A new method for the quantification of neutrophil and eosinophil cationic proteins in feces: establishment of normal levels and clinical application in patients with inflammatory bowel disease

Am J Gastroenterol. 2002 Jul;97(7):1755-62. doi: 10.1111/j.1572-0241.2002.05837.x.


Objectives: The aims of this study were 1) to develop a valid method for the measurement of the eosinophil proteins eosinophil cationic protein (ECP) and eosinophil protein X (EPX) and neutrophil proteins myeloperoxidase and human neutrophil lipocalin (HNL) in feces and 2) to investigate their potential role as disease activity markers in inflammatory bowel disease (IBD).

Methods: Feces samples were obtained from 44 apparently healthy individuals (HIs), 18 patients with IBD (11 with ulcerative colitis [UC] and seven with Crohn's disease [CD]), and three with collagen colitis. The granulocyte markers were measured using immunoassays in supernatants from processed feces.

Results: ECP, myeloperoxidase, and, to a lesser degree, EPX and HNL were bound to the solid part of feces. However, feces homogenized in an extraction buffer containing the cationic detergent N-cetyl-N,N,N-trimethylammonium bromide allowed an efficient recovery of the proteins (i.e., up to 100-fold increased levels compared to homogenization in saline). All four proteins were stable for at least 7 days at +6 degrees C and at least 3 days at +22 degrees C. The normal fecal geometric mean (95th percentile) levels of ECP, EPX, myeloperoxidase, and HNL were estimated to be, respectively, 1.69 microg/g (6.41), 0.57 microg/g (1.72), 3.54 microg/g (8.77), and 1.97 microg/g (4.91). Markedly increased feces levels of all markers (p < 0.0002), compared to HIs and CD patients, were observed in UC. However, the marker levels in CD patients were significantly increased relative to HIs (p < 0.05 to p < 0.0002). Increased levels of HNL and myeloperoxidase were also observed in the three collagen colitis patients. The discriminative capability between UC patients and HIs was somewhat superior for EPX and myeloperoxidase.

Conclusions: The method described here takes into account the molecular properties of the granule proteins and the heterogeneity in feces consistency, which is a prerequisite for a valid and reproducible measurement of cationic granule proteins. We suggest that EPX and myeloperoxidase, when applied in IBD, are the best eosinophil and neutrophil markers for studying GI inflammation.

Publication types

  • Comparative Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Acute-Phase Proteins*
  • Adolescent
  • Adult
  • Aged
  • Antimicrobial Cationic Peptides
  • Blood Proteins / analysis*
  • Carrier Proteins / analysis*
  • Chemistry Techniques, Analytical / methods
  • Eosinophil Granule Proteins
  • Eosinophil-Derived Neurotoxin
  • Feces / chemistry*
  • Humans
  • Inflammatory Bowel Diseases / metabolism*
  • Lipocalin-2
  • Lipocalins
  • Middle Aged
  • Oncogene Proteins*
  • Peroxidase / analysis*
  • Proto-Oncogene Proteins
  • Reference Values
  • Reproducibility of Results
  • Ribonucleases / analysis*


  • Acute-Phase Proteins
  • Antimicrobial Cationic Peptides
  • Blood Proteins
  • Carrier Proteins
  • Eosinophil Granule Proteins
  • LCN2 protein, human
  • Lipocalin-2
  • Lipocalins
  • Oncogene Proteins
  • Proto-Oncogene Proteins
  • neutrophil basic proteins
  • Peroxidase
  • Eosinophil-Derived Neurotoxin
  • Ribonucleases