was given in chow (1.6  g/kg food, provided by modulation of a wide range of neurotransmitters (Sanchez et al., H. Lundbeck A/S) for 7 d. This dosage has previously been shown 2015; Pehrson et al., 2016; Li et al., 2017; Pan et al., 2017; Smith to produce >80% SERT occupancy (Chen et al., 2016). Fluoxetine, et  al., 2017). Pathophysiological processes in MDD are intim- also provided by H. Lundbeck A/S, was delivered in the drinking ately linked to the biological underpinnings of cognitive loss water (160 mg/L) for 7 d. This dose was chosen to achieve >80% (Anderson, 2018). The various pathophysiological changes, such SERT occupancy with fluoxetine (Chen et al., 2016). Regular food as increased oxidative stress, occurring over the course of neu- pellets with the same composition were given to vehicle rats for roprogression in MDD may lead to suboptimal mitochondrial the same period. Drug delivery via the chow ensures a stable functioning (Anderson, 2018; Czarny et  al., 2018). Suboptimal target engagement and removed any stressors associated with mitochondrial functioning may be the primary driver of the dosing. Body weight, food intake, and fluid intake were meas- pathophysiological changes and emerging cognitive deficits in ured regularly. All procedures are approved by the Danish ani- MDD (Anderson, 2018). Substantial data support that MDD was mal ethics committee (2012-15-2934-00254; C –sheet 1). accompanied by oxidative stress dysregulation, which could indicate a dysfunction of mitochondria, and that antidepressant Tissue Preparation and Sampling treatment may reduce oxidative stress (Alcocer-Gomez et  al., Rats were perfused transcardially 24 h after treatment with fixa- 2014; Klinedinst and Regenold, 2015). Increased production of reactive oxygen species (ROS) associated with age- and disease- tives (4% paraformaldehyde and 2% glutaraldehyde in 0.1 M PBS). Brains were removed and postfixed in the same fixative, and dependent loss of mitochondrial function, altered metal homeo- stasis, and reduced antioxidant defense directly affect synaptic kept at 4°C until further processing. Hippocampi were isolated, and left or right hippocampus was selected randomly, manually activity and neurotransmission in neurons leading to cognitive dysfunction (Vavakova et al., 2015Czarn ; y et al., 2018). Synapses straightened along the septotemporal axis to diminish its natu- ral curvature, embedded in 5% agar, and cut into 65-µm-thick are dynamic, and mitochondria constantly move along axons and dendrites by dividing and fusing in response to synaptic sections perpendicularly to its longest axis on a Vibratome 3000. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Chen et al. | 605 Three sets of sections were chosen based on a systematic ran- from spines by a less densely stained cytoplas containing micro- dom sampling principle and a section sampling fraction of 1/15. tubules and mitochondria. The spine synapses may be subdivided One set was stained with thionin for estimating the volume of into perforated and nonperforated synapses. Perforated synapses hippocampus with light microscopy and immunohistochemi- displayed discontinuous or perforated PSD profiles, whereas non- cally staining for BDNF was performed on a second set. The last perforated synapses exhibited continuous PSD files in all con- set of tissue sections for electron microscopy was embedded in secutive sections (Figure 1) (Geinisman et al., 2001). TAAB 812 Epon (TAAB) for cutting 20 consecutive serial ultrathin The synapse number density was estimated using the PSD sections. The actual mean ultrathin section thickness (around as a counting unit. At least 2 neuropil fields were randomly pho- 67–72 nm) was determined according to Small’s method of mini- tographed on each ultrathin section. From each section series mal folds (Small JV, 1968). (16–20 sections), 10 consecutive sections were selected from sec- tion 2 (section 1 was look-up section) and used as the reference section of the disectors. The last 5 sections of each series were Ultrastructural Study of Synapses and Mitochondria used as look-up sections to ensure that all counted PSDs were The ultrathin sections were observed in a FEI Morgagni included in their entirety in the section series. Axo-spinous per - transmission electron microscope. Electron micrographs forated synapses and shaft synapses were counted with ~120 were taken with a digital camera at an initial magnification disectors and axo-spinous nonperforated synapses with ~48 of 10 500×, and digitally enlarged to a final magnification of disectors in each animal. The total synapse number was esti- 23 850×. The synapses and mitochondria were sampled by mated as the product of the synapse number density and vol- the physical disector and analyzed using the digitized elec- ume of the CA1 stratum radiatum. Detailed information can be tron micrographs via iTEM software (Olympus Soft Imaging found in our previous paper (Chen et al., 2010). Solutions GmbH) (Chen et al., 2010). Mitochondria were counted throughout the neuropil and The synapses were identified primarily on the basis of the specifically in the axon terminals and dendrites. The criteria presence of a postsynaptic density (PSD) with vesicles in close for identifying mitochondria were the presence of distinctive proximity to the presynaptic zone. Only spine and shaft syn- cristae and a double membrane (Figure  1). Neuropil structures apses of asymmetric synapses were analyzed in this study. In the were identified as axon terminals (presence of 3 or more synap- spine synapses, PSD located on the spines that were small spher - tic vesicles), dendrites (postsynaptic to a synapse or having an ical postsynaptic protrusions of the dendrite filled with a clear attached spine), or astroglial processes (presence of fibrils and cytoplasmic matrix and a distinct spine apparatus but without watery cytoplasm). The total number of mitochondria in neu- mitochondria and microtubules. The shaft synapses terminated ropil and the number of mitochondria in axon terminals and directly on the dendritic shaft. The dendrites were differentiated dendrites were determined. Figure 1. Count of synapses and mitochondria in serial sections. Electron micrographs of consecutive ultrathin sections (a–d) showed nonperforated synapses (black arrow), a perforated synapse (large white arrow), and shaft synapse (large black arrow). The postsynaptic spine exhibited PSD discontinuities (black stars). Mitochondria: Isolated particle (I), A branch dividing (white stars), or branches connection (B). Scale bar, 0.5 μm. