Calcineurin Signaling as a Target for the Treatment of Alcohol Abuse and Neuroinflammatory Disorders

Abstract

Converging lines of evidence point to a significant role of neuroinflammation in a host of psychiatric conditions, including alcohol use disorder, TBI, and PTSD. A complex interaction of both peripheral and central signaling underlies processes involved in neuroinflammation. Calcineurin is a molecule that sits at the nexus of these processes and has been clearly linked to a number of psychiatric disorders including alcohol use disorder (AUD). Like its role in regulating peripheral immune cells, calcineurin (CN) plays an integral role in processes regulating neuroimmune function and neuroinflammatory processes. Targeting CN or elements of its signaling pathways at critical points may aid in the functional recovery from neuroinflammatory related disorders. In this review we will highlight the role of neuroinflammation and calcineurin signaling in AUD, TBI and stress-induced disorder and discuss recent findings demonstrating a therapeutic effect of immunosuppressant-induced calcineurin inhibition in a pre-clinical model of binge alcohol drinking.

Introduction

Converging lines of evidence point to a role of neuroinflammation in a host of psychiatric conditions, including alcohol use disorder, TBI, and PTSD. A complex interaction of both peripheral and central signaling underlies processes involved in neuroinflammation. Calcineurin is a molecule that sits at the nexus of these processes and has been clearly linked to a number of psychiatric disorders including alcohol use disorder (AUD). Like its role in regulating peripheral immune cells, calcineurin (CN) plays an integral role in processes regulating neuroimmune function and neuroinflammatory processes. Calcineurin is a “fundamental link” between morphological changes in astrocytes and microglia and immune/inflammatory signaling in both acute and chronic neuroinflammation (for review, ¹). Calcineurin regulates glial cell activation and cytokine expression and is induced during disease, injury, and aging ^1,2. As such, targeting CN or elements of its signaling pathways at critical points may aid in the functional recovery from neuroinflammatory related disorders. ^2,3

Calcineurin

Calcineurin regulates immune/inflammatory processes in both peripheral T cells and brain glial cells through activation of the transcription factors nuclear factor of activated T-cells (NFAT) and nuclear factor kappa-light-chain-enhancer of activated B cells (NFκB). Calcineurin is a Ca2+ and calmodulin-dependent serine-threonine phosphatase also known as protein phosphatase 3 (PPP3). ⁴ Dephosphorylation of NFAT by CN causes activation and nuclear translocation where it stimulates transcription of several proinflammatory genes including interleukin-2 (IL-2). ^5-7 Calcineurin mediates activation and translocation of NFκB not through direct dephosphorylation of the transcription factor but through activation of a kinase that causes the removal of inhibitory subunits. ⁸ A host of pro-inflammatory cytokines strongly activate CN in primary astrocyte culture which leads to the expression of other immune/inflammatory molecules. Through its actions on NFAT and NFκB-dependent transcription, CN seems to be perfectly suited to drive the self-perpetuating “cytokine cycles” that have been implicated in chronic neuroinflammation (for review see,¹).

Alcohol and Neuroinflammation:

It is now widely accepted that immune mechanisms, and specifically neuroimmune mechanisms, play important, if poorly understood, roles in the etiology and progression of alcohol abuse ^3,9 The hypothesis that an overactive neuroimmune system promotes alcohol consumption is gaining widespread support among scientists. ^10-12 The first mention of a possible connection between immunosuppressants and alcohol consumption was published as a brief clinical report in 1990. ¹³ It was noted that AUD liver graft patients were more abstinent while on immunosuppressants.

