Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Apr 23;10(1):1835.
doi: 10.1038/s41467-019-09735-4.

Gut Microbiome-Derived Phenyl Sulfate Contributes to Albuminuria in Diabetic Kidney Disease

Affiliations
Free PMC article

Gut Microbiome-Derived Phenyl Sulfate Contributes to Albuminuria in Diabetic Kidney Disease

Koichi Kikuchi et al. Nat Commun. .
Free PMC article

Abstract

Diabetic kidney disease is a major cause of renal failure that urgently necessitates a breakthrough in disease management. Here we show using untargeted metabolomics that levels of phenyl sulfate, a gut microbiota-derived metabolite, increase with the progression of diabetes in rats overexpressing human uremic toxin transporter SLCO4C1 in the kidney, and are decreased in rats with limited proteinuria. In experimental models of diabetes, phenyl sulfate administration induces albuminuria and podocyte damage. In a diabetic patient cohort, phenyl sulfate levels significantly correlate with basal and predicted 2-year progression of albuminuria in patients with microalbuminuria. Inhibition of tyrosine phenol-lyase, a bacterial enzyme responsible for the synthesis of phenol from dietary tyrosine before it is metabolized into phenyl sulfate in the liver, reduces albuminuria in diabetic mice. Together, our results suggest that phenyl sulfate contributes to albuminuria and could be used as a disease marker and future therapeutic target in diabetic kidney disease.

