Hao Hong,1,* Suya Zhou,2,* Junyao Zheng,3 Haimin Shi,3 Yue Chen,4 Ming Li3

1Department of Intensive Care Unit, The First Affiliated Hospital of Soochow University, Soochow, Suzhou, People’s Republic of China; 2Laboratory Nephrology, Jinshan hospital of Fudan University, Shanghai, People’s Republic of China; 3Laboratory Nephrology, The First Affiliated Hospital of Soochow University, Soochow, Suzhou, People’s Republic of China; 4Laboratory Nephrology, The First People’s Hospital of Kunshan, Soochow, Suzhou, People’s Republic of China

Purpose: The aim of this study was to investigate the changes of different metabolites in the body fluids of non-dialysis patients with chronic kidney disease (CKD) using a metabolomics approach. The goal was to identify early biomarkers of CKD progression through metabolic pathway analysis.
Patients and Methods: Plasma samples from 47 patients with stages 1– 4 CKD not requiring dialysis and 30 healthy controls were analyzed by liquid chromatography-mass spectrometry (LC-MS). Using multivariate data analysis, specifically a partially orthogonal least squares discriminant analysis model (OPLS-DA), we investigated metabolic differences between different stages of CKD. The sensitivity and specificity of the analysis were evaluated using the Area Under Curve (AUC) method. Furthermore, the metabolic pathways were analyzed using the Met PA database.
Results: Plasma samples from CKD patients and controls were successfully differentiated using an OPLS-DA model. Initially, twenty-five compounds were identified as potential plasma metabolic markers for distinguishing CKD patients from healthy controls. Among these, six compounds (ADMA, D-Ornithine, Kynurenine, Kynurenic acid, 5-Hydroxyindoleacetic acid, and Gluconic acid) were found to be associated with CKD progression It has been found to be associated with the progression of CKD. Changes in metabolic pathways associated with CKD progression include arginine and ornithine metabolism, tryptophan metabolism, and the pentose phosphate pathway.
Conclusion: By analyzing the metabolic pathways of different metabolites, we have identified the significant impact of CKD progression. The main metabolic pathways involved are Arginine and Ornithine metabolism, Tryptophan metabolism, and Pentose phosphate pathway. ADMA, D-Ornithine, L-Kynurenine, Kynurenic acid, 5-Hydroxyindoleacetic acid, and Gluconic acid could serve as potential early biomarkers for CKD progression. These findings have important implications for the early intervention and treatment of CKD, as well as for further research into the underlying mechanisms of its pathogenesis.

Keywords: arginine and ornithine metabolism, tryptophan metabolism, pentose phosphate pathway, biomarkers

Introduction

Chronic kidney disease (CKD) is a prevalent condition in the general population and a significant global public health concern. According to Hill et al, the estimated global prevalence of CKD ranges from 10.6% to 13.4%.1 In the early stages of CKD, symptoms may not be obvious and can be easily overlooked. However, the progression of CKD towards end-stage renal disease (ESRD) is associated with high costs, mortality, and a significant decline in a patient’s quality of life, particularly when dialysis treatment is introduced. The widely used plasma biochemical marker of glomerular filtration rate (GFR), plasma creatinine concentration (sCr), is often considered insensitive for detecting early stages of CKD. Hence, early recognition of CKD is crucial in order to slow down disease progression, reduce morbidity, and improve survival.

Metabolomics is a systematic analysis of metabolites in a biological specimen, focusing on the dynamic changes, interactions, and responses of metabolites in various metabolic pathways. It has become a commonly used approach in systems biology research, particularly for identifying new diagnostic and prognostic biomarkers for human diseases.2,3

Liquid chromatography-tandem mass spectrometry (LC-MS) is a powerful tool that enables accurate identification and quantification of compounds due to its greater sensitivity and specificity. Currently, LC-MS is primarily utilized to uncover the complexity of plasma metabolome and provide various biomedical applications based on the obtained results. The primary objective of this study was to assess plasma metabolic status in non-dialysis patients with early CKD compared with controls using LC-MS.

