University of Groningen
Biological interactions in depression: Insights from preclinical studies Moraga Amaro, Rodrigo
DOI:
10.33612/diss.165782986
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2021
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Moraga Amaro, R. (2021). Biological interactions in depression: Insights from preclinical studies. University of Groningen. https://doi.org/10.33612/diss.165782986
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
200
Chapter 9
Discussion
1. This thesis.
The main objective of this thesis was to study interactions between biological systems in the brain that are associated with depression, thereby obtaining insights into the differential response to pharmacological treatment of patients suffering from major depressive disorder (MDD). To this end, a combination of several approaches was used, utilizing models of depression in rats, behavioral and molecular analyses and PET imaging. First, methodological approaches to study the neurobiology of depression were discussed. Then, interactions between biological systems were studied in different animal models of depression, with and without pharmacological interventions, and in animals prenatally exposed to an antidepressant drug. Together, this thesis provides a multifocal view of the biological mechanisms interacting in animal models of depression, which will be further discussed in the following sections.
2. The search for new methods to study the neurobiology of depression.
As Major Depressive Disorder (MDD) is one of the most disabling psychiatric disorders in the world, a huge number of resources have been invested in research in this field in order to discover new treatments. However, knowledge on the onset and pathophysiology of this disorder and on the proper treatment for each individual within the depression spectrum is far from complete. The main challenges in the research of depression are the heterogeneous nature of the disorder, leading to a large variability in symptoms, and the differential responsiveness of patients to currently available treatments. In this regard, it is evident that more research in the field is needed. The use of animal models for the study of the neurobiology of depression offers advantages over clinical trials, such as the ability to control environmental factors, and the possibility to test new potential pharmacological drugs and to perform ex-vivo analyses of biological processes in the brain. Nonetheless, translation of these models to the human situation is difficult, due to different outcomes as a result of the use of different models and techniques, and the fact that these models only mimic part of the processes found in patients. This means that the search of new approaches for animal studies in this area, such as improved animal models or refinement of the outcome measurements, are needed to move research on depression forward.
PET is a promising technique to study changes in molecular processes in the brain in a non-invasive manner. Chapter 2 reviews the potential use of this imaging technique in the
202
research on the role of steroidal sex hormones in the brain. The ability of these hormones to cross the blood-brain barrier and their receptors being present in several brain areas has led to the investigation of the role of these hormones in the central nervous system [1]. The contribution of these hormones to brain function could be of interest for psychiatric disorders, as steroidal sex hormones could have a great influence on the susceptibility/resilience to develop these disorders. Despite the advancements made in this research field, there is a technical limitation to study the neurobiology of sex hormones, since biological studies have been performed mainly in animal models and post-mortem brain studies. In this context, the use of non-invasive molecular imaging techniques may help to overcome these limitations, providing a technical platform to study longitudinal changes in these hormone systems during the development of psychiatric disorders such as depression. However, there are some issues to consider when using PET imaging.
The main challenges of PET imaging of sex hormones in the brain are the low receptor expression [2] and difficulties in obtaining radiotracers with sufficient affinity, metabolic stability and good penetration of the blood-brain barrier [3]. Several radiotracers for sex hormone receptors have been produced, whereas only the estrogen receptor tracer [18F]-FES
was able to detect changes in estrogen receptors in the pituitary of both humans and rodents [2,4]. However, more research is required for the development of suitable tracers to measure sex hormone receptors in the brain and it is not likely they will become available in the near future.
The effects of sex hormones in the brain have also been indirectly studied by measuring downstream biomarkers. These approaches include the use of tracers, that can detect changes in the metabolic pathways of sex hormones in the brain (e.g. aromatase), brain metabolism, or neurotransmitter markers, in healthy volunteers, during periods of fluctuations of sex hormones (i.e. menstrual cycle), and in patients with hormone replacement therapies. Correlations between sex hormone levels and brain metabolism [5–9], serotonin biomarkers [10–16] or dopaminergic biomarkers [17] have been found using PET. In this sense, the use of validated biomarkers, which can successfully measure changes in the brain due to hormone fluctuations, can offer an advantage over the direct measurement of sex hormone receptors in the brain.
In conclusion, due to the lack of radiotracers which could measure sex hormone receptors in the brain, research on the effects of these hormones on brain function should be focused on the measurement of changes of downstream biomarkers that could be affected by hormonal fluctuations. In addition, the use of animal models can complement these studies, as
postmortem analysis with molecular biology techniques, such as western blot and immunohistochemistry, can help to increase our understanding of the biological pathways involved in this issue.
3. Risk factors in depression: studying the interaction between stress and menopause
It is already known that MDD is a multifactorial disease and several risk factors may influence its development [18]. One of these factors is related to estrogen fluctuations. Studies have shown that women have an increased risk of developing depression during periods of hormonal changes, such as pre-menstrual, during pregnancy and post-partum, and both peri- and post-menopausal periods [19,20], positioning estrogens as neuroprotective hormones against MDD. However, the fact that not all women with decreased estrogen levels (i.e. post menopause) develop depression [21], implies that additional risk factors are needed in order to trigger this disorder. Furthermore, the fact that individuals respond differently to the same risk factors, and the interaction between different risk factors may give different outcomes, makes research on depression highly challenging. To overcome this hurdle, risk factors could be studied in specific animal models, in order to determine their contribution to the development of depression. In Chapter 3, a reduction in estrogen levels and chronic stress were studied as risk factors of MDD. For this purpose, brain metabolism was measured in an animal model of menopause in combination with a stress protocol.
In female rats, ovariectomy (OVX) is widely used as a model to induce a menopause-like state [22,23]. This model has the advantage of resembling the depressive-menopause-like and anxiety-like behavior that is found in some women during menopause [24]. However, in the human situation, not all postmenopausal women develop MDD or symptoms related to psychiatric disorders, and it is suggested that additional risk factors may be involved [21,25]. In this regard, the use of the chronic mild stress (CMS), mimicking daily stress, is a logical choice since it includes a variety of stress factors, simulating the challenges in the daily life of humans [26]. Chapter 3 showed that reduced levels of estrogen had an influence on behavior, physiology, and brain metabolism and that these effects were time dependent. However, the experimental approach used in this chapter showed that CMS did not worsen the behavioral and physiological changes induced by OVX, nor influenced brain metabolism. It is important to note that there are some limitations regarding the translation of the rat menopause model to peri-menopausal women: 1) Female rats do not go through menopause as humans, as the cycles last days in rats, compared to weeks in women, and 2) the transition of women through
204
menopause consists of a gradual change in levels of estrogens (perimenopause to menopause), which is not reproduced in our study because of the sudden reduction of estrogens due to OVX. Moreover, the lack of an effect of the CMS protocol does not exclude the possibility that stress in a certain population of women going through menopause may have a significant impact on MDD development. The use of different (and more severe) stress models, like for example the repeated social defeat (RSD) model, should be tested in order to exclude stress as a risk factor.
