Abstract

Background Previous studies have shown that the gut microbiome is associated with thyroid diseases, including Graves’ disease, Hashimoto's disease, thyroid nodules, and thyroid cancer. However, the association between intestinal flora and primary hypothyroidism remains elusive. We aimed to characterize gut microbiome in primary hypothyroidism patients.

Methods Fifty-two primary hypothyroidism patients and 40 healthy controls were recruited. The differences in gut microbiota between the two groups were analyzed by 16S rRNA sequencing technology. Fecal microbiota transplantation (FMT) was performed in mice using flora from both groups; changes in thyroid function were then assessed in the mice.

Results There were significant differences in α and β diversities of gut microbiota between primary hypothyroidism patients and healthy individuals. The random forest analysis indicated that four intestinal bacteria (Veillonella, Paraprevotella, Neisseria, and Rheinheimera) could distinguish untreated primary hypothyroidism patients from healthy individuals with the highest accuracy; this was confirmed by receiver operator characteristic curve analysis. The short chain fatty acid producing ability of the primary hypothyroidism patients’ gut was significantly decreased, which resulted in the increased serum lipopolysaccharide (LPS) levels. The FMT showed that mice receiving the transplant from primary hypothyroidism patients displayed decreased total thyroxine levels.

Conclusions Our study suggests that primary hypothyroidism causes changes in gut microbiome. In turn, an altered flora can affect thyroid function in mice. These findings could help understand the development of primary hypothyroidism and might be further used to develop potential probiotics to facilitate the adjuvant treatment of this disease.

Introduction

Hypothyroidism (underactive thyroid), of which the primary form is the most common type, is a condition in which the thyroid gland does not produce enough triiodothyronine (T3) and thyroxine (T4). Its causes are varied, including autoimmune disease, hyperthyroidism treatments, radiation therapy, iodine deficiency, congenital disease, pituitary disorder, thyroid surgery, and certain medications [1]. Hypothyroidism has become a global public health concern due to its increasing prevalence, numerous complications, and the need of lifelong treatment. Hypothyroidism is very common, affecting approximately 4–10% of the population worldwide [2]. In iodine-sufficient countries, the prevalence of hypothyroidism ranges from 1% to 2%, rising to 7% in individuals aged between 85 and 89 years. In the absence of age-specific reference ranges for thyrotropin or thyroid stimulating hormone (TSH), an aging population is likely to result in a higher prevalence of hypothyroidism. This condition, which is approximately 10 times more prevalent in women than men [3], can lead to a myriad of health problems including goiter, mental health issues, peripheral neuropathy, myxedema, infertility, and birth defects. It may also be associated with an increased risk of heart disease and heart failure, primarily due to high levels of low-density lipoprotein (LDL) cholesterol [2].

Recent technological advancements have allowed an extensive study of the relation between the composition of the human microbiome (including the gut microbiome) and many diseases [4]. The human intestinal microbiota is composed of 1013–1014 microorganisms whose collective genome (‘microbiome’) contains at least 100 times more genes than that of humans [5]. Human intestinal tract fosters a number of phyla, majorly including Firmicutes, Bacteroidetes, Actinobacteria, and Proteobacteria [6] which are closely related to human health. For example, the studies conducted by Stefanaki C and colleagues indicate that the gut microbiome is closely related to diabetes [7,8]. Recently, studies have reported gut microbiome composition changes in thyroid cancer, thyroid nodules, Hashimoto's thyroiditis, and Graves’ disease [9–12]. The gut microbiome could even affect the whole body health by the gut-thyroid axis [13]. Furthermore, a contribution of gut microbiome to lipid levels has been reported [14], indicating a potential link between hypothyroidism and gut microbiome. However, the specific changes of intestinal flora in patients with primary hypothyroidism and the relationship between the gut microbiome and primary hypothyroidism have not yet been reported.

Therefore, the present study aimed to analyze the gut microbiome of primary hypothyroidism patients and compare it to that of healthy controls by sequencing the 16S ribosomal RNA (rRNA) gene. Additionally, we aimed to discover the relationship between clinical characteristics of primary hypothyroidism and abnormalities in the composition of intestinal flora. Finally, we performed a fecal transplant from primary hypothyroidism patients to mice in order to assess the effects of the intestinal flora on thyroid function.

Materials and methods

Study cohort and recruitment of participants

We recruited 92 participants, including 52 untreated primary hypothyroidism patients and 40 healthy controls in our study at the Shandong Provincial Hospital, Jinan, China. There were no differences in terms of age, sex, and body mass index (BMI) of the healthy volunteers and the individuals of primary hypothyroidism groups. Patients with primary hypothyroidism caused by autoimmune thyroiditis and/or iodine deficiency based on the test of thyroglobulin antibody (TGAb), thyroperoxidase antibody (TPOAb), thyrotropin receptor antibody (TRAb), and urinary iodine (UI) were excluded from the present study, because these etiologies could seriously disturb intestinal flora. All patients had received no medical treatment for at least half a year. The following exclusion criteria were applied to all groups: pregnancy; cigarette smoking; alcohol addiction; diabetes mellitus; recent (<3 months prior) use of antibiotics, probiotics, prebiotics, symbiotics, hormonal medication, laxatives, proton pump inhibitors, insulin sensitizers, or Chinese herbal medicine; a known history of disease with an autoimmune component, such as multiple sclerosis (MS), rheumatoid arthritis, irritable bowel syndrome (IBS), or inflammatory bowel disease (IBD); and a history of malignancy or any gastrointestinal tract surgery. Informed consent was obtained from all participants. All procedures were performed in compliance with relevant laws and institutional guidelines.

Sample collection

Peripheral blood (5 ml) was collected from all participants in the morning after an overnight fast (≥8 h) for laboratory tests, including free T3 (FT3), free T4 (FT4), TSH, TGAb, TPOAb, TRAb, blood glucose (BG), total triglyceride (TG), total cholesterol (TC), high-density lipoprotein (HDL), and low-density lipoprotein (LDL). Urine samples were collected for urinary iodine (UI) test. Fecal samples were collected by the patients in the morning using disposable sterile forceps. Basic information, such as acquisition time and patient name, was recorded in the sample collection box. Upon collection, each fecal sample was immediately divided into aliquots and stored at −80°C until DNA extraction.

Assays of thyroid function and thyroid-related autoantibodies

Serum levels of FT3, FT4, TSH, TGAb, TPOAb, BG, TG, TC, HDL, and LDL were assayed using chemiluminescent immunoassays (ADVIA Centaur XP, Germany) and serum level of TRAb was assayed using immunochemistry cobas e 801 module (Roche, Mannheim, Germany) according to the manufacturer's instructions. UI was assayed using the colorimetric assay kit. Reference ranges were as follows: TSH, 0.55–4.78 μIU/ml; FT4, 11.5–22.7 pmol/l; FT3, 3.5–6.5 pmol/l; TPOAb, 0.00–60 IU/ml; TGAb, 0.00–60 IU/ml; TRAb, 0.00–1.58 IU/l; BG, 3.9–6.3 mmol/l; TG, 0.4–1.8 mmol/l; TC, 3.6–6.2 mmol/l; HDL, 0.8–1.5 mmol/l; LDL, 0.5–3.36 mmol/l; and UI, 100–300 μg/l.

16s rRNA gene sequencing

Total genome DNA from the fecal samples was extracted using the improved CTAB (cetyl trimethylammonium bromide) method [15]. We used Nanodrop 2000 (Thermo Scientific) spectrophotometer to determine the concentration of the extracted DNA. The V1–V2 regions of the 16S rRNA gene were amplified and sequenced on an Illumina HiSeq 2500 system. The PCR was conducted using the bacterial universal primers (Univ 27F/Univ 338R) (Supplemnetary Table S1) [16]. Then, the amplicons were sequenced on an Illumina HiSeq 2500 system and 250 bp paired-end reads were generated.

