Nico Smit (Presenter)
Leiden University Medical Center
Authorship: N.P.M. Smit, L.R. Ruhaak, F. Romijn, Y. Van der Burgt, A. van der Laarse, C.M. Cobbaert
Department of Clinical Chemistry and Laboratory Medicine, Leiden University Medical Center, the Netherlands
| Short Abstract Acute kidney injury (AKI) is an important cause of hospital related morbidity and mortality. Recently two promising urine biomarkers were reported that together offer optimal diagnostic performance for early detection of AKI: Tissue inhibitor of metalloproteinase 2 and insulin-like growth factor binding protein 7. We used quantitative proteomics for detection of the proteotypic peptides of both proteins. Candidate peptides were selected after tryptic digestion of recombinant proteins and conditions for MRM-LCMS were optimized. LOQ for peptides of both proteins were below the expected normal protein concentrations and both proteins could be detected in normal urine pool samples. |
Long Abstract
Introduction
Acute kidney injury (AKI) is an important cause of hospital related morbidity and mortality in case of major surgery and sepsis. Conventional kidney damage biomarkers are albumin and total protein in urine. These biomarkers are elevated when advanced kidney damage occurs, but are not suitable for early detection of AKI, at a stage where damage can still be reversed. Various biomarkers for AKI have been proposed in the last decades that may reflect damage to kidney structures (e.g. tubuli or glomerulus) at an earlier stage [1,2]. Two novel AKI biomarkers, tissue inhibitor of metalloproteinase 2 and insulin-like growth factor binding protein 7 (TIMP-2 and IGFBP7) have been reported recently with early diagnostic sensitivity for AKI (stages 1 -3) and being less confounded by comorbidities as compared to neutrophil gelatinase associated lipocalin (NGAL), kidney injury molecule (KIM-1), interleukin-18 (Il-18) and Liver fatty acid binding protein (L-FABP)[3]. TIMP-2 and IGFBP7 appeared as the top two biomarkers for AKI in a discovery study and showed good promise for early detection of AKI in the subsequent (Sapphire) validation study [3]. Our goal was to add clinical value through early detection of AKI by developing an immunoassay-independent test for detection and quantitation of TIMP-2 and IGFBP7 in urine by using multiple reaction monitoring (MRM) mass spectrometry. In the case of TIMP-2 this implies that concentrations in controls of < 85 pmol/L and for IGFBP7 < 1667 pmol/L should be easily detectable.
Methods
To first assess potential candidate peptides that could be targeted using MRM-LCMS, recombinant proteins of TIMP-2 and IGFBP7 (Novus Biologicals; 405 and 347 nmol/L, respectively) were digested with porcine trypsin (sequencing grade Promega). Presence of peptides for TIMP-2 and IGFBP7 was then investigated by measuring the specific MRM transitions of the top 8 peptides for both proteins in the Peptide Atlas Best SRM Transition (PABST) ranking (http://www.peptideatlas.org/) using an Agilent 1290 Infinity LC coupled to a 6490 QQQ-MS. Based on these results, six peptides for TIMP-2 and four peptides for IGFBP7 were synthesized to confirm peptide identification in the recombinant protein digests. The synthesized peptides were used for further optimization of MS measurement, defining MRM settings with regard to retention time (RT), collision energy(CE), and precursor and product ion (MRM transitions).
For the TIMP-2 peptides 109EYLIAGK115 and 122MHITLCcmDFIVPWDTLSTTQK141 (EYLIA and MHITL) and the IGFBP7 peptides 206TELLPGDR213 and 263ITVVDALHEIPVK275 (TELLP and ITVVD) heavy labelled internal standards were produced and used for linearity measurements (with concentrations up to 1 µmol/L). Proteins in urine were concentrated by ultrafiltration or precipitation with acetone or acetonitrile [4]. Proteins were digested using a standard digestion protocol and subsequently formed peptides were measured by MRM-LCMS as described previously [5]. Urine control (pool) samples were spiked with intermediate and high concentrations (81 and 405 pmol/L recTIMP-2 and 1736 and 3473 pmol/L recIGFBP7, respectively). External calibration was performed with recTIMP-2 in a range between 20 and 4050 pmol/L and for recIGFBP7 between 347 and 69400 pmol/L in pooled urine and in a solution with human serum albumin (40 mg/L).
Results
In digests of recombinant protein, six peptides for TIMP-2 and four for IGFBP7 were detected at either 3+ or 2+ charge of the precursor ions. The best responses were obtained for the peptides EYLIA and MHITL for TIMP-2 and ITVVD and TELLP for IGFBP7. Further optimization of the peptide measurement demonstrated the highest response for the TIMP-2 peptide EYLIA with excellent linearity for peak area (R2 = 0.999) and relative response (RR, peak area corrected for IS peak area; R2 = 0.998) and a LOQ of 8.5 pmol/L. Linearity was also found for the IGFBP7 peptides TELLP (RR; R2 = 0.998) and ITVVD (RR, R2 = 0.984) with LOQ values of 13.0 and 7.7 pmol/L, respectively. Peptides EYLIA, TELLP and ITVVD could be identified in urine protein digests of ultrafiltrated and acetone or acetonitrile precipitated urine samples. In urine spiked with recombinant protein best correlations of the relative response values and spiked protein concentrations were found for the acetonitrile precipitated samples both for the TIMP-2 peptide EYLIA and the IGFBP7 peptide TELLP. Results indicated that the intermediate spiked concentrations of 81 pmol/L TIMP-2 and 1736 pmol/L IGFBP7 can be detected in the urine samples. Using acetonitrile precipitation, linearity was found for the calibration ranges of TIMP-2 and IGFBP7 with the peptide EYLIA and TELLP, respectively.
