DOI: https://doi.org/10.21203/rs.3.rs-2707130/v1
The gut microbiota has been identified as a source of pathogenic mediators in chronic kidney disease (CKD). A gut microbiota-dependent metabolite, trimethylamine N-oxide (TMAO), has been reported to be closely related to CKD complications. This study aimed to investigate the changes in intestinal microbiota and circulating levels of TMAO in Chinese patients with CKD.
The study comprised 50 paticipants including 30 patients with CKD and a control group of 20 healthy controls. Plasma TMAO levels were detected by high-performance liquid chromatography-tandem mass spectrometry, and gut microbiota was analysed using High-throughput sequencing.
Compared to the healthy control group, the CKD patients had relatively lower albumin and hemoglobin levels and showed obviously impaired renal function and abnormal urine test results. Additionally, CKD patients showed increased plasma TMAO levels, especially those with a low glomerular filtration rate (GFR). Among the biochemical indices of the CKD patients, impaired renal function was the main contributor of the increased TMAO levels. High-throughput sequencing revealed obvious gut dysbiosis in CKD patients with biased community constitutions. Based on the Pearson correlation analysis, many bacteria positively or negatively correlated with TMAO production at the phylum and genus levels.
Our study demonstrated that gut microbiota dysbiosis and decreased GFR were the main causes of plasma TMAO level. Elevation, and inhibition of intestinal metabolite TMAO production may be the key to preventing CKD progression.
The prevalence of chronic kidney disease (CKD) is high in both developing and developed countries. A cross-sectional survey of a nationally representative sample of Chinese adults found that the overall prevalence of CKD was 10,8%[1]. Data showed that the global all-age prevalence of CKD continues to increase, and early intervention of kidney disease has a major effect on global health, both as a direct cause of global morbidity and mortality and as an important risk factor for cardiovascular disease. CKD should play a greater role in global health policy decision making [2]. Over recent years, the relationship between CKD and gut microbiota has gained growing interest. It is well-known that renal injury can disrupt the enteric microbial composition. Conversely, gut microbial composition and function directly influences the progression of renal disease[3–5]. Understanding the complex between these two components (gut microbiota and the kidneys) may provide novel nephroprotective interventions to prevent CKD progression by targeting the gut microbiota. However, the correlation between CKD and gut microbiota is unclear. A gut microbiota-dependent metabolite, trimethylamine N-oxide (TMAO) has recently gained attention because of its potential role in mediating various diseases, such as cardiovascular diseases, cardiorenal disorders, diabetes mellitus, metabolic syndrome, cancers (namely, stomach and colon), and neurological disorders[6, 7]. Circulating TMAO levels are elevated in patients with CKD and are associated with increased cardiovascular events[8]. Inhibition of TMAO production attenuates CKD development and cardiac hypertrophy in mice[9]. In our study, we aimed to investigate the relationship between intestinal microbiota and circulating levels of TMAO and investigate the role of TMAO in Chinese patients with CKD.
The study included 30 CKD patients and 20 healthy controls without CKD, designated the control group, from the Zhejiang Provincial People’s Hospital from August 2019 to March 2020. Patients with proven abnormal kidney structure or function lasting more than three months were included in this study. Patients were excluded from the study if they (1) had acute intercurrent illness and infections and (2) used antibiotics or probiotics within 1 month before admission. The median age was 50 years, and the male-to-female ratio was 1.3:1. All participants in the CKD group had CKD stage 1–4, none were diagnosed with end-stage renal disease.
The following biochemical levels were measured by the hospital central laboratory. Blood urea nitrogen (BUN), creatinine, albumin, uric acid, fasting glucose, calcium, phosphate, total cholesterol, triglyceride, alanine transaminase, homocysteine, alkaline phosphatase, hemoglobin, urine total protein-to-creatinine ratio, urine occult blood, and 24-hour urine protein.
