Neonatal exposure to morphine results in prolonged pain hypersensitivity during adolescence, driven by gut microbial dysbiosis and gut-brain axis-mediated inflammation

Abstract

Opioids, such as morphine, are used in the Neonatal Intensive Care Unit (NICU) for pain relief in neonates. However, the available evidence concerning the benefits and harms of opioid therapy in neonates remains limited. While previous studies have reported that neonatal morphine exposure (NME) results in long-term heightened pain sensitivity, the underlying mechanisms are not well understood. This study proposes that dysbiosis of the gut microbiome contributes to pain hypersensitivity following NME. Using an adolescent female murine model, pain sensitivity was evaluated using the tail flick and hot plate assays for thermal pain and the Von Frey assay for mechanical pain. Gut microbiome composition was assessed using 16S rRNA sequencing, while transcriptomic changes in midbrain samples were investigated using bulk RNA sequencing. NME induced prolonged hypersensitivity to thermal and mechanical pain in adolescence, accompanied by persistent gut microbial dysbiosis and sustained systemic inflammation, characterized by elevated circulating cytokine levels (e.g., IL-1α, IL-12p70, IFN-γ, IL-10). Transplantation of the microbiome from NME adolescents recapitulated pain hypersensitivity in naïve adolescent mice, while neonatal probiotic intervention with Bifidobacterium infantis (B. infantis) reversed the pain hypersensitivity by preventing gut dysbiosis and associated systemic inflammation. Furthermore, transcriptomic analysis of midbrain tissues revealed that NME upregulated several genes and key signaling pathways, including those related to immune activation and excitatory signaling, which were notably mitigated with neonatal B. infantis administration. Together, these findings highlight the critical role of the gut-brain axis in modulating pain sensitivity and suggest that targeting the gut microbiome offers a promising therapeutic strategy for managing neurobiological disorders following early opioid exposure.

1. Introduction

Opioid analgesics, such as morphine, are the gold standard in the United States for moderate to severe pain management. However, chronic use is associated with a paradoxical pain hypersensitivity referred to as opioid-induced hyperalgesia (OIH) (Lee et al., 2011). Currently, there are no general Centers for Disease Control and Prevention (CDC) guidelines on opioid use for analgesia in the neonatal population (Dowell et al., 2016; Dowell et al., 2022). However, neonates who need invasive procedures, such as surgery, are exposed to opioids postnatally for pain management and stabilization in the Neonatal Intensive Care Unit (NICU), with morphine being the most used opioid (Hall and Anand, 2014; Fong et al., 2024). The CDC recognizes that the available evidence concerning the benefits and harms of long-term opioid therapy in neonates is limited and that additional research is needed for the development of future guidelines for this critical population.

How neonatal morphine exposure (NME) affects pain sensitivity long-term has been relatively unexplored in humans. However, clinical studies have demonstrated a significant reduction in brain volume, neuronal packing density, and dendritic growth after exposure to morphine in the early postnatal period, which can impact pain thresholds, impair learning abilities, and alter locomotor activity later in life (Durrmeyer et al., 2010) (Fong et al., 2024). Preclinical models have further described hypersensitivity to pain persisting later in life following early postnatal exposure to morphine. In rodent models, NME has been shown to result in prolonged thermal hyperalgesia and mechanical allodynia during both adolescence and adulthood (Zhang and Sweitzer, 2008) (Craig and Bajic, 2015).

Although NME-induced alterations in pain processing have been reported, the mechanisms underlying these long-lasting alterations remain incompletely understood. This is especially relevant as uncovering mechanisms that play a role in altered pain responses may help mitigate the adverse effects of NME and provide insights into early intervention strategies. Recent research has highlighted the importance of the gut microbiome in neurodevelopment and behavior through the gut-brain axis. The gut microbiome consists of trillions of microorganisms, including bacteria, fungi, parasites, and viruses, along with their genes (Human Microbiome Project Consortium, 2013). Numerous signaling molecules derived from the gut microbiota, such as metabolites and neuromodulators, can act on pain receptors and regulate peripheral and central neuronal signaling (Guo et al., 2019). As a result, the gut microbiome can directly or indirectly influence the excitability of primary nociceptive neurons (Guo et al., 2019). Furthermore, morphine dependence in adults has been shown to induce dysbiosis in the gut microbiome, disrupting gut homeostasis (Wang et al., 2018). Previous studies from our group have demonstrated that chronic morphine-induced gut dysbiosis creates an inflammatory environment and compromises gut epithelial barrier function in the small intestine, leading to sustained systemic inflammation in adult mice (Meng, et al., 2013). Opioid-induced microbial dysbiosis has been implicated in the development of analgesic tolerance to morphine (Zhang, et al., 2019). Both analgesic tolerance and opioid-induced pain hypersensitivity contribute to the declining efficacy of opioids over time and are essential phenomena necessitating thorough investigation. Moreover, similarly as seen in adults, early postnatal opioid exposure has been shown to induce opioid dependency and withdrawal symptoms following abstinence or treatment with an opioid receptor antagonist (Jones and Barr, 1995); (Barr and Wang, 1992). Consequently, morphine dependence during the neonatal period may result in persistent alterations in pain behaviors through gut microbial dysbiosis and sustained systemic inflammation.

Although the pathophysiology of prolonged pain encompasses complex interactions between sensory, immune, and inflammatory responses within the nervous system (Dworsky-Fried et al., 2020), the role of the gut microbiome in prolonged hypersensitivity to pain following NME had not yet been investigated. NME was found to induce long-term dysbiotic alterations in the gut microbiome that persist later in life (Antoine et al., 2022). However, limited knowledge exists regarding the gut microbiome’s role in the long-term effects of early opioid exposure in the adolescent population, underscoring the need to characterize the gut microbiome’s involvement in sensory and immune functions in adolescents following NME. Based on previous studies, we hypothesized that a dysbiotic gut microbiome contributes to prolonged pain hypersensitivity in a murine model of NME. We assessed pain sensitivity, gut microbiome composition, gut epithelial barrier function in the small intestine, and circulating inflammatory markers during adolescence. Furthermore, using bulk RNA sequencing, we examined the unique transcriptional signatures induced by NME in the midbrain which contains the periaqueductal gray (PAG), a critical structure involved in the supraspinal modulation of pain. We also evaluated how modulation of the gut microbiome with probiotics influences these transcriptional changes.

2. Materials and methods

2.1. Experimental animals

Female and male C57BL/6J mice were obtained from Jackson Laboratory at 8–10 weeks of age. Their offspring were used in this study. A maximum of five mice were housed per cage, and cages were regularly changed. Each treatment group was handled separately to avoid contamination between their microbiomes. Power analysis was conducted using preliminary data to estimate the required sample size. A minimum of five mice per group was necessary to achieve 80% power at a significance level of p < 0.05. All animals were sacrificed using a CO₂ chamber to minimize pain and suffering during tissue collection.

2.2. Neonatal morphine exposure paradigm

Nulliparous female mice aged 10–12 weeks were mated with healthy males of the same age. Four female mice were housed with one male mouse. Females were assessed for pregnancy and housed individually upon confirmation of pregnancy. Upon delivery, postnatal day one was recorded, and pups were randomly assigned to a morphine group to be NME mice or a saline group to be saline-exposed controls. All pups were exposed to morphine or saline starting on postnatal days 6–7 for five days total. Morphine was administered subcutaneously at a dose of 5 mg/kg/day, alongside a saline control, with each delivered in a volume of 10 μL once daily. Previous research has explored morphine dosages as high as 10 to 15 mg/kg (Craig and Bajic, 2015) (Borrelli, et al., 2021); however, the minimum dose necessary to achieve analgesia in rodent neonates was established at 3 mg/kg (Zhang and Sweitzer, 2008) (Zissen et al., 2007). Consequently, the 5 mg/kg dosage was selected for this study as it balances efficacy in analgesia while minimizing excessive exposure. Pups were weighed and placed on a heated pad to ensure adequate warmth during injections and gavages. Separation from the dam was minimized to less than one minute for each procedure, after which the pups were promptly returned to their mother. Following the cessation of morphine treatment, the pups remained with their mothers until weaning at 21 days of age. After weaning, they were sexed and housed until behavioral testing during adolescence, which occurred at 4–5 weeks of age. All procedures were reviewed and approved by the University of Miami Institutional Animal Care and Use Committee (IACUC) and were in line with the guidelines set forth by the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

2.3. Behavioral paradigms

The adolescent mice underwent nociceptive testing at 4–5 weeks old.

Tail flick:

Thermal pain sensitivity at the spinal cord level was evaluated using the tail flick assay. A radiant heat stimulus, emitted by a high-intensity lamp, was applied to the mid-region of the tail, with a maximum cutoff of 10 seconds to minimize tissue damage. The latency to tail-flick was determined in one trial. Since tail-flick is affected by tail skin temperature, careful attention was paid to maintaining the ambient temperature at 22–23 °C.

Hot plate:

The hot plate assay was used to evaluate supraspinal pain pathways. Thermal pain sensitivity was assessed by measuring foot withdrawal latency to heat. Each animal was placed in a box containing a smooth, temperature-controlled surface. The intensity of the hot plate was maintained constant throughout all experiments at 52 °C. Animals were observed and removed at the first foot withdrawal movement or foot licking. Latency to withdrawal was recorded in one trial and the maximum time on the hot plate was 40 seconds to avoid tissue damage.

