Integrating Endocannabinoid Signaling and Cannabinoids into the Biology and Treatment of Posttraumatic Stress Disorder, Matthew N. Hill et al., 2018

Integrating Endocannabinoid Signaling and Cannabinoids into the Biology and Treatment of Posttraumatic Stress Disorder

Matthew N. Hill, Patrizia Campolongo, Rachel Yehuda and Sachin Pate

Neuropsychopharmacology REVIEWS, 2018, 43, 80–102.

doi : 10.1038/npp.2017.162


Exposure to stress is an undeniable, but in most cases surmountable, part of life. However, in certain individuals, exposure to severe or cumulative stressors can lead to an array of pathological conditions including posttraumatic stress disorder (PTSD), characterized by debilitating trauma-related intrusive thoughts, avoidance behaviors, hyperarousal, as well as depressed mood and anxiety. In the context of the rapidly changing political and legal landscape surrounding use of cannabis products in the USA, there has been a surge of public and research interest in the role of cannabinoids in the regulation of stress-related biological processes and in their potential therapeutic application for stress-related psychopathology. Here we review the current state of knowledge regarding the effects of cannabis and cannabinoids in PTSD and the preclinical and clinical literature on the effects of cannabinoids and endogenous cannabinoid signaling systems in the regulation of biological processes related to the pathogenesis of PTSD. Potential therapeutic implications of the reviewed literature are also discussed. Finally, we propose that a state of endocannabinoid deficiency could represent a stress susceptibility endophenotype predisposing to the development of trauma-related psychopathology and provide biologically plausible support for the self-medication hypotheses used to explain high rates of cannabis use in patients with trauma-related disorders.


Posttraumatic stress disorder (PTSD), while once characterized as a variant of an anxiety disorder, is now explicitly viewed as a separate entity and categorized as a trauma- or stressor-related disorder (APA, 2013). PTSD represents a pathological condition that emerges, sometimes after a period of incubation, following either direct or indirect exposure to a trauma. The original conceptualizations of PTSD viewed the disorder as more of a normative-type response that would occur following exposure to extremely stressful events, although more recent statistics indicate that it is only a proportion of individuals exposed to a trauma that actually meet diagnostic criteria for PTSD (Kilpatrick et al, 2013; Perkonigg et al, 2000). The biological mechanisms subserving the susceptibility to develop PTSD following exposure to a trauma remain elusive, although genetic factors, trauma load, and psychiatric co-morbidity are established risk factors (Almli et al, 2014; Pitman et al, 2012; Ross et al, 2017; Yehuda et al, 2015b).

In addition to trauma exposure, a diagnosis to PTSD requires presence of symptoms in four distinct clusters; intrusion, avoidance, arousal/reactivity, and negative cognitions/ mood (APA, 2013; Yehuda et al, 2015b). Exposure to trauma results in generation and consolidation of trauma memory via association of environmental and interoceptive cues with the negative physical and affective consequences of trauma exposure. Although such processes facilitate avoidance of potential future harms, dysregulation of these physiological processes are thought to be central to the development of PTSD. Thus, current conceptualizations describe PTSD fundamentally as a disorder of learning and memory processes (Bowers and Ressler, 2015b; Ross et al, 2017). Specifically, many of the predominant theories suggest that individuals that develop PTSD either have a greater propensity to consolidate or recall emotionally laden memories, or have impairments in the ability to appropriately extinguish associations between environmental cues and the negative effects and consequences of traumatic stress exposure (Careaga et al, 2016; Milad et al, 2009; Orr et al, 2000; Wicking et al, 2016). In addition, impairments in physiological habituation, or pathological sensitization mechanisms, are thought to contribute to the delayed onset of PTSD that often occurs (Lissek and van Meurs, 2015). Such dysregulations can result in intrusive re-experiencing symptoms in the forms of flashbacks or nightmares, and development of avoidance behaviors to minimize exposure to ‘triggers’, which predict danger and generate negative affective states. Hyperarousal and negative cognitive/mood states can be considered consequences of persistent re-experiencing and avoidance, as well as the associated functional decline often seen in PTSD patients. Overall, PTSD can be a highly debilitating illness often co-existing with anxiety disorders and substance use disorders, making effective treatment challenging with conventional approaches
such as SSRIs and cognitive-based psychotherapies.

