The effects of ketamine on BDNF-TrkB neurotransmission in animal models of depression When the brain loses TrkBactivation

(1)

When the brain loses TrkBactivation

The effects of ketamine on BDNF-TrkB

neurotransmission in animal models of depression

Author: Hanna Sädbom-Williams Supervisor: Sven Tågerud Examiner: Ravi Vumma Term: VT21

Subject: Examination Project Work in

Health Science

(2)

Abstract

Ketamine is a non-competitive N-methyl-D-aspartate (NMDA)-channel blocker that has recently shown promise in the treatment of major depressive disorder,

distinguishing itself from classical anti-depressants through its rapid and lasting effects when given at sub-anaesthetic doses. Animal models of depression are commonly used to research individual mechanisms of action and this literature review aimed to investigate how ketamine influences BDNF-TrkB

neurotransmission in the hippocampus and prefrontal cortex within animal models of depression. Reduced levels of BDNF and TrkB-transmission, as well as

downstream signalling, are common in both humans experiencing depression and in rodent models of depression, and ketamine was found to counteract this reduction in the majority of studies reviewed. In the majority of studies ketamine’s

anti-depressant actions were viewed to be at least partially connected to its effects on BDNF-TrkB neurotransmission. This was supported by the anti-depressant effects being readily blocked by pharmachological inhibition of TrkB. Inhibition also blocked the downstream neurobiological changes associated with ketamines anti-depressant effects.

Sammanfattning

Ketamin är en icke-kompetitiv N-methyl-D-aspartate (NMDA)-kanal antagonist som nyligen har visat lovande resultat i behandling av depression. Substansen särskiljer sig från klassiska antidepressiva läkemedel genom att dess effekt infinner sig snabbt och kvarstår under en längre period om det ges i låga doser. Djurmodeller av depression används för att undersöka individuella mekanismer relaterade till depression och denna litteraturstudie ämnade att undersöka hur ketamin påverkar BDNF-TrkB signallering inom hippocampus och prefrontala cortex i djurmodeller av depression. Minskade nivåer av BDNF och TrkB-signalering är vanligt

förekommande både hos männsikor med depression och i djurmodeller av depression. I majoriteten av studierna återställde ketamin nivåerna av BDNF och TrkB-signalering till normala värden. Dess antidepressiva effekt kopplades till denna signalväg eftersom farmakologisk inhibering av TrkB i majoriteten av studierna resulterade i att den anti-depressiva effekten uteblev. Inhiberingen blockerade även nedströms neurobiologiska förändringar som anses kopplade till ketamins antidepressiva effekter.

Key words

(3)

Acknowledgments

(4)

Appendices

(5)

1 Abbreviations

AMPA Alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid

AP2M Adaptor protein complex 2 medium light chain

BDNF Brain derived neurotrophic factor

CA1 & CA3 Cornu Ammonis (regions of the hippocampus)

CeA Central nucleus of the amygdala

CREB Cyclic adenosine monophosphate-response element binding protein CUMS/UCMS Unpredictable chronic mild stress model

DG Dentate gyrus (region of hippocampus)

DRN Dorsal raphe nucleus (nucleus in brainstem)

ECT Electro convulsive therapy

eEF2 Eukaryotic elongation factor 2

EEG Electroencephalography

EPM Elevated plus maze

ERK Extracellular signal-regulated kinases

FST Forced swim test

GABA Gamma-aminobutyric acid

GLT1 Glutamate transporter 1

GluA1 Subunit of AMPA receptor

GSK3β Glycogen synthase kinase 3 beta

HNK Hydroxynorketamine (active metabolite of ketamine)

KO Knock out (Gene knock out)

LC Locus coeruleus (nucleus in the brainstem)

MDD Major depressive disorder

mPFC Medial prefrontal cortex

mTor Mammalian target for rapamycin

NAc Nucleus Accumbens (region of the striatum)

NMDA (R) N-methyl-D-aspartate (Receptor)

NORT Novel object recognition test

NPC Neural progenitor cell

NRF2 KO Nuclear factor erythroid derived 2-related factor knock-out

OFT Open field test

p- (prefix) Phosphorylated

P70-S6K Ribosomal protein S6 kinase beta-1

PFC Prefrontal cortex

POD Post-operative depression (stress model of depression) PSD-95 Post synaptic density protein

SPT Sucrose preference test

TrkB Tropomyosin receptor kinase B

TST Tail suspension test

WB Western blot

2 Introduction

2.1 Depression today

(6)

Depression can consist of a variety of symptoms, ranging from chronic feelings of hopelessness or sadness to anhedonia, suicidal ideation and issues related to eating or sleeping (2). It is estimated that approximately 30% of patients with depression develop what is referred to as treatment resistant depression, (TRD), in which after two or more attempts of treatment with anti-depressant medications, the patient does not show an adequate response. With the potentially lethal outcome of major

depressive disorder, the lack of response in a large percentage of patients to current treatments is a concern, and research into novel treatments is underway (3).

Most of what is known about depression today is based on the actions of medications with anti-depressant effects, such as selective serotonin reuptake inhibitors, (SSRI), and other modulators of mono amine neurotransmission. The actions of these medications led to the formation of the so-called mono amine theory of depression, which postulates that, depression occurs because of abnormal mono amine levels in the brain (2,4).

The two main neurotransmitters that are implicated under the mono amine theory of depression are serotonin (5-HT) and noradrenalin. The direct neurochemical effects of anti-depressant medications on these two neurotransmitters occur within hours of administration. There is however a significant lag time between these effects and the actual anti-depressant effects, which is of concern in severely depressed patients. It may indicate that the anti-depressant effects are due to secondary mechanisms of adaptation that occur in the brain in response to the immediate effects (2–4). Studies of the brains of patients with depressive disorders show that reduced neurotransmission in the hippocampus and the prefrontal cortex, as well as a reduction in the size of these areas with a loss of neurons and dendrites is common. This means that ventricles are often enlarged in patients with depressive disorders. Several factors appear to be implicated in these changes, such as the levels of brain derived neurotrophic factor (BDNF) and abnormal glutamate transmission via the NMDA (N-methyl-d-aspartate) receptor (4).

