Measurement of Aß1-42 in cerebrospinal fluid is influenced by matrix effects

A beta 142 measurement in CSF is an important biochemical marker for Alzheimer disease (AD). However, our understanding of why this biomarker is predictive and why it is often difficult to measure in a reproducible fashion is still lacking. To study these questions, the concentration of A beta 142 in CSF was compared before and after denaturation with 6M guanidine and reverse-phase HPLC. Measurement of the A beta 142 after denaturation and reverse-phase HPLC demonstrated that considerably more A beta 142 was present in CSF than revealed when assaying non-denatured CSF. A comparison of A beta 142 concentrations before and after HPLC in AD CSF with that in normal controls suggested that matrix interference may affect the differentiation between the diagnostic groups. A similar effect was observed with dilutions of crude CSF. Together, these results suggested that at least part of the mechanism by which low A beta 142 concentrations in CSF function as a biomarker of AD is related to matrix components which preferentially hide a portion of the A beta 142 from detection in AD CSF. In contrast, we show that the association of the APOEe4 allele with lower A beta 142 concentrations in CSF is preserved even after denaturation and HPLC. A similar relationship between the presence of the APOEe4 allele and lower concentrations of A beta 140 was also apparent, thereby generating similar ratios of A beta 142/ A beta 140 across the APOE genotypes. The results from the present study suggested that A beta 142 in CSF functions as a biomarker of AD in tandem with other CSF matrix components that are increased in AD CSF. Further studies are needed to identify which matrix factors (e.g. binding of A beta to proteins) underlie the increased detection of A beta 142 concentrations after denaturation and HPLC. The data also suggested that denaturation and HPLC of CSF may be a useful approach for studies using A beta 142 as a pharmacodynamic marker or in other paradigms where measurement of total non-covalently bound A beta 142 is required.