Mark Duncan (Presenter)
Biodesix Inc.
Authorship: Ryan Walsh (1), Brian Feild (1), Max Steers (2), and Mark Duncan (2)
(1) Shimadzu Scientific Instruments Corp. (2) Biodesix Inc.
| Short Abstract The cost-effective, sensitive and practical identification and quantification of tear peptides and proteins is critical if we are to exploit the diagnostic potential of tear fluid. In the era of precision medicine, we will require methods that can precisely quantify multiple components simultaneously and deliver sensitive detection. The approach developed for this study is a non-targeted MALDI-MS based method for the analysis of human tear fluid. Our findings demonstrate that sensitive, practical and precise quantification of multiple components is achievable (CV values of 10% or less). This method described allows detection, identification, and quantification of potential biomarkers that relate to eye injury and/or disease. The results of this study further reinforce the view that tear analysis can serve as a diagnostic tool, and illustrate the Clinical potential of MALDI-TOF MS. |
Long Abstract
Introduction
Analysis of human tear reveals it to be a complex biological fluid comprised of components vital to the function, health, and defense of the ocular surface. In addition, there are also tear components arising from the general circulation that can serve as predictive, prognostic, preventive or diagnostic biomarkers. In this era of personalized medicine, measuring changes in tear fluid composition has the potential to deliver insights into crucial pathways relating to both ocular and systemic disease. Tear analysis is of particular interest to us because sampling is convenient and non-invasive, there is a diverse array of components in the fluid, and extraction/work-up is minimal.
Prior studies have employed LC-MS/MS to identify over 1,500 peptides and proteins in tear [1-4], but this approach is not well-suited to use in a routine clinical setting because it is costly, complex and time consuming. Alternatively, we and others have “profiled” the components of tear fluid in a practical and cost-effective way by using matrix assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS). While MALDI-MS profiling does not provide qualitative (structural) information on sample components, it is a robust, routinely adopted and user-friendly analytical platform that can be used to screen, quantify and/or identify components in complex biological samples. This presentation focuses on the ongoing development of a cost effective, high throughput approach to tear analysis that allows: 1) rapid generation of a tear peptide/protein profile by MALDI-MS; 2) precise quantification of specific tear components and 3) structural elucidation/verification by MS/MS analysis. We illustrate the potential of this analytical approach to be used in a routine, high-throughput clinical setting, not only for tear analysis, but also for the analysis of other biological fluids and tissue homogenates.
Methods
Polished glass capillaries were used to collect tear directly from the surface of the eye. Tear samples were frozen until analysis. Samples were then diluted 1:10 with water, mixed with matrix solution (á-cyano-4-hydroxycinnamic acid and sinapinic acid were used) in a 1:1 ratio and spotted (dried droplet method) directly on the MALDI plate. Samples were analyzed on a Shimadzu 7090 MALDI mass spectrometer and both external and internal calibration were employed to ensure optimal mass accuracy. Peaks within the MALDI-MS “profile” were selected for fragmentation to generate MS/MS spectra. All spectra were aligned, baseline corrected, and smoothed prior to analysis. The data files, .mzXML, were imported in MASS++ software (Shimadzu Scientific Instruments Inc.) for de novo sequencing and statistical analysis.
Results
Human tear samples (n=20) were collected and analyzed by MALDI-TOF(/TOF) MS. Peptides and proteins commonly found in tear and identified directly included proline-rich protein 4 (PRP4), polymer immunoglobulin receptor (PIGR), lipocalin, lysozyme, and lactoferrin. A subset of the tear samples was pooled and concentrated, and MS/MS analysis repeated to identify additional components. Multiple peptides, including PRP4 derived-peptides, and other novel peptides, were sequenced by this approach. Internal standards were added to samples to allow precise quantification and with the addition of a reference standard, and inclusion of a calibration curve, absolute quantification was possible (r2>0.999 over 3-4 orders of magnitude).
Conclusions & Discussion
MALDI-MS profiling has failed to gain wide acceptance for three main reasons: first, MALDI-TOF MS is still not widely regarded as a reproducible, quantitative tool; second, the identities of the component peaks is the profile are not readily accessible without recourse to orthogonal analytical strategies (e.g., LC-MS/MS) and third, the low abundance peptide & protein components of a complex sample are not accessible without some sample cleanup. In this study, we demonstrate comprehensive and reproducible coverage of the peptide and protein components of tear following minimal sample clean up. Further, precise quantification is possible, even from a complex mixture, provided attention is directed towards critical experimental and data acquisition parameters - the two most important variables when attempting to achieve reproducible results. Sample/matrix application was focused on obtaining a homogeneous layer of co-crystallized sample/matrix and “hot spots” were minimized by pre-mixing sample and matrix prior to spotting. Laser power, number of averaged profiles and raster pattern were all optimized to give reproducible spectra, while at the same minimizing acquisition time. Further, we illustrate practical MALDI-TOF/TOF peptide and protein identification. We demonstrate how these strategies can be used together to open up additional, practical and powerful opportunities to use MALDI-MS profiling in routine clinical chemistry.
References & Acknowledgements:
1. Li, N., Wang, N., Zheng, J., Liu, X. M., Lever, O. W., & Erickson, P. M. (2005). Characterization of Human Tear Proteome Using Multiple Proteomic Analysis Techniques research articles. Journal of Proteome Research, 4, 2052–2061.
2. de Souza, G.A., Godoy, L.M., & Mann, M. (2006). Identification of 491 proteins in the tear fluid proteome reveals a large number of proteases and protease inhibitors. Genome biology. 7:R72.
3. Hayakawa, E., Landuyt, B., Baggerman, G., Cuyvers, R., Lavigne, R., Luyten, W., & Schoofs, L. (2013). Peptidomic analysis of human reflex tear fluid. Peptides, 42, 63–69.
4. Zhou, L., Zhao, S.Z., Koh, S.K., Chen, L., Vaz, C., Tanavde, V., Li, R.X., & Beuerman R.W. (2012) In-depth analysis of the human tear proteome. Journal of Proteomics, 75:3877-3885.
| 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 |