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1
Fig. 1
Diabetic SLCO4C1-Tg rats showed reduced proteinuria. a Body weight, blood glucose, blood urea nitrogen (BUN), and creatinine clearance (Ccr) in diabetic SLCO4C1-Tg rats (white circles, n = 5) and WT rats (black circles, n = 6). b Representative PAS (left) stained sections of WT and SLCO4C1-Tg rat glomeruli. Bars = 100 μm. n = 50 in each group. Data are mean ± SEM. *p < 0.05 vs. WT. Student t test. c Variation in the five groups in the PLS-DA scores plot using the chemical features detected in plasma. Sample conditions are represented by color coded circles: WT-d7 (black), WT-d63 (blue), WT-d119 (cyan), Tg-d63 (red), and Tg-d119 (orange). Groups D7, D63, and D119 are surrounded by black, red, and orange solid rings, respectively. Groups WT-d63, WT-d119, Tg-d63, and Tg-d119 are surrounded by blue, cyan, red, and orange dotted rings, respectively. d S-plot analyses of orthogonal partial least square-discriminant analysis (OPLS-DA) of WT rats on days 7 and 119. The selected 94 features (surrounded by a dotted red rectangle) were significantly greater on day 119 than on day 7. e S-plot analyses of OPLS-DA of WT and Tg rats on day 63. f S-plot analyses of OPLS-DA of WT and Tg rats on day 119. g Changes in the concentration of the selected feature m/z 172.97 with the progression of diabetes. Wild-type rats (white column, n = 6); SLCO4C1-Tg rats (black column, n = 5). h Fragment ion mass spectrum of m/z 172.97. The main detected precursor and fragment ion was m/z 172.97. The product ions found in the corresponding extracted MSE high-energy spectrum at m/z 79.9, m/z 93.0, m/z 109.0, and m/z 121.0 were hypothesized to be [M-C6H5O-H]-, [M-SO3-H]-, [M-SO2-H]-, and [M-C4H4-H]-, respectively. i Chemical structure of PS. j SLCO4C1-mediated PS uptake by SLCO4C1/MDCKII cells (n = 3). Data are mean ± SEM. *p < 0.05 vs. control according to Student t test (a, b, g, j). Source data are provided as a Source Data file and untargeted metabolome data are provided as a Supplementary Data 1
Fig. 2
Fig. 2
Pathophysiological features of PS in vitro and in vivo. a Plasma levels of PS, PCS; IS, and TMAO in 11-week-old db/db mice in control (Ctrl; n = 5) or PS (50 mg/kg/day for 6 weeks, n = 6). b Plasma albumin and Cr levels before and after administering PS for 6 weeks to db/db mice. Control (white circles, n = 5) and PS-treated (black circles, n = 6) groups. c Glomeruli in db/db mouse with or without PS for 6 weeks. Bar, 80 μm (PAS) and 1 μm (EM). d Cell toxicity analysis (n = 6). e Cellular GSH level (n = 6). f Bioenergetic characterization of cultured human podocytes in terms of oxygen consumption rate (OCR). n = 4. g PS, albuminuria and Cr levels before and after administering PS for 6 weeks in HFD-KKAy mouse (6 weeks old). For PS, n = 3 (HFD-control) and n = 4 ((HFD-PS). For Cr and albuminuria, n = 4. Wilcoxon (PS) and Student’s t test (Cr and albuminuria). h Histological images of PAS and electron microscopic analysis of podocytes from control (Ctrl; top row) and PS-treated (PS; bottom row) groups of HFD-KKAy mice. Scale bar = 200 μm for PAS and 1 μm for electron microscopy. The effacement of podocytes (white arrows) and GBM thickness (yellow arrow heads) are shown. The inflammatory area (Elastica Masson) and macrophage infiltration (F4/80 immunostaining) around the vascular area are indicated (black arrow heads). Scale bar, 80 μm. i PS levels in eNOS knockout mice with or without diabetes (n = 5). j Histological examination stained with PAS of eNOS knockout mice with or without diabetes. Scale bar, 50 μm. *p < 0.05 vs. control according to Student t (a, b, d, e, I, j) or Tukey’s test (f). Source data are provided as a Source Data file
Fig. 3
Fig. 3
The clinical significance of PS in DKD patients. a The relationship between the plasma PS concentration (n = 362). b The relationship between the plasma PS concentration and the urinary albumin level (n = 362). c The relationship between the plasma PS concentration and the eGFR (n = 362). d The relationship between the plasma PS concentration and soluble urokinase-type plasminogen activator receptor (suPAR) (n = 362). e PS is a predictor of 2-year albuminuria. Adjusted odds ratio (95% confidence interval). Note that, among the known factors, PS was the only factor that served as a predictor of the progression of 2-year ACR in patients with microalbuminuria (n = 87). Receiver operating characteristic (ROC) curve analysis in the U-CARE Study, showing comparison of known factors. f AUC-ROC using known factors (0.713) and adjusted with PS (0.751) (n = 87). g AUC-ROC using known factors (0.713) and adjusted with suPAR (0.725). (n = 87). h AUC-ROC using known factors with suPAR (0.725) and adjusted with PS (0.752) (n = 87)
Fig. 4
Fig. 4
TPL inhibitors lowered PS concentrations and reduced renal damage in mouse models of diabetic kidney disease and adenine-induced renal failure. a Scheme of gut microbiome-derived PS, PCS, and IS generation. Dietary tyrosine is converted to phenol by tyrosine phenol-lyase (TPL) and further modified to produce PS or PCS. Dietary tryptophan is first converted to indole by tryptophan indole-lyase (TIL), further converted and finally modified into IS. Specific inhibitors of TPL are 2-aza-tyrosine and l-meta-tyrosine, while homo-BZI-Ala is a specific inhibitor of TIL. b Plasma concentrations of PS, IS, PCS, TMAO, and Cr in db/db mice before (Pre) and after treated with 2-aza-tyrosine (post, 10 mg/kg, n = 10). c Urinary albumin/creatinine ratio of db/db mice with either no intervention (Ctrl) or treatment with the TPL inhibitor 2-aza-tyrosine (AZA). N = 6 in each group. d Plasma concentrations of PS, IS, PCS, TMAO, and Cr in the adenine-induced renal failure mouse model treated with 2-aza-tyrosine (5 mg/kg) for 14 days. Four groups were measured: control mice (Ctrl), control mice treated with 2-aza-tyrosine (AZA), adenine-induced renal failure mice with no intervention (RF), and adenine-induced renal failure mice treated with 2-aza-tyrosine (RF-AZA). n = 5 in each group. e Representative imaging mass spectrometry spectra of PS and IS in kidney sections from renal failure mice (RF) and renal failure mice administered 2-aza-tyrosine (5 mg/kg) (RF + AZA). Mass spectrometry imaging shows the distribution of PS ([M-H], m/z 172) and IS ([M-H], m/z 212). Scale bar, 1000 μm. f Quantitative analysis of PS and IS tissue contents in mouse kidneys in RF (n = 4) and others (n = 5). g Morphometric analysis of the fractional cortical tubular area of Masson trichrome stained whole kidney images. n = 5 in each group. Data are presented as the percentage of the total cortex (n = 5). Scale bar, 500 μm. h Enzymatic activity of TIL. TIL activity under control conditions (Ctrl), with TIL inhibitor homo-BZI-Ala (100 μM) and a tyrosine phenol-lyase inhibitors (l-meta-tyrosine (LmT) or 2-aza-tyrosine (AZA), 1 mM), n = 3. *p < 0.05 vs. control according to Paired t test (b), data are mean ± SEM *p < 0.05 vs. control according to t test (c, d, g, f) or Tukey–Kramer test (h, *p < 0.05, **p < 0.01, ***p < 0.001). Source data are provided as a Source Data file
Fig. 5
Fig. 5
Effect of 2-aza-tyrosine (AZA) on the gut microbiome in adenine-induced renal failure mice. a Observed OTUs rarefaction analysis. OTU rarefaction curves of gut microbiome were used to estimate richness in the following groups: control mice (Ctrl); control mice treated with 5 mg/kg 2-aza-tyrosine (AZA); adenine-induced renal failure mice (RF); and adenine-induced renal failure mice treated with 5 mg/kg 2-aza-tyrosine (RF-AZA). b OTU-based α-diversity of each microbiome. No significant diversity was seen in any group by Chao1 (left) and Shannon (right) analyses. c Principal coordinates analysis of the microbiome profiles using weighted UniFrac. Scores are presented for the first principal component (PCo1) vs. both the second principal component (PCo2) and the third principal component (PCo3). d Relative abundance of microbiota based on the average number of each subgroup at order, family and genus levels. The major subgroups are indicated on the right. e The relative abundance of microbiota differed significantly among the four groups analyzed. The y-axis indicates the relative abundance of each microbe as a percentage. The proportional change in the minor groups correlated with renal failure and 2-aza-tyrosine treatment. The major microbial order (Bacteroidales, Clostridales, and Lactobacillales) were unchanged by either renal failure or 2-aza-tyrosine treatment. However, the minor populations (Coriobacteriales and Erysipelotrichales) were significantly changed in the correlation with the PS concentration by renal failure and 2-aza-tyrosine treatment. n = 5 in each group. Data were mean ± SEM. Statistical analysis was performed by the Tukey–Kramer test, followed by FDR correction of p values. *pFDR < 0.1 was treated as statistically significant. The following groups were tested: control group (Ctrl); control group administered 2-aza-tyrosine (AZA); the renal failure group (RF); and the renal failure group administered 2-aza-tyrosine (RF-AZA). Metagenome Source data are provided as in a Source Data file
Fig. 6
Fig. 6
A schematic model of the generation and toxicity of PS. In gut microbiota, TPL converts tyrosine to phenol and ammonia. PS is also generated in the liver. PS accumulates in plasma as a metabolite and has deleterious effects on the vasculature and kidneys. In diabetic kidney disease, PS damages podocytes, accelerates GBM thickening, and induces proteinuria. Treatment with TPL inhibitors reduces plasma PS levels and prevents the progression of renal failure in animal models