The biological underpinnings of PTSD have been difficult to establish, although disturbances in a wide array of biological systems that could contribute to the development and maintenance of PTSD have been proposed (Horn et al, 2016; Kelmendi et al, 2016; McFarlane et al, 2017; Pitman et al, 2012). Not surprisingly, distributed cortico-limbic circuits important for salience attribution, cognitive processes,
and emotion generation and modulation have been implicated in the pathophysiology of PTSD. The amygdala represents a key structure in this regard, given its importance in the processing of emotionally relevant information, consolidation and extinction of emotional memories (particularly those related to traumatic stress), generation of anxiety states, and its role in activation of the sympathetic nervous system (SNS) in the periphery (Duvarci and Pare, 2014; Janak and Tye, 2015; LeDoux, 2007). The amygdala is particularly relevant for both the recognition (often at a preconscious level) of threatening stimuli in the environment, as well as the assembly of a behavioral response to threat, such as the generation of states of vigilance (Duvarci and Pare, 2014; Janak et al, 2015; LeDoux, 2007). With respect to PTSD, the amygdala appears to be hyper-reactive, exhibiting elevated metabolic activity during periods of heightened symptom presentation and showing increased responsiveness to emotionally salient information, even stimuli unrelated to the trauma itself (Diamond and Zoladz, 2016; Hughes and Shin, 2011; Sheynin and Liberzon, 2016; Shin et al, 2006).

In addition to the amygdala, subregions of the prefrontal cortex (PFC) are also believed to be relevant to the development and maintenance of PTSD. Specifically, the ventromedial PFC (vmPFC) in humans has repeatedly been found to be hypofunctional in individuals with PTSD (Hughes and Shin, 2011; Pitman et al, 2012), particularly during processing of trauma-related information and during extinction related tasks. For example, deficient recruitment of the vmPFC during fear extinction is believed to relate to the impairments in extinction seen in PTSD, which is consistent with the established role of the vmPFC in
promoting the extinction process. In fact, the reduction in vmPFC activity inversely correlates with the severity of PTSD symptoms (Shin et al, 2004), and there are consistent inverse relationships between activation of the vmPFC and amygdala, such that hyperactivity of the amygdala in PTSD is related to hypoactivity of the vmPFC (Shin et al, 2004). This coupling, both functional and structural, between the vmPFC and amygdala is known to be very important for emotional regulation, in addition to emotional memory stability, and impaired coupling of these structures is reliably found in individuals with PTSD or anxiety-related disorders (Gilmartin et al, 2014; Harris and Gordon, 2015; Kim et al, 2011; Likhtik and Paz, 2015). That being said, there are specific variants and subsets of this disease, such as those which experience a high degree of dissociation, that may represent as unique subtype of PTSD, which exhibits opposite alterations in activation of these cortico-limbic circuits (Lanius et al, 2010). As such, the proceeding discussion more specifically relates to the classic and typical presentation of PTSD, which is characterized by re-experiencing and hyperarousal.

In addition to these alterations in functional patterns of activity within cortico-limbic circuits, there are also alterations in many neuroendocrine systems in PTSD. A more indepth discussion of these findings can be found in (Daskalakis et al, 2013). Many of the initial studies of neuroendocrine function in PTSD demonstrated that, while individuals with PTSD appear to exhibit elevated levels of catecholamines and corticotropin releasing hormone (CRH), circulating levels of cortisol were quite surprisingly reduced in PTSD (Yehuda et al, 1996; Mason et al, 1986; Yehuda, 2009). Since these early studies, reduced levels of cortisol have generally been found to be a consistent and widespread finding in PTSD patients; although it is unclear whether reduced levels of glucocorticoids represent a cause or consequence of the disease, or reflect early adverse experiences impacting HPA-axis function (Daskalakis et al, 2013). There is some indication that these alterations in cortisol levels may in fact be reflective of alterations in regulatory components of the HPA-axis, such as FKBP5, a chaperone protein for the glucocorticoid receptor for which gene variants have been explicitly linked to susceptibility to PTSD (Binder et al, 2008; Klengel et al, 2013; Yehuda et al, 2009). The current data would indicate that alterations in glucocorticoid receptor sensitivity to cortisol could be associated with PTSD, such that greater receptor sensitivity could result in enhanced negative feedback and consequential reductions in circulating levels of cortisol (Binder, 2009).  Ongoing work is attempting to further understand and characterize the nature of these HPA-axis disturbances.

