Neuroimmune and epigenetic mechanisms underlying persistent loss of hippocampal neurogenesis following adolescent intermittent ethanol exposure

Abstract

Alcohol abuse and binge drinking are common during adolescence – a maturational period characterized by heightened hippocampal neuroplasticity and neurogenesis. Preclinical rodent models of adolescent binge drinking (i.e., adolescent intermittent ethanol [AIE]) find unique vulnerability of adolescent hippocampal neurogenesis with reductions persisting into adulthood after ethanol cessation. Recent discoveries implicate increased neuroimmune signaling and decreased neurotrophic support through epigenetic mechanisms in the persistent AIE-induced loss of neurogenesis. Importantly, interventions aimed at rectifying the increased neuroimmune signaling and neurotrophic-epigenetic modifications through physical activity, anti-inflammatory drugs, and histone deacetylase inhibitors protect and recover the loss of neurogenesis and cognitive deficits. The mechanisms underlying the persistent AIE-induced loss of adult hippocampal neurogenesis could contribute to broader neurodegeneration, loss of hippocampal neuroplasticity, and cognitive dysfunction.

Introduction

The adolescent brain is rapidly maturing as neurons and glia interact to form the imprint of mature circuitry that persists throughout adulthood. In the rodent hippocampus, adolescent maturation is characterized by a burst of neurogenesis that progressively declines across adulthood [1,2]. For instance, employing the S-phase mitotic marker 5-bromo-2’-deoxyuridine (BrdU), Cameron and McKay [1] estimated that ~17,000 new cells are generated in the hippocampus of the naïve rat during adolescence (i.e., postnatal day [P]35) within a 24-hour time frame whereas ~9,000 cells are observed in the young adult (i.e., P70) hippocampus. These newly formed hippocampal granule neurons are incorporated into the granular cell layer of the dentate gyrus and have been implicated in hippocampal-mediated cognitive function [3,4]. While there is some evidence that the decline in hippocampal neurogenesis across age is also observed in humans [5], which may contribute to the age-associated decline in cognitive function, this concept is currently under debate due to recent reports with conflicting evidence [6,7] and a renewed debate of technical limitations, which are beyond the scope of this review [for an excellent review of this debate, see 8,9]. In contrast, hippocampal neurogenesis has repeatedly been identified as a predictor of cognitive function in rodents, and loss of neurogenesis is implicated in a variety of rodent models of neuropsychiatric disorders [for review, see 10]. Although many environmental factors have been found to affect hippocampal neurogenesis, alcohol (ethanol) has repeatedly been found to inhibit markers of hippocampal neurogenesis in rats and non-human primates [11]. As adult ethanol inhibition of neurogenesis has previously been reviewed [see 12,13 for review]. This review focuses on recent studies on the effect of AIE on hippocampal neurogenesis in rodents and explores potential mechanisms underlying the persistent AIE-induced reductions of neurogenesis.

Persistent Loss of Hippocampal Neurogenesis Following Adolescent Binge Alcohol Exposure

Over 50 years ago, Altman and Das [14] discovered neurogenesis occurred in the adult hippocampus overturning the long-held accepted theory that the adult brain lacks the capacity to form new neurons. Although this discovery was initially met with skepticism, ample evidence now exists to suggest that adult neurogenesis is well conserved across a variety of mammalian species where radial glial cells proliferation, differentiate, and functionally integrate as neurons into existing neurocircuitry. However, the extent of neurogenesis appears to have an inverted relationship with cognitive capacity at a species level such that rodents have much higher levels of neurogenesis than primates, and human neurogenesis remains a point of contention [8]. Within the rodent hippocampus, neurogenesis is restricted to the mitotically active subgranular zone of the dentate gyrus [15]. It involves cell proliferation and progression of neuroprogenitor cells (NPCs) from the precursor cell phase to the late post-mitotic maturation phase that can be distinguished based on various marker proteins [see e.g., 16]. Newborn hippocampal granule cells originate from a population of radial glia-like precursor cells (type 1 cells) that express sex determining region Y-box 2 (Sox2) which give rise to intermediate progenitor cells with initial glial (type 2a) and then neuronal (type 2b) phenotypes that express the T-box brain protein 2 (Tbr2) [17]. The migratory neuroblast-like stage (type 3) gives rise to immature neurons expressing doublecortin (DCX) [18] that exit the cell cycle and then enter the maturation stage, ultimately developing the mature excitatory granule cell phenotype. Hippocampal rodent neurogenesis is modulated by a host in internal and external stimuli, including neurotransmitters (e.g., acetylcholine [19]), neurotrophins (e.g., brain-derived neurotropic factor [BDNF [20]), physical activity [21,22], traumatic brain injury and other pathologies, and drugs of abuse. While the contributions of rodent hippocampal neurogenesis to learning and memory have not been fully elucidated, accumulating data support a role for this process in adaptation to novelty [23] as well as hippocampal-dependent cognitive and emotive function [4,24]. For instance, facilitation of neurogenesis increases cognitive function [21,25], whereas inhibition of neurogenesis decreases cognitive function [26].