2.2 Ketamine as a novel treatment

Ketamine is a non-competitive N-methyl-D-aspartate (NMDA)-channel blocker that was originally developed as an anaesthetic drug. It has dissociative side-effects and addictive properties. Antidepressant effects of the drug at sub-anaesthetic doses have been identified, these are rapid (occur within minutes) and are sustained for up to two weeks, even after a single dose. Anti-depressant ketamine treatment has a high success rate with patients diagnosed with treatment resistant depression. The anti-depressant effects of ketamine appear to be more complicated than simply being connected to the substance's main mechanism of action on the NMDA channel protein; suggesting that again, secondary mechanisms may be of interest (4).

2.3 Molecular pathways implicated in depression

2.3.1 Mono amine theory

Insufficient synaptic levels of 5-HT are still heavily implicated in depression, but do not tell the full story of the pathogenesis, and systems other than the mono amine system are being found to be relevant for the disorder. Disruption of homeostasis, in particularly through chronic stress, has been suggested to cause cascade-like

(7)

Chronic stress pathology involves a loss of glial cells and a reduced glutamate uptake capacity, leading to excitotoxicity, neuronal atrophy, loss of dendritic spine density and reduced glutamate transmission (6).

2.3.2 Brain derived neurotrophic factor

One of the major factors implicated in depression is the brain derived neurotrophic factor, (BDNF) (7). BDNF neurotransmission occurs through binding to a

transmembrane receptor known as TrkB (tropomyosin receptor kinase B), initiating the phosphorylation of three pathways starting respectively with: PLCγ

(phosphoinositide phospholipase Cγ), PI3K (phosphoinostide 3-kinase) and ERK (extracellular signal-regulated kinases). These pathways ultimately lead to the transcription and translation of genes required for neuronal differentiation and survival (8). The PI3K and ERK pathways lead to the activation of mTOR (mammalian target of rapamycin), a kinase of importance to functions such as “neurogenesis, axonal sprouting, dendritic spine growth, ionic and receptor channel expression, axonal regeneration and myelination” (9).

BDNF is of importance for glutamatergic transmission through its influence on presynaptic glutamate and GABA (gamma-aminobutyric acid) release. BDNF increases the availability of vesicles containing glutamate in presynaptic neurons. BDNF further modulates synaptic transmission by altering the activational patterns of NMDA receptors and GABA receptors as well as upregulating the surface expression of AMPA (alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid) receptors (6,9). BDNF levels have been found to be reduced in rodent depression models as well as in post-mortem studies of the PFC (prefrontal cortex) and hippocampus of depressed patients when compared to healthy controls. Various anti-depressant medications have been found to increase BDNF levels in plasma when given over an extended period, correlating with improvement of symptoms (7,8).

Ketamine stands out when compared to classical anti-depressants such as SSRI’s (selective serotonin reuptake inhibitors) when it comes to its regulation of BDNF levels. For instance, ketamine has been shown to upregulate BDNF levels within 15 minutes of administration to cortical cultures (10). A proposed explanation for the differences in BDNF regulation between SSRI and ketamine is demonstrated in Figure 1. In this model, ketamine works more rapidly by increasing translation of BDNF whilst SSRI’s act by increasing transcription of BDNF, a route which in theory would be more time consuming (9,11).

(8)

BDNF. Increased availability of BDNF and its release increases TrkB activation on NPCs, again promoting neurogenesis and anti-depressant effects(11) (Image available via CC BY 4.0 licence, Numakawa, 2018).

2.3.3 Dual pathology theory

An alternative explanation for the differences between the actions of traditional anti-depressants and ketamine is that two different mechanisms altogether are of

relevance. This theory of a dual pathology is based on the assumption that ketamine reverses chronic stress pathology by upregulating neurotrophic factors and protein synthesis in the PFC and hippocampus within 24 hrs of administration, creating a long lasting and increased synaptic connectivity. In contrast, traditional anti-depressants would reverse chronic stress pathology by attenuating the mono amine regulation, for instance by reversing hyperconnectivity in dopaminergic pathways of the NAc (6).

2.4 Background into the methods behind the reviewed studies

2.4.1 Models of depression in rodents

Animal models of depression aim to simulate a depressive-like phenotype in which a combination of certain behavioural endpoints are met. For example, the animal should exhibit some of the following symptoms (12):

•_Anhedonia

• Apathy or depressed mood • Irritability • Disturbances in: o Sleep o Appetite o Psychomotor behaviour o Social behaviour • Anxiety • Cognitive impairment

It should be noted that these models are unable to simulate or give the researchers the ability to measure certain aspects of depression accurately. Feelings that are associated with depression such as worthlessness, guilt, suicidal ideation or attempts to commit suicide are not applicable to any of the animal-based models of

depression (12).

2.4.1.1 Stress-based models

Stress-based models are the most commonly used models for depression and their validity is considered fairly high, firstly because stress is seen as an established trigger for depression in humans, and secondly as these models result in similar neurobiological changes to what has been seen when studying depression in humans (12).

(9)

Unpredictable chronic mild stress (UCMS/CUMS) is a model in which the rodents are exposed to a variety of mild but chronic and unpredictable socio-environmental stressors such as "changes in lighting, contention in a small tube, introduction of

rats faeces, cage tilting, social stress, cage changes, sawdust changes, no sawdust, humid sawdust, water in the cage” (12).

Depression-like response to the treatment in this model is associated with neurobiological changes such as:

• Dysregulation of hypothalamus-pituitary-adrenal axis • Hippocampal reduction of

o BDNF levels o Neurogenesis o Dendritic branching • Increased microglial activation

• Reduced 5-HT transmission in the forebrain

• Reduction of BDNF levels and dendritic branching in certain frontal regions Social defeat stress model of depression involves the test rodents during a ten day-period being placed in the home cage of an aggressive rodent for ten minutes per day, then having to spend the rest of the day within range to see, hear and smell the aggressor, each day it is introduced to another aggressor. This leads to depressive-like symptoms in a substantial proportion of the rodents, with some individuals showing resilience. The main symptoms associated with this model are social withdrawal and anhedonia, and these symptoms correlate with dysregulation in the PFC as well as changes in the levels of neurotrophic factors (12).

