Very good question you raised! Though many subjects achieve remission without any particular antidepressant treatment, it does not necessarily indicate that the neurobiological changes associated with a depressive episode resolve. We need to bear in mind that patients with a previous depressive episode has a higher risk of a new episode in the future and the risk exponentially increases after additional depressive episodes. This is probably due to the persistence of mild neurobiological changes in these subjects. This is in line with several hypotheses regarding the neurobiology of affective disorders such Kindling and Sensitization (Robert Post), Neuroprogression (Flavio Kapczinski), Scars in Depression (Wichers, M.).

Therefore, I think that we need novel conceptual approaches to tackle this question. One approach may be the distinction from "symptomatic remission" and the "resolution of neurobiological changes in depression". Possibly, there is a distinct temporal pattern between symptomatic remission and the resolution of neurobiological changes (the latter taking longer to occur). As current studies focus solely on the "symptomatic remission" and assume that neurobiological changes simultaneously resume with symptomatic remission, data on the neurobiology of remission may be biased and leading to conflicting results.

Do you have descriptive data of cases of remission? Are you including psychosocial factors as well?

Yes, Peter makes a good point. It would seem you need a baseline of sorts, Simon, probably consisting of descriptive data of cases of spontaneous remission. Also, I am concerned that we haven't defined depression yet. There are many definitions of it available, and unless it is defined (and measured) at baseline and followup with a high degreee of precision (characteristics of descriptive cases will help), then results of this exploration may not be generizable from one study cohort to another.

dear colleague, you are raising a very important point; there has to be a neurobiology of recovery that most likely is intertwined with the neurobiology of resiliency and of individual variations in stress tolerability.

From a practical point of view this question could probably be best addressed in animal models, at least in the first place. Rodents could be exposed to repeated stressors or other models of depression (such as hopelessness). This has been done many times; what has not been done is NOT to sac the animals at the end of the exposure but to follow them for a while with behavioral and endocrine end-points of "recovery". That should at least give some targets for human studies in terms of which systems recover first (if they do) and at which pace.

Yes Giovanni; also to study the animals most resilient, most resistant to developing stress/depression in the first place, and see in what ways they differ from more vulnerable animals. Pavlov I believe noticed that some animals were almost impossible to drive crazy.

However: although this is a good starting place, in humans there may be cognitive strategies in stress recovery; if rats use such strategies they have not yet told us about it. My own bias is that depression should be looked at both from 'mind' and 'body'. I strongly believe from my own clinical experiences that the capacity to recover from depression can be strongly influenced by various learnt voluntary strategies,

Gosh! Thank you again Peter... when the rats tell us about their strategies for stress recovery, can you ask them to share their secrets with me :)

Guys,

even I am fully aware than rats are not humans and that there are limits to animal models of depression. BUT, the question started citing the BDNF system in the brain of humans; unless one does post-mortem studies (which are done in post-mortem often from sucide samples). there is no non-invasive ways to get to the BDNF of "recovered depressed".

Evolutionary mechanisms of resiliency are pretty much preserved and repetitive throughout phylogenesis, (NOT identical but similar enough that animal models may point us (with relatively low cost and quite fast) in the right direction for human studies. Yes it is true that humans via the neocortex may have additional cognitive strategies that do not exist in rodents.

Yes Giovanni, sorry, I was not disagreeing with you.

Thank you all for your valuable inputs!

The neurobiological/molecular mechanisms in depression in many ways ressemble a vicious circle.

That is, stress (cortisol) leads to inflammation (and vice versa), leading to an activation of the kynurenine pathway and glutamate excitotoxicty/NMDA receptor stimulation. So, a sort of neurodegenerative mechanism. Further, cortisol and inflammation reduce BDNF thus hampering hippocampal neurogenesis. The pathological changes involved here are seems out of control, that is, it seems like there's no real negative feedback systems working and the pathological changes just become more and more pronounced. Eg. the more hippocampal dysfunction we see the less it can regulate the HPA axis, leading to further cortisol build-up and more harm to the hippocampal neurogenesis.

But something has to happen, since a significant proportion recover without treatment. My question was more about which mechanisms could work as a "brake" and reverse the changes above in order to mediate recovery. At least hypothetically.

