Stress synapses and psychopathology

Maurizio Popoli is professor of Pharmacology at the Department of Pharmacological and Biomolecular Sciences, University of Milano, Italy.

In the last several years, the role of stress in neuropsychiatric disorders has been a central object of investigation in his research group. Stress-related neuropsychiatric disorders, such as depression, anxiety, and post-traumatic stress disorder (PTSD), represent one of the great therapeutic challenges for the 21st century, which announces itself as a phase of great social, economic and geopolitical changes. The stress response is a physiological reaction to environmental changes that can be positive and proadaptive (in most cases) or negative and maladaptive. Everyone reacts differently: for some people, stress is an incentive to do better, and may enhance one’s cognitive capability.

Others, due to both different genetic characteristics and personal story, may have trouble reaching the necessary adaptation to environmental changes. This negative response can become more frequent when an individual is exposed to particularly strong (as is often so in PTSD), or repeated stressful stimuli. In this case we speak of maladaptive responses, which can favour the onset of diseases, mental but also cardiovascular, metabolic and of different types.

You mention in recent literature the time-dependent nature of the changes in synaptic function and brain architecture that occur as a result of stress¹. Could you describe these changes?
It has been shown by several studies that the neuronal architecture of select brain areas (hippocampus, prefrontal cortex) is altered in humans affected by some mental disorders, such as depression and PTSD. The volume of these areas is smaller in patients, and in a number of studies also reduced density of dendrites and synaptic spines has been found in the same areas. A large number of studies with chronic stress models in rodents characterised the changes in equivalent brain areas of animals, finding reduced volume, reduced density of dendrites/spines, as well as reduced synaptic connectivity and behavioural changes. Human and rodent data taken together strongly suggest that these changes are the result of a maladaptive stress reaction, and that this is a main component of pathophysiology, characterised by changes and/or disconnection of key brain networks. In rodents, a few weeks of stress are enough to induce structural changes, which makes sense if one considers that one month of adult rat/mouse life roughly corresponds to three years of human life. However, so far most studies have just assessed the changes between stressed and control rodents; a growing body of literature is showing that, as for humans, rodents have different individual reactions which allow us to distinguish subjects with greater vulnerability to the effects of stress from more resilient individuals. Understanding what distinguishes a vulnerable (maladaptive) from a resilient (proadaptive) response is a major goal of research on the pathophysiology of stress-dependent disorders.

You also discuss the short-term and longer-term after-effects of acute stress. Could you take us through those, and what we know about the differences between acute and (what becomes) the chronic or prolonged response to stress?
Traditionally, animal models for neuropsychiatric disorders are based on repeated or chronic stress, although it is known that in some cases (e.g., PTSD) even a single trauma may be enough to trigger a pathology. There is at present no clear answer to the question: how does a single stressful event trigger a potentially life-long pathology? Most animal studies use chronic stress protocols and just look at the endpoint of numerous adaptations occurring during the brain and body stress response, somewhat overlooking the most important issue: how does the system get to the point where a physiological stress response turns into a maladaptive pathway that may favour psychopathology? For a number of reasons, it is relatively easier to address this question in acute stress protocols. However, only a handful of studies have investigated the effects of acute stress, or of a few closely spaced stressors, on neuronal architecture. Surprisingly, it turns out that acute stress may also induce atrophy/remodelling of dendrites, suggesting that even a single stressful event may cause rapid morphological changes in the brain.

We recently characterised the synaptic response to acute stress by using different methods, including the superfusion of purified synaptic terminals (synaptosomes). Acute stress rapidly enhances the release of glutamate and excitatory transmission in prefrontal cortex. This was measured soon after completion of the stress protocol, and it is mainly due to local, synaptic, action of corticosteroids (increased by stress). However, we were puzzled when we found that in our stressed rats we could observe atrophy/retraction of apical dendrites already after 24 hours, as observed by many studies after weeks of stress². This morphological change is also sustained, because it was measured for up to two weeks after the stress. We wondered whether the enhancement of glutamate release observed soon after acute stress could be prolonged for longer time periods and indeed found that enhanced glutamate release could be measured for up to 24 hours following stress.

What are the goals of your work? It was quite interesting to read about the ‘turning point’ that may distinguish vulnerable and resilient individuals, and the therapeutic target this might present.
Overall, the results obtained with acute stress protocols completely change our traditional distinction between the effects of acute versus chronic stress. It appears that a single exposure to stress may have long-term functional (glutamate release) and structural (dendrite atrophy) consequences. These sustained stress-related changes may be relevant for pathophysiology of stress-related disorders, including PTSD, thus suggesting a way whereby acute stress may affect the extended stress response. A complete dissection of the short- and long-term effects of acute stress is required to understand when and how a physiological brain and body response turns into pathology-related maladaptive changes.

