Impact of Adolescent Ethanol Binge on Serotonin Signaling and Pain Sensitivity Post-Withdrawal

Abstract

Adolescence is a critical neurodevelopmental period characterized by heightened neuroplasticity. While the acute effects of binge ethanol (EtOH) consumption are documented, its long-term impact on both pain sensitivity and microglial activation during adolescence remains unclear. Given serotonin’s (5-HT) known involvement in pain processing and sensitivity to EtOH, this study examined the effects of adolescent EtOH binge on microglia-induced neuroinflammation in serotonergic nuclei, downstream 5-HT signaling, and pain sensitivity at different time points after EtOH withdrawal.

Adolescent male C57BL/6J mice received triweekly oral gavage of 20% EtOH or water for 4 weeks and were assessed after 24 hours and 3 weeks post-withdrawal. We used immunohistochemistry to assess neuroinflammation in the dorsal raphe, median raphe, and raphe magnus by labeling 5-HT, CD68, and P2Y12. Further analyses examined downstream signaling via 5-HT and serotonin transporter (SERT) expression in the nucleus accumbens, anterior cingulate cortex, thalamus, amygdala, hypothalamus, and raphe magnus. Pain sensitivity was then assessed using the Hargreaves test.

EtOH exposure led to widespread serotonergic and neuroinflammatory changes. Significant increases in microglia-induced neuroinflammation were observed in the dorsal raphe nucleus, median raphe nucleus, and raphe magnus nucleus after both 24 hours and 3 weeks post-withdrawal, along with significant deficits in 5-HT. Similar 5-HT deficits were observed in downstream regions—notably in the anterior cingulate cortex, thalamus, amygdala, and hypothalamus—at varying time points post-withdrawal. EtOH-exposed mice also showed lasting hyperalgesia at both 24 hours and 3 weeks post-withdrawal that persisted for up to 9 weeks. These results suggest that persistent hyperalgesia following adolescent EtOH binge may be driven by changes in serotonergic function and microglial activation.

Graphical Abstract

Introduction

Clinical relevance

Adolescence represents a critical period of development characterized by heightened neuroplasticity. During this time, significant exposure to substances such as ethanol (EtOH) can contribute to lasting effects on the brain and behavior (Broadwater et al., 2011; Peñasco et al., 2019). In numerous preclinical studies, gavage administration of EtOH has been used as a controlled and effective method to examine EtOH’s impacts on behavior and neurobiology, including measures of pain sensitivity, microglial-induced neuroinflammation, and serotonin (5-HT) signaling in various brain regions (Carson & Pruett, 1996; Khan et al., 2023; Stuckey & Berry, 1984). Binge drinking of ethanol-based products is particularly concerning due to its prevalence among adolescent populations and its associations with increased risks of adverse outcomes, such as liver damage and alcohol dependence, which may persist into adulthood (Carson & Pruett, 1996; Chung et al., 2018; Crews et al., 2016; Peñasco et al., 2019). Despite a decline in prevalence over the last decade, binge drinking continues to be a public health concern (SAMHSA, 2019–2023), with a high proportion of youth still engaging in the activity (Chung et al., 2018; Patrick et al., 2013).

Neuroinflammation and Microglial Activation

A major mechanism through which adolescent EtOH exposure may exert lasting effects is through its capacity to drive neuroinflammatory responses. In particular, EtOH-induced neuroinflammation has been linked to activation of microglia, the resident immune cells of the central nervous system (CNS). Neuroinflammation resulting from adolescent ethanol exposure may be linked to microglial activation in the dorsal raphe nucleus (DRN) and other serotonergic regions, with increased expression of the lysosomal marker CD68 (Khan et al., 2023). Elevated CD68 expression is associated with increased microglia lysosomal activity and has been observed in mouse models for alcohol use disorder (AUD) (Alfonso-Loeches et al., 2016; Lowe et al., 2020; Marshall et al., 2013; Socodato et al., 2020; Walter & Crews, 2017).

Microglial activation is also known to be associated with decreased expression of the P2Y12 purinergic receptor, which reflects a transition from a surveillant to an activated, pro-inflammatory state (Haynes et al., 2006). Upon activation, microglia secrete proinflammatory cytokines and chemokines, including numerous interferons, interleukins (ILs), and tumor necrosis factors (TNFs) (Alfonso-Loeches et al., 2016; Crews et al., 2021; Fernandez-Lizarbe et al., 2009; Lowe et al., 2018, 2020; Marshall et al., 2013; Pascual et al., 2015; Qin et al., 2008; Qin & Crews, 2012; Socodato et al., 2020; Walter & Crews, 2017). This cascade has been shown to be initiated in part by the activation of Toll-like receptor 4 (TLR4), a pattern-recognition receptor that mediates EtOH-induced cytokine release in the CNS (Fernandez-Lizarbe et al., 2009, 2013; Holloway et al., 2023).

Since serotonergic nuclei such as the DRN are vulnerable to neuroinflammation, this raises the possibility that EtOH-induced immune activation may disrupt 5-HT signaling and alter pain sensitivity.

Alcohol Use, Serotonin Signaling, and Pain Sensitivity

Previous EtOH studies in rodents have demonstrated differential pain sensitivity following chronic intermittent EtOH exposure, with the level of hyperalgesia observed to be dependent on blood alcohol concentration (Brandner et al., 2023; Gatch, 2009).