Material and MethodsStudy Design

This study included 47 patients with non-Dialysis CKD stage 1–4 (13, 11, 11, 12, respectively) and 30 healthy volunteers. CKD was diagnosed from patient medical records and CKD severity was assessed by estimated glomerular filtration rate (eGFR) calculated using the CKD-EPI formula.4 Inclusion criteria: 1, Patients diagnosed with non-dialysis CKD for the first time and who have not undergone any drug treatment; 2, All patients were adult. The studied groups were matched based on certain exclusion criteria, such as secondary glomerulonephritis, severe heart failure, cardiac arrhythmias, liver cirrhosis, hematopathy, infectious diseases, vegetarianism, solitary kidney, professional sports involvement, pregnancy, lactation, menstruation, and mental illness. This study was approved by the Ethics Research Society of the First Affiliated Hospital of Soochow University (no. 079). All participants signed an informed consent form. Our study was complied with the Declaration of Helsinki.

Sampling and Sample Preparation

Plasma samples were collected according to the ethical protocol of the Ethics Research Society of the First Affiliated Hospital of Soochow University. Informed consent was obtained from all participants. The study was carried out according to the principles set forth by the World Medical Association in Helsinki. Blood samples were collected under fasting conditions and then centrifuged at 3000 rpm for 10 minutes. The supernatant was separated and stored at −80°C until assayed. Prior to analysis, all samples were thawed at 4°C and vortexed for 60 seconds after adding 800 µL of methanol (OKA, China). The supernatant was then centrifuged for 10 minutes at 12,000 rpm at 4°C. The prepared sample was dissolved in 300L methanol aqueous solution (4:1, 4°C), filtered through a 0.22m filter membrane, and then used for LC-MS analysis. Quality control (QC) samples were prepared to evaluate the bias in the results of the analysis of the cell mixture caused by the analyzer itself.

LC-MS Conditions

Chromatographic separations were performed on a Thermo Ultimate 3000 system with an ACQUITY UPLC®HSS T3 column (150 x 2.1 mm, 1.8 µm, Waters) maintained at 40 °C. The autosampler temperature was 8 °C. Gradient elution was performed at a flow rate of 0.25 mL/min using 0.1% formic acid in water (Merck, Germany) (A) and 0.1% formic acid in acetonitrile (Thermo, USA) (B) or 5 mM ammonium formate in water (Sigma, Germany) (C) and acetonitrile (D). After equilibration, inject 2 µL of each sample. Solvent B (v/v) linear gradient 0–1min, 2% B/D; 1–9 minutes, 2–50% B/T; 9–14 minutes, 50–98% B/D; 14–15 minutes, 98% B/D; 15–15.5 minutes, 98–2% B/D; 15.5–17 minutes, 2% B/D.

ESI-MSn experiments were performed using a fine-focus Thermo Q mass spectrometer. The spray voltages for positive and negative modes were set to 3.8 kV and −2.5 kV, respectively. The sheath gas and auxiliary gas were set to 45 and 15 arbitrary units, respectively. The capillary temperature was maintained at 325 °C. The Orbitrap analyzer scans the mass range from m/z 81 to 1000 with a resolution of 70,000. Correlative Acquisition (DDA) mass spectrometry experiments were performed by HCD scanning. The normalized collision energy is 30 eV. Redundant spectral information is removed by dynamic subtraction.

Data Preprocessing and Statistical Comparison

Data sets for LC-MS analysis were converted using Proteowizard (v3.0.8789). R (v3.3.2) xcms was used for peak identification, filtering and alignment. Batch normalization was employed to compare orders of magnitude. Subsequently, multivariate statistical analysis was conducted using R (ropls). Principal component analysis (PCA) was used to assess data quality, while orthogonal projections to latent structures discriminant analysis (OPLS-DA) was used to observe differences between groups studied. Both one-way ANOVA p-value (P < 0.01) were applied.

