Animal models of PTSD: a challenge to be met

Abstract

Recent years have seen increased interest in psychopathologies related to trauma exposure. Specifically, there has been a growing awareness to posttraumatic stress disorder (PTSD) in part due to terrorism, climate change-associated natural disasters, the global refugee crisis, and increased violence in overpopulated urban areas. However, notwithstanding the increased awareness to the disorder, the increasing number of patients, and the devastating impact on the lives of patients and their families, the efficacy of available treatments remains limited and highly unsatisfactory. A major scientific effort is therefore devoted to unravel the neural mechanisms underlying PTSD with the aim of paving the way to developing novel or improved treatment approaches and drugs to treat PTSD. One of the major scientific tools used to gain insight into understanding physiological and neuronal mechanisms underlying diseases and for treatment development is the use of animal models of human diseases. While much progress has been made using these models in understanding mechanisms of conditioned fear and fear memory, the gained knowledge has not yet led to better treatment options for PTSD patients. This poor translational outcome has already led some scientists and pharmaceutical companies, who do not in general hold opinions against animal models, to propose that those models should be abandoned. Here, we critically examine aspects of animal models of PTSD that may have contributed to the relative lack of translatability, including the focus on the exposure to trauma, overlooking individual and sex differences, and the contribution of risk factors. Based on findings from recent years, we propose research-based modifications that we believe are required in order to overcome some of the shortcomings of previous practice. These modifications include the usage of animal models of PTSD which incorporate risk factors and of the behavioral profiling analysis of individuals in a sample. These modifications are aimed to address factors such as individual predisposition and resilience, thus taking into consideration the fact that only a fraction of individuals exposed to trauma develop PTSD. We suggest that with an appropriate shift of practice, animal models are not only a valuable tool to enhance our understanding of fear and memory processes, but could serve as effective platforms for understanding PTSD, for PTSD drug development and drug testing.

Fig. 1

Guiding principles for selecting an animal model of PTSD. There is no single animal model of PTSD which as such is more adequate than others. Different models have their pros and cons. There are, however, several principles which are, in view of the authors, critical and should be incorporated in any model in order to increase the ecological validity of the models and their relevance to the human psychopathology. a Humans exposed to trauma are of heterogeneous genetic background. It should be natural to select outbred strains with a similar heterogeneous genetic background. Selecting an inbred line should be considered as a manipulation which defines the scope of the outcome of the study. As is indicated by the pale blue arrowed line (h), the genetic disposition may mediate effects at multiple levels, i.e., at the juvenile adversity, the trauma perception, and/or the individual outcome. b Males and females are sensitive to stress and trauma in different ways. The preference should be to examine both males and females. However, it should be borne in mind that specific manipulations may affect more either males or females. Likewise, males and females may differ in the behavioral aspects which are affected and thus different behavioral tests may be required in order to identify affected males or females. As is indicated by the pale blue arrowed line (i), sex differences may mediate effects at multiple levels, i.e., at the juvenile adversity, the trauma perception, and/or the individual outcome. c Most individuals exposed to trauma will not develop PTSD, indicating that the trauma will only be effective if it interacts with some additional pre-disposing factors. Studying the neurobiology of pre-disposing factors is thus a fundamental part of understanding the neurobiology of PTSD. Adding potential risk factors and examining their contribution should be considered, regardless of which trauma model is employed. d As is indicated in the text, there is no right or wrong with regards to which trauma should be employed in animal models of PTSD. Also, in humans different types of trauma may lead to the development of the disorder in some individuals. However, the choice of trauma and its parameters should be carefully considered and described, since this choice defines the relevance of the outcome of the study to exposure of similar nature in humans. Furthermore, researchers should address the question of the assumed severity of the traumatic experience. A clearer dissociation between stressful experiences (which are within the coping abilities of the animal) and traumatic experiences (which are beyond the coping abilities of the animal) is needed and should become part of the discussion of any study. e The age of exposure to trauma as well as the time after exposure for testing the impact of the exposure are important factors to consider. There is no right or wrong here but those choices define the relevance of the outcome of the study to exposure of similar nature and to the stage of evaluation in humans. f Because only some individuals exposed to a trauma will develop psychopathology it is critical to move away from analyzing the averages of the exposed and non-exposed groups. Instead, a more individual characterization of each animal as being pathologically affected or not is required. Towards that end, it seems important to aim for examining animals over batteries of tests that cover several behavioral faculties, in order to achieve a more reliable profiling of the individual animals. g Individual profiling of the animals could then be translated into defining individual animals as affected or non-affected, in a similar way to diagnosis in humans. With that type of analysis, effects of drugs can be examined as the impact of the proportion of affected/non-affected individuals, rather than on the averaged severity of specific symptoms