Although the translation of results from animals to humans is challenging, the results described in this thesis provide additional information about the temporal changes in brain metabolism when there is a large reduction of circulating estrogens in women. This knowledge can be taken into account for further studies on the influence of estrogen fluctuations in MDD, such as estrogen replacement therapy, and other therapies implying changes in sex hormone levels. Furthermore, the fact that the CMS protocol did not induce changes in behavior or brain metabolism in an estrogen depletion animal model of 10 weeks, highlights the limitations of animal models, which cannot be generalized to clinical conditions. In this regard, more studies using animal models of menopause with different periods of estrogen depletion, in combination with other chronic stress models could help to gain knowledge about the effects of hormonal fluctuations, and the specific stress factors that may be a risk factor for women for the development of MDD.
4. Pharmacological interventions to study modulation of brain inflammation by other biological systems.
Because the RSD model is a more severe stress model than CMS and was proven to induce neuroinflammation, this model was used to study the interaction between changes in biological systems in the brain and neuroinflammation using PET scans with a tracer targeting the 18kD translocator protein (TSPO) that is overexpressed in activated microglia. To this end, two studies were performed, using pharmacological interventions. Chapter 4 aimed to determine if the fast antidepressant effects of ketamine [27] induced changes in neuroinflammation in the RSD model of depression. Chapter 6 aimed to determine whether caffeine, known for its neuroprotective effect in MDD [28], is able to exert an antidepressant effect in this model through modulation of neuroinflammation.
Chapter 4 aimed to link the antidepressant effects of a single dose of the NMDA receptor antagonist ketamine [27] with changes in neuroinflammation, using the RSD model.
Contrary to what was expected, no effects of ketamine on depressive-like behavior, in particular the sucrose preference test (SPT) measured anhedonia induced by RSD, were found. Considering that the data obtained in this project was not sufficient to explain the absence of an effect of ketamine on behavior, it can be argued that the 5-day RSD model does not induce depressive-like behavior with the same intensity as other longer stress protocols [26]. In addition, most of the studies on ketamine in preclinical models measure the short-term effects (from hours to one day after the last stress session). This means that no information is available about how long does this effect lasts in animal models. Clinical studies have shown that the antidepressant effect of a single dose of ketamine in MDD patients, can last for more than a week. In this respect, the RSD model used in this chapter proved to be useful, as the antidepressant effects were still present (but weaker) one week after the RSD, which has not been shown with other models. In this regard, the use of models which induce long-lasting depressive-like behavior should be preferred in order to improve the translation of preclinical results to the clinical situation. An important limitation of our study is that only a single behavioral test was used to assess depressive-like behavior, while it is known that the heterogeneity in symptoms of MDD in patients is broad. Therefore, it is not possible to conclude whether ketamine treatment did have an effect on other symptoms in rodents. To improve the study design of the study described in Chapter 4, additional behavioral test should have been added to measure other symptoms as well, such as helplessness, which can be measured by the forced swim test (FST) and the tail suspension test (TST). Thus, the results described in Chapter 4 should only be interpreted for the effects of ketamine on anhedonic behavior.
The main objective of the study in Chapter 6 was to determine whether caffeine induces its antidepressant effect, as measured by the sucrose preference test (SPT), via modulation of neuroinflammation and microglial response. Caffeine is a component of beverages like coffee and acts as a non-selective adenosine receptor antagonist (A1R/A2AR). In chapter 6, it was observed that caffeine induces an antidepressant effect when administered for 2 weeks after the RSD protocol. In clinical studies, the effects of caffeine on depression have been studied in MDD patients. In these studies, it has been shown that caffeine acts as a prophylactic against MDD, but can also cause modulation of inflammatory markers in the brain [29]. In Chapter 6, the antidepressant effects of caffeine were measured, instead the prophylactic effects, by orally administering caffeine after the induction of depressive-like behavior. This study showed for the first time an antidepressant effect of caffeine in the RSD
206
rat model. However, in this study, although caffeine did induce antidepressant effects, it was not possible to link them with changes in neuroinflammation.
In Chapter 4, [11C]-PK11195 PET showed an increased uptake of TSPO one week
after RSD, as expected [30]. Conversely, the [11C]-PBR28 volume of distribution (V T)
analyses performed in Chapter 6 did not show changes in tracer binding due to RSD, measured 2 weeks after the last RSD session. An explanation for these differences could be the time point at which neuroinflammation was measured, as the timeline of the neuroinflammatory process can depend on the intensity of the psychiatric/neurological model and can last from days in mild models to months in the case of the most severe disease models [31]. In another study [30], neuroinflammation was found 1 week after RSD in several brain areas, but was not detected in the third week. Therefore, it can be argued that the lack of neuroinflammation two weeks after RSD is possibly caused by a remission of the inflammatory effect induced by RSD. In addition, differences in the PET tracers and analysis methods used made a direct comparison between studies difficult.
When animals were treated with either a single injection of ketamine or repeated caffeine administration for two weeks, no effect on neuroinflammation was found. In fact, both treatments enhanced the RSD-induced neuroinflammation, as observed by the increased tracer uptake in some specific brain regions. Nonetheless, the possible influence of these drugs decreasing neuroinflammation in the RSD model cannot be discarded, since it is possible that the time point at which neuroinflammation is measured may not the best time points to show the largest decrease in neuroinflammation. Additionally, the fact that the immunohistological analysis performed in Chapter 6 showed an increase in microglial density in both the frontal and occipital cortex, suggests that the tracers used did not have sufficient sensitivity to detect such small changes in neuroinflammation. In this respect, it is possible that the increased tracer uptake after the pharmacological intervention with ketamine or caffeine is related to an increased microglia density in the brain, and not due to microglial activation. Yet, studying the progression of neuroinflammation over time, using PET with a TSPO tracer, could be useful in determining how the neuronflammatory process evolves in these two studies, and what would be the optimal timing for the pharmacological treatments.
To conclude, several points must be considered when interpreting the data of these two intervention studies. One is the timing of neuroinflammation. When comparing these two chapters, our results suggest that even if the same RSD stress model is used, the neuroinflammation peak, as measured with a TSPO tracer for PET, may be different. Furthermore, even if the time of measurement is the same using the same animal model (when
comparing chapters with the same timing in this thesis), slight changes in the procedures can lead to different results (further discussed in subsection 9). Although the RSD model has been found to induce neuroinflammation in previous studies, including those described in this thesis, a decrease in neuroinflammation as a result of modulation of glutamatergic and adenosinergic transmission was not found under the experimental conditions tested in Chapter 4. Moreover, results obtained using TSPO tracers for PET to assess neuroinflammation can be misleading, since in MDD patients and animal models of depression, the intensity of inflammation processes may not be strong enough to be detected with this technique. This may be reflected in the high variability in results of both clinical and preclinical studies on depression, using TSPO as an inflammatory marker [32]. To overcome the possible lack of sensitivity of TSPO PET, it is recommended to measure additional markers in order to determine the degree of inflammation occurring in the brain, such as proinflammatory cytokines and oxidative stress. All this information together may help to understand the temporal dynamics of the inflammatory process, and may even help to find a temporal window, in which the use of anti-inflammatory drugs may be used as adjuvant therapy for MDD.