Von Frey:

Mechanical pain sensitivity was determined using calibrated Von Frey filaments (IITC Life Science Inc, CA) introduced serially to the hind paws in ascending order of strength. Animals were placed in non-transparent plastic cubicles on a mesh floor for an acclimatization period of at least 30 minutes on the morning of the test day. A positive response was defined as a rapid withdrawal and/or licking of the paw immediately upon application of the stimulus. The “ascending stimulus” method was used in which monofilaments with increasing force were applied until a withdrawal response was elicited. The force of the Von Frey filament that elicited two or more out of five applications was designated as the mechanical withdrawal threshold. The weights of the von Frey hairs used were the following: 0.4, 0.8, 1.3, 1.8, 2.4, 2.9, 7.5, 11, and 12 grams.

2.4. Fecal microbiota Transplant (FMT)

Fecal Microbiota Transplant (FMT) was performed following the stepwise procedures for FMT in mice (Bokoliya et al., 2021) and following the established procedure from previous publications with modifications (Zhang, et al., 2019); (Thomaz et al., 2021). The fecal contents of NME and saline-exposed adolescent mice were collected and pooled. Briefly, 200 mg of the fecal extract were suspended in 10 mL sterile PBS, filtered through a 70 μm cell strainer, and centrifuged at 6000xg for 20 minutes. About 10¹⁰ CFU/mL fecal bacteria were suspended in 6% NaHCO₃ buffer with 20% sucrose. Naïve recipient mice were 4–5 weeks old and were gavaged 200 μL of an antibiotic cocktail for three days to deplete most of their natural microbiome to provide an ecological niche for colonization of the transplanted stool microbiota. The antibiotics cocktail contained 10 mg/mL Bactricin, 10 mg/mL Metronidazole, 40 mg/mL Neomycin, 4 mg/mL Vanomycin, and 24 ug/mL Primaricin. Following the antibiotic treatment, the recipient mice were orally gavaged with 200 μL of freshly prepared fecal suspension for seven consecutive days. Pain sensitivity was measured at baseline and after FMT. Following the FMT treatment, stool samples were collected to perform 16S rRNA sequencing to assess the gut microbiome.

2.5. Neonatal probiotic administration

Probiotics and vehicle were administered to neonates during the NME period. The Bifidobacterium longum subspecies infantis (B. infantis) was obtained as a powder and was dissolved in water and given at a maximum concentration of 1 × 10⁹ CFU/day based on previous studies (Hiraku, et al., 2023); (McKernan et al., 2010). The probiotic solution or a vehicle were administered to neonates via oral drops once a day for five days while exposing them to morphine or saline. Following neonatal treatments, pain sensitivity was assessed, and stool samples were collected to assess the gut microbiome during adolescence.

2.6. Statistical analysis

Data were plotted as means +/− standard error of means (SEM) and analyzed using GraphPad Prism (GraphPad Software, Inc). A student’s t-test was used to compare parametric data and the Mann-Whitney test was used for non-parametric data. For multiple-group comparison, data were analyzed with one- or two-way ANOVA analysis, followed by Sidak’s multiple comparison test. Comparisons over multiple time points were assessed with repeated measures ANOVA. The criterion of statistical significance was p < 0.05.

2.7. 16S rRNA gene sequencing

To investigate the gut microbiome, DNA was extracted from all collected stool samples using a DNeasy 96 PowerSoil Pro QIAcube HT Kit, along with the QIAcube HT liquid-handling machine (Qiagen, Maryland, USA). Sequencing on the extracted samples was performed by the University of Minnesota Genomic Center, MN. The hypervariable V4 region of the 16S rRNA gene was used for PCR amplification using the forward primer 515F (GTGCCAFCMGCCGCGGTAA) and reverse primer 806R (GGACTACHVGGGTWTCTAAT). Utilizing the Illumina MiSeq v.3 platform, the amplicons were sequenced and generated 300-bp paired-end reads.

2.8. Bioinformatic analysis

Amplicon sequence variants (ASVs) were generated with the DADA2 package (Callahan et al., 2016), and the abundance values from the V4 region were used. The QIIME2 (Bolyen, et al., 2019) pipeline was used for diversity analyses and taxonomic assignment. To analyze the alpha diversity, the student’s t-test or an ANOVA was used to detect if alpha diversity differed between the treatment groups. To analyze beta diversity, Permutational multivariate analysis of variance (PERMANOVA) tests were used to assess if the microbial communities significantly differed between the treatment groups, and the Nonmetric multidimensional scaling (NMDS) plots were used for data visualization. Linear Discriminant Effect Size (LEfSe) (Segata, et al., 2011) was used to detect taxa differentially enriched across treatment groups.

2.9. RT² profiler

To evaluate inflammatory and nociceptive mediators in the neonatally exposed mice, midbrain tissues were flash-frozen and stored at − 80 °C for mRNA determination. The tissues were homogenized in TRIzol (Invitrogen) and centrifuged to extract the supernatant. cDNA from the supernatant was synthesized using the RT² First Strand Kit (Qiagen) following the manufacturer’s protocol. The quantitative real-time polymerase chain reaction was performed using the Mouse Pain: Neuropathic & Inflammatory Profiler PCR Array (Gene Globe ID: PAMM-162Z). A total of 84 key genes related to the transduction, maintenance, and modulation of pain responses were simultaneously profiled according to the manufacturer’s instructions. The fold change and relative expression levels to reference genes were calculated using the online data analysis RT² Profiler^™ PCR Array Data Analysis software at https://geneglobe.qiagen.com/us/analyze. A 1.2-fold change was set as a cutoff for upregulation and downregulation and a p value less than 0.05 was set for significance.

2.10. Multiplex cytokines bead array

To investigate the levels of cytokines circulating systematically, blood was collected by cardiac venipuncture. Plasma was isolated and stored at − 80 °C after centrifugation at 10,000 rpm for 15 minutes. Blood plasma concentration of a panel of cytokines was measured using the LEGENDplex^™ Mouse Inflammation Panel (13-Plex) (Cat # 740446) according to the manufacturer’s instructions. The panel included the following proteins: IL1-α, IFN-γ, IFN-β, IL-12p70, IL-1β, IL-6, TNF-α, IL-10, IL-17A, IL-23, IL-27, MCP-1, and GM-CSF.

2.11. Histological evaluation

To investigate the morphology of the gut, small intestine tissues were collected, paraffin-embedded, and cut into 5-μm sections. The sections were dewaxed in xylene for five minutes twice and dehydrated in a gradient of alcohol and water. The Division of Comparative Pathology at the University of Miami performed the hematoxylin and eosin (H&E) staining according to standard protocols. Each section was blindly examined and scored using a histological scoring system previously described (Jalodia, et al., 2022) based on four main criteria of villus architecture, epithelial damage, inflammation, and epithelial expulsion to the lumen. The villus and crypt cell lengths were measured using a calibrated ruler in ImageJ.

2.12. FITC dextran permeability assay

To access in vivo intestinal permeability, the Fluorescein isothiocyanate (FITC) dextran permeability assay was used. 500 mg/kg of FITC-dextran (wt 4000; Sigma-Aldrich) was orally gavaged into mice four hours before blood collection. The blood was centrifugated at 10,000 rpm for 15 minutes to separate the plasma. FITC dextran fluorescence intensity in the plasma was immediately measured by SpectraMax^® i3 × . (Excitation wavelength 488 nm; emission 520 nm).

2.13. Bulk RNA sequencing

The midbrains of the neonatally exposed mice were collected for transcriptomic analysis. Upon collection, the tissues were immediately flash-frozen in liquid nitrogen and stored at − 80 °C. RNA extraction and sequencing were performed by Novogene Bioinformatics Technology Co., Ltd., in Beijing, China. Gene reads were aligned to Genome Browser assembly ID: Mus Musculus (GRCm38/mm10).

2.14. Differential expressed genes (DEG) screening, and pathway enrichment analysis

The DESeq2 package (ver. 1.38.2) with R version 4.2.2 (2022-10-31) was used to calculate differentially expressed genes (DEGs) between pairwise groups. The p values were adjusted using the Benjamini and Hochberg approach to control false discovery rates (FDRs). Genes with adjusted p < 0.05 and log2 fold change ≥ 0.5 were considered differentially expressed. DESeq2 was also used to compare the differential expression of transcripts between the treatment groups. DEGs were subsequently analyzed using QIAGEN IPA to predict functional annotations and canonical pathways with a threshold set to |log2Fold-Change| ≥ 0.5 & p-value < 0.05.

3. Results

3.1. Neonatal morphine exposure (NME) results in prolonged pain hypersensitivity during adolescence independent of sex

The long-term effects of NME on development were investigated in a murine model during adolescence. Newborn mice were exposed to morphine at a concentration of 5 mg/kg/day or saline for five days. All behavioral assays were performed during adolescence at 4–5 weeks old (Fig. 1a). During the neonatal period, neonates exposed to morphine exhibited significantly reduced weight gain compared to their saline-exposed counterparts (Fig. 1b). This weight disparity persisted into adolescence, with both female and male NME adolescents displaying significantly lower body weights compared to their respective saline-exposed controls (Fig. 1c).

