Elsevier

Brain, Behavior, and Immunity

Volume 58, November 2016, Pages 237-247
Brain, Behavior, and Immunity

Full-length Article
Deficient adolescent social behavior following early-life inflammation is ameliorated by augmentation of anandamide signaling

https://doi.org/10.1016/j.bbi.2016.07.152Get rights and content

Highlights

  • Early-life inflammation on P14 decreases social behavior in adolescent rats.

  • LPS leads to an acute increase in anandamide (AEA) 6 h after LPS injection on P14.

  • P14 LPS reduces CB1 binding and alters FAAH activity and AEA in adolescent animals.

  • FAAH inhibitor PF-04457845 (orally or in the amygdala) rescues social behavior.

Abstract

Early-life inflammation has been shown to exert profound effects on brain development and behavior, including altered emotional behavior, stress responsivity and neurochemical/neuropeptide receptor expression and function. The current study extends this research by examining the impact of inflammation, triggered with the bacterial compound lipopolysaccharide (LPS) on postnatal day (P) 14, on social behavior during adolescence. We investigated the role that the endocannabinoid (eCB) system plays in sociability after early-life LPS. To test this, multiple cohorts of Sprague Dawley rats were injected with LPS on P14. In adolescence, rats were subjected to behavioral testing in a reciprocal social interaction paradigm as well as the open field. We quantified eCB levels in the amygdala of P14 and adolescent animals (anandamide and 2-arachidonoylglycerol) as well as adolescent amygdaloid cannabinoid receptor 1 (CB1) binding site density and the hydrolytic activity of the enzyme fatty acid amide hydrolase (FAAH), which metabolizes the eCB anandamide. Additionally, we examined the impact of FAAH inhibition on alterations in social behavior. Our results indicate that P14 LPS decreases adolescent social behavior (play and social non-play) in males and females at P40. This behavioral alteration is accompanied by decreased CB1 binding, increased anandamide levels and increased FAAH activity. Oral administration of the FAAH inhibitor PF-04457845 (1 mg/kg) prior to the social interaction task normalizes LPS-induced alterations in social behavior, while not affecting social behavior in the control group. Infusion of 10 ng PF-04457845 into the basolateral amygdala normalized social behavior in LPS injected females. These data suggest that alterations in eCB signaling following postnatal inflammation contribute to impairments in social behavior during adolescence and that inhibition of FAAH could be a novel target for disorders involving social deficits such as social anxiety disorders or autism.

Introduction

Inflammation, particularly during the prenatal period, has been associated with increased incidence of disorders such as autism (Patterson, 2011) and schizophrenia (reviewed in Green and Nolan, 2014, Meyer et al., 2011, Miller et al., 2013) in the offspring. This phenomenon has been intensively investigated using a variety of animal models where inflammatory agents such as lipopolysaccharide (LPS) and Poly inositol:cytosine (Poly I:C) have been administered to pregnant dams. However, in the protected fetal environment, it is unlikely that these substances enter the fetal circulation and the effects on the offspring are most likely secondary to inflammatory changes occurring in the placenta (Hsiao and Patterson, 2012). Cultivable bacteria in the amniotic fluid are present in less than 1% of normal births (Romero et al., 2002), whereas after birth infants may be exposed directly to a variety of pathogens. Nevertheless, developmental windows beyond prenatal and very early postnatal stages are less well studied in the context of inflammation.

Inflammatory insults with live bacteria on postnatal day (P) 4 have been associated with altered fear memory (Bilbo et al., 2005) and motor coordination (Lieblein-Boff et al., 2013), as well as potentiation of glial and cytokine responses in the adult rat (reviewed in Bilbo and Schwarz, 2012), particularly in response to a second inflammatory challenge (Bilbo et al., 2006). In a similar time frame, rats that were given LPS at P3 and P5 also show increased anxiety and hypersensitivity to stressful stimuli as adults (Sominsky et al., 2013), with hyper-responsiveness of the hypothalamus-pituitary-adrenal axis and decreased glucocorticoid receptor expression and abundance (Shanks et al., 1995, Shanks et al., 2000, Sominsky et al., 2013). These changes are complemented by increases in corticotropin releasing hormone (CRH) mRNA expression in the hypothalamic paraventricular nucleus, but not the amygdala (Amath et al., 2012) as well as decreases in cannabinoid type 1 receptor (CB1) density in the hippocampus (Amath et al., 2012) and amygdala (Zavitsanou et al., 2013).

