1、Vol.:(0123456789)1 3Journal of Endocrinological Investigation https:/doi.org/10.1007/s40618-019-1010-9ORIGINAL ARTICLEAlterations intheintestinal microbiota ofpatients withsevere andactive Graves orbitopathy: acrosssectional studyT.T.Shi1 Z.Xin1 L.Hua2 R.X.Zhao1 Y.L.Yang1 H.Wang3 S.Zhang4 W.Liu1 R.R
2、.Xie1Received: 11 August 2018 / Accepted: 14 January 2019 Italian Society of Endocrinology (SIE) 2019AbstractBackground The intestinal microbiota was linked to autoimmune diseases. Graves orbitopathy (GO) is an autoimmune dis-ease that is usually associated with Graves disease. However, information
3、on the microbiome of GO patients is yet lacking.Objectives To investigate whether GO patients differ from healthy controls in the fecal microbiota.Design A cross-sectional study.Setting 33 patients with severe and active GO and 32 healthy controls of Han nationality were enrolled between March 2017
4、and March 2018.Methods The Gut microbial communities of the fecal samples of GO patients and healthy controls were analyzed and com-pared by 16S rRNA gene sequencing.Results Community diversity (Simpson and Shannon) was significantly reduced in fecal samples from patients with GO as compared to cont
5、rols (p 0.05). The similarity observed while assessing the community diversity (PCoA) proposed that the microbiota of patients with GO differ significantly from those of controls (p 0.05). At the phyla levels, the proportion of Bacteroidetes increased significantly in patients with GO (p 0.05), whil
6、e at the genus and species levels, significant differences were observed in the bacterial profiles between the two groups (p 0.05).Limitations Single-centered study design and limited fecal samples.Conclusions The present study indicated distinctive features of the gut microbiota in GO patients. The
7、 study provided evi-dence for further exploration in the field of intestinal microbiota with respect to the diagnosis and treatment of GO patients by modifying the microbiota profile.Keywords Graves orbitopathy (GO) Gut microbiota 16S rRNA gene Thyrotropin receptor antibody (TRAb)IntroductionGraves
8、orbitopathy (GO) is an autoimmune disease, usually associated with Graves disease, which damages the appear-ance and visual acuity and decreases the quality of life sig-nificantly 13. About 30% of the patients with Graves disease (GD) suffer from mild to severe, unilateral or bilat-eral GO during th
9、e disease 4. The clinical course of GO involves two stages 5. The active phase is characterized by orbital inflammation and tissue remodeling, following which the condition stabilizes and eventually trends into an inactive phase. However, the pathogenesis underlying GO is not yet completely understo
10、od. Several factors have been demon-strated to be involved in the development of GO, including genetic susceptibility and multiple environmental factors, for example, gender, age, and smoking 6.Electronic supplementary material The online version of this article (https :/doi.org/10.1007/s4061 8-019-
11、1010-9) contains supplementary material, which is available to authorized users. * Z. Xin * L. Hua 1 Department ofEndocrinology, Beijing Tongren Hospital, Capital Medical University, 1 Dong Jiao Min Xiang, Beijing100730, China2 Department ofMathematics, School ofBiomedical Engineering, Capital Medic
12、al University, Beijing100069, China3 Department ofEmergency, Beijing Tongren Hospital, Capital Medical University, Beijing, China4 Department ofPharmacy, Beijing Tongren Hospital, Capital Medical University, Beijing, China Journal of Endocrinological Investigation1 3Data from human and animal studie
13、s demonstrated that the gut microbiota influences various autoimmune diseases, including type 1 diabetes (T1D) 7, systemic lupus erythe-matosus (SLE) 8, inflammatory bowel disease (IBD) 9, and rheumatoid arthritis (RA) 10. Furthermore, some stud-ies showed the correlation between the autoimmune thyr
14、oid diseases (AITD), especially Hashimoto thyroiditis (HT) and GD, and the gastrointestinal tract 11. Kahaly etal. reported the positive association between GD and celiac disease, and a positive association of celiac disease and ocular involve-ment 12. In addition, HT patients demonstrated altered g
15、ut microbiota, which was related to clinical parameters 13. A significant difference was noted in the gut micro-biota between hyperthyroid and healthy groups 14. Recent studies also demonstrated that in experimental murine GO model established in different environments, the gut micro-biota modulates
16、 the clinical development and presentation of the disease 15. However, to date, there are only a few clinical studies exploring the role of gut microbiota in GO patients, while none of the studies have compared the bacte-rial communities between GO patients and healthy controls.In the current study,
17、 the gut microbiota-related changes were characterized by 16S rRNA-based microbial profil-ing among patients with severe and active GO, as well as matched healthy volunteers. The primary objectives of the present study are to provide evidence for researching the field of intestinal microbiota in GO
