The human gut microbiome and sleep across adulthood: associations and therapeutic potential

Abstract

Sleep is an essential homeostatic process that undergoes dynamic changes throughout the lifespan, with distinct life stages predisposed to specific sleep pathologies. Similarly, the gut microbiome also varies with age, with different signatures associated with health and disease in the latest decades of life. Emerging research has shown significant cross-talk between the gut microbiota and the brain through several pathways, suggesting the microbiota may influence sleep, though the specific mechanisms remain to be elucidated. Here, we critically examine the existing literature on the potential impacts of the gut microbiome on sleep and how this relationship varies across adulthood. We suggest that age-related shifts in gut microbiome composition and immune function may, in part, drive age-related changes in sleep. We conclude with an outlook on the therapeutic potential of microbiome-targeted interventions aimed at improving sleep across adulthood, particularly for individuals experiencing high stress or with sleep complaints.

Introduction

Sleep is an essential biological process encompassing behavioral, experiential, and physiological dimensions (Andrillon and Oudiette 2023). Disrupted sleep results in adverse health outcomes (Nayak and Anilkumar 2024, Patel et al. 2024), and prolonged sleep deprivation (SD) leads to death (Vaccaro et al. 2020). Both external and internal factors, including psychological stress, influence sleep, with emerging research showing a connection between the gut microbiome and sleep (Matenchuk et al. 2020, Han et al. 2022, Wang et al. 2022, Dos Santos and Galiè 2024). Importantly, improving sleep quality across the lifespan offers significant health benefits and reduces mortality rates (Ramar et al. 2021).

Sleep and the gut microbiome are strongly linked in the first few months of life, with gut microbiome profiles and sleep/wake patterns predicting behavioral development (Schoch et al. 2022). Indeed, evidence suggests that optimizing sleep via the gut, starting in early life and extending into adulthood, may provide long-lasting benefits (Schoch et al. 2022). For example, targeting the gut microbiome with pre- and probiotic interventions appears to be effective in supporting sleep in adults (Nishida et al. 2019, Fei et al. 2023, Santamarina et al. 2024). Given the lifelong interplay of lifestyle, stress, environmental, and physiological factors that can affect sleep (Li et al. 2018, Yang et al. 2018, Silva-Costa et al. 2021), understanding how sleep can be improved throughout the lifespan remains crucial. In the present review, we first provide a brief overview of how sleep, the gut microbiome, and the immune system change across adulthood, highlighting key future research directions that could further our mechanistic understanding of the gut–sleep relationship. Although the gut microbiome–sleep relationship appears to be bidirectional (Sen et al. 2021), we focus primarily on how the gut microbiome influences sleep, rather than vice versa, with the aim of advancing microbiome-based therapies to improve sleep. We discuss observational studies examining the gut microbiome and sleep of young, middle-aged, and older adults, and expand on this discussion with intervention studies showing the potential of pre- and probiotic interventions in supporting sleep quality.

Quality sleep is characterized by sufficient duration, lack of disruption, and absence of sleep disorder (Ramar et al. 2021), and faces different threats throughout adulthood. Early adulthood (late teens to mid-forties) is marked by substantial developmental milestones, including the establishment of autonomy, career paths, and intimate relationships, all of which shape evolving life structure (Aktu and İlhan 2017). However, the competing demands of work and family roles can disrupt sleep patterns. Gao et al. reported that 85% of 534 surveyed students slept <7 h per night, indicating a substantial burden of suboptimal sleep during early adulthood (Gao et al. 2022). Work–family conflicts are associated with sleep complaints (Silva-Costa et al. 2021) and, along with high job stress, increased insomnia risk (Yang et al. 2018). Stress disrupts normal sleep, with the extent of disruption varying based on individual sleep reactivity to stress (Kalmbach et al. 2018). Interestingly, 72.85% of Gao et al.’s participants reported frequent insomnia alongside gastrointestinal symptoms, and consistent sleep schedules appear to be a critical predictor of gut microbiome health (Kado 2024). In older adults, polypharmacy and multimorbidity increase the prevalence of sleep disorders (Li et al. 2018, Kim et al. 2022). Additionally, the sleep of older adults tends to be shorter and more easily disrupted compared to younger adults (Ohayon et al. 2004), with the timing of these sleep changes coinciding with age-related shifts in gut microbiome composition (Ghosh et al. 2022a).