Inflammation and the endocannabinoid (eCB) system share a variety of interconnections. First, cyclooxygenase-2 (COX-2) is an important enzyme in the response to inflammation. This enzyme can degrade both anandamide (AEA) and 2-arachindoyl glycerol (2-AG) and convert them to pro-inflammatory prostaglandins (Kozak et al., 2000, Yu et al., 1997). Secondly, the cytokines interleukin (IL)-1 and tumor necrosis factor (TNF)-α play an important role in synaptic transmission; IL-1 has been identified as a regulator of the effects of synaptic CB1, while CB1 in turn regulates the TNF-α mediated potentiation of excitatory synapses (reviewed by Rossi et al., 2015). In turn, inflammation has been shown to affect cannabinoid function. TNF-α has been suggested to increase AEA synthesis in hypothalamic fragments (Fernandez-Solari et al., 2006) and activation of toll-like receptor 4, the main signaling target of LPS, has been shown to increase AEA production in peripheral macrophages (Liu et al., 2006).

Long-term effects of LPS-induced inflammation at P14 in rodents have been investigated in the past (e.g. Galic et al., 2008, Mouihate et al., 2010, Spencer et al., 2005). P14 is comparable to a human infant of approximately 0.5–2 years of age (Gottlieb et al., 1977). Inflammatory insults experienced at P14 are more comparable to infections a young child would experience, while earlier ages likely mirror insults experienced during the late intrauterine period. As such, it is important to determine if the effects that occur during a fetal developmental stage are also seen postnatally, at a time when gene expression, electrophysiological and synaptic properties are undergoing experience-dependent maturation (Dehorter et al., 2012, reviewed in Semple et al., 2013). P14 represents a time of regional differentiation in gene expression (Stead et al., 2006) and high synaptic turnover, when microglia engulf synaptic elements actively (Paolicelli et al., 2011, Zhan et al., 2014). Within the first 14 postnatal days microglia are proliferating and migrate into the grey matter. Alterations in microglial function during the early postnatal period results in deficits in synaptic pruning, and ultimately less efficient neural transmission (Zhan et al., 2014). Behaviorally, this is associated with deficits in social interaction and increased repetitive-behavior phenotypes (Zhan et al., 2014). In adult rodents (Madore et al., 2013) and on P14 (Dinel et al., 2014), LPS administration leads to an upregulation of cytokines in the circulation and the brain. P14 further characterizes the end of the stress hypo-responsive period (Levine, 1994) and around this time, the brain reacts to LPS with an increase in paraventricular nucleus Fos, CRH mRNA and subsequent HPA axis activation (Dent et al., 1999). There is evidence that a variety of neuroinflammatory responses to LPS can be quite different around the second week of age, when compared to perinatal time points (P1) (Brochu et al., 2011). Rodents that have experienced early-life inflammation show alterations in anxiety (novelty induced suppression of feeding) and depression related tests (Dinel et al., 2014).

Amongst possible regions mediating changes in response to early-life inflammation, the amygdala stands out due to its involvement in emotional regulation. This structure is known to be affected by LPS (Frenois et al., 2007, Prager et al., 2013); on P14 the amygdala shows increased concentrations in proinflammatory (TNF-α, IL-1, IL-6) but not anti-inflammatory (IL-10) cytokines, which distinguishes it remarkably from other structures such as hypothalamus, hippocampus or prefrontal cortex (Dinel et al., 2014).

Altered excitability, activation and structure of the amygdala have been implicated in a variety of disorders including post traumatic stress disorder (PTSD) (Weston, 2014), schizophrenia (van Erp et al., 2014) and autism spectrum disorder (ASD) (reviewed by Zalla and Sperduti (2013)). Interestingly, in recent years a growing body of work emerged suggesting the eCB system to be altered in ASD. In particular CB1 appears decreased in post-mortem brains of autistic individuals, and polymorphisms in the CB1 coding gene CNR1 appears to be a predictor for an autistic phenotype. Furthermore, individuals with ASD show altered FAAH activity and AEA (reviewed by Chakrabarti et al. (2015)).