Bioinformatic analysis of sequencing data

The raw sequencing data were merged and quality filtered using the Laser FLASH method, described by Magoc and Salzberg [17]. Paired-end reads were assigned to each sample according to the unique barcodes. The 16S rRNA OTUs (operational taxonomic units) were selected from the combined reads using the Quantitative Insights into Microbial Ecology (QIIME) software package (http://qiime.org/) and were annotated for taxonomic information by the Greengenes (13.8 release) ribosomal database. Core_diversity_analyses.py scripts were used to analyze alpha (α) diversity (within samples) and beta (β) diversity (among samples). KEGG pathways and COG (clusters of orthologous groups) functions were analyzed using PICRUSt software [18,19]. Spearman correlation was calculated using R package psych. To determine if taxonomic differences in the microbiota could be used to classify samples into different cohorts and the ability of prediction, a machine learning algorithm named random forest (RF) was applied to analyze the genus-level abundances of gut bacteria by R package randomForest [20]. We applied the receiver operating characteristic (ROC) analysis to assess the performance of the primary hypothyroidism metagenomic biomarkers by using the ‘pROC’ package in R software [20]. Figures were mainly painted by R package ‘ggplot2’ (v 3.1.0). Heatmaps were made by R package ‘pheatmap’ (v 1.0.10).

Fecal microbiota transplantation

Twenty male specific pathogen-free (SPF) BALB/c mice (6–8 weeks old) were purchased from SPF (Beijing) Biotechnology Co., Ltd. and raised in a temperature and humidity controlled laminar flow room under a 12 h light/12 h dark cycle with free access to water and food in Animal Center of Shandong Provincial Hospital. All animal experiments were conducted in accordance with the principles and procedures outlined in the Institutional Animal Care and Use Committee in Shandong Provincial Hospital approved by the Animal Ethics Committee of Shandong First Medical University (2019088). After 3 days of adaptive feeding, all mice were then administered antibiotics treatment before FMT. Then, all the mice were divided into two groups randomly on average: n(FMT-C) = 10 and n(FMT-H) = 10. The mice were transplanted fecal microbiota from healthy individuals (FMT-C) and primary hypothyroidism patients (FMT-H), respectively. During this process, the environments in which the mice lived were kept clean as far as possible. The method of anesthesia in the present study was as follows: a 50 mg/ml solution of pentobarbital in sterile saline was administered IP (intraperitoneal) at a dose of 60 mg/kg. Finally, mice were killed by cervical dislocation. The methods of antibiotics treatment and FMT were described as below.

All mice were given the following cocktail for 2 weeks: 500 mg ampicillin, 250 mg vancomycin, 500 mg neomycin-sulfate, 500 mg metronidazole, and 10 g grape Kool-Aid in 500 ml water, which was sterile filtered through a 0.22 μm filter. Water bottles were changed once per week [21,22].

Feces from healthy individuals and primary hypothyroidism patients were collected as described before. Each fecal sample was diluted in phosphate buffered saline (PBS) to obtain a concentration of 120 mg/ml, and the solution was homogenized by vortexing and minimally clarified by low-speed centrifugation for 5 min at 300 g to get the supernatants. Glycerol was added to each supernatant and the concentration of supernatant was adjusted to 20%. Then, all supernatants were stored at −80°C. Before FMT, all the supernatants were mixed in one container. Mice were then gavaged three times with 200 μl of the mixture each time. Gavage occurred on day 1, day 2, and day 5 post antibiotic treatments [22].

Measurement of serum T4 in mice and lipopolysaccharide assay

The blood of mice was collected from the tail vein. The level of serum total T4 (TT4) was measured using T4 kits (SenBeiJia Biological Technology, Nanjing, China) by SpectraMax Microplate Reader (Molecular device, U.S.A.) according to the instructions. The serum level of lipopolysaccharide (LPS), the Gram-negative bacterial endotoxin, was quantified by EC Endotoxin Test Kit (Bioendo, Xiamen, China) according to the manufacturer's manual.

Quantitative real-time PCR

The gDNA from the fecal samples were obtained by the method described previously. Then, SYBR Green quantitative real-time PCR (qPCR) technology was used to analyze the changes of phyla of Firmicutes, Bacteroidetes, and the genes of ButA (Butyryl-CoA CoA transferase), LcdA (lactoyl-CoA dehydratase), PduP (propionaldehyde dehydrogenase), and MmdA (methylmalonyl-CoA decarboxylase). To assess mRNA expression of tight junction molecules including zonula occludens-1 (ZO-1), occludin and junctional adhesion molecule-A (JAM-A), total RNA was extracted from colon tissue samples using the TRIzol™ reagent (Invitrogen, U.S.A.) and was reverse-transcribed using the PrimeScript RT Reagent Kit (Takara, Japan). Then, quantitative PCR was performed using TB Green Premix Ex Taq (Takara, Japan) and specific primers (Supplementary Table S1) on a CFX96 Real-Time PCR System (Bio-Rad, U.S.A.). Relative gene expression was calculated by the 2−∆∆Ct method using 16S rRNA gene (Univ 337F/Univ 518R) or β-actin gene as an internal control.

Statistical analysis

The Mann–Whitney U test or unpaired ­t-test with appropriate correction for multiple comparisons was used to detect significances for continuous outcomes. For categorical variables, the Fisher's exact test was performed. P<0.05 was considered as statistically significant. FDR (false discovery rate) corrected P-value (q-value) <0.05 was deemed to be significant for some specific analysis. Analyses were performed using R software (version 3.5.1), STAMP, and GraphPad Prism 7 software.

Results

Differences between the gut microbiota of primary  hypothyroidism patients and healthy individuals

In the present study, 52 initial patients with primary hypothyroidism (hypothyroidism group) and 40 healthy individuals (control group) were recruited. The demographic characteristics of all the participants are presented in Table 1. The 16S rRNA gene sequencing depth was saturated (Goods coverage>0.99, Supplementary Table S2) and was not significantly different in both groups (Supplementary Figure S1A,B). The Chao1 and Ace, which reflected community richness, were significantly increased in primary hypothyroidism patients (Figure 1A,B). However, the Shannon and Simpson index of the intestinal flora, which reflected community diversity, was significantly lower in primary hypothyroidism patients than in healthy controls (Figure 1C,D). The details of above results are presented in Supplementary Table S2. The Bray–Curtis distance-based community analysis showed an obvious separation between the samples of the healthy individuals and primary hypothyroidism patients, revealing that the microbiota composition differed significantly between the two groups (P<0.001) (Figure 1E).

Composition of gut microbiota is significantly altered in primary hypothyroidism patients

Figure 1
Composition of gut microbiota is significantly altered in primary hypothyroidism patients

(A–D) The α diversity indices (Chao1, Ace, Shannon, and Simpson index) of intestinal flora in healthy individuals and primary hypothyroidism patients. (E) A principal component (PCoA) score plot based on Bray–Curtis distance matrix for all participants. Each point represents the composition of the intestinal microbiota of one participant. The ellipses do not represent statistical significance but rather serve as a visual guide to illustrate group differences. The line in the middle of each box represents the median. The Mann–Whitney U test was used to detect significant changes for A–D. *: P<0.05; **: P<0.01. C: healthy individuals, nC = 40; H: primary hypothyroidism patients, nH = 52.

Figure 1
Composition of gut microbiota is significantly altered in primary hypothyroidism patients

(A–D) The α diversity indices (Chao1, Ace, Shannon, and Simpson index) of intestinal flora in healthy individuals and primary hypothyroidism patients. (E) A principal component (PCoA) score plot based on Bray–Curtis distance matrix for all participants. Each point represents the composition of the intestinal microbiota of one participant. The ellipses do not represent statistical significance but rather serve as a visual guide to illustrate group differences. The line in the middle of each box represents the median. The Mann–Whitney U test was used to detect significant changes for A–D. *: P<0.05; **: P<0.01. C: healthy individuals, nC = 40; H: primary hypothyroidism patients, nH = 52.

