The Effect of Sleep Apnea and Hypertension on Gut Microbiome

Overview

Obstructive sleep apnea (OSA) and hypertension are closely associated diseases. Here we characterized the differences in the gut microbiome which is affected by the two diseases, when the two diseases coexist or are present alone. Fifty-two consecutive patients who underwent polysomnography (PSG) were enrolled and divided into four groups: without OSA or hypertension (OSA0HT0), OSA without hypertension (OSA1HT0), hypertension without OSA (OSA0HT1), and with OSA and hypertension (OSA1HT1). Fecal specimens were collected for 16S rRNA sequencing and the characteristics of community richness, diversity, and composition of the gut microbiome and their relationship with disease were analyzed using bioinformatics methods.

1. Participants Participants who underwent overnight polysomnography (PSG) in the sleep laboratory of Peking University First Hospital from July 2017 to November 2017 were included in the study. This study was approved by the ethics committee of the Peking University First Hospital. The study adhered to the Declaration of Helsinki, and patient confidentiality was maintained. All participants signed the informed consent form. The inclusion criteria were as follows: (1) age 18 years or older; (2) outpatients and inpatients who underwent a PSG study for snoring; (3) those who volunteered for this study and signed an informed consent form. Exclusion criteria were: (1) patients who were treated for OSA (regular CPAP therapy, oral appliances, maxillofacial surgery, etc.) prior to enrolment; (2) patients with secondary HT with a clear primary cause; (3) those with a history of dyspeptic disease (history of gastrointestinal surgery, peptic ulcer, inflammatory bowel disease, chronic pancreatitis, etc.); (4) those who had organ insufficiency, were receiving immune agents or glucose; (5) those who had received antibiotic treatment in the last 2 months or had taken probiotic products (yogurt, milk, cheese, etc.) continuously (daily) for the last 2 months. (6) Alcohol or drug dependency. 2. Questionnaire survey A uniformly designed questionnaire including questions regarding general information, previous history of HT, coronary heart disease, diabetes mellitus, recent history of infection, medication, smoking and drinking history, and dietary habits was used. 3. Obstructive sleep apnea assessment and HBI calculation Patients underwent regular overnight PSG (Compumedics, E-Series). Standard PSG was conducted according to the American Academy of Sleep Medicine manual (AASM)\[11\]; that is, six electroencephalography (EEG) channels (C3-M2, C4-M1, F3-M2, F4-M1, O1-M2, O2-M1), two electrooculography channels (E1-M2, E1-M2), chin electromyography (EMG1-EMG2, EMG1-EMG3), electrocardiography, respiration (nasal pressure, airflow), SpO2, abdominal and chest movements, and leg movements were recorded . Sleep stages were divided into three non-REM (N1, N2, N3), REM (R), and wake (W) stages. Respiratory events included obstructive apnoea, central apnea, mixed apnea, and hypopnea. The apnea and hypopnea index (AHI) (sum of the number of apnea and hypopnea events per hour) was calculated. Sleep stage, apnea, and hypopnea events were scored according to the American Academy of Sleep Medicine manual 2.3. The HBI (the hypoxia burden index) was calculated as the integral area under the desaturation curve divided by TST. The area under the desaturation curve was obtained by calculating the integral of the oxygen saturation reduction below 90% and the corresponding time. Higher HBI values are related to a higher hypoxic load (duration and degree). Calculations were performed with MATLAB 2016 for Windows (The Mathworks, Inc., Natick, USA). Specific methods refer to our previous report. 4. Blood pressure measurement and specimen collection Blood pressure was measured before going to sleep and immediately after waking up. Fecal samples of 1-3 g was collected from the patients in the morning of the following day and immediately frozen in a -80℃ refrigerator by the researcher for later use. 5. Information collection, entry, and participant grouping. The enrolment number, sex, age, height, weight, Epworth sleep scale (ESS), HT and medication, other major medical and personal histories of all participants in the questionnaire, and the PSG report parameters, mainly AHI, were compiled and entered into an electronic form. Participants were divided into four groups according to whether they had a confirmed diagnosis of essential HT and OSA (AHI ≥15 beats/h (International Classification of Sleep Disorders (3rd edition), as follows. Individuals without OSA and HT belonged to group OSA0HT0; individuals with both OSA and HT were in group OSA1HT1; individuals without OSA but with HT were in group OSA0HT1; and individuals with OSA but without HT were in the group OSA1HT0. 