Acute ethanol treatment of rats dose-dependently suppresses hippocampal neurogenesis with adolescents exhibiting increased sensitivity relative to adults [27]. Studies in adult rats have found acute and chronic alcohol dependency models suppress hippocampal neurogenesis while abstinence in these models recovers neurogenesis to age-appropriate levels [28,29]. In contrast, adolescent intermittent ethanol (AIE) exposure, which models human adolescent intermittent binge drinking, causes long-lasting reductions of neurogenesis that persist through adulthood despite abstinence [see e.g., 2]. This suggests that adolescent neurogenesis is particularly vulnerable to the detrimental effects of ethanol exposure, resulting in alterations to the developmental trajectory of the hippocampus. Since alcohol is one of the most prevalently used and abused drugs during adolescence, investigations into the mechanisms underlying the long-term consequences of adolescent binge drinking on hippocampal neurogenesis are critical.

Adolescent drinking habits are typically characterized by weekend binge drinking, which the National Institute on Alcohol Abuse and Alcoholism (NIAAA) defines as the consumption of five or more consecutive alcoholic beverages in males (≥4 in females) in a two-hour period. These higher rates of drinking may be due to adolescents’ diminished sensitivities to the sedative effects of alcohol [for review, see 30]. Alcohol binge drinking is common during adolescence as 4% of 8^th grade, 9% of 10^th grade, and 14% of 12^th grade students report binge drinking in the past two weeks [31]. This heavy drinking pattern continues through the college years as 44% of students report binge drinking every two weeks, and 19% report more than three binge drinking episodes per week [32,33]. The heightened neuroplasticity that characterizes the hippocampus during adolescence also increases its vulnerability to the deleterious effects of alcohol binge drinking. Indeed, greater levels of neurogenesis are observed in the adolescent dentate gyrus relative to adults, with levels of neurogenesis declining with age across mammalian species [1,2,34]. Using preclinical rodent models wherein ethanol is administered intermittently during adolescence to model human weekend binge drinking, our laboratory and others find persistent reductions of hippocampal neurogenesis across rat strains (i.e., Sprague-Dawley and Wistar rats) and ethanol administration paradigms (i.e., intragastric, intraperitoneal, vapor, and voluntary self-administration) [2,17,20,27,35–38] (see Table 1). For instance, our laboratory found that AIE exposure to rats during adolescence (i.e., P25 – P55) reduced expression of DCX, which is a neuroprogenitor microtubule-associated protein expressed specifically by immature neurons [18], in late adolescence (i.e., P56; 24 hr post-AIE) that persisted into adulthood (P220; 165 days post-AIE) in both the dorsal and ventral dentate gyrus. Similar reductions of DCX-immunoreactive (+IR) cells are observed in post-mortem dentate gyrus of humans with alcohol use disorder (AUD) [39]. The persistent loss of neurogenesis in the AIE model appears to be specific to adolescence as adult ethanol exposure transiently reduces hippocampal neurogenesis that recovers over a period of abstinence [28,29]. In fact, Broadwater and colleagues [37] found that the persistent loss of hippocampal neurogenesis is specific to binge ethanol exposure during adolescence and not adulthood, indicating a unique vulnerability of the adolescent hippocampus to the detrimental effects of ethanol [27].

Table 1.

Effect of adolescent intermittent ethanol (AIE) exposure on hippocampal neurogenesis.