Social behavior and reward are strongly dependent on eCB signaling across many developmental epochs (Manduca et al., 2015, Wei et al., 2015). In adolescent rats, social experiences are associated with increased AEA levels within the amygdala and striatum (Marco et al., 2011, Trezza et al., 2012). During adolescence, and into adulthood, inhibition of eCB signaling can induce social withdrawal and social anxiety (Trezza et al., 2012, Seillier et al., 2013, Litvin et al., 2013), while systemic or intra-amygdala amplification of AEA signaling can enhance social interaction and contact (Seillier et al., 2013, Trezza et al., 2012).

Given the relationship between eCB signaling and social behavior, as well as the fact that early-life inflammation has been associated with alterations in the eCB system within the amygdala (Zavitsanou et al., 2013), we hypothesize that P14 LPS will decrease adolescent social behavior and that these changes will be sensitive to enhancement of AEA signaling through the inhibition of its degradative enzyme fatty acid amide hydrolase (FAAH).

Section snippets

Animals

Sprague Dawley (SD) rats obtained from Charles River Laboratories were maintained under standard specific pathogen-free environmental conditions and bred locally. For all behavioral experiments, litters were culled to no more than 12 pups and efforts were made to obtain roughly equal numbers of males and females. In all studies, no more than 2 pups per litter/gender, per treatment group were used for any measure. Endocannabinoid levels were determined at 2 different time points (P14 and P40)

Social behavior

P14 LPS injected adolescent animals displayed significantly less combined social behavior (Fig. 3A), when interacting reciprocally with a non-cagemate of equal age, treatment and sex (n = 48, F (1, 44) = 20.55, p < 0.0001). Post-hoc testing localized showed this effect of early-life inflammation in males (t (22) = 3.328, p < 0.01) and females (t (22) = 3.038, p < 0.01). No significant effect of sex was found (n = 48, F (1, 44) = 1.064, p = 0.3079). When distinguishing social play and social non-play behaviors (

Discussion

With this research, we have shown that a single immune challenge during a sensitive developmental window can affect adolescent social behavior. Furthermore, early-life inflammation acutely altered AEA levels and evoked a multi-level reprogramming of the amygdalar eCB system. In adolescent animals, P14 LPS injections were associated with reduced CB1 receptor binding site densities, elevated levels of AEA and, surprisingly, increased AEA hydrolysis by FAAH. Interestingly, pharmacological

Acknowledgments

This work supported by CIHR, NSERC and Brain Canada grants to QJP and MNH; SP’s contribution is supported by NIH Grants MH090412 and MH103515. PF-04457845 was donated by Pfizer Inc. VMD and JMG are supported by Hotchkiss Brain Institute trainee awards; JMG holds an AIHS postdoctoral fellowship; CMS holds a AIHS undergraduate research studentship, MNH is a CRC Tier 2 Research Chair; QJP is an AHFMR Medical Scientist.

References (62)

  • S. Levine

    The ontogeny of the hypothalamic-pituitary-adrenal axis. The influence of maternal factorsa

    Ann. N. Y. Acad. Sci.

    (1994)
  • C. Madore et al.

    Early morphofunctional plasticity of microglia in response to acute lipopolysaccharide

    Brain Behav. Immun.

    (2013)
  • A. Manduca et al.

    Distinct roles of the endocannabinoids anandamide and 2-arachidonoylglycerol in social behavior and emotionality at different developmental ages in rats

    Neuropsychopharmacol.

    (2015)
  • E.M. Marco et al.

    Social encounter with a novel partner in adolescent rats: activation of the central endocannabinoid system

    Behav. Brain Res.

    (2011)
  • M.J. Meaney et al.

    A descriptive study of social development in the rat (Rattus norvegicus)

    Anim. Behav.

    (1981)
  • B.J. Miller et al.

    Prenatal inflammation and neurodevelopment in schizophrenia: a review of human studies

    Prog. Neuropsychopharmacol. Biol. Psychiatry

    (2013)
  • P.H. Patterson

    Maternal infection and immune involvement in autism

    Trends Mol. Med.

    (2011)
  • M. Qin et al.

    Endocannabinoid-mediated improvement on a test of aversive memory in a mouse model of fragile X syndrome

    Behav. Brain Res.

