Article pubs.acs.org/jpr
Variability of Protein and Phosphoprotein Levels in Clinical Tissue Specimens during the Preanalytical Phase Sibylle Gündisch,†,◆ Stefanie Hauck,‡ Hakan Sarioglu,‡ Christina Schott,†,◆ Christian Viertler,§,◆ Marcel Kap,∥,◆ Tibor Schuster,⊥ Bilge Reischauer,†,◆,# Robert Rosenberg,⊗,¶ Cornelis Verhoef,○ Hans-Joerg Mischinger,∇ Peter Riegman,∥,◆ Kurt Zatloukal,§,◆ and Karl-Friedrich Becker*,†,◆ †
Institute of Pathology, Technische Universität München, Trogerstrasse 18, D-81675 Munich, Germany Research Unit for Protein Science, Helmholtz Zentrum München, Ingolstaedter Landstrasse 1, D-85764 Neuherberg, Germany § Institute of Pathology, Medical University of Graz, Auenbruggerplatz 25, A-8036 Graz, Austria ∥ Department of Pathology, Erasmus Medical Center, P.O. Box 2040, 3000 CA Rotterdam, The Netherlands ⊥ Institute of Medical Statistics and Epidemiology, Technische Universität München, Ismaninger Strasse 22, D-81675 Munich, Germany ⊗ Department of Surgery, Klinikum rechts der Isar, Technische Universität München, Ismaninger Strasse 22, D-81675 Munich, Germany ○ Division of Surgical Oncology, Daniel den Hoed Cancer Center, Erasmus Medical Center, P.O. Box 5201, 3008 AE Rotterdam, The Netherlands ∇ Department of Surgery, Division of General Surgery, Medical University of Graz, Auenbruggerplatz 29, A-8036 Graz, Austria ◆ The SPIDIA Consortium, QIAGEN GmbH, QIAGEN Strasse 1, 40724 Hilden, Germany ‡
S Supporting Information *
ABSTRACT: The quality of human tissue specimens can have a significant impact on analytical data sets for biomarker research. The aim of this study was to characterize fluctuations of protein and phosphoprotein levels in human tissue samples during the preanalytical phase. Eleven intestine and 17 liver specimens were surgically resected, aliquoted, and either snap-frozen or fixed in formalin immediately or exposed to different ischemic conditions before preservation. Protein levels in the resultant samples were investigated by reverse phase protein array, Western blot analysis, and liquid chromatography−tandem mass spectrometry. Our data revealed that the degree of sensitivity of proteins and phosphoproteins to delayed preservation varied between different patients and tissue types. For example, up-regulation of phospho-p42/44 MAPK in intestine samples was seen in some patients but not in others. General trends toward up- or down-regulation of most proteins were not evident due to pronounced interpatient variability but signal intensities of only a few proteins, such as cytokeratin 18, were altered from baseline in postresection samples. In contrast, glyceraldehyde 3-phosphate dehydrogenase was found to be stable during periods of cold ischemia. Our study represents a proper approach for studying potential protein fluctuations in tissue specimens for future biomarker development programs. KEYWORDS: tissue, preanalytical phase, protein, biomarker, biobank
■
and standardized.1 In the past few years, however, researchers have increasingly recognized the fact that the preanalytical history of a tissue sample is of high importance in the field of biomarker research. Because of the difficulty in controlling all preanalytical variables extant in a clinical setting, only a few publications address the impact of these variables on tissue specimen quality. In recent key publications, researchers have maintained that tissue is still alive after resection, and that gene
INTRODUCTION
Accurate detection and quantification of biomolecules in blood and tissue is fundamental to the development of personalized medicine and improvement in the quality of human health care. Within this context, a basic prerequisite for scientific progress is that researchers study clinical tissue samples of the highest quality possible. In particular, molecular and proteomic profiles in tissue samples should represent the in vivo state rather than a modified or artifactual state induced by preanalytical variables. Unfortunately, the preanalytical phase of clinical tissue samples and its effect on biomolecules has yet to be systematically studied © 2012 American Chemical Society
Received: June 22, 2012 Published: November 8, 2012 5748
dx.doi.org/10.1021/pr300560y | J. Proteome Res. 2012, 11, 5748−5762
Journal of Proteome Research
Article
ischemia times on tissue quality should be investigated, it is obvious that warm ischemia time is dependent on a number of uncontrollable factors including the nature of the underlying disease and the surgical technique used in the procedure. Although one cannot control warm ischemia time for the purpose of preserving surgical specimen quality, it is an important variable to monitor and analyze its influence on proteins. To study interpatient variability of protein levels, we collected quantities of homogeneous, nonmalignant tissue from patients undergoing intestine and liver resection. We reasoned that the use of nonmalignant tissue avoids the well-known heterogeneity in cellular composition of, for example, tumor tissue.8 Liver tissue is known to be especially homogeneous with respect to cell composition which facilitates comparative studies. As ultimate effectors of cellular function, proteins and phosphoproteins play a fundamental role in the emerging field of personalized medicine. They serve as diagnostic