Unlike the genome, the microbiome’s composition and function can be modulated through dietary, prebiotic, and probiotic interventions (Rothschild et al. 2018, Hitch et al. 2022, Ross et al. 2024). Although our understanding of the gut microbiome’s functional outputs is still emerging, it is increasingly clear that these microbial products are critical for human health (Postler and Ghosh 2017, Krautkramer et al. 2021, Zheng et al. 2022, Zhang et al. 2023). The human gut microbiome produces thousands of bioactive compounds, such as neurotransmitters, organic acids, lipid vesicles, and proteins, that influence the host immune system, interact with host receptors, and circulate throughout the body (Postler and Ghosh 2017, Cryan et al. 2019). One study found that the human gut microbiome significantly contributes to cross-sectional variation in 44% of measured blood metabolites, independently of host genetics (Diener et al. 2022). Microbial metabolism of amino acids (e.g. tryptophan) has been linked to numerous central nervous system (CNS) functions, including sleep (Gao et al. 2020). Indeed, research on the gut–brain axis has revealed multiple communication routes between microbial outputs and the CNS, including vagal afferents, immune responses, and the hypothalamic–pituitary–adrenal (HPA) axis, giving rise to the concept of “psychobiotics,” or pre/probiotic interventions supporting brain health (Cryan et al. 2019). These findings suggest that the gut microbiome may influence sleep through a variety of biological pathways, making it a promising target for interventions that improve sleep quality.

Quantifying sleep

Several tools are used to quantify physiological and neurological parameters of sleep. Physiologically, sleep is divided into different stages—rapid eye movement (REM) and non-REM (NREM) stages 1, 2, and 3—each characterized by specific brain waveforms recorded by scalp electroencephalogram (EEG). Changes in muscle tone, eye movement, heart and breathing rates, and blood oxygen levels also characterize these sleep stages, and can be monitored simultaneously alongside EEG with polysomnography (PSG; Andrillon and Oudiette 2023, Nayak and Anilkumar 2024, Patel et al. 2024). Sleep cycles through the different sleep stages over the course of a night, and this pattern of sleep stages is known as sleep architecture (Nayak and Anilkumar 2024), where alterations are associated with a variety of health and disease indicators (Ujma and Bódizs 2024). Delta power, a measure of delta wave activity derived from EEGs using Fourier analysis, serves as an indicator of homeostatic sleep drive, and it strongly correlates with NREM sleep intensity and duration (Davis et al. 2011).

Actigraphs—wrist-worn devices monitoring movement—are another objective sleep assessment tool. These devices estimate total sleep time, number of wake events, wake after sleep onset (WASO), sleep efficiency (the ratio of total sleep time to time spent in bed), and sleep latency (Martin and Hakim 2011). Actigraphy offers a key advantage over gold-standard EEG or PSG by enabling continuous at-home sleep assessment (Martin and Hakim 2011). Limitations include overestimation of total sleep time due to reliance on movement and less accuracy in estimating sleep latency than EEG, particularly in individuals with sleep disorders (Martin and Hakim 2011). Compared to PSG, clinical-grade actiwatches overestimate total sleep time and sleep efficiency for shorter nights while underestimating sleep latency and WASO, particularly when these measures are large (Chinoy et al. 2021).

Beyond objective measurements, validated questionnaires like the Pittsburgh Sleep Quality Index (PSQI) and Epworth Sleepiness Scale (ESS) are widely used to assess perceived sleep quality and daytime sleepiness, respectively, in both clinical and research settings (Fabbri et al. 2021). These questionnaires measure sleep parameters distinct from each other (Mondal et al. 2013) and from those measured by objective tools (Buysse et al. 2008), highlighting the importance of utilizing both tools for a comprehensive assessment (Fabbri et al. 2021).

Sleep, the gut microbiome, and the immune system across the lifespan

The gut microbiome is in a dynamic, reciprocal relationship with the host immune system, starting from birth (Fig. 1) (Postler and Ghosh 2017, Wernroth et al. 2022). Early gut microbial colonization is essential for normal immune development, influenced by factors such as delivery mode, birth weight, and antibiotics (Wernroth et al. 2022). Colonization disruptions before 2–3 years of age have been linked to an increased risk of chronic immune diseases later in life (Wernroth et al. 2022). Similarly, early life sets the stage for future sleep quality, with one study showing that sleep disturbances at 16 years old predicted sleep disturbances as late as 42 years old, independently from depression (Dregan and Armstrong 2010). Furthermore, the gut microbiomes of children with sleep-disordered breathing show a pro-inflammatory signature (Collado et al. 2019, Valentini et al. 2020). Together, these findings point to early life as a critical window that shapes future sleep, marked by the concurrent maturation of the gut microbiome, immune system, and sleep.