The serotonergic system is a key modulator of this pain sensitivity—influencing both central and peripheral nociceptive pathways (Wang & Nakai, 1994; Wang et al., 2013). Peripherally, 5-HT influences nociceptor sensitivity via acid-sensing ion channels, which are activated under the acidic conditions that arise from injury and inflammation (Deval et al., 2008; Mamet et al., 2002; Morgan et al., 2020; Wang et al., 2013). Additionally, the 5-HT released from peripheral nerve endings helps facilitate neurotransmitter release in the dorsal horn, allowing for pain signal amplification during states of inflammation (Sommer, 2004).

However, pain is more broadly, and primarily, regulated by the CNS through serotonergic projections originating from raphe nuclei in the brainstem (Gaspar et al., 2003). In particular, the DRN, which is often referred to as the brain’s major “pain inhibitory nucleus” (Wang & Nakai, 1994). The DRN houses the majority of the brain’s serotonergic cell bodies (Charnay & Leger, 2010). From the raphe nuclei, 5-HT acts on various downstream brain regions involved in pain processing—including the nucleus accumbens, anterior cingulate cortex, thalamus, amygdala, hypothalamus, and raphe magnus—all of which are key sites of serotonergic modulation (Apkarian et al., 2005; Bliss et al., 2016; Bocchio et al., 2016; Bourne et al., 2014; Brown & Molliver, 2000; Bushnell et al., 2013; Hanley & Van de Kar, 2003; Leknes & Tracey, 2008; Liang et al., 2013; Monckton & McCormick, 2002; Pan et al., 1993; Price, 2000; Tian et al., 2017; Zhuo, 2008, 2014). Because many of these downstream projection sites are innervated by 5-HT neurons that may be compromised by EtOH-induced inflammation, it is important to assess whether serotonergic signaling is disrupted in these areas following EtOH exposure.

Disruptions to serotonergic circuitry have been linked with EtOH exposure in previous rodent studies (Deehan et al., 2016; Luessen et al., 2019). EtOH dependence, for instance, has been associated with alterations in 5-HT signaling within the DRN and median raphe nucleus (MRN) in mice (Castle & Flanigan, 2024; Kelaï et al., 2008), impairing their normal regulatory roles in pain processing. Adolescent EtOH exposure has been shown to induce 5-HT deficits in the DRN and other brain regions associated with pain processing, notably in the thalamus, anterior cingulate cortex, and amygdala (Apkarian et al., 2005; Bushnell et al., 2013; Leknes & Tracey, 2008; Liang et al., 2013). 5-HT deficits, or loss of serotonergic neurons outright, have been linked to both increased neuroinflammation and heightened pain sensitivity (Khan et al., 2023; Zhao et al., 2007). Genetic animal models have provided additional evidence regarding the role of 5-HT dysfunction in pain processing alterations, with mice possessing serotonin transporter knockout (SERT^−/−) observed to exhibit significant decreases in mechanical and thermal sensitivity (Perrin & Noristani, 2019; Zhao et al., 2007). Findings like these reinforce the notion that changes in 5-HT tone may underlie the persistent hyperalgesia observed following EtOH exposure.

Goal

This study investigates the long-term consequences of adolescent EtOH binge exposure, focusing on pain sensitivity and its potential mediation by serotonergic signaling and microglial activation. To test this, we first used immunostaining to assess microglial activation within major serotonergic nuclei. We then quantified 5-HT and SERT localization in several downstream regions that receive serotonergic input. Lastly, we assessed behavioral changes in pain sensitivity using the Hargreaves thermal nociception test. By examining region-specific changes in serotonergic markers and microglial activation state markers in pain-related brain regions, we aim to elucidate the interplay between neuroinflammation, serotonergic dysfunction, and persistent alterations in pain sensitivity.

Materials and Methods

Animals

All procedures were performed in accordance with the ethical guidelines research regarding animals and approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Iowa. Male C57BL/6J mice (catalog #000664, Jackson Labs, Bar Harbor, ME, USA; RRID:IMSR_JAX:000664) were purchased for arrival at postnatal day 21 (P21) and aged to P25 before use. A total of 60 mice were used: 20 C57BL/6J mice for Hargreaves pain assessment and 40 for immunohistochemistry (IHC) experiments.

All mice were housed in a temperature- and humidity-controlled AAALAC-approved vivarium at the University of Iowa. Vivaria were on a standard 12-h/12-h dark/light (reverse) cycle to abide by institutional requirements. All mice were housed in conventional-style rodent cages containing shredded paper bedding and had ad libitum access to water and standard chow composed of 14% kcal fat, 60% kcal carbohydrate, and 26% kcal protein (catalog #5P76l Land O’Lakes, Arden Hills, MN, USA).

Mice were assessed experimentally in the following manner, following the completion of 4 weeks oral gavage, 3 days per week. Cohort 1: Brains harvested for IHC 24 hours post-withdrawal. Cohort 2: Brains harvested for IHC 3 weeks post-withdrawal. Cohort 3: Hargreaves pain assessment performed 24 hours, 3 weeks, 6 weeks, and 9 weeks post-withdrawal (Figure 1).

Fig. 1. Experimental timeline and cohort assignments for adolescent binge ethanol exposure and downstream analyses.