Identification of Metabolites and Signaling Pathways

Identification of statistically significant metabolites in positive and negative ionization modes was performed through LC-MS analyses using databases such as HMDB, Metlin, Massbank, LipidMaps, and Mzcloud. MetPA was then applied to identify possible disturbed metabolic pathways.

Statistical Analysis of Clinical Data

For nonnormal variables, the median and interquartile range were given, while for normal variables, the mean and standard deviation were given. Subnormal variables were log-transformed and tested for normality. Normally distributed continuous variables were analyzed using one-way ANOVA with Fisher’s least significant test, and skewed continuous variables were analyzed using the Kruskal–Wallis test. Categorical variables were analyzed using the chi-squared test. Data analysis was performed using SPSS software, version 22.0. P < 0.05 was considered significant.

ResultsClinical and Biochemical Parameters

Clinical and biochemical parameters of recruited subjects were summarized in Table 1. BMI, Creatinine, blood urea nitrogen (BUN) and total cholesterol (TC) were transformed as logarithmic data to normalize the data allowing the use of parametric tests.

Table 1 Clinical Characteristics of Control Group and 47 Non-Dialysis Patients with Stage 1 to Stage 4

Plasma Metabolomics and Analytical Reproducibility Testing

The chromatogram of the plasma peak measured by LC-MS is shown in Figure 1. The OPLS-DA model was created using filtered and normalized data sets obtained from LC-MS analytical measurements (Figure 2). Check the stability of the analytical system, the reproducibility of the procedure and the test method using a QC sample of the PCA plot (Figure 2).

Figure 1 Base peak chromatogram of CKD patients and control group obtained with LC-MS positive (A), negative (B)ionization mode.

Notes: E control group; A-D: CKD stage 1 to stage 4.

Figure 2 PCA models built on QC data obtained with LC-MS analysis in (A) positive ionization mode, (B) negative ionization mode; OPLS-DA models built on plasma data obtained with LC-MS analysis in (C) positive ionization mode, (D) negative ionization mode, respectively.

Statistical Comparison and Metabolic Changes Associated with CKD Patients and Controls

Among the variables measured by LC-MS positive and negative ionization modes, 147 metabolites were significantly distributed in CKD patients with stage 1 to stage 4 and controls according to one-way ANOVA (P < 0.05), 78 metabolites according to VIP value > 1 and 67 metabolites were selected according to AUC value > 0.7 (Figure 3).

Figure 3 Heatmap of discriminating metabolites between CKD patients with stage 1 to stage 4 and controls with positive (A), negative (B) ionization mode.

Notes: E control group; A-D: CKD stage.

25 metabolites were selected after sifting chemicals and unmatched metabolites in KEGG pathways.6 metabolites showed a continuous upward trend period (Figure 4). The identified statistically significant metabolites and the differences between groups were shown in Table 2. The AUC values of 6 metabolites between control group and CKD patients were > 0.7 (P <0.01) (Figure 5).

Table 2 Statistical Analysis of 6 Main Discriminating Metabolites

Figure 4 Differences in normalized intensity of selected metabolites between 6 compared groups.

Notes: Based on One-way ANOVA and the Fisher least significant difference test. (A) ADMA, (B) D-Ornithine, (C) L-Kynurenine, (D) Kynurenic acid, (E) 5-Hydroxyindoleacetic acid, (F) Gluconic acid. 1–4, CKD stage 1 to stage 4; *: P <0.05, **: P <0.01.

Figure 5 ROC curve of 6 metabolites between 6 between control group and CKD patients.

Metabolic Pathway

The main metabolic pathways were illustrated in Figure 6. Notably, the arginine and proline metabolism, tryptophan metabolism, and pentose phosphate pathway were identified as the most significant pathways, which were enhanced in the early stages and the progression of CKD.

Figure 6 Pathway identification between CKD patients and controls.

Correlation Between 6 Discriminating Metabolites and Various Clinical Parameters

The correlation analysis between the six metabolites and CKD-related clinical parameters revealed significant associations (P <0.05) (Figure 7). These metabolites were mainly related to inflammation and oxidative stress in CKD.