The gut–immune–sleep connection remains relevant beyond early life. In young to middle-aged adults (18–45), psychological stress strongly contributes to sleep disruptions (Kalmbach et al. 2018), and insomnia is frequently reported alongside gastrointestinal symptoms (Gao et al. 2022). Psychological stress increases stress-related hormones, hyperactivating the HPA axis, and triggering a systemic cascade of pro-inflammatory cytokines such as Tumor Necrosis Factor α (TNF-α) and Interleukin 6 (IL-6). This inflammatory cascade downregulates tight junctions in the gut lining and blood–brain barrier, and facilitates gut dysbiosis and lipopolysaccharide (LPS) translocation (Liu et al. 2017b, Geng et al. 2020). These changes can disrupt sleep by amplifying inflammation and stress responses in a feedback loop (Fig. 1). For instance, LPS can elevate TNF-α by binding to Toll-like receptor 4 on macrophages, which further stimulates the release of corticotropin-releasing factor in the hypothalamus where sleep-regulatory circuits reside (Liu et al. 2017b, Geng et al. 2020). Microbial metabolites such as short-chain fatty acids (SCFAs) also play a critical role in balancing pro- and anti-inflammatory cytokines (Vinolo et al. 2011). In turn, the immune system can shape the gut microbiome through inflammation, lumenal redox disruption, secreted antibodies, and antimicrobial peptide production (Bosco and Noti 2021).

Dysregulation of the host immune system can occur in otherwise healthy older adults (Franceschi et al. 2018, Nagpal et al. 2018). So-called “inflammaging” is a chronic, low-grade inflammatory state marked by age-associated increases in systemic levels of pro-inflammatory cytokines such as TNF-α, IL-1β, and IL-6 (Franceschi et al. 2018, Nagpal et al. 2018). Coinciding with the timing of the onset of inflammaging signatures, the gut microbiome shifts, as early as 40–50 years old, depending on the host’s health (Fig. 1) (Bosco and Noti 2021, Wilmanski et al. 2022). Changes to sleep architecture also occur through this period and do not stabilize until around 60 years old (Ohayon et al. 2004). Age-associated changes in the gut microbiome can be partitioned into healthy versus unhealthy, disease-associated aging trajectories (Fig. 1). On one hand, healthy gut microbiome aging signatures are associated with disease-independent health metrics, and include increasing beta-diversity uniqueness (a measurement of inter-individual microbiome difference) (Wilmanski et al. 2021, Ghosh et al. 2022b) and a decline in core Gram-negative anaerobes, such as Bacteroides and Prevotella (Wilmanski et al. 2021). On the other hand, signatures of unhealthy gut microbiome aging are broadly characterized by increased abundances of pathobionts and a maintenance of certain core taxa that tend to dominate in younger individuals (Biagi et al. 2016, Wilmanski et al. 2021, Ghosh et al. 2022a). Unhealthy gut microbiome aging may contribute to inflammaging (Bosco and Noti 2021), and although this relationship is not clear, work in animal models has shown that gut dysbiosis induces inflammaging signatures (Clark et al. 2015, Thevaranjan et al. 2017).

Interestingly, many cytokines play important roles in sleep regulation (Krueger 2008). In particular, nearly three decades of preclinical research establishes TNF-α as a sleep-relevant molecule that acts directly on sleep regulatory circuits in the brain (Rockstrom et al. 2018). Human studies mirror preclinical findings on the role of TNF-α in sleep regulation. For example, in HIV+ men, who frequently present with fragmented sleep and fatigue, nocturnal serum TNF-α levels positively correlate with delta power, and this correlation becomes weaker with infection progression (Darko et al. 1995). Furthermore, TNF-α inhibitor treatment in obstructive sleep apnea (OSA) patients greatly reduced excessive daytime sleepiness (Vgontzas et al. 2004), and treating rheumatoid arthritis patients with recombinant Tumor Necrosis Factor Receptor (TNFR) improved their percentage of N2 sleep, total sleep time, and sleep efficiency (Detert et al. 2016).