Conventional treatment for alcoholism results in a 35-45% abstinence rate after 1 year. ¹⁴ Liver transplant for AD patients is controversial because of the concern that the patients may relapse and not only reject the organ, but also exhaust medical and financial resources ¹⁵. However, over the past 15 years, AD liver graft recipients have been noted as a clinically distinct group with much higher abstinence rates than AUD individuals taking FDA approved treatment. ^16-22 Most transplant patients one year following transplant showed about 90% abstinence rate and about 75-80% after three years. ²³ These abstinence rates are significantly higher than the abstinent rates seen in non-liver graft patients using conventional treatment. At first, it was assumed that selection of patients, environmental and psychosocial factors contributed to such high abstinence rates. However, detailed studies showed that none of these factors corresponded with the long-term abstinence rates shown in AD liver graft recipients. ²⁴ However, these studies did not rule out the effect of medications frequently used after a transplant. High abstinence rates in AD liver graft recipients following treatment with the immunosuppressant cyclosporine-A (CsA) were noted but did not get much attention. ¹³

Our group was the first to test a preclinical strategy to investigate the use of immunosuppressants to treat excessive drinking. ²⁵ It has taken many years for the research community to embrace and investigate this possible connection. But now multiple lines of converging evidence now point to neuroinflammation as playing a fundamental role in the development and maintenance of alcohol abuse. In our initial preclinical investigation, mice treated (IP) with the immunosuppressant cyclosporine A (CsA) drank significantly less ethanol in a controlled, continuous-access, two-bottle choice model while there was no effect on total liquid intake. ²⁵

We then sought to determine the mechanism of this effect by comparing three different immunosuppressants acting though different modes of action in a modified “drinking in the dark” (DID) paradigm of excessive drinking. ²⁶ Again, immunosuppressants were highly effective in decreasing alcohol consumption, but only those acting through inhibition of the phosphatase calcineurin (CN) and those that crossed the blood-brain barrier. CsA and tacrolimus (TRL or FK-506) inhibit CN activity through different mechanisms and both decreased ethanol consumption. A third immunosuppressant, sirolimus (SRL), that acts through a CN-independent pathway had no effect on ethanol consumption. Though inhibition of CN was clearly implicated, it was still unclear whether the anti-drinking effects were due to CN inhibition-mediated suppression of peripheral immune responses or direct inhibition of CN in brain or via another mechanism in brain. Both CsA and TRL crossed the blood brain barrier in this study. They also have known inhibitory effects on CN in brain where immunophilins and calcineurin are abundant and widespread. ²⁷ Sirolimus did not effectively penetrate brain ^26,28,29 suggesting that a general suppression of peripheral immune function was not enough to decrease excessive alcohol consumption.

We next sought to clarify if the CN-mediated inhibitory effect on drinking is acting via suppression of peripheral immune response or acting directly in the central nervous system (CNS). We hypothesized that central administration of CsA via intracerebroventricular (ICV) injection would be sufficient to inhibit excessive drinking behavior similar to what was observed after peripheral administration (IP). This was the case implicating neuroregulatory and neuroinflammatory processes in binge drinking behaviors ³ These findings add to a growing body of work implicating neuroinflammation in alcoholism.

Alcohol use leads to a range of neuroinflammatory responses and multiple compounds or genetic approaches that inhibit neuroimmune function have been shown to limit ethanol intake to varying degrees in animal models. ^10-12 These studies target different specific mechanisms, but all converge on the common outcome of reducing neuroimmune activation. Of these compounds, CsA has one of the strongest effects on attenuation of binge drinking; perhaps because along with reducing neuroinflammation, CsA also plays a key role in complementary neuronal signaling processes related to brain reward and addiction.

Calcineurin and Extended Brain Reward Pathways:

The effectiveness of calcineurin inhibitors in reducing ethanol consumption. ^3,25,26 may be due to their complementary roles in both neuroinflammatory pathways and directly in neuronal signaling pathways related to brain reward and addiction. During the past few decades, the diverse and important roles CN plays in neuronal signaling processes have been more and more appreciated. Calcineurin is one of the most abundant phosphatases in brain and is known to play important roles in a wide variety of signal transduction pathways. Not only involved in neuroimmune processes, CN is also expressed at especially high levels in brain neurons. ¹ There are particularly high concentrations of CN in the hippocampus, neocortex, striatum, and amygdala. ¹ where CN modulates diverse cellular functions including receptor and ion channel trafficking, ion channel function, apoptosis, and gene regulation. ³⁰ Calcineurin expression is very similar to that of calmodulin kinase II (CamKII) and it is abundantly expressed and in all of the regions of the brain where CaMKII is found including the dopaminergic reward pathway of the ventral tegmental area (VTA) and the nucleus accumbens (NAc). ³¹