2.4.1.2 Genetic models

Due to depression having a genetic component, various genetic lines of rodents have been bred with abnormal sensitivity to developing a depressive-like phenotype. In addition, genetic knockout techniques may be used to study the impact of specific genes and the interactions of these genes with various treatments (12).

The transcription factor Nrf2 (nuclear factor erythroid derived 2-related factor) is of importance to the cellular defence against inflammatory processes. Rodents in the Nrf2 knock-out model have increased susceptibility to inflammation and

depression-like symptoms (13).

Heterozygous BDNF knock-out mice have a reduced production capacity for BDNF with a production between 40-60% of BDNF compared to wild type mice (14).

2.4.2 Main methods used for studying effects of ketamine treatment

2.4.2.1 Behavioural tests

Rodent behavioural tests in studies of depression assist in evaluating the

effectiveness of treatments. The tests are not developed with a particular pathology in mind, but can be useful in evaluating the changes in cognition and emotion of the rodents (15). As the results of these tests only serve to lend support to the

biochemical findings that are looked at in this review, they will not be described in detail here, but the tests involved are the following:

•_{Tail suspension test (TST) – Animal is suspended by the tail, after some} time it ceases its attempts to free itself, the longer it remains inactive the more it is deemed to show depressive-like behaviour (15)

(10)

•_{Evaluated plus maze – The ratio between avoidance behaviour and}

explorative behaviour in a four-armed maze is used to gauge anxiety levels • Novel object recognition test (NORT) – In order to assess the anxiety and

cognitive function, the test animal is introduced into the same cage two days in a row for ten-minute intervals. They are then introduced to two identical objects in the cage for ten minutes, the next day one of the objects is swapped for a novel object and the time spent on investigating the new object is used to measure the parameters mentioned (16)

• Forced swim test (FST) – Rodent is forced to swim in a container that cannot be escaped, after some time it starts showing despair-like behaviour, only performing the absolute necessity of movement for survival, length of time to reach this despair-like behaviour is used as a measure of depressive-like phenotype (15)

• Sucrose preference test (SPT) – Using the test animal’s sensitivity to rewards by its preference of water with different levels of sucrose, the level of depressive behaviour is assessed (15)

2.4.2.2 Western blotting (WB)

The western blot is a technique that allows researchers to separate and identify specific proteins from tissue samples or cell cultures. The WB is performed through a combination of separation by weight (using gel electrophoresis) and antibody binding using antibodies specific to the protein of interest (17).

2.4.2.3 Golgi stain

Golgi stain is a technique that involves using silver nitrate to stain whole neurons, including their dendrites and axons. The technique makes it possible to study dendrite branching and density (18).

2.4.2.4 Pharmacological manipulation

To determine the importance of BDNF-TrkB neurotransmission response to ketamine treatment, the primary method is by blocking the TrkB receptor; the substances used in the studies covered in this review are ANA-12 (13,19,20) and K252 (21).

2.5 Statement of purpose

The purpose of this study was to investigate how ketamine influences BDNF-TrkB neurotransmission and downstream signalling in the hippocampus and prefrontal cortex in rodent models of depression.

3 Method

The initial literature search was performed using PubMed's search function using the following search terms: search words “anti-depressant”, “ketamine”, “TrkB” and results were filtered by “animal studies” as these enable the investigation of specific molecular mechanisms in a way that is not possible with human studies. The search yielded 52 results, the abstracts of these were read to determine their suitability in answering the question posed, and exclusion from the results section of this paper was done according to the following:

• Review papers

• Papers on other substances than ketamine or its enantiomers

• Papers with the main focus on the combination of ketamine and another substance

(11)

•_{Papers focused on verifying already established facts such as ketamine's} rapid anti-depressant effects

• Papers covering non-rodent species

Following exclusion, 14 papers remained, see Appendix 1 for detailed summary, including excluded articles.

4 Results

4.1 Prefrontal cortex

Wu et al. tested the impact of three different doses of ketamine, (10 mg/kg, 30 mg/kg and 100 mg/kg) on behavioural parameters in mice and WB analysis to measure BDNF expression of PFC and hippocampus. In the behavioural tests, the mice were clearly impaired by the higher dose, they showed no anti-depressant-like effect at the lowest dose, only the 30mg/kg showed anti-depressant-like effect. By blocking TrkB with ANA-12, a selective TrkB receptor antagonist, the anti-depressant effect of ketamine was removed. For the PFC WB showed that BDNF expression was increased by the two higher doses, but in the hippocampus the higher dose lowered BDNF expression, while the lower doses had no impact. Interestingly the researchers found that the change in BDNF expression occurred without altering GluA1 (AMPA receptor), eEF2 (eukaryotic elongation factor 2), mTOR and PSD-95 (postsynaptic density protein 95) at the time of testing, 24 hrs post treatment. This result indicates that the cascade effects of BDNF have not occurred at the time when anti-depressant like effect (anti-depressant like effect) was identified, and as the TrkB antagonist blocked anti-depressant like effect of ketamine it is possible that BDNF may act through another receptor when TrkB is blocked (19).

Li et al. investigated the impact of ketamine vs ketamine combined with a TrkB antagonist on post-operative depression (POD) following abdominal surgery in mice by performing behavioural testing and WB on mPFC (medial PFC), hippocampus, liver and muscles. It was found that regardless of if the mice showed post-operative depressive symptoms or not, they had a reduced expression of BDNF in medial PFC when compared to the control group of mice that have not undergone surgery. The mice that were susceptible to POD showed significantly lower TrkB

phosphorylation in the medial PFC as well as lower TrkB and BDNF presence in liver and muscles compared to mice that showed POD resistance. Ketamine (10 mg/kg) was found to reverse or normalize the above effects and was blocked by inhibition of TrkB with ANA-12 (0,5 mg/kg) (22).