Figure 7 Heatmap of correlation analysis between 6 discriminating metabolites and various clinical parameters.

Discussion

In this study, we employed an untargeted metabolomic approach to identify biomarkers for patients with non-dialysis CKD. We observed distinct differences in plasma metabolic profiles between patients with non-dialysis CKD and healthy controls. Notably, ADMA, D-Ornithine, L-Kynurenine, Kynurenic acid, 5-Hydroxyindoleacetic acid, and Gluconic acid exhibited significant changes.

One advantage of our study was the prospective inclusion of patients with non-dialysis CKD. This approach reduced the metabolic deviation caused by dialysis and allowed for the exploration of the original metabolic changes in CKD patients at an early stage. This provides valuable insights for the early diagnosis, prevention, and progression of CKD. Our OPLS-DA analysis revealed a significant difference between patients with non-dialysis CKD and healthy controls, highlighting the presence of a distinct metabolic pattern in non-dialysis CKD patients.

ADMA is an endogenous endothelial nitric oxide synthase inhibitor derived from arginine residues and catalyzed by protein arginine methyltransferase. Approximately 80% of ADMA is degraded via dimethylarginine dimethylaminohydrolase (DDAH), while the remaining ADMA is excreted by the kidney.5 Fliser et al reported an association between plasma ADMA and early-stage CKD patients, while Ravani et al identified plasma ADMA as an independent risk factor for patients with end-stage renal disease.6,7 Our results support these findings and suggest that this could be attributed to decreased renal excretion or reduced DDAH activity in CKD. ADMA reduces the level of nitric oxide (NO) by decreasing the activity of endothelial nitric oxide synthase (NOS), leading to impaired vascular tone and endothelial dysfunction. This could result in decreased renal perfusion and increased cardiovascular disease (CVD) events, which may further complicate CKD.8 Some studies have indicated that drugs such as EPO, vitamin E, statins, and ACE inhibitors can potentially interfere with the level of ADMA. However, these studies have not been able to explain whether the benefit observed in CKD patients is due to the reduction in ADMA or other aspects of these drugs.9–11 Further research is needed to elucidate the underlying mechanism. Age plays an important role in vascular aging. An imbalance between oxidative and antioxidant systems leads to increased reactive oxygen species production, NO inactivation, increased nitrosative stress, and ultimately age-related endothelial dysfunction.12 Our findings demonstrate a positive correlation between ADMA and age (r=0.471, P <0.01). Accumulation of ADMA with age leads to endothelial damage in CKD patients. The Neutrophil-to-Lymphocyte Ratio (NLR) is currently recognized as a biomarker for inflammation and atherosclerosis progression.13,14 The simultaneous increase of ADMA and NLR may be associated with inflammatory activity.15 In this study, ADMA showed a positive correlation with NLR (r=0.274, P <0.05). Similar findings were reported by Ibrahim et al, who found that NLR was significantly different between ESRD hemodialysis patients and controls, and NLR was positively correlated with ADMA.16 These results suggest that ADMA could potentially serve as a predictive marker for the early occurrence, progression, and end-stage of CKD. Further investigation into the role of ADMA in inflammation and endothelial injury is crucial for the diagnosis and treatment of CKD. In addition, it may be an interesting clinical intervention to explore the potential impact of relatively safe drugs, such as vitamin C consumption at different ages, on ADMA levels to potentially improve outcomes for patients with CKD.

D-Ornithine, an intermediate in the urea cycle, has been identified as a significant factor in hyperphosphatemia. This condition leads to increased expression of arginase 2, resulting in elevated levels of D-Ornithine.17 Several studies have reported the negative effects of elevated D-Ornithine levels in hyperphosphatemia on endothelial cells, including reduced production of nitric oxide and disruption of mitochondrial function.18,19 The concentration of blood phosphorus gradually increases with the progression of chronic kidney disease (CKD) (P <0.01) and shows a positive correlation with D-Ornithine (r=0.62, P <0.01). We speculate that the increase in blood phosphorus concentration in CKD patients may enhance the activity of arginase 2, leading to an elevation in ornithine concentration. However, we have not yet measured the activity of arginase 2 or other potential factors that may interfere with the enzyme. The relationship between phosphorus-arginase 2 and D-ornithine still requires further elaboration. This is of great significance in understanding the mechanism of ornithine as a potential therapeutic target for treating hyperphosphatemic-related cardiovascular events andCKD. The mechanism of cardiorenal syndrome in CKD patients is not yet fully understood. Given the role of D-Ornithine in both the kidneys and heart, investigating the impact of interfering with D-Ornithine on the development of cardiorenal syndrome in CKD patients could be an interesting research topic.

