核心提示：由于许多成人疾病起源于儿童早期，因此在婴儿期利用充足的睡眠和与年龄相适应的肠道微生物群的保护因素可能构成整个人类生命周期的健康促进因素。

  一．摘要

  睡眠障碍与成人和动物模型中肠道微生物群组成的改变有关，但尚不清楚这种联系是如何发展的。通过对 162 名健康婴儿的纵向评估，我们提出了迄今为止未被认识的睡眠-大脑-肠道相互关系。首先，我们报告了睡眠习惯和肠道微生物群之间的联系：白天睡眠与细菌多样性有关，而夜间睡眠的碎片化和变异性与细菌成熟度和肠型有关。其次，我们证明了睡眠-大脑-肠道的联系：细菌多样性和肠型与睡眠神经生理学有关。第三，我们表明睡眠-大脑-肠道联系与发育相关：睡眠习惯和细菌标志物可预测行为发育结果。我们的研究结果证明了睡眠、肠道微生物群、以及婴儿期大脑和行为的成熟，这与新出现的睡眠-脑-肠轴概念一致。重要的是，睡眠和肠道微生物群代表了有希望的健康目标，因为两者都可以通过非侵入性方式进行修改。由于许多成人疾病起源于儿童早期，因此在婴儿期利用充足的睡眠和与年龄相适应的肠道微生物群的保护因素可能构成整个人类生命周期的健康促进因素。

  二．亮点

  1.婴儿的睡眠习惯和神经生理与他们的肠道微生物群有关

  2.这种睡眠-大脑-肠道的联系对婴儿的发育很重要

  3.睡眠和肠道微生物群是有希望的早期干预目标

  三．结论

  1、婴儿白天的睡眠与alpha多样性有关

  Fig. 2.The association between alpha diversity and Sleep Day. Alpha diversity of bacteria in stool samples is lower in infants with more daytime sleep. a) Observed diversity was significantly associated with Sleep Day across all ages (multilevel model, p =0.03) and at 3 months (p =0.004), but not at 6 and 12 months. b) Chao1 diversity was significantly associated with Sleep Day across all ages (multilevel model, p =0.02) and at 3 months (p =0.008), but not at 6 and 12 months. c) The Shannon Index was significantly associated with Sleep Day at 3 months (p =0.04), yet not at other ages. Data were averaged across 100 imputations for plotting (n =162). Data were analyzed pooled across 100 imputations using multilevel models with individual intercepts for each participant across all time points and generalized linear models for each time point. All the models controlled for participant age, sex, breastfeeding status, and batch.

  以α多样性为结果，以睡眠组合为自变量(控制年龄、母乳喂养、性别、批次)。我们发现白天睡眠(表征白天小睡的持续时间和次数及其规律性)与α多样性之间存在负相关：α多样性较高的婴儿白天睡眠较少。

  2、细菌成熟指数较高的婴儿夜间醒来次数较多

  Fig. 3. Associations between sleep habits as measured by sleep composites with gut microbial maturity and enterotype profiles. a) Infant Sleep Activity across all assessment timepoints positively correlates with bacterial maturation index (p = 0.03), however, the association is not significant for single assessment timepoints (p> 0.05. n = 162). b) Sleep Activity at 12 months correlates with enterotype evolution. Significantly more nighttime awakenings are found in infants with a “Bifidobacterium” (n = 19) pattern compared to “Switchers” (p = 0.048. n = 82). c) Sleep Variability at 12 months correlates with enterotype evolution. Infants with more variability in sleep show a “Bacteroides” (n = 44) pattern more often than a “Switcher” pattern (p = 0.02. n = 82). Data were averaged across 100 imputations for plotting. Data were analyzed pooled across 100 imputations using multilevel models with individual intercepts for each participant across all time points (A) and generalized linear models for each time point (A, B & C). All the models controlled for participant age, sex, breastfeeding status, batch. * p < 0.05.

  以细菌成熟指数作为结果和以睡眠组合(控制确切的年龄、性别、母乳喂养和批次)作为自变量预测因子。研究发现睡眠活动与细菌成熟指数呈正相关，表明肠道微生物组更成熟的婴儿在夜间表现出更多的活动和觉醒。 这种效应只有在结合所有婴儿年龄时间才明显，而在单独的年龄时间段这种因果关系不显著。

  3、婴儿的睡眠习惯与肠道婴儿微生物的肠型关系

  Note. Standard errors of the mean are reported. Breastfeeding was not included in the model at 3 months because all infants were (over 50%)breastfed at that age. Run could not be included for models at 3 and 12 months because not all combinations of Run/Enterotype existed at these timepoints. Positive estimates signify associations with enterotype A. Enterotype A is linked with higher abundance of Bifidobacterium while enterotype B is linked with higher abundance of Bacteroides. Bold marks significant associations (P < 0.05).