(12)

Kohtala et al. 2019 investigated how the brain oscillations in PFC were altered after ketamine treatment by performing EEG (electroencephalography) studies after administration of different doses of ketamine. Subanaesthetic doses (10mg/kg) increased gamma high oscillations slightly, but it was found that only higher doses (100mg/kg) caused a significant increase in delta, theta, beta as well as both high and low gamma oscillations whilst also lowering alpha oscillations. Similarities between these results and EEG studies of brain oscillations in patients post ECT (electroconvulsive therapy) or general anaesthesia led the researchers to speculate that the altered brain state may be of importance to the anti-depressant effect of ketamine in a similar way to ECT (24)

Further, the researchers looked at how TrkB activation was altered post ketamine infusion using WB of the PFC, and in order to determine if the metabolite HNK (hydroxynorketamine) was of relevance to the anti-depressant effect, they also tested this independently of ketamine. Ketamine administration at high dosage resulted in an increase of TrkB activation, as judged by TrkB-phosphorylation and GSK3ß-phosphorylation (glycogen synthase kinase 3 beta), as well as downstream mTOR activation, as judged by p70S6K-phosphorylation (ribosomal protein S6 kinase), as early as thirty minutes post infusion. Lower dosage showed no effect at this time point, nor did direct HNK treatment. This was interpreted as indicating that ketamine's rapid anti-depressant effect and effect on TrkB could be seen as

independent of the metabolism of the drug, indicating that HNKs anti-depressant effect could be attributed to another mechanism altogether. The researchers suggest that the combination of results achieved through WB and EEG may indicate that ketamine triggers a homeostatic rebound through increasing cortical excitability and then altering the oscillations of the PFC, leading to an increase in TrkB signalling and the neurotrophic effects that follow (24).

Fernandes et al. attempted to map the distribution pattern of ERK phosphorylation in the rodent brain post ketamine treatments that held a range of different doses. The study was performed by utilising microwave irradiation to fixate brain chemistry at the set time point. Dissection was performed to extract samples of mPFC and central amygdala nucleus, yielding 4-10 slides per animal. For the control group of animals, which were not treated with ketamine a few scattered phosphorylated ERK-ir (immunoreactive) neurons were found in various regions but for the ketamine groups a significant increase in phosphorylated ERK-ir neurons was found in mPFC, CeA (central nucleus of the amygdala), DRN (dorsal raphe nucleus) and LC (locus coeruleus). This finding was dose dependent, with higher doses yielding higher phosphorylation rates, but was also time dependent – at 5 to 10 minutes the rates were at their highest, dropping quickly to near baseline at 20 minutes. No significant effect was found on hippocampus, striatum, thalamus or hypothalamus (25).

(13)

Yang et al. compared how R-ketamine and S-ketamine impacted BDNF-TrkB neurotransmission, the anti-depressant effect and impact that the enantiomers had on synaptogenesis in prelimbic PFC, DG and CA3. Both enantiomers significantly improved BDNF levels in social defeat stress model, but R-ketamine had

significantly higher effect than S-ketamine. This was also the case for attenuating the reduced dendritic spinal density at 8 days post treatment, both had significant effect, but R-ketamine had significantly higher effect (23).

4.2 Hippocampus and the hippocampal – mPFC pathway

Carreno et al. investigated how the ventral hippocampal and medial PFC pathway relates to ketamine's anti-depressant like response (anti-depressant like response). Study was first performed in rodents where ventral hippocampal neurotransmssion was temporarily deactivated by administering lidocaine prior to ketamine treatment and then performing behavioural testing at 30 mins post-treatment and one-week post-treatment. The anti-depressant like response was not blocked at 30 mins post treatment, however it was blocked at one-week post treatment. The researchers concluded that activity in the ventral hippocampus is required for sustained anti-depressant effect of ketamine but not for the acute anti-anti-depressant effects (21). Secondly, the study investigated whether blocking the pathway between the ventral hippocampus and medial PFC, by performing asymmetric disconnection with lidocaine, altered the anti-depressant like response to ketamine. The temporary blockage of the ventral hippocampus and medial PFC pathway was found to block the anti-depressant like response one-week post treatment, this was not the case for 30 mins post-treatment nor for symmetric disconnection. Sustained anti-depressant like response appeared to be dependent on the pathway (21).

Thirdly, the study investigated how activation of TrkB within the ventral

hippocampus occurred in response to ketamine. This was done through WB analysis of the ventral hippocampus at 30 minutes post ketamine treatment and at one-week post treatment. A significant effect in the activation of TrkB was found at 30 mins post treatment but not at one-week post treatment. When blocking the activation (phosphorylation) of TrkB using K252a there was a significant blocking of the anti-depressant like response at one week post ketamine treatment:this indicates that TrkB activation in the ventral hippocampus is also essential for sustained anti-depressant effect (21).

Overall, the study indicated that activity in the ventral hippocampus – medial PFC pathway is essential for the sustained anti-depressant like response of ketamine, but that different mechanisms may be of relevance for the acute anti-depressant like response to ketamine. The researchers suggest that ketamine may initiate a

molecular cascade via BDNF-TrkB activation leading to neuroplastic changes and a sustained anti-depressant effect (21).

The main focus of a study by Fred et al. was how the traditional anti-depressant fluoxetine impacted the interaction between TrkB and AP-2M (AP-complex 2, medium chain) on cell surfaces of hippocampal neurons, but they also looked at other substances including ketamine. AP2-2M complex is responsible for the clathrin-dependent endocytosis which downregulates TrkB receptors on the cell surface, ketamine was along with the other anti-depressant substances found to interrupt the interaction between TrkB and AP-2M, “resulting in an increased exposure of TrkB at the cell surface”, in turn increasing the probability of TrkB activation by BDNF (27).