Although more work is needed to characterize chronically elevated levels of systemic TNF-α on sleep (Rockstrom et al. 2018), several studies show that systemic TNF-α can affect sleep. TNF-α can cross the blood–brain barrier in rodents (Gutierrez et al. 1993), and systemic administration of TNF-α induces brain TNF-α expression that disrupts sleep (Kubota et al. 2001, Zielinski et al. 2013). These effects are reduced by subdiaphragmatic vagotomy, highlighting the role of vagal afferents (Kubota et al. 2001, Zielinski et al. 2013). Importantly, the gut microbiome can influence systemic cytokine levels, such as TNF-α (Mossad and Blank 2021), and impact microglia activity in the brain, although the detailed mechanisms remain unclear (Wang et al. 2018). Additionally, gut microbes can produce neurotransmitters, such as serotonin, GABA, dopamine, and histamine, which are implicated in sleep regulation, though how these molecules interact with the CNS from the gut to influence sleep remains poorly understood (Wang et al. 2022, Lin et al. 2024). As reviewed in Lin et al. and Wang et al., gut serotonin may signal to the brain via serotonergic vagal afferents, but how this signaling connects to sleep-regulatory circuits is unclear. Similarly, GABA-producing microbes, such as Lactobacilli, have shown promise in improving sleep quality, yet the mechanisms underpinning these effects remain elusive. Microbially produced dopamine and histamine may influence sleep by interacting with sleep-regulatory circuits, potentially through systemic circulation to the brain via the hepatoportal system, highlighting the need for targeted mechanistic research to clarify these pathways (Wang et al. 2022, Lin et al. 2024).

Gut–sleep observational studies across adulthood

Observational studies in young to middle-aged adults (18–45) report alterations to the gut microbiome associated with different sleep parameters. For example, one study showed that insomniac adults had decreased F/B ratio (ratio of the relative abundances of the Firmicutes and Bacteroidetes phyla), alpha-diversity (measures of intra-individual microbiome diversity), and beta-diversity (measures of inter-individual differences in microbiome composition) compared to healthy adults (Liu et al. 2019). Similarly, poor sleep quality (PSQI) has been associated with a lower F/B ratio (Grosicki et al. 2020), and lower F/B ratios have been linked to intestinal inflammation (Stojanov et al. 2020). An actigraphy study found that sleep efficiency, fewer WASO events, and total sleep time were positively associated with alpha-diversity (Smith et al. 2019). Smith et al. found that time in bed and total sleep time were positively associated with IL-6, and that IL-6 was positively associated with alpha-diversity. Consistent with gut–immune–sleep interplay, both Smith et al. and Grosicki et al. found that specific Lachnospiraceae genera, including Blautia, Oribacterium, and Ruminococcus, were associated with sleep quality, though the directionality of these associations varies across studies (Table 1) (Smith et al. 2019, Grosicki et al. 2020). Notably, Lachnospiraceae contains major butyrate producers (Singh et al. 2023), and butyrate has anti-inflammatory effects (Siddiqui and Cresci 2021). Although most butyrate is metabolized in the colon, some can enter systemic circulation through the hepatoportal vein, where it can then cross the blood–brain barrier (Siddiqui and Cresci 2021). Butyrate can directly inhibit histone deacetylases and downregulate NF-κB signaling in various cell types, including intestinal and immune cells, which allows it to modulate the production of pro-inflammatory cytokines, such as TNF-α (Siddiqui and Cresci 2021). Furthermore, butyrate can activate G-protein coupled receptors (GPCRs), including GPR41, GPR43, and GPR109A (Siddiqui and Cresci 2021). These GPCRs are distributed across several cell types and tissues where they regulate immune responses and maintain intestinal homeostasis (Siddiqui and Cresci 2021).

Levels of butyrate may be influenced by sleep (e.g. through temporally aberrant production due to eating during times when humans normally sleep, like in the case of shift workers), and butyrate itself may directly influence sleep. For instance, in healthy students, SD and sleep restriction (SR) disrupted butyrate-producing bacteria, including taxa within the Lachnospiraceae family and the Faecalibacterium genus, reducing butyrate levels by 53.1% in SD and 30.7% in SR (Gao et al. 2022). Interestingly, SD mice that were fed butyrate were protected from adverse gut microbiome and intestinal changes (Gao et al. 2022). When butyrate was administered to healthy mice, NREM sleep increased by 50% and 70%, respectively, depending on oral versus intraportal administration (Szentirmai et al. 2019). One study in humans reported that sodium butyrate supplementation improved PSQI scores, decreased inflammatory biomarkers, and upregulated circadian clock genes in patients with ulcerative colitis (Firoozi et al. 2024). These studies suggest that while SD and SR can adversely alter the gut microbiome, butyrate supplementation can protect against these changes and promote better sleep, possibly by modulating TNF-α signaling in the brain via entering systemic circulation through the hepatoportal system. Further research is needed to elucidate the specific mechanisms involved.