First, it would be interesting to understand if the sustained excitatory activation induced by acute stress is a main reason for dendritic retraction, as suggested by earlier works. This may well be an adaptive measure taken by neuronal networks facing hyperactivation. Dendritic retraction/expansion may normally occur, as for instance in hibernating animals. However, what represents a proadaptive response in an overstimulated system might at some point, because the environmental stimulation is too strong or too long, become a maladaptive change, addressing the brain towards pathological states. We are currently exploring this problem.

Second, it would also be interesting to understand whether and how the stress response is differently regulated between resilient and vulnerable individuals. Assuming there is a continuum between physiological and pathological response, as in many biological phenomena, and that the difference is also a function of individual vulnerability, our studies should investigate the determinants of resilient versus vulnerable trajectory in the brain stress response. The identification of these determinants, through a thorough dissection of the stress response at the cellular/molecular level, could single out neuronal mechanisms and effectors that go wrong during the stress response, and allow the identification of novel targets for treatment. Therefore, the study of the long-term outcome of acute stress may allow the identification of critical ‘turning points’ in the stress response and shed light on pathophysiology of stress-related disorders.

Looking at your recent publications it was also curious to see your work in an animal model of ALS³. Could you describe this work?
These are studies we carry out in collaboration with colleagues at the University of Genoa. We found that abnormal glutamate release occurs in the spinal cord of SOD1G93A mice, a transgenic animal model for human ALS based on a mutation of superoxide dismutase. This excessive release of glutamate is exocytotic in nature, is linked to molecular changes in the presynaptic release apparatus and to the number of presynaptic vesicles available for release. Remarkably, this phenomenon is detectable as early as at postnatal day 30 in the ALS model, still in the pre-symptomatic disease stage, thus suggesting that the dysregulation of glutamate release represents a pivotal feature for pathology onset and development. Altered presynaptic mechanisms leading to the abnormal and excessive release of glutamate may represent targets for novel pharmacological approaches to reduce excitotoxicity in ALS. In this case there is no direct relationship with stress because this is a genetic model of ALS. However, although we still know little about this, undoubtedly stress and other environmental factors have a crucial role in pathophysiology of ALS, just as in other neurodegenerative disorders (e.g., Alzheimer, Parkinson). We used here the same method to measure glutamate release as we do in stress studies – that is, purified synaptosomes in superfusion.

What are the current projects you are working on, and where do you anticipate them leading in the future?
We have a number of projects running. A main project, as addressed before, deals with the search for determinants of resilience/vulnerability to acute and chronic stress. This is paralleled by the search for mechanism of rapid acting antidepressants (ketamine and other drugs acting on the glutamate system). A main aim here is to understand the nature of antidepressant mechanisms (not only drugs) beyond the classical monoamine-based hypothesis. Both in the action of stress and of drugs, we look at epigenetic mechanisms⁴, in particular posttranslational histone modifications and microRNAs. Many of these epigenetic targets are in principle druggable, and we hope to get some interesting and novel targets.

In addition, we study the role of the neurotrophic factor BDNF in brain pathophysiology, response to stress or physical exercise, and drug action. This is done also by using a mutant mouse carrying the BDNF Val66Met polymorphism⁵. Among other studies on this, we just completed a genome-wide epigenetic study of the mutant mice subjected to chronic psychosocial stress, an example of Gene X Environment study in a model of genetic vulnerability. The results of this study may supply more interesting putative therapeutic targets.

In 2016 Maurizio Popoli gave a lecture in a symposium on social stress and psychopathology (view the webcasts of the symposium).

In this video he talks about the ECNP Congress, but also about his daily work on the effects of stress.

References:
1. Musazzi L, Tornese P, Sala N, Popoli M. (2016) Acute stress is not acute: sustained enhancement of glutamate release after acute stress involves readily releasable pool size and synapsin I activation. Mol Psychiatry. [Epub ahead of print]
2. Nava N, Treccani G, Alabsi A, Kaastrup Mueller H, Elfving B, Popoli M, Wegener G, Nyengaard JR. (2015) Temporal Dynamics of Acute Stress-Induced Dendritic Remodeling in Medial Prefrontal Cortex and the Protective Effect of Desipramine. Cereb Cortex (Epub ahead of print).
3. Bonifacino T, Musazzi L, Milanese M, Seguini M, Marte A, Gallia E, Cattaneo L, Onofri F, Popoli M, Bonanno G. (2016) Altered mechanisms underlying the abnormal glutamate release in amyotrophic lateral sclerosis at a pre-symptomatic stage of the disease. Neurobiol Disease 95:122-133.
4. Rusconi F, Grillo B, Ponzoni L, Bassani S, Toffolo E, Paganini L, Mallei A, Braida D, Passafaro M, Popoli M, Sala M, Battaglioli E. (2016) LSD1 modulates stress-evoked transcription of immediate early genes and emotional behavior. Proc Natl Acad Sci USA 113:3651-3656.
5. Ieraci A, Madaio AI, Mallei A, Lee FS, Popoli M. (2016) Brain Derived Neurotrophic Factor Val66Met human polymorphism impairs the beneficial exercise-induced neurobiological changes in mice. Neuropsychopharmacol 41:3070-3079.