Administration Paradigm	Species/Strain	Age of Assessment	Effects of AIE
			Cell Proliferation	Neurogenesis	Cell Death
Intragastric administration
Liu and Crews [9]	Wistar rat	Adolescent	-- Ki67, -- BrdU, ↓ Sox2, ↓ Tbr2	-- DCX	↑ CC3
		Adult	↓ Ki67, ↓ BrdU, -- Sox2, -- Tbr2	↓ DCX, ↓ BrdU/NeuN	↑ CC3
Vetreno and Crews [2]	Wistar rat	Adult	↓ Ki67	↓ DCX (P56 - P220)	↑ CC3
Vetreno et al [27]	Wistar rat	Adolescent	↓ Ki67, ↓ Nestin	↓ DCX	↑ CC3
		Adult	↓ Ki67, ↓ Nestin	↓ DCX	↑ CC3
Broadwater et al [29]	SD rat	Adult	-- Ki67	↓ DCX	↑ CC3
Intraperitoneal administration
Sakharkar et al [28]	SD rat	Adult	↓ Ki67	↓ DCX	Not assessed
Vapor administration
Ehlers et al [30]	Wistar rat	Adolescent	↓ Ki67	↓ DCX	↑ CC3, ↑ FJB
		Adult	-- Ki67	↓ DCX	↑ CC3
Self-administration
Briones & Woods [12]	SD rat	Adolescent	Not assessed	↓ DCX	Not assessed
Taffe et al [33]	Macaque primate	Adult	↓ Ki67	↓ PSA-NCAM	-- CC3, ↑ FJB

BrdU, 5′-bromo-2-deoxyuridine; CC3, cleaved caspase-3; DCX, doublecortin; FJB, FluoroJade B; NeuN, neuronal nuclei; P, postnatal day; PSA-NCAM, polysialylated-neural cell adhesion molecule; SD rats, Sprague-Dawley rat; Sox2, sex determining regions Y-box 2; Tbr2, T-box transcription 2. -- indicates no change, relative to CONs.

The loss of hippocampal neurogenesis following AIE exposure likely involves a combination of decreased NPC proliferation, disrupted progression of NPCs through the neurogenesis cell cycle, and increased cell death. Consistent with this hypothesis, our laboratory and others find AIE treatment reduces expression of Ki67, an endogenous nuclear protein expressed in dividing cells [40], in the rat dentate gyrus [2,17,35,36,38]. This reduction in progenitor cell proliferation is accompanied by transient reductions in expression of Sox2, which is expressed in non-radial and horizontal type-1 cells, and Tbr2, which is expressed by type-2a and -2b intermediate cells [17]. Interestingly, the reduction in Sox2 and Tbr2 that is observed immediately following the conclusion of AIE emulates the age-associated reductions over time observed in control subjects. Similar reductions of NPC proliferation and development markers are observed in nonhuman primate models of adolescent binge ethanol exposure [11] as well as the postmortem human AUD dentate gyrus [39]. Furthermore, increased expression of cell death markers, including cleaved caspase-3 and FluoroJade B, are observed in the dentate gyrus of AIE-treated rodents and in nonhuman primate models of adolescent binge drinking [2,11,17,35,37,38] that may contribute to reduced pools of NPCs. Thus, the loss of NPCs coupled with increased cell death likely contribute to the observed persistent reduction of hippocampal neurogenesis following AIE exposure.

Reductions of Neurogenesis Following AIE are Associated with Cognitive Deficits

Hippocampal neurogenesis has been implicated in mediating cognitive function. Adolescent binge drinking-induced persistent reductions of neurogenesis may contribute to hippocampal-related cognitive deficits. For instance, human adolescents with heavy episodic drinking history present with reduced hippocampal task activation and impairments on a simple verbal learning task [41]. Similar cognitive impairments are observed in rodent AIE models, with several of these studies providing associations between reductions of hippocampal neurogenesis and impaired behavioral performance [2,38]. For instance, the reduction of adult hippocampal neurogenesis following AIE is associated with persistently increased disinhibitory behavior on an open-field conflict test [38]. Similarly, our laboratory found that expression of DCX-immunopositive cells in the dorsal dentate gyrus is correlated with performance on a novel object recognition task, such that lower expression levels of DCX were associated with diminished object recognition memory [2]. On the same behavioral test, decreased neurogenesis in the ventral hippocampus correlated with increased thigmotaxia during open field habituation, suggesting anxiety-like behavior [42]. While the exact role of reduced neurogenesis in AIE-induced cognitive deficits remains to be fully elucidated, reductions of hippocampal neurogenesis in a large number of studies are associated with an increased risk for the development of adult psychopathology (e.g., depression) and cognitive dysfunction [see e.g., 43,44].

