Plasma proteomic analysis is a complicated undertaking due to factors ranging from the dynamic range of protein analytes to the sheer variation in physical properties ofthe individual proteins comprising the human proteome. The examples in this article all illustrate further complications, including aspects inherent to the sample itself, in addition to sample handling, transport, storage, and processing. These variables are all pre-analytical processes, and influence the quality or reproducibility ofthe data, in addition to the myriad of analytical process variables themselves.
The examples cited in this article include many aspects of pre-analytical variation, from the phlebotomy event itself, to the addition and selection of protease inhibitors. These studies were conducted independently to address the many different issues. It is difficult to draw detailed sample handling recommendations that can be applied in broad situations. However, several important outcomes can be derived from this assembly of data. First, it is clear that particular analytical techniques may have certain sample and/or pre-analytical requirements that are transparent to other methods. Thus, it is advisable for researchers to consider all aspects of the sample and its handling, in relation to the analysis that is performed. Although it may be an obvious statement, certain variables may only be discovered to be important in retrospect, or may easily be overlooked entirely by those not familiar with the details of clinical sample acquisition and handling (Tab. 2). The data presented herein suggest that it is important to match the sample type, e.g., selection of plasma versus serum, to the analytical method, whatever that may be. For example, rigorous peptidomic analyses may be incompatible with the use of serum samples. In addition, sample aging during storage can alter analytical data, possibly in unanticipated ways.
Second, and perhaps more important, is the fact that, in the nascent field of proteomics, there are a large number of factors, maybe even currently unidentified, that could alter data or its reproducibility. Our most important collective suggestion is to carefully track every possible variable, and to eliminate as many as possible, during all steps between phlebotomy and the final proteomic data generation. Thus, full annotation of the donor's history information, the sample, and the numerous processes in its handling are recommended. Use of reference materials for calibration, control, traceability, and commutability can help improve reproducibility and strengthen comparisons between experiments and laboratories. Proper specimen storage, including minimizing the exposure of samples to elevated temperatures and for extended periods of time, is critical to maintaining specimen integrity. Protease inhibitors may be protecting plasma proteins, as early as during the phlebotomy process itself, and protease inhibitor use seems likely to provide a more reproducible sample. However, some inhibitors, e.g., AEBSF, have the potential to alter proteins and thus careful consideration of the desired analytical outcome is important. Protease inhibitor presence has also been demonstrated to have a beneficial effect on one analytical method outcome itself (Fig. 6), and thus protection against proteolysis can have benefits beyond merely preserving the original sample proteins. Careful consideration of all aspects of sample handling and the pre-analytical process will strengthen studies, improve reproducibility, and accelerate the growth in the utility of proteomics for further characterization ofthe plasma proteome.
We would like to thank Lynn Echan at the Wistar Institute for performing the 1-D and 2-D gel analyses associated with the stability studies, and Randall Orchekowski at the Van Andel Institute for performing the antibody microarray experiments. A portion of this work was supported by a grant from the HUPO to A.J.R.
Was this article helpful?