Fig. 1.

NME results in decreased weight gain and prolonged pain hypersensitivity during adolescence independent of sex. a. Neonates were exposed to morphine (5 mg/kg/day) or saline for five days. Behavioral assays and microbiome analysis were performed during adolescence at 4–5 weeks old. b. During the neonatal period, neonates exposed to morphine showed significantly reduced weight gain compared to those exposed to saline (RM ANOVA; Drug: F (1, 23) = 104.6; p < 0.0001). c. Female and male NME adolescents had significantly lower body weight compared to saline-exposed controls (two-way ANOVA; Drug: F (1, 45) = 15.95, p = 0.0002) d. Tail flick (seconds), e. Hot plate latency (seconds), and f. Von Frey withdrawal thresholds (grams) of female and male adolescents exposed to morphine or saline neonatally. Data were analyzed using a two-way ANOVA followed by Sidak’s correction and expressed as mean +/− SEM. n = 8–15, Tail flick (Drug: F (1, 50) = 16.31, p = 0.0002); Hot plate (Drug: F (1, 48) = 21.17, p < 0.0001); Von Frey (Drug: F (1, 36) = 64.39, p < 0.0001).

To determine the long-term impact of NME on nociceptive behavior, thermal thresholds were assessed using the tail flick and hot plate assays, while mechanical thresholds were assessed using calibrated Von Frey filaments during adolescence. Both female and male NME adolescents demonstrated significantly lower tail flick latencies (Fig. 1d), hot plate latencies (Fig. 1e), and Von Frey withdrawal thresholds (Fig. 1f) compared to their respective saline-exposed controls. These results indicate that NME induces prolonged hypersensitivity to thermal and mechanical pain, independent of sex, during adolescence. The existing literature suggests that women are generally more sensitive to pain, experiencing it more frequently, more intensely, and for longer durations (Paller et al., 2009); (Fillingim and Ness, 2000). As females experience chronic pain at a higher prevalence than males during adolescence (Chambers et al., 2024), this study concentrated on the female animal model to further explore the underlying mechanisms of NME-induced pain hypersensitivity. All subsequent experiments were conducted on female adolescent mice.

3.2. NME results in prolonged gut dysbiosis contributing to pain hypersensitivity during adolescence

Previous studies have established that opioid use induces gut dysbiosis (Jalodia, et al., 2022); (Wang and Roy, 2017); (Abu, et al., 2021). Specifically, NME has been shown to result in long-term alterations to the gut microbiome, observable during both adolescence and adulthood (Antoine et al., 2022). This study thus aimed to investigate the causal role of the gut microbiome in driving prolonged hypersensitivity to pain following NME.

To evaluate the changes in gut microbiome diversity and composition, Illumina sequencing of microbial 16S rRNA genes was performed on fecal samples from both NME and saline-exposed adolescent mice. After quality control, unique ASVs were identified among all samples (Sup Data 1). Analysis of alpha diversity, using the Chao1 index (Fig. 2a) and the Shannon index (Fig. 2b), indicated no significant differences between the NME and saline-exposed groups. However, significant differences in beta diversity were observed (Fig. 2c), with distinct clustering observed between fecal samples of NME and saline-exposed adolescents (PERMANOVA, F:4.1134, R2: 0.177, p = 0.003). To assess the composition of the gut microbiome, LEfSe was used to identify differentially enriched microbial taxa from the phylum to the genus levels. At the phylum level, NME adolescents showed an increase in Bacteroidota and a decrease in Proteobacteria and Firmicutes compared to saline-exposed controls (Fig. 2d). At the genus level, NME adolescents showed an increase in Enterorhabdus, Eubacterium_ventriosum_group, Intestinimonas, and ASF356, alongside a decrease in several genera, including Family_XIII_UCG_001, UCG_005, Romboutsia, Ruminococcus, Anaerotruncus, Incertae_Sedis, Oscillibacter, Turicibacter, Lachnospiraceae_UCG_001, and Lachnospiraceae_NK4A136_group, when compared to saline-exposed controls (Fig. 2e).

Fig. 2.

NME results in significant dysbiotic alterations in the gut microbiome during adolescence. Alpha diversity measured by a. the Chao1 index (unpaired t-test, t = 1.420, df = 19, p = 0.1719) and b. the Shannon diversity index (unpaired t-test, t = 0.3969, df = 19, p = 0.6959) of female adolescents following NME. c. NMDS analysis for beta diversity revealed distinct and significant clustering in the NME adolescents compared to the saline-exposed adolescents (PERMANOVA, F:4.1134, R2: 0.177, p = 0.003). d. Lefse analysis of bacterial taxa at the phylum level. e. Lefse analysis of bacterial taxa at the genus level. f. BugBase analysis predicting the proportions of biofilm-forming, gram-negative, gram-positive, and potentially pathogenic bacteria between the microbial communities. n = 9–11, Data were analyzed using a student’s t-test and expressed as mean +/− SEM. *p < 0.05; **p < 0.01; ***p < 0.001.

Potential phenotypic changes in the gut microbiome were predicted using BugBase. NME adolescents displayed a significant increase in the relative abundance of biofilm-forming bacteria, gram-negative bacteria, and potentially pathogenic bacteria as well as a significant decrease in the relative abundance of gram-positive bacteria compared to saline-exposed controls (Fig. 2f). No significant differences were predicted in the relative abundance of aerobic bacteria, anaerobic bacteria, bacteria containing mobile elements, or facultative bacteria between the groups (Sup Fig. 4). These findings demonstrate that NME induces significant, long-term dysbiotic alterations in the gut microbiota, observable during adolescence.

To investigate the causal role of the gut microbiome in prolonged hypersensitivity to pain, FMT was performed. Processed fecal samples from NME adolescents depicted as “morphine” and saline-exposed controls depicted as “saline” were gavaged daily for one week into aged-matched naïve mice (Fig. 3a). Recipient mice were treated with antibiotics for three days prior to FMT to deplete their natural microbiome, creating an ecological niche for colonization of the transplanted stool microbiota (Fig. 3a). Thermal nociceptive assessments conducted before and after transplantation revealed that the “morphine” microbiome resulted in significantly reduced tail flick (Fig. 3b) and hot plate withdrawal latencies (Fig. 3c) relative to the “saline” microbiome in the recipient mice. Consequently, the “morphine”-assigned recipients displayed thermal hyperalgesia compared to the “saline”-assigned recipients.

Fig. 3.

Transplantation of the microbiome from NME adolescents results in hypersensitivity to thermal pain in naïve adolescent mice during adolescence. a. Recipient adolescents underwent fecal microbiota transplantation (FMT) with samples isolated from NME adolescents (“morphine”) and from saline-exposed adolescents (“saline”). b. Percent change in tail flick latency (unpaired t-test, t = 2.252, df = 18, p = 0.0370) and of c. Hot plate latency (unpaired t-test, t = 2.930, df = 20, p = 0.0083) from before and after colonization with either “saline” or “morphine” fecal samples. d. Alpha diversity measured by the Chao1 index and e. the Shannon diversity index of recipient mice before antibiotics treatment (naïve), after antibiotics treatment (ABX), after FMT of “saline”, and of “morphine” microbiome. Alpha diversity data were analyzed using a one-way ANOVA and expressed as mean +/− SEM. (n = 9–14, *p < 0.05, ****p < 0.0001, ns: non-significant). f. NMDS analysis for beta diversity revealed distinct and significant clustering in the “morphine” recipients compared to “saline” recipients (PERMANOVA, F:3.3202, R2: 0.126, p = 0.007). g. Lefse analysis of bacterial taxa in the recipient mice at the genus level.

Microbiome analysis of the recipient mice revealed the persisting NME-induced dysbiotic alterations post-transplantation. Pre-antibiotic treatment (ABX) successfully decreased the number of different bacterial species in the naïve age-matched recipients as per the Chao1 (Fig. 3d) and Shannon index (Fig. 3e). Following transplantation, recipients of the “morphine” microbiome showed no significant change in alpha diversity according to the Chao1 index (Fig. 3d) but showed a significant decrease in alpha diversity per the Shannon index (Fig. 3e) when compared to the “saline” microbiome recipients. Significant differences in beta diversity were also observed (Fig. 3f), with distinct clustering between the fecal samples of the “morphine” and “saline” microbiome recipients (PERMANOVA, F:3.3202, R2: 0.126, p = 0.007). At the genus level, recipients of the “morphine” microbiome showed a persisting increase in Enterorhabdus and ASF356 and a decrease in the commensal genera, Turicibacter, compared to “saline” microbiome recipients (Fig. 3g). Taken together, the FMT experiment during adolescence demonstrated that the transplantation of the microbiome from NME adolescents recapitulated pain hypersensitivity in naïve recipient mice.