markers used in routine pathology for guiding therapy decisions as well as targets for individualized therapy. Moreover, it is known that the correlation between the presence and quantities of protein and that of corresponding mRNAs in mammalian cells is often poor. The reasons for this include pre- and post-translational processes and the different mRNA and protein half-lives.9 In particular, post-translational modifications such as phosphorylation require careful, systematic investigation in that phosphorylated protein levels indicate the activation status of signal transduction pathways controlled by kinases, the inhibition of which has recently been a focus of new drug development. Cryopreservation is accepted as the gold standard for preserving tissue for molecular diagnostics but is often not feasible in a routine clinical setting. Surgical specimens, therefore, have historically been preserved in formalin and stored as formalin-fixed, paraffin embedded (FFPE) tissues which for decades have been the only available material for histological analysis. Nevertheless, there is still debate concerning the quantitative preservation of post-translational protein modifications in FFPE samples. In recent years, protocols for extraction of proteins from FFPE tissue have been optimized, and researchers have shown that immunoreactive, intact proteins and phosphoproteins can be successfully extracted from FFPE tissues.10−12 Our main goal in conducting this study was to investigate the effect of ischemia on protein levels of tissue specimens and whether these effects were independent of the preservation method used. We sought to obtain statistically relevant results concerning these effects on protein profiles by focusing on only two tissue types but employing several patient specimens per tissue type. To eliminate institutional bias, using the same experimental protocol, three hospitals participated in the study. Reverse phase protein array (RPPA) was applied as a targetspecific approach to analyze single proteins and levels of certain phosphorylated proteins,13,14 and tandem mass spectrometry (LC−MS/MS) as a nontargeted approach to obtain broad insight into the proteome of ischemic samples.15,16 The results of the present study give insights into the interpatient variability as well as the fluctuations of protein profiles in clinical samples during the preanalytical phase.
expression and protein profiles, especially phosphoproteins, change dramatically during ischemia.2−4 A major difference to our study is that previous studies were performed with small numbers of patient samples of a certain tissue rather than with multiple tissue replicates of the same tissue from different patients. Surprisingly, some studies have shown that RNA appears to be stable over long time periods in unfixed surgical specimens,5−7 a finding which seems to contradict current scientific belief concerning one of the most unstable biomolecules. These confusing and often contradictory findings indicate a strong need for further studies that expand upon the findings of earlier work. One of the most important preanalytical factors affecting tissue analysis is the intra- and postsurgical delay to fixation or cryopreservation of clinical tissue samples. This delay comprises both warm and cold ischemia times. Warm ischemia time is defined as the time between vessel ligation and resection of a tissue specimen from the body, while cold ischemia time is defined as the time from resection of the specimen until fixation or cryopreservation (Figure 1). While the impact of both types of
Figure 1. Overview of investigated preanalytical factors and resected specimens (A) A biopsy was taken at the beginning of the surgical resection close to the original first incision site and before vessel ligation, if possible, and another tissue specimen was collected after surgical resection. The time between vessel ligation and surgical resection is defined as warm ischemia. The time between surgical resection and preservation is defined as postresection time. A time course experiment was performed on aliquoted samples stored in a humidified chamber at room temperature in which at specified intervals samples were fixed in formalin or snap-frozen in LN2 (procedural delay to preservation, range 0−360 min). (B) Human nonmalignant intestine (*large intestine n = 8, small intestine n = 3) and liver tissue specimens including biopsies were collected and either snap-frozen (LN2) or formalin-fixed and paraffinembedded (FFPE). Proteins were analyzed by reverse phase protein array (RPPA), tandem mass spectrometry (LC−MS/MS) or Western blot analysis.
■
EXPERIMENTAL SECTION
Tissue Samples and Processing: Intestine Tissues
Nonmalignant human intestine tissue specimens (eight colon and three small intestine specimens from a total of 11 patients) 5749
dx.doi.org/10.1021/pr300560y | J. Proteome Res. 2012, 11, 5748−5762
Journal of Proteome Research
Article
Table 1. Overview Sample Characteristics of Intestine and Liver Specimensa Intestine (n = 11)
Average Liver (n = 17)
Average a
patient ID
B
WI [min]
PRT [min]
1 2 3 4 5 6 7 8 9 10 11
no no no no no no no no no no no
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
no no no no no no yes yes yes yes yes yes yes yes yes yes yes
45 60 80 25 U 5 45 5 3 5 30 30 15 15 45 15 15
