Transferrin microheterogeneity in human perilymph.

OBJECTIVES/HYPOTHESIS: Assay for beta2-(asialo-) transferrin has been advocated for use in diagnosis of cerebrospinal fluid (CSF) leak or perilymphatic fistula based on the fact that it is present in these fluids but not in serum. Quantitation of the sensitivity of transferrin assays has not been reported previously. The present study was undertaken to quantify the sensitivity of a microelectrophoretic assay of beta2-transferrin and assess its potential applicability to clinical diagnosis of perilymphatic fistula.

STUDY DESIGN

The initial part of the study was a prospective bench biochemistry assessment of assay sensitivity and reliability. Subsequent application of the assay was a blinded prospective clinical trial.

METHODS

Transferrin is a ubiquitous monomeric glycoprotein consisting of 679 amino acids, two iron-binding sites, and two N-linked complex glycan chains. The N-glycan chains branch in variable degree, carrying from zero to eight sialic acid residues. This variation in sialylation has been termed "microheterogeneity." When both iron-binding sites are saturated, the microheterogeneity of sialic acid content results in isoelectric points ranging from pH 5 to pH 6. Thus these nine transferrin variants can be distinguished by isoelectric focusing. Samples of transferrin solution or body fluids (serum, CSF, and perilymph) were incubated in iron-loading buffer to saturate both iron-binding sites and then subjected to isoelectric focusing (IEF). The separated proteins were immunoprecipitated in the IEF gel and silver stained for visualization. Serial dilutions of pure transferrin solution were used to determine assay sensitivity. Neuraminidase was used to digest sialic acid side chains from pure transferrin in solution, and the reaction product was used as a reference standard for comparison to assay of unknown fluids. Patient inner ear fluid samples obtained during stapedectomy or cochlear implantation were used to assess clinical applicability of the assay.

RESULTS

This microelectrophoretic technique, using only 0.3 microL of iron-loaded sample, was able to consistently detect less than 250 pg of transferrin in solution and separate the different sialylation variants based on their isoelectric points. Assay of patient serum samples clearly demonstrated transferrin microheterogeneity. Assay of CSF consistently showed the predicted beta2-(asialo-) transferrin band. Assay of inner ear fluid samples also demonstrated transferrin microheterogeneity. However, no inner ear fluid samples had detectable levels of beta2-transferrin. Presumably, perilymph sample dilution during iron loading and by admixture with serum, local anesthetic, or middle ear secretions lowered the beta2-transferrin concentration below the detection limen of the assay.

CONCLUSIONS

Microelectrophoretic assay of iron-loaded transferrin can detect as little as 250 pg of protein and can identify microheterogeneity in serum, CSF, and perilymph. However, dilutional effects of sample handling and preparation can lower the beta2-transferrin concentration of inner ear fluid samples below the detection limen of the assay. Thus, depending on the relative amounts of serum and perilymph (or CSF) in a mixed sample, electrophoretic separation of transferrin variants may not be diagnostic.