3.3. Neonatal probiotic administration of B. infantis rescues NME-induced pain hypersensitivity through the rescue of gut dysbiosis during adolescence

To investigate whether NME-induced hypersensitivity to pain can be mitigated, the probiotic Bifidobacterium longum subspecies infantis (B. infantis) was administered to the neonates to rescue NME-induced gut dysbiosis. B. infantis was used in this study as it is a species that has been found to dominate the gut microbiota of breastfed infants and benefits the host by accelerating the maturation of the immune response, balancing the immune system to suppress inflammation, and improving intestinal barrier function (Underwood et al., 2015). Notably, multiple studies have demonstrated its efficacy in pain modulation. Specifically, B. infantis has been shown to alleviate irritable bowel syndrome (IBS)-associated visceral pain in two distinct rat strains (McKernan et al., 2010) as well as in pediatric patients with IBS (Giannetti, et al., 2017). Following B. infantis administration, its subsequent impact on the nociceptive system and gut microbiome during adolescence was assessed (Fig. 4a). Upon B. infantis administration, NME adolescents and saline-exposed controls exhibited no significant differences in tail flick latency (Fig. 4b), hot plate latency (Fig. 4c), and Von Frey withdrawal thresholds (Fig. 4d). Taken together, NME adolescents given B. infantis no longer exhibited pain hypersensitivity as they demonstrated significantly increased latencies and thresholds in all three assays compared to NME adolescents administered the vehicle.

Fig. 4.

Neonatal administration of the probiotic, B. infantis, rescues NME-induced pain hypersensitivity and facilitates the growth of beneficial bacteria during adolescence in the context of NME. a. Neonates were exposed to morphine (5 mg/kg/day) or saline and gavaged with probiotic, B. infantis, or vehicle for 5 days. Behavioral assays and microbiome analysis were performed during adolescence at 4–5 weeks old. b. Tail flick latency (seconds), c. Hot plate latency (seconds), and d. Von Frey withdrawal thresholds (grams) of NME and saline-exposed adolescents administrated probiotics or vehicle neonatally. Data were analyzed using a two-way ANOVA followed by Sidak’s correction and expressed as mean +/− SEM. Tail flick (Probiotic; F (1, 83) = 5.709, p = 0.0191); Hot plate (Probiotic: F (1, 77) = 53.56, p < 0.0001; Von Frey (Probiotic: F (1, 28) = 28.01, p < 0.0001). Alpha diversity measured by e. the Chao1 index and f. the Shannon index of adolescents neonatally exposed to morphine or saline and gavaged with probiotics or vehicle. g. NMDS analysis for beta diversity revealed distinct and significant clustering between NME adolescents given probiotics and those given vehicle. (PERMANOVA, F:5.7684; R2: 0.21549; p = 0.001). h. Lefse analysis of bacterial taxa at the genus level comparing NME adolescents given probiotics and those given vehicle.

Microbial analysis during adolescence revealed that the administration of B. infantis attenuated dysbiotic alterations in the gut microbiome of NME adolescents and saline-exposed controls (Sup Fig. 5a–e). In particular, high-level potential phenotypes of the gut microbiome predicted using BugBase indicated no significant differences in the relative abundance of biofilm-forming bacteria, gram-negative bacteria, gram-positive bacteria, and potentially pathogenic bacteria between NME and saline-exposed adolescents administered B. infantis (Sup Fig. 5f). The administration of one bacterial species, B. infantis, significantly decreased alpha diversity in both NME and saline-exposed adolescents when compared to those given the vehicle, as per the Chao1 index (Fig. 4e) and the Shannon index (Fig. 4f). To investigate the direct role of B. infantis in attenuating NME-induced alterations, comparisons were made between NME adolescents given probiotics and NME adolescents given the vehicle. Notably, significant differences in bacterial communities were observed via NMDS analysis for beta diversity (PERMANOVA, F:5.7684; R2: 0.21549; p = 0.001) (Fig. 4g). At the genus level, NME adolescents given B. infantis showed an increase in mainly the commensal genera of Lactobacillus, Bacteroides, and Turicibacter, whereas NME adolescents given the vehicle showed an increase in several genera, including Candidatus_soleaferrea, Colidextribacter, Incertae_Sedis, Eubacterium_ventriosum_group, Monoglobus, Lachnospiraceae_UCG_006, Enterorhabdus, Oscillibacter, Roseburia, Intestinimonas, and Lachnospiraceae_NK4A136_group (Fig. 4h). These findings demonstrate that neonatal administration of B. infantis mitigates NME-induced pain hypersensitivity by attenuating gut dysbiosis, primarily through the significant growth of several commensal bacteria during adolescence.

3.4. NME results in sustained systematic inflammation during adolescence, which is rescued with neonatal probiotic administration

The integrity and function of the intestinal barrier in the small intestine of NME adolescents compared to saline-exposed controls were assessed, as the small intestine was found to be involved in gut leakiness and increased inflammation following chronic morphine administration in adults (Meng, et al., 2013). Given that an activated immune response is thought to be a key mechanism by which NME-induced gut dysbiosis impacts pain behavior, we selected the small intestine as the focus of our analysis. Histological analysis of formalin-fixed paraffin-embedded (FFPE) tissue sections stained with H&E revealed significant disruptions in gut morphology and integrity in the NME adolescents (Fig. 5a). Specifically, these adolescents exhibited significantly shorter and fragmented intestinal villi (Fig. 5b), reduced crypt cell height (Fig. 5c), and an overall higher histological damage score (Fig. 5d) compared to saline-exposed controls. Additionally, in vivo, intestinal permeability was evaluated using the FITC permeability assay. Given that intestinal permeability decreases with age, with younger mice at 2 weeks of age having a more permeable gut than adults as the gut matures (Patel et al., 2012); (Gleeson et al., 2021), naïve adults were included in the assay to account for age-related differences. NME adolescents demonstrated significantly elevated levels of FITC-Dextran in their blood plasma compared to the naïve adults (Fig. 5e). However, the difference did not reach significance when compared to the saline-exposed adolescents, which also exhibited elevated FITC levels in their plasma (Fig. 5e). Consequently, these results indicate that NME causes long-term disruptions of gut morphology and they suggest enhanced intestinal permeability in mice during adolescence, as supported by previous studies.

Fig. 5.

NME disrupts the morphology of the small intestine and enhanced gut permeability occurs during adolescence. a. H&E staining of the small intestine of adolescent mice (4–5 weeks old) exposed to saline or morphine (5 mg/kg/day) neonatally. b. Villi lengths, c. crypts lengths, and d. the histological damage score of the small intestine of adolescent mice exposed to saline or morphine neonatally. Data were analyzed with an unpaired t-test and expressed as mean +/− SEM. n = 5, (***p < 0.001, **p < 0.01, *p < 0.05). e. FITC-Dextran fluorescence transfer from the gut to the blood in naïve adults and adolescent mice exposed to saline or morphine neonatally. Data were analyzed with one-way ANOVA and expressed as mean +/− SEM. n = 5–8, (F (2, 18) = 6.1, P = 0.0090, **p < 0.01, ns: non-significant). Scale bar = 200 μm.

Enhanced intestinal permeability facilitates the translocation of bacterial products from the gut to other parts of the body (Meng, et al., 2013). Consequently, we hypothesized that inflammatory and bacterial products derived from NME-induced gut dysbiosis translocate out of the gut, result in sustained inflammation, and mediate prolonged pain hypersensitivity. To investigate the levels of inflammatory cytokines circulating systematically following NME, the LEGENDplex^™ Mouse Inflammation Panel (Cat # 740446) was used to determine the concentration of 13 cytokines in the blood plasma of both NME and saline-exposed adolescents simultaneously. Additionally, we included NME and saline-exposed adolescents given probiotics to assess the impact of neonatal probiotic intervention on circulating cytokine levels. The results revealed a significant increase in the concentration of Interleukin 1 alpha (IL-1α) (Fig. 6a), Interferon-gamma (IFN-γ) (Fig. 6b), Interleukin 12 (IL-12p70) (Fig. 6c), and Interleukin 10 (IL-10) (Fig. 6d) in the plasma of NME adolescents compared to saline-exposed controls when given the vehicle. The levels of Interleukin-23 (IL-23), a member of the IL-12 family of cytokines, were also elevated, but the difference between the NME and saline-exposed adolescents given the vehicle did not reach significance (p = 0.0564) (Fig. 6e). Notably, these cytokine levels were no longer elevated with probiotic intervention, as no significant differences were observed between the NME and saline-exposed adolescents given probiotics (Fig. 6a–e). No changes were detected in all four groups for the other tested cytokines (Sup Fig. 6). Overall, these results demonstrate that NME causes an increase in the levels of several cytokines, including IL-1α, IL-10, IFN-y, and IL-12p70, leading to sustained systematic inflammation which can be attenuated with probiotic intervention. Systemic inflammation can sensitize nociceptive neurons, lowering their thresholds of activation and driving prolonged hypersensitivity to pain (Pinho-Ribeiro et al., 2017). Interestingly, neonatal probiotic intervention with B. infantis decreased the levels of these markers, indicating a reduction in the sustained inflammation following NME.

Fig. 6.

NME results in an increase of several cytokines in systemic circulation during adolescence, which is attenuated with neonatal probiotic administration. Blood plasma concentration of a. Interleukin 1 alpha (IL-1α), b. Interferon-gamma (IFN-y), c. Interleukin 12 (IL-12p70), d. Interleukin 10 (IL-10), and e. Interleukin-23 (IL23) in adolescent mice exposed to saline or morphine neonatally and given a vehicle or the probiotic, B. infantis, neonatally. Data were analyzed using a two-way ANOVA followed by Sidak’s correction and expressed as mean +/− SEM. (n = 7–10, *p < 0.05, ns: non-significant).