Neuroimmune Contributions to AIE-induced Loss of Neurogenesis

While the mechanism(s) underlying the persistent loss of hippocampal neurogenesis remain to be fully elucidated, converging lines of evidence implicate activation of the neuroimmune system (i.e., innate immune agonists and their receptors) as a contributing factor in the persistent AIE-induced loss of hippocampal neurogenesis. For instance, AIE treatment increases expression of proinflammatory Toll-like receptors (TLRs), the endogenous cytokine-like TLR ligand high-mobility group box 1 (HMGB1), and several downstream proinflammatory cytokines (e.g., tumor necrosis factor alpha [TNFα], monocyte chemoattractant protein-1 [MCP-1]) in the adult hippocampus [2,35]. In addition, AIE treatment increases phosphorylation (activation) of the canonical neuroimmune gene transcription factor nuclear factor kappa-light-chain-enhancer of activated B cells p65 (pNF-κB p65) in the adolescent hippocampus that persists into adulthood [35]. Similar to the preclinical AIE models, induction of neuroimmune signaling is also observed in post-mortem human AUD hippocampus, with the greatest magnitude observed in individuals with an earlier adolescent age of drinking onset [35,45,46]. Expression of pNF-κB p65 in the AIE model is also positively correlated with expression of cleaved caspase-3, suggesting that neuroimmune activation contributes to increased cell death [35]. The persistent induction of neuroimmune signaling in the adult rat hippocampus is accompanied by microgliogenesis and partial activation of microglia [17,47]. Adolescent intermittent ethanol exposure also increases Iba-1+IR microglial expression of pNF-κB p65 in the adult hippocampus [35]. It also increases expression of β2-microglobulin (β2M), which is a major histocompatibility complex I protein expressed on microglia known to inhibit hippocampal neurogenesis [48] in adulthood (P95), but not late adolescence (P57). These data suggest that AIE causes a progressive induction of neuroimmune signaling across aging [17]. Thus, interventions aimed at targeting neuroimmune signaling may be an effective approach to prevent AIE-induced neuroimmune activation and loss of hippocampal neurogenesis.

Consistent with this hypothesis, blockade of AIE-induced neuroimmune signaling prevents the loss of hippocampal neurogenesis. For example, our laboratory found that treatment with the non-steroidal anti-inflammatory drug indomethacin during AIE blocked the increase of pNF-κB p65 in the dentate gyrus and prevented the loss of neurogenesis as well as increased expression of cleaved caspase-3 [35] supporting a role for neuroimmune signaling in the AIE-induced loss of neurogenesis. This hypothesis is further supported by our laboratory’s findings that lipopolysaccharide (LPS) induction of neuroimmune signaling in hippocampus decreases neurogenesis, and increases expression of pNF-κB p65 and the cell death marker cleaved caspase-3, mimicking the AIE-induced loss of neurogenesis [35]. Blockade of LPS-induced neuroimmune signaling with indomethacin, like AIE, blocked the loss of DCX+IR cells in the rat hippocampus [49]. Similarly, treatment with the histone deacetylase (HDAC) inhibitor trichostatin A (TSA), which blocks neuroimmune activation and promotes BDNF expression, reversed AIE-induced loss of neurogenesis in rat hippocampus [36] and LPS-induced neuroimmune signaling (e.g., TNFα, MCP-1) in mouse hippocampus [50], suggestive of epigenetic mechanisms regulating neuroimmune induction and AIE-induce loss of neurogenesis.

Culture studies provide further insight into the role of neuroimmune signaling in the AIE-induced loss of hippocampal neurogenesis. For example, Monje and colleagues [49] report that application of recombinant IL-6 and TNFα to in vitro hippocampal precursor cells reduced neurogenesis. Studies from our laboratory employing an ex vivo model of organotypic hippocampal-entorhinal cortex (HEC) brain slice culture replicated the in vivo ethanol-induced loss of neurogenesis and provide further evidence that neuroimmune signaling contributes to the loss of neurogenesis. For instance, inhibition of NF-κB target genes in HEC slice culture prevents ethanol-induced loss of neurogenesis [45]. Taken together, accumulating evidence suggests that AIE-induced induction of neuroimmune signaling and lasting partial microglial activation contribute to the generation of positive feedback loops of neuroimmune activation that reduce NPC cell proliferation and increase cell death, contributing to the persistent loss of neurogenesis in adulthood [30].