5. Refinement of the measurement of neuroinflammation with PET: which tracer should be used?
PET imaging is a useful technique to study molecular changes inside the body in a non-invasive way, but the reliability of this measurement depends on the properties of the tracers used. Ideal PET tracers should have certain properties, such as a high specificity and affinity for the biological target (low non-specific binding), limited metabolism to have enough tracer available for target binding, suitable kinetics for the duration of the PET scan (ability to reach relevant body compartments within the duration of the scan), and for the specific case of the central nervous system, it should be able to cross the blood brain barrier. In other words, tracers should have low non-specific binding and low binding to plasma proteins, and should generate low amounts of metabolites marked with radioactivity, in order to reduce the background signal. We used different TSPO PET tracers in this thesis for measuring of neuroinflammation in the RSD model of depression. While in Chapter 4 the first-generation tracer [11C]-PK11195 was used, Chapter 6 was performed using the second-generation tracer [11C]-PBR28. In Chapter 4, [11C]-PK11195 was selected because good results were obtained previously with this tracer in the same animal model [30]. However, this
208
PET tracer has some limitations, including low affinity and poor signal-to-noise ratio [33]. Therefore, a new generation of TSPO radiotracers was developed with improved affinity and signal-to-noise ratio. These new tracers showed good results in animal models of neurodegeneration, which is accompanied by severe neuroinflammation. Nonetheless, testing whether these improved tracers are also suitable for the use in specific models with only mild neuroinflammation is still necessary. Therefore, Chapter 5 compared the ability of [11
C]-PK11195 and [11C]-PBR28 scans, representing first and second generation tracers
respectively, to detect mild neuroinflammation in the repeated social defeat (RSD) rat model of depression.
In this study, [11C]-PBR28 was able to detect changes in TSPO expression in the brain
after RSD, whereas [11C]-PK11195 could not detect any significant TSPO variation in the
same animals. Yet, a high correlation between the brain uptake of the two tracers was observed, but the linear regression showed an intercept of >0 and a slope <1. This suggests that [11C]-PBR28 has a larger dynamic range and less non-specific binding than [11
C]-PK11195 and therefore should be more sensitive to detect mild inflammation in the brain. Therefore, Chapter 5 showed evidence suggesting that [11C]-PBR28 is a more suitable tracer
to study subtle changes in neuroinflammation than [11C]-PK11195, at least for the RSD model
of depression. Nonetheless, translation of these results to humans should be considered with care, especially since the affinity of [11C]-PBR28 was found to depend on a TSPO
polymorphism observed in humans [34], which complicates the interpretation of the results in humans, as well as the translation of results from animals to humans.
6. Studying molecular changes in the brain associated with the natural remission of depressive-like behavior: a new approach to study the neurobiology of depression.
When studying depression, multiple approaches and different points of view should be considered. Understanding the differences in pathophysiology between MDD patients who respond well and patients who do not respond to the treatment may be crucial for the improvement of current pharmacological therapies. Clinical studies in patients with treatment resistant depression (TRD) have suggested that abnormalities in several biological systems may be responsible for treatment resistance, including dysregulation in glutamatergic transmission, synaptic plasticity, neurotropic factors, the hypothalamic-pituitary-adrenal (HPA) axis, the immune system, neuroinflammation, as well as epigenetic changes [35]. However, it is difficult to determine the specific contribution of each biological system,
because of the heterogeneity of MDD in terms of symptomatology. Preclinical studies can help to increase our understanding of the mechanisms and interactions involved. An interesting approach in this respect could be to investigate the remission of depressive symptoms that occurs naturally in many animal models. One of the disadvantages of animal models is that depressive-like behavior does not persist for a long period. This apparent disadvantage, however, could be considered as a potential opportunity, as this spontaneous remission could be applied as a translatable model of remission in humans. Chapter 7 therefore aimed to determine changes in serotonergic and dopaminergic neurotransmission associated with the spontaneous remission of depressive-like behavior in the RSD model.
In general, many components of the serotonergic and dopaminergic transmission system (e.g. neurotransmitter concentration and their receptors, transporters and enzymes) are known to be affected by stressful events [36,37]. In this context, it is important to make a distinction between the mechanisms that cause short-term and long-term changes. It has been shown that a single social defeat trial induces an acute increase in dopamine and serotonin [38], leading to receptor desensitization and changes in their availability after repeated exposures [39]. In our study, long-term changes in D2R/D3R and SERT availability (14 days post RSD) were found, whereas short-term changes (1 day after RSD) were not found. A plausible explanation for the lack of short-term effects is the fact that changes in the availability of biomarkers of the serotonergic and dopaminergic system, including D2R/D3R and SERT, depend on the number of stressful events that animals are exposed to [40]. The number of stressful events was low in our study (5 trials) compared to most chronic stress protocols (from 10 to 40 trials). In contrast, appearance of long-term changes in our study may result from compensatory mechanisms due to changes in dopamine and serotonin concentration, resulting in desensitization of receptors. As a result of the desensitization, the same dose of a drug would produce a lower physiological response and thus limit the therapeutic effect of the treatment. Furthermore, the desensitization may also influence the development of other symptoms or other mood disorders. Moreover, these changes that are accompanying the natural remission of depressive-like behavior in rats may qualify as suitable candidates for biomarkers of remission.
In this regard, the natural remission of depressive-like behavior in stress-based animal models offers an opportunity to study biological markers of remission. Although it can be argued that this does not represent the pharmacological remission in MDD patients, it can help to understand which specific biological changes occur during remission that could facilitate the advance to a more personalized treatment. However, when translating our results
210
to the clinical situation they have to be carefully interpreted, since the observed changes in dopaminergic and serotonergic biomarkers may be specific for this particular preclinical depression model, and it cannot be excluded that the same risk factor may lead to different outcomes in different models. Therefore, generalization must be avoided, until a better understanding of the differences between animal models is available. This would offer the possibility to link a specific animal model with specific features in MDD and thus could improve the translation of preclinical data.
7. The risk of antidepressant treatment during pregnancy: long-lasting effects on mood and cognition in the offspring.
The importance of doing research on the neurobiology of depression has been highlighted in this thesis. However, research in the field of depression should also focus on the treatment of this disorder, as it may be an additional risk factors that could increase the psychological burden to society. The study described in Chapter 8 investigated the long-term effects of antidepressant exposure in utero [41]. Thus, the antidepressant most frequently prescribed to pregnant women (fluoxetine) was administered to female rats before and during pregnancy, and behavior and cognition was assessed in the in utero exposed offspring.
