Proteomics: a novel tool to unravel the patho-physiology of uraemia

Uraemic toxicity results in the dysfunction of many organ systems, provoking an increase in morbidity and mortality. To date, only approximately 90 uraemic retention solutes have been described. To examine unknown uraemic substances thoroughly, the identification of as many compounds as possible in the ultrafiltrate and/or plasma of patients would lead to a less biased definition of the uraemic retention process compared with what is proposed today.

We describe the application of a novel proteomic tool for the identification of a large number of molecules present in ultrafiltrate from uraemic and normal plasma obtained with high- or low-flux membranes. Separation by capillary electrophoresis was coupled on-line to a mass spectrometer, yielding identification of polypeptides based on their molecular weight.

Between 500 and >1000 polypeptides with a molecular weight ranging from 800 to 10,000 Da could be detected in individual samples, and were identified via their mass and their particular migration time in capillary electrophoresis. In ultrafiltrate from uraemic plasma, 1394 polypeptides were detected in the high-flux vs 1046 in the low-flux samples, while in ultrafiltrate from normal plasma, 544 polypeptides vs 490 were found in ultrafiltrate from normal plasma obtained from membranes with comparable cut-off. In addition, polypeptides >5 kDa were virtually only detected in the uraemic ultrafiltrate from the high-flux membrane (n = 28 vs n = 5 with the low-flux membrane). To demonstrate the feasibility of further characterizing the detected molecules, polypeptides present exclusively in uraemic ultrafiltrate were chosen for sequencing analyses. A 950.6 Da polypeptide was identified as a fragment of the salivary proline-rich protein. A 1291.8 Da fragment was derived from alpha-fibrinogen.

The data presented here strongly suggest that the application of proteomic approaches such as capillary electrophoresis and mass spectrometry will result in the identification of many more uraemic solutes than those known at present. This could enable the introduction of more direct elimination strategies, since it is possible to obtain an extended appreciation of the removal capacities of particular dialyser membranes.