The repeated cross fostering protocol as a mouse model of panic disorder: suggestions for new treatments from behavioral and molecular characterization

Panic disorder (PD) is a common psychiatric illness with highly stereotyped symptoms including a sense of shortness of breath or feelings of suffocation. PD is characterized by spontaneous and recurrent panic attacks (PAs) that consist of incapacitating periods of acute-onset respiratory, cardiovascular, gastrointestinal, autonomic and cognitive symptoms. According to the DSM-5, recurrent panic attacks in PD are categorized as being either spontaneous (unexpected) or cued (expected). Accumulating evidence suggests that spontaneous PAs may be provoked by interoceptive sensory triggers caused by fluctuations in the internal homeostatic milieu. An important internal homeostatic trigger for the genesis of PAs, supported by an emerging body of work, is acid–base imbalance and associated pH chemosensory mechanisms. Largely founded on panic provocation studies with agents promoting homeostatic pH imbalance (e.g. sodium lactate or CO2) and related to the false suffocation alarm theory the role of acid–base and chemosensory systems in panic provides strong scientific insights on the genesis of PAs. Heightened sensitivity to carbon dioxide (CO2) is an established biological correlate of PD. Indeed inhaled CO2 triggers PAs in most individuals with PD but only a minority of unaffected controls. More recently the acid sensing ion channel-1 (ASIC1) has been proposed as a candidate gene responsible for these phenomena. Indeed this channel is expressed in central nervous system and in particular in amygdala, brainstem, structures involved in chemosensory detection and breathing. According to twin studies, shared genetic determinants appear to be the major underlying cause of the developmental of adult PD and of altered sensitivity to CO2. Moreover, in addition to genetic determinants, environmental risk factors affect the liability to these traits, and several life events that influence the susceptibility to PD also predict heightened CO2 reactivity. There is evidence that genetic and environmental determinants interact to influence human responses to CO2 suggesting also epigenetic mechanisms to underlie the development of PD. Valuable animal models for PD are lacking due to the difficulty to measure “panic” in animals. Indeed models of PD used in pre-clinical research measure the defensive behaviors showed by the animals in response to a real aversive stimulus and not spontaneously and/or in the absence of real dangerous situation, as in PD. Not being able to interview the animal, about its symptoms, such as fear of dying or going crazy as in human PD patients, CO2 hypersensitivity, observed in patients with panic and their unaffected relatives, represents a valid endophenotype to model this disorder in animals. PD is a chronic disorder with variable course and treatments available until now are not specific and are often only modestly efficacious. Typical pharmacologic treatments are antidepressants (SSRI) or anxiolytics (benzodiazepines). An alternative strategy is psychotherapy (cognitive behavioral therapy) and often patients are treated with a combination of psycho- and pharmaco-therapies. In some cases therapies (e.g., benzodiazepines) may be associated with treatment-specific side effects or risks such as sedation or the risk of dependence or tolerance. For these reasons is important to find a therapy specific for PD, possibly with the fewest side effects. The aims of my PhD Thesis were: 1) to validate the RCF protocol in mice as a useful manipulation procedure affecting individual emotionality, different from the classical maternal separation (Handling), usually applied in rodents, to evaluate the effects of an early adverse environment. I evaluated the short and long-term effects of these early manipulations on several behavioral, molecular and physiological parameters (mother-pups interaction; stress response; emotionality; CO2 panic-related response; gluco- and mineral-corticoid receptors mRNA expression; etc.) 2) to analyze possible molecular mechanisms underlying the panic-related CO2 hypersensitivity showed by RCF animals and evaluate different pharmacological treatments (chloridiazepoxide, chlorogenic acid and amiloride) able to recover the normal respiratory response to hypercapnia, on the basis of molecular suggestions 3) to verify the cognitive capability of RCF animals trough learning tests (such as active avoidance test and novel recognition test) and investigate the capability of exposure to 6% CO2 on animals’ behavioral conditioning, in RCF and Control subjects. Indeed, humans with PD show behavioral conditioning to panic attacks and develop PA also in absence of unconditioned stimulus. 4) to investigate whether the CO2 hypersensitivity showed by RCF animals was a transgenerational transmissible trait. First of all, results reported in this study suggest that the behavioral and physiological phenotypes observed during development and adulthood depend on characteristics and timings of early adversities capable of activate different biological processes. Reasonably, the response of the animal to the early manipulations is different and aimed at maximizing individual fitness: the early environment could exert its programming role during this developmental plastic period, through specific epigenetic modifications. Short, even if repeated, separations from the mother (Handling protocol) induce habituation to a relatively low stressing environment, enhancing the capability of the subject to face new stressful situations. By contrast, the disruption of the infant attachment bond (RCF protocol) is associated to a modification in the respiratory response to high CO2 in breathing air, an endophenotype these animals share with PD patients. The disruption of infant-mother bond in RCF animals suggested by the enhanced separation anxiety at 8 days age supports the relation between SAD and PD already suggested in literature. In addition the CO2 reactivity showed by these animals represents a useful tool to study PD in pre-clinical research. Molecular alterations found in RCF animals (experiment 2a) supported the involvement of acid-base balance dysregulation in development of CO2 hypersensitivity. Indeed RCF animals showed a higher expression in ASIC1 gene that codifies for acid sensing ion channels. These channels are sensitive to lower levels of pH being able to detect changes in CO2 concentration in the body and adjust the respiratory function to receive enough O2 not to compromise biological processes. Molecular investigations in addition revealed alterations in GABAergic transmission in RCF animals supporting the idea of an involvement of this neurotransmitter in the development of PD. RCF animals showed an increased expression of Dbi which is an inhibitor of GABAergic transmission. These molecular findings have provided indications suggesting that a possible rescue treatment for PD patients should consist in reducing CO2 hypersensitivity. Lowering of this increased respiratory response to modest increase in CO2 could reduce the negative feeling associated to condition, reducing the conditioning potentiality that favor the development of panic disorder, after repeated panic attacks. The use of benzodiazepine such as chlordiazepoxide was able to restore the normal respiratory response to CO2 as well, giving pharmacological validation to RCF model. However, benzodiazepines have several contraindications, especially for chronic treatments and their sedative effect should also be taken into consideration. Even if I only present few data on the effects of chlorogenic acid and amiloride on RCF animals, I think these results are very interesting and need further and deeper evaluation. Both these compounds interacted with the pH sensitive channels (asics) and their administration was able to restore the respiratory response observed in control animals. It is well known, that panic attacks are able to condition behaviors of PD patients. They indeed tend to avoid situations and places similar to those where a panic attack previously occurred. Similarly RCF animals showed, in experiment 3, behavioral conditioning to the situation previously paired with CO2 (tone exposure). It should be now explored whether RCF animals generalize the conditioned fear, suggesting how an initial panic attack can evolve into panic disorder in humans. Finally RCF model demonstrated a transgenerational transmission of the respiratory endophenotype (experiment 4) supporting the hypothesis of gene-enviroment interplay role to predisposition to panic disorder (Spatola et al., 2011). The epigenetic mechanisms responsible for this trans-generational transmission are under investigation as well as possible strategies to prevent this phenomenon. In conclusion, the Repeated Cross-Fostering protocol seems a valid mouse model of Panic Disorder in humans: RCF mice show typical features of this disorder such as separation anxiety during childhood, CO2 hypersensitivity and CO2 conditioned and avoidance behaviors. Acid sensing ion channels are interesting molecular markers which can be used as new targets for pharmacological treatments and can help to explain hyper-responsiveness to CO2 in PD patients as well.