Proof that model organisms can “suffer” from psychiatric illness is at very best modest. However, the use of animal models gets around this either through symptom modeling or studying endophenotypes (see recent SRF endophenotype discussion and Gould and Gottesman, 2005). Thus, only facets, whether they be face valid “re-creations” of symptoms or models of inherent and quantifiable measures of brain functions, are utilized.
The recent paper by Svennignsson, Greengard, and colleagues takes advantage of these approaches to describe a novel function of p11, namely, the modulation of depression-like states. This includes increased tail suspension test (TST) immobility in mice where p11 has been removed (knockout; KO mice), and decreased TST immobility in mice that overexpress p11. Further, p11 KO mice spent more time along the “safer” sides of an open field, while mice overexpressing p11 tended to move away from the sides. These data are consistent with evidence that p11 is involved in modulating cellular pathways involved in depression-like and anxiety-like behavior. There are a number of additional behavioral tasks related to depression-like (e.g., forced swim test, learned helplessness) and anxiety-like (e.g., the elevated plus maze and black/white box) behavior, which could be studied in these mice. Additionally, the researchers used modern molecular and cellular biology, in addition to electrophysiological techniques, to strongly make the case that the behavioral changes likely involve interactions with the 5-HT1B receptor.
The authors “link” their basic science and rodent behavior data to humans by showing that both mRNA and protein levels of p11 are decreased in the postmortem brain of depressed patients compared to control subjects. Their finding that both ECT and imipramine increase p11 levels in the mouse brain suggests that similar effects could occur in humans. However, caution is warranted: The present field of psychiatric genetics was initiated with the understanding that human diseases are complex in nature—needing multiple genes working in disharmony with nongenetic contributors for the human syndrome. Thus, while a single gene disruption (e.g., p11) in the mouse may result in “human-related” psychiatric phenotypes, in humans, any involvement of p11 in the pathophysiology or treatment of mood disorders undoubtedly requires complex interactions with other susceptibility genes.
References:Gould TD, Gottesman II (in press): Psychiatric endophenotypes and the development of valid animal models. Genes, Brain, and Behavior. (full text courtesy of Genes, Brain, and Behavior)
View all comments by Todd Gould
The recent manuscript from Per Svenningsson’s laboratory at the Karolinska Institute in Stockholm, and Paul Greengard’s laboratory at the Rockefeller University in New York has identified a molecule—p11—as a putative mediator of depressive states, and as a target of antidepressant drug action.
The major strength of the study is the diverse array of methodologies/paradigms utilized to provide convergent data. Thus, they used a yeast two-hybrid system, transfected cells, selectively bred animals, and transgenic and knockout animals, and even human postmortem brain samples. They report that the p11 protein mediates 5-HT1B receptor surfacing, 5-HT1B receptor-induced inhibition of the cAMP pathway and the ERK pathway, fEPSP, and 5-HT turnover. Importantly, they show that mRNA and/or protein levels of p11 are different in groups of sham versus antidepressant treated animals, non-helpless versus helpless selectively bred animals, and control versus depression patients’ postmortem brain tissue. Furthermore, transgenic mice potentially overexpressing p11 proteins exhibit increased locomotion and reduced thigmotaxis in the open field test and reduced immobility in the tail suspension test. Finally, p11 knockout mice exhibit increased thigmotaxis in the open field test, increased immobility in the tail suspension test, and reduced consumption of sweetened water in the sucrose preference test.