3.5. NME significantly impacts midbrain gene expression during adolescence, which is rescued with neonatal probiotic administration

To elucidate the molecular mechanisms by which NME-induced gut dysbiosis modulates hypersensitivity to pain, we assessed the impact of NME on gene expression of the midbrain region, which is an essential region within the neural circuit for pain processing and opioid-mediated analgesia (Ossipov et al., 2010). In particular, the midbrain PAG is a key structure involved in the supraspinal modulation of pain where neurons within this region descend via the descending neural pathway to the rostroventromedial medulla (RVM) into the spinal cord to inhibit pain signals (Ossipov et al., 2014). The PAG receives ascending signals and provides descending opioid-mediated inhibition of these nociceptive inputs due to its dense expression of mu-opioid receptors (Kalyuzhny et al., 1996). Therefore, the transcriptomic expression of the midbrain from NME (Mor) adolescents and saline-exposed (Sal) controls was assessed by bulk RNA sequencing. Gene reads were analyzed across different treatment groups (Sup Data 2). Principal component analysis (PCA) of the samples, based on normalized mRNA expression levels, revealed a distinct separation between the transcriptome profiles of the midbrain samples from NME and saline-exposed adolescents (Fig. 7a). A total of 2988 differentially expressed genes (DEG) were identified, with 1596 upregulated genes and 1392 downregulated genes in the NME adolescents compared to saline-exposed controls (Fig. 7b, c). The heat map illustrated the expression pattern of the top DEGs between the samples (Fig. 7b), and the volcano plot highlighted the upregulated and downregulated genes (Fig. 7c). To determine the biological relevance of the DEGs associated with NME, canonical pathways associated with the DEGs were analyzed by IPA. NME significantly altered several canonical pathways in the midbrain. The top upregulated pathways included opioid signaling, CREB signaling in neurons, G-protein coupled receptor (GPCR) signaling, neuropathic pain signaling in dorsal horn neurons, calcium signaling, synaptic long-term potentiation, synaptic long-term depression, and multiple sclerosis signaling pathway (Fig. 7d). Notably, genes heavily involved in immune activation and inflammation, such as nuclear factor kappa-B subunit 1 (Nfkb1), interleukin-16 (Il16), and interleukin-22B (Il22b), were significantly upregulated following NME (Fig. 7e). Furthermore, genes implicated in increased glutamatergic, calcium, and GPCR signaling, which contribute to heightened pain signaling, including glutamate metabotropic receptor 4 (Grm4), glutamate ionotropic receptor AMPA type subunit 4 (Gria4), and calcium/calmodulin-dependent protein kinase IV (Camk4), were also upregulated in the midbrain (Fig. 7e). Overall, the activation of these pathways, which are central to synaptic plasticity, immune responses, and neuronal communication, likely underlies the disruptions in sensory perception and pain processing observed following NME. In contrast, the top downregulated pathways included the transcriptional regulatory network in embryonic stem cells and the adrenomedullin signaling pathway (Fig. 7d).

Fig. 7.

NME results in significant transcriptional changes in the midbrain during adolescence. A. PCA of the samples based on normalized mRNA expression level. b. Heat map showing the expression pattern of all DEGs from the sequenced samples. c. Volcano plot of significant DEGs. Red dots indicate upregulated genes; green dots indicate downregulated genes. d. The top significantly altered canonical pathways predicted by Qiagen Ingenuity pathway analysis. The orange bars indicate upregulated pathways, and the blue bars represent downregulated pathways in the NME adolescents’ midbrain tissues. e. Normalized counts of several significant DEGs from DESeq2. Data were analyzed using a student’s t-test and expressed as mean ± SEM. *p < 0.05. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

To validate the involvement of neuronal and immune signaling in mediating pain responses, markers of neuropathic and inflammatory pain were evaluated in the midbrain of NME and saline-exposed adolescents using an RT² Profiler PCR Array (Qiagen, PAMM-162Z). This array profiles the expression of 84 genes involved in the transduction, maintenance, and modulation of pain responses. Consistent with findings in the blood plasma, IL-1α and IL-10 were significantly upregulated in the midbrain of the NME adolescents compared to the saline-exposed controls (Sup Data 3). Additional elevated pro-inflammatory markers included Chemokine receptor 2 (Ccr2) and Colony Stimulating Factor 1 (Csf1). Genes coding for potassium inwardly rectifying channel subfamily J member 6 (Kcnj6) and 5-hydroxytryptamine (serotonin) receptor 1A (Htr1a), involved in neuronal transmission, were significantly downregulated in the midbrain of the NME adolescents compared to the saline-exposed controls (Sup Data 3). These data demonstrate significant upregulation of inflammatory markers IL-1α, IL-10, Ccr2, and Csf1, alongside a downregulation of neuronal transmission genes Kcnj6 and Htr1a in the midbrain of NME adolescents. In conclusion, these results validate the involvement of both neuronal and immune signaling in mediating pain responses following NME. The observed changes in gene expression and pathway activation underscore the complex interplay between gut dysbiosis, systemic inflammation, and neural mechanisms that contribute to prolonged pain hypersensitivity.

To investigate the role of the gut microbiota in mediating gene expression, bulk RNA sequencing was performed on the midbrains of the NME (Mor_pro) and saline-exposed (Sal_pro) adolescents treated with B. infantis. Remarkably, B. infantis administration completely mitigated all NME-induced transcriptional changes in the midbrain. PCA of normalized mRNA expression levels revealed no distinct separation between the transcriptomic profiles of the midbrain samples from NME and saline-exposed adolescents that received probiotics during the neonatal period (Fig. 8a). The heatmap further demonstrated the similarity expression pattern of the genes between the samples showing no DEGs between the two groups (Fig. 8b). Notably, genes associated with increased inflammation (e.g., Nfkb1, Il16, and Il22b), glutamatergic signaling, and GPCR neurotransmission (e.g., Grm4, Gria4, Camk4) which were significantly upregulated in response to NME, were no longer upregulated following B. infantis treatment (Fig. 8c).

Fig. 8.

Neonatal administration of the probiotic, B. infantis, mitigates all NME-induced transcriptional changes in the midbrain during adolescence. a. PCA of the samples based on normalized mRNA expression level. b. Heat map showing the similar expression pattern of the genes from the sequenced samples. c. Normalized counts of several genes of interest from DESeq2. Data were analyzed using a student’s t-test and expressed as mean ± SEM, ns: non-significant.

Lastly, to assess the impact of B. infantis on the midbrain transcriptome in the context or absence of NME, the NME (Mor_pro) and saline-exposed (Sal_Pro) adolescents treated with B. infantis were also compared to the NME (Mor) and saline-exposed (Sal) adolescents given vehicle. Distinct transcriptome profiles were observed across these groups, indicating unique gene expression patterns (Fig. 9a). Interestingly, neonatal administration of the probiotic, B. infantis, modulated midbrain gene expression differently in saline-exposed animals compared to NME-exposed animals during adolescence. First, to assess the impact of B. infantis on gene expression in the absence of NME, we compared the DEGs of the saline-exposed adolescents given probiotics (Sal_pro) and those given vehicle (Sal) (Fig. 9b). IPA revealed that B. infantis administration primarily resulted in an overall upregulation of several canonical pathways, including oxidative phosphorylation, neutrophil extracellular trap (NET) signaling, tRNA charging, D-myo-inositol-5-phosphate metabolism, and the endocannabinoid neuronal synapse pathway (Fig. 9d). These canonical pathways are integral to several crucial biological processes. For instance, oxidative phosphorylation, which occurs in the mitochondria, generates ATP through the electron transport chain, providing essential energy for cellular activities (Wang, 1967). The NET signaling pathway is a key immune response mechanism where neutrophils release DNA and antimicrobial proteins to trap and kill pathogens (Huang et al., 2022). tRNA charging, or aminoacyl-tRNA synthesis, involves attaching amino acids to their corresponding tRNA molecules, a critical step in protein synthesis (Gomez and Ibba, 2020); (Yao and Fox, 2013). Consequently, B. infantis administration is sufficient to upregulate processes that ensure energy production, immune defense, protein synthesis, and effective cellular communication. Next, to evaluate the impact of B. infantis on gene expression in the presence of NME, we compared the DEGs of Mor_pro and Mor (Fig. 9c). In the NME adolescents, B. infantis administration predominantly led to a significant downregulation of several canonical pathways, including immunogenic cell death (ICD) signaling pathway, telomerase signaling, unfolded protein response, phosphatidylinositol 3-kinase (PI3K)/Akt signaling, and 14-3-3 mediated signaling (Fig. 9e). These pathways are involved in cellular stress responses, inflammation, and survival signaling, and their downregulation suggests that B. infantis may mitigate the pathological effects of NME by reducing maladaptive cellular processes and restoring homeostasis.

Fig. 9.