Table 1
Demographic and clinical characteristics of primary hypothyroidism patients and healthy individuals
ControlHypothyroidismP value
Number 40 52  
Sex (M/F) 15/25 19/33 ns 
Age (Y) 43.13 ± 8.27 42.56 ± 9.86 ns 
BMI (kg/m222.28 ± 1.76 22.63 ± 2.07 ns 
FT3 (pml/l) 4.87 (0.55) 2.56 (0.80) **** 
FT4 (pmol/l) 15.89 (1.69) 10.23 (4.81) **** 
TSH (μIU/ml) 1.71 (1.26) 13.6 (75.55) **** 
TPOAb (IU/ml) 27.35 (18.78) 27.97 (16.45) ns 
TGAb (IU/ml) 24.75 (26.48) 25.74 (29.7) ns 
TRAb (IU/l) 0.91 (0.48) 0.91 (0.72) ns 
BG (mmol/l) 5.01 (0.8) 5.11 (0.73) ns 
TG (mmol/l) 0.96 (0.9) 1.13 (0.79) ns 
TC (mmol/l) 4.99 (0.99) 5.53 (1.66) ns 
HDL (mmol/l) 1.44 (1.25) 1.35 (0.89) ns 
LDL (mmol/l) 2.24 (0.68) 2.4 (1.0) ns 
UI (μg/l) 165 (77.3) 171 (98.2) ns 
ControlHypothyroidismP value
Number 40 52  
Sex (M/F) 15/25 19/33 ns 
Age (Y) 43.13 ± 8.27 42.56 ± 9.86 ns 
BMI (kg/m222.28 ± 1.76 22.63 ± 2.07 ns 
FT3 (pml/l) 4.87 (0.55) 2.56 (0.80) **** 
FT4 (pmol/l) 15.89 (1.69) 10.23 (4.81) **** 
TSH (μIU/ml) 1.71 (1.26) 13.6 (75.55) **** 
TPOAb (IU/ml) 27.35 (18.78) 27.97 (16.45) ns 
TGAb (IU/ml) 24.75 (26.48) 25.74 (29.7) ns 
TRAb (IU/l) 0.91 (0.48) 0.91 (0.72) ns 
BG (mmol/l) 5.01 (0.8) 5.11 (0.73) ns 
TG (mmol/l) 0.96 (0.9) 1.13 (0.79) ns 
TC (mmol/l) 4.99 (0.99) 5.53 (1.66) ns 
HDL (mmol/l) 1.44 (1.25) 1.35 (0.89) ns 
LDL (mmol/l) 2.24 (0.68) 2.4 (1.0) ns 
UI (μg/l) 165 (77.3) 171 (98.2) ns 

Data are shown as mean±SD and median (interquartile range, IQR). The Chi-squared test (sex), t-test (age and BMI) and Mann–Whitney U test (FT3, FT4, TSH, TPOAb, TGAb, TRAb, BG, TG, TC, HDL, LDL, and UI) were used to detect significant changes. ns: not significant; ****: P<0.0001.

Control, healthy individuals; hypothyroidism, primary hypothyroidism patients; M, male; F, female; Y, year; BMI, Body Mass Index; FT3, free T3; FT4, free T4; TSH, thyrotropin; TGAb, thyroglobulin antibody; TPOAb, thyroperoxidase antibody; TRAb, thyrotropin receptor antibody; BG, blood glucose; TG, total triglyceride; TC, total cholesterol; HDL, high density lipoprotein; LDL, low-density lipoprotein; UI, urinary iodine.

At the phylum level, more than 95% of bacteria were mainly composed of Firmicutes, Bacteroidetes, and Proteobacteria (Figure 2A), where there were common in the two groups in terms of dominating composition. Abundance of Bacteroidetes in primary hypothyroidism patients was decreased compared with that in the healthy individuals’ group (Figure 2B), which was further confirmed by qPCR (Figure 2C). Thereby, the ratio of Firmicutes/Bacteroidetes was significantly increased in primary hypothyroidism patients (Figure 2D). At the genus level (abundance>0.2%), the abundances of 13 bacterial genera were significantly decreased, while four bacterial genera were markedly increased in the primary hypothyroidism patients; they all belonged to Firmicutes, Proteobacteria, and Bacteroidetes, except Fusobacterium, which belonged to Fusobacteria (Figure 2E).

Gut microbiota of primary hypothyroidism patients differed from that of healthy individuals at the phylum and genus levels

Figure 2
Gut microbiota of primary hypothyroidism patients differed from that of healthy individuals at the phylum and genus levels

(A) Comparison of the intestinal flora at the phylum level. (B) Relative abundances of main intestinal bacteria at the phylum level in the control and primary hypothyroidism groups. (C) Fold changes in the abundance of Firmicutes and Bacteroidetes in the intestinal flora. They were determined by qPCR. nC = 10; nH = 10. (D) The Firmicutes/Bacteroidetes ratio in primary hypothyroidism patients and healthy individuals calculated using the 16S rRNA gene sequences data. (E) Heatmap of the relative abundances of intestinal bacteria with significant differences (corrected P<0.05 by Mann–Whitney U test) at the genus level in the control and primary hypothyroidism groups. The color bar indicates the Z score, which represents the relative abundance. Z scores <0 and >0 indicate that the relative abundance is lower and higher than the mean, respectively. Data are represented as median (IQR). The Mann–Whitney U test was used to detect significant changes. ns: not significant; **: P<0.01; ***: P<0.001. C: healthy individuals, nC = 40; H: primary hypothyroidism patients, nH = 52.

Figure 2
Gut microbiota of primary hypothyroidism patients differed from that of healthy individuals at the phylum and genus levels

(A) Comparison of the intestinal flora at the phylum level. (B) Relative abundances of main intestinal bacteria at the phylum level in the control and primary hypothyroidism groups. (C) Fold changes in the abundance of Firmicutes and Bacteroidetes in the intestinal flora. They were determined by qPCR. nC = 10; nH = 10. (D) The Firmicutes/Bacteroidetes ratio in primary hypothyroidism patients and healthy individuals calculated using the 16S rRNA gene sequences data. (E) Heatmap of the relative abundances of intestinal bacteria with significant differences (corrected P<0.05 by Mann–Whitney U test) at the genus level in the control and primary hypothyroidism groups. The color bar indicates the Z score, which represents the relative abundance. Z scores <0 and >0 indicate that the relative abundance is lower and higher than the mean, respectively. Data are represented as median (IQR). The Mann–Whitney U test was used to detect significant changes. ns: not significant; **: P<0.01; ***: P<0.001. C: healthy individuals, nC = 40; H: primary hypothyroidism patients, nH = 52.

In order to find bacteria most closely related to primary hypothyroidism at the genus level, we conducted a variety of analyses. The RF analysis indicated that four intestinal bacteria (Veillonella, Paraprevotella, Neisseria, and Rheinheimera) could distinguish untreated primary hypothyroidism patients from healthy individuals with the highest accuracy (Figure 3A). These findings indicated that these four bacteria might play an important role in the process of primary hypothyroidism. To evaluate primary hypothyroidism based on these biomarkers, the two group data were assessed by an ROC curve. We found that the area under the receiver operating characteristic (AUC) curves of each of the four bacteria was more than the 80% (Figure 3B–E), indicating that each of the four bacteria could be used to classify primary hypothyroidism patients with a high accuracy. Among the four bacteria, Veillonella and Paraprevotella were significantly reduced in the primary hypothyroidism patients, while Neisseria and Rheinheimera were significantly increased. A further ROC analysis based on the merger of these four bacteria showed a 92% AUC (Figure 3F). This also verifies the results of our RF.