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 606 | International Journal of Neuropsychopharmacology, 2018 Combining the disector principle with the object’s 0.3% H O in TBS for 7  min and washed by TBS 3 times for 2 2 3D Euler number estimates the number of mitochondria 10  min. Sections were then mounted on the gelatin-coated (Kroustrup and Gundersen, 2001). In terms of disector section slides and dehydrated with alcohol gradient and cleared with pairs, mitochondria profiles in one section plane are com- xylene. pared with mitochondria profiles on the next section plane. Images of whole immunostained sections were taken by the Mitochondria are identified in each section plane, and a Virtual Slides System VS120 (Olympus) including the software change between planes is deduced as being 1 of 2 significant VS ASW OIL 2.7 (Olympus Soft Imaging Solutions GmbH). ImageJ possibilities: a new isolated part, a so-called “Island, , or a ” I software was used to analyze the images of immunostained sec- new connection between isolated mitochondria, a “Bridge,” tions and calculated the mean optical density of BDNF-positive B, in an existing mitochondrion. The total Euler number, areas in subregions of hippocampus (Figure 2). Σx, contribution from all disectors is obtained as the signed sum of Islands, and Bridges (Figure  1). The total mitochon- Measurement of mRNA Levels with Real-time dria number was estimated as the product of the mitochon- Quantitative Polymerase Chain Reaction dria number density and volume of CA1 stratum radiatum. (Real-Time qPCR) Detailed information can be found in our previous paper The rats were decapitated and the brain collected. Briefly, (Chen et al., 2013). coronal brain sections (200  µm) were cut using a cryostat at -20ºC. Using a scalpel, the CA1 area was dissected and stored Immunohistochemistry at -80°C until extraction of RNA with the Qiagen Rneasy Free-floating 8–9 vibratome sections from each animal were Mini kit (Qiagen) as previously described (Silva Pereira et al., washed 3 times for 10  min in Tris-buffered saline (TBS) (pH 2017). The dissected CA1 tissues were homogenized in Lysis 7.4), immersed in endogenous peroxidase blocking solution buffer (Qiagen) with a mixer-mill (Retsch; twice for 40 s at 30 for 30 min, and were incubated in preheated Target Retrieval Hz/s). Total RNA was isolated following the manufacturer’s solution at 85°C for 40  min (Dako, EnVision System HRP). instructions (Qiagen). The RNA concentration, A260/280, and Tissue sections were then incubated at 4°C overnight in a A260/230 ratios were determined with a NanoDrop spec- solution containing the rabbit anti-BDNF polyclonal antibody trometer (Thermo Fisher Scientific). Afterwards, RNA was (diluted 1: 500) (AB1779, Merck Millipore). The next day, sec- reversely transcribed using random primers and Superscript tions were washed 3 times for 10  min with buffer (1% BSA IV Reverse Transcriptase (Invitrogen) following the manu- and 0.3% Triton-X in TBS) and incubated in buffer (1% BSA in facturer’s instructions. The input RNA concentration was TBS) added goat anti-rabbit IgG (1:200) for 2  h at room tem- adjusted to 29  ng/µL in the individual samples. The cDNA perature. Sections were washed 3 times for 10 min in TBS and samples were stored undiluted at -80°C until real-time qPCR then visualized with 0.1% 3, 3’-diaminobenzidine containing analysis. Figure  2. (A) BDNF expression levels were examined by immunohistochemistry in the subregions of hippocampus. (B–D) Immunohistochemistry-examined BDNF expression levels in each group. Mean optical density (MOD) was calculated with the following formula: = lo OD g10(max pixel intensity/mean pixel intensity), where max pixel intensity = 255. MOD in the vortioxetine-treated group significantly increased compared with the vehicle and fluoxetine groups in DG (A) and CA1 (C) sub- regions of hippocampus. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Chen et al. | 607 Signaling #2250; 1:500), and rabbit anti-β-actin (Licor 926–42212; Real-Time qPCR 1:3000) overnight at 4°C followed by incubation with the appro- The samples were diluted 1:10 with DEPC water before being priate IRDye conjugated secondary antibody for 1 h at RT: IRDye used as a qPCR template. The real-time qPCR reactions were 800CW goat anti-mouse IgG, IRDye 680RD goat anti-rabbit, or carried out in 96-well PCR-plates using an Mx3005P (Stratagene) IRDye 800CW goat-anti rabbit IgG, at 1:15 000 dilution (Licor). and SYBR Green. The gene expression of synapse-related genes Infrared signals were detected using the Odyssey CLx infrared (Arc, Cdc42, c-fos, Cofilin1, Homer1, Psd95, Rac1, RhoA, Spinophilin, imaging system, and bands were quantified using Image Studio Synapsin1), mitochondria-related genes (Drp1F , is1, Mfn1, Mfn2, software (LI-COR Biosciences). Band intensities were normalized Opa1), as well as 8 different reference genes (18s rRNA, ActB, to β-actin levels within the same lane. CycA, Gapdh, Hmbs, Hprt1, Rpl13A, Ywhaz) was investigated. The reference genes were selected as previously described Statistical Procedures (Bonefeld et  al., 2008). The primers were designed and tested as in our previous work (Elfving et  al., 2008). Essential gene- All data were subjected to 1-way ANOVA to compare treatment specific data about primer sequence and amplicon sizes is responses followed by posthoc tests (Tukey and least signifi- given in Supplement Table  1. The primers were obtained from cant difference) and Dunnett’s test for multiple comparisons Sigma-Aldrich. Each SYBR Green reaction (10  µL total volume) (Immunoblotting). P < .05 was considered statistically significant. contained 1x SYBR Green master mix (Sigma-Aldrich), 0.5  µM Statistical analyses and graphical representations of the find- primer pairs, and 3 µL of diluted cDNA and were carried out as ings were carried out using SPSS11 (SPSS Corp) and Sigmaplot described previously (Elfving et al., 2008 2010 , ). The mixture was 10 (SYSTAT Inc.) software. heated initially to 95°C for 3 min to activate hot-start iTaq DNA polymerase and then 40 cycles with denaturation at 95°C for Results 10  s, annealing at 60°C for 30  s, and extension at 72°C for 60  s were applied. To verify that only one PCR product was detected, BDNF Expression Levels by Immunohistochemistry the samples were subjected to a heat dissociation protocol; after the final cycle of the PCR, the reactions were heat-denatured by BDNF immunoreactivity was mainly localized in the intra-cyto- increasing the temperature from 60°C to 95°C. All samples and plasm of neurons. Mean optical density of BDNF in DG and CA1 the standard curve were run in duplicates. A standard curve was (P < .05) of hippocampus was significantly higher in rats treated generated on each plate. with vortioxetine compared with vehicle rats (Figure 2). There was Initially, the mRNA levels were determined for the 8 refer - no difference between the fluoxetine-treated rats and vehicle rats. ence genes. Stability comparison of the expression of the refer - ence genes was then conducted with the Normfinder software. The Volume of Hippocampus and Hippocampal CA1 Values of the target genes were subsequently normalized with Stratum Radiatum the geometric mean