Three cohorts of mice underwent triweekly oral gavage of EtOH or water from postnatal day 25 (P25) to P56. (Cohort 1) At 24 hours post-withdrawal, brain tissue from mice in cohort 1 was collected for immunohistochemistry (IHC) analysis. (Cohort 2) Same as (Cohort 1), brain tissue was collected 3 weeks post-withdrawal. (Cohort 3) Mice in cohort 3 were assessed for pain sensitivity using the Hargreaves test at 24 hours, 3 weeks, 6 weeks, and 9 weeks post-withdrawal

Ethanol Gavage

20% EtOH solutions (%v/v) were prepared from tap water and 95% ethyl alcohol (catalog #2801TP, Decon Labs). Starting at age P25, mice received 4 weeks of either EtOH or water gavage triweekly (i.e., Monday, Wednesday, Friday). 10 mL/kg of the 20% EtOH solution was administered to achieve a dose of 2 mg/kg for experimental groups; control groups correspondingly received 10 mL/kg water. Mice were first restrained in a vertical position by pulling up the loose skin on the back of the neck to extend the head and straighten the spine. The gavage needle was then inserted into the mouth, between the incisors and molars, and guided along the roof of the mouth. Using the needle as a lever, the head was extended back to achieve vertical alignment of the neck and esophagus. The needle was then inserted into the esophagus and the solution was administered. Once administered, the needle was gently removed following the same angle as insertion. Animals were then returned to their original home cages, that were maintained at 38 °C on a heating pad to avoid hypothermia.

Tissue Preparation

Following completion of their respective post-withdrawal period, mice from cohorts 1 and 2 were transcardially perfused with PBS, followed by 4% paraformaldehyde prior to brain extraction. Brains were cryoprotected using a sucrose gradient before being embedded in OCT and stored at −80 °C. Before cryosectioning, all brains were equilibrated at −20 °C for one full day. Brains were cryosectioned coronally at 25 μm (CM3050S, Leica, Germany).

5-HT Neuroinflammation Immunohistochemistry and Image Analysis

Tissue sections from cohorts 1 and 2 that contained the DRN, MRN, and raphe magnus (RMg) were selected—specifically one coronal section at each rostro-caudal level for the DRN and MRN (rostral, mid, and caudal), and one section for the RMg—yielding a total of four raphe slices per animal. These were then washed in PBS while free-floating. Sections were then permeabilized in a 0.5% TritonX-100 PBS solution for 30 minutes at room temperature, followed by another wash with PBS. Sections were then blocked for 60 minutes at room temperature in a buffer containing 0.1% TritonX-100 and 10% normal donkey solution in PBS. Sections were incubated overnight at 4 °C in the previous blocking buffer with the following primary antibodies: goat anti-5-HT (ImmunoStar, catalog #20079, lot #2203001, 1:2000), rat anti-CD68 (Abcam, catalog #AB53444, lot #1080382–14, 1:500), and rabbit anti-P2Y12 (AnaSpec, catalog #AS-55043A, lot #UB1701, 1:2000). The sections were then washed in PBS following overnight incubation, and treated with the following secondary antibodies in PBS for 2 hours at room temperature: donkey anti-goat/Alexa-488 (Jackson, catalog #705–545-147, lot #155707, 1:500), donkey anti-rat/Alexa-647 (Abcam, catalog #AB150155, lot #GR3420807–1, 1:500), and donkey anti-rabbit/Alexa-555 (Invitrogen, catalog #A31572, lot #2831376, 1:500). After washing with PBS, the sections were mounted on HistoBond+ glass slides (Marienfeld Superior, Lauda-Königshofen, Germany) in PBS and cover-slipped with Vectashield Vibrance mounting medium (catalog #H-1700–10, Vector Laboratories, Newark, CA, USA).

The slides were later imaged using an Olympus VS200 slide scanner (Evident Scientific, Waltham, MA, USA) at 20X. Images were then analyzed using QuPath v0.5.1 (University of Edinburgh, Edinburgh, United Kingdom). The 5-HT stain was used in tandem with Paxinos and Franklin’s mouse brain reference atlas (Paxinos and Franklin, 2001) to annotate the DRN, MRN, and RMg regions. Within these regions of interest (ROIs), cell counts and percent immunoreactive area (%IR) were determined with pixel classifiers for 5-HT, CD68, and P2Y12 markers. The estimated percentage of activated microglia within each ROI was also determined—first by calculating the average microglial size based on immunoreactive area for P2Y12 for an individual microglia. This average was then used to approximate the total number of microglia based on the total P2Y12 immunoreactive area within a given ROI. Microglia co-expressing CD68 were then manually counted, and the percentage of activated microglia was calculated using the proportion of CD68+/P2Y12+ cells relative to the estimated total microglia population.

