Simultaneous Detection of Six Isoforms of Tau Protein in Human Cerebrospinal Fluid by Multidimensional Mass Spectrometry-Based Targeted Proteomics.

Abnormal expression of Tau protein can cause the development of Alzheimer's disease (AD). So far, much evidence has demonstrated that Tau has multiple isoforms. These isoforms are suggested to have distinct physiological roles and contribute unequally to the progress of AD. Thus, detection of individual Tau isoforms may be helpful to better understand the link between clinical outcome and Tau status and to further improve AD diagnosis and treatment. However, few studies have been conducted on absolute quantification of Tau isoforms, probably due to high sequence homology and also low abundance of these isoforms in biofluids such as cerebrospinal fluid (CSF). Therefore, mass spectrometry-based targeted proteomics was attempted here. This targeted proteomics approach can principally measure a protein of interest at the surrogate peptide level, yet little has been done to detect protein isoforms, probably due to lack of isoform-specific surrogate peptides in mass spectrometry. In this study, separations in more dimensions were added, including immunoprecipitation (IP) and sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) for sample pretreatment and systems of linear equations for post-lab data extraction. Moreover, the reliability of the approach including IP enrichment, gel separation, and linear algebra algorithms was discussed. As a result, each isoform of Tau protein can be individually detected and quantified. Using IP enrichment, ∼250-fold enhancement of sensitivity was achieved. The ultimate LOQ was 0.50 nM. Finally, this multidimensional mass spectrometry-based targeted proteomics assay was validated and applied to simultaneous quantitative analysis of six Tau isoforms in CSF of AD patients.

Organic contaminant sorption parameters should only be compared across a consistent system of linear functions.

Modeling contaminant sorption data using a linear model is very common; however, the rationale for whether the y-intercept should be constrained or not remains a subject of debate. This article justifies constraining the y-intercept in the linear model to zero. By doing so, one imposes consistency on the system of linear equations, allowing for direct comparison of the sorption coefficients.