Association of suboptimal health status with intestinal microbiota in Chinese youths

Abstract Suboptimal health status (SHS), a physical state between health and disease, is a subclinical and reversible stage of chronic disease. Previous studies have shown alterations in the intestinal microbiota in patients with some chronic diseases. This study aimed to investigate the association between SHS and intestinal microbiota in a case‐control study with 50 SHS individuals and 50 matched healthy controls. Intestinal microbiota was analysed by MiSeq 250PE. Alpha diversity of intestinal microbiota in SHS individuals was higher compared with that of healthy controls (Simpson index, W = 2238, P = .048). Beta diversity was different between SHS and healthy controls (P = .018). At the phylum level, the relative abundance of Verrucomicrobia was higher in the SHS group than that in the controls (W = 2201, P = .049). Compared with that of the control group, nine genera were significantly higher and five genera were lower in abundance in the SHS group (all P < .05). The intestinal microbiota, analysed by a random forest model, was able to distinguish individuals with SHS from the controls, with an area under the curve of 0.79 (95% confidence interval: 0.77‐0.81). We demonstrated that the alteration of intestinal microbiota occurs with SHS, an early stage of disease, which might shed light on the importance of intestinal microbiota in the primary prevention of noncommunicable chronic diseases.

been validated in various populations, including Chinese, 3 African 4 and European populations. 5 Although several items of SHS are similar to some diseases, SHS cannot reach the diagnosable condition of any current defined diseases. SHS is not a disease state, but several studies have suggested that SHS, as an overall assessment of human body, might precede the occurrence of noncommunicable chronic diseases (NCDs), including type 2 diabetes mellitus (T2DM), cardiovascular diseases and hypertension. [3][4][5][6] A cross-sectional study conducted among workers in urban Beijing demonstrated that SHS was associated with cardiovascular risk factors. 6 In Russia, a community-based cross-sectional study showed that SHS was related to endothelial dysfunction, suggesting that the integration of SHS and endothelial dysfunction can be applied to the elevated risk of cardiovascular diseases. 5 In Ghana, a cross-sectional study demonstrated that SHS might be involved in the development of T2DM. 4 In a cross-sectional study of Chinese students, SHS was shown to be correlated with stress management, psychological states and physical activity. 7 More recently, Anto et al 8 found that high SHS score is associated with increased incidence of preeclampsia.
In China, due to modern lifestyles, increased work pressures, changes in diet and other factors, the prevalence of chronic diseases such as hypertension, heart disease, stroke, cancer, chronic obstructive pulmonary disease and diabetes is increasing. 9 As a subclinical and reversible stage of chronic disease, SHS might play a very important role in the pathogenesis of chronic diseases. 2 The aetiology and mechanism of SHS are still poorly understood, and SHS is diagnosed based on subjective questionnaires and lacks objective biomarkers. Shortened relative telomere length, N-glycosylation and high levels of plasma cortisol were associated with SHS, suggesting that objective biomarkers have the potential to diagnose SHS. [10][11][12] Therefore, novel diagnostic markers based on objective measurements are urgently needed.
In recent years, many studies have shown that intestinal microbiota play important roles in host health, 13 as well as the immune system, 14,15 nervous system, 16 digestive system 17 and cardiovascular system, 18 all of which are components of SHS. 2 Emerging evidence suggests a link between the intestinal microbiota and various diseases, such as atherosclerosis, 19,20 hypertension, 21 T2DM 22 and acquired immune deficiency syndrome. 23 This association between microbiota and diseases was also validated through faecal transfer experiments. [24][25][26] Furthermore, altered gut microbiota was observed in patients suffering from chronic fatigue syndrome (CFS), a disease resembling SHS. 27