5-HT Signaling Immunohistochemistry and Image Analysis

Tissue sections were selected from cohorts 1 and 2 that contained specific downstream regions. The following coronal sections were collected from each animal: one section for the anterior cingulate cortex and nucleus accumbens, one section for the thalamus, one section containing both the amygdala and hypothalamus, and one containing the RMg, for a total of 4 sections. The free-floating sections containing these regions were then washed in PBS and then blocked for 60 minutes in a buffer consisting of 10% NDS, 0.5% TritonX-100, and 0.2% Tween-20 in PBS. The sections were then rinsed with PBS and incubated at 4 °C overnight in buffer consisting of 2% NDS, 0.5% TritonX-100, and 0.2% Tween-20 in PBS, as well as the following primary antibodies: goat anti-5-HT (ImmunoStar, catalog #20079, lot #2203001, 1:500), rabbit anti-SERT (ImmunoStar, catalog #24330, lot #1620001, 1:500), and guinea pig anti-NeuN (Millipore, catalog #ABN90, lot #4155262, 1:1000). The free-floating sections were then incubated with the following secondary antibodies: donkey anti-guinea pig/Alexa-488 (Jackson, catalog #706–545-148, lot #160322, 1:500), donkey anti-rabbit/Cy3 (Jackson, catalog #711–165-152, lot #160467, 1:500), and donkey anti-goat/Alexa-647 (Jackson, catalog #705–605-003, lot #160139, 1:500) secondary antibodies for 90’ at room temperature. The sections were mounted on HistoBond+ glass slides (Marienfeld Superior, Lauda-Königshofen, Germany) in PBS before being cover-slipped with Vectashield Vibrance mounting medium (catalog #H-1700–10, Vector Laboratories, Newark, CA, USA).

The slides were later imaged using an Olympus VS200 slide scanner (Evident Scientific, Waltham, MA, USA) at 20X. %IR was determined with QuPath (University of Edinburgh, Edinburgh, United Kingdom) by creating pixel classifiers for 5-HT and SERT within each of the ROIs—drawn according to the shape of the region based on Paxinos and Franklin’s mouse brain reference atlas (Paxinos and Franklin, 2001).

Hargreaves Pain Assessment

Thermal sensitivity was assessed via the Hargreaves test, as described by Hargreaves et al., 1988. Mice in cohort 3 were first acclimated on the Hargreaves test apparatus for 3 hours on 2 consecutive days prior to experimenting. The test apparatus was situated inside a clear acrylic box (100 × 100 × 150 mm) and consisted of a heated base (Model 400; IITC Life Science, Woodland Hills, CA, USA) with its temperature maintained at a constant 30 °C. A heat-generation light beam was then focused onto the plantar surface of each hind paw for each mouse. The time taken to elicit a paw withdrawal response was measured 3 times for each paw. The results were averaged to determine the latency of paw withdrawal. This was done at 24 hours, 3 weeks, 6 weeks, and 9 weeks post-withdrawal.

Statistical Analyses

All data analyses and statistical tests were performed using Prism version 10 (GraphPad Software, Inc., La Jolla, CA, USA; RRID:SCR_002798). Outliers were first determined using a ROUT test (Q = 1 %). Data were then checked for normal distribution and homogeneity of variance using the Fmax test. Where data violated homogeneity of variance, a non-parametric analysis was used. 1-dimensional plots (interleaved bars) report individual data points with mean and standard deviation. 2-dimensional plots (Hargreaves data) report the average of all data points per experimental condition, plungers representing standard error of the mean. Statistical difference was defined at α < 0.05, and all statistical tests used a two-tailed hypothesis. Exact p values, statistical tests, and specific variables required for independent replication are found in the supplemental Statistical Summary Table document.

Results

Neuroinflammatory and Serotonergic Changes in the Dorsal Raphe Nucleus

Given the known effects of EtOH on the modulation of serotonin (5-HT) neurons (Lowery-Gionta et al., 2015; Marcinkiewcz, 2015), we first assessed percent immunoreactive area (%IR) for 5-HT. Compared to water-treated controls, EtOH-exposed animals exhibited widespread reductions in 5-HT %IR within the dorsal raphe nucleus (DRN). Significant decreases in 5-HT %IR ranging from 28–50% were observed across the DRN at both 24 hours (rostral **p = 0.0070; caudal **p = 0.0098; aggregate ***p = 0.0002) and 3 weeks post-withdrawal (rostral *p = 0.0289; caudal *p = 0.0400; aggregate **p = 0.0017) (Fig. 2A–C, E). No significant changes were found in 5-HT neuron density at either post-withdrawal time point (Fig. 2D, F).

Fig. 2. Serotonergic and neuroinflammatory changes in the dorsal raphe nucleus (DRN) following adolescent EtOH exposure.

(A–B) Representative images of 5-HT immunostaining in the DRN at 24 hours (A) and 3 weeks (B) post-withdrawal. (C–D) Graphs for 5-HT percent-immunoreactivity (%IR) (C) and 5-HT neuron density (D) across rostral, caudal, and aggregate DRN at 24 hours. (E–F) 3-week post-withdrawal counterparts for 5-HT %IR (E) and 5-HT neuron density (F). (G–H) Representative images of P2Y12 and CD68 staining at 24 hours (G) and 3 weeks (H) post-withdrawal. (I–J) Graphs for P2Y12 (I) and CD68 (J) %IR at 24 hours post-withdrawal. (K–L) Average microglial size (K) and percent microglial activation (L) at 24 hours post-withdrawal. (M–P) 3-week post-withdrawal counterparts for P2Y12 %IR (M), CD68 %IR (N), average microglial size (O) and percent microglial activation (P). *p < 0.05, **p < 0.01, ***p < 0.001, ***p < 0.0001, ns = not significant. Interleaved bar graphs report individual data points with mean and standard deviation. “Agg. DRN” refers to pooled data from rostral and caudal DRN sections to better capture nucleus-wide effects.