Alcohol, Calcineurin, and Corticotropin Releasing Factor:

Calcineurin signaling regulates a host of pathways known to be involved in addictive behaviors. One key molecular mediator that has been implicated in ethanol consumption and the development of alcohol abuse is corticotropin releasing factor. ^32-34 A primary mediator of central and peripheral stress responses, CRF signaling in brain through CRF1 receptors has been consistently demonstrated to play a key role in many aspects alcohol abuse including binge-like ethanol drinking in the DID model in mice. ^34-36 Particularly, the CRF system has been shown to be central to the processes driving higher levels of intake, withdrawal-induced anxiety, relapse after withdrawal and sensitivity to stress. We have also recently demonstrated that increased conditioned fear cue-induced ethanol consumption is blocked by a CRF1 antagonist (unpublished results).

Calcineurin regulates CRF gene expression and CRF-mediated signaling. Transcription of CRF is mediated via cAMP response element binding protein (CREB) phosphorylation. Phosphorylation of CREB is necessary but not sufficient for CRF transcription. ³⁷ A CN-regulated coactivator called CRTC (cAMP-regulated transcriptional co-activators, formerly known as “TORC”, transducer of regulated CREB activity, not to be confused with “TORC” - mTOR signaling complex) is also required for CREB-induced CRF transcription ^38,39 Dephosphorylation of CRTC by CN facilitates translocation to the nucleus where it interacts with the dimerization and DNA binding domain of CREB and enables the recruitment of the preinitiation complex. This facilitates transcriptional activation of CREB-dependent genes including CRF. Stress alone induces nuclear translocation of CRTC and transcription of CRF in rodents and knockdown of CRTC isoforms completely blocks cyclic AMP-stimulated CRH promoter activity without affecting CREB phosphorylation. ^38,40 This CRTC mediated mechanism of stress induced CRF expression may underlie the enhanced sensitivity to stressors seen in models of ethanol abuse. Calcineurin is also involved in CRF receptor mediated signaling through a CN-NFAT pathway. Cyclosporine blocks CRF-receptor mediated catecholamine synthesis and release through inhibition of NFAT activation in cell culture. ⁴¹

Monoamine activity is also integral to processes related to reward, stress and addiction and CN has been shown to be involved in monoamine signaling. Along with direct effects on monoamine synthesis and release it has been shown that CN activity regulates an important dopamine and addiction related signal transduction protein - dopamine and cAMP-regulated neuronal phosphoprotein (DARPP-32). This protein regulates both dopaminergic and glutamatergic (NMDA) receptor responses in extended brain reward pathways. ^42-44 and is critically involved in regulating responses to ethanol and other drugs of abuse. ^45,46 DARPP-32 is a direct CN substrate and is a critical factor for dopamine dependent striatal synaptic plasticity. ⁴⁷ Dopamine causes increased phosphorylation and CN acts to maintain low levels of phosphorylated DARPP in neostriatum. ⁴⁸ It has been hypothesized that CN inhibition (i.e. CsA) acts as a dopamine mimetic via dopamine-like increases in phosphorylated DARPP-32. ⁴⁹ In this way, CN inhibition could potentially dampen temporally linked dopamine reward cues elicited by alcohol and lessen reinforcement.