Using a medically relevant dose of fluoxetine, this model aimed to simulate a human situation in which treatment with antidepressant is prescribed before pregnancy, and lasts until the first period of pregnancy. In agreement with previous studies, in utero exposure to fluoxetine resulted in an increase in anxiety-like and depressive-like behavior [42–44], and a decreased cognition performance in the young adult male offspring. Interestingly, adverse changes were found only in cortical-related cognitive performance, inducing remote memory storage deficits in the Morris water maze, and learning deficits in non-hippocampal dependent memory, in particular object recognition memory. On one hand, studies have suggested that fluoxetine in utero exposure induced alterations in serotonergic and dopaminergic transmission and in synaptic plasticity, although the mechanisms underlying the behavioral changes are not known [45–47]. On the other hand, the memory impairments observed in this study could be the result of changes in glutamatergic transmission through the ionotropic receptors AMPA and NMDA receptors, known to be key players in synaptic plasticity [48]. Therefore, we performed additional western blot analysis, which revealed changes in the NMDA receptor subunits stoichiometry (NR2B/NR2A) in the temporal cortex, but not
hippocampus. No changes in AMPA receptor subunits in either of the brain areas analyzed
were observed. In early post-natal stages, the number of NR2B subunits present in NMDA receptors are higher than NR2A subunits, but during development the NR2B subunit is outnumbered by the NR2A subunit. Hence, it is possible that gestational fluoxetine exposure interferes with this process. Despite both NR2A and NR2B subunits are necessary for cognitive processes, an increased NR2B/NR2A ratio has been reported to be more beneficial for memory acquisition. Therefore, the decreased NR2B/NR2A ratio found only in the temporal cortex of animals exposed in utero to fluoxetine could explain why exclusively non-hippocampal dependent memory deficits were observed.
These results support the idea that in utero exposure to antidepressants, such as selective serotonin reuptake inhibitors (SSRIs), can induce long-term emotional and cognitive deficits that could remain obscured until adulthood [49]. They also support the existence of an interaction between serotonergic and glutamatergic neurotransmission during development [50]. Given that most of the clinical information related to the use of antidepressants during pregnancy is obtained from epidemiological studies, the use of preclinical studies, like the one present in this thesis, could help to improve our knowledge in this field. However, the human situation may differs from that in animal studies, in terms of the time that the fetus was exposed to the drug, the drug dose, and if the treatment started before or after conception. Therefore, the results obtained from protocols studying the in utero exposure to antidepressants should be interpreted carefully, since timing matters. Furthermore, differences in the exposure to different environmental stimuli (e.g. life style, nutrition, physical activity, medication) increase the spectrum of possible outcomes in humans, which cannot always be mimicked in preclinical studies. For this reason, it is important to determine to which extend results can be generalized from preclinical studies, since they may only be applicable for specific situations in humans.
8. A brief overview of the neuroanatomical network involved in depression.
Unraveling the interaction of biological systems is important for the improvement of currently used pharmacotherapy for MDD. However, as important as the molecular processes, identification of the brain areas in which these processes are taking place is vital for the understanding of the neuroanatomical networks associated to MDD. Knowledge of the networks involved may be used for diagnosis and differentiation of MDD subtypes with brain imaging techniques in the future. As shown in this thesis and in a variety of clinical and preclinical studies, imaging techniques can be useful tools to determine which biological
212
processes in specific brain areas are involved in MDD pathophysiology. An overview of the most relevant imaging findings from studies on the effects of stress and interventions on depressive-like behavior in rats, as observed in this thesis, is shown in Table 1.
Chapter Treatment/
intervention PET tracer Process Regional changes in tracer uptake
Chapter 3 Ovariectomy [18F]-FDG
(SUVmean, voxel-based analysis
Brain
metabolism Decreased: Midbrain, hippocampus, brainstem, thalamus, frontal cortex, parietal cortex, striatum, insular cortex, cerebellum, bed nucleus stria terminalis.
Chapter 3 Chronic mild
stress [ 18F]-FDG (SUVmean, voxel-based analysis Brain metabolism No changes Chapter 4 RSD [11C]-PK11195
(SUVmean) TSPO availability Increased: Enthorhinal cortex, insular cortex. Chapter 4 RSD +
ketamine [
11C]-PK11195
(SUVmean) TSPO availability Increased: Enthorhinal cortex, insular cortex, basal ganglia.
Chapter 5 RSD [11C]-PBR28
(SUVmean) TSPO availability Increased: Cerebellum, temporal cortex.
Decreased:
Bed nucleus stria terminalis Chapter 5 RSD [11C]-PK11195
(SUVmean) TSPO availability No changes Chapter 6 RSD +
caffeine [
11C]-PBR28
(VT)
TSPO
availability Increased: Amygdala, enthorhinal cortex, insular cortex, hypothalamus, orbitofrontal cortex, visual cortex, whole brain. Chapter 6 Caffeine [11C]-PBR28
(VT)
TSPO
availability Increased: Olfactory cortex Chapter 6 RSD [11C]-PBR28 (VT) TSPO availability No changes Chapter 7 RSD [11C]-raclopride (BPND) D2R/D3R
availability Increased: Caudate putamen Chapter 7 RSD [11C]-DASB
(BPND)
SERT
availability Increased: Brainstem, insular cortex, midbrain, thalamus.
Table 1. Description of changes in biological markers in brain specific areas. This table summarizes the
results found in the different chapters of this thesis which used PET to determine molecular changes. All changes described are compared to the control groups in each chapter (Chapter 3=EST+CTL; Chapter 4=CTL+VEH; Chapter 5=CTL, Chapter 6=CTL+VEH; Chapter 7=CTL)..
In neuroanatomical research of MDD, the prefrontal cortex is one of the most studied brain areas, since imaging studies have shown volume reductions, sulcal widening, decreased glucose metabolism, and alterations in blood flow in this brain region in MDD patients [51– 53]. In line with these observations, we found in this thesis a decrease in glucose metabolism in the prefrontal cortex due to estrogen depletion and an increase in TSPO availability due to RSD and caffeine treatment. On the other hand, no changes in SERT availability were found in the frontal cortex 1 day or two weeks after RSD. Studies showing changes in SERT and TSPO in the prefrontal cortex in chronic stress models of depression [30,54–56] suggest that differences in these biomarkers are time dependent, and can be different for each specific model (for further discussion, see subsection 9).
Another important system that is dysregulated in MDD is the limbic system. The limbic system contains brain structures, such as the amygdala, thalamus, hypothalamus and hippocampus, and is in charge of emotional processing and memory formation [57]. Results described in this thesis show increased TSPO binding in the amygdala and hypothalamus after RSD and caffeine exposure, and increased SERT in the thalamus, and decreased brain metabolism in the hippocampus and thalamus after RSD. The involvement of these areas in pathophysiology of depression has already been described [58–62]. However, the results described in this thesis also show an important contribution of the insular cortex. Changes in different biological systems, such as increased TSPO and SERT availability due to RSD and decreased metabolism in ovariectomized female animals were found in the insular cortex. These data support several findings, suggesting this area is involved in emotional processing in MDD [63]. The entorhinal cortex also showed increased TSPO availability after RSD. However, only few studies demonstrate participation of this area in MDD, in particular in cognitive processes [64,65].