Plasma TMAO levels were detected by high-performance liquid chromatography-tandem mass spectrometry(HPLC-MS). The HPLC-MS system used was an LC-30AT HPLC system(Shimadzu Corp.) equipped with an ACQUITY UPLC HILIC column (length 100 mm; internal diameter 2.1 mm; particle size 1.7 µm; Waters Corporation). The analytes were separated using gradient elution at an HPLC-MS/MS solvent flow rate of 300 µL/min. The solvents were 0.1% aqueous formic acid (mobile phase A) and acetonitrile (mobile phase B), the temperature was 35°C, and the injection volume was 10 µL. Identification and quantification of TMAO were performed using a TripleTOF 5600 + mass spectrometer (SCIEX Corp.) in electrospray ionization (ESI) positive ion mode. The mass spectrometry parameters were as follows: ion source gas one, 50 psi; ion source gas, two 50 psi; source temperature, 500°C; curtain gas, 25 psi; the declustering potential was set at 60 V; collision energy (CE), 35 V. Quantitation was performed using multiple reaction monitoring of the transitions of m/z 76.0657–76.0857 for TMAO. All reagents used in this study were of analytical grade unless otherwise specified.
Fecal samples (50–200 mg) were collected and frozen at -80°C. Total DNA was extracted from fecal samples, then genomic DNA integrity and quality were detected and measured using agarose gel electrophoresis and NanoDrop 2000, respectively. The amplification primers used were based on the target regions. Sequencing adaptors were attached to the ends of the primers for polymerase chain reaction (PCR) amplification. The PCR products were purified, quantified, and homogenized to form a sequencing library that was subsequently sequenced using the MiSeq platform and a 2×250 bp double-ended sequencing strategy, followed by bioinformatics analysis. To obtain high-quality sequencing data and improve the accuracy of subsequent bioinformatics analysis, the DADA2 plug-in of the QIIME2 software was used to filter, denoise, merge, and de-chimer data.
All values were expressed as means ± standard errors, semi-quantitative results of proteinuria and hematuria were converted into numerical values, negative, weak positive, +, ++, +++, ++++ convert to 0, 0.5, 1, 2, 3, 4 respectively. Statistical analyses were performed using SPSS (version 22.0; IBM Corp). Values were compared using The Mann–Whitney U test and one-way analysis of variance. The relationships between TMAO and gut microbiome taxa were assessed using Pearson correlation. Differences were considered statistically significant at P < 0.05.
Basic clinical characteristics were collected for analysis, and the data are presented in Table 1. The CKD patients presented with impaired renal function and abnormal urine test results, including significantly elevated BUN, hematuria, and proteinuria. The CKD patients also had relatively lower levels of albumin and hemoglobin compared to the control group. Based on the GFR values, we divided the CKD patients into two groups, namely a high GFR and a low GFR group. The high GFR group had 19 patients with GFR values ≥ 60 mL/min/1.73 m2, which represented a relatively low level of renal impairment, whereas the low GFR group with GFR values < 60 mL/min/1.73 m2) had 11 patients. As demonstrated in Table 2, the low GFR group had relatively higher levels of fasting glucose, urinary protein creatinine ratio and relatively lower levels of albumin and hemoglobin compared to the patients in the high GFR group.