of the 2 optimal reference genes (Ywhaz and Hmbs), based on the NormFinder mathematical algorithm The volume of whole hippocampus and hippocampal CA1-SR (Andersen et al., 2004). in the vortioxetine group was significantly increased compared with the vehicle group (P < .05) (Figure 3; Table ).1 There was no difference in fluoxetine-treated rats compared with vehicle rats. Immunoblotting Aliquots (20 µg total protein) of hippocampal CA1 tissue homog- The Number of Synapses in Serial Sections enized in Cell Lysis Buffer (Bio-Rad) containing 1x protease inhibitor cocktail (Roche) were separated on 10% criterion TGX The total number of synapses (P < .001), nonperforated spine gels (Bio-Rad), transferred to nitrocellulose membranes, blocked synapses (P < .01), and perforated spine synapses (P < .01) was in odyssey blocking buffer (Licor), and probed with the primary significantly increased in the vortioxetine group compared with antibodies: mouse anti-Rac1 (Abcam ab33186; 1:500), mouse the fluoxetine and vehicle groups. No changes in the number of anti-Drp1 (Cell Signaling #14647; 1:1000), rabbit anti-c-Fos (Cell shaft synapses were observed (Figure 4; Table ). 1 Figure 3. The volume of hippocampus and CA1 stratum radium. The volume of hippocampus and CA1-SR in vortioxetine group is significantly increased compared with the vehicle and fluoxetine groups (*P < .05). Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 608 | International Journal of Neuropsychopharmacology, 2018 Figure 4. The number of synapses including subtypes of synapse in CA1 (*P < .05; ** P < .01; ***P < .001) (∆: vehicle group; ▲: vortioxetine; ▽: fluoxetine). (A) The total num- ber of synapses was significantly higher in the vortioxetine group compared with the vehicle and fluoxetine groups. (B) The number of nonperforated spine synapses was significantly higher in vortioxetine group compared with the vehicle and fluoxetine groups. (C) The number of perforated spine synapses was significantly higher in the vortioxetine group compared with the vehicle and fluoxetine groups. (D) The number of shaft synapses was not any different in the vortioxetine group compared with the vehicle and fluoxetine groups. treatment relative to vehicle. There was a tendency towards The Number of Mitochondria in Serial Sections and increase of protein expression levels of Rac1 in the vortioxetine Mitochondria Volume treatment group compared with vehicle group. No changes of The mitochondria number in total neuropil (P < .05) and axon the protein levels of Rac1 in the fluoxetine treatment group terminal (P < .05) was significantly increased in the vortioxetine were observed. There were no significant differences in protein group compared with fluoxetine and vehicle groups. There were expression levels of Drp1 in the vortioxetine and fluoxetine no significant differences in mitochondria volume and the den- treatment groups compared with vehicle group (Figure 7). drites’ mitochondria number (Figure 5; Table). 1 Correlations between Synapses, Mitochondria, Synapse- and Mitochondria-Related Gene Volume, and BDNF Level in the Hippocampus Expression and Protein Levels following Antidepressant Treatment Real-time qPCR analysis revealed that Rac1 in the vortioxetine Correlations of morphological data and BDNF values are shown treatment group was significantly increased compared with the in Figure 8. BDNF levels were positively correlated with the vol- vehicle group (P < .05) (Figure 6). There was no significant differ - ume of hippocampal CA1-SR (r = 0.80; P < .001), synapse number ence in mRNA levels of Rac1 in the fluoxetine treatment group (r = 0.78; P < .001), and mitochondrial number (r = 0.76; P < .001). compared with the vehicle group. There was a tendency towards Moreover, the volume of hippocampal CA1-SR was positively increased mRNA levels of Drp1 in the vortioxetine treatment correlated with synapse number (r = 0.78; P < .001) and mitochon- group compared with vehicle group. The mRNA expression lev- drial number (r = 0.61; P < .001). Furthermore, there was a positive els of the immediate early gene (IEG), c-fos, were significantly correlation between total mitochondria number and total num- reduced following vortioxetine (P < .01) and fluoxetine (P < .05) ber of synapses (r = 0.81; P < .001). treatment relative to vehicle; however, neither vortioxetine nor fluoxetine had any effect on Arc mRNA expression. The other Discussion selected synapse-related and mitochondria-related genes given in Supplement Table 1 were not regulated. Here, we have shown that vortioxetine significantly increased The protein expression levels of c-Fos were significantly the number of synapses and mitochondria accompanied by reduced following vortioxetine (P < .05) and fluoxetine (P < .01) BDNF level elevation, whereas fluoxetine had no effect after 7 Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Chen et al. | 609 Figure 5. The number of mitochondria in the various structures (neuropil, axons, and dendrites) and the mean volume of mitochondria in CA1 (*, < .05) (∆: v P ehicle group; ▲: vortioxetine; ▽ : fluoxetine). (A) The total number of mitochondria in neuropil displayed a significant increase in the vortioxetine group compared with the vehicle and fluoxetine groups. (B) The number of mitochondria in the axon terminal also showed a significant increase in the vortioxetine group compared with the vehicle and fluox- etine groups. (C) The number of mitochondria in dendrites was not significantly different in the vortioxetine group compared with the vehicle and fluoxetine groups. (D) The mean volume of mitochondria in CA1 stratum radiatum was not significantly different in the vortioxetine group compared with the vehicle and fluoxetine groups. Figure  6. mRNA expression investigated with real-time qPCR after 1-week treatment with fluoxetine and vortioxetine. Plotted data show normalized mean group values± SEM and are expressed as percent of control rats. Asterisks represent significant differences from control rats (*P < .05). = 8 in eac n h group. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 610 | International Journal of Neuropsychopharmacology, 2018 Figure  7. Protein expression levels of Rac1 and Drp1 in hippocampal total lysate. (A) Representative immunoblots of Rac1, Drp1, and c-Fos protein expression in response to vortioxetine and fluoxetine treatment. (B–D) Bar graphs representing the β-actin normalized density of Rac1, Drp1, and c-Fos (n = 8 rats/gr oup). Data are expressed as mean percentage± SEM of control mean values (*P < .05, **P < .01, 1-way ANOVA followed by Dunnett’s test for multiple comparisons). Figure 8. The correlations between BDNF, synaptic plasticity, and mitochondria plasticity of hippocampus (*** < .001). P The volume of hippocampal CA1-SR correlated positively with synapse number (A) and mitochondrial number (B). The levels of BDNF expression correlated positively with the volume of hippocampal CA1-SR (D), synapse number (E), and mitochondrial number (F). Furthermore, there was a strong significant positive correlation between the total mitochondria number density and total number of synapses (C). d treatment. The gene expression level of Rac1 was significantly correlation between BDNF level and mitochondria/synapse increased and Drp1 showed a tendency towards an increase in number. Based on our previous observation that vortioxetine the vortioxetine treatment group. Moreover, there was a ten- increases spine number and dendritic branching in the hippo- dency towards increase of protein expression levels of Rac1 campus CA1 faster than the SSRI fluoxetine, the present study after vortioxetine treatment by immunoblotting. However, supports the hypothesis that the new spines induced by vorti- both gene and protein expression levels of c-Fos were signifi- oxetine can rapidly form functional synapses by mitochondrial cantly decreased. Furthermore, there was a significant positive support, accompanied by increased BDNF-signaling. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Chen et al. | 611 Table 1. Summary of the Overall Measurement Results of Ultrastructure Veh vs Vor Veh vs Flu Vehicle Vortioxetine Fluoxetine P P Total volume (hippocampus) 34.870 (6.907) 42.373 (4.162) 35.553 (6.301) <.05 .86 V(CA1) 3.961 (0.748) 4.736 (0.403) 4.233 (0.393) <.05 .44 Total N(syn/CA1) x10 (9) 8.684 (1.490) 12.4511 (0.472) 9.498 (1.488) <.001 .36 N(np-syn) x10 (9) 6.296 (1.135) 8.9710 (1.055) 6.978 (0.907) <.01 .28 N(p-syn) x10 (9) 1.982 (0.361) 3.027 (0.779) 2.242 (0.907) <.05 .45 N(sh-syn) x10 (9) 0.406 (0.109) 0.513 (0.1511) 0.277 (0.153) .11 .12 Total N(mit/CA1) x10 (9) 3.438 (1.483) 5.657 (1.105) 3.802 (1.307) <.05 .66 N(a-mit)x10 (9) 2.366 (0.957) 3.895 (0.656) 2.752 (1.052) <.01 .52 N(d-mit)x10 (9) 1.065 (0.521) 1.745 (0.556) 1.048 (0.349) .05 .95 VN(mit/CA1) μm 0.080(0.018) 0.062(0.015) 0.069(0.013) .11 .29 Total V(mit/CA1) mm 0.257(0.071) 0.255(0.012) 0.252(0.043) <.05 .880 Abbreviations: N(a-mit), number of mitochondria in axons; N(d-mit), number of mitochondria in dendrites; N(mit/CA1), number of mitochon- dria in CA1; N(np-syn), number of nonperforated synapses; N(p-syn), the number of perforated synapses; N(sh-syn), the number of shaft syn- apses; N(syn/CA1), total number of synapses in CA1; V(CA1), volume of CA1 stratum radiatum; V(mit/CA1), total volume of mitochondria in CA1; V (mit/CA1), mean volume of mitochondria. documented a role for Rac-regulators in hippocampal neurogen- Effect of Vortioxetine on Hippocampal Synaptic esis and synaptogenesis and shown that dysregulation leads to Plasticity and Synapse-Related Gene Expression cognitive impairment. Thus, Rho GTPases and their downstream Several lines of evidence suggest that, in addition to neurogen- effectors may represent important therapeutic targets for dis- esis, more rapid synaptic plasticity may play an important role orders associated with cognitive dysfunction. Interestingly, the in the neurobiology of depression and effects of antidepressant role of Rac1 in modulation of synapses is not limited to hippo- therapy (Popoli et  al., 2002Duman, ; 2004). The current results campus and reported in other brain regions including nucleus showed that the number of synapses (including both nonper - accumbens (Golden et  al., 2013). A  recent study indicated the forated and perforated synapses) significantly increased in the impact of vortioxetine on synaptic integration in prefrontal-sub- vortioxetine group compared with the vehicle group. However, cortical circuits by showing diminished mPFC-msNAc afferent there were no changes in the fluoxetine group compared with drive (Chakroborty et al., 2017). the vehicle group at this time point. Previously, we have reported Emerging findings have suggested that BDNF plays a critical that 1-week vortioxetine treatment induced changes in spine role as a regulator of neuronal development and function, through number and density and also dendritic morphology, whereas an Rac1-mediated remodeling of the actin cytoskeleton (Zhou et al., equivalent dose of fluoxetine had no effects (Chen et al., 2016). 2007) and the defect in BDNF-induced spine morphogenesis is a Our recent findings implied that induction of synapses was asso- result of impaired activation of Rac1 (Lai et al., 2012). Consistent ciated with the creation of new spines at adjacent sites in the with these findings, our results indicate that the morphological dendrite. Therefore, the increased number of synapses is corre- changes of synapses to some extent may rely on Rac1-dependent lated with changes of spine number and dendritic morphology. signaling cascade accompanied by BDNF elevation. Previous experiments indicated that vortioxetine enhanced LTP The IEGs, such as c-fos, has been widely used as a molecular in the CA1 subregion of the hippocampus (Dale et al., 2014). marker tightly associated with synaptic plasticity (Minatohara Furthermore, these morphological changes were associated et al., 2015). The transcription factor IEGs c-fos is considered as with the improvement of depression-like behavior in the forced a good candidate for the initial steps of learning inducing long- swim test (FST) according to our previous behavioral results. term synaptic plasticity (Abraham et  al., 1991). This indicates Li et  al. (Li et  al., 2013) used a progesterone withdrawal model that the threshold of synaptic activation inducing the expres- in female Long-Evans rats to investigate the effects of several sion of transcription factor genes in particular brain regions classes of antidepressants (including vortioxetine and fluoxet- can be closely linked to synaptic plasticity. In the present study, ine) in the FST as a measure of antidepressant activity. Acute c-fos was downregulated in response to vortioxetine treatment. (3 injections over 2  days) treatment results showed that vorti- Consistent with these results, 4 weeks chronic vortioxetine can oxetine significantly reduced immobility in the FST, whereas promote a decrease in c-fos expression in the hippocampus fluoxetine was ineffective. Therefore, our findings support that (Waller et  al., 2016). Although several lines of evidence reveal synaptic plasticity may play an important role in the effects of that vortioxetine can promote expression of various genes that antidepressant therapy. play a role in synaptic plasticity (Li et al., 2015 du J ; ardin et al., Evidence suggests that Rac1, a well-known Rho GTPase, con- 2016; Waller et al., 2017), these studies measured mRNA expres- tributes to the regulation of dendritic spine formation, excitatory sion and were not focused on a particular layer of hippocampus, synapses, and synaptic function (Penzes et al., 2003Cala ; brese as well as by species differences in the response to vortioxetine. et al., 2006; Oh et al., 2010). The longer