These lower-than-expected levels of cortisol have also been found to associate with a disinhibition of SNS activity in PTSD, which then results in persistent and steady-state increases in catecholamine secretion (Daskalakis et al, 2013). This chronic elevation in catecholamines, in turn, is associated with many PTSD symptoms such as hyperarousal and distress (Daskalakis et al, 2013; Krystal and Neumeister, 2009). Interestingly, human imaging work has found that glucocorticoids play an important role in tempering
amygdala responses to threatening cues and can sculpt functional connectivity between the amygdala and frontal cortical regions during emotional processing (Henckens et al, 2010, 2012; Joels et al, 2011). Preclinical studies have generally supported these findings, such that animal models of traumatic stress find lower levels of HPA responses to traumatic stressors are associated with greater long-term maladaptive changes (Bowens et al, 2012; Krishnan et al, 2007; Whitaker and Gilpin, 2015), and that glucocorticoids are required for normative fear extinction (Bitencourt et al, 2014; Yang et al, 2006, 2007). More so, preclinical research indicates that glucocorticoid administration in the immediate aftermath of traumatic stress exposure can restrict the development of long-term maladaptive changes from emerging (Whitaker et al, 2016; Zohar et al, 2011), which is paralleled by clinical evidence that elevating glucocorticoid levels could be ameliorative in the treatment of PTSD (Aerni et al, 2004; Yehuda et al, 2015a). As such, reduced levels of glucocorticoids, coupled to elevated levels of catecholamines, could very well be a primary contributing factor in PTSD, possibly working through impairing fear extinction processes, sensitization of the amygdala, and reduced functional coupling of the amygdala and vmPFC.

Another recent advance in the field of PTSD is the increased recognition of the role the immune system and inflammatory processes could play in the development of the disease (Michopoulos et al, 2017;Wieck et al, 2014). Elevated markers of inflammation such as C-reactive protein and proinflammatory cytokines have been identified in both the CSF and circulation of individuals with PTSD, both at rest and in response to an immune challenge (Michopoulos et al, 2017). Similarly, gene network and genome wide association studies have implicated immune-related genes in PTSD (Breen et al, 2015; Nievergelt et al, 2015; O’Donovan et al, 2011; Yehuda et al, 2009). At a functional level, it is interesting that cytokine signaling is known to promote emotional memory expression and impair fear extinction (Bi et al, 2016; Yu et al, 2017). In addition, preclinical studies have demonstrated that pro-inflammatory cytokines can facilitate glutamatergic transmission onto, and promote activation of, amygdalar pyramidal neurons (Chen et al, 2013; Engler et al, 2011; Prager et al, 2013). Similarly, in humans, inflammation and pro-inflammatory cytokines are related to enhanced activation of the amygdala in response to threatening stimuli (Inagaki et al, 2012; Swartz et al, 2017), further implicating immune dysregulation in the pathophysiology of PTSD.

Taken together, these findings indicate a prominent role of an imbalance in cortical-amygdala coupling, with hyperactivity of the amygdala and hypoactivity of the vmPFC in PTSD. In addition, PTSD is associated with reduced levels of cortisol, excess levels of catecholamines, and a state of persistent inflammation. These processes likely exist in a reciprocal feed-forward situation, where reduced levels of cortisol and elevated levels of norepinephrine and pro-inflammatory cytokines sensitize the amygdala and impair its coupling to the vmPFC, which in turn may provide additional drive on both the SNS and immune system (Muscatell et al, 2015; Tawakol et al, 2017). Because many of these biological and behavioral processes are influenced by cannabinoids and endocannabinoid (eCB) signaling, the overarching aim of this review is to provide a comprehensive summary of the current state of knowledge
regarding how cannabinoids and eCB signaling influence these processes in PTSD, and how eCB signaling could both represent a substrate for etiology of PTSD as well as a target for the development of novel therapeutics.


Cannabis is the most commonly used illicit recreational drug around the world, and contains over 80 terpeno-phenol molecules, which fall under the umbrella of ‘cannabinoids’ (Izzo et al, 2009). These plant-derived cannabinoids are typically referred to as phytocannabinoids, the most wellknown
of which is Δ9-tetrahydrocannabinol (THC), the primary psychoactive constituent of cannabis (Izzo et al,
2009). In addition to THC, the phytocannabinoid cannabidiol (CBD) is potentially relevant for the effect of cannabis on PTSD. While THC is known to exert its effects through direct activation of cannabinoid receptors, the pharmacology of CBD is more enigmatic and involves interactions with a series of neurochemical systems, most notably through interactions with serotonergic and adenosine signaling (Carrier et al, 2006; Izzo et al, 2009; Rock et al, 2012).