Exploring the Effect of Urine Matrix Constituents on Recovery of Kidney Damage Markers in Urine Using a Targeted LC-MRM-MS Approach Tirsa van Duijl (Presenter) Leiden University Medical Center >> POSTER (PDF) Presenter Bio: After finishing my Master’s Degree in Bio-Pharmaceutical Sciences at Leiden University I started my PhD at the clinical laboratory of the Leiden University Medical Center (Netherlands). Under supervision of prof. dr. C.M. Cobbaert, I am currently working on the development of an LC-MRM-MS assay for biomarker measurement in urine. My research is focused on the search for kidney injury biomarkers and the translation of biomarkers to the clinical laboratory

Authors: T.T. van DUIJL (1), L.R. RUHAAK (2), N.P.M. SMIT (3), C.M. COBBAERT (4)
Department of Clinical Chemistry and Laboratory Medicine, Leiden University Medical Center, Leiden, The Netherlands

Urine is a challenging and variable matrix for test development using a bottom-up proteomics strategy. We aimed to evaluate the effect of pre-analytical factors and matrix constituents on the quantification of urinary biomarkers TIMP-2 and IGFBP7 using LC-MRM-MS. Tryptic peptides from abundant non-target proteins induce ion suppression in ESI, resulting in reduced assay sensitivity of >400 % in samples with a protein concentration of >0.5 g/L. Non-specific proteolytic degradation of the peptide measurands by urinary proteases affected test recovery. Yet, non-specific degradation of urinary proteins could be prevented by thermal denaturation or by the addition of protease inhibitors. As the protease activity and total protein concentration is variable between patient urine samples, attention should be given to the pre-analysis in order to develop a robust and quantitative LC-MRM-MS test.

Introduction

Urinary levels of tissue inhibitor of metalloprotease 2 (TIMP-2) and insulin-like growth factor binding protein 7 (IGFBP7) are promising biomarkers elevated in patients with acute kidney injury after major surgery (e.g. cardiac surgery) and in patients with sepsis.1,2 These biomarkers are not yet routinely implemented in medical labs. So far, research-use only immunoassays and an FDA-cleared NephroCheck™ test have been developed. We aimed to develop an in house LC-MRM-MS test with high specificity and multiplexing capability. As urine matrices are highly variable among patients and diseases, we here investigated the effect of different urinary protein concentrations and urinary proteases on the recovery of tryptic peptides of TIMP-2 and IGFBP7.

Methods

An Agilent 1290 infinity LC system coupled to an Agilent 6495 QQQ-MS was used to develop a targeted LC-MS method for TIMP-2 and IGFBP7. First, theoretical peptide characteristics based on the protein primary structure in combination with an MS2 scan of recombinant TIMP-2 and IGFBP7 tryptic digest were used to select 3-4 candidate proteotypic peptides.3 Peptide were synthesized in our in-house facility and product-ion scans were used to identify specific fragments. The collision energies were optimized to obtain MRM transitions for each peptide, and the LC gradient was optimized for fast but specific quantitation. Urine samples (150 µL) were spiked with SIL-peptides ([13C6, 15N2]-lysine or [13C6, 15N4]-arginine) (0.1 nM) as internal standard. Upon denaturation, reduction and alkylation, urinary proteins were tryptically digested. Samples were desalted by HLB SPE prior to quantification by LC–MRM-MS. For protease inactivation, the urines were heated to 95 °C in a water bath or a mixture of protease inhibitors (cOmplete™ Protease Inhibitor Cocktail, Roche, with and without EDTA) was added, each immediately at the start of the sample preparation, prior to the addition of SIL peptides. Ion suppression by abundant non-target proteins was studied by adding SIL-peptides (0.025 µM) from TIMP-2 and IGFBP7 just before LC-MS measurement to digests of human albumin at concentrations of 0.5, 1, 2, 4 and 6 g/L. For a preliminary method comparison, urinary TIMP-2 and IGFBP7 concentrations were measured in samples from 20 patients who underwent major surgery (kidney-pancreas transplantation or coronary artery bypass grafting) using both ELISA assays and the preliminary developed LC-MRM-MS method.

Results

Proteotypic peptides for TIMP-2 (IQYEIK, EYLIAGK and FFACcamIK) and IGFBP7 (DNLAIQTR, TELLPGDR, HEVTGWVLVSPLSK and ITVVDALHEIPVK) were selected for LC-MRM-MS based on the theoretical characteristics and the performance in MS. For each peptide, three product ions were identified for quantification and qualification. Peptide recoveries in HLB SPE were 30-98%, dependent on the peptide, but were equal for endogenous and SIL peptides. In one LC-ESI-MRM-MS run, peptides elute within 8 min at a flow rate of 0.3 mL/min. Very large variation (CV >100 %) was observed in the ion abundances of the spiked SIL peptides between urine samples. SIL peptide recovery could be increased up to 8 times, depending on the urine sample, by thermal denaturation of urine samples or the addition of protease inhibitors. In addition, an inverse relationship was observed between the total urinary protein and the MS-signal of endogenous and SIL-peptide. Especially for samples with high protein content (> 0.5 g/L), ion suppression diminished the MS response of the target peptides (10-50 times decrease). When results from urinary TIMP-2 and IGFBP7 concentrations from ELISA were compared to relative responses obtained by LC-MS measurement from the same samples after thermal denaturation, the correlation was poor for urine samples with high protein content (> 0.5 g/L) (R2 = 0.01 - 0.30), but promising for urine samples without proteinuria (R2 = 0.71 - 0.93).

Conclusions & Discussion

Pre-analytical factors and matrix effects were found to be pitfalls in the development of an LC-MRM-MS assay for the quantification of urinary proteins TIMP-2 and IGFBP7. Specifically, non-specific proteases and the urinary protein concentration were shown to influence quantitative results. Using thermal denaturation or the addition of protease inhibitors, we were able to limit non-specific protein degradation during tryptic digestion. However, the integrity of urinary proteins TIMP-2 and IGFBP7 in freshly collected urine, and the optimal collection procedure still is a current topic of study. Increased urinary protein concentrations rapidly diminish test sensitivity, even though SIL-peptide as internal standard corrects for ion suppression in ESI due to co-eluting compounds. Therefore, we are currently evaluating the need to reduce protein complexity by protein depletion or immunoprecipitation to ensure accurate protein quantitation in diseased individuals.

Overall, we here brought to light major issues with regards to pre-analytical conditions for collection and sample preparation of urine. However, we also identified potential solutions, and believe it will be feasible to overcome these hurdles, which is needed to achieve a robust LC-MRM-MS/MS based assay for targeted protein quantification in the clinical laboratory.

1. Kashani, K. et al. Discovery and validation of cell cycle arrest biomarkers in human acute kidney injury. Critical care (London, England) 17, R25, doi:10.1186/cc12503 (2013).

2. Gunnerson, K. J. et al. TIMP2*IGFBP7 biomarker panel accurately predicts acute kidney injury in high-risk surgical patients. J Trauma Acute Care Surg 80, 243-249, doi:10.1097/ta.0000000000000912 (2016).

3. Lange, V., Picotti, P., Domon, B. & Aebersold, R. Selected reaction monitoring for quantitative proteomics: a tutorial. Molecular systems biology 4, 222, doi:10.1038/msb.2008.61 (2008).