Night-to-night sleep duration variability and gut microbial diversity: more evidence for a brain-gut microbiome-sleep connection

Sleep disorders and gut microbiome dysbiosis can both lead to increased risk of obesity, diabetes, and cardiovascular diseases such as heart attack and stroke. The gut-brain axis connections have been apparent for centuries, but the role of the gut microbiome in regulating possible interconnections is a relatively new science, starting with the advent of high-throughput bacterial sequencing in the second half of the 2000s [1]. Holzhausen et al.’s article in press, adds important evidence to a growing literature that supports a significant association between sleep characteristics and the composition of the gut microbiome [2]. The authors studied 720 study participants, aged 18–94 from the Survey of the Health of Wisconsin, who provided stool samples and 7 days of accelerometry data information on sleep duration, within-person night-to-night sleep duration standard deviation, sleep efficiency, wake-after sleep onset (WASO), and average sleep latency. They also collected self-reported sleep data over the past 30 days.

The study sample included 417 (58%) women, with 83% being non-Hispanic white, with mean age of 55 (SD 16) years. Wrist actigraphy measures recorded an average nightly sleep duration of 8.15 (SD 1.1) hours, with 72.3 minutes (SD 37.8) of night-to-night sleep duration variability, and WASO of 50.1 minutes (SD 26.1). The authors found that there were significant within-person compositional differences in the gut microbiome using three different commonly used measures of alpha diversity: Chao 1, Inverse Simpson’s Diversity, and Shannon’s Diversity. Regardless of the alpha diversity measure used, they reported that greater night-to-night sleep duration variability and increased WASO were significantly associated with decreased richness and diversity. With regards to beta-diversity, or between-person differences in microbiome composition, they reported only one metric, the Bray-Curtis dissimilarity index. There were significant differences between persons in sleep latency, sleep efficiency, self-reported sleep quality, and napping at least once per week, with the strongest association being night-to-night sleep duration variability.

The authors also reported that there were several taxa that were associated with increased night-to-night sleep duration variability, longer average sleep latency, higher sleep efficiency, and better self-reported sleep quality, some of which (e.g. Christensenellaceae and Mogibacteriaceae families) have been previously reported in association with metabolic health. While another recently published study of 606 older men identified several butyrate-producing bacteria as being associated with better sleep regularity [3], Holzhausen’s study reported the opposite effect with just one butyrate-producing bacteria, Coprococcus being associated with increased sleep latency that can be an indicator of poor sleep. However, sleep latency in this study ranged from 0 to 7.8 minutes, with an average latency of only 3 minutes, which is in the range of good sleep quality [4].

The strongest effect on gut microbial composition across all sleep measures was the objective measure of night-to-night sleep duration variability. This was the sole measure that retained statistical significance in all three alpha-diversity measures, beta-diversity, and taxonomic analyses. In another study of subjective and objective measures of sleep and the gut microbiome involving over 600 older men, it was a measure of sleep regularity that retained significance in beta-diversity and taxonomic analyses [3], suggesting that consistency of sleep behaviors may be the most important predictor of overall gut microbial health.

It is known that bacteria and humans both have circadian rhythms in gene expression and behavior, with over 40% of the mouse genome dependent on the circadian clock [5]. The adult gastrointestinal tract contains on the order of 10-fold more bacteria than human cells, with about 100-fold more gene functions than its human host [6]. Thus, in considering what sleep characteristics may most affect the gut microbiome, it is perhaps not surprising that objective measures of sleep regularity might explain the most variability in microbiome composition.

Circadian variation in humans and health outcomes have been well-described and range from reports of circadian misalignment and increase in cardiovascular disease risk factors [7] to an increased risk of adverse health outcomes reported in shift workers [8]. Concomitantly, scientists have convincingly shown that the intestinal microbiota possesses their own functional oscillations that can affect the host's circadian transcriptional, epigenetic, and metabolic oscillations [9]. It should perhaps come as no surprise that host behaviors should be able to affect bacterial health and optimal metabolism. Likewise, it is conceivable that bacterial metabolites may affect human behavior. As an example, Yano et al. published that indigenous bacteria from the gut microbiota regulate host serotonin biosynthesis [10]. Serotonin itself has long been implicated as an important factor in sleep and is one of the precursors of melatonin biosynthesis [11].

In a cross-sectional study such as Holzhausen’s, we are unable to infer whether it is night-to-night sleep variability or diversity of the gut microbiome that may be the causal link to adverse health outcomes such as obesity, diabetes, or cardiovascular disease. However, their work in a relatively large sample of adults across a broad spectrum of ages provides confirmatory evidence that the gut microbiome may play an important role in the gut-brain axis, with bidirectional communication likely occurring.

In the gut microbiome literature, much emphasis has been placed on butyrate-producing bacteria that are associated with better human health. However, with regards to the gut microbiome and sleep, reporting of specific bacterial families, genus, and species identified by 16S rRNA methods have not coalesced upon common organisms. Given the complexity of next-generation sequencing that can be affected by sampling, storage, and DNA extraction methods that may in turn influence the specific bacteria identified, it is plausible that stool collection and processing could be the explanation for disparate results [12]. Alternatively, the study populations reported to date have differed in size, geographic location, and age groups, so this may also account for the differences in bacterial families identified.

Holzhausen’s study results provide a strong rationale for continuing to delve into the complex interactions between human guts and brains, likely potentiated by the gut microbiome. For some, good quality sleep is considered a fifth vital sign, and crucial for healthy aging. The field is particularly promising because human behaviors such as timing of eating, dietary choices, and sleep behaviors all have the potential to affect health in positive ways. To maximize field impact, taxonomic signatures, functional and metabolic approaches in addition to interventions conducted using longitudinal study design with repeated measures should have great promise.

Disclosures Statement

The author has no financial disclosures to report with regard to the content covered in this editorial. The author also has no nonfinancial disclosures/conflicts of interest.