Although the additional two peptides MHITL and ITVVD to some extent confirmed the results obtained with EYLIA for TIMP-2 and with TELLP for IGFBP7 these peptides were shown to be less suitable, as they exhibited lower responses and sensitivity in the urine protein digests.
Discussion
MS-based urine protein diagnostics has the advantage that urine can be obtained non-invasively in relatively high quantities. In the case of renal diseases the urine proteome may reflect damage to specific parts of the kidney [1]. Nevertheless, for clinical use the quantification of urinary proteins poses a real challenge due to variability of urinary protein concentrations and because the composition is influenced by e.g. diet, medication, disease and exercise [6]. We here demonstrate that MS-based quantification of two kidney damage proteins that possess good diagnostic performance for early stage AKI, is feasible. The best performing proteotypic peptides for TIMP-2 and IGFBP7, EYLIA respectively TELLP, can be detected by MRM mass spectrometry at concentrations below that of the expected values in control urine samples (85 pmol/L and 1667 pmol/L, respectively). Furthermore, these peptides can be identified and accurately detected in urine samples after (acetonitrile) precipitation and tryptic digestion of the proteins. Next to protein precipitation for concentration of the urinary proteins ultrafiltration also showed promise in our preliminary investigations. Absolute and reproducible quantification of the proteins demands further optimization of the method. Results obtained by quantification peptides in this preliminary study require confirmation by additional peptides for both TIMP-2 and IGFBP7. Special attention will be paid to commutability of both internal standards and external calibrators and to metrological traceability of test results. We developed an in house test for acute kidney injury biomarkers in urine. Additional refinement is needed to make the test fit-for-clinical purpose. Subsequently, extensive analytical and clinical validation studies are planned according to the framework of the EFLM Test Evaluation working group [7] to determine test purpose and test role in the AKI clinical pathway.
References & Acknowledgements:
References
[1] Alge, J. L., and Arthur, J. M. (2015) Biomarkers of AKI: a review of mechanistic relevance and potential therapeutic implications, Clin.J.Am.Soc.Nephrol. 10, 147-155.
[2] Bonventre, J. V., Vaidya, V. S., Schmouder, R., Feig, P., and Dieterle, F. (2010) Next-generation biomarkers for detecting kidney toxicity, Nat Biotechnol 28, 436-440.
[3] Kashani, K., Al-Khafaji, A., Ardiles, T., Artigas, A., Bagshaw, S. M., Bell, M., Bihorac, A., Birkhahn, R., Cely, C. M., Chawla, L. S., Davison, D. L., Feldkamp, T., Forni, L. G., Gong, M. N., Gunnerson, K. J., Haase, M., Hackett, J., Honore, P. M., Hoste, E. A., Joannes-Boyau, O., Joannidis, M., Kim, P., Koyner, J. L., Laskowitz, D. T., Lissauer, M. E., Marx, G., McCullough, P. A., Mullaney, S., Ostermann, M., Rimmele, T., Shapiro, N. I., Shaw, A. D., Shi, J., Sprague, A. M., Vincent, J. L., Vinsonneau, C., Wagner, L., Walker, M. G., Wilkerson, R. G., Zacharowski, K., and Kellum, J. A. (2013) Discovery and validation of cell cycle arrest biomarkers in human acute kidney injury, Crit Care 17, R25.
[4] Beretov, J., Wasinger, V. C., Schwartz, P., Graham, P. H., and Li, Y. (2014) A standardized and reproducible urine preparation protocol for cancer biomarkers discovery, Biomark.Cancer 6, 21-27.
[5] van den Broek, I., Romijn, F. P., Nouta, J., van der Laarse, A., Drijfhout, J. W., Smit, N. P., van der Burgt, Y. E., and Cobbaert, C. M. (2016) Automated Multiplex LC-MS/MS Assay for Quantifying Serum Apolipoproteins A-I, B, C-I, C-II, C-III, and E with Qualitative Apolipoprotein E Phenotyping, Clin.Chem. 62, 188-197.
[6] Albalat, A., Mischak, H., and Mullen, W. (2011) Clinical application of urinary proteomics/peptidomics, Expert.Rev.Proteomics 8, 615-629.
[7] Horvath, A. R., Lord, S. J., St John, A., Sandberg, S., Cobbaert, C. M., Lorenz, S., et al. (2014) From biomarkers to medical tests: the changing landscape of test evaluation. Clin Chim Acta 427, 49-57.
| Description | Y/N | Source |
| Grants | no | |
| Salary | no | |
| Board Member | no | |
| Stock | no | |
| Expenses | no |
IP Royalty: no
| Planning to mention or discuss specific products or technology of the company(ies) listed above: | no |