| Study population and design
Previously, we conducted a cross-sectional survey among undergraduates at Weifang University, which is located in Shandong Province, China, in November 2017. In total, 5219 participants completed the questionnaire, of which 442 (8.50%) were considered suffering from SHS. 29 Among these undergraduates, 50 cases and 50 controls were included in this case-control study. All students attended a standardized examination protocol in Weifang Hospital, including an interview about the history of previous diseases, family income, physical activity, smoking status and history of drinking.  (Table S1). 2 Participants were enrolled if they met the following inclusion criteria: (a) signed informed consent; and (b) a completed questionnaire. They were excluded if they met the following exclusion criteria: (a) a history of somatic or psychiatric abnormalities; (b) a history of mental illness or drug abuse; and (c) a history of antibiotic consumption in the previous 2 months. After enrolment, propensity score matching was used to match a subset of 50 SHS cases (SHS score ≥35) to a subset of 50 ideal healthy controls (SHS score <35) by age, gender and body mass index (BMI). The score was calculated from the SHSQ-25. The details of SHSQ-25 and the characteristics of the subjects between SHS group and control group are shown in Tables 1 and S2.

| Faecal sample collection and DNA extraction
Stool samples were obtained through a specimen collection kit and immediately stored at −80°C. Bacterial DNA was extracted at the Beijing Municipal Key Laboratory of Clinical Epidemiology using a TIANGEN kit based on the manufacturer's recommendations. The concentration and purity were evaluated using a Nanodrop ® spectrophotometer (Thermo Scientific). Extracted DNA was stored at −80°C.

| DNA sequencing of the 16S rRNA gene
The V4 region of the bacterial 16S rRNA gene was ampli-  TA B L E 1 Characteristics of subjects in SHS and healthy groups concatenate forward and reverse reads, paired-end sequences and barcodes were sorted and matched.

| Bioinformatic analysis
Quality filtering and analyses of microbial community diversity were performed with the QIIME1 software package (Quantitative Insights into Microbial Ecology). Sequences were removed from the analysis if they were <200 nt, had a quality score <20, contained ambiguous characters, contained an uncorrectable barcode or did not contain the primer sequence. Sequences were assigned to samples by exam-

| Statistical analysis
The Kolmogorov-Smirnov test was used for normality test. If the data conformed to a normal distribution, continuous variables were presented as the mean ± standard deviation (SD), and an independent t test was used for analysis. If data were not normally distributed, continuous variables were presented as a median (interquartile range) and analysed by Wilcoxon rank-sum test. The χ 2 test was used to examine differences in categorical variables between the cases and controls.
Wilcoxon rank-sum tests were used to assess associations of individ-

| Ethics statement
This study was approved by the ethics committee of Capital Medical University, Beijing, China. All subjects gave written informed consent in accordance with the Declaration of Helsinki.  Figure S1. There were no significant differences between the SHS and control groups in demographic variables (age, sex, family income, alcohol consumption, smoking history and physical activity), anthropometric measurements (BMI and WHR), blood pressure or biochemical measurements (Table 1).  Table S3).

| Microbial diversity
Weighted and unweighted UniFrac distance metric matrices were used to evaluate overall differences between the two groups through PCoA. Weighted UniFrac distance metrics in some extent distinguished the intestinal microbiota of SHS subjects from that of the control subjects (P = .018). There appeared to be a difference in beta diversity between the SHS group and healthy group (Figure 2).
On the weighted UniFrac distance metrics figure, there is a cluster of nine healthy and one SHS at the top right corner that is clearly separated from the rest of the individuals. The aspartate aminotransferase (AST) and alanine aminotransferase (ALT) were found to be increased in the 10 individuals (Table S4). In order to find the relationship between ALT/AST and intestinal microbiota, a Spearman's correlation analysis was performed. In phylum level, ALT had a weak positive correlation with Euryarchaeota (r = .28, P = .01), and AST had weak positive correlation with Actinobacteria (r = .26, P = .01) and Synergistetes (r = .23, P = .03; Table S5).