Calcineurin also plays a regulatory role in other neurotransmitter systems related to alcohol reward and addiction, including NMDA and GABAergic neurons. We have demonstrated that quantitative NMR metabolomics can reliably determine brain levels of major neurotransmitters (such as glutamate, GABA, aspartate) and their changes upon CN inhibition. ⁵⁰ We demonstrated that CsA, but not SRL, reduces brain levels of glutamate and GABA. ^28,29,51 We have also shown that CsA and TRL treatment in the brain led to a specific decrease in mitochondrial metabolism of neurons, while their requirements for carbohydrate consumption (glucose) were elevated. ^28,52 Interestingly, we and other have shown that only CN-inhibitors (but not SRL) are known to produce a unique protective effects against ischemia, so-called ischemic preconditioning, by preserving/ slowing down mitochondrial metabolism and glutamate release in the brain, heart and kidney. ^53,54 The proposed protective mechanisms of CsA are directly linked to its CN-inhibition including increased phosphorylation of ERK and activation of AMP-activated protein kinase AMPK, specifically in rat hippocampus. ^55-57 AMPK signaling has long been recognized as an important regulator of brain, heart and body metabolism and energy homeostasis. There is a need to further establish the precise metabolic crosstalk between alcohol-induced changes in the brain and CN-inhibition using a metabolomic approach.

Corticotropin Releasing Factor, Ethanol and Neuroimmune Processes:

There is a reciprocal functional interaction between neuroimmune and CRF systems in brain. CRF is a critical mediator of neuroimmune responses including those seen after chronic ethanol administration. Enhanced cytokine expression (CCL2, IL1β, and TNFα mRNA) following acute withdrawal from chronic ethanol treatment is induced by CRF through CRF1 activation ⁵⁸ and cytokines drive anxiety during ethanol withdrawal via CRF signaling in the CeA. ^58,59 There is also a significant complex interaction between proinflammatory cytokines and the HPA axis. Cytokines such as IL-1β and IL-6 stimulate hypothalamic CRF transcription, CRF-induced ACTH release, and subsequent glucocorticoid release that functions to dampen further inflammatory responses. ^60,61

Calcineurin Signaling in Other Neuroinflammatory Disorders

Along with neurocognitive impairments, other persistent debilitating effects of TBI are common. These include chronic relapsing conditions such as PTSD and substance abuse, especially alcohol. ⁶² These problems have been perhaps best identified in military populations. Alcohol has been identified in a recent Institute of Medicine report as the key substance use problem in need of intervention and/or treatment among military personnel ⁶³ The rate of alcohol use disorder (AUD) among veterans is 32%, significantly higher than the general population. ⁶⁴ and there is evidence that Veterans with TBI have almost twice that as Veterans without TBI. ⁶⁵ Comorbid mental health conditions occur in 37% of people with a drinking problem. ⁶⁶ One of the most debilitating comorbid conditions is PTSD. Estimates of PTSD rates in Veterans who served in OEF/OIF range from double to almost ten times that of community rates ^67,68 Veterans from other conflicts also have a higher incidence. Vietnam era Veterans have lifetime rates estimated as high as 30%. ⁶⁹ Complicating the matter, individuals with PTSD, in the absence of TBI, have very high rates of comorbid AUD and substance abuse ranging from 28% to 75%. ^69,70 These conditions are inextricably intertwined and the causal mechanisms underlying them are likely shared, at least in part.

One key mechanism that connects these conditions is dysregulated neuroinflammatory processes (Figure 1). Neuroinflammation is a key feature associated with TBI, stress, and alcohol abuse. As such this is an area ripe for the investigation and development of therapies. As we have recently shown, an anti-inflammatory immunosuppressant acting in brain is sufficient to decrease binge alcohol drinking. ³ Evidence supports the notion that this general strategy targeting various aspects of neuroimmune signaling will be effective in treating some TBI comorbid conditions. Presently, there are no FDA approved medications for the treatment of comorbid TBI/AUD/PTSD.

Fig. 1.

Dysregulated neuroinflammatory signaling is a common factor linking multiple comorbid conditions.