Compared to controls, adolescent EtOH exposure also induced significant changes in expression of microglial state markers in the DRN at 24 hours post-withdrawal. Specifically, significant reductions in P2Y12 percent immunoreactive area (%IR) were observed in the rostral (**p = 0.0010), caudal (*p = 0.0435), and aggregate DRN (****p < 0.0001), with no significant changes observed at 3 weeks post-withdrawal (Fig. 2G–I, M). Average microglia size, as determined by P2Y12 immunoreactive area for singular microglia, was significantly decreased at 24 hours post-withdrawal (rostral *p = 0.0147; caudal *p = 0.0489; aggregate ***p = 0.0001), with some persistence at 3 weeks post-withdrawal in the caudal (**p = 0.0092) and aggregate (****p < 0.0001) DRN (Fig. 2K, O). CD68 %IR showed significant increases in the rostral (*p = 0.0468), caudal (*p = 0.0359), and aggregate DRN (**p = 0.0023) at 24 hours post-withdrawal, but remained mostly unchanged at 3 weeks post-withdrawal (Figure 2G–H, J, N). Microglial activation, which we defined as the percentage of P2Y12+ cells that were dually CD68+, was significantly elevated in the DRN of EtOH-exposed animals at both 24 hours (*p = 0.0329 rostral; *p = 0.0189 caudal; ***p = 0.0002 aggregate) and 3 weeks post-withdrawal (**p = 0.0078 caudal; **p = 0.0043 aggregate) (Fig. 2L, P); no significant changes were seen in the rostral DRN at 3 weeks post-withdrawal. Exact p values and statistical tests for each variable can be found in supplementary Table S1.

Neuroinflammatory and Serotonergic Changes in the Median Raphe Nucleus

Similar to the DRN, EtOH exposure resulted in significant reductions in 5-HT %IR in the median raphe nucleus (MRN). These deficits were seen in the rostral (*p = 0.0244) and aggregate (**p = 0.0056) MRN at 24 hours post-withdrawal and in caudal (**p = 0.0079) and aggregate (*p = 0.0235) MRN at 3 weeks post-withdrawal (Fig. 3A–C, E). As we observed in the DRN, there were no significant changes in 5-HT neuron density for either time point in the MRN (Fig. 3D, F).

Fig. 3. Serotonergic and neuroinflammatory changes in the median raphe nucleus (MRN) following adolescent EtOH exposure.

(A–B) Representative images of 5-HT immunostaining in the MRN at 24 hours (A) and 3 weeks (B) post-withdrawal. (C–D) Graphs for 5-HT percent-immunoreactive area (%IR) (C) and 5-HT neuron density (D) across rostral, caudal, and aggregate MRN at 24 hours post-withdrawal. (E–F) 3-week post-withdrawal counterparts for 5-HT %IR (E) and 5-HT neuron density (F). (G–H) Representative images of P2Y12 and CD68 staining at 24 hours (G) and 3 weeks (H) post-withdrawal. (I–J) Graphs for P2Y12 (I) and CD68 (J) %IR at 24 hours post-withdrawal. (K–L) Average microglial size (K) and percent microglial activation (L) at 24 hours post-withdrawal. (M–P) 3-week post-withdrawal counterparts for P2Y12 %IR (M), CD68 %IR (N), average microglial size (O) and percent microglial activation (P). *p < 0.05, **p < 0.01, ***p < 0.001, ***p < 0.0001, ns = not significant. Interleaved bars graphs report individual data points with mean and standard deviation. “Agg. MRN” refers to pooled data from rostral and caudal MRN sections to better capture nucleus-wide effects.

A neuroinflammation analysis of the MRN revealed that EtOH-exposed animals had significantly diminished P2Y12 %IR in the rostral (*p = 0.0105), caudal (*p = 0.0495), and aggregate regions (***p = 0.0005) at 24 hours post-withdrawal; no significant changes were observed at the 3-week timepoint (Fig. 3G–I, M). Additionally, average microglial size in the MRN showed no significant changes at either time points (Fig. 3K, O). Similarly, no significant increases or decreases were seen in CD68 %IR at either 24 hours or 3 weeks post-withdrawal (Fig. 3G–H, J, N). However, microglial activation—based on co-expression of P2Y12 and CD68 in microglia—was observed to be significantly elevated in the caudal (**p = 0.0047) and aggregate (**p = 0.0099) MRN at 24 hours post-withdrawal, and rostral (**p = 0.0032) and aggregate (***p = 0.0004) MRN at 3 weeks post-withdrawal (Fig. 3L, P). Exact p values and statistical tests for each variable can be found in supplementary Table S2.

Neuroinflammatory and Serotonergic Changes in the Raphe Magnus Nucleus

We next assessed the raphe magnus (RMg). Analysis of 5-HT %IR in the raphe magnus showed there was a significant deficit (*p = 0.0165) in 5-HT at 24 hours post-withdrawal, but none at 3 weeks post-withdrawal (Fig. 4A–B, D) for EtOH-exposed animals; this data is also shown in Fig. 5B, D due to the RMg’s dual function as both an origin and recipient of pain-related 5-HT signaling (Beitz, 1982). As with the other serotonergic nuclei, no significant changes in 5-HT neuron density in the RMg were detected for either time point (Fig. 4C, E).

Fig. 4. Serotonergic and neuroinflammatory changes in the raphe magnus (RMg) following adolescent EtOH exposure.

