Metabonomic analysis of potential biomarkers and drug targets involved in diabetic nephropathy mice

Abstract

Diabetic nephropathy (DN) is one of the lethal manifestations of diabetic systemic microvascular disease. Elucidation of characteristic metabolic alterations during diabetic progression is critical to understand its pathogenesis and identify potential biomarkers and drug targets involved in the disease. In this study, (1)H nuclear magnetic resonance ((1)H NMR)-based metabonomics with correlative analysis was performed to study the characteristic metabolites, as well as the related pathways in urine and kidney samples of db/db diabetic mice, compared with age-matched wildtype mice. The time trajectory plot of db/db mice revealed alterations, in an age-dependent manner, in urinary metabolic profiles along with progression of renal damage and dysfunction. Age-dependent and correlated metabolite analysis identified that cis-aconitate and allantoin could serve as biomarkers for the diagnosis of DN. Further correlative analysis revealed that the enzymes dimethylarginine dimethylaminohydrolase (DDAH), guanosine triphosphate cyclohydrolase I (GTPCH I), and 3-hydroxy-3-methylglutaryl-CoA lyase (HMG-CoA lyase) were involved in dimethylamine metabolism, ketogenesis and GTP metabolism pathways, respectively, and could be potential therapeutic targets for DN. Our results highlight that metabonomic analysis can be used as a tool to identify potential biomarkers and novel therapeutic targets to gain a better understanding of the mechanisms underlying the initiation and progression of diseases.

Figure 1. Histological examination of kidney tissues.

Representative HE and PAS stains of kidney tissues from 17-week-old wildtype mice and db/db mice.

Figure 2. NMR spectra of urine samples.

Representative ¹H NMR spectra of urine samples obtained from the wildtype mice (A) and db/db mice (B), respectively.

Figure 3. Pattern recognition analysis of urine samples.

(A) PLS trajectory based on the mean ¹H NMR spectra of urine samples collected from the db/db mice () at various time points (9-wk, 11-wk, 13-wk, 15-wk and 17-wk), and the age-matched wildtype mice (). (B) The PLS-DA score plot based on ¹H-NMR spectra of urine samples from db/db mice 9-wk (), 11-wk (), 13-wk (), 15-wk () and 17-wk (). (C) is the loading plot revealing the metabolites with large intensities responsible for the discrimination of the corresponding score plot shown (A).

Figure 4. Quantitative analysis of urinary metabolites.

Relative abundances of metabolites obtained from ¹H NMR spectra of urine samples collected from the db/db mice and the wildtype mice at 9-wk, 11-wk, 13-wk, 15-wk and 17-wk, respectively. Keys: MNA, 1-methylnicotinamide; ^*P < 0.05 and ^**P < 0.01 compared with age-matched wildtype mice; ^#P < 0.05 and ^##P < 0.01 compared with the db/db mice at 9-wk.

Figure 5. NMR spectra of kidney samples.

Representative ¹H NMR spectra of kidney samples obtained from the wildtype mice (A) and db/db mice (B), respectively.

Figure 6. Pattern recognition analysis of kidney samples.

The PLS-DA score plot (A) and validation plot (B) based on the ¹H NMR spectra of kidney samples obtained from the wildtype mice () and db/db mice (). The coefficient-coded loading plot (C) corresponding to PLS-DA revealing the metabolites with large intensities responsible for the discrimination of the corresponding score plots.

Figure 7. Correlation analysis of urinary and renal metabolites.

Pearson’s correlations of UACR and quantities of the metabolites determined from 17-week-old mice urine samples (A, wildtype mice; B, db/db mice) and kidney samples (C, wildtype mice; D, db/db mice). Red and blue represent positive and negative correlations, respectively, the colour scale represents Pearson’s correlation coefficients. Keys: UACR, urinary albumin to creatinine ratio; LAC, lactate; PYR, pyruvate; SUCC, succinate; 2-OX, 2-oxoglutarate; CIT, citrate; C-AC, cis-aconitate; FUM, fumarate; Ma, methylamine; DMA, dimethylamine; TMA, trimethylamine; ACE, acetate; 3-HB, 3-hydroxybutyrate; AC, acetone; ACA, acetoacetate; CRE, creatine; CRT, creatinine; ALLA, allantion; HIP, hippurate; MNA, 1-methylnicotinamide; 3-IS, 3-indoxylsulfate; FOR, formate; ALA, alanine; PHE, phenylalanine; CHO, choline; M-INS, myo-inositol; GLU, glutamate; GLY, glycine; TYR, tyrosine; TAU, taurine; ASP, aspartate; VAL, valine; LEU, leucine; ILE, isoleucine; URA, uracil; URI, uridine; GTP, guanosine triphosphate; NIA, niacinamide.

Figure 8. Disturbed metabolic pathways related to pathogenic process of diabetic nephropathy.

The metabolite changes detected by ¹H NMR urine and kidney analysis and the pathway referenced to the KEGG database show the interrelationship of the identified metabolic pathways involved in the db/db mice. Metabolites in red and green represent increase and decrease in levels, respectively, compared with wildtype mice. Stars represent the potential targets of drugs for diabetic nephropathy.