  我们研究了发现肠道菌群A型和B型与婴儿睡眠习惯无显著关系

  4、婴儿的睡眠习惯与肠道微生物的肠型进化有关

  6-12个月期间肠型进化分为：转换型(6个月时为A型肠型，12个月时为B型)、双歧杆菌型(6个月和12个月时为A型肠型)、拟杆菌型(6个月和12个月时为B型肠型)和反转型(6个月时为B型，12个月时为A型肠型)。

  研究发现双歧杆菌型的婴儿在12个月时的睡眠活动明显高于转换型婴儿。拟杆菌型的婴儿睡眠变化性高于转换型肠型婴儿。

  5、婴儿可能存在的睡眠-肠-脑轴

  Fig. 4. Sleep neurophysiology is linked to gut microbiota enterotype in 6-month-old infants.a) Topographical distribution of slow wave activity (SWA, 0.75 – 4.25 Hz) for the first 30 min of artifact-free NREM sleep at 6 months of age depending on enterotype at 6 months of age (N= 32 total, enterotype A N = 18. enterotype B N= 14). Maps are based on 109 EEG electrodes and averaged for each enterotype. Values are color-coded and scaled to maximum (red) and minimum (blue). While infants in both enterotypes show an occipital maximum of SWA, the power averaged across all electrodes is significantly greater in enterotype B (272.82 ± 84.68 μV2) than enterotype A (206.81 ± 59.86 μV2. p= 0.02 generalized linear model controlling for age, sex, breastfeeding, and other gut microbial markers). b) Distribution of t-values across the scalp resulting from electrode-wise student’s unpaired t-test comparing the infants in enterotype A with enterotype B. Significant electrodes are indicated as white dots. SWA was significantly increased in enterotype B in a cluster of 51 globally distributed electrodes (Statistical nonparametric Mapping clustercorrected)

  在 6 个月大的婴儿中，我们发现了睡眠-肠脑的联系。根据高密度脑电图记录，肠道微生物肠型与不同的睡眠神经生理学相关。 特别是，具有肠道微生物肠型 A 的婴儿表现整体慢波活性降低。该效应与SWA频率有关，而θ或σ功率无关。脑电图功率与阿尔法多样性之间没有相关性。

  6、6个月大时的睡眠神经生理学预测了之后的alpha多样性

  Fig. 5. Infant neurophysiology assessed during sleep predicts their later gut microbial diversity. a) Theta power averaged across all 109 electrodes at 6 months plotted against alpha diversity (Shannon) at 12 months (N = 22. p = 0.03 in a model corrected for age, sex,breastfeeding, alpha diversity at 6 months, and slow wave activity and sigma power). b) Electrode-wise partial correlation between theta power (4.5 – 7.5 Hz) at 6 months of age and alpha diversity at 12 months of age, corrected for age at EEG assessment, illustrated on a scalp model. Significant electrodes are indicated as white dots. The association between theta at 6 months and alpha diversity at 12 months is significant in a large global cluster of 75 electrodes (Statistical nonparametric Mapping cluster correction for multiple comparisons).

  θ频率的脑电图功率预测了之后的alpha多样性:① 6个月时θ功率增加的婴儿在12个月时显示出较低的alpha多样性;② 在3个月或12个月的肠道微生物标志物与6个月的SWA或sigma功率之间没有发现其他关联。因此，婴儿睡眠神经生理学预测了之后的alpha多样性，代表了睡眠脑肠道轴框架内自上而下的动态。

  7、睡眠习惯和肠道微生物群预测发育行为结果

  Fig. 6. Associations between 5 composites of sleep habits, the three gut microbiota markers,and 3 domains of behavioral development at 3.6. and 12 months of age (β ± Standard Error).Left columns contain statistics for concurrent associations (e.g., 3 months in relation to 3 months), right columns indicate predictive associations lagged by one assessment timepoint(e.g., 3 months in relation to 6 months). Colors indicate significant associations in a random intercept structural equation model controlling for exact age and sex and breastfeeding for gut microbiota markers (p < 0.05. N = 147 for 3 months, N = 157 at 6 months, N = 141 at 12 months and N = 133 at 24 months). Blue color denotes positive association, while red color denotes negative associations.

  在6个月时，睡眠变异性和睡眠时间预测了12个月时的集体和个人社会得分;

  12个月大的睡眠日预测了24个月大肌肉运动的发展;

  从肠道微生物群标记来看，只有6个月时的肠型可以预测之后的行为发展，因为肠型A与12个月时的肌肉运动的发展评分有关;

  因此，总的来说，与肠道微生物群和行为发育相比，睡眠习惯和后期行为发育之间存在更多的关联，这表明睡眠比肠道微生物群更能预测婴儿的行为发育。

  四 结论

  纳入162名健康婴儿，评估其睡眠状况与肠道菌群、神经发育之间的关系：

  ① 白天睡眠状况与菌群α多样性存在关联，而夜间睡眠的碎片化和变异性与肠道菌型及成熟度存在关联;

  ② 肠道菌型和多样性与睡眠神经生理学信号存在关联，提示存在睡眠-脑-肠轴;

  ③ 婴儿睡眠模式和细菌生物标志物可以预测行为相关的神经发育结局;

  ④ 重要的是，睡眠和肠道菌群都是可以被无创地改变，是具有重要意义的干预靶点。