Detection of viral proteins in human cells lines by xeno-proteomics: elimination of the last valid excuse for not testing every cellular proteome dataset for viral proteins.

Alexey L Chernobrovkin,Roman A Zubarev

doi:10.1371/journal.pone.0091433

Alexey L Chernobrovkin, Roman A Zubarev

Open Access

https://doi.org/10.1371/journal.pone.0091433

Copy DOI

Journal: PloS one	Publication Date: Mar 11, 2014
Citations: 38	License type: CC BY 4.0

Affiliation: Karolinska Institutet, Science for Life Laboratory

Abstract

Cell cultures used routinely in proteomic experiments may contain proteins from other species because of infection, transfection or just contamination. Since infection or contamination may affect the results of a biological experiment, it is important to test the samples for the presence of “alien” proteins. Usually cells are tested only for the most common infections, and most of the existing tests are targeting specific contaminations. Here we describe a three-step procedure for reliable untargeted detection of viral proteins using proteomics data, and recommend this or similar procedure to be applied to every proteomics dataset submitted for publication.

Highlights

Cell lines are widely used as model systems in biology, especially in cancer research [1]
Whereas the pioneering panel introduced by National Cancer Institute in 1990 contained 60 cancer cell lines (NCI-60), large-scale studies reported nowadays are dealing with panels as large as 639 [2] and 947 [3] cell lines
Data Sets Mass-spectrometric data were provided: deep proteomics of eleven common cell lines [18] by Mann’s group; sixty proteome datasets for NCI-60 cancer cell lines - by Kuster’s group [29]

Summary

Introduction

Cell lines are widely used as model systems in biology, especially in cancer research [1]. The number of available cell lines grows rapidly. One of the main motivations for working with cell lines is their supposedly identical behavior in different labs, provided the experimental conditions are similar or identical. This presumed reliable behavior is critical for reproducibility of scientific experiments, which is the backbone of the scientific method. One of the known issues leading to irreproducible results is the cell line contamination. The German Collection of Microorganisms and Cell Cultures (DSMZ) found that 18% of 252 submitted cell lines were cross-contaminated with more than half of the contamination sources located within only six laboratories. Subsequent testing performed by DSMZ with extended number of cell lines showed that of 598 leukemia–lymphoma cell lines, 187 (31%) were contaminated with mycoplasma and/or a second cell line, with 38 (6%) of cell lines contaminated with both [5]

Methods

Results

Conclusion