Observational studies in older adults (45+) with OSA highlight specific gut bacterial taxa with different age-dependent effects. These studies offer insights into the interplay between the gut microbiome, immune system, and sleep, given that OSA has immunological components (Nadeem et al. 2013). One study found that OSA was associated with several taxa including depletion of Alistipes and Ruminococcus gnavus, along with enrichment of Coprococcus (Lu et al. 2022). Similarly, Baldanzi et al. showed that Bacteroidales (including the genus Alistipes) negatively correlated with two OSA parameters: the percentage of time asleep with oxygen saturation below 90% (T90) and the number of times per hour that oxygen saturation falls at least 4% below baseline (ODI). In contrast, Coprococcus comes, R. gnavus, and Collinsella aerofaciens positively correlated with both T90 and ODI (Baldanzi et al. 2023). Coprococcus is a core butyrate-producing genus, typically associated with health in youth, and is depleted with age (Ghosh et al. 2022a). Interestingly, the decline of Coprococcus in older healthy adults is typically less severe than in those with health issues (Ghosh et al. 2022a), indicating that there might be an optimal level of Coprococcus for healthy aging. Alternatively, the enrichment of Coprococcus observed in the OSA studies (Lu et al. 2022, Baldanzi et al. 2023) may result from the sleep disorder itself, rather than underpinning its etiology. More work is needed to understand the roles Coprococcus plays in host sleep quality with age. Similarly, the role of R. gnavus in host physiology is complex, as shown by its conflicting associations with OSA (Table 1) (Lu et al. 2022, Baldanzi et al. 2023), and links to both anti- and pro-inflammatory effects (Crost et al. 2023). In the context of Crohn’s disease, R. gnavus produces a pro-inflammatory polysaccharide that induces TNF-α expression in dendritic cells (Henke et al. 2019). In contrast, another study found that R. gnavus was positively associated with longevity, suggesting it may have beneficial health effects (Wang et al. 2019). The odds ratio increased significantly after controlling for smoking, body mass index (BMI), food preference, and alcohol consumption, which suggests that these lifestyle factors may dampen the effect of R. gnavus on longevity (Wang et al. 2019). Further studies that consider R. gnavus within the broader context of the gut microbiome ecology, host physiology, and environmental factors will need to be conducted to understand its contradictory roles.

The presence/absence of bacterial taxa and their directionality of associations with sleep are inconsistent across studies. For instance, the directionality of associations of Lachnospiraceae genera are opposite between the Smith et al. and Grosicki et al. studies, and there are differences in taxa across the OSA studies (Table 1). These discrepancies may reflect distinct pathological mechanisms or symptoms of different sleep conditions (i.e. insomnia vs. OSA), or may be due to differences in study design (i.e. cross-sectional vs. case-control), cohort size (e.g. Baldanzi et al. study had N = 4045, whereas other studies had around N = 50 or fewer), demographics (e.g. Smith et al. focused exclusively on younger male adults), health status (e.g. Lu et al. looked at hypertensive patients with OSA), taxonomic resolution, or stool sample collection and analysis protocols. Furthermore, different sleep outcome measurements (such as the use of PSQI vs. actigraphy and differing definitions of OSA), different covariate adjustments, and a lack of multiple test corrections in most studies might contribute to the heterogeneity of findings. Nonetheless, these studies highlight significant associations between the gut microbiome, immune system, and sleep across adulthood, and underscore butyrate and specific bacterial taxa as promising targets for gut-based interventions to improve sleep, warranting further investigation (Fig. 1).

Probiotic effects on sleep in young and middle-aged adults (18–45 years old)