18、patients and open novel avenues for the diagnosis and treatment of patients based on modifying the microbiota profile.Subjects andmethodsStudy design andsettingIn this cross-sectional study, 16S rRNA gene sequencing was used to analyze and compare the gut microbial com-munities of the fecal samples
19、of GO patients and healthy controls. The present study was performed at the Depart-ment of Endocrinology, Beijing Tongren Hospital, Capital Medical University, Beijing, China. According to our previ-ous data for -diversity indices (Simpson), the difference in means of the two groups is 0.03 and the
20、standard deviation is 0.02 and 0.05, respectively; a total of 28 casecontrol sets were need (1:1) with the 90% power (1-) and 5% signifi-cant level (). Between March 2017 and March 2018, 33 patients with severe and active GO and 32 healthy volunteers were enrolled for the comparison of gut microbiot
21、a in fecal samples.The Ethics Committee of Beijing Tongren Hospi-tal, Capital Medical University, approved this study (no.: TRECKY2016-003). Written informed consent was obtained from all patients and healthy volunteers prior to the collec-tion of clinical data and samples.SubjectsAll the enrolled 3
22、3 patients with severe and active GO and 32 healthy volunteers were Han Chinese. Among GO patients, 31 cases were hyperthyroid, one patient was euthyroid, and one patient was hypothyroid. All patients with hyperthy-roidism received only antithyroid drugs (Thyrozol, Merck Company). The diagnosis of G
23、O was established according to the EUGOGO Guidelines 3. These enrolled GO patients had not received any treatment for ocular discomfort. The activity and severity of GO were established according to the EUGOGO Guidelines 3, 16. The active GO was defined by a clinical activity score (CAS) 3/7, and th
24、e severe GO was defined by NOSPECS score IV. CT or MRI was used to exclude any orbital space-occupying disease such as tumor or extraocular myositis. Furthermore, no significant differ-ences were observed in age, gender, and BMI. The charac-teristics of the study cohort are summarized in Table1.The
25、exclusion criteria for the study were age 65years, therapy of probiotics or antibiotics over the previous 4weeks, use of hormonal medication or Chinese herbal medicine ( 3months prior), history of chronic diar-rhea or constipation such as inflammatory bowel disease (IBD), severe disease (acute infec
26、tions, diabetes, stroke, heart disease, renal or hepatic dysfunction, cancer, or auto-immune disease), history of gastrointestinal tract surgery, pure vegetarian, pregnancy and lactation, and alcohol or substance addiction 17.Clinical characteristicsThe baseline characteristics of all subjects, incl
27、uding age, gender, BMI calculated as weight (kg)/height (m2), ethnic-ity, current treatment, past medical history, drug history, and dietary habits, were recorded.To evaluate the thyroid function and examine the thyroid autoantibody, peripheral blood was collected from the sub-jects after an overnig
28、ht fast. The levels of serum FT3, FT4, and thyroid stimulating hormone (TSH) were assayed using ADVIA Centaur XP Immunoassay System (Siemens Diag-nostics), and that of thyroperoxidase antibody (TPOAb), antithyroglobulin antibody (TgAb), thyrotropin receptor antibody (TRAb) were evaluated by electroc
29、hemilumines-cence immunoassay using Cobas e601 (Roche Diagnostics). The quality control was assessed before, during, and after examination. The reference ranges (3.506.50pmol/L for FT3, 11.5022.70pmol/L for FT4, 0.554.78IU/mL for TSH, 0.0040.00IU/mL for TPOAb, 0.00115.00IU/mLfor TgAb, and 0.001.75IU
30、/L for TRAb) were provided based Journal of Endocrinological Investigation 1 3on the recommendations by the manufacturer. The missing data for each variable were no more than 20%.Stool sample collectionApproximately, 2.5g of a fresh fecal sample (from mid-stream stool) was collected from each partic
31、ipant collected using a plastic tube pre-filled with stool DNA stabilizer and including a measuring spoon for sample collection (PSP Spin Stool DNA Plus Kit, STRATEC Molecular, Berlin, Germany). The stool samples were preserved at 80C within 1day and stored until DNA extraction.Microbiota sequencing
32、Bacterial DNA from fecal samples was extracted accord-ing to the manufacturers instructions (PSP Spin Stool DNA Plus Kit, STRATEC Molecular). DNA concentrations were assayed using a NanoDrop 2000 BioAnalyzer at 260nm (Thermo Fisher Scientific Inc., Waltham, MA, USA). 16S rRNA genes of distinct regio
33、ns (16S V4/16S, 515F: GTG CCA GCMGCC GCG GTAA; 806R: GGA CTA CHVGGG TWT CTAAT) were amplified using specific primers 17. The PCR reaction consisted of Phusion High-Fidelity PCR Master Mix (New England Biolabs) and SYBR green. The electrophoresis was carried out on 2% agarose gel, and the resulting P