6. 16S rRNA sequencing analysis of the gut microbiome DNA extraction and sequencing of 16S rRNA coding genes were performed on all fecal samples. This part of the experiment and analysis was conducted at Novogene Bioinformatics Technology Co. Ltd. (Beijing, China) . 7. Fecal analysis 7.1. Extraction of genome DNA Total genomic DNA from human fecal samples was extracted using cetyltrimethylammonium bromide (CTAB) /sodium dodecyl sulfonate (SDS), according to the manufacturer's instructions. DNA concentration and purity was monitored on 1% agarose gels. 7.2. Amplicon generation 16S rRNA/18S rRNA/ITS genes of distinct regions (16S V4/16S V3/16S V3-V4/16S V4-V5, 18S V4/18S V9, ITS1/ITS2, and Arc V4) were amplified using specific primers (for example 16S V4: 515F-806R, 18S V4: 528F-706R, 18S V9: 1380F-1510R, et. al) with a barcode. All polymerase chain reactions (PCRs) were performed using Phusion® High-Fidelity PCR Master Mix (New England, Biolabs). 7.3. Polymerase chain reaction products mixing and purification Mix same volume of 1X loading buffer (contained SYB green) with PCR products and operate electrophoresis on 2% agarose gel for detection. PCR products was mixed in equidensity ratios. Then, mixture of PCR products was purified with GeneJETTM Gel Extraction Kit (Thermo Scientific). 7.4. Library preparation and sequencing Sequencing libraries were generated using Ion Plus Fragment Library Kit 48 rxns (Thermo Scientific), according to the manufacturer's recommendations. The library quality was assessed on a fluorometer (Qubit 2.0; Thermo Scientific). Finally, the library was sequenced on an Ion S5TM XL platform and 400 bp/600 bp single-end reads were generated. 8. Bioinformatics, and statistical analysis 8.1. Single-end reads assembly and quality control 8.1.1 Data split Single-end reads were assigned to samples based on their unique barcode and truncated by cutting off the barcode and primer sequence. 8.1.2 Data Filtration Quality filtering on the raw reads were performed under specific filtering conditions to obtain the high-quality clean reads according to the Cutadapt（V1.9.1）quality controlled process. 8.1.3 Chimera removal The reads were compared with the reference database (Gold database), using the UCHIME algorithm to detect chimeric sequences, which were then removed. Then the Effective Tags were finally obtained. 8.2. Operational taxonomic unit clustering and species annotation 8.2.1 Operational taxonomic unit production Sequences analyses were performed by Uparse software (Uparse v7.0.1001). Sequences with ≥97% similarity were assigned to same operational taxonomic units (OTUs). A representative sequence of each OTU was screened for further annotation. 8.2.2 Species annotation For each representative sequence, the Silva Database was used based on RDP classifier (Version 2.2) algorithm to annotate taxonomic information. 8.2.3 Phylogenetic relationship Construction In order to study the phylogenetic relationship among different OTUs, and the difference of the dominant species in different samples (groups), multiple sequence alignments were conducted using the multiple sequence comparison by log- expectation (MUSCLE) software (Version 3.8.31). 8.2.4 Data Normalization OTUs abundance information were normalized using standard sequence numbers corresponding to the sample with the least sequences. Subsequent analysis of alpha diversity and beta diversity were all performed based on this output normalized data. The Chao1, Shannon, and Simpson indices were calculated to estimate alpha diversity and principal coordinate analysis (PCoA) was used to represent beta diversity. 8.3. Alpha Diversity Alpha diversity was applied to analyze the complexity of species diversity in a sample through six indices, including observed species, Chao1, Shannon, Simpson, ACE, Good-coverage. All these indices were calculated with QIIME (Version1.7.0) and displayed using the R software (Version 2.15.3). Chao1 and ACE were selected to identify community richness. Shannon and Simpson indices were used to identify the community diversity. 8.4. Beta Diversity Beta diversity analysis was used to evaluate the differences of samples in species complexity. Beta diversity on both weighted and unweighted unifractions were calculated by QIIME software (Version 1.7.0). PCoA was used to represent Beta diversity.