INTRODUCTION: C-peptide is a widely used marker of endogenous insulin secretion. It is used for assessing residual beta cell function and in the diagnostic workup of hypoglycemia. C-peptide is routinely measured in the clinical laboratory by immunoassays, which are sensitive but prone to limitations such as cross-reactivity, between-lot variability, and a lack of concordance across different platforms. LC-MS/MS methods are more specific and can be multiplexed. Previous LC-MS/MS methods developed for serum c-peptide measurement detected intact peptide. Unfortunately, intact peptide is poorly ionized, requiring immunoaffinity extraction or solid phase extraction with two-dimensional chromatography.
OBJECTIVE: The objective of this study was to develop and validate a novel enzyme (Glu-C) digestion-based LC-MS/MS assay for the quantification of serum c-peptide.
METHODS: The internal standard was isotopically labeled at two leucine residues: EAED*LQVGQVELGGGPGAGSLQP*LALEGSLQ. Serum proteins were precipitated using acetonitrile (ACN) and the precipitated proteins were resuspended in 10% ACN in 90mM ammonium bicarbonate (AMBIC). Solid phase extraction was performed, followed by reduction and alkylation using dithiothreitol (DTT) and indoacetamide (IAA) respectively. Glu-C was used to digest c-peptide into smaller peptides and sample injected into LCMS for quantitative analysis of c-peptide. Using a Waters Xevo TQ-MS tandem mass spectrometer coupled with Waters Acquity UPLC, two peptide fragments, EAEDLQVGQVE (EAE1 Peptide; +1 Precursor m/z 1216.5692) and LGGGPGAGSLQPLALE (LGG1 Peptide; +1 Precursor m/z 1436.69), were monitored using MRM in positive ion mode. One peptide (EAE1) was used for quantitation and the other (LGG1) as a confirmatory ion.
RESULTS: The assay was linear between 0.1 to 15 ng/mL using an eleven-point dilution series (0.1-15 ng/mL), which was prepared by diluting a high patient pool (spiked with c-peptide certified reference material) with horse serum (negative serum). Within-batch and between-batch imprecision was 4.6 %CV and 5.8 %CV, respectively, with a total error 7.4 %CV. The limit of quantification was determined to be 0.05 ng/mL. No significant carryover was observed up to 50 ng/mL. Spiking 5 ng/mL c-peptide into patient samples showed 99 % (± 8 %) mean recovery, with <15 % bias from expected mean results.
CONCLUSION: We validated a robust LC-MS/MS assay which employs solid phase extraction and enzyme (Glu-C)-mediated peptide digestion for the quantification of c-peptide in human serum.