Statistically significant differences were observed in tryptophan metabolism between CKD patients and the control group. Tryptophan, an essential amino acid in humans, is primarily metabolized through two pathways: the kynurenine pathway, which produces L-Kynurenine and Kynurenic acid, and the serotonin pathway, which produces 5-Hydroxyindoleacetic acid. The kidney plays a crucial role in the metabolism and excretion of tryptophan. L-Kynurenine and Kynurenic acid act as endogenous agonists of the aromatic compound receptor (AHR), which is considered to be a significant factor in inducing immune response tolerance and promoting the reprogramming of Th17 cells into regulatory T cells.20,21 L-Kynurenine has the ability to interact with AHR in dendritic cells, leading to an upregulation in the expression of Indoleamine 2.3-dioxygenase. This, in turn, results in the further decomposition of tryptophan and the production of L-Kynurenine.22 In an inflammatory environment, the L-Kynurenine and AHR pathway exhibit protective effects.23 Increased levels of L-Kynurenine and Kynurenic acid have been observed in various complications of CKD, such as thrombosis, oxidative stress, mesangial cell proliferation, CKD-MBD, and dyslipidemia.24–29 These findings suggest a potential association between inflammation, immune disorders, and CKD. In this study, we found that the levels of L-Kynurenine, Kynurenic acid, and 5-Hydroxyindoleacetic acid increased with the progression of CKD (P <0.05). Additionally, L-Kynurenine and 5-Hydroxyindoleacetic acid showed a positive correlation with triglycerides (r=0.259, P <0.05, r=0.301, P <0.05). Furthermore, the key rate-limiting enzyme of the kynurenine pathway, Indoleamine 2.3-dioxygenase, was found to be overexpressed in CKD.30 Previous studies by Sekula P et al have demonstrated that Indoleamine 2.3-dioxygenase serves as a reliable indicator of impaired renal function.31 Our previous study has identified the alterations in Indoleamine 2.3-dioxygenase levels among CKD patients.32,33 However, further investigation is required to explore the impact of these changes on the tryptophan metabolic pathway and its role in the inflammatory and immune regulation mechanisms of CKD. Moreover, Clinicians have recognized the significance of depressive symptoms in certain CKD patients. Exploring the potential impact of modulating the serotonin metabolism pathway of tryptophan on the depressive state of CKD patients could be an intriguing topic.

Gluconic acid is enzymatically formed from glucose by the activity of glucose oxidase, which releases hydrogen peroxide. Previous research has shown an association between gluconate and decreased kidney function in patients with end-stage chronic kidney disease.34 In a rat nephrotoxicity model, gluconic acid levels changed in a dose- and time-dependent manner, suggesting that they may be related to the severity of renal injury.35 Additionally, gluconic acid has been found to be correlated with early Diabetic retinopathy and type 2 diabetes.36,37 Uric acid, on the other hand, is the end product of purine metabolism in the body. It has extracellular antioxidant effects and has been implicated in oxidative stress.38,39 Our study revealed a significant increase in gluconic acid levels in stage 2 of CKD (P <0.05), indicating its association with disease progression. Furthermore, we observed a positive correlation between gluconic acid and uric acid (r=0.251, P <0.05). Interestingly, our findings also showed an enhancement of the pentose phosphate pathway, which is primarily involved in gluconic acid metabolism. This pathway is considered to have potential protective effects, as its metabolite NADPH can effectively reduce reactive oxygen species levels, thereby counteracting oxidative stress in the kidneys.40 Weng et al demonstrated that increased NADPH production through the pentose phosphate pathway can alleviate oxidative stress in renal ischemia-reperfusion injury.41 These findings suggest that oxidative stress occurs in early CKD stages, triggering a protective mechanism. Metabolic pathways were analyzed using LC-MS, revealing changes in arginine and ornithine metabolism, tryptophan metabolism, and the pentose phosphate pathway. Metabolites identified in these pathways included ADMA, D-ornithine, L-kynurenine, kynurenic acid, 5-hydroxyindoleacetic acid, and Gluconic acid. These findings have potential implications for the early detection of CKD in patients. Furthermore, exploring interventions involving common clinical drugs targeting these metabolic pathways and metabolites may hold promise for the treatment of CKD.