Several studies have used EEG alongside stress and subjective sleep measurements to specifically examine the effects of Lactobacillus probiotics on sleep in young to middle-aged adults (18–45 years old). For example, a 4-week administration of Lactobacillus gasseri CP2305 prevented sleep disturbances in medical students during exams, reduced afternoon cortisol levels, inhibited Enterobacteriaceae, and showed trends toward Veillonella suppression and increased Lactobacillus abundance (Sawada et al. 2017). Better PSQI scores were also found following a 24-week intervention with heat-inactivated L. gasseri CP2305, along with less stress-induced reductions in Bifidobacterium and an attenuated rise in Streptococcus. The probiotic also increased EEG delta power ratio (the delta power of the first sleep cycle to the delta power of the total sleep period) and shortened sleep latency of the first N3 stage and WASO, with no effects on total REM and NREM sleep times (Nishida et al. 2019). Similar findings were reported following an 11-week intervention with L. casei strain Shirota YIT 9029 in students under stress, where the probiotic significantly suppressed sleep latency and N3 sleep reduction, increased delta power by 20%, but did not affect EEG-sleep total time and efficiency. The probiotic improved self-reported “sleepiness on rising” and “sleep length” factors of the Oguri–Shirakawa–Azumi sleep quality assessment (Takada et al. 2017). These studies show that Lactobacillus probiotics may improve sleep in young to middle-aged adults by attenuating stress-induced sleep disruptions, possibly by reducing stress-induced perturbations of the gut microbiome, and by increasing NREM sleep, without a significant impact on REM sleep.

Pre- and probiotic effects on sleep in older adults (45+ years old)

In a recent study, overweight, older adult volunteers took a nutraceutical blend with or without milk thistle extract (Santamarina et al. 2024). Depletion of Bacteroides and enrichment of Collinsella, Lachnospiraceae Ruminococcus, Faecalibacterium, and Alistipes onderdonkii were associated with improvement in daytime sleepiness (ESS). Additionally, both groups had less inflammatory profiles at the endpoint, with decreases in blood IL-6/IL-10 ratio and increases in IL-4 compared to their respective pretreatment baselines. The milk thistle group showed a decrease in blood TNF-α compared to the pretreatment baseline (Santamarina et al. 2024). Another prebiotic study found that a preparation of traditional Chinese medicine plant granules ameliorated perimenopausal insomnia (PSQI). The treatment group’s endpoint gut microbiomes were depleted in Ruminococcus, Blautia obeum, Roseburia faecis, Prevotella copri, and Fusicatenibacter saccharivorans, and enriched in Bacteroides, Faecalibacterium, Bifidobacterium, and Lactobacillus. The change in abundance of Ruminococcus and Bacteroides is opposite to that observed by Santamarina et al., possibly highlighting sex-specific differences that warrant further investigation. However, both studies found positive effects of Faecalibacterium on improving sleep, in terms of PSQI and ESS scores (Table 1), possibly due to Faecalibacterium’s role as a core butyrate producer (Vital et al. 2017).

Considering the Santamarina et al. study in light of the observational OSA findings above (Table 1) (Lu et al. 2022, Baldanzi et al. 2023), where Bacteroidales (including the genus Alistipes) were depleted in older adult OSA patients, it appears that Alistipes may play a favorable role in improving daytime sleepiness. Promising results in mice show that oral administration of Alistipes onderdonkii can sufficiently repress systemic TNF-α expression to allow mismatched skin engraftment (Li et al. 2023), hinting at its positive effects on sleep through TNF-α regulation. Additionally, Alistipes has been positively associated with longevity in centenarians in East China (Wang et al. 2019), and older adults living in longevity villages in South Korea (Park et al. 2015). Both Faecalibacterium and Alistipes are associated with healthy aging, with findings from these prebiotic and OSA studies suggesting they may improve sleep quality and support healthy aging trajectories in older adults, possibly mediated by the immune system. However, further investigation of Alistipes is required given its association with depression (Parker et al. 2020). It may prove useful to identify the metabolic underpinnings of these taxon-phenotype associations in order to apply them effectively across different contexts.

Bacterial probiotics have been shown to improve sleep in older adults (45+), particularly in individuals with sleep complaints or a sleep-related condition. Murakami et al. showed that a Bifidobacterium adolescentis probiotic increased total sleep time, time spent in REM sleep, and daytime wakefulness in healthy Japanese men and women with sleep complaints (Murakami et al. 2024). Similarly, Ben Othman et al. found that daytime sleepiness (ESS) decreased in a study of predominantly female, obese patients on a low-carb diet given a prebiotic intervention of carob beans or a probiotic cocktail, but not in those on the low-carb diet alone (Ben Othman et al. 2023). Kikuchi-Hayakawa et al. reported that Lacticaseibacillus paracasei strain Shirota fermented milk improved daytime sleepiness (EEG) in individuals with sleep complaints (Kikuchi-Hayakawa et al. 2023), while Yamamura et al. found that Lactobacillus helveticus fermented milk improved sleep efficiency and reduced waking episodes in healthy older adults (Yamamura et al. 2009). Finally, a study in older adults with mild cognitive impairment found that a probiotic mixture of 18 different strains from the genera Lactobacillus, Lactococcus, and Bifidobacterium significantly improved the sleep quality, latency, duration, and disorder components of the PSQI (Fei et al. 2023). These sleep improvements were accompanied by better gastrointestinal health, with greater abundances of several taxa, including Blautia, Coprococcus, L. Ruminococcus, and Lachnospiraceae observed in the treatment group compared to controls. Additionally, the probiotic group showed an increase in serum BDNF, an anti-inflammatory protein that downregulates NF-κB, which is a key player in TNF-α signalling (Liu et al. 2017a) in brain microglia (Charlton et al. 2023).

