Diagnostic biomarkerGut MicrobiotaHippurateKetone bodiesMetabolomics

A Urinary Metabolic Signature for Multiple Sclerosis and Neuromyelitis Optica

  • 2016
  • Teklab Gebregiworgis, Helle H. Nielsen, Chandirasegaran Massilamany, et al.
Why this matters for your health

Multiple sclerosis (MS) and neuromyelitis optica spectrum disorder (NMO-SD) can look alike, especially early on. Mixing them up can lead to wrong treatment and more disease activity. Urine is easy and non-invasive to collect, yet it is rarely studied for MS biomarkers compared with spinal fluid or blood. This study suggests that a urine test based on NMR metabolite profiling could help diagnose MS, tell it apart from NMO-SD, and shed light on disease mechanisms. The authors stress that it is only an early proof of principle.

Summary

The researchers collected 38 urine samples: 14 from 7 healthy volunteers, 15 from 8 patients with relapsing-remitting MS, and 9 from 9 patients with aquaporin-4-positive NMO-SD. Samples were analyzed by ¹H NMR spectroscopy, multivariate statistics (OPLS-DA), ANOVA and t-tests, and metabolite pathway analysis. Urine from MS patients showed a metabolic signature distinct from healthy controls and from NMO-SD patients. A total of 27 metabolites were altered across MS and NMO-SD. They were linked to ketone body metabolism, amino acids, propionate and pyruvate metabolism, the TCA cycle and glycolysis. The authors relate these to energy and fatty acid metabolism, mitochondrial activity and the gut microbiota. They call their findings a “working hypothesis” because of the small sample.

Key findings

  • MS versus healthy controls: the model separated the groups well (R²Y = 0.77, Q² = 0.39, CV-ANOVA p = 7.8 × 10⁻⁴), and 26 metabolites differed.
  • The strongest MS-versus-healthy contributors were creatinine, hippurate, 3-hydroxybutyrate, malonate, oxaloacetate and trimethylamine N-oxide (TMAO).
  • Using ANOVA, a t-test with false-discovery correction, and the multivariate model together, a set of eight metabolites may separate MS from healthy controls: acetate, creatinine, hippurate, 3-hydroxybutyrate, 3-hydroxyisovalerate, malonate, oxaloacetate and TMAO.
  • NMO-SD versus healthy controls: a stronger model (R²Y = 0.93, Q² = 0.68, CV-ANOVA p = 7.3 × 10⁻³), with 20 metabolites different. Acetate, creatinine, 3-hydroxybutyrate, methylmalonate, oxaloacetate and succinate had the highest contributions.
  • MS versus NMO-SD: creatinine (p = 0.015), 3-hydroxybutyrate (p = 0.008), 3-hydroxyisovalerate (p = 8.4 × 10⁻⁵) and methylmalonate (p = 1.1 × 10⁻⁴) differed.
  • Metabolites specific to MS (SUS-plot analysis) were 2-hydroxyisovalerate, alanine, hippurate, isovalerate and TMAO. Metabolites specific to NMO-SD were acetate, creatine, lactate, methylmalonate and citrate.
  • Top affected pathways: synthesis and degradation of ketone bodies, propionate metabolism, pyruvate metabolism, the TCA cycle, alanine/aspartate/glutamate metabolism, and glycolysis.
  • In MS urine, creatine increased and creatinine decreased. Hippurate, propionate and TMAO are gut-microbiota-related, which the authors say hints at a gut microbiota role in MS, though this is speculative.
  • Limitations: very small groups (7 healthy, 8 MS, 9 NMO-SD), so the findings need validation in larger cohorts. Healthy controls were much younger (mean 26.7 years) than MS (44.6) and NMO-SD (39.3) patients, and patients were on different treatments (azathioprine, interferon, glatiramer, natalizumab, or none). Both could affect urine metabolites. Most participants provided two samples, and no independent test group was used.
  • Some details in the PDF are inconsistent. The conclusions say eight metabolites are associated with NMO-SD but list seven. The oxaloacetate p-value in MS versus healthy is given as 5.0 × 10⁻², which is borderline. Some metabolite names contain typos (“3-hydroxyisovalerae”). Check the original before quoting exact figures.
  • The authors declare no competing financial interest. Funding came from NIH grants, the University of Nebraska Research Council, and Danish sources (Scleroseforeningen, Lundbeckfonden, Odense University Hospital).

Read the full paper
VIEW PDF
Citation

Gebregiworgis T, Nielsen HH, Massilamany C, Gangaplara A, Reddy J, Illes Z, Powers R. A Urinary Metabolic Signature for Multiple Sclerosis and Neuromyelitis Optica. J Proteome Res. 2016;15(2):659–666. doi:10.1021/acs.jproteome.5b01111

Related Research Papers
n