Neonatal administration of probiotic, B. infantis, modulates midbrain gene expression differently in saline-exposed animals compared to NME-exposed animals during adolescence. a. Venn diagram of unique gene counts expressed in the midbrains of all NME and saline-exposed adolescents; those given B. infantis and those not given B. infantis neonatally. b. Volcano plot of the significant DEGs comparing saline-exposed adolescents given B. infantis and those not given B. infantis (Sal_Pro vs Sal). Red dots indicate upregulated genes; green dots indicate downregulated genes. c. Volcano plot of the significant DEGs comparing NME adolescents given B. infantis and those not given B. infantis (Mor_Pro vs Mor). d. The top significantly altered canonical pathways predicted by Qiagen Ingenuity pathway analysis comparing saline-exposed adolescents given B. infantis and those not given B. infantis (Sal_Pro vs Sal). The orange bars indicate upregulated pathways, and the blue bars represent downregulated pathways. e. The top significantly altered canonical pathways predicted by Qiagen Ingenuity pathway analysis comparing NME adolescents given B. infantis and those not given B. infantis (Mor_Pro vs Mor). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

4. Discussion

Previous preclinical studies have shown NME to result in prolonged pain hypersensitivity (Zhang and Sweitzer, 2008) (Craig and Bajic, 2015). However, the current study is the first to implicate the gut microbiome as a mediator of this effect. Using a murine model, we showed that the transplantation of the microbiome from NME adolescents was sufficient to induce pain hypersensitivity in naïve recipients. We also demonstrated that neonatal probiotic administration of B. infantis rescues both NME-induced gut dysbiosis and prolonged pain hypersensitivity during adolescence. Furthermore, our study demonstrates for the first time that NME-induced gut dysbiosis drives persistent systemic inflammation during adolescence, contributing to prolonged pain hypersensitivity, which can be mitigated with neonatal probiotic administration. Using bulk RNA sequencing, we further elucidated the impact of NME and gut microbiome modulation on the midbrain transcriptome, revealing specific alterations in genes and pathways involved in synaptic plasticity, neuronal communication, immune responses, and pain modulation in NME adolescents.

First, a clinically relevant murine model of NME was established, in which neonates were administered morphine or saline postnatally for one week. Accumulating evidence suggests that NME adversely affects neonatal neurodevelopment (Dührsen, et al., 2012); (Attarian et al., 2014). Consistent with this, our study demonstrated that NME in this model leads to reduced weight gain during the neonatal period, persisting into adolescence, irrespective of sex. Reduced weight gain represents a phenotype of delayed development, often clinically referred to as failure to thrive (FTT) in infants (Franceschi et al., 2021). Previous preclinical studies have corroborated these findings, reporting similar weight gain deficits associated with delayed development following early opioid exposure (Barr and Wang, 1992); (Robinson, et al., 2020). Similarly, clinical studies have linked NME in preterm infants to adverse outcomes, including lower weight and smaller head circumference observed at 5 to 7 years of age (Ferguson et al., 2012). Moreover, we confirmed that our murine model exhibited hypersensitivity to thermal and mechanical pain following NME, aligning with previous reports. For instance, Zhang et al., 2008 reported increased sensitivity to both thermal and mechanical pain in male and female Sprague Dawley (SD) rats during adolescence at postnatal day 30 following NME (Zhang and Sweitzer, 2008). Similarly, Craig et al., 2015 documented prolonged hypersensitivity to thermal pain in adult male and female SD rats following NME (Craig and Bajic, 2015). In accordance with these studies, our NME adolescents exhibited prolonged hypersensitivity to thermal and mechanical pain, warranting further investigation as heightened pain sensitivity is a strong motivator for substance self-administration.

Previous research has shown that NME induces gut dysbiosis that persists later in life (Antoine et al., 2022). To explore the role of the gut microbiome in driving prolonged pain hypersensitivity following NME, we analyzed gut microbiome diversity and composition in neonatally exposed female mice during adolescence. While no differences in alpha diversity were observed, significant differences in beta diversity were noted, with distinct clustering between NME and saline-exposed adolescents. These findings are consistent with several studies that have reported lasting alterations in the gut microbiome following both prenatal and postnatal opioid exposure (Abu, et al., 2021); (Antoine et al., 2022); (Grecco et al., 2021); (Lyu, et al., 2022); (Abu, et al., 2024). Although these studies have reported varying effects on alpha diversity—some have noted increased alpha diversity (Antoine et al., 2022); (Grecco et al., 2021), while others have reported no change (Abu, et al., 2021); (Lyu, et al., 2022); (Abu, et al., 2024) —all consistently identify profound changes in microbial composition that persisted into later life, including altered beta diversity and a reduction in commensal bacteria such as Lactobacillus (from phylum Firmicutes), (Abu, et al., 2021); (Abu, et al., 2024); Bifidobacterium (phylum Actinobacteria) (Abu, et al., 2024) and Turicibacter (phylum Firmicutes) (Antoine et al., 2022). In this study, we identified several differentially enriched bacterial taxa between NME and saline-exposed adolescents, including an increased abundance of Enterorhabdus, Eubacterium_ventriosum_group, Intestinimonas, and Clostridium ASF356, alongside a notable decrease in Romboutsia, Oscillibacter, and Turicibacter. The relative abundances of biofilm-forming bacteria, gram-negative bacteria, and potentially pathogenic bacteria were predicted to be significantly elevated in the NME adolescents relative to the saline-exposed adolescents. Several genera, such as Enterorhabdus, Eubacterium_ventriosum_group, Clostridium ASF356, and Oscillibacter, exhibited differences compared to previous studies, likely due to variations in the opioid used, treatment paradigms, sample collection, and the age at which animals were tested. Nevertheless, our data revealed the development of a dysbiotic microbiome following NME during adolescence.

Notably, serotonin (5-HT), a critical neurotransmitter for pain modulation, is primarily synthesized in the gut, with over 90 % of total body serotonin being produced by enterochromaffin cells (ECs) in the gastrointestinal tract (Mawe and Hoffman, 2013). Dysbiosis, as observed in NME, has been shown to significantly impact serotonin synthesis via gut microbial metabolites that signal EC cells to generate 5-HT through tryptophan hydroxylase 1 (TPH1) (Yano, et al., 2015). Germ-free mice and antibiotic-treated mice display significant reductions in peripheral 5-HT levels, which are reversed upon re-colonization with a normal microbiome, highlighting the gut microbiota’s critical role in modulating serotonergic pathways (Yano, et al., 2015). Dysregulation of peripheral 5-HT has been implicated in pain hypersensitivity and inflammatory diseases such as inflammatory bowel diseases (IBD), where increased EC cell activity and 5-HT production exacerbate inflammation (Shajib, et al., 2019); (Kwon, et al., 2019); (Haq, et al., 2021). Peripheral 5-HT, while unable to cross the blood–brain barrier (Agus et al., 2018), may influence central pain processing through peripheral sensitization (Loyd et al., 2013). In our study, the observed dysbiosis in NME adolescents, including an increase in genera like Enterorhabdus and Eubacterium_ventriosum_group, may contribute to altered serotonergic signaling, thereby driving heightened pain sensitivity.

Indeed, NME increased several genera. The genus Enterorhabdus includes species, such as Enterorhabdus mucosicola and Enterorhabdus caecimuris, which are gram-positive, non-spore-forming rods found in the mammalian gut that have been isolated from models of colitis and intestinal inflammation (Kakodkar and Mutlu, 2017); (Shlisky, et al., 2017). Moreover, an increased abundance of the Enterorhabdus genus has been implicated in the modulation of gut inflammation in a murine model of a Tryptophan-Deficient Diet (Yusufu, et al., 2021), suggesting a potential role for Enterorhabdus in immune activation and pain conditions observed in IBD. Another increased genus was the Eubacterium_ventriosum_group, which is part of a subcategory within the genus Eubacterium, a group of gram-positive bacteria in the family Eubacteriaceae. Several studies have linked Eubacterium_ventriosum_group to inflammation, IBD, and obesity (Kasai, et al., 2015); (Mukherjee et al., 2020). This group’s association with gut dysbiosis and systemic inflammation underscores its potential contribution to increased intestinal permeability and pain. Additionally, Intestinimonas, another genus increased by NME, was identified as an anti-inflammatory bacterium that produces butyrate, a short-chain fatty acid (SCFA) that maintains gut health and reduces inflammation (Cai, et al., 2020); (Zhuang, et al., 2020). However, the anti-inflammatory effects of Intestinimonas may be overwhelmed by the increase in the other genera driving inflammation. Finally, Clostridium ASF356 plays a role in metabolic interactions within the altered Schaedler flora (ASF) community (Biggs, et al., 2017). Clostridium ASF356 depends on certain amino acids provided by other community members to thrive, highlighting its reliance on a healthy gut microbiome diversity and its contribution to the community’s overall metabolic network (Brand, et al., 2015). Disruptions in the population of Clostridium ASF356 could potentially contribute to alterations in essential gut-brain axis pathways.