| characteristics | CON | CKD | p -value |
|---|---|---|---|
| numbers | 20 | 30 | Not available |
| albumin(g/L) | 43.75(2.54) | 33.23(7.61) | <0.001* |
| Creatinine(µmol/L) | 77.91(15.25) | 114.38(72.83) | 0.107 |
| BUN(mmol/L) | 4.75(1.18) | 7.28(3.60) | 0.002* |
| uric acid(µmol/L) | 371.75(116.19) | 372.37(97.44) | 0.913 |
| fast glucose(mmol/L ) | 5.18(1.45) | 6.43(3.72) | 0.010* |
| TC(mmol/L ) | 5.50(1.13) | 5.33(2.08) | 0.285 |
| TG(mmol/L ) | 1.53(1.04) | 1.62(0.65) | 0.172 |
| ALT(U/L ) | 30.90(30.76) | 23.30(18.57) | 0.422 |
| ALP (U/L ) | 75.80(21.84) | 81.76(32.13) | 0.654 |
| hemoglobin(g/L) | 146.85(15.96) | 127.27(24.35) | 0.003* |
| Semi-quantification of proteinuria | 0.00(0.00) | 1.68(1.12) | <0.001* |
| Semi-quantification of hematuria | 0.00(0.00) | 1.13(0.96) | <0.001* |
| characteristics | High GFR group | Low GFR group | p -value | |
|---|---|---|---|---|
| numbers | 19 | 11 | Not available | |
| albumin(g/L) | 36.00(5.86) | 28.46(8.17) | 0.014* | |
| BUN( mmol/L) | 5.40(1.47) | 10.55(3.90) | <0.001* | |
| Creatinine(µmol/L) | 74.51(16.72) | 183.22(81.51) | <0.001* | |
| uric acid(µmol/L) | 345.58(80.61) | 418.64(110.10) | 0.077 | |
| fast glucose(mmol/L ) | 5.35(0.93) | 8.31(5.70) | 0.018* | |
| TC(mmol/L ) | 5.63(2.33) | 4.83(1.53) | 0.307 | |
| TG(mmol/L ) | 1.61(0.59) | 1.65(0.78) | 0.866 | |
| ALT(U/L ) | 21.58(10.95) | 26.27(27.78) | 0.767 | |
| ALP (U/L ) | 79.00(30.84) | 86.27(35.19) | 0.674 | |
| calcium(mmol/L ) | 2.22(0.13) | 2.08(0.18) | 0.055 | |
| phosphate(mmol/L ) | 1.16(0.15) | 1.21(0.27) | 0.774 | |
| Hcy(µmol/L) | 13.62(3.73) | 15.40(3.69) | 0.306 | |
| hemoglobin(g/L) | 135.16(18.24) | 113.64(28.25) | 0.037* | |
| Semi-quantification of proteinuria | 1.53(1.33) | 1.96(0.57) | 0.232 | |
| Semi-quantification of hematuria | 1.00(1.09) | 1.36(0.64) | 0.171 | |
| UPCR(g/g) | 1.11(1.52) | 3.70(2.11) | 0.001* | |
| 24-hour urine protein (mg/24h) | 1940.66(3074.90) | 3651.01(3035.08) | 0.050 | |
In our study, the plasma TMAO levels of the CKD patients were 15.4-fold higher than the control group (9.22 ug/mL vs. 141.91 ug/mL, p < 0.0001, Fig. 1A). In the CKD subgroup, patients in the low GFR group had higher plasma levels of TMAO than those in the high GFR group (Fig. 1B). We also divided the CKD patients into two subgroups according to their plasma TMAO levels. The high TMAO level group had 15 patients with TMAO levels higher than the median level of the CKD group, whereas the low level group with TMAO levels lower than or equal to the median level also had 15 patients. As demonstrated in Table 3, patients in the high TMAO group had relatively higher levels of BUN, creatinine, and urine occult blood than those in the low TMAO group.