Acknowledgments

We would like to thank Boxi Youth Natural Science Foundation.

Disclosure

All authors report no conflicts of interest in this work.

References

1. Hill NR, Fatoba ST, Oke JL, et al. Global prevalence of chronic kidney disease – a systematic review and meta-analysis. PLoS One. 2016;11(7):e0158765. doi:10.1371/journal.pone.0158765

2. Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate, ann. Intern Med. 2009;150:604.

3. Zhang ZH, Chen H, Vaziri ND, et al. Metabolomic signatures of chronic kidney disease of diverse etiologies in the rats and humans. J Proteome Res. 2016;15(10):3802–3812. doi:10.1021/acs.jproteome.6b00583

4. Andrassy KM. Comments on ‘KDIGO 2012 clinical practice guideline for the evaluation and management of chronic kidney disease’. Kidney Int. 2013;84(3):622–623. doi:10.1038/ki.2013.243

5. Chen J, Hamm LL, Mohler ER, et al. Interrelationship of multiple endothelial dysfunction biomarkers with chronic kidney disease. PLoS One. 2015;10(7):e0132047. doi:10.1371/journal.pone.0132047

6. Fliser D, Kronenberg F, Kielstein JT, et al. Asymmetric dimethylarginine and progression of chronic kidney disease: the mild to moderate kidney disease study. J Am Soc Nephrol. 2005;6(8):2456–2461. doi:10.1681/ASN.2005020179

7. Ravani P, Tripepi G, Malberti F, et al. Asymmetrical dimethylarginine predicts progression to dialysis and death in patients with chronic kidney disease: a competing risks modeling approach. J Am Soc Nephrol. 2005;16(8):2449–2455. doi:10.1681/ASN.2005010076

8. Emrich IE, Zawada AM, Martens-Lobenhoffer J, et al. Symmetric dimethylarginine (SDMA) outperforms asymmetric dimethylarginine (ADMA) and other methylarginines as predictor of renal and cardiovascular outcome in non-dialysis chronic kidney disease. Clin Res Cardiol. 2018;107(3):201–213. doi:10.1007/s00392-017-1172-4

9. Fujiwara N, Nakamura T, Sato E, et al. Renovascular protective effects of erythropoietin in patients with chronic kidney disease. Intern Med. 2011;50(18):1929–1934. doi:10.2169/internalmedicine.50.5145

10. Nanayakkara PW, Kiefte-de Jong JC, ter Wee PM, et al. Randomized placebo-controlled trial assessing a treatment strategy consisting of pravastatin, vitamin E, and homocysteine lowering on plasma asymmetric dimethylarginine concentration in mild to moderate CKD. Am J Kidney Dis. 2009;53(1):41–50. doi:10.1053/j.ajkd.2008.06.016

11. Hov GG, Sagen E, Hatlen G, et al. Arginine/asymmetric dimethylarginine ratio and cardiovascular risk factors in patients with pre-dialytic chronic kidney disease. Clin Biochem. 2011;44(8–9):642–646. doi:10.1016/j.clinbiochem.2011.03.024

12. Mozos I, Luca CT. Crosstalk between oxidative and nitrosative stress and arterial stiffness. Curr Vasc Pharmacol. 2017;15(5):446–456. doi:10.2174/1570161115666170201115428