RF and ROC curves analysis

Figure 3
RF and ROC curves analysis

(A) Results of the RF analysis. The bacterial genera that could significantly discriminate between primary hypothyroidism patients and healthy individuals are presented in descending order. Four bacterial genera that could distinguish untreated primary hypothyroidism patients from healthy individuals with the highest accuracy are indicated by rectangles. (B–E) Accuracy of the four gut bacteria-based RF predictive model is measured by the area under receiver-operating characteristic curve (AUC) in the primary hypothyroidism patients and healthy individuals, and the violin plot of bacteria for all participants in the primary hypothyroidism and the healthy individuals are shown. (B) Veillonella; (C) Paraprevotella; (D) Neisseria; (E) Rheinheimera. (F) ROC curve of the RF model using the relative abundances of the four genera together. ****: P<0.0001. nC = 40; nH = 52.

Figure 3
RF and ROC curves analysis

(A) Results of the RF analysis. The bacterial genera that could significantly discriminate between primary hypothyroidism patients and healthy individuals are presented in descending order. Four bacterial genera that could distinguish untreated primary hypothyroidism patients from healthy individuals with the highest accuracy are indicated by rectangles. (B–E) Accuracy of the four gut bacteria-based RF predictive model is measured by the area under receiver-operating characteristic curve (AUC) in the primary hypothyroidism patients and healthy individuals, and the violin plot of bacteria for all participants in the primary hypothyroidism and the healthy individuals are shown. (B) Veillonella; (C) Paraprevotella; (D) Neisseria; (E) Rheinheimera. (F) ROC curve of the RF model using the relative abundances of the four genera together. ****: P<0.0001. nC = 40; nH = 52.

Correlations among intestinal bacteria or clinical features of primary hypothyroidism with bacteria

Our results demonstrated that the correlations among many intestinal bacteria of primary hypothyroidism patients were significantly changed (Figure 4). Compared with those in healthy individuals, almost all correlations between bacteria in the primary hypothyroidism patients had changed, except that between Streptococcus and Haemophilus, Blautia and Dorea, Lachnobacterium and Coprococcus; however, in the disease group, the positive correlation was still stronger than that in the healthy group. Overall, compared with healthy individuals, the association and correlation between the intestinal flora in the primary hypothyroidism patients was more intense and the negative correlation almost disappeared, while the positive correlation was significantly increased, which was consistent with the results of α diversity. In healthy individuals, Bacteroides, which is the numerically predominant genera in the human gut [23,24], was negatively correlated with many bacteria (such as Dorea, Faecalibacterium, and Prevotella), but these correlations disappeared in the primary hypothyroidism patients. Bacteroides was at the core of the network in healthy human gut, but isolated in the primary hypothyroidism group, indicating that the inhibition of Bacteroides exerting on other bacteria in the intestinal tract disappeared. Additionally, Veillonella and Paraprevotella were positively correlated with some bacteria in healthy individuals, while this relationship disappeared in the primary hypothyroidism patients, which also explained the two bacteria being significantly reduced in the primary hypothyroidism group and becoming biomarkers. However, the positive correlation between Neisseria, Rheinheimera, and other bacteria in the intestine was increased in the primary hypothyroidism patients, which showed that the two bacteria were enriched in the primary hypothyroidism group and became biomarkers for primary hypothyroidism.

Correlation network of gut microbiota

Figure 4
Correlation network of gut microbiota

Correlation network of gut microbiota in healthy individuals (control) and primary hypothyroidism patients (hypothyroidism). The significant strong correlations (|r|>0.3 and corrected P<0.05 by t-test) among the intestinal bacterial genera are presented in the network. The red and green edges represent the positive and negative correlations, respectively. The spot colors represent different bacterial phyla. The thickness of the edges represents the strength of the correlation. The dotted rectangles indicate the bacteria that are emphasized in results section.

Figure 4
Correlation network of gut microbiota

Correlation network of gut microbiota in healthy individuals (control) and primary hypothyroidism patients (hypothyroidism). The significant strong correlations (|r|>0.3 and corrected P<0.05 by t-test) among the intestinal bacterial genera are presented in the network. The red and green edges represent the positive and negative correlations, respectively. The spot colors represent different bacterial phyla. The thickness of the edges represents the strength of the correlation. The dotted rectangles indicate the bacteria that are emphasized in results section.

To clarify the pathological significance of the altered intestinal bacteria in primary hypothyroidism patients, the correlations between their relative abundances and clinical features of primary hypothyroidism (serum levels of FT3, FT4 and TSH) were analyzed through the Spearman's rank correlation coefficient method. At the phylum level, the results showed that Spirochaetae, Thermi, Actinobacteria, TM7, and SR1 were negatively correlated with the serum levels of FT3 and FT4, while these were positively correlated with the serum level of TSH. With respect to Tenericutes, Lentisphaerae, Verrucomicrobia, Bacteroidetes, and Synergistetes, the results were exactly the opposite of the above (Figure 5A). At the genus level, intestinal bacteria except Neisseria, Streptococcus, Ruminococcus, and Dorea were positively correlated with the serum levels of FT3 and FT4, but negatively correlated with the serum level of TSH (Figure 5B). Among the altered bacteria with significant association to the clinical indicators of primary hypothyroidism, Veillonella and Paraprevotella, which were enriched in healthy people, were positively associated with FT3 and FT4, and negatively associated with TSH. However, Neisseria, enriched in the primary hypothyroidism group, had exactly the opposite result.

Correlations among intestinal bacteria and clinical indicators of primary hypothyroidism

Figure 5
Correlations among intestinal bacteria and clinical indicators of primary hypothyroidism

(A) Heatmap of correlations between three clinical indicators of primary hypothyroidism and the abundances of the intestinal bacterial phyla. The color bar with numbers indicates the correlation coefficients. The important bacterial phyla are marked with rectangles. (B) Heatmap of the correlations between three clinical indicators of primary hypothyroidism and the top 17 most abundant bacterial genera whose abundances significantly changed in primary hypothyroidism patients. The color bar with numbers indicates the correlation coefficients. The false discovery rate (FDR)-adjusted P-values by t-tests are shown by asterisks (*: corrected P<0.05; **: corrected P<0.01; ***: corrected P<0.001).

Figure 5
Correlations among intestinal bacteria and clinical indicators of primary hypothyroidism

(A) Heatmap of correlations between three clinical indicators of primary hypothyroidism and the abundances of the intestinal bacterial phyla. The color bar with numbers indicates the correlation coefficients. The important bacterial phyla are marked with rectangles. (B) Heatmap of the correlations between three clinical indicators of primary hypothyroidism and the top 17 most abundant bacterial genera whose abundances significantly changed in primary hypothyroidism patients. The color bar with numbers indicates the correlation coefficients. The false discovery rate (FDR)-adjusted P-values by t-tests are shown by asterisks (*: corrected P<0.05; **: corrected P<0.01; ***: corrected P<0.001).

Imputed microbiome function and the declining SCFAs-producing ability

Given the structural differences in intestinal bacteria of both groups, we subsequently examined whether primary hypothyroidism would cause functional changes within the microbiome. In the absence of shotgun metagenomic sequencing data, we applied PICRUSt to our 16S rRNA gene survey to predict metagenome functional content. PICRUSt is a computational approach that uses evolutionary modeling to predict the present gene families from 16S data in relation to a reference genome database. The imputed relative abundances of KEGG pathways in each respective sample were used to predict changes in metabolic function within the microbiomes of the primary hypothyroidism patients compared with the healthy individuals (Figure 6A,B). The Bray–Curtis distance-based community analysis showed an obvious separation between the samples of the healthy individuals and the primary hypothyroidism patients, revealing that the metabolic function composition differed significantly between the two groups (P = 0.006) (Figure 6A). The KEGG pathways that exhibited the greatest statistical difference in primary hypothyroidism and healthy individuals were oxidative phosphorylation, protein digestion and absorption, citrate cycle (TCA cycle) and butanoate, propanoate, inositol phosphate, purine, tryptophan metabolism, with inositol phosphate and tryptophan metabolism having a higher predicted relative abundance while oxidative phosphorylation, protein digestion and absorption, citrate cycle (TCA cycle) and butanoate, propanoate, and purine metabolism having lower predicted relative abundance in primary hypothyroidism patients (Figure 6B).