Lactic acid bacterial probiotics have been shown to exert immunomodulatory effects that restore immune homeostasis in the host (Mazziotta et al. 2023), and above, we see several instances of their positive effects on sleep improvement. Notably, Fei et al. found both increases in anti-inflammatory BDNF and enrichment of several butyrate producers associated with improved PSQI in their probiotic intervention trial. Santamarina et al. found decreased TNF-α and enrichment of Faecalibacterium alongside improved ESS in the milk thistle group of their prebiotic study. Together, these intervention studies complement the observational studies, highlighting gut–immune–sleep interplay.

As in the observational studies, heterogeneity in findings can be attributed to differences in sample size, demographics, study duration, treatments, dosages, placebo controls, outcome metrics, and study designs. Furthermore, few studies collected both gut microbiome sequencing and sleep data (Table 1), limiting insights into how gut interventions change the microbiome to affect sleep. Most studies reported improvements in self-assessed sleep (i.e. PSQI or ESS), while those using objective measures like EEG were constrained by small sample sizes and limited recordings per participant. The high costs and logistical challenges of collecting objective sleep data likely contribute to these studies being underpowered to detect microbiome-related effects. Despite these limitations, intervention studies show promising results (Fig. 1), but further research is needed to clarify the causal mechanisms by which these pre- and probiotics improve sleep.

Conclusions and future perspectives

In the present review, we show that PSQI in observational cohorts is generally associated with lower gut microbiome alpha-diversity and perturbations in butyrate-producing bacteria, particularly taxa within the Lachnospiraceae family (Smith et al. 2019, Grosicki et al. 2020, Gao et al. 2022). In animal models, butyrate supplementation promotes sleep and protects against the detrimental consequences of poor sleep (Szentirmai et al. 2019, Gao et al. 2022), possibly through its immunomodulatory actions (Chang et al. 2014, Siddiqui and Cresci 2021). In most of the pre- and probiotic intervention trials reviewed here, participants were often experiencing some form of stress (Sawada et al. 2017, Takada et al. 2017, Nishida et al. 2019), disrupted sleep (Kikuchi-Hayakawa et al. 2023, Murakami et al. 2024), or some other sleep-related condition (Yang et al. 2021, Ben Othman et al. 2023, Fei et al. 2023). In contrast, Yamamura et al.’s study (2009) was conducted in healthy older adults without sleep complaints or disorders. Yamamura et al. reported weak effects from the probiotic, in terms of sleep improvement, suggesting there may be a “ceiling effect” in populations with optimal sleep (Yamamura et al. 2009). Overall, it seems prebiotics and probiotics may be helpful in mitigating the impact of psychological stress on sleep or correcting an underlying biological dysregulation of systems that impact sleep, such as immune or endocrine signaling.

Although both human observational and intervention studies show promising results, many microbiome–sleep associations are unclear or contradictory when integrated across studies. This may be due, in part, to the context dependencies of microbiota–sleep interactions, especially those associated with age, diet, and lifestyle. Future studies should include a demographically diverse mixture of young, middle-aged, and older adults to tease apart this heterogeneity. Furthermore, to better understand the mechanisms by which bacterial taxa affect sleep, it may be necessary to look beyond compositional shifts and consider the metabolic outputs of the gut microbiome (Wilmanski et al. 2022). More research is needed to identify the specific microbial metabolites, such as butyrate, that influence sleep, as well as their mechanisms of action. While cytokines, such as TNF-α, may mediate some of the gut microbiome's effect on sleep quality, there are also likely many unknown microbially derived molecules that interact with host pathways. Indeed, many microbially derived molecules, such as N-acyl amides, SCFAs, tryptophan derivatives, and secondary bile acids, are known to interact with host GPCRs expressed by intestinal, immune, and neuronal cells (Aleti et al. 2023). However, about one-third of human GPCRs remain orphan receptors, while numerous microbial metabolites either remain functionally uncharacterized or undiscovered (Postler and Ghosh 2017, Aleti et al. 2023). To fully unravel these interactions and their role in gut–brain crosstalk in the context of sleep, work that integrates blood metabolomics, blood proteomics, gut microbiome sequencing, and sleep assessment is needed. One approach would be to leverage large, deeply phenotyped human cohorts for hypothesis generation (Melamud et al. 2020). These datasets offer integrated multi-omic, longitudinal data that enable high-powered statistical analyses to detect bacterial taxa and metabolic outputs associated with phenotypes of interest (Melamud et al. 2020). Associations identified in human cohorts can then be used to generate mechanistic hypotheses for preclinical experiments, where the implicated bacterial taxa or bacterial consortia can be cultured in vitro to examine their metabolic outputs (Rudi and Zhao 2021). The metabolites, or the relevant taxa, can then be introduced into animal models to assess their effects on sleep.