13. Buonacera A, Stancanelli B, Colaci M, Malatino L. Neutrophil to lymphocyte ratio: an emerging marker of the relationships between the immune system and diseases. Int J Mol Sci. 2022;23(7):3636. doi:10.3390/ijms23073636

14. Mozos I, Malainer C, Horbańczuk J, et al. Inflammatory markers for arterial stiffness in cardiovascular diseases. Front Immunol. 2017;8:1058. doi:10.3389/fimmu.2017.01058

15. Yuksel M, Yildiz A, Oylumlu M, et al. Novel markers of endothelial dysfunction and inflammation in Behçet’s disease patients with ocular involvement: epicardial fat thickness, carotid intima media thickness, serum ADMA level, and neutrophil-to-lymphocyte ratio. Clin Rheumatol. 2016;35(3):701–708. doi:10.1007/s10067-015-2907-0

16. Dogan I, Dogan T, Yetim M, et al. Relation of serum ADMA, apelin-13 and LOX-1 levels with inflammatory and echocardiographic parameters in hemodialysis patients. Ther Apher Dial. 2018;22(2):109–117. doi:10.1111/1744-9987.12613

17. Zhou R, Kang X, Tang B, et al. Ornithine is a key mediator in hyperphosphatemia-mediated human umbilical vein endothelial cell apoptosis: insights gained from metabolomics. Life Sci. 2016;146:73–80. doi:10.1016/j.lfs.2016.01.003

18. Shuto E, Taketani Y, Tanaka R, et al. Dietary phosphorus acutely impairs endothelial function. J Am Soc Nephrol. 2009;20(7):1504–1512. doi:10.1681/ASN.2008101106

19. Di Marco GS, Hausberg M, Hillebrand U, et al. Increased inorganic phosphate induces human endothelial cell apoptosis in vitro. Am J Physiol Ren Physiol. 2008;294(6):F1381–F1387. doi:10.1152/ajprenal.00003.2008

20. Marszalek-Grabska M, Walczak K, Gawel K, et al. Kynurenine emerges from the shadows - current knowledge on its fate and function. Pharmacol Ther. 2021;225:107845. doi:10.1016/j.pharmthera.2021.107845

21. Huang T, Song J, Gao J, et al. Adipocyte-derived kynurenine promotes obesity and insulin resistance by activating the AhR/STAT3/IL-6 signaling. Nat Commun. 2022;13(1):3489. doi:10.1038/s41467-022-31126-5

22. Sharma MD, Pacholczyk R, Shi H, et al. Inhibition of the BTK-IDO-mTOR axis promotes differentiation of monocyte-lineage dendritic cells and enhances anti-tumor T cell immunity. Immunity. 2021;54(10):2354–2371.e8. doi:10.1016/j.immuni.2021.09.005

23. Bessede A, Gargaro M, Pallotta MT, et al. Aryl hydrocarbon receptor control of a disease tolerance defence pathway. Nature. 2014;511(7508):184–190. doi:10.1038/nature13323

24. Addi T, Dou L, Burtey S. Tryptophan-derived uremic toxins and thrombosis in chronic kidney disease. Toxins. 2018;10(10):412. doi:10.3390/toxins10100412

25. Pawlak K, Kowalewska A, Mysliwiec M, et al. Kynurenine and its metabolites--kynurenic acid and anthranilic acid are associated with soluble endothelial adhesion molecules and oxidative status in patients with chronic kidney disease. Am J Med Sci. 2009;338(4):293–300. doi:10.1097/MAJ.0b013e3181aa30e6

26. Yoshimura H, Sakai T, Kuwahara Y, et al. Effects of kynurenine metabolites on mesangial cell proliferation and gene expression. Exp Mol Pathol. 2009;87(1):70–75. doi:10.1016/j.yexmp.2009.02.002

27. Kalaska B, Pawlak K, Domaniewski T, et al. Elevated levels of peripheral kynurenine decrease bone strength in rats with chronic kidney disease. Front Physiol. 2017;8:836. doi:10.3389/fphys.2017.00836