| Composition of faecal bacteria
At the phylum level, the majority of the 16S reads were classi- and Synergistetes were also detected, representing 1% of the total reads analysed. At the phylum level, the relative abundance of Verrucomicrobia was considered to be higher in the SHS group than in the controls (P = .049; Table S6). There was no significant difference in ratio of Firmicutes/Bacteroidetes between SHS and control group (P = .11,  Figure S3). LEfSe analysis, an algorithm for high-dimensional biomarker discovery, further confirms these significant differences. Twenty-four discriminative features (LDA score ≥2) with relative abundances varied significantly between the SHS group and the control group (Figures 4   and 5). Taxon annotation of OTUs from all samples was shown in Table S7.

| Using a random forest model to discriminate between the SHS group and the control group
A random forest approach was used to classify samples into SHS or control groups. The random forest diagnosis model was validated by 10-fold cross-validation. The AUC for the random forest diagnostic model of SHS was 0.79 (95% CI: 0.77-0.81; Figure 6). The confusion matrix of the model is shown in Table 2.

| D ISCUSS I ON
In the present study, alpha diversity of intestinal microbiota in SHS individuals was significantly higher compared with that of healthy controls. In addition, beta diversity was significantly different between SHS and healthy controls, but the difference is not big enough. Sixteen intestinal bacterial biomarkers of SHS were identified by LEfSe analysis. The intestinal microbiota can effectively discriminate individuals with SHS from the controls, with an AUC of 0.79 (95% CI: 0.77-0.81). This is the first attempt to explore the association between SHS and intestinal microbiota.
Alpha diversity, the mean species diversity of habitats at a local scale, includes two levels: richness and evenness. 31  Actinobacteria increased in acute-on-chronic alcohol mice while decreased in cirrhotic patients. 35,36 Increased Synergistetes was found in bile from individuals with opisthorchiasis. 37 As mentioned above, these bacterial abundances are altered in several diseases, but it is difficult to clarify how these bacteria affect the level of ALT/AST.

Alterations of intestinal microbiota composition are suspected
to affect the host inflammatory and metabolic responses, and several studies have shown that disruption of intestinal microbiota equilibrium was associated with chronic diseases. [38][39][40][41] In our study, the large majority of all sequences identified belonged to one of four phyla, Firmicutes, Bacteroidetes, Proteobacteria or Actinobacteria, consistent with previous reports from human intestinal microbiota studies. 42 However, the relative proportions of Verrucomicrobia were significantly higher in SHS individuals than in healthy controls. Previous study found that the relative proportions of Verrucomicrobia increased in the intestine following broad-spectrum antibiotic treatment. 43 This difference may be attributed to a medication history of antibiotic use. Participants with SHS are susceptible to diseases, and they have poor resistance to disease. Therefore, the frequency of antibiotic consumption might be higher than those of healthy. Although participants have a history of antibiotic consumption in the previous 2 months were excluded, participates used antibiotic before 2 months might affect the intestinal microbiota. In another study, the abundance of Verrucomicrobia was found to be modulated by the immune system, 44  There are several limitations in our study that may influence the results. First, in this study, we included more females than males, which might introduce selection bias in the association between intestinal microbiota and SHS. The prevalence of SHS is higher in females than that in males, and the response rate in faecal sample collection is also higher in females than in males. To increase the comparability between groups, we matched SHS cases and controls using propensity score matching based on age, gender and BMI.
Second, the design of a case-control study based on a cross-sectional study makes it inevitable to avoid selection bias and may overestimate the accuracy of diagnosis for SHS. Third, the diet of participants, the main associated factor of the intestinal microbiota, 66 was not controlled, which might introduce confounding bias.
Forth, the cortisol and adrenaline/noradrenaline, the only known measurable biochemical parameter of SHS, were not included in our study. Finally, the sample size of our study is relatively small, resulting in the absence of statistical power. Therefore, further cohort studies with larger sample sizes are needed to explore the causal association between SHS and the intestinal microbiota.

| CON CLUS ION
In this case-control study with a relatively small sample size, we first demonstrated that the intestinal microbiota is associated with SHS.
Alteration of intestinal microbiota occurs with SHS, an early stage of disease, which might shed light on the importance of intestinal microbiota in the primary prevention of NCDs.

CO N FLI C T O F I NTE R E S T
The authors declare no conflicts of interest associated with this study.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.