dendritic spines in hip- pocampal CA1 pyramidal neurons of Srgap 3-/- mice were found Effect of Vortioxetine on Hippocampal Mitochondrial through increasing Rac1 activity by negatively regulating Srgap3 Plasticity and Mitochondria-Related Gene Expression (Waltereit et al., 2012). The definitive evidence showed compel- ling links between Rho GTPases signaling and cognitive function Synaptic transmission requires mitochondrial ATP generation (De et al., 2014). Zamboni et al. (Zamboni et al., 2016) have also for neurotransmitter exocytosis, vesicle recruitment, activation Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 612 | International Journal of Neuropsychopharmacology, 2018 of ion conductance, signaling at metabotropic receptors, potenti- Interestingly, several studies have demonstrated that fluox- ation of neurotransmitter release, and synaptic plasticity (Li etine induces mitochondria triggering apoptosis and/or inter - et  al., 2010; Jiao and Li, 2011). Mitochondrial dynamics, a cor - feres with mitochondrial function by modulating the activity porate term for the processes of mitochondrial fission, fusion, of respiratory chain components and enzymes of the Krebs and transport, controls mitochondrial function and localization cycle. Furthermore, fluoxetine alters mitochondria-related within the cell (Detmer and Chan, 2007). Mitochondria con- redox aspects in different experimental models (de Oliveira, stantly move along axons and dendrites, dividing and fusing 2016). However, there was no significant effect of fluoxetine in response to synaptic changes (Mattson et  al., 2008 P; almer on hippocampal mitochondrial plasticity and mitochondria- et  al., 2011), and help to regulate synaptic activity and conse- related genes expression in this study. This discrepancy may quently learning and memory (Li et al., 2010 Jiao and Li, ; 2011). be explained by the fact that the dosage and time of fluoxetine Mitochondria, therefore, not only provide dynamic energy sup- administration in the present study was different compared port for normal synaptic functioning but also directly modulate with the previous reports (Adzic et al., 2013). synaptic structural and functional plasticity (Cheng et al., 2010). Our results suggest that the increased mitochondrial number 5-HT Receptor Activities of Vortioxetine provided ATP to support features of synaptogenesis. Therefore, mitochondrial biogenesis may play an important physiological Recent studies demonstrated that vortioxetine prevented the 5-HT-induced increase in inhibitory post-synaptic potentials role in synaptic plasticity in the hippocampus. Mitochondria are dynamically transported in and out of recorded from CA1 pyramidal cells, most likely by 5-HT recep- tor antagonism (Dale et  al., 2014). Furthermore, 5-HT receptor axons and dendrites to maintain neuronal and synaptic func- tion. It is shown that the movement of mitochondria in axons antagonists exhibited memory-enhancing properties in a num- ber of preclinical cognition models (Arnsten et al., 1997 Pitsikas ; and dendrites are different due to the movement of mitochon- dria in axons with a consistently rapid velocity than are those and Borsini, 1996). A  selective 5-HT R agonist (flesinoxan) and 1A a selective 5-HT receptor antagonist (ondansetron) reversed in dendrites (Ligon and Steward, 2000). Therefore, the difference in motility and metabolic properties of mitochondria in axons memory impairments in 5-HT depleted rats (Jensen et al., 2014). Thus, accumulating evidence suggests that both 5-HT and and dendrites reflects alterations in energy metabolism during 1A synaptic plasticity. Since most metabolic activity takes place in 5-HT receptors play an important role in the effect of vortiox- etine on cognitive function. axon terminals (Zinsmaier et al., 2009), an increased number of mitochondria in axon terminals after vortioxetine treatment in A selective 5-HT receptor antagonist, Tropisetron, protected PC12 cells against high glucose-induced apoptosis and oxida- our study implies that the increased metabolism may support high synaptic activity by the generation of action potentials and tive stress by preventing mitochondrial pathways (Aminzadeh, 2017). Oxidative stress results from an imbalance between the trafficking of synaptic vesicles (neurotransmitter exocytosis and vesicle recruitment). production of ROS and antioxidant capacity (Uttara et al., 2009). ROS could increase the gating potential of mitochondrial pores Regulation of mitochondrial dynamics is partly accomplished through posttranslational modification of mitochondrial fission and contribute to cytochrome c release (Sharifi et al., 2007). The and fusion enzymes, in turn influencing mitochondrial bioen- 5-HT receptor agonist promoted axonal transport of mitochon- 1A ergetics and transport (Cheng et  al., 2010). The importance of dria in hippocampus neurons by activating the Akt- GSK3β path- posttranslational regulation is highlighted in numerous neuro- way (Chen et  al., 2007). GSK3β is a common therapeutic target degenerative disorders associated with posttranslational modi- of many antidepressant drugs (Jope and Roh, 2006). Thus, it is fication of the mitochondrial fission enzyme Drp1 (Li et al., 2004). suggested that aberrant 5-HT-induced mitochondrial movement One study showed that mitochondrial abnormality including may be linked to depression, and effects of 5-HT on mitochon- morphological impairment is due to the imbalance between dria functionality may be mediated by both 5-HT and 5-HT 1A 3 mitochondrial fusion and fission via a glycogen synthase kinase receptors. Furthermore, we suggest that vortioxetine’s 5-HT 1A 3β (GSK3β)/dynamin-related protein-1 (Drp1)-dependent mech- receptor agonism and 5-HT receptor antagonism may con- anism (Huang et al., 2015). tribute to cognitive function through mitochondrial biogenesis Furthermore, increasing mitochondrial fragmentation by based on our present findings. overexpression of Drp1 enhances synapse formation, whereas dominant-negative inhibition of Drp1 has the opposite effect The Relationship between BDNF, Synaptic Plasticity, in cultured hippocampal neurons, indicating a prominent and Mitochondria Plasticity of Hippocampus in role for mitochondrial dynamics in synaptogenesis (Dickey Depression and Strack, 2011). Our finding showing a tendency towards increased levels of Drp1 gene is consistent with their findings. Emerging findings suggest roles for mitochondria as media- But, there was no significant change of protein levels of Drp1. tors of at least some of the effects of glutamate and BDNF on However, it was not certain whether a vortioxetine-induced synaptic plasticity (Shuttleworth et  al., 2003 Lu et  ; al., 2009). increase in Drp1 gene expression was associated with a simi- Several neurotrophic factors including BDNF have been shown lar enhancement on protein level of Drp1 and the functional to promote