Imputed metagenomic differences between primary hypothyroidism patients and healthy individuals

Figure 6
Imputed metagenomic differences between primary hypothyroidism patients and healthy individuals

(A) The relative abundance of metabolic pathways encoded in each imputed sample metagenome was analyzed for a principal component (PCoA) using R. (B) The significant differences in metagenomic functions of primary hypothyroidism patients compared with healthy individuals (corrected P<0.05 and confidence intervals = 95%). The important metagenomic functions are marked with rectangles. This analysis was performed by software STAMP. nC = 40; nH = 52. (C) The abundance changes of the key enzymes for producing butyrate and propionate in the intestinal flora of primary hypothyroidism patients and healthy individuals by qPCR. nC = 8; nH = 8. (D) Serum levels of LPS in primary hypothyroidism patients and healthy individuals measured by the LAL test. Data are represented as median (IQR). The Mann–Whitney U test was used to detect significant changes. **: P<0.01; ***: P<0.001; ****: P<0.0001. C: healthy individuals; H: primary hypothyroidism patients.

Figure 6
Imputed metagenomic differences between primary hypothyroidism patients and healthy individuals

(A) The relative abundance of metabolic pathways encoded in each imputed sample metagenome was analyzed for a principal component (PCoA) using R. (B) The significant differences in metagenomic functions of primary hypothyroidism patients compared with healthy individuals (corrected P<0.05 and confidence intervals = 95%). The important metagenomic functions are marked with rectangles. This analysis was performed by software STAMP. nC = 40; nH = 52. (C) The abundance changes of the key enzymes for producing butyrate and propionate in the intestinal flora of primary hypothyroidism patients and healthy individuals by qPCR. nC = 8; nH = 8. (D) Serum levels of LPS in primary hypothyroidism patients and healthy individuals measured by the LAL test. Data are represented as median (IQR). The Mann–Whitney U test was used to detect significant changes. **: P<0.01; ***: P<0.001; ****: P<0.0001. C: healthy individuals; H: primary hypothyroidism patients.

Some key enzymes produce butyrate and propionate in gut flora, such as ButA, LcdA, PduP, and MmdA. We tested the expression levels of these genes by qPCR. The abundances of these key enzyme genes were all significantly decreased in intestinal flora of primary hypothyroidism patients, confirming the significant decrease in the short chain fatty acids (SCFAs)-producing ability (Figure 6C). Additionally, the serum level of LPS was significantly increased in primary hypothyroidism patients (Figure 6D).

Potential role of the microbiome in primary hypothyroidism

The above results illustrated the effects of primary hypothyroidism on intestinal flora. FMT was performed in mice to evaluate whether this altered intestinal flora could have a pathogenic role on primary hypothyroidism. The progress of FMT of the two cohorts is shown in Figure 7A. Before FMT, there was no difference in serum TT4 levels between the two groups. After the procedure, serum TT4 levels in the FMT-H group started to decrease in the second and fourth weeks (no statistical difference), and became significantly lower in the sixth week after transplantation, compared with those of FMT-C group (P = 0.01) (Figure 7B). Additionally, the abundances of SCFAs-producing key enzyme genes were significantly decreased in the intestinal flora of FMT-H mice at the sixth week after FMT (Figure 7C). The mRNA expression of tight junction molecules including zonula occludens-1 (ZO-1), occludin and junctional adhesion molecule-A (JAM-A), was lower in the colon of FMT-H mice than FMT-C mice at the sixth week after FMT (Figure 7D). Moreover, serum LPS level was significantly increased in the FMT-H group at the sixth week after FMT (Figure 7E).

FMT in mice

Figure 7
FMT in mice

(A) Mice were treated with antibiotics to eliminate the previous microbiota and then FMT was performed. (B) Dynamic changes in serum total thyroxine (TT4) in both groups. (C) Abundance changes of the key enzymes for producing butyrate and propionate in the intestinal flora of both groups of mice at the sixth week after FMT by qPCR. (D) Relative mRNA expression of the tight junction molecules occludin, ZO-1, and JAM-A are shown at the sixth week from FMT. (E) Serum levels of LPS of two groups of mice at the sixth week from FMT. Data are represented as median (IQR). The Mann–Whitney U test was used to detect significant changes. ns: not significant; *: P<0.05; **: P<0.01; ***: P<0.001. FMT-C: mice received fecal microbiota from healthy individuals, nFMT-C = 10; FMT-H: mice received fecal microbiota from primary hypothyroidism patients, nFMT-H = 10.

Figure 7
FMT in mice

(A) Mice were treated with antibiotics to eliminate the previous microbiota and then FMT was performed. (B) Dynamic changes in serum total thyroxine (TT4) in both groups. (C) Abundance changes of the key enzymes for producing butyrate and propionate in the intestinal flora of both groups of mice at the sixth week after FMT by qPCR. (D) Relative mRNA expression of the tight junction molecules occludin, ZO-1, and JAM-A are shown at the sixth week from FMT. (E) Serum levels of LPS of two groups of mice at the sixth week from FMT. Data are represented as median (IQR). The Mann–Whitney U test was used to detect significant changes. ns: not significant; *: P<0.05; **: P<0.01; ***: P<0.001. FMT-C: mice received fecal microbiota from healthy individuals, nFMT-C = 10; FMT-H: mice received fecal microbiota from primary hypothyroidism patients, nFMT-H = 10.

Discussion

To date, only two studies had involved the association between hypothyroidism and intestinal bacterial dysbiosis [25,26]. In the present study, we revealed the composition and relationship between intestinal flora and primary hypothyroidism. In comparison with healthy subjects, primary hypothyroidism patients showed both an altered composition and inter-relation of intestinal bacteria, which related to clinical features of primary hypothyroidism as well. Moreover, transplantation of intestinal flora from primary hypothyroidism patients resulted in a decrease in thyroid function in mice.

Regarding the composition of the gut microbiome, there was a significant difference in α diversity between both groups. First, Chao1 and Ace were higher in the primary hypothyroidism patients, which indicated that the community richness in the primary hypothyroidism patients’ guts was increased. This finding is consistent with a study from Lauritano et al. [25], which showed that hypothyroidism was more likely to cause intestinal bacterial dysbiosis manifested as small intestinal bacterial overgrowth. Moreover, the results of network analysis indicated the existence of bacterial overgrowth in the intestinal tract of primary hypothyroidism patients. These changes in microflora might be closely related to the progression of primary hypothyroidism. Substrate availability and physicochemical conditions in the intestinal tract are key factors that affect the composition and activity of the gut microbiota [27]. The slow peristalsis of gut in primary hypothyroidism patients, which could change the substrate availability and physicochemical conditions in the intestinal tract, may be the reason why intestinal bacterial dysbiosis occurred. Secondly, the lower Shannon and Simpson index of the intestinal flora in primary hypothyroidism patients revealed a decreased community diversity, which is in accordance with the concept that greater overall diversity implies better health [28].