This thesis also supports the idea that that the striatum (also part of the limbic system) is involved in depression, as dopaminergic transmission was affected by RSD and glucose metabolism was decreased after ovariectomy. This specific brain area is important for MDD since it is one of the major dopaminergic nuclei in the brain, projecting to several areas and modulating several higher brain functions, including emotion processing [66]. Other interesting findings include changes in biological markers in the temporal cortex and cerebellum. Although these brain regions do not seem to be major players in the pathophysiology of MDD, these areas have been suggested to play a role in emotional and cognitive processing [67–69]. Although this thesis does not provide a plausible link between neuroinflammation and monoamine neurotransmission, research in the field of MDD should
214
move in that direction to be able to create a time line of the changes that are occurring during the progression of the disorder.
Research on the neuroanatomical pathophysiology of depression has shown several brain structures which are impaired in both MDD patients and animal models of depression. Although most of the studies focused on areas related to emotional processing and cognition, studies have also proposed areas that may be involved in other aspects of depression, such as the basal ganglia and the brainstem, which participate in the reward system and regulating basic functions like feeding and sleeping behavior, respectively [70]. Current research in this field has shown the diversity of brain areas that are somehow involved in the development and progression of MDD. The different brain areas involved and the different interactions between biological systems highlight the complexity of MDD and emphasize the fact that we are far from understanding the mechanisms and brain connectivity that underlies this disorder. In this regard, an effort should be made to correlate the neuroanatomical networks with changes in and interactions between biological systems.
9. Reproducibility of results in preclinical research: influence of the experimental variables.
Preclinical and basic research in the biomedical field is the first step and possible foundation for further advances towards clinical research. In this regard, it is important to acknowledge that great advances in science were founded on preclinical research [71]. However, preclinical research is facing important challenges, such as the existence of different models for the same illnesses/diseases, variabilities in the follow-up period between species and differences in responses to different drugs and their concentration [72]. Possibly, the most critical challenge is the poor reproducibility of the majority of findings presented in high-impact journals [73,74]. Consequently, preclinical research can be poorly predictive for clinical outcomes. Therefore, more robust and reproducible preclinical research is essential to enable a better translation to the human condition. In this subsection, the reproducibility of our results in different studies using the same chronic stress model (RSD) will be discussed.
RSD is a social stress model widely used because of its high ethological validity [26]. The RSD model was used in this thesis because it was previously shown to induce biological and behavioral changes after 5 days of resident-intruder exposure [30]. In this thesis Chapters 4 to 7 used the same RSD protocol, making it possible to directly compare results.
Both Chapter 4 and Chapter 5 explored the changes in neuroinflammation one week after the RSD protocol. Although the obtained anhedonic behavior (SPT) was similar, the results were not consistent between the studies described in this thesis, as an RSD-induced increase in [11C]-PK11195 uptake was found in Chapter 4, but not in Chapter 5. The main explanation for this discrepancy lies in the differences in the protocols used. While in Chapter 4 an acute injection of ketamine/vehicle together with an extra SPT and OFT were performed between RSD and the PET scan, animals used in Chapter 5 did not have any intervention or exposure to the OFT. Therefore, the difference between the studies suggests that the stress caused by the injection of vehicle and the extra behavioral test may have cause additional stress and thus resulted in such an increase in neuroinflammation that it could be detected by [11C]-PK11195 PET. A previous study from our group [30] also found increased neuroinflammation with [11C]-PK11195 PET, but in more areas than in Chapter 4. In that study, the animals also underwent [11C]-PK11195 and [18F]-FDG PET scans at baseline (before RSD) and one day after RSD. These 4 additional PET scans may have resulted in sensitization to stress. Taken together, these data suggest that the 5-day RSD protocol by itself may not be able to induce sufficient neuroinflammation to be detected by [11C]-PK11195 PET, but the combination of RSD with the stress of additional procedures might tip the scale, making the activation of glial cells detectable.
A similar situation was found for Chapter 6 and Chapter 7, as the results obtained from the SPT two weeks after RSD were different between both studies. While depressive-like behavior was still present two weeks after RSD in Chapter 6, this was not the case in Chapter 7. It should be noted that the animals in Chapter 7 were submitted to three PET scans instead of one PET scan for animals in Chapter 6, so apparently the number of PET scans did not seem to have had a strong influence in the duration of depressive-like behavior in rats. Therefore, other differences in the protocol must have been responsible for this discrepancy between studies. One explanation could be the type of separation between the intruder and the resident after defeat in the RSD sessions (wire mesh cage versus Plexiglas wall). In the wire mesh case, the intruder is still experiencing close contact with the resident, whereas intruder can move a little away from the resident behind the Plexiglas wall, making this procedure less stressful. Another explanation for the differences between both studies could be the additional stress caused by drug administration. Animals used in Chapter 6 were daily administered with either caffeine or vehicle by oral gavage for two weeks, which may have caused the additional stress responsible for the longer duration of the depressive-like behavior. However, two weeks of intraperitoneal drug administration after RSD did not result
216
in long-lasting depressive behavior in a previous study [75], but this discrepancy between studies may be explained by the differences in drug administration routes or the effect of additional triggers.
Another important point of discussion is the data distribution and the statistical analyses. As seen in this thesis, the distribution of PET and behavioral data one day after RSD shows high variability. This suggests that individual differences in coping with stress could have affected the outcome. It also highlights the need to develop methods to differentiate between stress-susceptible and stress-resistant subpopulations within a study to improve the veracity of measured outcomes. Part of the high inter-subject variability could also be overcome by selecting a statistical analysis method that focusses on changes within subjects by including baseline measurements in studies with a longitudinal design. By doing this, translation to the clinical situation, in which some individuals are more susceptible to develop MDD than others, may improve.
All these findings together highlight the need to refine experimental approaches, ensuring that even the smallest experimental details are reproducible, and to be careful with the generalization of results, as they may vary depending of small variations in the protocols. 10. Final remarks – a considerable amount of research is still needed
Depressive disorders are disabling diseases that are likely to remain a major concern in the future in terms of public health and economic burden. Despite all research in this field, we are far from understanding these diseases sufficiently to have high remission rates. Moreover, modern day society is characterized by increasing social and work-related stress, together with an increasing general anxiety for the use of technology, which will probably lead to an increase in the incidence of depressive disorders, such as MDD. Thus, it is imperative to improve current treatment for these disorders in order to maintain patients quality of life.
Because the pathophysiology of MDD is highly heterogeneous, more clinical and preclinical studies are needed in order to elucidate how different biological systems interact in the brain. This knowledge is necessary to determine how risk factors may affect the development, progression and treatment response in patients. Because many biological systems are involved in MDD and genetics can differentially influence the response of subjects at-risk to environmental stimuli, a huge spectrum of possible outcomes can occur. This could explain why the treatment efficacy is low, and why we are still far from understanding the complete etiology and pathophysiology of these disorders. In this regard,
this thesis presented several biomarkers and interactions which can be useful to develop an improved vision of the pathophysiology of MDD.