(14)

model. Through WB they found that BDNF levels were significantly increased at both 4 days post treatment and 8 days post treatment, in the PFC, DG, CA1 and CA3 of the hippocampus. The precursor pro-BDNF was significantly higher in the PFC, the researchers also noted that GluA1 in the PFC was upregulated in the PFC and suggested that the long-lasting upregulation of GluA1 could be responsible for the sustained anti-depressant like effect (28).

Ma et al. examined general effects of ketamine on hippocampal neurogenesis but also specifically investigated the role of TrkB in the anti-depressant effect of ketamine by using mice lacking the gene for TrkB. These mice did not show increased depressive like behaviour compared to wild type mice when exposed to a pharmacologically induced model of depression, but critically they showed no response to ketamine treatment as measured by behavioural parameters. As another part of the study, ketamine’s effect on neural progenitor cell differentiation through BDNF-TrkB neurotransmission and downstream signalling was investigated and ERK phosphorylation was found to increase within 6 hours of treatment. This was assessed as being essential for the differentiation neural progenitor cells. The researchers suggest that the anti-depressant effect of ketamine may depend on BDNF-TrkB downstream signalling causing activation of ERK, followed by progenitor differentiation (29).

Liu et al. investigated if apoptosis in the hippocampus had relevance in

understanding ketamine's anti-depressant effect via BDNF-TrkB signalling. Using a UCMS rodent model they combined behavioural testing with WB, and staining techniques, to observe how ketamine treatment impacted apoptotic processes of astrocytes in the hippocampus and levels of BDNF, pCREB (phosphorylated cyclic adenosine monophosphate-response element binding protein), GLT1 (glutamate transporter 1) and PSD-95. Ketamine treatment significantly reduced apoptotic processes and improved dendritic spine density within 24 hrs of treatment. This effect was readily blocked by TrkB inhibition by pre-treatment with K252. Ketamine was also found to attenuate the reduction in levels of BDNF, pCREB, GLT1 and PSD-95 at this time point, again the effect was readily blocked by K252 administration pre-treatment (30).

As astrocytes are the main cells responsible for glutamate regulation in the hippocampus, impairment of these cells can lead to glutamate induced

neurotoxicity, and the researchers point out that this can lead to disturbances in cell signalling and neurotransmission – causing and perpetuating depressive symptoms. They further note that the more traditional anti-depressant fluoxetine also has been found to increase GLT1 to normal levels, although not at the same speed as ketamine (30).

(15)

4.3 Ketamine in cortical cultures

Using primary cortical cultures from rodent embryos Lepack et al. analysed ketamine's effect on levels of BDNF and phosphorylated ERK. WB and visual analysis were performed, AMPA and TrkB inhibition prior to ketamine treatment was also tested. Ketamine significantly attenuated levels of pTrkB at 60 min p.t. (effect was readily blocked by both TrkB and AMPA inhibition), it also increased neuronal complexity and the quantity of branch crossings during the 24 hr

incubation. For the different doses of ketamine an inverted u-shaped dose response was found for phosphorylated ERK at one hr post-treatment. The authors suggest that the reason why ketamine at high doses fails to produce heightened

phosphorylation of ERK could be due to ketamine at higher dosages blocking the post-synaptic NMDAR, rather than the pre-synaptic NMDAR on GABA

interneurons. They further discuss that while mono amine anti-depressants do not impact mTOR activation in PFC cultures post infusion unless given repeatedly, ketamine does. The authors also question whether the anti-depressant effect of ketamine seen in vivo occurs independently of an NMDAR dependent homeostatic response (10).

4.4 Summary of reviewed studies

(16)

Table 1. Summary of reviewed articles including models used, tests, time points, markers, dosages, pharmacological inhibitors and results.

1st Author Model Behavioral _tests Biochemical test collected at Tissue

time point Marker Area

Substance and

dosage Inhibitor Findings

Wu (19) Unspecified/ _{No model} EPM, NORT, TST, OFT,

FST WB BDNF PFC and Hipp Ketamine 10mg/kg, 30 mg/kg, 100mg/kg TrkB ANA-12 30 mg/kg dosage indicated as optimum dosage in TST, EPM, FST.

Effects readily blocked in FST by ANA-12

Li (20) Post surgical _depression TST, OFT, _{FST, SPT} WB BDNF, TrkB, pTrkB PFC and peripheral _tissue Ketamine _10mg/kg TrkB ANA-₁₂ Ketamine attenuated the reduction in BDNF and TrkB levels overall. Effects readily blocked by ANA-12 Qu (13) Nrf2 KO _{SPT, LMF}TST, FST, WB 8 days p.t. GluA1, PSD-95 PFC R-ketamine _10mg/kg TrkB ANA-₁₂ reduction in GluA1 and PSD-95. R-ketamine attenuated the

Effect readily blocked by ANA-12.

Kohtala

(24) Unspecified/ No model None WB

3 minutes and 30 minutes pTrkB, pGSK3ß, p70S6K PFC Ketamine 10mg/kg- 200 mg/kg HNK 20 mg/kg

Ketamine yielded significant, dose-dependent, increases in pTrkB, pGSK3ß, p70S6K at 30 mins p.t. at >100 mg/kg. HNK no effect at same

time point. Ketamin 200mg/kg increased markers at 3 mins p.t. - prior to significant metabolism of

drug to HNK. Fernandes

(25) Unspecified/ No model None Immunohistochemistry pERK

mPFC, CeA, DRN, LC, Hipp, striatum, thalamus, hypothalamus Ketamine 10mg/kg, 30 mg/kg

Significant increase in pERK within PFC, CeA, DRN and LC at 5 and 10 mins p.t., levels nearly back to

baseline at 20 mins.