In healthy mammals, the Firmicutes/Bacteroidetes (F/B) ratio is relatively stable, and an increase in the F/B ratio often means a disease state [29]. The F/B ratio of the primary hypothyroidism patients was increased, which is similar to most of the current studies. Many conditions, such as obesity and metabolic syndrome, are associated with a higher intestinal F/B ratio [30–32]. Ley et al. [30] found that the abundance of Bacteroidetes is decreased in obese mice and obese people, while the abundance of Firmicutes is increased. The ratio of the two bacteria was related to obesity. The increased F/B ratio may be a reason why blood lipids are elevated in hypothyroidism patients. In our study, blood lipids were higher in primary hypothyroidism patients than in healthy individuals, although this difference did not reach statistical significance. We speculate the increases in blood lipids caused by abnormal bacterial flora require a long time to develop, and our patients had a short illness history. Further longitudinal research is needed to elucidate the underlying mechanism.

Through a variety of analytical methods, we found that four intestinal bacteria (Veillonella, Paraprevotella, Neisseria, and Rheinheimera) could represent biomarkers for distinguishing untreated primary hypothyroidism patients from healthy individuals. The clinical diagnosis method for primary hypothyroidism is well-established and convenient, so we do not recommend the intestinal marker as a new diagnostic method for primary hypothyroidism. A previous study has shown that the relative abundance of Neisseria was significantly higher in thyroid cancer and thyroid nodules [11]. Our results documented the abundance of Neisseria increased in primary hypothyroidism patients and Neisseria was negatively associated with FT3 and FT4, while positively associated with TSH, which suggest that Neisseria may be closely related to thyroid disease. Another study found that thyroxine supplementation in Neisseria meningitidis-infected mice enhanced bacterial clearance, attenuated the inflammatory responses, and promoted survival [33]. Therefore, it may be possible that correcting Neisseria abundance improves primary hypothyroidism symptoms. Rheinheimera, one of four biomarkers, was exclusively found in primary hypothyroidism patients’ guts in our study. In a previous research, this species was also found only in thyroid cancer patients [11]. These data suggest that changes in Rheinheimera abundance may be closely related to thyroid disease, but more studies are needed to clarify this association.

SCFAs are important metabolites of bacteria in the human gut. Increasing evidence shows that SCFAs play an important role in health maintenance and disease development. Diseases such as IBD, colorectal cancer, obesity, and MS are associated with decreased levels of SCFAs in the intestine [34–37]. Although there are no data regarding intestinal SCFAs changes in primary hypothyroidism patients, it has been shown that SCFAs, including acetate, propionate and butyrate, may influence T3 functions [38]. SCFAs can regulate the endocrine function in different organs, including the anterior pituitary gland, where they inhibit growth hormone (GH) secretion and enhance T3-induced stimulation of prolactin expression [39,40]. In our study, Veillonella and Paraprevotella, that can produce propionate and butyrate [41,42], were significantly decreased in primary hypothyroidism patients. According to network analysis, the promotion of these two genera to other SCFA-producing bacteria, such as Lachnospira, Roseburia, and Collinsella disappeared in the primary hypothyroidism group and the abundance of Lachnospira and Roseburia was markedly reduced in primary hypothyroidism patients (Supplementary Figure S1C). Further functional predictions also revealed a decrease in propionate and butyrate metabolism in the primary hypothyroidism group. Finally, our qPCR results showed that the ability to produce SCFAs, especially propionic and butyric acid, was reduced in the primary hypothyroidism group. In those patients, SCFAs reduction may be associated with some of the symptoms of T3 deficiency. Research has shown that apart from being a major source of energy for colonocytes, SCFAs perform various physiological functions, including regulation of colonic motility and blood flow, and regulation of gastrointestinal pH, which can influence electrolytes and nutrients uptake and absorption [43]. Decreased SCFAs may explain the gastrointestinal symptoms of primary hypothyroidism. As SCFAs increase the barrier function of the gut, their reduction in primary hypothyroidism patients allows many endotoxins to enter the body, triggering a wide range of symptoms [44,45]. In our study, the mRNA expression of intestinal barrier related molecules such as ZO-1, occludin and JAM-A were significantly decreased in the colon of FMT-H mice, showing that intestinal dysbiosis in primary hypothyroidism could decrease the intestinal barrier in mice. The increased serum LPS levels in primary hypothyroidism patients also point to a breach in the intestinal barrier, probably caused by the decrease in SCFAs. LPS has been shown to inhibit the enzyme iodothyronine deiodinase, thus decreasing the amount of active T3 in the circulation [46,47]. In addition, FMT resulted in an increased LPS and decreased TT4 levels in mice. Based on the above, we speculate that the reduction of SCFAs-producing bacteria in the intestinal tract of primary hypothyroidism patients leads to decreased SCFAs; with an altered intestinal barrier, blood LPS levels increase and induce primary hypothyroidism. Certainly, further experiments are needed to elucidate the mechanism.

Our study not only confirmed that patients with hypothyroidism are prone to intestinal bacterial dysbiosis, but also revealed the specific compositional changes in the intestinal flora. The existing literature indicates a complex interrelationship between gut flora and the host [48]. We previously considered that primary hypothyroidism caused these changes in the flora. However, through the FMT results, we showed that the microflora might actually have a pathogenic role in primary hypothyroidism; this may be a strong proof of its impact on the gut-thyroid axis. In accordance with the results of the study by Virili and Centanni [49], the flora from primary hypothyroidism patients caused TT4 levels to decrease in mice.

There are several limitations of the present study. We did not study the changes in trace elements such as selenium, iron, and zinc, which not only affect thyroid function but also are closely related to the intestinal microflora. Furthermore, we did not address the mechanisms by which primary hypothyroidism caused the alterations in microflora. Finally, although we demonstrated a decreased ability to produce SCFAs in the intestine of primary hypothyroidism patients, we were not able to measure levels of SCFAs in feces. Further mechanistic studies are necessary to understand the pathogenic implications of microbiota composition and function in patients with primary hypothyroidism.

Conclusion

In conclusion, our study confirms that primary hypothyroidism has a significant impact on gut microbiome, and strongly supports the hypothesis of an association between primary hypothyroidism and intestinal bacterial dysbiosis. Our study suggests that differences in gut microbiome may be one of the pathogenic phenomena of primary hypothyroidism. In turn, the altered flora can also affect thyroid function in mice. This finding has important implications for understanding the occurrence and development of primary hypothyroidism. These results might be further used to develop potential probiotics as adjuvants in the treatment of primary hypothyroidism. Finally, further research is needed to clarify the mechanisms of gut-thyroid interaction and to confirm these results in a larger population.

Clinical perspectives

  • The association between intestinal flora and primary hypothyroidism is not known.

  • Gut microbiome in patients with primary hypothyroidism significantly changed and the altered flora can also affect thyroid function in mice.

  • These findings could help understand the development of primary hypothyroidism and might be further used to develop potential probiotics to facilitate the adjuvant treatment of this disease.

Competing Interests

The authors declare that there are no competing interests associated with the manuscript.

Funding

This work was funded from the project ZR2019BH023 supported by Shandong Provincial Natural Science Foundation and the National Natural Science Foundation of China [grant number 81900716].

Author Contribution

Conception and design of experiments: S.Z.M. and Z.W. Experiments conducted by: X.H.S., Y.Z., and Y.L. Data analysis: X.H.S. Manuscript writing: X.H.S. and S.Z.M. All authors approved the final submitted version of the report.

Ethics Approval

The present study was approved by the Ethics Committee of Shandong Provincial Hospital and conformed to the principles of the Declaration of Helsinki (NO 2016-KY-102).