Shift in Neuroimmune/Neurotrophic Balance through Epigenetic Mechanisms Contributes to AIE-induced Loss of Neurogenesis

While strong evidence supports neuroimmune signaling as contributing to the AIE-induced loss of neurogenesis that persists into adulthood, the mechanisms are complex. Interestingly, AIE treatment also diminishes expression of neurotrophins critical for neurogenesis. One such neurotrophin, BDNF, has been implicated in regulating synaptic plasticity and hippocampal neurogenesis [51,52]. BDNF is a CREB target gene that plays a critical role in neurogenesis, neuroplasticity, and cognition. Reductions of BDNF may contribute to the lasting reductions of neurogenesis observed following AIE. Indeed, AIE exposure decreases protein levels of BDNF as well as the BDNF receptor tropomyosin receptor B (TrkB) in the hippocampus that parallels the loss of neurogenesis [20,36]. Interestingly, administration of the BDNF receptor TrkB agonist 7,8-dihydroflavone during AIE recovered the reduction of BDNF and TrkB expression as well as the loss of hippocampal NPC proliferation and neurogenesis [20], implicating reductions of BDNF trophic support in the loss of neurogenesis. Although neuroimmune markers were not assessed in these studies, neuroimmune gene expression is likely inversely related to BDNF expression, with both impacting hippocampal neurogenesis.

A recent study linked epigenetic mechanisms to long-term reductions of hippocampal BDNF levels. Epigenetics involves histone acetylation and histone or DNA methylation that cause activation or repression of gene transcription without changing the underlying DNA sequence, resulting in an environmentally triggered change in gene expression leading to a relatively stable phenotype [53,54]. Specifically, the lasting AIE-induced reduction of hippocampal BDNF expression was accompanied by decreased BDNF exon IV-specific histone 3 lysine 9 (H3K9) acetylation that paralleled the loss of neurogenesis. The reduction of BDNF exon IV-specific H3K9 acetylation was accompanied by increased nuclear HDAC activity as well as reductions of the cAMP responsive element binding protein (CREB)-binding protein (CBP). CBP has been found to regulate hippocampal neurogenesis and histone acetylation [55,56], and decreased levels of CBP are correlated with reductions of H3K9 acetylation and BDNF expression in the adult hippocampus following AIE [36]. The AIE-induced loss of hippocampal BDNF appears to be associated with suppression of H3K9 acetylation, as treatment with the HDAC inhibitor TSA in adulthood restored the H3K9 acetylation at the BDNF exon IV promoter and recovered the AIE loss of hippocampal neurogenesis [36]. Further, voluntary wheel running, which has been shown to reduce proinflammatory signaling and increase BDNF levels in the hippocampus [52,57], prevented the AIE-induced loss of cell proliferation and neurogenesis as well as the increased cleaved caspase-3 cell death and pNF-κB p65 in adult dentate gyrus [35]. The persistent induction of neuroimmune signaling, coupled with the lasting reductions of BDNF-neurotrophic support, are consistent with an AIE-induced shift in the neuroimmune/neurotrophic balance that may contribute to the persistent loss of hippocampal neurogenesis.

Ex vivo HEC culture studies further support an ethanol-induced shift in the neuroimmune/neurotrophic balance through dysregulation of the reciprocal relationship between CREB protein transcriptional activation of BDNF and NF-κB-induced activation of proinflammatory gene expression [58]. Using the HEC model, Zou and Crews [59] found that ethanol treatment dose-dependently decreased CREB DNA binding and increased NF-κB DNA binding, which provide an index of CREB and NF-κB activation, respectively. In the same study, application of the proinflammatory cytokine TNFα to HEC increased NF-κB activation and reduced BDNF mRNA levels. Importantly, administration of rolipram, a phosphodiesterase 4 inhibitor known to increase CREB activation [60], and butylated hydroxytoluene, an anti-oxidant that blocks NF-κB activation [61,62], reversed ethanol-induced inhibition of neurogenesis. Taken together, these data suggest that AIE may disrupt the neuroimmune/neurotrophic balance in the hippocampus through epigenetic mechanisms resulting in lasting suppression of hippocampal neurogenesis and hippocampal-dependent cognitive dysfunction.

