Multiplexed MRM with Internal Standards for Cerebrospinal Fluid Candidate Protein Biomarker Quantitation

Abstract

Multiplexed quantitation is essential for discovering, verifying, and validating biomarkers for risk stratification, disease prognostication, and therapeutic monitoring. The most promising strategy for quantifying unverified protein biomarkers in biofluids relies on selected/multiple reaction monitoring (SRM or MRM) technology with isotopically labeled standards employed within a bottom-up proteomic workflow. Since cerebrospinal fluid (CSF) is an important fluid for studying central nervous system (CNS) related diseases, we sought to develop a rapid, antibody- and fractionation-free MRM-based approach with a complex mixture of peptide standards to quantify a highly multiplexed panel of candidate protein biomarkers in human CSF. Development involved peptide transition optimization, denaturation/digestion protocol evaluation, transition interference screening, and protein quantitation via peptide standard curves. The final method exhibited excellent reproducibility (average coefficient of variation of <1% for retention time and <6% for signal) and breadth of quantitation (130 proteins from 311 interference-free peptides) in a single 43-min run. These proteins are of high-to-low abundance with determined concentrations from 118 μg/mL (serum albumin) to 550 pg/mL (apolipoprotein C-I). Overall, the method consists of the most highly multiplexed and broadest panel of candidate protein biomarkers in human CSF reported thus far and is well suited for subsequent verification studies on patient samples.

Multiple Reaction Monitoring Enables Precise Quantification of 97 Proteins in Dried Blood Spots

“…5A shows that the intra-assay precision was excellent as Ͼ97% of peptide targets had a CV Ͻ20% and the average CV ranged between 6.2-7.7% for these six biological samples. These intra-assay precision values are similar to highly multiplexed LC/ MRM-MS assays that we have previously developed for protein quantification, such as in plasma (average 6.9% CV) and in cerebrospinal fluid (average 5.8% CV) (33,55). Fig.…”

“…The average protein concentration fold-change between pairs of peptides was 2.8 and a histogram of these values is included as supplemental S1. Similar discrepancies for protein concentrations calculated from multiple peptides have commonly been reported in biofluids and cell cultures when using peptide-level internal standards (31,33,34,48,49), and it is generally accepted that the main source of divergence is incomplete trypsin digestion (50,51). One potential solution to this fundamental challenge in all bottom-up proteomic experiments is the use of fulllength SIS proteins that could be spiked into the sample to account for digestion efficiency (52).…”

Precise quantitation of 136 urinary proteins by LC/MRM-MS using stable isotope labeled peptides as internal standards for biomarker discovery and/or verification studies

“…In our laboratory, it is our current practice to quantify proteins in a control sample from standard curves using constant NAT and variable SIS peptide concentrations across a minimum of 6 levels, as demonstrated in our recent methods for alternative biofluids (i.e., cerebrospinal fluid [43] and plasma [39]). This approach was used again here with 8 levels (spanning a 9000-fold SIS concentration range) used in the quantitative method development and 6 levels (spanning a 360-fold range) used in the subsequent application to urine from prostate cancer (PC) patients.…”

Current trends in quantitative proteomics – an update

“…This has led to interest in biofluids with less‐invasive collection methods, such as urine and saliva . Although its collection is even more invasive than plasma, cerebrospinal fluid may still be the best option for biomarkers of brain‐related diseases . Multiplexed MRM assays for animal models of human diseases, such as the mouse, have also been developed for use in medical research …”