Abbreviations

     
  • BG

    blood glucose

  •  
  • BMI

    body mass index

  •  
  • ButA

    Butyryl-CoA CoA transferase

  •  
  • F/B

    Firmicutes/Bacteroidetes

  •  
  • FMT

    fecal microbiota transplantation

  •  
  • FT3

    free triiodothyronine

  •  
  • FT4

    free thyroxine

  •  
  • HDL

    high-density lipoprotein

  •  
  • IBD

    inflammatory bowel disease

  •  
  • JAM-A

    junctional adhesion molecule-A

  •  
  • LcdA

    lactoyl-CoA dehydratase

  •  
  • LDL

    low-density lipoprotein

  •  
  • LPS

    lipopolysaccharides

  •  
  • MmdA

    methylmalonyl-CoA decarboxylase

  •  
  • OTU

    operational taxonomic unit

  •  
  • PduP

    propionaldehyde dehydrogenase

  •  
  • RF

    random forest

  •  
  • ROC

    receiver operator characteristic

  •  
  • SCFA

    short chain fatty acid

  •  
  • TGAb

    thyroglobulin antibody

  •  
  • TPOAb

    thyroperoxidase antibody

  •  
  • TRAb

    thyrotropin receptor antibody

  •  
  • TSH

    thyroid stimulating hormone

  •  
  • TT4

    total thyroxine

  •  
  • UI

    urinary iodine

  •  
  • ZO-1

    zonula occludens-1

References

References
1.
Chakera
A.J.
,
Pearce
S.H.
and
Vaidya
B.
(
2012
)
Treatment for primary hypothyroidism: current approaches and future possibilities
.
Drug Des. Dev. Ther.
6
,
1
11
[PubMed]
2.
Udovcic
M.
,
Pena
R.H.
,
Patham
B.
,
Tabatabai
L.
and
Kansara
A.
(
2017
)
Hypothyroidism and the heart
.
Methodist. Debakey Cardiovasc. J.
13
,
55
59
[PubMed]
3.
Taylor
P.N.
,
Albrecht
D.
,
Scholz
A.
,
Gutierrez-Buey
G.
,
Lazarus
J.H.
,
Dayan
C.M.
et al.
(
2018
)
Global epidemiology of hyperthyroidism and hypothyroidism
.
Nat. Rev. Endocrinol.
14
,
301
316
[PubMed]
4.
Gilbert
J.A.
,
Blaser
M.J.
,
Caporaso
J.G.
,
Jansson
J.K.
,
Lynch
S.V.
and
Knight
R.
(
2018
)
Current understanding of the human microbiome
.
Nat. Med.
24
,
392
400
[PubMed]
5.
Gill
S.R.
,
Pop
M.
,
Deboy
R.T.
,
Eckburg
P.B.
,
Turnbaugh
P.J.
,
Samuel
B.S.
et al.
(
2006
)
Metagenomic analysis of the human distal gut microbiome
.
Science
312
,
1355
1359
[PubMed]
6.
Dethlefsen
L.
,
McFall-Ngai
M.
and
Relman
D.A.
(
2007
)
An ecological and evolutionary perspective on human-microbe mutualism and disease
.
Nature
449
,
811
818
[PubMed]
7.
Stefanaki
C.
,
Peppa
M.
,
Mastorakos
G.
and
Chrousos
G.P.
(
2017
)
Examining the gut bacteriome, virome, and mycobiome in glucose metabolism disorders: are we on the right track?
Metabolism
73
,
52
66
[PubMed]
8.
Stefanaki
C.
,
Michos
A.
,
Mastorakos
G.
,
Mantzou
A.
,
Landis
G.
,
Zosi
P.
et al.
(
2019
)
Probiotics in adolescent prediabetes: a pilot RCT on glycemic control and intestinal bacteriome
.
J. Clin. Med.
8
,
[PubMed]
9.
Zhao
F.
,
Feng
J.
,
Li
J.
,
Zhao
L.
,
Liu
Y.
,
Chen
H.
et al.
(
2018
)
Alterations of the gut microbiota in Hashimoto's thyroiditis patients
.
Thyroid
28
,
175
186
[PubMed]
10.
Ishaq
H.M.
,
Mohammad
I.S.
,
Shahzad
M.
,
Ma
C.
,
Raza
M.A.
,
Wu
X.
et al.
(
2018
)
Molecular alteration analysis of human gut microbial composition in Graves' disease patients
.
Int. J. Biol. Sci.
14
,
1558
1570
[PubMed]
11.
Zhang
J.
,
Zhang
F.
,
Zhao
C.
,
Xu
Q.
,
Liang
C.
,
Yang
Y.
et al.
(
2019
)
Dysbiosis of the gut microbiome is associated with thyroid cancer and thyroid nodules and correlated with clinical index of thyroid function
.
Endocrine
64
,
564
574
[PubMed]
12.
Frohlich
E.
and
Wahl
R.
(
2019
)
Microbiota and thyroid interaction in health and disease
.
Trends Endocrinol. Metab.
30
,
479
490
[PubMed]
13.
Lerner
A.
,
Jeremias
P.
and
Matthias
T.
(
2017
)
Gut-thyroid axis and celiac disease
.
Endocr. Connect.
6
,
R52
R58
[PubMed]
14.
Allayee
H.
and
Hazen
S.L.
(
2015
)
Contribution of gut bacteria to lipid levels: another metabolic role for microbes?
Circ. Res.
117
,
750
754
[PubMed]
15.
Wang
X.
,
Zhang
L.
,
Wang
Y.
,
Liu
X.
,
Zhang
H.
,
Liu
Y.
et al.
(
2018
)
Gut microbiota dysbiosis is associated with Henoch-Schonlein Purpura in children
.
Int. Immunopharmacol.
58
,
1
8
[PubMed]
16.
Carvalho
F.A.
,
Koren
O.
,
Goodrich
J.K.
,
Johansson
M.E.
,
Nalbantoglu
I.
,
Aitken
J.D.
et al.
(
2012
)
Transient inability to manage proteobacteria promotes chronic gut inflammation in TLR5-deficient mice
.
Cell Host Microbe
12
,
139
152
[PubMed]
17.
Magoc
T.
and
Salzberg
S.L.
(
2011
)
FLASH: fast length adjustment of short reads to improve genome assemblies
.
Bioinformatics
27
,
2957
2963
[PubMed]
18.
Parks
D.H.
,
Tyson
G.W.
,
Hugenholtz
P.
and
Beiko
R.G.
(
2014
)
STAMP: statistical analysis of taxonomic and functional profiles
.
Bioinformatics
30
,
3123
3124
[PubMed]
19.
Langille
M.G.
,
Zaneveld
J.
,
Caporaso
J.G.
,
McDonald
D.
,
Knights
D.
,
Reyes
J.A.
et al.
(
2013
)
Predictive functional profiling of microbial communities using 16S rRNA marker gene sequences
.
Nat. Biotechnol.
31
,
814
821
[PubMed]
20.
Ren
Z.
,
Li
A.
,
Jiang
J.
,
Zhou
L.
,
Yu
Z.
,
Lu
H.
et al.
(
2019
)
Gut microbiome analysis as a tool towards targeted non-invasive biomarkers for early hepatocellular carcinoma
.
Gut
68
,
1014
1023
[PubMed]
21.
Thackray
L.B.
,
Handley
S.A.
,
Gorman
M.J.
,
Poddar
S.
,
Bagadia
P.
,
Briseno
C.G.
et al.
(
2018
)
Oral antibiotic treatment of mice exacerbates the disease severity of multiple flavivirus infections
.
Cell Rep.
22
,
3440e6
3453e6
22.
Steed
A.L.
,
Christophi
G.P.
,
Kaiko
G.E.
,
Sun
L.
,
Goodwin
V.M.
,
Jain
U.
et al.
(
2017
)
The microbial metabolite desaminotyrosine protects from influenza through type I interferon
.
Science
357
,
498
502
23.
Salyers
A.A.
(
1984
)
Bacteroides of the human lower intestinal tract
