Comprehensive Metabolic Signature of Renal Dysplasia in Children. A Multiplatform Metabolomics Concept

Abstract

Renal dysplasia is a severe congenital abnormality of the kidney parenchyma, which is an important cause of end-stage renal failure in childhood and early adulthood. The diagnosis of renal dysplasia relies on prenatal or postnatal ultrasounds as children show no specific clinical symptoms before chronic kidney disease develops. Prompt diagnosis is important in terms of early introduction of nephroprotection therapy and improved long-term prognosis. Metabolomics was applied to study children with renal dysplasia to provide insight into the changes in biochemical pathways underlying its pathology and in search of early indicators for facilitated diagnosis. The studied cohort consisted of 72 children, 39 with dysplastic kidneys and 33 healthy controls. All subjects underwent comprehensive urine metabolic profiling with the use of gas chromatography and liquid chromatography coupled to mass spectrometry, with two complementary separation modes of the latter. Univariate and multivariate statistical calculations identified a total of nineteen metabolites, differentiating the compared cohorts, independent of their estimated glomerular filtration rate. Seven acylcarnitines, xanthine, and glutamine were downregulated in the urine of renal dysplasia patients. Conversely, renal dysplasia was associated with higher urinary levels of dimethylguanosine, threonic acid or glyceric acid. This is the first metabolomic study of subjects with renal dysplasia. The authors define a characteristic urine metabolic signature in children with dysplastic kidneys, irrespective of renal function, linking the condition with altered fatty acid oxidation, amino acid and purine metabolisms.

Keywords: GC-MS; LC-MS; metabolomics; multiplatform approaches; pediatric nephrology; renal dysplasia.

FIGURE 1

Chromatograms of urine metabolic fingerprints measured by means of RP LC-TOF-MS in positive (A) and negative (B) ionization modes, HILIC LC-TOF-MS in both positive (C) and negative (D) ionization modes and GC-QQQ/MS (E) in a child with dysplastic kidneys.

FIGURE 2

PCA models built on data obtained from RP-LC-TOF-MS analysis in positive (A) and negative (B) ionization modes, HILIC-LC-TOF-MS analysis in positive (C) and negative (D) ionization modes and from GC-MS analysis (E). Red circles correspond to samples from disease group, blue ones represent healthy controls and yellow ones QC samples.

FIGURE 3

PLS-DA models built on data obtained from RP-LC-TOF-MS analysis in positive (A) (R ² = 0.25, Q² = 0.282, CV-ANOVA p < 0.001) and negative (B) (R ² = 0.453, Q² = 0.236, CV-ANOVA p = 0.013) ionization modes, HILIC-LC-TOF-MS analysis in positive (C) (R ² = 0.227, Q² = 0.505, CV-ANOVA p < 0.001) and negative (D) (R ² = 0.264, Q² = 0.271, CV-ANOVA p < 0.001) ionization modes and from GC-MS analysis (E) (R ² = 0.164, Q² = 0.174, CV-ANOVA p = 0.017). Red circles correspond to samples from disease group, while blue ones represent healthy controls.

FIGURE 4

The main biochemical pathways altered only in renal dysplasia (*) and renal dysplasia accompanied by decreased eGFR (**). Decreased levels of urinary metabolites are marked in green. Increased urinary metabolite levels are marked in red.

FIGURE 5

Relationship between metabolic reprogramming of mitochondrial metabolism (FAO, OXPHOS) and kidney interstitial fibrosis.