(A) Representative images of 5-HT immunostaining in the RMg following 24 hours and 3 weeks post-withdrawal. (B–C) Graphs for 5-HT percent-immunoreactive area (%IR) (B) and 5-HT neuron density (C) at 24 hours post-withdrawal. (D–E) 3-week post-withdrawal counterparts for 5-HT %IR (D) and 5-HT neuron density (E). (F) Representative images of P2Y12 and CD68 IHC staining for both 24 hours and 3 weeks post-withdrawal. (G–H) Graphs for P2Y12 (G) and CD68 (H) %IR at 24 hours post-withdrawal. (I–J) Average microglial size (I) and percent microglial activation (J) at 24 hours post-withdrawal. (K–N) 3-week post-withdrawal counterparts for P2Y12 %IR (K), CD68 %IR (L), average microglial size (M) and percent microglial activation (N). *p < 0.05, **p < 0.01, ns = not significant. Interleaved bars graphs report individual data points with mean and standard deviation.

Figure 5. Region-specific disruptions in serotonergic signaling downstream of raphe nuclei following adolescent EtOH exposure.

(A–B) Representative images of 5-HT immunostaining at 24 hours (A) and 3 weeks (B) post-withdrawal in downstream projection regions: nucleus accumbens (NAcc), anterior cingulate cortex (ACC), thalamus (Thal), amygdala (Amyg), hypothalamus (Hyp), and raphe magnus (RMg). (C–D) Representative images of SERT immunostaining at 24 hours (C) and 3 weeks (D) post-withdrawal. (E–F) Graphs for 5-HT (E) and SERT (F) percent-immunoreactive area (%IR) at 24 hours post-withdrawal. (G–H) 3-week post-withdrawal counterparts for 5-HT (G) and SERT (H) %IR. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, ns = not significant. Interleaved bars graphs report individual data points with mean and standard deviation. Some images have been color-adjusted for demonstration purposes only. All analyses were performed on raw, unedited images. RMg 5-HT %IR data in the figure are repeated from Fig. 4 for comparison purposes.

Analysis of neuroinflammation in the RMg revealed no significant changes in P2Y12 %IR, CD68 %IR, or average microglial size at 24 hours post-withdrawal (Fig. 4F–I). However, microglial activation via assessment of CD68+/P2Y12+ cells showed a significant increase (**p = 0.0016) at this time point (Fig. 4J) compared to water-treated controls. At 3 weeks post-withdrawal, we observed significant decreases (**p = 0.0029) in P2Y12 %IR as well as for average microglial size (*p = 0.0308) (Fig. 4F, K, M). While no significant changes were observed for CD68 %IR (Fig. 4L), we did see a significant elevation (*p = 0.0155) in microglial activation for this time point (Fig. 4N) for EtOH-exposure animals. Exact p values and statistical tests for each variable can be found in supplementary Table S3.

Disruption of Serotonin Signaling in Downstream Projection Regions

Since we found alterations of 5-HT immunolabeling in major raphe nuclei, where we also observed an increase in signs of neuroinflammation, we next assessed 5-HT signaling in several downstream regions—including the RMg, which receives downstream 5-HT input in addition to being a serotonin-producing nucleus (Beitz, 1982). One of these downstream projection regions, the nucleus accumbens (NAcc), showed no significant changes for either 5-HT or SERT %IR at either time point (Fig. 5A–H; NAcc). In the anterior cingulate cortex (ACC), we observed a significant decrease (**p = 0.0010) in 5-HT %IR for EtOH-exposed animals at 3 weeks post-withdrawal, but none at 24 hours (Fig. 5A–B, E, G; ACC). No significant changes in SERT %IR were detected in this region (Fig. 5C–D, F, H; ACC).

In the thalamus (Thal), EtOH-exposed animals exhibited significant reductions (*p = 0.0206) in 5-HT %IR at 24 hours with none at 3 weeks post-withdrawal (Fig. 5A–B, E, G; Thal); SERT in the Thal was not significantly different at either time point (Fig. 5C–D, F, H; Thal). The RMg, whose 5-HT data are repeated from Fig. 4 for cross-regional comparison, showed a similar pattern to the Thal, with significant deficits (*p = 0.0165) in 5-HT %IR at 24 hours post-withdrawal only (Fig. 5A–B, E, G; RMg). Similarly, there were no significant changes in SERT %IR for either post-withdrawal time point (Fig. 5C–D, F, H; RMg). In the amygdala (Amyg), we observed significant decreases in 5-HT %IR at both 24 hours (**p = 0.0044) and 3 weeks post-withdrawal (***p = 0.0006) (Fig. 5A–B, E, G; Amyg) compared to water-treated controls. We also found a significant deficit in SERT %IR at 24 hours (****p < 0.0001) but none for 3 weeks post-withdrawal (Fig. 5C–D, F, H; Amyg). In the hypothalamus (Hyp), 5-HT %IR was significantly reduced at both 24 hours (**p = 0.0037) and 3 weeks post-withdrawal (**p = 0.0019) (Fig. 5A–B, E, G; Hyp) for EtOH-exposed mice. While no significant changes in SERT %IR at 24 hours post-withdrawal were observed, there was a significant decrease at 3 weeks post-withdrawal (*p = 0.0185) (Fig. 5C–D, F, H; Hyp). Exact p values and statistical tests for each variable can be found in supplementary Table S4.

