Tumors exhibit a variety of strategies to dampen antitumor immune responses. With an aim to identify factors that are secreted from tumor cells, we performed an unbiased mass spectrometry-based secretome analysis in lung cancer cells. Interleukin-6 (IL-6) has been identified as a prominent factor secreted by tumor cells and cancer-associated fibroblasts isolated from cancer patients. Incubation of dendritic cell (DC) cultures with tumor cell supernatants inhibited the production of IL-12p70 in DCs but not the surface expression of other activation markers which is reversed by treatment with IL-6 antibody. Defects in IL-12p70 production in the DCs inhibited the differentiation of Th1 but not Th2 and Th17 cells from naïve CD4+ T cells. We also demonstrate that the classical mitogen-activated protein kinase, ERK5/MAPK7, is required for IL-6 production in tumor cells. Inhibition of ERK5 activity or depletion of ERK5 prevented IL-6 production in tumor cells, which could be exploited for enhancing antitumor immune responses.

The analysis of the secretome provides important information on proteins defining intercellular communication and the recruitment and behavior of cells in specific tissues. Especially in the context of tumors, secretome data can support decisions for diagnosis and therapy. The mass spectrometry–based analysis of cell-conditioned media is widely used for the unbiased characterization of cancer secretomes in vitro. Metabolic labeling using azide-containing amino acid analogs in combination with click chemistry facilitates this type of analysis in the presence of serum, preventing serum starvation-induced effects. The modified amino acid analogs, however, are less efficiently incorporated into newly synthesized proteins and may perturb protein folding. Combining transcriptome and proteome analysis, we elucidate in detail the effects of metabolic labeling with the methionine analog azidohomoalanine (AHA) on gene and protein expression. Our data reveal that 15–39% of the proteins detected in the secretome displayed changes in transcript and protein expression induced by AHA labeling. Gene Ontology (GO) analyses indicate that metabolic labeling using AHA leads to induction of cellular stress and apoptosis-related pathways and provide first insights on how this affects the composition of the secretome on a global scale. Key messages Azide-containing amino acid analogs affect gene expression profiles. Azide-containing amino acid analogs influence cellular proteome. Azidohomoalanine labeling induces cellular stress and apoptotic pathways. Secretome consists of proteins with dysregulated expression profiles.

The analysis of the secretome provides important information on proteins defining intercellular communication and the recruitment and behavior of cells in specific tissues. Especially in the context of tumors, secretome data can support decisions for diagnosis and therapy. The mass spectrometry-based analysis of cell-conditioned media is widely used for the unbiased characterization of cancer secretomes in vitro. Metabolic labeling using azide-containing amino acid analogs in combination with click chemistry facilitates this type of analysis in the presence of serum, preventing serum starvation-induced effects. The modified amino acid analogs, however, are less efficiently incorporated into newly synthesized proteins and may perturb protein folding. Combining transcriptome and proteome analysis, we elucidate in detail the effects of metabolic labeling with the methionine analog azidohomoalanine (AHA) on gene and protein expression. Our data reveal that changes in transcript and protein expression levels of 15% – 39% of the proteins detected in the secretome were induced by AHA labeling. Performing Gene Ontology (GO) analyses, our results reveal that metabolic labeling using AHA leads to induction of cellular stress and apoptosis-related pathways and provide first insights on how this affects the composition of the secretome on a global scale.