In addition to the increase of several genera, NME resulted in a significant decrease in other genera, including Romboutsia, Oscillibacter, and Turicibacter. Both Romboutsia and Oscillibacter have been associated with probiotic-induced changes in the gut microbiota. Elevated levels of Romboutsia have been linked to decreased severity of pancreatitis and sepsis in experimental models of acute pancreatitis (Geitsen and Gerritsen, 2015). Higher levels of Oscillibacter have been associated with lower cholesterol levels in both stool and blood, as well as with other markers indicative of reduced cardiovascular disease (CVD) risk, such as lower triglycerides and glucose, and higher high-density lipoprotein (HDL) (Li, 2024). These findings suggest that Oscillibacter plays a critical role in gut health and metabolic regulation. Another notable genus decreased by NME is Turicibacter, which possesses potential anti-inflammatory properties (Zhuang et al., 2018). Reduced Turicibacter abundance has been linked to inflammation and IBD (Zhuang et al., 2018). Furthermore, Turicibacter has been shown to modulate bile acids and host lipid metabolism, significantly altering lipid profiles in mice and reducing cholesterol, triglycerides, and fat tissue, which are crucial for managing conditions like obesity and CVD (Lynch, et al., 2023); (Jung, 2016). Overall, the balance of these gut bacteria is crucial in maintaining health and preventing disease, emphasizing the significant impact of an NME-induced dysbiotic microbiome on host health and the importance of maintaining a well-balanced gut microbiome.

Based on our hypothesis that the gut microbiome mediates hypersensitivity to pain, we sought to modulate the gut microbiome through FMT and probiotic intervention and assess the impact on somatic pain sensitivity. FMT from NME adolescents into naïve age-matched control mice resulted in decreased thermal thresholds, as assessed by both the tail flick and hot plate assays, demonstrating that the transplanted microbiome can influence pain sensitivity. Following antibiotic treatment and NME-microbiome transplantation, the microbiome profile of the recipient mice showed a persisting increase in Enterorhabdus and Clostridium ASF356 as well as a persisting decrease in the commensal genera Turicibacter similar to that observed in the donor NME adolescents. These findings suggest that these specific genera may predominantly drive NME-induced pain hypersensitivity.

Other murine models have also linked pathological changes in the gut microbiome to somatosensory pain, with studies demonstrating that targeting the gut microbiome through FMT can either induce or attenuate pain hypersensitivity. For example, Wang et al., 2018, from our group, identified Enterococcus faecalis as a biomarker of morphine-induced alteration of the gut microbiome following chronic morphine administration in adult wild-type mice (Wang et al., 2018). Infection with Enterococcus faecalis was sufficient to significantly increase pain sensitivity and augment morphine analgesic tolerance in infected mice compared to those treated with morphine alone (Wang et al., 2018). Lee et al., 2018 explored whether opioid-induced changes in the gut microbiota could influence inflammation-driven pain hypersensitivity and impaired reward behavior in adult wild-type mice (Lee et al., 2018). They found that in antibiotic-treated mice FMT from a saline control microbiota, but not a morphine-associated microbiota, restored microglial activation state and behaviors, including a rescue in pain hypersensitivity (Lee et al., 2018). Additionally, mice fed a Western diet that developed obesity-induced pain hypersensitivity showed attenuation through FMT of the gut microbiome from lean mice (Bonomo, et al., 2020). This improvement was accompanied by increased nerve density in the skin and elevated circulating butyrate levels in the blood (Bonomo, et al., 2020). Beyond somatosensory pain, studies have also implicated the gut microbiome to visceral pain hypersensitivity. For instance, Lucarini et al 2022 demonstrated in a 2,4-dinitrobenzene sulfonic acid (DNBS)-induced colitis mouse model that persistent abdominal pain could be either recapitulated in naïve rats or rescued in the DNBS-treated rats via FMT (Lucarini, et al., 2022). While visceral pain can share mechanisms with somatic pain syndromes, crucial differences exist in central nociceptive processing between these pain types (Sikandar and Dickenson, 2012). However, gut inflammation following NME could plausibly result in abdominal pain that contributes to the priming of somatic pain sensitivity. Collectively, these studies highlight the ability of gut microbiome modulation via FMT to influence pain sensitivity, regardless of pain type or origin.

In our study, we also modulated the gut microbiome through the neonatal administration of the probiotic Bifidobacterium infantis (B. infantis), a dominant member of the gut microbiota of breastfed infants known to produce several SCFAs, including acetate and lactate (Chichlowski et al., 2020). B. infantis plays a crucial role in modulating host physiology by accelerating immune responses, balancing the immune system to suppress inflammation, and improving intestinal barrier function (Chichlowski et al., 2020; Underwood et al., 2015). Specifically, acetate produced by B. infantis enhances the defense functions of host epithelial cells (Fukuda, et al., 2011) and serves as a carbon source for the growth and function of other microbiota that produce beneficial SCFAs, such as butyrate (Duncan et al., 2004). Beyond their effects in the gut, SCFAs enter systemic circulation, influencing other organs, including the brain, and inducing overall metabolic benefits (Koh et al., 2016). For instance, lactate can cross the blood–brain barrier and modulate neuronal activity in the cortex (Bozzo et al., 2013). Moreover, probiotic interventions in adults using formulations like the VSL#3 mixture, which contains multiple strains of Lactobacillus and Bifidobacterium (including B. infantis), have demonstrated anti-inflammatory effects and attenuation of abdominal pain in chronic IBD (Cheng, et al., 2020); (Boonma, 2021). In our study, neonatal administration of B. infantis rescued NME-induced gut dysbiosis and resulted in a significant increase in several commensal genera, including Lactobacillus, Bacteroides, and Turicibacter, in NME adolescent mice. This suggests sufficient and lasting engraftment from early probiotic administration. Additionally, SCFAs, such as acetate and propionate, are produced by strains of the Lactobacillus genera (Sivieri et al., 1014); (Meimandipour, et al., 2010) and the Bacteroides genera (O’Riordan, et al., 2022), which could contribute to the observed therapeutic effects. More importantly, neonatal B. infantis administration significantly increased thermal and mechanical pain thresholds in NME adolescent mice, effectively rescuing NME-induced pain hypersensitivity. These findings align with prior evidence linking gut microbiota interventions to the mitigation of pain hypersensitivity (Lei et al., 2017); (Pedersini et al., 2021); (Crawford et al., 2024). For instance, Crawford et al (2024), demonstrated that a ketogenic diet that restored SCFAs-producing bacteria, such as Bacteroides, Lachnospiraceae, and Ruminococcaceae, mitigated OIH in adult mice (Crawford et al., 2024). Furthermore, supplementation with SCFAs—including acetate, butyrate, and propionate—delayed the onset of OIH, highlighting the role of microbial metabolites in modulating pain signaling (Crawford et al., 2024). Collectively, these findings emphasize the critical role of the gut microbiome in driving NME-induced prolonged pain hypersensitivity and suggest that probiotic interventions targeting SCFA-producing bacteria may offer therapeutic potential.

While this work has established a critical role for the gut microbiome in NME, we sought to further explore the mechanisms by which the gut microbiome mediates central and peripheral pain processes. Chronic opioid exposure in adult mice is well documented to cause gut dysbiosis, disrupt the intestinal barrier, and promote systemic inflammation (Banerjee, et al., 2016), which can drive pain conditions (Guo et al., 2019). This phenomenon occurs through a disruption of the intestinal epithelium, increased gut permeability, and translocation of bacterial products, which in turn induce sustained inflammatory factor release, decreasing nociceptive thresholds (Guo et al., 2019); (Meng, et al., 2013); (Banerjee, et al., 2016). The intestinal barrier is a critical player in gut homeostasis maintenance as it enables the communication between gut microbes, both commensals and pathogens, and the host immune system. The intestinal barrier consists of a self-renewing epithelium organized into crypts and villi, including both stem cells and differentiated cells that allow the epithelium to perform multiple physiological activities (Stolfi et al., 2022). Previous reports have shown that the intestinal barrier of adolescent mice is significantly more permeable than that of adults due to altered villi architecture and tight junction localization in the developing gut tissue (Gleeson et al., 2021). We investigated gut morphology and gut permeability in our studies to confirm these findings and to determine whether NME further exacerbated such permeability. Our results confirmed enhanced permeability during adolescence, which can facilitate the translocation of bacterial products out of the gut into other parts of the body. Although NME did not significantly exacerbate gut permeability, it significantly disrupted the morphology of the small intestine, manifested mainly by shortened and broken villi as well as shortened crypt cells, indicative of gut inflammation. These alterations signal a disruption in the intestinal epithelium and a resultant decrease in its ability to protect the host from luminal antigens, pathogens, and toxins. Consequently, sustained immune activation leading to systemic inflammation may occur during adolescence.

Indeed, using a multiplex assay, our studies uncovered NME-induced systemic inflammation, manifested by significant elevations in the cytokines IL-1α, IL-12p70, IFN-y, and IL-10 in the blood plasma of NME adolescent mice, which was rescued with neonatal probiotic administration. Accumulating studies have shown a strong correlation between increased levels of cytokines and heightened pain states across various diseases. Key cytokines implicated in these processes include IL-1α, IL-1β, TNF-α, IL-12, IFNy, and IL-10. In particular, IL-1β and TNF-α have been found to play crucial roles in neuropathic pain and chronic pain-related depression through microglial activation in the brain, such as in the hippocampus (Mokhtari et al., 2019). In a mouse model of spared nerve injury, the administration of an IL-1 receptor antagonist alleviated the effects of neuropathic pain on depression-like behaviors (Norman et al., 2010). Interestingly, IL-1α is associated with inflammation at mucosal barrier surfaces, as IL-1α-deficient mice, but not IL-1β-deficient mice, were found to be protected from acute colitis induced by oral feeding of cytotoxic agent dextran sodium sulfate (DSS) (Bersudsky, et al., 2014); (Malik, et al., 2017).