neuronal differentiation, survival, and modifying activity of the protein. Moreover, correlation analysis in large- synaptic plasticity (Duman and Monteggia, 2006 K; ermani and scale data sets reported approximately 50% correspondence Hempstead, 2007). BDNF promotes synaptic transmission and between messenger RNA and protein levels (Tian et al., 2004). plasticity, in part by increasing mitochondrial energy produc- Thus, the gene expression of Drp1 in the present study may tion (Markham et  al., 2014). Consequences of abnormalities in possibly offer a mechanistic explanation for increased mito- the BDNF and glucocorticoid receptor signaling are impairment chondria number induced by vortioxetine. Furthermore, a link of mitochondrial respiration efficiency and synaptic plastic- between Drp1 and mitochondria plasticity may indicate that ity (Jeanneteau and Arango-Lievano, 2016). A  study of BNDF- there is an increase of mitochondrial fission after vortioxetine induced mitochondrial motility arrest and presynaptic docking treatment. suggests that mitochondrial transport and distribution play Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Chen et al. | 613 an essential role in BDNF-mediated synaptic transmission (Su Stochastic Geometry and Advanced Bioimaging. Dr. Wegener et al., 2014). The neuroprotective effect of BDNF shares molecu- reported having received lecture/consultancy fees from lar signaling pathways (MEK-Bcl-2 pathway) with mitochondrial H.  Lundbeck A/S, Servier SA, Astra Zeneca AB, Eli Lilly A/S, respiratory coupling (Markham et al., 2012). Sun Pharma Pty Ltd, Pfizer Inc, Shire A/S, HB Pharma A/S, Arla More importantly, the ability of mitochondria in regulation Foods A.m.b.A., Alkermes Inc, and Mundipharma International 2+ of Ca clearance in connection with BDNF signaling pathways Ltd., and research funding from the Danish Medical Research is a key point of modulating the synaptic plasticity, especially Council, Aarhus University Research Foundation (AU-IDEAS ini- synaptogenesis (Markham et  al., 2014). Consistent with these tiative (eMOOD)), the Novo Nordisk Foundation, the Lundbeck findings, our results also indicated that BDNF has a significant Foundation, and EU Horizon 2020 (ExEDE). Drs Ardalan, Danladi, positive correlation with synapse and mitochondrial number. Elfving and Müller reported no biomedical financial interests or The mitochondria are essential components in synaptic trans- potential conflicts of interest. mission, since they are a major source of energy required for main- tenance and restoration of ion gradients (Todorova and Blokland, References 2017). High levels of monocarboxylate (Gerhart et  al., 1998) and glucose (Gerhart et al., 1991) transporters have been observed in Abraham WC, Dragunow M, Tate WP (1991) The role of immedi- CA1, suggesting elevated metabolic and synaptic activity. Several ate early genes in the stabilization of long-term potentiation. studies suggest that physical proximity between mitochondria Mol Neurobiol 5:297–314. and synapses is regulated by neuronal activity (Courchet et  al., Adzic M, Lukic I, Mitic M, Djordjevic J, Elaković I, Djordjevic A, 2013; Sheng, 2014). Our results showed that increased total syn- Krstic-Demonacos M, Matić G, Radojcic M (2013) Brain region- apse numbers are strongly positively correlated with total mito- and sex-specific modulation of mitochondrial glucocorticoid chondria number. The mitochondrial changes may be crucially receptor phosphorylation in fluoxetine treated stressed rats: linked to changed energy metabolism and, therefore, may have effects on energy metabolism. Psychoneuroendocrinology consequences for cell plasticity, resilience, and survival in patients 38:2914–2924. with MDD. Antidepressants might ultimately enhance energy Alcocer-Gómez E, de Miguel M, Casas-Barquero N, Núñez-Vasco metabolism and reduce the damage of oxidative stress. However, J, Sánchez-Alcazar JA, Fernández-Rodríguez A, Cordero MD because our experimental model is in vivo, it is difficult to show (2014) NLRP3 inflammasome is activated in mononuclear directly the role of mitochondria in synaptic plasticity. Further in blood cells from patients with major depressive disorder. vitro studies would be necessary to elucidate how mitochondria Brain Behav Immun 36:111–117. modulate synaptic plasticity, e.g., changes of dendritic mitochon- Aminzadeh A (2017) Protective effect of tropisetron on high glu- dria content or mitochondrial activity. cose induced apoptosis and oxidative stress in PC12 cells: In conclusion, our study suggests that mitochondria play a roles of JNK, P38 mapks, and mitochondria pathway. Metab critical role in synaptic plasticity accompanied by increasing Brain Dis 32:819–826. BDNF levels. In particular, the coincidence of rapid changes in the Andersen CL, Jensen JL, Ørntoft TF (2004) Normalization of real- BDNF levels and synaptic/mitochondria plasticity of hippocam- time quantitative reverse transcription-PCR data: a model- pus following vortioxetine compared with fluoxetine may be based variance estimation approach to identify genes suited ascribed to vortioxetine’s modulation of one or more serotonin for normalization, applied to bladder and colon cancer data receptors. A newly published patent (US 9820984 B1) documents sets. Cancer Res 64:5245–5250. that a treatment regimen of a single i.v. dose of vortioxetine fol- Anderson G (2018) Linking the biological underpinnings of lowed by oral vortioxetine achieves a faster onset of therapeutic depression: role of mitochondria interactions with melatonin, action in MDD patients (Sanchez et al., 2017). Further studies are inflammation, sirtuins, tryptophan catabolites, DNA repair necessary to clarify how BDNF levels trigger changes in mito- and oxidative and nitrosative stress, with consequences for chondria and synapses. These findings might provide insights in classification and cognition. Prog Neuropsychopharmacol to how to target specific 5-HT receptors to induce mitochondrial Biol Psychiatry 80:255–266. biogenesis and may provide new avenues for development of Arnsten AF, Lin CH, Van Dyck CH, Stanhope KJ (1997) The effects novel therapeutics targeting the biogenesis response. of 5-HT3 receptor antagonists on cognitive performance in aged monkeys. Neurobiol Aging 18:21–28. Bonefeld BE, Elfving B, Wegener G (2008) Reference genes for nor - Supplementary Material malization: a study of rat brain tissue. Synapse 62:302–309. Supplementary data are available at International Journal of Calabrese B, Wilson MS, Halpain S (2006) Development and regu- lation of dendritic spine synapses. Physiology (Bethesda) Neuropsychopharmacology online. 