Comment in

Cited by 7 articles

See all "Cited by" articles

References

    1. Ritz E, Orth SR. Nephropathy in patients with type 2 diabetes mellitus. N. Engl. J. Med. 1999;341:1127–1133. doi: 10.1056/NEJM199910073411506. - DOI - PubMed
    1. Fineberg D, Jandeleit-Dahm KA, Cooper ME. Diabetic nephropathy: diagnosis and treatment. Nat. Rev. Endocrinol. 2013;9:713–723. doi: 10.1038/nrendo.2013.184. - DOI - PubMed
    1. Pavkov ME, et al. Early renal function decline in type 2 diabetes. Clin. J. Am. Soc. Nephrol. 2012;7:78–84. doi: 10.2215/CJN.07610711. - DOI - PMC - PubMed
    1. Mikkaichi T, et al. Isolation and characterization of a digoxin transporter and its rat homologue expressed in the kidney. Proc. Natl Acad. Sci. USA. 2004;101:3569–3574. doi: 10.1073/pnas.0304987101. - DOI - PMC - PubMed
    1. Menni C, et al. Biomarkers for type 2 diabetes and impaired fasting glucose using a nontargeted metabolomics approach. Diabetes. 2013;62:4270–4276. doi: 10.2337/db13-0570. - DOI - PMC - PubMed

Publication types

MeSH terms

Feedback