Further, in line with our findings, others have observed IL-12 and IFN-γ to be prominent pro-inflammatory cytokines that positively regulate inflammation and pathological pain (Verri, et al., 2005) (Ullrich et al., 2020); (Tsuda et al., 2009). IL-12 is produced mainly by monocytes and macrophages as an early host response to bacteria, bacterial products, and intracellular parasites and induces IFN-γ production by CD4⁺ Th1 cells, CD8⁺ cytotoxic T cells, and activated natural killer (NK) cells (Trinchieri, 1995); (D’Andrea, et al., 1992). IFN-γ, in turn, acts on antigen-presenting cells (APCs) to augment IL-12 secretion in a positive feedback loop (Grohmann, et al., 2001). In a rat model, IL-12 administration alone was sufficient to induce mechanical hyperalgesia, but not thermal hyperalgesia, mediated by action on the endothelin receptor (Verri, et al., 2005). Similarly, intrathecal administration of IFN-γ alone in naïve rats induced long-lasting mechanical hyperalgesia by activating spinal microglia (Jung, 2016). Furthermore, we found a significant increase in IL-10, which is an anti-inflammatory cytokine that is produced by almost all cells of the innate and adaptive arms of the immune system. Previous research has supported the role of IL-10 in modulating pain through its interaction with immune cells and neurons, further linking immune responses to nociceptive pathways (Vanderwall and Milligan, 2019). The major role of IL-10 is to limit the extent of activation of both the innate and adaptive immune cells to maintain a homeostatic state and prevent tissue damage (Iyer and Cheng, 2012); (Ma, et al., 2015). In cases where invading pathogens resist clearance, IL-10 may be produced to help reduce inflammation and thereby minimize pathology, as demonstrated in cases of infection by the virulent SD strain of L. major that drives excessive IL-10 response (Anderson et al., 2007). Thus, elevated levels of IL-10 in our mice further signal the presence of invading bacterial products and sustained immune responses following NME.

Notably, the gut microbiota can regulate the development and function of a variety of immune cells including regulatory T cells and cytotoxic T cells, which produce the key cytokines referenced here (Zheng et al., 2020); (Kabat et al., 2014). Interestingly, in our model, probiotic administration of B. infantis decreased the levels of these cytokines and restored their expression to near-basal levels. Indeed, colonization by B. infantis EVC001 in preterm babies has been shown to modulate enteric inflammation and decrease levels of several proinflammatory cytokines such as IL-1β, IL-22, and IFN-y (Henrick, et al., 2019). Various studies have shown that B. infantis in infants reduces susceptibility to diseases driven by unwanted T-cell responses (Vogel, et al., 2023), suggesting that B. infantis constitutes a promising preventive and interventional treatment option. Collectively, our findings and others suggest that a disrupted intestinal barrier combined with a dysbiotic microbiome, resulting in sustained systemic inflammation, underlie the increased pain observed following NME and that targeting cytokines like IL-1α, IL-1β, IFN-y, IL-12, and IL-10 to homeostatic levels could be viable strategies for managing chronic pain.

Lastly, we assessed transcriptomic changes in the midbrains of NME and saline-exposed adolescents, including those given B. infantis neonatally. NME resulted in the upregulation of genes associated with immune activation and inflammation (e.g. Nfkb1, Il16, Il22b) in the midbrain. These findings are consistent with previous studies reporting increased central inflammation in prolonged pain conditions (Ji et al., 2018) and following chronic opioid use in the human midbrain (Wei, et al., 2023). The upregulated inflammation-related genes in the midbrain may impair PAG function, as chronic inflammation has been shown to alter the efficacy of its descending inhibitory pathways (Zhuang, et al., 2016). Additionally, we observed the upregulation of genes involved in glutamatergic signaling, calcium signaling, and GPCR signaling (e.g., Grm4, Gria4, Camk4) in the midbrain of NME adolescents. Glutamate, the primary excitatory neurotransmitter in the CNS, plays a critical role in transmitting pain signals from the periphery to the brain. Upon neuronal stimulation, glutamate is released into the synaptic cleft, activating both fast-acting ionotropic and slower-acting metabotropic receptors (Aiyer et al., 2018); (Temmermand et al., 2022). Following NME-induced gut dysbiosis and the resulting persistent systemic inflammation, initial peripheral changes, such as the upregulation of inflammatory mediators and nociceptor sensitization, can lead to abnormal neuronal firing and ectopic activity in peripheral afferent neurons. This peripheral sensitization can contribute to central sensitization, characterized by increased glutamatergic signaling and ongoing pain disorders. Chronic pain conditions are linked to the dysregulation of glutamate receptors and transporters (Temmermand et al., 2022), and elevated extracellular glutamate levels have been associated with worsened pain symptoms (Osikowicz et al., 2013). Consistent with the gene expression changes, IPA revealed that NME resulted in the significant upregulation of canonical pathways, such as opioid signaling pathway, CREB signaling in neurons, GPCR signaling, neuropathic pain signaling in dorsal horn neurons, calcium signaling, synaptic long-term potentiation, synaptic long-term depression, and multiple sclerosis signaling pathway. These findings are supported by previous studies that identified similar pathway enrichment in the forebrains of neonates on PND 14 such as opioid signaling pathway, GPCR signaling, CREB Signaling pathway, and phagosome formation (Dunn, 2023), as well as in the brainstems of neonates on PND 15 following NME, such as oxidative phosphorylation, focal adhesion, protein digestion, and absorption (Borrelli, et al., 2021). Furthermore, a recent study investigating the transcriptomic profile of the nucleus accumbens (NAc) at PND 14 following in-utero and postnatal oxycodone exposure found that pathways involved in the regulation of neurological system processes, forebrain and neural crest development, and synaptic transmission, were affected (Odegaard et al., 2020).

Interestingly, all the dysregulated genes and pathways in our model following NME, including those associated with increased central inflammation and excitatory signaling, were rescued with neonatal probiotics administration of B. infantis, highlighting the critical role of the gut microbiome in these processes. In the context of NME, B. infantis administration resulted in significant changes in midbrain gene expression, which likely underlie its ability to mitigate persistent pain hypersensitivity. In NME adolescents, B. infantis administration led mainly to a significant downregulation of several canonical pathways, including ICD signaling pathway, telomerase signaling, unfolded protein response, PI3K/Akt signaling, and 14-3-3 mediated signaling. ICD, a regulated cell death pathway, involves the release of damage-associated molecular patterns (DAMPs) by dying cells that interact with the host immune system to activate both innate and adaptive immune responses (Kroemer et al., 2022). ICD is also activated in response to unfolded protein stress to restore endoplasmic reticulum homeostasis. Additionally, the PI3K/Akt signaling pathway and 14-3-3 mediated signaling are critical for cell growth, survival, and apoptosis, particularly with 14-3-3 proteins implicated in various neurological disorders due to their release from damaged neurons (Foote and Zhou, 2012). Thus, B. infantis administration in NME adolescents primarily led to the inhibition of pathways associated with cell death, unfolded protein response, and immune activation, which facilitated cell survival, decreased inflammation, and rescued homeostatic functioning. Taken together, these findings highlight the long-term neurobiological deficits associated with early opioid exposure and underscore the therapeutic potential of the gut microbiome in promoting brain plasticity and functional recovery.

Despite our significant findings, this study has limitations that should be acknowledged. First, while the murine model provides valuable insights into the effects of NME and gut microbiome alterations, the translatability of these results to humans is vital. Therefore, clinical studies in human populations are essential to validate these preclinical findings and explore the feasibility of probiotic interventions as a preventive or therapeutic measure for individuals exposed to opioids neonatally. Second, while we identified significant changes in midbrain transcriptome, the precise cellular and molecular mechanisms within specific midbrain cell types remain unclear. Future research will aim to elucidate these pathways through targeted molecular and functional studies. Lastly, while accumulating evidence suggests that early opioid exposure can negatively impact cognitive and emotional behaviors (Al-Mouqdad et al., 2020) (Luzzati, et al., 2023), investigations into the role of the gut microbiome in these alterations are limited. Our current study investigated NME-induced gut dysbiosis, which we linked to heightened pain sensitivity. Given the well-established role of the gut-brain axis in modulating cognitive and emotional processes, we hypothesize that NME-induced dysbiosis may similarly contribute to the other behavioral changes following NME. Thus, future studies will aim to explore the involvement of the gut microbiome in altered cognitive and emotional behaviors, expanding our understanding of the broader consequences of NME.

5. Conclusions

In conclusion, our findings highlight the significant role of the gut microbiome in mediating prolonged pain hypersensitivity following NME. The observed changes in microbial composition and the persistent peripheral and central inflammation during adolescence underscore the importance of targeting the gut microbiome for therapeutic intervention. Probiotic administration offers a promising approach to mitigate these effects, emphasizing the potential of gut-brain axis modulation in improving long-term outcomes for individuals exposed to opioids neonatally.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.bbi.2025.01.021.

Data availability

The authors confirm that the data supporting the findings of this study are available within the article and its supplementary materials.