Dong (26) Social defeat _{stress model} TST, FST, SPT WB and Golgi stain 7 days p.t. BDNF, TrkB, GluA1, PSD-₉₅ PFC, DG, CA1, CA3, _NAc Ketamine _10mg/kg

Ketamine significantly attenuated the reduction in BDNF, TrkB, GluA1

and PSD-95 in PFC, DG and CA3. Reduction in dendritic spinal density

was attenuated in pre-limbic PFC, DG and CA3. Yang (23) Social defeat stress model and learned helplessness model

TST, FST, SPT WB, Golgi stain, parvoalbumin

immunohistochemistry 8 days p.t. BDNF PFC, DG, CA1, CA3, NAc R-ketamine 10mg/kg, S-ketamine 10mg/kg ANA-12

Both enantiomers significantly improved BDNF levels in pre-limbic

PFC, DG and CA3. Effect readily blocked by ANA-12. Both significantly attenuated reduced dendritic spinal density in PFC, DG and CA3. R-Ketamine in both cases had significantly higher attenuation

than S-Ketamine. Carreno

(21) Unspecified/ No model FST WB 30 min and 7 days p.t. TrkB, pTrkB VHipp Ketamine 10mg/kg K252

Ketamine significantly increased pTrkB at 30 min p.t. but no significant effect remained at 7 days

p.t.K252 readily blocked ketamines effect on pTrkB Fred (27) Unspecified/ _{No model} None MS 15 min p.t. TrkB Hipp _{(dose n.s.)}Ketamine Ketamine was found to interrupt the interaction between TrkB and

AP-2M Zhang (28) Social defeat _{stress model} TST, FST WB _{days p.t.}4 and 8 BDNF, proBDNF, PSD-95, _GluA1 PFC, DG, CA1, CA3, _NAc Ketamine _10mg/kg

Ketamine significantly attenuated the reduction in BDNF in PFC, DG CA1 and CA3. proBDNF significantly

higher in PFC only. Ma(29) Pharmacolo gically induced depression

FST Immunohistochemistry 6 hrs p.t. TrkB, ERK Hipp (DG) Ketamine 7 _mg/kg

Ketamine was only able to alleviate depressive behaviours in mice with

TrkB-gene. TrkB-dependent activation of ERK was suggested to

be essential for ketamines anti-depressant effecT

Liu (30) Unpredictable chronic

mild stress FST, OFT

WB, Golgi stain,

immunoflourescent stain 24 hrs p.t. BDNF, pCREB, GLT1, PSD-95 Hipp Ketamine 10mg/kg TrkB K252

Ketamine significantly reduced apoptotic processes attenuated reduction in dendritic spine density

24 hrs p.t. Ketamine attenuated reduction in levels of BDNF, pCREB,

GLT1 and PSD-95. Both effects readily blocked by K252. Lindholm (14) Genetic BDNF reduction model

FST, OFT Immunoblot detection and enzyme-linked immunosorbent assay 60 min or 7 days p.t. BDNF, pTrkB Hipp Ketamine 20mg/kg, 50mg/kg Significant attenuation of behavioural parameters but no significant effect on BDNF or pTrkB

at either time point Lepack (10) No model - study of cortical culture

None WB, Immunoblot 60 min BDNF, pERK Cotrical culture

Ketamine 100 nM or 500 nM/ 0.6 million cells TrkB K252 AMPA NBQX

(17)

5 Discussion

5.1 BDNF and TrkB neurotransmission

The purpose of this study was to investigate ketamine’s effect on BDNF-TrkB neurotransmission and downstream signalling in animal models of depression. Overall, in the reviewed studies, ketamine at the mode dosage of 10mg/kg bodyweight, was found to attenuate behavioural parameters and the reduction in levels of BDNF and TrkB neurotransmission, in the PFC and specific regions of the hippocampus at various time points post treatment. Time points extended from 3 minutes to 8 days. Also, in studies that did not use a depression model, upregulation occurred in response to ketamine. The main exception from these results was in the study by Lindholm et al. (14) where the mice with genetically reduced BDNF levels showed improvement in behavioural parameters, but no effect on levels of BDNF or pTrkB either at 1- or 7-days post treatment. It is worth noting that the Lindholm et

al. study used an alternative method to WB for analysing proteins, and that they

utilised doses of 20 mg/kg and 50 mg/kg (14).

The Lindholm et al. study and the Qu et al. study both used genetic models to increase susceptibility to depression-like phenotypes in the animals tested (13,14). In both studies, ketamine/R-ketamine significantly attenuated behavioural

parameters but only in the case of Qu et al. was there an effect on BDNF-TrkB neurotransmission and downstream signalling, as measured by an increase of the downstream marker’s PSD-95 and GluA1 (13). The difference between these two studies, taken together with ketamine in the other studies generally attenuating the reduction in BDNF-TrkB signalling, shows that although BDNF-TrkB

neurotransmission is of importance to ketamine's anti-depressant effects it is not the only mechanism of relevance.

Liu et al. found that ketamine reduced both astrocyte apoptosis and upregulated BDNF levels, thereby helping to break the vicious cycle of chronic stress with increased glucocorticoid levels, exitotoxicity due to excess glutamate, cell damage etc (30). Ma et al. found that TrkB and downstream activation of ERK was essential for the anti-depressant effect of ketamine, as well as for the differentiation of neural progenitor cells (29). Kohtala et al. proposed that ketamine triggers a homeostatic rebound through increased TrkB activation if preceded by cortical excitation, in a similar way to ECT treatment (24). These studies point toward the possibility that ketamine gives the right brain areas a “nudge”, leading to cascade-like changes that last long after the substance has been eliminated from the system.

5.2 Brain regions and pathophysiology

In terms of the brain regions implicated in depression, this review has focused on studies looking at the prefrontal cortex and hippocampus. Ketamine's impact on BDNF-TrkB neurotransmission in these two areas, their sub regions and joint pathway, illustrate that ketamine has different effects on neurons in different regions. As a part of the introduction to this review, the recently suggested dual pathology theory of depression was mentioned; in which dysfunction of mono amine transmission and glutamatergic transmission represent two distinct

(18)

Whether ketamine’s effects on depression indicate a dual model of pathology or not, it certainly cements that the mono amine theory alone cannot explain depression. Assuming that BDNF-TrkB neurotransmission and downstream signalling is essential for ketamine’s anti-depressant effects, perhaps the mono amine theory is just one piece of the puzzle, rather than an altogether separate pathology from one related to glutaminergic transmission. The effect of ketamine on BDNF-TrkB neurotransmission provides insight into how complex the pathology of depression is, and that better animal models are needed to improve our understanding of the disorder.