Thermal Hypersensitivity Following Adolescent Ethanol Exposure

With the observed alterations in 5-HT signaling in downstream areas, and their involvements in pain processing, a Hargreaves paw test was conducted to assess pain sensitivity. The Hargreaves assessment was performed for 24-hours, 3-weeks, 6-weeks, and 9-weeks post-withdrawal. Using a 2-way rm mixed-model ANOVA, we found a significant reduction in paw withdrawal latency in both the left (****p < 0.0001) and right (****p < 0.0001) paws, as well as the average for both paws (****p < 0.0001) regarding the factor of EtOH vs. water (Fig 6A–C). Time was not found to be a factor; therefore, we did not pursue assessment at additional timepoints (p = 0.1628 for avg both paws). Exact p values and statistical tests for each variable can be found in supplementary Table S5.

Fig. 6. Heightened thermal sensitivity measured by Hargreaves assay following adolescent EtOH exposure.

Paw withdrawal latency for left paw (A), right paw (B), and average latency (C) at 24 hours, 3 weeks, 6 weeks, and 9 weeks post-withdrawal. 2w rm mixed-model ANOVA with Sidak’s multiple comparisons test; *p < 0.05, **p < 0.01, ****p < 0.0001. Plungers represent SEM.

Discussion

Region-Specific Reductions in Serotonergic Tone

Our immunohistochemistry findings reveal widespread reductions in 5-HT %IR across major serotonergic nuclei and their downstream projection targets after adolescent EtOH binge exposure. We observed serotonergic deficits in the DRN, MRN, and RMg as 5-HT-producing nuclei, and the ACC, Thal, Amyg, and Hyp as 5-HT projection sites (Figs. 2–5). These findings align with those of Khan et al. (2023), who similarly found reductions in serotonergic markers (5-HT and SERT) in the DRN, ACC, Thal, and Amyg following adolescent intermittent access to ethanol (AIE) via intraperitoneal injection. Furthermore, an AIE study performed by Vetreno et al. (2017) reported similar 5-HT deficits in the DRN, Hyp, and Amyg, which is in alignment with the regional patterns we observed. While our exposure models and withdrawal timelines differ slightly from these other studies, the overlap with the impacted regions highlights the generalizability of adolescent EtOH’s impact on serotonergic circuits.

In contrast to Khan et al. (2023), however, our 5-HT reductions occurred in the absence of significant losses of 5-HT neurons. This indicates that our observed 5-HT dysfunction resulted from impairments to neurotransmitter production, release, or reuptake, rather than cell loss. This also aligns with prior studies that highlight chronic EtOH exposure’s ability to disrupt serotonergic transmission through synaptic mechanisms—such as altered 5-HT receptor expression or glutamate release in the anterior insular cortex (Alexander et al., 2012), and transcriptional mechanisms—such as those that involve reduced expression of the 5-HT markers Tph2 and Vmat2 in raphe nuclei (Vetreno et al., 2017). Our observed reductions in SERT %IR in the Amyg and Hyp suggest additional disruptions to 5-HT reuptake mechanisms.

Increased Microglial Activation in Serotonergic Nuclei

In parallel with the serotonergic deficits, we observed clear neuroinflammatory changes within key 5-HT nuclei following adolescent EtOH binge exposure. Specifically, we found reductions in P2Y12 %IR, increases in CD68 %IR, and elevated microglial activation (CD68+/P2Y12+ cells)—indicators of a pro-inflammatory, phagocytic microglial phenotype (Haynes et al., 2006; Socodato et al., 2020; Walter & Crews, 2017). These changes have similarly been seen in other adolescent EtOH exposure studies, wherein elevated neuroinflammatory markers were co-observed with deficits of 5-HT following AIE in rodents. In particular, findings by Vetreno et al. (2017) lend support to the idea that 5-HT nuclei are especially vulnerable to alcohol-induced neuroimmune responses. Following AIE in rats, they reported significant increases in microglial activation via Iba-1 and CD11b markers in the DRN, which is a region they also found to have significant reductions in 5-HT %IR. Khan et al. (2023) similarly observed heightened microglial activation in mice, measured using CD68 and P2Y12 expression, coinciding with 5-HT deficits in the DRN, ACC, and Amyg. Together these studies reinforce a broader model in which adolescent EtOH exposure initiates a neuroimmune response that converges with 5-HT dysfunction across multiple brain circuits.

Microglial Activation as a Mediator of Serotonergic Dysfunction and Pain

Given the overlap in timing and localization between 5-HT deficits and neuroinflammation, we propose that microglial activation may mediate EtOH-induced 5-HT dysfunction. Previous literature has shown links between microglia and 5-HT neurons in multiple ways: cytokine secretion (Sălcudean et al., 2025; Wohleb, 2016), synapse elimination via synaptic pruning (Miller & Broadie, 2024; Socodato et al., 2020), and direct contact with neuronal soma and terminals (Colonna & Butovsky, 2017). Sălcudean et al. (2025) describe how activated microglia release numerous cytokines (i.e., TNF-α, IL-1β, and IL-6) that upregulate the indoleamine 2,3-dioxygenase (IDO) enzyme, which diverts tryptophan metabolism away from the synthesis of 5-HT—ultimately reducing its availability and contributing to serotonergic deficits. Given that TNF-α production can be initiated via the TLR4 cascade (Crews et al., 2016), which mediates EtOH-induced cytokine release in the CNS (Fernandez-Lizarbe et al., 2009, 2013; Holloway et al., 2023), this pathway may represent an important bridge between increased microglial activation and the observed 5-HT deficits.