Now, if the stressor (internal/external) that lead to the depression disappears, that could be one potential mechanism, but not all patients who recover had their depression due to a stressor, further underlining the role of kindling mechanisms. Second, when the whole system of molecular changes associated with depression is "set off", which mechanisms provide negative feedback?

The stress response reminds me in a way of the mechanisms that control appetite in the sense that in both cases there are stimulatory and inhibitory mechanisms but they are not equally developed.

The response is non-symmetrical:

A) in the case of the stress response that are of course mechanisms that initiate it; they are also mechanisms that stop it. For example the cytokine cascade, which is part of the stress response is initiated by pro-inflammatory cytokines but then is followed in time by anti-inflammatory cytokines that have the function to stop it before it becomes maladaptive.

B) Similarly we have hormone that stimulate appetite and hormones that stop appetite, but the first are more powerful than the second otherwise people would never become obese.

In both cases the evolutionary ratio is obvious: evolution favors and preserves mechanisms that allow the survival of the individual in the shorter-term (i.e. until the individual is able to reproduce and raise the offspring) so that the species is preserved in the long term.

Según sus comentarios pareciera que la neurología de la depresión sería casi la misma que de otros trastornos emocionales y la demencia.Se podría hablar de un continium patológico.Qué sabéis de la neurobiología de la resiliencia?.Gracias y discúlpenme la pregunta me surge de sus importantes comentarios

Well, at the risk of making a fool of myself, let me add another 2 cents. Disclaimer: this is not my area of specialty; I only know what the rats told me. ; )

2 points:

First, as to the neurochemical mechanisms for negative feedback. Endorphins? Cannabinoids? I would imagine either of these would have inhibitory effects on the HPA axis and cortisol secretion.

Also: referencing Porges' polyvagal theory--a shift from dorsovagal or sympathetic activation to ventrovagal, involving increased social interaction, would also have a significant stress-reducing effect. Oxytocin is released under these circumstances. I know the above mechanism is very much involved in human recovery from trauma.

Second, with respect to humans: role of cortical mechanisms, especially rhose areas acting as super ordinate control of ANS: orbitofrontomedial, anterior cingulate?

Possible inhibition of amygdala?

A related question: do rats subject to stress-induced depression recover spontaneously in approximately the same percentages as humans? This could be suggestive as to where one might look for the ways of encouraging recovery in humans. Is recovery (self-regulation) largely a lower brain process or is this capacity enhanced by cortical functions?

Apologies if my comments are naive.

Peter

My thought about Cannabinoids has proven fruitful. here is an excerpt from an article at

http://www.maedica.org/articole/nr3_2007/236-250_About.pdf

"Recent studies have shown that the endocannabinoid signaling system, expressed in neural progenitors, participates in the control of progenitor cell proliferation and differentiation (29) in the normal brain. Endocannabinoids are produced by neural progenitors leading to intracellular calcium increase, through CB1 receptor activation and triggering of hippocampus neural progenitors proliferation and neurogenesis (29). Thus, the CB1 cannabinoid receptor plays a neuroprotective role against brain excitotoxicity and stroke and in different neurodegenerative (29). In this context, chronic pharmacological administration of the synthetic cannabinoid induces neurogenesis. Dexanabinol is a synthetic cannabinoid analogue devoid of psychotropic activity, with strong neuroprotective potential. The drug differs from other neuroprotective drugs because it targets various pathophysiological mechanisms, including glutamate excitotoxicity, free-radical damage, and inflam- matory response. Dexanabinol was shown to be highly neuroprotective in an animal model of traumatic brain injury. Overall, the capacity of cannabinergic drugs to modulate neurogenesis after brain damage may open new approaches to improve the insufficient neuro- genic response of endogenous neural pro- genitors caused by excitotoxicity or ischemia mechanisms."

Suggestive, no?

Peter

Very good question you raised! Though many subjects achieve remission without any particular antidepressant treatment, it does not necessarily indicate that the neurobiological changes associated with a depressive episode resolve. We need to bear in mind that patients with a previous depressive episode has a higher risk of a new episode in the future and the risk exponentially increases after additional depressive episodes. This is probably due to the persistence of mild neurobiological changes in these subjects. This is in line with several hypotheses regarding the neurobiology of affective disorders such Kindling and Sensitization (Robert Post), Neuroprogression (Flavio Kapczinski), Scars in Depression (Wichers, M.).