The 11 kDa p11 protein, which is also known as S100A10, is a member of the S100 family of Ca2+-binding proteins expressed in the CNS. The function of p11 in the CNS has hitherto been largely unknown. Traditional antidepressants have long been known to exert their initial effects on increasing the synaptic levels of serotonin and/or norepinephrine. However, a major problem has been the delay in therapeutic effects (often several weeks) observed in patients. This has led to the suggestion that the delayed therapeutic effects may be due to delayed, adaptive changes in serotonergic receptors and post-receptor signaling cascades. Along with the 5-HT1A and 5-HT2A receptors, the 5-HT1B receptor has been postulated to represent a delayed, therapeutically relevant target for the actions of antidepressants. In this study, the authors focused upon the 5-HT1B receptor. In an open-ended search for the molecules directly interacting with 5-HT1B receptor using yeast two-hybrid screen, these investigators found that p11 selectively controls the number of 5-HT1B receptor at the cell surface; indeed, trafficking of important receptors to the cell surface appears to represent a major form of regulating various forms of neural plasticity. Concomitant with the effects on surface 5-HT1B receptors, p11 alters 5-HT1B receptor-mediated signaling processes, including inhibition of cAMP production and inactivation of the extracellular signal-regulated kinase (ERK) pathways. At a more “systems level,” they found that p11 attenuates corticoaccumbal glutamatergic synaptic transmission.
They then investigated the effects of manipulating the p11 protein in the context of depressive-like behaviors, and antidepressant models. While animal models of depression/antidepressant drug action do have some limitations, they can often provide useful information in the gene-to-human behavior causal chain. One such rodent model is the learned helplessness model, in which the rodents develop some of the behavioral/biochemical changes of depression, many of which are reversed by antidepressant treatment. Notably, the researchers found lower levels of p11 mRNA in forebrain of rodents that exhibited learned helplessness. Interestingly, the investigators found that chronic antidepressants (either chemical or electroconvulsive seizures) increased forebrain p11 mRNA and protein levels.
Svenningsson and colleagues then investigated the effects of genetically decreasing/increasing p11 levels in brain by generating knockout/transgenic mice. Mice lacking p11 exhibited increased thigmotaxis (the tendency to stay close to the walls; a putative measure of anxiety-like behavior) in the open field test, increased immobility in the tail suspension test (another test of depression-like behavior), and attenuated consumption of sweetened water (a putative test of hedonic drive). By contrast, transgenic mice overexpressing p11 in cortex, hippocampus, and striatum showed increased locomotion and reduced attenuated immobility in the tail suspension test.
A study of postmortem human brain tissue from depressed subjects also revealed a small, but statistically significant, decrease in p11 mRNA levels.
In toto, these novel, intriguing convergent data from multiple sources raise the distinct possibility that p11 may play a role in depressive behavior, and that directly targeting p11 may represent a novel strategy for the development of improved therapeutics. A number of issues still need to be addressed:
1. Interestingly, even the p11 knockout mice respond to imipramine and anpirtoline (a 5-HT1B receptor agonist) in the tail suspension test (albeit more modestly), suggesting the existence of the functional homologs of p11 for mediating behavioral effects of antidepressants.
2. A major issue in behavioral animal models of antidepressant action is a temporal one. Thus, while chronic (weeks) of antidepressant administration are required to produce changes in p11 levels, the drugs exert rapid effects in the animal models used here. This suggests that in the tail suspension test, p11 is unlikely to be mediating the effects of acute antidepressants. However, this is a problem faced by the field with respect to target validation, and hardly unique to this study.
3. The issue of how p11 is up-regulated by antidepressant treatments and down-regulated in depression at transcription level remains completely unknown.
4. Does p11 modulate other G protein-coupled receptors on the cell surface, especially those which have been potentially more implicated in the pathophysiology of depression (e.g., 5-HT1A or 5-HT2A)?
5. The 5-HT1B receptor has been shown to play a role in regulating hippocampal neurogenesis. Does p11 play a role in the hippocampal neurogenesis produced by antidepressants?
View all comments by Guang ChenView all comments by Husseini K. Manji
As the 5-HT system is involved in the pathology and pharmacological treatment of depression, any new evidence for genes or molecules regulating its function has putative implication for mechanisms and/or treatment of depression. Here, Svenningsson et al. provide compelling evidence about the identification and role of p11 in mediating some of the downstream effects of 5-HT1B receptor signaling. In particular, the authors, using several complementary approaches, demonstrate that p11 levels correlate with 5-HT1B receptor level and function at the membrane. Thus, p11 levels may be considered an “index” of 5-HT1B receptor function. Furthermore, manipulations that increase 5-HT function (i.e., chronic antidepressant treatment and ECT) up-regulate p11 levels, possibly responding to increased demand on 5-HT1B function to regulate presynaptic release. The fact that these up-and-down manipulations of p11 correlate with behaviors in the mouse that have been associated either with changes that occur after antidepressant treatment or in “depression-like” or increased fearfulness states strengthens the link between serotonin function and depressive states. In turn, the small changes observed in the human depressed subjects could correspond to a balance between decreased levels due to depression, which are partially reversed by antidepressant exposure in some subjects. These results will await confirmation and clarifications.
