Metabolic profiling reveals new serum biomarkers for differentiating diabetic nephropathy

Akiyoshi Hirayama, Eitaro Nakashima, Masahiro Sugimoto, Shin Ichi Akiyama, Waichi Sato, Shoichi Maruyama, Seiichi Matsuo, Masaru Tomita, Yukio Yuzawa, Tomoyoshi Soga

Research output: Contribution to journalArticlepeer-review

134 Citations (Scopus)


Capillary electrophoresis coupled with time-offlight mass spectrometry was used to explore new serum biomarkers with high sensitivity and specificity for diabetic nephropathy (DN) diagnosis, through comprehensive analysis of serum metabolites with 78 diabetic patients. Multivariate analyses were used for identification of marker candidates and development of discriminative models. Of the 289 profiled metabolites, orthogonal partial leastsquares discriminant analysis identified 19 metabolites that could distinguish between DN with macroalbuminuria and diabetic patients without albuminuria. These identified metabolites included creatinine, aspartic acid, γ-butyrobetaine, citrulline, symmetric dimethylarginine (SDMA), kynurenine, azelaic acid, and galactaric acid. Significant correlations between all these metabolites and urinary albumin-to-creatinine ratios (p<0.009, Spearman's rank test) were observed. When five metabolites (including γ-butyrobetaine, SDMA, azelaic acid and two unknowns) were selected from 19 metabolites and applied for multiple logistic regressionmodel, AUC value for diagnosing DN was 0.927 using the whole dataset, and 0.880 in a cross-validation test. In addition, when four known metabolites (aspartic acid, SDMA, azelaic acid and galactaric acid) were applied, the resulting AUC was still high at 0.844 with the whole dataset and 0.792 with cross-validation. Combination of serum metabolomics with multivariate analyses enabled accurate discrimination of DN patients. The results suggest that capillary electrophoresis-mass spectrometry based metabolome analysis could be used for DN diagnosis.

Original languageEnglish
Pages (from-to)3101-3109
Number of pages9
JournalAnalytical and bioanalytical chemistry
Issue number10
Publication statusPublished - 2012 Dec


  • Biomarker
  • Capillary electrophoresismass spectrometry
  • Diabetic nephropathy
  • Metabolome
  • Multiple logistic regression
  • Orthogonal partial least-squares discriminant analysis

ASJC Scopus subject areas

  • Analytical Chemistry
  • Biochemistry


Dive into the research topics of 'Metabolic profiling reveals new serum biomarkers for differentiating diabetic nephropathy'. Together they form a unique fingerprint.

Cite this