5.3 Animal models of depression

Most animal models of depression show a clear correspondence in symptoms between depression in humans and in the animal models, and there is evidence of appropriate neurobiological changes taking place in testing animals. Animal models of depression are however not perfect, as they are only able to model specific subtypes of depression (12).

"... depression lies on a putative combination of factors and thus, there is no a unique, specific neuronal circuit involved in its pathophysiology, but rather a multiple alteration of intricate networks" (12)

The complexity of the disease makes it impossible to completely mimic its

pathology entirely in rodents, making the models usefulness extend only to studying fairly specific pathologies. The basis for the validity of animal models of MDD is that they show a replicable response to traditional anti-depressants. They have been developed to provide predictive validity. This makes some of the models uncertain when it comes to studying TRD as other target mechanisms may be involved (12). This has implications for using the models when studying the effects of ketamine on animal models. If we acknowledge the possibility of depression having a dual pathology based on response vs nonresponse to traditional anti-depressant

medications, then the animal models used need to be representative of response vs response. Ketamine's ability to alleviate depressive symptoms in

non-responders needs to be replicable in animal models in order to enable studying alternative target mechanisms.

The UCMS model as well as the social defeat model makes it possible to split rodent groups into responders and non-responders, and in the social defeat model response to ketamine is similar to what is seen in TRD humans (12).

Depression's complexity is clearly difficult to capture in animal models, but they do provide a good starting point when it comes to studying specific target mechanisms. In the studies reviewed several different animal models where used, making the results difficult to compare but also demonstrating that certain points are made repeatedly across different methodologies.

The large variation between the reviewed studies, in terms of methodology, poses a challenge for interpreting the results and their meaning. Several of the studies were performed without utilizing a model of depression. It is questionable if the results obtained, when treating healthy rodents with ketamine, are transferable to how the rodents respond in a depression model, not to mention human studies of actual depression.

5.4 Future perspective and conclusions

(19)

of depression in the future. An interesting approach would be to investigate if there is a difference to how the symptoms of responders and non-responders to SSRIs are affected by ketamine. Looking for similarities and differences between

non-responders to both drugs could also lend important insights to treatment resistance and help identify other mechanisms to target.

In conclusion, this work aimed to review research into ketamine’s effect on BDNF-TrkB transmission in rodent models of depression. Reliable animal models of depression are able to reproduce similar neurobiological changes to those that occur in human depression and downregulation of BDNF-TrkB neurotransmission

(20)

6 References

1. World Health Organization. Depression [Internet]. 2020 [cited 2021 Mar 8]. Available from:

https://www.who.int/news-room/fact-sheets/detail/depression

2. Banich MT, Compton RJ. Cognitive neuroscience. Cambridge, United Kingdom ; Camebridge University Press; 2018.

3. Marcatili M, Sala C, Dakanalis A, Colmegna F, D’Agostino A, Gambini O, et al. Human induced pluripotent stem cells technology in treatment resistant depression: novel strategies and opportunities to unravel ketamine’s fast-acting antidepressant mechanisms. Therapeutic advances in

psychopharmacology. 2020;10:2045125320968331.

4. Ritter J. Rang and Dale’s Pharmacology. Ninth edit. Rang HP, Flower RJ (Rod J), Henderson G, Loke YK, Rang HP, Dale MM, editors.

Pharmacology. Amsterdam: Amsterdam : Elsevier; 2019.

5. Athira KV, Bandopadhyay S, Samudrala PK, Naidu VGM, Lahkar M, Chakravarty S. An Overview of the Heterogeneity of Major Depressive Disorder: Current Knowledge and Future Prospective. Current

neuropharmacology [Internet]. 2020;18(3):168–87.

6. Mihaljević S, Pavlović M, Reiner K, Ćaćić M. Therapeutic mechanisms of ketamine. Psychiatria Danubina. 2021;32(3–4):325–33.

7. Duman RS, Li N. A neurotrophic hypothesis of depression: role of

synaptogenesis in the actions of NMDA receptor antagonists. Philosophical transactions of the Royal Society of London Series B, Biological sciences [Internet]. 2012 Sep 5;367(1601):2475–84.

8. Cunha C, Brambilla R, Thomas K. A simple role for BDNF in learning and memory? [Internet]. Vol. 3, Frontiers in Molecular Neuroscience . 2010. p. 1.

9. Ignácio ZM, Réus GZ, Arent CO, Abelaira HM, Pitcher MR, Quevedo J. New perspectives on the involvement of mTOR in depression as well as in the action of antidepressant drugs. British journal of clinical pharmacology [Internet]. 2016/01/08. 2016 Nov;82(5):1280–90.

10. Lepack AE, Bang E, Lee B, Dwyer JM, Duman RS. Fast-acting antidepressants rapidly stimulate ERK signaling and BDNF release in primary neuronal cultures. Neuropharmacology [Internet]. 2016/09/12. 2016 Dec;111:242–52.

11. Numakawa T, Odaka H, Adachi N. Actions of Brain-Derived Neurotrophin Factor in the Neurogenesis and Neuronal Function, and Its Involvement in the Pathophysiology of Brain Diseases. International Journal of Molecular Sciences. 2018 Nov 19;19:3650.

12. Planchez B, Surget A, Belzung C. Animal models of major depression: drawbacks and challenges. Journal of Neural Transmission [Internet]. 2019;126(11):1383–408.

(21)

14. Lindholm JSO, Autio H, Vesa L, Antila H, Lindemann L, Hoener MC, et al. The antidepressant-like effects of glutamatergic drugs ketamine and AMPA receptor potentiator LY 451646 are preserved in bdnf+_/−_{heterozygous null}

mice. Neuropharmacology. 2012 Jan;62(1):391–7.