An ongoing debate in the depression field relates to the cellular and regional sites of action and/or role of serotonin receptors in the etiology of depression, but also on the nature of the networks supporting therapeutic effects after chronic antidepressant treatment. And here an interesting dissociation is worth noting. Indeed, the work of Svenningsson et al. provides evidence for “postsynaptic”, or “non-serotonergic” neurons mediating some of the depression-related behavior. Chronic antidepressant treatment or ECT increase p11 mRNA in frontal cortex but not in raphe, indicating that changes occur in non-serotonergic cortical neurons (mRNAs for presynaptic 5-HT1B receptor in serotonergic neurons are located in the raphe nucleus). Furthermore, due to the expression pattern of CAMK2, increased p11 levels in non-serotonergic neurons of Tg-p11 mice correlate with antidepressant-like state, suggesting that the neural mechanisms that are responsible for the expression of the depression-like and antidepressant-like phenotypes in this study are probably modulated, but not mediated, by 5-HT (i.e., 5-HT1B inhibition/modulation of other neurotransmitter system). In other words, the nature of the neurons/network that are regulated by 5-HT1B presynaptic inhibition and that may mediate the behavioral phenotypes are non-serotonergic.
Evidence is limited on the role of the 5-HT1B receptor in depression, especially when compared to the 5-HT1A receptor or the 5-HT transporter, for instance. So it is not known at this point whether p11 represents a novel candidate for dysfunction in depression. However, this study identified an important new element in the complexity of serotonergic neurotransmission, and may guide future studies towards specific cortical networks that either express presynaptic 5-HT1B receptors or that are targeted by these neurons. As the authors judiciously noted in their conclusion, their work sheds light on “molecular adaptations occurring in neuronal networks that are dysfunctional in depression-like states.”
View all comments by Etienne Sibille
It's most interesting that Paul Greengard and colleagues report lower levels of p11 in brain samples from depressed patients. Renegunta et al. report that knockdown of p11 with siRNA enhanced trafficking of TASK-1 to the surface membrane. Hopwood et al. find that present data suggest that the excitatory effects of 5-HT on DVN are mediated in part by inhibition of a TASK-like, pH-sensitive K+ conductance, and the Perrier group reports that 5-HT1A receptors inhibit TASK-1-like K+ current in the adult turtle. Might we suspect that a specific inhibitor of TASK-1 conductance would be beneficial in depression, and might this in part explain the benefit reported by SSRIs and agents with 5-HT1A receptor agonist activity in the treatment of depression?
References:Renigunta V, Yuan H, Zuzarte M, Rinne S, Koch A, Wischmeyer E, Schlichthorl G, Gao Y, Karschin A, Jacob R, Schwappach B, Daut J, Preisig-Muller R. The Retention Factor p11 Confers an Endoplasmic Reticulum-Localization Signal to the Potassium Channel TASK-1.
Traffic. 2006 Feb;7(2):168-81.
Hopwood SE, Trapp S. TASK-like K+ channels mediate effects of 5-HT and extracellular pH in rat dorsal vagal neurones in vitro.
J Physiol. 2005 Oct 1;568(Pt 1):145-54. Epub 2005 Jul 14.
Perrier JF, Alaburda A, Hounsgaard J. 5-HT1A receptors increase excitability of spinal motoneurons by inhibiting a TASK-1-like K+ current in the adult turtle.
J Physiol. 2003 Apr 15;548(Pt 2):485-92. Epub 2003 Mar 7.
View all comments by Mary Reid