Our findings of microglial activation paired with 5-HT deficits were consistent with findings by Khan et al. (2023), who reported increased expression of microglial cytokines TNF-α and IL-1β in the DRN, in addition to downregulation of key serotonergic markers in the DRN and in downstream areas implicated in pain modulation. As in our study, these changes were associated with hyperalgesia that persisted into adulthood, which lends support to the idea that serotonergic impairment following adolescent EtOH exposure may be immune-mediated. Taken together, these findings highlight a potential immune-driven mechanism for 5-HT disruption and its effects on pain sensitivity.

While both serotonergic deficits and microglial activation may occur independently in response to adolescent EtOH exposure, our findings point to a more connected mechanism—especially with the observed temporal and regional overlaps. These overlaps, seen in the raphe nuclei, support the idea that microglial activation may be behind 5-HT dysfunction rather than just a consequence. This interpretation is consistent with work previously discussed, with activated microglia known to be able to influence 5-HT neurons via cytokine signaling, synaptic pruning, and diverting tryptophan metabolism away from neurotransmitter synthesis (Colonna & Butovsky, 2017; Sălcudean et al., 2025; Socodato et al., 2020). Regardless, it would be beneficial to incorporate selective inhibition of microglial activation in future studies in order to help determine whether it is necessary for causing the serotonergic impairments that we observed.

Adolescent Ethanol Exposure Induces Persistent Hyperalgesia

Our behavioral findings (Fig. 6) demonstrate that adolescent EtOH binge-like exposure results in significant and persistent hyperalgesia at not only 24 hours and 3 weeks post-withdrawal, but persisting at least as late as 9 weeks post-withdrawal. Our data is consistent with the results from Khan et al. (2023), who found that mice exposed to AIE exhibited significant thermal and mechanical hyperalgesia that persisted into adulthood. Our results are further supported by the results of Brandner et al. (2023), who demonstrated elevated mechanical and heat hyperalgesia following EtOH exposure via chronic intermittent EtOH vapor. This study, however, took into account sex and blood alcohol concentration—variables not directly controlled in our study, but worth considering for future directions.

When combined with the widespread 5-HT deficits and increased microglial activation observed in key pain-processing raphe nuclei (i.e., DRN, MRN, and RMg) and their downstream projection regions (i.e., Amyg and Hyp), these behavioral findings suggest that there is a potential mechanistic link. Specifically, they suggest that adolescent alcohol-induced neuroinflammation and 5-HT dysfunction may contribute to the development of persistent hyperalgesia into adulthood. While we did not directly investigate the mechanism in the present study, the temporal and regional overlaps with the observed changes fit a model where alcohol-induced neuroinflammation is responsible for 5-HT disruptions and subsequent prolonged hyperalgesia. To explore this model and directly assess causality, studies in the future should incorporate targeted interventions—such as ones that can inhibit microglial activation (e.g., minocycline, TLR4 antagonists) or restore 5-HT signaling via supplementation of 5-HT precursors.

Limitations, Future Directions, and Public Health Implications

It is worth addressing that our study used only male mice. While initially viewed as a limitation, the exclusion of females is justified in the current context given the focus of this study on the effects of alcohol on serotonergic systems and the confounding role of sex hormones in the modulation of that circuitry. In particular, 5-HT neurons in the DRN differentially express estrogen receptor beta (ERβ) in a sex-dependent manner (He et al., 2023; Murakawa et al., 2024). The activation of ERβ has been shown to regulate the expression of Tph2—the rate-limiting enzyme for the production of 5-HT—via an estrogen response element in the gene’s promoter region (Hiroi & Handa, 2013). Additionally, estrogens influence the expression of the monoamine oxidase B enzyme, which is important for the degradation of 5-HT, in both a region- and sex-dependent manner (Holschneider et al., 1998). Mechanisms such as these engender significant confounds when it comes to interpreting EtOH-mediated changes in 5-HT signaling. As such, estrogens’ effects on 5-HT production and turnover would have likely obscured EtOH’s effect on 5-HT circuitry. While it would be important for future studies to incorporate estrus-matched female cohorts in order to accurately determine if similar neuroinflammatory and pain-related responses are observed in both males and females—with sex hormones clearly playing a nuanced, yet critical, role in modulating these serotonergic systems -- this was beyond the scope of the current study. Secondly, the functional consequences of 5-HT dysfunction were not assessed in our study; these could be more directly assessed via in vivo electrophysiology of serotonergic nuclei and/or downstream projection regions in the future.

The findings from our study possess vital implications for public health, since adolescent binge drinking remains prevalent in American society, and our data suggest that it may have lasting consequences for pain sensitivity. Therapeutic strategies that target 5-HT signaling and/or neuroinflammation mechanisms could offer promising results. Novel approaches like these could be crucial for determining which of them is necessary or sufficient to drive the observed and prolonged changes in pain sensitivity. In future studies, these same pathways may also serve as viable therapeutic targets in hopes of mitigating some of the long-term effects of adolescent EtOH binge exposure.

Data Availability Statement

The raw and processed data supporting the findings of this study are available from the corresponding author upon reasonable request.