Conclusions

Preclinical studies of AIE exposure reveal persistent reductions of NPC proliferation and hippocampal neurogenesis, induction of neuroimmune signaling, and a decline in cognitive function. Moreover, whereas adult ethanol exposure causes a transient reduction of neurogenesis that recovers after a period of abstinence, AIE exposure causes deficits in hippocampal neurogenesis that persist well into adulthood despite later abstinence from ethanol exposure. These findings, coupled with evidence that AIE also alters epigenetic regulation of neurotrophic factors (i.e., BDNF), suggest lasting shifts in the neuroimmune/neurotrophic-BDNF balance (see Fig. 1). However, more work is needed to elucidate the mechanisms linking AIE-induced dysregulation of the neuroimmune/neurotrophic balance to the persistent loss of hippocampal neurogenesis. For instance, there is ample evidence suggesting relationships between AIE induction of neuroimmune signaling, loss of trophic support, and neurodegeneration, but it is unclear whether these events occur simultaneously or sequentially. Elucidating the order of events would aid in providing more specific targets for the development of treatment strategies to reverse AIE-induced deficits. In addition, the majority of preclinical work to date has utilized male subjects, and it is critical to verify that similar neurodevelopmental shifts due to AIE exposure occur in females. Lastly, while ample evidence links neurogenesis to cognitive and emotive function, it is unclear whether the loss and restoration of hippocampal neurogenesis following AIE exposure is directly responsible for shifts in hippocampal-mediated behaviors or whether restoration of neurogenesis reflects a general state of brain health.

Figure 1. Adolescent binge ethanol-induced shift in neuroimmune/neurotrophic balance in the adult hippocampus contributes to persistent loss of neurogenesis.

Adolescent binge ethanol exposure causes lasting reductions of hippocampal neurogenesis in rodents due in part to decreased neuroprogenitor cell (NPC) proliferation, disrupted progression of NPCs through the neurogenesis cell cycle, and increased cell death. Accumulating evidence implicates increased neuroimmune signaling in the adolescent binge ethanol-induced loss of neurogenesis. Adolescent binge ethanol exposure upregulates proinflammatory Toll-like receptors (TLRs) and the endogenous cytokine-like TLR ligand high-mobility group box 1 (HMGB1) in the hippocampus, resulting in increased phosphorylation (activation) of the canonical neuroimmune gene transcription factor nuclear factor kappa-light-chain-enhancer of activated B cells p65 (pNF-κB p65) [2,35]. This in turn results in induction of neuroimmune signaling molecules, including tumor necrosis factor alpha (TNFα), that activate positive loops of amplification that persist into adulthood [63]. Further, AIE increases microgliogenesis and lasting partial activation of microglia in the hippocampus following adolescent binge ethanol exposure [17,47]. The lasting upregulation of neuroimmune signaling is accompanied by diminished hippocampal levels of the pro-neurogenic trophic factor brain-derived neurotrophic factor (BDNF) in association with reduced BDNF exon IV-specific histone 3 lysine 9 (H3K9) acetylation [20,36]. Interestingly, exposure to aerobic exercise, anti-inflammatory drugs, histone deacetylase inhibitors, and BDNF receptor TrkB agonists all recover the adolescent binge ethanol-induced loss of hippocampal neurogenesis [20,35,36], suggesting that a shift in the neuroimmune/neurotrophic balance may contribute to the persistent loss of hippocampal neurogenesis.

As alcohol is one of the most commonly abused drugs during adolescence and adolescents engage in episodic binge drinking, identification of treatment strategies for alcohol-induced neuropathology is critical. Fortunately, preclinical studies suggest that interventions aimed at reducing hippocampal neuroimmune induction and loss of BDNF-neurotrophic support can prevent and/or recover the loss of neurogenesis following AIE. As adolescent binge drinking is predictive of developing an alcohol use disorder later in life, increasing educational efforts that focus on the long-term effects of alcohol consumption during adolescence with the aim of decreasing binge drinking is a critical goal for the scientific field.