34、CR resolved products were purified from the gel using GeneJET Gel Extraction Kit (Thermo Scientific). The sequencing libraries were built using Ion Plus Frag-ment Library Kit 48 reactions (Thermo Scientific) following manufacturers instructions (Novogene, Beijing, China). The quality of the sequenci
35、ng libraries was assessed and ana-lyzed on the Qubit 2.0 Fluorometer (Thermo Scientific). Lastly, the library was sequenced on an Ion S5 XL plat-form, and 400bp/600bp single-end reads were generated.Singleend reads quality controlSingle-end reads were assigned to each sample based on the unique barc
36、ode and abbreviated by cutting off the barcode and the primer sequence. Then, the raw reads were subjected to obtain clean and qualified reads based on the Cutadapt (V1.9.1) quality control process under specific filtering con-ditions. The chimera sequences were detected and removed after aligning t
37、he reads with the reference database using the UCHIME algorithm.OTU cluster andspecies annotationThe sequences were analyzed using Uparse software (Uparse v7.0.1001) and those with 97% similarity were classified into the same OTUs. The sequence representative for each OTU was screened out for furthe
38、r annotation. Next, the Silva Table 1 Clinical characteristics of case and control subjectsValues are means (SD), median (range), or n (%)Comparison between any two groups by MannWhitney U test*p 0.05SBP systolic blood pressure, DBP diastolic blood pressure, CAS clinical activity score, TSH thyroid
39、stimulating hormone, TPOAb thyroperoxidase antibody, TGAb antithyroglobulin antibody, TRAb thyrotrophin receptor antibodySex(M/F)Age(year)Height(cm)Weight(cm)GO dura-tion(months)CASNOSPECSFT3(pmol/L)FT4(pmol/L)T3(pmol/L)T4(pmol/L)TSH(uIU/mL)TPOAb(IU/mL)TGAb(IU/mL)TRAb(IU/L)Case (33)17/1646.0 11.7116
40、7.9 9.366.9 10.77.4 4.13.3 0.5IV (30)V (3)5.32(4.77, 6.64)16.41(12.82, 18.74)1.91*(1.74, 2.31)106.00(86.50, 123.55)0.42*(0.06, 2.16)12.43(8.06, 125.15)14.63*(10.26, 75.36)4.50*(1.699.64)Control(32)16/1643.4 9.7166.8 8.166.4 12.75.01(4.39, 5.33)15.12(11.55, 17.20)1.42(1.24, 1.77)93.45(90.35, 105.90)1
41、.60(1.17, 2.32)10.98(1.08, 17.23)11.76(0.20, 15.46) 1%, p values 0.05, and Q values 99.5% for the two groups, suggesting a satisfactory sequencing depth of the fecal microbiota. The rank-abundance curves of the two groups exhibited similar patterns based on the OTU results (Fig.1a). The rarefaction
42、curve analysis dis-played similar OTU number by the intestinal microbiota in the two groups (Fig.1b). The Simpson and Shannon diversi-ties were significantly reduced in samples from patients with GO as compared to healthy controls (both p 0.05, Fig.1c, d); however, no significant difference was dete
43、cted in Chao and Ace indexes. The Venn diagram showing the overlap-ping OTU data of GO and controls was analyzed and dem-onstrated to evaluate the shared community abundance. The diagram revealed that 1344/1575 OTUs were common to all fecal samples. Furthermore, 120 and 111 OTUs were iden-tified in
44、the fecal samples of patients with GO and healthy volunteers, respectively (Fig. S1).Journal of Endocrinological Investigation 1 3-diversity values were used to evaluate the difference or similarity in communities between fecal samples. The PCoA based on the weighted UniFrac method was used to revea
45、l the similarity of microbiome community structures. The results demonstrated that the microbiota of patients with GO differed significantly from those of the healthy controls. The PCoA results demonstrated that the fecal microbiota of the patients with GO and healthy controls were significantly sep
46、arated despite the inter-individual variation (Fig.2a). The analysis of group similarities (ANOSIM) demonstrated that the differences between patients with GO and healthy con-trols were significant (p 0.05, Fig.2b). These findings sug-gested that the grouping design was reliable.Taken together, the
47、bacterial community diversity (Simp-son and Shannon)was significantly reduced in fecal samples from patients with GO as compared to the controls. The PCoA results demonstrated that the microbiota of patients with GO differed significantly from those of controls.Fecal bacterial composition inpatients
48、 withGOTo explore the specific bacterial taxa associated with GO, we analyzed the compositions of the intestinal microbiota of GO patients and controls at different levels in this study. At the phylum level, Bacteroidetes, Firmicutes, Proteobacteria, and Actinobacteria were the predominant phyla in
49、patients Fig. 1 lpha diversity is decreased in GO patients as compared to healthy controls. The uniformity of bacterial species was analyzed by rank abundance curves. The bacterial communities of the GO patients and healthy controls exhibited similar patterns (a). The abundance of bacterial species was analyzed by rarefaction curves. The abundance of microbiota of the GO patients was lower than healthy controls (b). lpha diversity indexes were analyzed in gut microbiota from GO patients and healthy controls. Simpson index (c), Shannon index (d). Data represent the means with standard error. *