Traumatic Brain Injury and Alcohol Abuse

By far, most studies investigating a link between alcohol abuse and TBI center on alcohol as a risk factor not the opposite. It is established that alcohol abuse is a risk factor for TBI but, importantly, alcohol abuse after TBI has been linked to increased seizures, psychiatric disorders including mood and anxiety. ⁷¹ There is not a clear answer to the question of whether TBI increases alcohol use due to many factors. First of all, many TBI patients abused alcohol before their injury which is highly predictive of abuse after injury. Secondly, after an acute decline in drinking found in many studies many return to drinking, or start drinking, so long-term follow-up studies are needed, and are lacking. Finally, other comorbid conditions such as PTSD and not TBI may account for increased use. There is an enhanced risk of increased alcohol abuse after TBI in military populations (for review see ⁷¹). Given the high incidence of TBI among Veterans in Iraq and Afghanistan, post-TBI drinking has been the subject of many recent studies. A link between combat brain injury and increased alcohol abuse has been reported by numerous studies. ^72-76 However, in some other studies, comorbid PTSD was suggested as the cause. ^77,78 Regardless, TBI increases the risk for PTSD which, in turn, increases the risk for alcohol abuse. Even when PTSD is accounted for, TBI has been found to be a significant risk factor for increased drinking. ^72,73 In one large study of over 13,000 Veterans who served in Afghanistan and/or Iraq, confirmed active duty TBI was associated with a 3-fold increase in substance abuse. ⁷⁹ Other studies of soldiers have also found increased association between TBI and alcohol abuse. A study of over 4,000 British soldiers who served in the same theaters found that TBI more than doubled the risk of later alcohol abuse. ⁷⁶ A third study of active duty soldiers also found an association between TBI and increased binge-drinking (odds ratio = 1.48). ^72,73

There is not a clear understanding why there is a greater link between TBI and alcohol abuse in military personnel but certainly, there is an interaction with stressors and a link with enhanced PTSD. It is likely that increased neuroinflammatory processes caused by chronic stressors may exacerbate the effects of TBI and that stressors after TBI may not only prevent healing but may act on a sensitized neuroinflammatory system yielding greater negative outcomes. There is evidence of this from a recent study. A single exposure to a predator odor delivered 8 months after blast-induced TBI produces elevated, persistent anxiety-like behavior compared to controls. ⁸⁰

Traumatic Brain Injury and Neuroinflammation

There is evidence that TBI injuries cause chronic dysregulated neuroinflammatory processes (for reviews see, ^81,82). Much evidence suggests that after well-documented acute increases in neuroinflammatory processes after TBI insult, there is a more protracted increase as well. Inflammation following TBI involves a complex interaction between and within peripheral and central systems. There is early activation of microglia and peripheral neutrophil recruitment followed by infiltration of lymphocytes and monocyte-derived macrophages along with a complex array of both pro- and anti-inflammatory cytokines. Post-TBI inflammation can be beneficial as it promotes both clearance of debris and regeneration. Acute inflammatory activity likely is also involved in secondary injury but later, more delayed sub-acute inflammatory responses may be beneficial for healing. Broad spectrum immunosuppressants are effective in preventing injury in models of ischemia but may not be as effective in the stages after insult. In fact, neuroinflammation may subside in the later stages of TBI. In rats 40 weeks after exposure to a model of low-energy blast-induced TBI chronic neuroinflammation (40 weeks) does not occur in the brain regions examined though they did find that focal hemorrhage may trigger chronic focal neuroinflammation. ⁸³ It is likely that while TBI does not necessarily lead to long-term overt measures of neuroinflammation, it sensitizes neuroinflammatory systems such that responses to future insults such as those caused by stress and alcohol are more robust.

Stress and Neuroinflammation

Stress also has a major effect on neuroinflammatory processes. Both acute and chronic stressors stimulate neuroinflammatory systems. Pro-inflammatory activation of microglia and increased infiltrated monocyte-derived brain macrophages result from psychosocial stressors. ^84-86 There is also an interaction between anxiety and neuroinflammation. High anxiety is associated with higher pro-inflammatory activation of microglia, especially after a lipopolysaccharide (LPS) challenge that drives a pro-inflammatory polarization of macrophages and microglia. ^87-89 Microglial changes in morphology, indicative of functional responses, are induced by both classic inflammatory signals such as LPS, but many of these same morphological changes are induced by psychosocial stress (for review see,⁹⁰). Overall, a variety of rodent models of stress have demonstrated increased microglial activity as determined by increased Iba-1 expression in key limbic and brain reward regions that regulate neural responses to both stress and substance abuse such as amygdala, nucleus accumbens, paraventricular nucleus, hippocampus, and prefrontal cortex. ⁹⁰