21:38–47. Chakroborty S, Geisbush TR, Dale E, Pehrson AL, Sánchez C, West Acknowledgments AR (2017) Impact of vortioxetine on synaptic integration in prefrontal-subcortical circuits: comparisons with escitalo- Herdis Krunderup and Lone Lysgaard are gratefully acknowl- pram. Front Pharmacol 8:764. edged for their skillful EM technical assistance. Chang DT, Honick AS, Reynolds IJ (2006) Mitochondrial traf- ficking to synapses in cultured primary cortical neurons. J Statement of Interest Neurosci 26:7035–7045. Dr. Chen reports having received salary support from Chen F, Madsen TM, Wegener G, Nyengaard JR (2010) Imipramine H.  LundbeckA/S. Dr. Ardalan reported having salary support treatment increases the number of hippocampal synapses from Lundbeck Foundation. Connie Sanchez was full-time and neurons in a genetic animal model of depression. employee at H.  Lundbeck A/S when the study was conducted. Hippocampus 20:1376–1384. Dr. Nyengaard reports having received research funding from Chen F, Wegener G, Madsen TM, Nyengaard JR (2013) Sino-Danish Center and the Villum Foundation via Centre for Mitochondrial plasticity of the hippocampus in a genetic rat Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 614 | International Journal of Neuropsychopharmacology, 2018 model of depression after antidepressant treatment. Synapse AJ, Gonzalez-Maeso J, Neve RL, Turecki G, Ghose S, Tamminga 67:127–134. CA, Russo SJ (2013) Epigenetic regulation of RAC1 induces Chen F, du Jardin KG, Waller JA, Sanchez C, Nyengaard JR, synaptic remodeling in stress disorders and depression. Nat Wegener G (2016) Vortioxetine promotes early changes in Med 19:337–344. dendritic morphology compared with fluoxetine in rat hippo- Huang S, Wang Y, Gan X, Fang D, Zhong C, Wu L, Hu G, Sosunov campus. Eur Neuropsychopharmacol 26:234–245. AA, McKhann GM, Yu H, Yan SS (2015) Drp1-mediated mito- Chen S, Owens GC, Crossin KL, Edelman DB (2007) Serotonin chondrial abnormalities link to synaptic injury in diabetes stimulates mitochondrial transport in hippocampal neurons. model. Diabetes 64:1728–1742. Mol Cell Neurosci 36:472–483. Jeanneteau F, Arango-Lievano M (2016) Linking mitochondria to Cheng A, Hou Y, Mattson MP (2010) Mitochondria and neuroplas- synapses: new insights for stress-related neuropsychiatric ticity. ASN Neuro 2:e00045. disorders. Neural Plast 2016:3985063. Courchet J, Lewis TL Jr, Lee S, Courchet V, Liou DY, Aizawa S, Jensen JB, du Jardin KG, Song D, Budac D, Smagin G, Sanchez Polleux F (2013) Terminal axon branching is regulated by the C, Pehrson AL (2014) Vortioxetine, but not escitalopram or LKB1-NUAK1 kinase pathway via presynaptic mitochondrial duloxetine, reverses memory impairment induced by cen- capture. Cell 153:1510–1525. tral 5-HT depletion in rats: evidence for direct 5-HT receptor Czarny P, Wigner P, Galecki P, Sliwinski T (2018) The interplay modulation. Eur Neuropsychopharmacol 24:148–159. between inflammation, oxidative stress, DNA damage, DNA Jiao S, Li Z (2011) Nonapoptotic function of BAD and BAX in repair and mitochondrial dysfunction in depression. Prog long-term depression of synaptic transmission. Neuron Neuropsychopharmacol Biol Psychiatry 80:309–321. 70:758–772. Dale E, Zhang H, Leiser SC, Xiao Y, Lu D, Yang CR, Plath N, Jope RS, Roh MS (2006) Glycogen synthase kinase-3 (GSK3) in Sanchez C (2014) Vortioxetine disinhibits pyramidal cell func- psychiatric diseases and therapeutic interventions. Curr tion and enhances synaptic plasticity in the rat hippocam- Drug Targets 7:1421–1434. pus. J Psychopharmacol 28:891–902. Kermani P, Hempstead B (2007) Brain-derived neurotrophic De Filippis B, Romano E, Laviola G (2014) Aberrant rho gtpases factor: a newly described mediator of angiogenesis. Trends signaling and cognitive dysfunction: in vivo evidence for a Cardiovasc Med 17:140–143. compelling molecular relationship. Neurosci Biobehav Rev Klinedinst NJ, Regenold WT (2015) A mitochondrial bioenergetic 46:285–301. basis of depression. J Bioenerg Biomembr 47:155–171. de Oliveira MR (2016) Fluoxetine and the mitochondria: a review Kroustrup JP, Gundersen HJ (2001) Estimating the number of of the toxicological aspects. Toxicol Lett 258:185–191. complex particles using the conneulor principle. J Microsc Detmer SA, Chan DC (2007) Functions and dysfunctions of mito- 203:314–320. chondrial dynamics. Nat Rev Mol Cell Biol 8:870–879. Lai KO, Wong AS, Cheung MC, Xu P, Liang Z, Lok KC, Xie H, Palko Dickey AS, Strack S (2011) PKA/AKAP1 and PP2A/bβ2 regulate ME, Yung WH, Tessarollo L, Cheung ZH, Ip NY (2012) Trkb neuronal morphogenesis via drp1 phosphorylation and phosphorylation by cdk5 is required for activity-depend- mitochondrial bioenergetics. J Neurosci 31:15716–15726. ent structural plasticity and spatial memory. Nat Neurosci du Jardin KG, Müller HK, Sanchez C, Wegener G, Elfving B (2016) 15:1506–1515. A single dose of vortioxetine, but not ketamine or fluoxetine, Li Y, Raaby KF, Sánchez C, Gulinello M (2013) Serotonergic recep- increases plasticity-related gene expression in the rat frontal tor mechanisms underlying antidepressant-like action in cortex. Eur J Pharmacol 786:29–35. the progesterone withdrawal model of hormonally induced Duman RS (2004) Neural plasticity: consequences of stress and depression in rats. Behav Brain Res 256:520–528. actions of antidepressant treatment. Dialogues Clin Neurosci Li Y, Abdourahman A, Tamm JA, Pehrson AL, Sánchez C, Gulinello 6:157–169. M (2015) Reversal of age-associated cognitive deficits is Duman RS, Monteggia LM (2006) A neurotrophic model for accompanied by increased plasticity-related gene expression stress-related mood disorders. Biol Psychiatry 59:1116–1127. after chronic antidepressant administration in middle-aged Elfving B, Bonefeld BE, Rosenberg R, Wegener G (2008) Differential mice. Pharmacol Biochem Behav 135:70–82. expression of synaptic vesicle proteins after repeated elec- Li Y, Sanchez C, Gulinello M (2017) Distinct antidepressant- troconvulsive seizures in rat frontal cortex and hippocam- like and cognitive effects of antidepressants with different pus. Synapse 62:662–670. mechanisms of action in middle-aged female mice. Int J Elfving B, Plougmann PH, Wegener G (2010) Differential brain, Neuropsychopharmacol 20:510–515. but not serum VEGF levels in a genetic rat model of depres- Li Z, Okamoto K, Hayashi Y, Sheng M (2004) The importance of sion. Neurosci Lett 474:13–16. dendritic mitochondria in the morphogenesis and plasticity Frampton JE (2016) Vortioxetine: a review in cognitive dysfunc- of spines and synapses. Cell 119:873–887. tion in depression. Drugs 76:1675–1682. Li Z, Jo J, Jia JM, Lo SC, Whitcomb DJ, Jiao S, Cho K, Sheng M (2010) Geinisman Y, Berry RW, Disterhoft JF, Power JM, Van der Zee EA Caspase-3 activation via mitochondria is required for long- (2001) Associative learning elicits the formation of multiple- term depression and AMPA receptor internalization. Cell synapse boutons. J Neurosci 21:5568–5573. 