| characteristics | Low TMAO group | High TMAO group | p -value |
|---|---|---|---|
| numbers | 15 | 15 | Not available |
| albumin(g/L) | 36.25(4.85) | 30.21(8.77) | 0.089 |
| BUN( mmol/L) | 5.77(1.90) | 8.80(4.28) | 0.033* |
| Creatinine(µmol/L) | 81.50(26.34) | 147.23(89.32) | 0.016* |
| uric acid(µmol/L) | 345.20(91.71) | 399.53(98.37) | 0.305 |
| fast glucose(mmol/L ) | 5.42(1.01) | 7.45(5.04) | 0.148 |
| calcium(mmol/L ) | 2.24(0.10) | 2.11(0.19) | 0.093 |
| phosphate(mmol/L ) | 1.14(0.20) | 1.22(0.21) | 0.270 |
| TC(mmol/L ) | 5.27(1.82) | 5.40(2.37) | 0.967 |
| TG(mmol/L ) | 1.64(0.70) | 1.61(0.63) | 0.870 |
| ALT(U/L ) | 24.07(22.67) | 22.53(14.18) | 0.935 |
| Hcy(µmol/L) | 13.57(3.66) | 14.98(3.83) | 0.363 |
| ALP (U/L ) | 77.79(29.37) | 85.47(35.12) | 0.683 |
| hemoglobin(g/L) | 133.73(18.09) | 120.80(28.49) | 0.187 |
| UPCR(g/g) | 1.23(1.31) | 3.04(2.49) | 0.054 |
| Semi-quantification of hematuria | 0.80(1.00) | 1.47(0.81) | 0.033* |
| Semi-quantification of proteinuria | 1.53(1.17) | 1.83(1.08) | 0.461 |
| 24-hour urine protein (mg/24h) | 1822.50(2225.14) | 3394.22(3689.31) | 0.350 |
We further analyzed the composition of the gut microbiota. Microbiome diversity is typically defined in terms of within (i.e., α-diversity) and between community/sample (i.e., β-diversity). We determined α-diversity as the mean species richness, quantified by the Chao1 and Shannon indices, and found no significant difference between the control and CKD groups (Fig. 2A, 2B). β-diversity was measured by calculating the unweighted UniFrac distances between each pair of samples. The unweighted UniFrac distance matrix was measured and visualized using partial least squares discriminant analysis (PLS-DA). The score plots of the PLS-DA analysis showed that the two groups were well separated (Fig. 2C). This was supported by the ANOSIM analysis, which showed a significant difference between the two groups(p < 0.001).
To determine significantly different taxa, we applied linear discriminant analysis effect size (LEfSe) to compare samples between groups. LEfSe uses linear discriminant analysis to estimate the effect size of each differentially abundant feature. The threshold of the linear discriminant was set to two (Fig. 3A). At the phylum level, both the CKD patients and healthy controls had a typical gut microbiota structure; the main phyla were Firmicutes, Bacteroidetes, Proteobacteria, Actinobacteria, and Fusobacteria (Fig. 3B). The CKD patients showed a significantly higher abundance of Firmicutes and Actinobacteria, and a significantly lower abundance of Bacteroidetes and Fusobacteria (Fig. 3C). Moreover, the Firmicutes to Bacteroidetes ratio was comparable between the CKD patients and the control group (Fig. 3D). Pearson correlation analysis was used to identify the correlation between TMAO and these phyla ( correlation coefficient absolute value > 0.3 and P < 0.05). Bacteroidetes and Fusobacteria were negatively associated with TMAO production; however, Firmicutes and Actinobacteria were positively associated with TMAO production (Fig. 3E). At the genus level, the genera Bacteroidetes, Clostridium_XIVa, Parasutterella, Clostridium_XIVb, Roseburia, Fusobacterium, Megamonas, and Bilophila were more abundant in the healthy controls, whereas the genera Blautia, Bifidobacterium, Romboutsia, Dorea, Collinsella, Alloprevotella, Streptococcus, Lactobacillus, Eggerthella, Intestinibacter, Holdemanella, Anaerovorax, Clostridium_XVIII, and Clostridium_sensu_stricto were more abundant in patients with CKD (Fig. 3F). Among the genera with differences in the control and CKD groups, we used Pearson correlation analysis to determine the correlation between TMAO production and found that Collinsella, Alloprevotella, Streptococcus, Lactobacillus, and Clostridiumsensustricto were positively associated with TMAO production; however, Bacteroidetes, Clostridium_XIVa, Parasutterella, ClostridiumXIVb, Roseburia, Fusobacterium, and Bilophila were negatively associated with TMAO production( Fig. 3G).
Gut microbiota have both beneficial and adverse effects on human health. Recently, associations between the gut microbiota and CKD have gained attentions. Gut-derived metabolites and toxins affect CKD progression, and uremic toxins affect the microbiota[10, 11]. However, the relationship between gut microbiota and CKD is unclear, and further studies are necessary to explore the causal relationships between gut dysbiosis and kidney disease. Our study verified that TMAO level was correlated with both kidney damage and intestinal microflora disorder in patients with CKD, and providing a new perspective for better understanding the relationship between intestinal microflora and CKD. Here, we report the changes in gut microbiota and intestinal metabolite TMAO production in CKD patients.