A practical approach to investigate how biological systems interact in MDD could be to choose one biological system and see if pharmacological interventions in this system can influence other systems. Examples of this approach are given in Chapters 4 and 6, in which the interaction between interventions targeting the adenosine or glutamate system and neuroinflammation was the main focus. Measuring neuroinflammation with PET could be a useful approach to study how pharmacological interventions interact with neuroinflammation. For example, PET could help determine whether successful treatment with antidepressants (i.e.imipramine or fluoxetine) induce anti-inflammatory effects in specific brain areas associated to MDD [76]. Although TSPO PET measurements present some limitations, these can be compensated for by additional measurements of biomarkers, such as pro-inflammatory and oxidative stress molecules, by post-mortem analysis or preclinical models. In addition, being a translational technique, PET can be directly implemented from preclinical results into the clinic. Moreover, PET allows studying changes in biomarkers in a longitudinal manner, it could be used to establish time patterns of molecular changes at different stages of depression.
It is also important to discuss the benefits and limitations of the knowledge obtained from preclinical models. The main advantages of using animal studies lies in the ability to control environmental factors, the possibility of using pharmacological interventions, and the possibility of studying biological markers in simplified in-vivo and ex-vivo models. Nevertheless, besides the possible biological and physiological differences found between species, simplified models of diseases decrease the translational power of these studies to the human situation, and it does not reflect the time progression present in MDD patients. Moreover, results obtained from depression models should be assessed with care, since depressive-like behavior is only transient, and they cannot be diagnosed as depressed in the same way as patients are diagnosed. However, the transient behavioral effect observed in animal models can also be an advantage, since it may allow investigating differences in biomarkers between animals with and without depressive-like behavior, as shown in Chapter 7.
To conclude, the use of preclinical models in the research of biological interactions in depression is useful, but the translation of the results to humans should be assessed with care. The results obtained in animals could eventually be useful, as they could provide biomarkers to stratify different types of depression, and advance to a more personalized treatment. However, it would be recommended to always test a hypothesis using different animal
218
models, in order to establish the robustness or the limitations of preclinical studies. Therefore, results obtained from preclinical studies must be analyzed with care, as they only represent a small part of the picture of the real pathophysiological condition in humans. Since MDD shows a wide variability in symptoms and treatment response, the view of MDD as a single psychiatric disorder may need to be reconsidered. It may be possible that instead of a specific disorder, MDD is a spectrum of multiple disorders with similar characteristics, but different underlying mechanisms.
References
1. Albert KM, Newhouse PA. Estrogen, Stress, and Depression: Cognitive and Biological Interactions. Annu Rev Clin Psychol. 2019;15:399–423.
2. Moresco RM, Casati R, Lucignani G, Carpinelli A, Schmidt K, Todde S, et al. Systemic and Cerebral Kinetics of 16α[ 18 F]Fluoro-17β-Estradiol: A Ligand for the in vivo Assessment of Estrogen Receptor Binding Parameters. J Cereb Blood Flow Metab. 1995;15:301–11.
3. Moraga-Amaro R, van Waarde A, Doorduin J, de Vries EFJ. Sex steroid hormones and brain function: PET imaging as a tool for research. J Neuroendocrinol. 2018;30.
4. Khayum MA, de Vries EFJ, Glaudemans AWJM, Dierckx RAJO, Doorduin J. In Vivo Imaging of Brain Estrogen Receptors in Rats: A 16 -18F-Fluoro-17 -Estradiol PET Study. J Nucl Med. 2014;55:481–7.
5. Reiman EM, Armstrong SM, Matt KS, Mattox JH. The application of positron emission tomography to the study of the normal menstrual cycle. Hum Reprod. 1996;11:2799–805.
6. Eberling JL, Wu C, Tong-Turnbeaugh R, Jagust WJ. Estrogen- and tamoxifen-associated effects on brain structure and function. Neuroimage. 2004;21:364–71.
7. Ottowitz WE, Siedlecki KL, Lindquist MA, Dougherty DD, Fischman AJ, Hall JE. Evaluation of prefrontal– hippocampal effective connectivity following 24 hours of estrogen infusion: An FDG-PET study. Psychoneuroendocrinology. 2008;33:1419–25.
8. Silverman DHS, Geist CL, Kenna HA, Williams K, Wroolie T, Powers B, et al. Differences in regional brain metabolism associated with specific formulations of hormone therapy in postmenopausal women at risk for AD. Psychoneuroendocrinology. 2011;36:502–13.
9. Miller KK, Deckersbach T, Rauch SL, Fischman AJ, Grieco KA, Herzog DB, et al. Testosterone administration attenuates regional brain hypometabolism in women with anorexia nervosa. Psychiatry Res Neuroimaging. 2004;132:197–207.
10. Moser U, Wadsak W, Spindelegger C, Mitterhauser M, Mien L-K, Bieglmayer C, et al. Hypothalamic serotonin-1A receptor binding measured by PET predicts the plasma level of dehydroepiandrosterone sulfate in healthy women. Neurosci Lett. Elsevier Ireland Ltd; 2010;476:161–5.
11. Stein P, Baldinger P, Kaufmann U, Christina R-M, Hahn A, Höflich A, et al. Relation of progesterone and DHEAS serum levels to 5-HT1A receptor binding potential in pre- and postmenopausal women. Psychoneuroendocrinology. 2014;46:52–63.
12. Kranz GS, Rami-Mark C, Kaufmann U, Baldinger P, Hahn A, Höflich A, et al. Effects of hormone replacement therapy on cerebral serotonin-1A receptor binding in postmenopausal women examined with [carbonyl-11C]WAY-100635. Psychoneuroendocrinology. 2014;45:1–10.
13. Moses EL, Drevets WC, Smith G, Mathis C a, Kalro BN, Butters M a, et al. Effects of estradiol and progesterone administration on human serotonin 2A receptor binding: a PET study. Biol Psychiatry. 2000;48:854–60.
14. Perfalk E, Cunha-Bang S da, Holst KK, Keller S, Svarer C, Knudsen GM, et al. Testosterone levels in healthy men correlate negatively with serotonin 4 receptor binding. Psychoneuroendocrinology. Elsevier Ltd; 2017;81:22–8.
15. Frokjaer VG, Pinborg A, Holst KK, Overgaard A, Henningsson S, Heede M, et al. Role of Serotonin Transporter Changes in Depressive Responses to Sex-Steroid Hormone Manipulation: A Positron Emission Tomography Study. Biol Psychiatry. Elsevier; 2015;78:534–43.
16. Jovanovic H, Kocoska-Maras L, Rådestad AF, Halldin C, Borg J, Hirschberg AL, et al. Effects of estrogen and testosterone treatment on serotonin transporter binding in the brain of surgically postmenopausal women – a PET study. Neuroimage. Elsevier Inc.; 2015;106:47–54.
220
17. Nordström AL, Olsson H, Halldin C. A PET study of D2 dopamine receptor density at different phases of the menstrual cycle. Psychiatry Res. 1998;83:1–6.