28. Zinellu A, Sotgia S, Mangoni AA, et al. Impact of cholesterol lowering treatment on plasma kynurenine and tryptophan concentrations in chronic kidney disease: relationship with oxidative stress improvement. Nutr, Metab Cardiovasc Dis. 2015;25(2):153–159. doi:10.1016/j.numecd.2014.11.004

29. Pawlak D, Oksztulska-Kolanek E, Znorko B, et al. The association between elevated levels of peripheral serotonin and its metabolite - 5-hydroxyindoleacetic acid and bone strength and metabolism in growing rats with mild experimental chronic kidney disease. PLoS One. 2016;11(10):e0163526. doi:10.1371/journal.pone.0163526

30. Mohit K, Wang S, Guan Q, et al. Jevnikar, indoleamine 2,3-dioxygenase expression promotes renal ischemia-reperfusion injury. Am J Physiol Physiol. 2008;295(1):F226–F234. doi:10.1152/ajprenal.00567.2007

31. Sekula P, Goek ON, Quaye L, et al. A metabolome-wide association study of kidney function and disease in the general population. J Am Soc Nephrol. 2016;27(4):1175–1188. doi:10.1681/ASN.2014111099

32. Hong H, Zhou S, Chen Y, et al. Analysis of tryptophan-kynurenine pathway in 47 non-dialysis patients with chronic kidney disease. Chin J Kidney Dis Invest. 2020;9(6):247–252.

33. Hong H, Zhou S, Shi H, et al. Plasma and urine indoleamine 2,3-dioxygenase activity: promising biomarkers for chronic kidney disease and inflammation status. J Inflamm Res. 2022;15:5129–5139. doi:10.2147/JIR.S378594

34. Kimura T, Yasuda K, Yamamoto R, et al. Identification of biomarkers for development of end-stage kidney disease in chronic kidney disease by metabolomic profiling. Sci Rep. 2016;6(1):26138. doi:10.1038/srep26138

35. Hu X, Shen J, Pu X, et al. Urinary time- or dose-dependent metabolic biomarkers of aristolochic acid-induced nephrotoxicity in rats. Toxicol Sci. 2017;156(1):123–132. doi:10.1093/toxsci/kfw244

36. Chen L, Cheng CY, Choi H, et al. Plasma metabonomic profiling of diabetic retinopathy. Diabetes. 2016;65(4):1099–1108. doi:10.2337/db15-0661

37. Mercer KE, Yeruva L, Pack L, et al. Xenometabolite signatures in the UC Davis type 2 diabetes mellitus rat model revealed using a metabolomics platform enriched with microbe-derived metabolites. Am J Physiol Gastrointest Liver Physiol. 2020;319(2):G157–G169. doi:10.1152/ajpgi.00105.2020

38. Ament Z, Bevers MB, Wolcott Z, Kimberly WT, Acharjee A. Uric acid and gluconic acid as predictors of hyperglycemia and cytotoxic injury after stroke. Transl Stroke Res. 2021;12(2):293–302. doi:10.1007/s12975-020-00862-5

39. Corry DB, Eslami P, Yamamoto K, Nyby MD, Makino H, Tuck ML. Uric acid stimulates vascular smooth muscle cell proliferation and oxidative stress via the vascular renin-angiotensin system. J Hypertens. 2008;26(2):269–275. doi:10.1097/HJH.0b013e3282f240bf

40. Pácal L, Tomandl J, Svojanovsky J, et al. Role of thiamine status and genetic variability in transketolase and other pentose phosphate cycle enzymes in the progression of diabetic nephropathy. Nephrol Dial Transplant. 2011;26(4):1229–1236. doi:10.1093/ndt/gfq550

41. Weng XF, Li ST, Song Q, et al. Protective effect of nicotinamide adenine dinucleotide phosphate on renal ischemia-reperfusion injury. Kidney Blood Press Res. 2018;43(3):651–663. doi:10.1159/000489620