Therefore, I think that we need novel conceptual approaches to tackle this question. One approach may be the distinction from "symptomatic remission" and the "resolution of neurobiological changes in depression". Possibly, there is a distinct temporal pattern between symptomatic remission and the resolution of neurobiological changes (the latter taking longer to occur). As current studies focus solely on the "symptomatic remission" and assume that neurobiological changes simultaneously resume with symptomatic remission, data on the neurobiology of remission may be biased and leading to conflicting results.

Thank you Breno your comment was very well done and raises an interesting question. Are the neurobiological changes you mention present in symptomatic remission? What I mean to say is, how do we differentiate between symptomatic remission and long-term resolution of depressive neurobiological changes? Is there something about symptomatic remission that signals the likelihood of a repeat depressive episode? If there is something like this, we might be able to predict repeat episodes, perhaps not necessarily at a specific time. And are these repeat episodes of depression separated by periods of something like mania? I'm wondering what Post, Kapczinski, and Wichers say of this.

So conceptually, the depressive disease proces pushed a homeostatic equilibrium in a pathological direction, and as G. Cizza pointed out in the case of inflammation, anti-inflammatory cytokines counteract this disequilibrium. The same holds true for endocannabinoids. Perhaps, kindling/scaring/neuroprogression is driven by this equilibrium not returning completely to its initial state? A bit like Le Chatelier's principle in chemistry.

Then the interesting question arises why some remit spontaneously while other don't - and here perhaps parallels can be drawn to resilience research?

Simon, you raised a very good point. Your comment is in line with the hypothesis of allostasis and allostatic load. The body adapts (in several different ways) to challenges to "keep going" (allostasis); nonetheless, the persistence of of these adaptative changes over time may become maladaptive (e.g. persistence of elevated pro-inflammatory status), leading to negative outcomes.

Take a look at the interesting works form Bruce S. McEwen.

I think we may be deluding ourselves if we think we understand why an attack of depression starts at any particular time, let alone why one stops. Many start suddenly out of the blue for no apparent reason, though clearly there is an underlying disposition.

I agree with the learned response of Breno. In doing research on Treatment Resistant Depression, I assessed the outcomes of over 56 studies and spontaneous remission needs to be viewed with great caution. Remission is defined as either full or partial based on the dissipation of symptomologies that are defined as somatic, cognitive, affective and so forth, as well as the instruments used to measure symptomologies. Dependent on the type of Depression and the accurate diagnosis of sub-type this can affect the determination of remission. In Morphology studies it has often been found that for instance, there is a 19% reduction in left hippocampal volume and begs the question as to what the physiological cut off point is for recovery from Depression. Symptomologies have not, as far as I am aware been causally related to physiological events during recovery from Depression, but the correlational studies indicate lowering of activity, increase in heart rate variability lowering of cortisol levels etc, in the HPA Axis with an increase in activity in the left prefrontal lobe, specifically in those areas in which cognitive control is purported to occur, which has also been shown to reduce activity in the HPA Axis. However, spontaneous remission seems to me to be a mystery due to the fact that not only do physiological activities need to have returned to baseline pre-incident level, but so would cognitive processes. Research definitely points to residual physiological and cognitive vulnerabilities that would then exist after an episode and hence resource allocation, kindling and allostatic theoretical premises provide insight into the mechanisms behind the recurrence of Depression. A return to baseline before episode would theoretically mean that cognitive processes which occurred during the epsiode, would not in general have altered neural pathways, or reactions to events, whilst physiological processes would have been only been temporarily altered by the episode, which from the studies I have spent time going through appears to not be possible, unless an individual was incorrectly diagnosed, or symptomologies had not been clearly separated from the underlying pathophysiological and psychological processes underlying this debilitating disorder. I would treat spontaneous remission with great caution and with the assessment of multi-dimensional variables, from physiological, to psychological, to cognitive, to genetic predisposition, to organic medical causes, to symptomology to determine the existence of spontaneous remission.