18. Bentley SM, Pagalilauan GL, Simpson SA. Major Depression. Med Clin North Am. 2014;98:981–1005. 19. Schmidt PJ, Ben Dor R, Martinez PE, Guerrieri GM, Harsh VL, Thompson K, et al. Effects of estradiol withdrawal on mood in women with past perimenopausal depression: A randomized clinical trial. JAMA Psychiatry. 2015;72:714–26.
20. Whooley MA, Grady D, Cauley JA, M.A. W, D. G, J.A. C, et al. Postmenopausal estrogen therapy and depressive symptoms in older women. J Gen Intern Med. 2000;15:535–41.
21. Soares CN. Mood disorders in midlife women: understanding the critical window and its clinical implications. Menopause. 2014;21:198–206.
22. Xu J, Yao M, Ye J, Wang G, Wang J, Cui X, et al. Bone mass improved effect of icariin for postmenopausal osteoporosis in ovariectomy-induced rats. Menopause. 2016;23:1152–7.
23. Bailey ME, Wang ACJ, Hao J, Janssen WGM, Hara Y, Dumitriu D, et al. Interactive effects of age and estrogen on cortical neurons: implications for cognitive aging. Neuroscience. 2011;191:148–58.
24. Gogos A, McCarthy M, Walker AJ, Udawela M, Gibbons A, Dean B, et al. Differential effects of chronic 17β-oestradiol treatment on rat behaviours relevant to depression. J Neuroendocrinol. 2018;30:e12652.
25. Cohen JL, Ata AE, Jackson NL, Rahn EJ, Ramaker RC, Cooper S, et al. Differential stress induced c-Fos expression and identification of region-specific miRNA-mRNA networks in the dorsal raphe and amygdala of high-responder/low-responder rats. Behav Brain Res. 2017;319:110–23.
26. Czéh B, Fuchs E, Wiborg O, Simon M. Animal models of major depression and their clinical implications. Prog Neuro-Psychopharmacology Biol Psychiatry. Elsevier Inc.; 2016;64:293–310.
27. DeWilde KE, Levitch CF, Murrough JW, Mathew SJ, Iosifescu D V. The promise of ketamine for treatment-resistant depression: current evidence and future directions. Ann N Y Acad Sci. 2015;1345:47–58.
28. Grosso G, Micek A, Castellano S, Pajak A, Galvano F. Coffee, tea, caffeine and risk of depression: A systematic review and dose-response meta-analysis of observational studies. Mol Nutr Food Res. 2016;60:223– 34.
29. Wadhwa M, Chauhan G, Roy K, Sahu S, Deep S, Jain V, et al. Caffeine and Modafinil Ameliorate the Neuroinflammation and Anxious Behavior in Rats during Sleep Deprivation by Inhibiting the Microglia Activation. Front Cell Neurosci. 2018;12:1–16.
30. Kopschina Feltes P, de Vries EF, Juarez-Orozco LE, Kurtys E, Dierckx RA, Moriguchi-Jeckel CM, et al. Repeated social defeat induces transient glial activation and brain hypometabolism: A positron emission tomography imaging study. J Cereb Blood Flow Metab. 2019;39:439–53.
31. Schwartz M, Baruch K. The resolution of neuroinflammation in neurodegeneration: leukocyte recruitment via the choroid plexus. EMBO J. 2014;33:7–22.
32. van der Doef TF, Doorduin J, van Berckel BNM, Cervenka S. Assessing brain immune activation in psychiatric disorders: clinical and preclinical PET imaging studies of the 18-kDa translocator protein. Clin Transl Imaging. Springer Milan; 2015;3:449–60.
33. Alam MM, Lee J, Lee S-Y. Recent Progress in the Development of TSPO PET Ligands for Neuroinflammation Imaging in Neurological Diseases. Nucl Med Mol Imaging (2010). Nuclear Medicine and Molecular Imaging; 2017;51:283–96.
34. Rojas C, Stathis M, Coughlin JM, Pomper M, Slusher BS. The Low-Affinity Binding of Second Generation Radiotracers Targeting TSPO is Associated with a Unique Allosteric Binding Site. J Neuroimmune Pharmacol. 2018;13:1–5.
35. Caraci F, Calabrese F, Molteni R, Bartova L, Dold M, Leggio GM, et al. International Union of Basic and Clinical Pharmacology CIV: The Neurobiology of Treatment-resistant Depression: From Antidepressant
Classifications to Novel Pharmacological Targets. Ohlstein EH, editor. Pharmacol Rev. 2018;70:475–504. 36. Firk C, Markus CR. Review: Serotonin by stress interaction: a susceptibility factor for the development of depression? J Psychopharmacol. 2007;21:538–44.
37. Sinclair D, Purves-Tyson TD, Allen KM, Weickert CS. Impacts of stress and sex hormones on dopamine neurotransmission in the adolescent brain. Psychopharmacology (Berl). 2014;231:1581–99.
38. Ahnaou A, Drinkenburg WHIM. Simultaneous Changes in Sleep, qEEG, Physiology, Behaviour and Neurochemistry in Rats Exposed to Repeated Social Defeat Stress. Neuropsychobiology. 2016;73:209–23. 39. Elhwuegi AS. Central monoamines and their role in major depression. Prog Neuro-Psychopharmacology Biol Psychiatry. 2004;28:435–51.
40. Avgustinovich DF, Alekseenko O V, Bakshtanovskaia I V, Koriakina LA, Lipina T V, Tenditnik M V, et al. [Dynamic changes of brain serotonergic and dopaminergic activities during development of anxious depression: experimental study]. Usp Fiziol Nauk. Russia (Federation); 2004;35:19–40.
41. Susser LC, Sansone SA, Hermann AD. Selective serotonin reuptake inhibitors for depression in pregnancy. Am J Obstet Gynecol. Elsevier Inc.; 2016;215:722–30.
42. Ansorge MS, Morelli E, Gingrich JA. Inhibition of Serotonin But Not Norepinephrine Transport during Development Produces Delayed, Persistent Perturbations of Emotional Behaviors in Mice. J Neurosci. 2008;28:199–207.
43. Noorlander CW, Ververs FFT, Nikkels PGJ, van Echteld CJA, Visser GHA, Smidt MP. Modulation of Serotonin Transporter Function during Fetal Development Causes Dilated Heart Cardiomyopathy and Lifelong Behavioral Abnormalities. Albertson DN, editor. PLoS One. 2008;3:e2782.
44. Olivier JDA, Vallès A, van Heesch F, Afrasiab-Middelman A, Roelofs JJPM, Jonkers M, et al. Fluoxetine administration to pregnant rats increases anxiety-related behavior in the offspring. Psychopharmacology (Berl). 2011;217:419–32.
45. Cabrera-Vera TM, Garcia F, Pinto W, Battaglia G. Effect of prenatal fluoxetine (prozac) exposure on brain serotonin neurons in prepubescent and adult male rat offspring. J Pharmacol Exp Ther. 1997;280:138–45. 46. Yu W, Yen Y-C, Lee Y-H, Tan S, Xiao Y, Lokman H, et al. Prenatal selective serotonin reuptake inhibitor (SSRI) exposure induces working memory and social recognition deficits by disrupting inhibitory synaptic networks in male mice. Mol Brain. Molecular Brain; 2019;12:29.