Evidence to date suggests an important role of neuroinflammatory processes in the development and expression of depression and PTSD. Different animal models of mild TBI have shown have been shown to increase anxiety and PTSD-like phenotypes. ^91,92 Recently, we have shown that inducing neuroinflammation with LPS (ICV) in lieu of a traumatic stressor dose-dependently causes a depression/PTSD-like phenotype (unpublished data). Together these studies suggest that traumatic stressors are acting, at least in part, through neuroinflammatory mechanisms to cause PTSD. Stressors also cause increased cytokine expression such as IL-1β in brain which is thought to, at least in part, mediate the neurochemical and behavioral consequences of stress such as PTSD. ⁹³

Given the complexity of variables surrounding the connection between TBI and alcohol abuse there is a clear need for good animal models to gain better insight into the complex connections between these disorders. Most rodent studies have focused solely on relatively acute alcohol use after TBI, none have investigated the interaction of TBI and stressors on drinking behaviors. In order to understand the persistent, more protracted effects of TBI most commonly seen in our VA clinics to better develop strategies for functional recovery. To accomplish this, researchers need to utilize model of TBI, combined with stress and PTSD models and alcohol paradigms to better understand their relationship and to decipher interactive mechanisms as well as test putative pre-clinical treatment strategies.

Genetic Approach to Understanding Calcineurin Function

Identification of the specific pathways and substrates involved immunosuppressant effects on drinking and other neuroinflammatory disorders needs to be studied taking a variety of approaches. One line of inquiry among many that we are pursuing is to identify specific CLN mediated effects using a genetic approach. We are using Cre-LoxP conditional knockouts to determine the role of specific calcineurin-mediated pathways. We have mouse line with a floxed regulatory subunit of calcineurin (a generous gift of S. Tonegawa, MIT) that we are using to determine cell types and regions in brain contributing to this response.⁹⁴ For example, to test the hypothesis that calcineurin pathways in the dopamine reward system are regulating rewarding properties of ethanol, we are knocking out CLN specifically in dopaminergic neurons and testing mice in drinking models. To accomplish this, floxed CLN mice are crossed with another line expressing cre recombinase in tyrosine hydroxylase expressing neurons. Likewise, to determine in general if CLN in glia or neurons mediate the anti-drinking effect of CsA we have crossed our floxed mice with mice expressing cre recombinase only in glia or neurons. Another strategy we are employing involves the use of viral vectors expressing cre or vectors to induce RNA interference mediated silencing of CLN. ^95,96 This allows for the focal knockout/silencing of CLN in specific brain regions. These studies and others will yield information that will allow us to target specific aspects of CLN signaling to better enable us to develop effective treatments for AUD and other neuroinflammatory disorders.

Conclusions

Calcineurin signaling provides a rich avenue for the pursuit of therapeutic targets for treatment of AUD, TBI, and PTSD in humans. Calcineurin is key mediator of neuroinflammation an is directly involved in neuronal signaling pathways known to be associated with a host of psychiatric disorders. A growing body of evidence suggests that various calcineurin mediated neuro-immuno-modulatory pathways could provide novel targets for the development of pharmacotherapies. ^3,12,97 Calcineurin acts in both glia and neurons to control signaling related to neuroinflammation, reward, and stress. These pathways include NFAT, DARPP-32, CREB via its co-activator CRTC (TORC), monoamine synthesis and release, CRF, GABA, and NMDA signaling, among others. Given the relative ubiquity of calcineurin regulated signaling in brain it is imperative to understand mechanisms of its signaling in order to determine effective targets and strategies for precision therapies. Because calcineurin signaling is so diffuse in brain, global inhibition with antagonists, though effective as an anti-drinking agent, is associated significant side effects, both neurological and peripheral. The ultimate goal is to understand the specific calcineurin-mediated pathways involved in psychiatric disorders and provide novel treatments.