.
Annu. Rev. Microbiol.
38
,
293
313
[PubMed]
24.
Wexler
H.M.
(
2007
)
Bacteroides: the good, the bad, and the nitty-gritty
.
Clin. Microbiol. Rev.
20
,
593
621
[PubMed]
25.
Lauritano
E.C.
,
Bilotta
A.L.
,
Gabrielli
M.
,
Scarpellini
E.
,
Lupascu
A.
,
Laginestra
A.
et al.
(
2007
)
Association between hypothyroidism and small intestinal bacterial overgrowth
.
J. Clin. Endocrinol. Metab.
92
,
4180
4184
[PubMed]
26.
Almandoz
J.P.
and
Gharib
H.
(
2012
)
Hypothyroidism: etiology, diagnosis, and management
.
Med. Clin. North Am.
96
,
203
221
[PubMed]
27.
Woting
A.
and
Blaut
M.
(
2016
)
The intestinal microbiota in metabolic disease
.
Nutrients
8
,
202
[PubMed]
28.
Le Chatelier
E.
,
Nielsen
T.
,
Qin
J.
,
Prifti
E.
,
Hildebrand
F.
,
Falony
G.
et al.
(
2013
)
Richness of human gut microbiome correlates with metabolic markers
.
Nature
500
,
541
546
[PubMed]
29.
Mariat
D.
,
Firmesse
O.
,
Levenez
F.
,
Guimaraes
V.
,
Sokol
H.
,
Dore
J.
et al.
(
2009
)
The Firmicutes/Bacteroidetes ratio of the human microbiota changes with age
.
BMC Microbiol.
9
,
123
[PubMed]
30.
Ley
R.E.
,
Turnbaugh
P.J.
,
Klein
S.
and
Gordon
J.I.
(
2006
)
Microbial ecology: human gut microbes associated with obesity
.
Nature
444
,
1022
1023
[PubMed]
31.
Furet
J.P.
,
Kong
L.C.
,
Tap
J.
,
Poitou
C.
,
Basdevant
A.
,
Bouillot
J.L.
et al.
(
2010
)
Differential adaptation of human gut microbiota to bariatric surgery-induced weight loss: links with metabolic and low-grade inflammation markers
.
Diabetes
59
,
3049
3057
[PubMed]
32.
Louis
S.
,
Tappu
R.M.
,
Damms-Machado
A.
,
Huson
D.H.
and
Bischoff
S.C.
(
2016
)
Characterization of the gut microbial community of obese patients following a weight-loss intervention using whole metagenome shotgun sequencing
.
PLoS One
11
,
e0149564
[PubMed]
33.
Chen
Y.
,
Sjolinder
M.
,
Wang
X.
,
Altenbacher
G.
,
Hagner
M.
,
Berglund
P.
et al.
(
2012
)
Thyroid hormone enhances nitric oxide-mediated bacterial clearance and promotes survival after meningococcal infection
.
PLoS One
7
,
e41445
[PubMed]
34.
Wang
G.
,
Yu
Y.
,
Wang
Y.Z.
,
Wang
J.J.
,
Guan
R.
,
Sun
Y.
et al.
(
2019
)
Role of SCFAs in gut microbiome and glycolysis for colorectal cancer therapy
.
J. Cell. Physiol.
234
,
17023
17049
[PubMed]
35.
Parada Venegas
D.
,
De la Fuente
M.K.
,
Landskron
G.
,
Gonzalez
M.J.
,
Quera
R.
,
Dijkstra
G.
et al.
(
2019
)
Short chain fatty acids (SCFAs)-mediated gut epithelial and immune regulation and its relevance for inflammatory bowel diseases
.
Front. Immunol.
10
,
277
[PubMed]
36.
den Besten
G.
,
Bleeker
A.
,
Gerding
A.
,
van Eunen
K.
,
Havinga
R.
,
van Dijk
T.H.
et al.
(
2015
)
Short-chain fatty acids protect against high-fat diet-induced obesity via a PPARgamma-dependent switch from lipogenesis to fat oxidation
.
Diabetes
64
,
2398
2408
[PubMed]
37.
Zeng
Q.
,
Junli
G.
,
Liu
X.
,
Chen
C.
,
Sun
X.
,
Li
H.
et al.
(
2019
)
Gut dysbiosis and lack of short chain fatty acids in a Chinese cohort of patients with multiple sclerosis
.
Neurochem. Int.
129
,
104468
[PubMed]
38.
Ejtahed
H.S.
,
Angorani
P.
,
Soroush
A.R.
,
Siadat
S.D.
,
Shirzad
N.
,
Hasani-Ranjbar
S.
et al.
(
2019
)
Our little friends with big roles: alterations of the gut microbiota in thyroid disorders
.
Endocr. Metab. Immune Disord. Drug Targets
20
,
39.
Ishiwata
H.
,
Katoh
K.
,
Chen
C.
,
Yonezawa
T.
,
Sasaki
Y.
and
Obara
Y.
(
2005
)
Suppressing actions of butyrate on growth hormone (GH) secretion induced by GH-releasing hormone in rat anterior pituitary cells
.
Gen. Comp. Endocrinol.
143
,
222
230
[PubMed]
40.
Stanley
F.
and
Samuels
H.H.
(
1984
)
n-Butyrate effects thyroid hormone stimulation of prolactin production and mRNA levels in GH1 cells
.
J. Biol. Chem.
259
,
9768
9775
[PubMed]
41.
Polansky
O.
,
Sekelova
Z.
,
Faldynova
M.
,
Sebkova
A.
,
Sisak
F.
and
Rychlik
I.
(
2015
)
Important metabolic pathways and biological processes expressed by chicken cecal microbiota
.
Appl. Environ. Microbiol.
82
,
1569
1576
[PubMed]
42.
Ríos-Covián
D.
,
Ruas-Madiedo
P.
,
Margolles
A.
,
Gueimonde
M.
,
de Los Reyes-Gavilán
C.G.
and
Salazar
N.
(
2016
)
Intestinal short chain fatty acids and their link with diet and human health
.
Front. Microbiol.
7
,
185
[PubMed]
43.
Tazoe
H.
,
Otomo
Y.
,
Kaji
I.
,
Tanaka
R.
,
Karaki
S.I.
and
Kuwahara
A.
(
2008
)
Roles of short-chain fatty acids receptors, GPR41 and GPR43 on colonic functions
.
J. Physiol. Pharmacol.
59
,
251
262
[PubMed]
44.
Kelly
C.J.
,
Zheng
L.
,
Campbell
E.L.
,
Saeedi
B.
,
Scholz
C.C.
,
Bayless
A.J.
et al.
(
2015
)
Crosstalk between microbiota-derived short-chain fatty acids and intestinal epithelial HIF augments tissue barrier function
.
Cell Host Microbe
17
,
662
671
[PubMed]
45.
Tulkens
J.
,
Vergauwen
G.
,
Van Deun
J.
,
Geeurickx
E.
,
Dhondt
B.
,
Lippens
L.
et al.
(
2020
)
Increased levels of systemic LPS-positive bacterial extracellular vesicles in patients with intestinal barrier dysfunction
.
Gut
69
,
191
193
[PubMed]
46.
Yu
J.
and
Koenig
R.J.
(
2000
)
Regulation of hepatocyte thyroxine 5′-deiodinase by T3 and nuclear receptor coactivators as a model of the sick euthyroid syndrome
.
J. Biol. Chem.
275
,
38296
38301
[PubMed]
47.
Kahl
S.
,
Elsasser
T.H.
and
Blum
J.W.
(
2000
)
Effect of endotoxin challenge on hepatic 5′-deiodinase activity in cattle
.
Domest. Anim. Endocrinol.
18
,
133
143
[PubMed]
48.
Thursby
E.
and
Juge
N.
(
2017
)
Introduction to the human gut microbiota
.
Biochem. J.
474
,
1823
1836
[PubMed]
49.
Virili
C.
and
Centanni
M.
(
2015
)
Does microbiota composition affect thyroid homeostasis?
Endocrine
49
,
583
587
[PubMed]