TMAO is a gut microbiota-derived metabolite of choline, betaine, and carnitine. The plasma level of TMAO is determined by several factors including diet, gut microbial flora, drug administration, and liver flavin monooxygenase activity. Our data demonstrated that the plasma TMAO levels of the CKD patients were 15.4-fold higher than that of the control group, which was consistent with previous reports. Several studies reported that TMAO levels in CKD patients were elevated, and that renal function was also a major determinant of TMAO levels[12, 13]. Many studies have shown that circulating TMAO levels are linked to a wide range of health conditions including all-cause mortality, hypertension, cardiovascular disease, diabetes, cancer, and kidney function[14, 15]. Moreover, in mice choline and TMAO feeding is associated with increased renal fibrosis and impaired renal function, and inhibition of TMAO production attenuates CKD, suggesting that TMAO is not merely a marker of reduced renal function, but a potential cause of CKD[9]. In our study, CKD patients with low GFRs had significantly elevated TMAO levels, and those with high TMAO levels showed relatively higher levels of BUN and creatinine. Our results confirmed that renal function is a key contributor to TMAO concentration, and a high TMAO level might be an important risk factor affecting renal function in CKD patients.
Numerous studies have indicated that gut dysbiosis or toxic metabolite accumulation contribute to the onset and progression of CKD[16, 17]. The two most important bacterial phyla in the gastrointestinal tract, Firmicutes and Bacteroidetes, have gained considerable attention in recent years. The phyla Firmicutes and Bacteroidetes have been associated with metabolic diseases in humans[18]. Our data indicates that the relative abundance of Firmicutes phyla increased, while that of Bacteroidetes phyla decreased in CKD patients. The Firmicutes/Bacteroidetes (F/B) ratio is widely accepted to have an important influence in maintaining normal intestinal homeostasis. An increased or decreased F/B ratio is regarded as dysbiosis, whereby the former is usually observed with obesity and the latter is associated with inflammatory bowel disease (IBD)[19]. However, their association with disease is not always consistent between studies[20]. Accordingly, the F/B ratio was significantly increased in patients with CKD. Owing to the diversity within and across phyla, the mechanisms by which specific members of the microbiota can affect human health remains to be elucidated. Intestinibacter, Clostridium, and Romboutsia, belonging to Firmicutes, were negatively correlated with hypoglycemic effects[21]. Blautia and Intestinibacter increased significantly in Crohn's disease[22]. In our study, the abundance of genera Blautia, Romboutsia, Dorea, Lactobacillus, Intestinibacter, Holdemanella, Anaerovorax, Clostridium_XVIII, and Clostridium_sensu_stricto of the phylum Firmicutes was significantly exhausted in CKD patients. Bacteria-associated metabolites, such as butyrate, are the main energy sources for colon cells and play an important role in maintaining mucosal barrier function[23, 24]. Herein, we observed that the genera Roseburia and Megamonas, butyrate-producing bacteria from the phylum Firmicutes, were significantly reduced in CKD patients. Bacteroidetes are the largest phylum of gram-negative bacteria inhabiting the gastrointestinal tract and are considered the leading players in the healthy state and sophisticated homeostasis safeguarded by gut microbiota[25]. Our data indicated that the relative abundance of Bacteroidetes phyla decreased in CKD patients. Bacteroides and Prevotella are the two main Bacteroidetes genera. In our study, the genus Bacteroidetes was more abundant in the healthy controls, and the abundance of the genera Alloprevotella and Streptococcus of the phylum Bacteroidetes was significantly increased in CKD patients. The most frequently reported changes in the gut microbiome in stage 3–5 CKD, especially those undergoing dialysis, are related to lower levels of Bifidobacteriaceae and Lactobacillaceae and higher levels of Enterobacteriaceae[26]. Recently, a study investigated the microbial communities in the feces of CKD patients at different disease stages and healthy controls, suggested that the taxonomic composition of microbial communities differed between CKD patients and healthy controls, with the observed changes in the gut microbiota related to disease severity[27]. The patients in our study had higher GFRs and were not taking many phosphorus-lowering drugs or applying dietary restrictions. Data is lacking on how to fully characterize gut dysbiosis in CKD and understanding its physiological impact.