141:859–871. Gerhart DZ, Djuricic B, Drewes LR (1991) Quantitative immuno- Ligon LA, Steward O (2000) Movement of mitochondria in the cytochemistry (image analysis) of glucose transporters in the axons and dendrites of cultured hippocampal neurons. J normal and postischemic rodent hippocampus. J Cereb Blood Comp Neurol 427:340–350. Flow Metab 11:440–448. Lu CW, Lin TY, Chiang HS, Wang SJ (2009) Facilitation of glutam- Gerhart DZ, Enerson BE, Zhdankina OY, Leino RL, Drewes LR ate release from rat cerebral cortex nerve terminal by suban- (1998) Expression of the monocarboxylate transporter MCT2 esthetic concentration propofol. Synapse 63:773–781. by rat brain glia. Glia 22:272–281. Markham A, Cameron I, Bains R, Franklin P, Kiss JP, Schwendimann Golden SA, Christoffel DJ, Heshmati M, Hodes GE, Magida J, Davis L, Gressens P, Spedding M (2012) Brain-derived neurotrophic K, Cahill ME, Dias C, Ribeiro E, Ables JL, Kennedy PJ, Robison factor-mediated effects on mitochondrial respiratory Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018 Chen et al. | 615 coupling and neuroprotection share the same molecular sig- expression in the prefrontal cortex of adult male rats. Eur J nalling pathways. Eur J Neurosci 35:366–374. Pharmacol 815:304–311. Markham A, Bains R, Franklin P, Spedding M (2014) Changes in Small JV (1968) Measurement of section thickness. [Fourth mitochondrial function are pivotal in neurodegenerative and European Conference on Electron Microscopy]:609–610. psychiatric disorders: how important is BDNF? Br J Pharmacol Smith J, Browning M, Conen S, Smallman R, Buchbjerg J, Larsen 171:2206–2229. KG, Olsen CK, Christensen SR, Dawson GR, Deakin JF, Hawkins Mattson MP, Gleichmann M, Cheng A (2008) Mitochondria in neu- P, Morris R, Goodwin G, Harmer CJ (2017) Vortioxetine reduces roplasticity and neurological disorders. Neuron 60:748–766. BOLD signal during performance of the N-back working Minatohara K, Akiyoshi M, Okuno H (2015) Role of immediate- memory task: a randomised neuroimaging trial in remitted early genes in synaptic plasticity and neuronal ensembles depressed patients and healthy controls. Mol Psychiatry. underlying the memory trace. Front Mol Neurosci 8:78. Su B, Ji YS, Sun XL, Liu XH, Chen ZY (2014) Brain-derived neuro- Oh D, et  al. (2010) Regulation of synaptic rac1 activity, long-term trophic factor (BDNF)-induced mitochondrial motility arrest potentiation maintenance, and learning and memory by BCR and and presynaptic docking contribute to BDNF-enhanced syn- ABR rac gtpase-activating proteins. J Neurosci 30:14134–14144. aptic transmission. J Biol Chem 289:1213–1226. Palmer CS, Osellame LD, Stojanovski D, Ryan MT (2011) The regu- Tian Q, Stepaniants SB, Mao M, Weng L, Feetham MC, Doyle MJ, lation of mitochondrial morphology: intricate mechanisms Yi EC, Dai H, Thorsson V, Eng J, Goodlett D, Berger JP, Gunter B, and dynamic machinery. Cell Signal 23:1534–1545. Linseley PS, Stoughton RB, Aebersold R, Collins SJ, Hanlon WA, Pan Z, Grovu RC, Cha DS, Carmona NE, Subramaniapillai Hood LE (2004) Integrated genomic and proteomic analyses M, Shekotikhina M, Rong C, Lee Y, McIntyre RS (2017) of gene expression in mammalian cells. Mol Cell Proteomics Pharmacological treatment of cognitive symptoms in 3:960–969. major depressive disorder. CNS Neurol Disord Drug Targets Todorova V, Blokland A (2017) Mitochondria and synaptic 16:891–899. plasticity in the mature and aging nervous system. Curr Pehrson AL, Hillhouse TM, Haddjeri N, Rovera R, Porter JH, Mørk Neuropharmacol 15:166–173. A, Smagin G, Song D, Budac D, Cajina M, Sanchez C (2016) Uttara B, Singh AV, Zamboni P, Mahajan RT (2009) Oxidative Task- and treatment length-dependent effects of vortioxe- stress and neurodegenerative diseases: a review of upstream tine on scopolamine-induced cognitive dysfunction and hip- and downstream antioxidant therapeutic options. Curr pocampal extracellular acetylcholine in rats. J Pharmacol Exp Neuropharmacol 7:65–74. Ther 358:472–482. Vavakova M, Durackova Z, Trebaticka J (2015) Markers of oxida- Penzes P, Beeser A, Chernoff J, Schiller MR, Eipper BA, Mains RE, tive stress and neuroprogression in depression disorder. Oxid Huganir RL (2003) Rapid induction of dendritic spine morpho- Med Cell Longev 2015:898393. genesis by trans-synaptic ephrinb-ephb receptor activation Waller JA, Chen F, Sánchez C (2016) Vortioxetine promotes mat- of the rho-GEF kalirin. Neuron 37:263–274. uration of dendritic spines in vitro: a comparative study in Pitsikas N, Borsini F (1996) Itasetron (DAU 6215)  prevents age- hippocampal cultures. Neuropharmacology 103:143–154. related memory deficits in the rat in a multiple choice avoid- Waller JA, Tamm JA, Abdourahman A, Pehrson AL, Li Y, Cajina M, ance task. Eur J Pharmacol 311:115–119. Sánchez C (2017) Chronic vortioxetine treatment in rodents Popoli M, Gennarelli M, Racagni G (2002) Modulation of synap- modulates gene expression of neurodevelopmental and plas- tic plasticity by stress and antidepressants. Bipolar Disord ticity markers. Eur Neuropsychopharmacol 27:192–203. 4:166–182. Waltereit R, et al (2012) Srgap3⁻/⁻ mice present a neurodevelop- Sanchez C, Asin KE, Artigas F (2015) Vortioxetine, a novel anti- mental disorder with schizophrenia-related intermediate depressant with multimodal activity: review of preclinical phenotypes. Faseb J 26:4418–4428. and clinical data. Pharmacol Ther 145:43–57. Zamboni V, Armentano M, Sarò G, Ciraolo E, Ghigo A, Germena Sanchez C, Søby KK, Bang-Andersen B (2017) Dosing regimens G, Umbach A, Valnegri P, Passafaro M, Carabelli V, Gavello for fast onset of antidepressant effect. United States Patent US D, Bianchi V, D’Adamo P, de Curtis I, El-Assawi N, Mauro A, 9820984 B1 Nov. 21, 2017. Priano L, Ferri N, Hirsch E, Merlo GR (2016) Disruption of arh- Sharifi K, Mostaghni K, Maleki M, Badiei K (2007) Ischaemia/rep- gap15 results in hyperactive rac1, affects the architecture and erfusion injury in experimentally induced abomasal volvulus function of hippocampal inhibitory neurons and causes cog- in sheep. Vet Res Commun 31:575–590. nitive deficits. Sci Rep 6:34877. Sheng ZH (2014) Mitochondrial trafficking and anchoring in neu- Zhou P, Porcionatto M, Pilapil M, Chen Y, Choi Y, Tolias KF, Bikoff rons: new insight and implications. J Cell Biol 204:1087–1098. JB, Hong EJ, Greenberg ME, Segal RA (2007) Polarized signaling Shuttleworth CW, Brennan AM, Connor JA (2003) NAD(P)H fluor - endosomes coordinate BDNF-induced chemotaxis of cerebel- escence imaging of postsynaptic neuronal activation in m-ur lar precursors. Neuron 55:53–68. ine hippocampal slices. J Neurosci 23:3196–3208. Zinsmaier KE, Babic M, Russo GJ (2009) Mitochondrial transport Silva Pereira V, Elfving B, Joca SRL, Wegener G (2017) Ketamine dynamics in axons and dendrites. Results Probl Cell Differ and aminoguanidine differentially affect Bdnf and Mtor gene 48:107–139. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/6/603/4921128 by Ed 'DeepDyve' Gillespie user on 21 June 2018

Journal

International Journal of NeuropsychopharmacologyOxford University Press

Published: Mar 5, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off