15. Belovicova K, Bogi E, Csatlosova K, Dubovicky M. Animal tests for

anxiety-like and depression-like behavior in rats. Interdisciplinary toxicology [Internet]. 2017 Sep;10(1):40–3.

16. Lueptow LM. Novel Object Recognition Test for the Investigation of Learning and Memory in Mice. Journal of visualized experiments : JoVE [Internet]. 2017 Aug 30;(126):55718.

17. Mahmood T, Yang P-C. Western blot: technique, theory, and trouble shooting. North American journal of medical sciences [Internet]. 2012 Sep;4(9):429–34.

18. Binder MD, Hirokawa N, Windhorst U, editors. Golgi Stain BT - Encyclopedia of Neuroscience. In Berlin, Heidelberg: Springer Berlin Heidelberg; 2009. p. 1756.

19. Wu C, Wang Y, He Y, Wu S, Xie Z, Zhang J, et al. Sub-anesthetic and anesthetic ketamine produce different long-lasting behavioral phenotypes (24 h post-treatment) via inducing different brain-derived neurotrophic factor (BDNF) expression level in the hippocampus. Neurobiology of learning and memory. 2020 Jan;167:107136.

20. Li S, Luo X, Hua D, Wang Y, Zhan G, Huang N, et al. Ketamine Alleviates Postoperative Depression-Like Symptoms in Susceptible Mice: The Role of BDNF-TrkB Signaling. Frontiers in pharmacology [Internet]. 2020 Feb 6;10:1702.

21. Carreno FR, Donegan JJ, Boley AM, Shah A, DeGuzman M, Frazer A, et al. Activation of a ventral hippocampus–medial prefrontal cortex pathway is both necessary and sufficient for an antidepressant response to ketamine. Molecular Psychiatry [Internet]. 2016;21(9):1298–308.

22. Li S, Luo X, Hua D, Wang Y, Zhan G, Huang N, et al. Ketamine Alleviates Postoperative Depression-Like Symptoms in Susceptible Mice: The Role of BDNF-TrkB Signaling. Frontiers in pharmacology. 2019;10:1702.

23. Yang C, Shirayama Y, Zhang J, Ren Q, Yao W, Ma M, et al. R-ketamine: a rapid-onset and sustained antidepressant without psychotomimetic side effects. Translational psychiatry [Internet]. 2015 Sep 1;5(9):e632–e632. 24. Kohtala S, Theilmann W, Rosenholm M, Müller HK, Kiuru P, Wegener G,

et al. Ketamine-induced regulation of TrkB-GSK3β signaling is

accompanied by slow EEG oscillations and sedation but is independent of hydroxynorketamine metabolites. Neuropharmacology. 2019

Oct;157:107684.

25. Fernandes A, Li Y-W. Focused microwave irradiation-assisted immunohistochemistry to study effects of ketamine on phospho-ERK expression in the mouse brain. Brain Research [Internet]. 2017;1670:86–95. 26. Dong C, Zhang J-C, Yao W, Ren Q, Ma M, Yang C, et al. Rapid and

(22)

The international journal of neuropsychopharmacology. 2017 Mar;20(3):228–36.

27. Fred SM, Laukkanen L, Brunello CA, Vesa L, Göös H, Cardon I, et al. Pharmacologically diverse antidepressants facilitate TRKB receptor activation by disrupting its interaction with the endocytic adaptor complex AP-2. The Journal of biological chemistry. 2019 Nov;294(48):18150–61. 28. Zhang J, Yao W, Dong C, Yang C, Ren Q, Ma M, et al. Comparison of

ketamine, 7,8-dihydroxyflavone, and ANA-12 antidepressant effects in the social defeat stress model of depression. Psychopharmacology. 2015 Dec;232(23):4325–35.

29. Ma Z, Zang T, Birnbaum SG, Wang Z, Johnson JE, Zhang C-L, et al. TrkB dependent adult hippocampal progenitor differentiation mediates sustained ketamine antidepressant response. Nature communications [Internet]. 2017 Nov 21;8(1):1668.

(23)

Appendix 1

Ref no Exclusion by _criteria PMID Title First _Author DOI

OS 28798343 Isoflurane produces antidepressant effects and _{induces TrkB signaling in rodents} Antila H 10.1038/s41598-017-08166-9

Review 26519901 BDNF - a key transducer of antidepressant _effects Björkholm _C 10.1016/j.neuropharm.2015.10.034

OS 32688367 Mechanisms associated with the antidepressant-_{like effects of L-655,708} Bugay V 10.1038/s41386-020-0772-2

21 Included 26619811

Activation of a ventral hippocampus-medial prefrontal cortex pathway is both necessary and sufficient for an antidepressant response to ketamine

Carreno

FR 10.1038/mp.2015.176

Computational

model 33606976 Antidepressant drugs act by directly binding to TRKB neurotrophin receptors Casarotto PC 10.1016/j.cell.2021.01.034 Human study 33370585 Treatment response to low-dose ketamine infusion for treatment-resistant depression: A

gene-based genome-wide association study Chen MH 10.1016/j.ygeno.2020.12.030

Rapid and Sustained Antidepressant Action of the mGlu2/3 Receptor Antagonist MGS0039 in the Social Defeat Stress Model: Comparison with Ketamine

Dong C 10.1093/ijnp/pyw089

Focused microwave irradiation-assisted immunohistochemistry to study effects of ketamine on phospho-ERK expression in the mouse brain

Fernandes

A 10.1016/j.brainres.2017.05.008

Pharmacologically diverse antidepressants facilitate TRKB receptor activation by disrupting its interaction with the endocytic adaptor complex AP-2

Fred SM 10.1074/jbc.RA119.008837

The effects of ketamine on BDNF-TrkB neurotransmission in animal models of depression When the brain loses TrkBactivation

When the brain loses TrkBactivation