In terms of sleep assessment in humans, future studies should more often strive to include objective sleep measures. Indeed, effective microbiome-mediated therapies for sleep might require precise targeting of sleep architecture, where improper alterations to the gut may have harmful collateral consequences. For example, a study of healthy older adults (60+) found that those with sleep latency over 30 min, sleep efficiency below 80%, or an REM sleep percentage outside a range of 16%–25% showed a nearly 2-fold higher risk of all-cause mortality, even after controlling for age, sex, and medical burden (Dew et al. 2003). Additionally, chronotype and circadian misalignment (e.g. social jet lag) should be accounted for in experimental designs, as this could affect sleep while both being affected by and affecting the gut microbiome (Bermingham et al. 2023, Yue et al. 2023). Importantly, advancements in wearable technologies now enable high-throughput, continuous monitoring of sleep and activity. Several consumer wearable devices perform as well as, or even better than, clinical-grade actiwatches at recording sleep/wake measures when benchmarked against the gold-standard PSG (Chinoy et al. 2021). In addition to their accuracy, consumer wearable devices are significantly more cost-effective than EEG, PSG, and clinical-grade actiwatches (Martin and Hakim 2011, Zambotti et al. 2019). These technological advancements will make larger longitudinal studies capturing objective sleep measures more feasible.

Recognizing the technical, ethical, and cost challenges of microbiome (and, more broadly, multi-omic) research underscores key challenges that should be carefully navigated going forward. Standardizing gut microbiome (and multi-omic) sample collection and storage is essential to minimize biases, such as changes in stool microbial composition that can occur without prompt freezing or the use of chemical preservatives (Kim et al. 2017). Bias can also be introduced from sample contamination and batch effects due to the use of different kits or reagents across experiments (Kim et al. 2017). Additionally, microbiome sequencing can reveal sensitive personal information, such as host genetic material in fecal metagenomes, highlighting the need for robust privacy protections similar to those implemented in genomic research (Ejtahed et al. 2023). Furthermore, human interventions demand careful oversight, as they may have unanticipated effects on participants and their broader microbiomes, raising safety and community impact concerns (Ejtahed et al. 2023). Finally, we should focus on applying expensive microbiome (and multi-omic) research more equitably to populations outside the developed world so that the resulting medical advances can benefit all of humankind (Abdill et al. 2022). Carefully navigating these challenges will enable responsible advancement of gut–sleep axis research and reveal opportunities for therapeutic interventions.

Acknowledgments

We thanks members of the Gibbons and Kadosh labs for helpful discussions. Figure 1 was created using BioRender under a Lab-Academic license (Agreement Number: UL27FSGW9T).

Author contributions

Jacob Cavon (Conceptualization [equal], Investigation [equal], Methodology [equal], Writing – original draft [equal], Writing – review & editing [equal]), Melissa Basso (Conceptualization [equal], Investigation [equal], Methodology [equal], Writing – original draft [equal], Writing – review & editing [equal]), Kathrin Cohen Kadosh (Project administration [equal], Resources [equal], Supervision [equal], Writing – review & editing [equal]), and Sean M. Gibbons (Conceptualization [equal], Project administration [equal], Resources [equal], Supervision [equal], Writing – review & editing [equal])

Conflict of interest

The authors declare no conflicts of interest.

Funding

Research reported in this publication was supported by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) of the National Institutes of Health (NIH) under award number R01DK133468 (to S.M.G.) and by the Global Grants for Gut Health from Yakult and Nature Portfolio (to S.M.G.).

Data availability

There are no new data associated with this article.

References

Author notes