Detection of reduced orexin-A/hypocretin-1 and its fragments by orexin-A "gold-standard" radioimmunoassay.

BACKGROUND

Orexin-A/hypocretin-1 level is determined in the cerebrospinal fluid samples as a part of clinical narcolepsy diagnostics utilizing a specific commercial radioimmunoassay (RIA); this assay is also widely used in research of many other conditions. The specificity of RIAs is in general variable, and little has been firmly disclosed about the specificity of this RIA assay. Thus, the validity of many research results obtained using the kit is unclear. At least metabolites of orexin-A have been proposed as potential interfering substances.

METHODS

Since this issue has not been systematically assessed, we decided to investigate it using synthetic variants of orexin-A and -B (intact peptides and peptide fragments as well as reduced orexin-A).

RESULTS

Our synthetic orexin-A bound correspondingly to the orexin-A standard included in the kit while orexin-B did not bind even at 10000-fold higher concentrations. Reduction of the disulfide bridges in orexin-A (giving orexin-A-SS) decreased its binding 25-fold. C-terminal truncation of orexin-A-SS was well tolerated - some of the fragments actually bound better than orexin-A-SS - while N-terminal truncation was not allowed.

CONCLUSIONS

The results demonstrate that the RIA kit is fairly selective for intact orexin-A among the peptides tested. However, this does not as such prove that it measures intact orexin-A in the physiological samples, and further studies including identification of physiological orexin-A metabolites are thus required. We also suggest that the redox milieu of cerebrospinal fluid - that has been suggested to vary in different diseases - may have an impact on what is measured with the kit.