47. De Ceballos ML, Benedi A, Urdin C, Del Rio J. Prenatal exposure of rats to antidepressant drugs down-regulates beta-adrenoceptors and 5-HT2 receptors in cerebral cortex. Neuropharmacology. 1985;24:947–52. 48. Sachser RM, Haubrich J, Lunardi PS, de Oliveira Alvares L. Forgetting of what was once learned: Exploring the role of postsynaptic ionotropic glutamate receptors on memory formation, maintenance, and decay. Neuropharmacology. 2017;112:94–103.
49. Glover ME, Clinton SM. Of rodents and humans: A comparative review of the neurobehavioral effects of early life SSRI exposure in preclinical and clinical research. Int J Dev Neurosci. 2016;51:50–72.
50. Ghasemi M, Phillips C, Fahimi A, McNerney MW, Salehi A. Mechanisms of action and clinical efficacy of NMDA receptor modulators in mood disorders. Neurosci Biobehav Rev. Elsevier; 2017;80:555–72.
51. Cohen R. Evidence for common alterations in cerebral glucose metabolism in major affective disorders and schizophrenia. Neuropsychopharmacology. England; 1989;2:241–54.
52. Drevets WC, Price JL, Simpson JR, Todd RD, Reich T, Vannier M, et al. Subgenual prefrontal cortex abnormalities in mood disorders. Nature. 1997;386:824–7.
53. Elkis H, Friedman L, Buckley PF, Lee HS, Lys C, Kaufman B, et al. Increased prefrontal sulcal prominence in relatively young patients with unipolar major depression. Psychiatry Res Neuroimaging. Ireland; 1996;67:123–34.
222
54. Dong X-Z, Li Z-L, Zheng X-L, Mu L-H, Zhang G, Liu P. A representative prescription for emotional disease, Ding-Zhi-Xiao-Wan restores 5-HT system deficit through interfering the synthesis and transshipment in chronic mild stress-induced depressive rats. J Ethnopharmacol. Elsevier; 2013;150:1053–61.
55. Couch Y, Anthony DC, Dolgov O, Revischin A, Festoff B, Santos AI, et al. Microglial activation, increased TNF and SERT expression in the prefrontal cortex define stress-altered behaviour in mice susceptible to anhedonia. Brain Behav Immun. 2013;29:136–46.
56. Garro-Martínez E, Vidal R, Adell A, Díaz Á, Castro E, Amigó J, et al. β-Catenin Role in the Vulnerability/Resilience to Stress-Related Disorders Is Associated to Changes in the Serotonergic System. Mol Neurobiol. 2020;57:1704–15.
57. Park C, Rosenblat JD, Lee Y, Pan Z, Cao B, Iacobucci M, et al. The neural systems of emotion regulation and abnormalities in major depressive disorder. Behav Brain Res. Elsevier; 2019;367:181–8.
58. Wilson MA, Grillo CA, Fadel JR, Reagan LP. Stress as a one-armed bandit: Differential effects of stress paradigms on the morphology, neurochemistry and behavior in the rodent amygdala. Neurobiol Stress. Elsevier Ltd; 2015;1:195–208.
59. Gobinath AR, Mahmoud R, Galea LAM. Influence of sex and stress exposure across the lifespan on endophenotypes of depression: focus on behavior, glucocorticoids, and hippocampus. Front Neurosci. 2015;8:272–88.
60. Cernackova A, Durackova Z, Trebaticka J, Mravec B. Neuroinflammation and depressive disorder: The role of the hypothalamus. J Clin Neurosci. Elsevier Ltd; 2020;75:5–10.
61. Ehlert U, Gaab J, Heinrichs M. Psychoneuroendocrinological contributions to the etiology of depression, posttraumatic stress disorder, and stress-related bodily disorders: the role of the hypothalamus–pituitary–adrenal axis. Biol Psychol. 2001;57:141–52.
62. Hsu DT, Kirouac GJ, Zubieta J-K, Bhatnagar S. Contributions of the paraventricular thalamic nucleus in the regulation of stress, motivation, and mood. Front Behav Neurosci. 2014;8:272–88.
63. Mutschler I, Ball T, Wankerl J, Strigo IA. Pain and emotion in the insular cortex: evidence for functional reorganization in major depression. Neurosci Lett. Elsevier Ireland Ltd; 2012;520:204–9.
64. Kanner AM. Is major depression a neurologic disorder with psychiatric symptoms? Epilepsy Behav. 2004;5:636–44.
65. Östlund H, Keller E, Hurd YL. Estrogen Receptor Gene Expression in Relation to Neuropsychiatric Disorders. Ann N Y Acad Sci. 2003;1007:54–63.
66. Grace AA. Dysregulation of the dopamine system in the pathophysiology of schizophrenia and depression. Nat Rev Neurosci. 2016;17:524–32.
67. Zhang F-F, Peng W, Sweeney JA, Jia Z-Y, Gong Q-Y. Brain structure alterations in depression: Psychoradiological evidence. CNS Neurosci Ther. 2018;24:994–1003.
68. Mayberg HS. Frontal lobe dysfunction in secondary depression. J Neuropsychiatry Clin Neurosci. 1994;6:428–42.
69. Minichino A, Bersani FS, Trabucchi G, Albano G, Primavera M, Chiaie RD, et al. The role of cerebellum in unipolar and bipolar depression: A review of the main neurobiological findings. Neurosci Biobehav Rev. 2018;84:272–88.
70. Pandya M, Altinay M, Malone DA, Anand A. Where in the Brain Is Depression? Curr Psychiatry Rep. 2012;14:634–42.
71. Varga OE, Hansen AK, Sandøe P, Olsson IAS. Validating Animal Models for Preclinical Research: A Scientific and Ethical Discussion. Altern to Lab Anim. 2010;38:245–8.
72. Pound P, Ebrahim S, Sandercock P, Bracken MB, Roberts I. Where is the evidence that animal research benefits humans? BMJ. 2004;328:514–7.
73. Prinz F, Schlange T, Asadullah K. Believe it or not: how much can we rely on published data on potential drug targets? Nat Rev Drug Discov. Nature Publishing Group; 2011;10:712–712.
74. Peers IS, Ceuppens PR, Harbron C. In search of preclinical robustness. Nat Rev Drug Discov. Nature Publishing Group; 2012;11:733–4.
75. Giacobbo BL, Doorduin J, Moraga-Amaro R, Nazario LR, Schildt A, Bromberg E, et al. Chronic harmine treatment has a delayed effect on mobility in control and socially defeated rats. Psychopharmacology (Berl). Psychopharmacology; 2020;237:1595–606.
76. Alboni S, Benatti C, Montanari C, Tascedda F, Brunello N. Chronic antidepressant treatments resulted in altered expression of genes involved in inflammation in the rat hypothalamus. Eur J Pharmacol. Elsevier; 2013;721:158–67.
224