    (2015)
  • R. Romero et al.

    Infection and prematurity and the role of preventive strategies

    Semin. Neonatol.

    (2002)
  • S. Rossi et al.

    The interplay between inflammatory cytokines and the endocannabinoid system in the regulation of synaptic transmission

    Neuropharmacol.

    (2015)
  • B.D. Semple et al.

    Brain development in rodents and humans: identifying benchmarks of maturation and vulnerability to injury across species

    Prog. Neurobiol.

    (2013)
  • S.J. Spencer et al.

    Early life immune challenge—effects on behavioural indices of adult rat fear and anxiety

    Behav. Brain Res.

    (2005)
  • T.G.M. van Erp et al.

    A multi-scanner study of subcortical brain volume abnormalities in schizophrenia

    Psychiatry Res.

    (2014)
  • L.J.M.J. Vanderschuren et al.

    The neurobiology of social play behavior in rats

    Neurosci. Biobehav. Rev.

    (1997)
  • M. Yu et al.

    Synthesis of prostaglandin E2 ethanolamide from anandamide by cyclooxygenase-2

    J. Biol. Chem.

    (1997)
  • D.G. Amaral

    The amygdala, social behavior, and danger detection

    Ann. N. Y. Acad. Sci.

    (2003)
  • S.D. Bilbo et al.

    Neonatal infection-induced memory impairment after lipopolysaccharide in adulthood is prevented via caspase-1 inhibition

    J. Neurosci.

    (2005)
  • M.-E. Brochu et al.

    Developmental regulation of the neuroinflammatory responses to LPS and/or hypoxia-ischemia between preterm and term neonates: an experimental study

    J. Neuroinflammation

    (2011)
  • E. Dainese et al.

    Membrane lipids are key modulators of the endocannabinoid-hydrolase FAAH

    Biochem. J.

    (2014)
  • N. Dehorter et al.

    Timing of developmental sequences in different brain structures: physiological and pathological implications

    Eur. J. Neurosci.

    (2012)
  • A.-L. Dinel et al.

    Inflammation early in life is a vulnerability factor for emotional behavior at adolescence and for lipopolysaccharide-induced spatial memory and neurogenesis alteration at adulthood

    J. Neuroinflammation

    (2014)
  • Cited by (59)

    • Effects of endocannabinoid system modulation on social behaviour: A systematic review of animal studies

      2022, Neuroscience and Biobehavioral Reviews
      Citation Excerpt :

      While an unequivocal animal model of social impairment is not available, several paradigms have been developed to model various neuropsychiatric disorders involving social dysfunction. Amongst the studies that met inclusion for this review (i.e., by measuring social interaction behaviours following manipulation of the ECS), social anxiety was induced by social peer rejection, restraint stress, maternal separation and foot-shock exposure (Toth and Neumann, 2013); ASD-like phenotypes were modelled by Fmr1 gene knock-out (KO) mice, BTBR and 129S1/SvImJ (129 S) mice, or prenatal exposure to valproic acid (pVPA; (Kerr et al., 2013; Zamberletti et al., 2017)) or lipopolysaccharides (pLPS; (Doenni et al., 2016)); negative symptoms of schizophrenia, which include deficits in social behaviour, were modelled using phencyclidine (PCP) or methylazoxymethanol acetate (MAM)-induced social withdrawal, (Haller et al., 2005; Stark et al., 2019) or genetic manipulations (see Table 3). The synthesis of results followed a narrative approach.

    • The role of the endocannabinoid system as a therapeutic target for autism spectrum disorder: Lessons from behavioral studies on mouse models

      2022, Neuroscience and Biobehavioral Reviews
      Citation Excerpt :

      Interestingly, this effect was independent from changes in anxiety-levels and was prevented by CB1r blockade (Wei et al., 2016). Enhancing AEA signaling through inhibition of its degradation was also effective in mitigating the behavioral phenotypes induced by prenatal VPA (Kerr et al., 2016; Servadio et al., 2016) or LPS (Doenni et al., 2016) exposure in environmental rat models of ASD. In line with the relevance of the endocannabinoids as a therapeutic target in ASD, some phytocannabinoids, such as CBD have been suggested as treatments for this syndrome (Poleg et al., 2019), as previously mentioned.

    View all citing articles on Scopus
    View full text