CKD is a global health problem leading to high rates of morbidity and mortality. One of the major causes of death in CKD patients is cardiovascular disease. Elevated levels of gut microbiome-derived TMAO are associated with tubulointerstitial fibrosis, atherosclerosis, and increased risk of major cardiovascular events. A study with 19,256 patients found that high concentrations of TMAO and its precursors were associated with an increased risk of major adverse cardiovascular events and all-cause mortality, independent of traditional risk factors[28]. Our data demonstrated that the plasma TMAO level of CKD patients was significantly higher than that of the control group. Pearson correlation analysis was used to study the correlation between TMAO and the gut microbiome. Primarily, two different bacterial species (Firmicutes and Proteobacteria) have been identified as responsible for the metabolism of choline to produce TMAO[29]. Our data showed that Firmicutes and Actinobacteria were positively associated with TMAO production. At the genus level, Collinsella, Alloprevotella, Streptococcus, Lactobacillus, and Clostridiumsensustricto were positively associated with TMAO production, however, Bacteroidetes, ClostridiumXIVa, Parasutterella, ClostridiumXIVb, Roseburia, Fusobacterium, and Bilophila were negatively associated with TMAO production. Recent studies have reported that the genus Lactobacillus contributes to TMAO production, however, the genus Bacteroides shows a significant negative correlation with plasma TMAO levels[30], which is consistent with our results. Further research is needed to explore the relationship between TMAO levels and gut microbiota.
Our study had certain limitations. First, this was a single-center study, and a relatively small number of individuals were included. The research findings need further validation in a larger cohort to confirm them. Second, there is a lack of in vitro studies to validate the effect of bacterial flora alteration on increased TMAO levels and CKD progression. Furthermore, TMAO levels and gut microbiota were measured only once, which might not capture the long-term levels of the gut microbiota and its metabolites. Despite these limitations, our results provide information regarding altered microbial diversity and increased TMAO levels in Chinese CKD patients. The influence of intestinal flora and increased TMAO levels require further investigation.
Our results indicate that CKD patients have increased plasma TMAO levels due to contributions from both impaired renal function and dysbiosis of the gut microbiota. Based on these findings, further research should be performed to identify the roles played by the gut microbiota and its metabolites in the progression of CKD. Moreover, gut microbiota-related metabolite regulation may be a potential biomarker for the diagnosis, prevention and treatment of CKD.
Ethics approval and consent to participate
This study was approved by the institutional review board of Zhejiang Provincial People’s Hospital , and all experiments were performed in accordance with the Declaration of Helsinki’s ethical principles. All patients provided informed consent.
Consent for publication
Not applicable.
Availability of data and materials
The data used in the study is from a publicly available database. The sequences from biofilm field samples are available at NCBI SRA: BioProject: PRJNA949558, which can be downloaded from https://www.ncbi.nlm.nih.gov/bioproject/PRJNA949558
Competing interests
The authors have no conflicts of interest to declare.
Funding
This work was supported by grants from the General Project of the Medical and Health of Zhejiang Province (Grant Number: 2020KY048, 2023KY052) and the Project of Scientific Research Foundation of Chinese Medicine(2022ZB035).
Author Contributions
Research idea and study design: Bo Lin, Wenli Zou; data acquisition: Wei zhang, Wei Shen; data analysis/interpretation: Yueming Liu, Wei zhang; statistical analysis: Wenli Zou ,Yueming Liu; supervision or mentorship: Wenli Zou and Bo Lin. Each author contributed important intellectual content during manuscript drafting or revision, accepts personal accountability for the author’s own contributions, and agrees to ensure that questions pertaining to the accuracy or integrity of any portion of the work are appropriately investigated and resolved.
Acknowledgements
We would like to thank Editage (www.editage.cn) for English language editing.