Background Loss of skeletal muscle mass in cancer cachexia is recognized as an independent predictor of mortality. Mechanisms involved in this wasting process and parameters for early diagnosis are not yet clearly defined. As skeletal muscle is considered as a secretory organ, the aim of this present experimental work was to characterize the changes in the putative muscle secretome associated with cancer-induced cachexia to gain a better understanding of cellular mechanisms involved and to identify secreted proteins which might reflect this wasting process. Methods We investigated first the changes in the muscle proteome associated with cancer-induced cachexia by using differential label-free proteomic analysis on muscle of the C26 mouse model. The differentially abundant proteins were then submitted to sequential bioinformatic secretomic analysis in order to identify potentially secreted proteins. Multiple reaction monitoring and Western blotting were used to verify the presence of candidate proteins at the circulating level. Finally, we investigated the regulation of the production of these secreted proteins by muscle in vitro and in vivo. Results Our results revealed a dramatic increased muscular production (2-to 25-fold) of several acute phase reactants (APR: haptoglobin, serpina3n, complement C3, serum amyloid A1) which are released in the circulation during C26 cancer cachexia. This observation was confirmed in two other preclinical models of cancer cachexia as well as in cancer patients. The muscular origin of these APR was demonstrated by their increased expression in skeletal muscle and myotubes. Our results showed also that IL-6 plays a major role in the muscular induction of these APR in vivo, while glucocorticoids and pro-inflammatory cytokines stimulate directly their increased expression in muscle cells in vitro. Conclusions Muscle wasting caused by cancer is associated with marked changes in muscle secretome. Our study demonstrates a marked increased production of APR by skeletal muscle in pre-clinical models of cancer cachexia and in cancer patients. Further studies are required to unravel the potential role of these proteins in muscle atrophy and their interest as biomarkers of cancer cachexia.

The plasma of 2 mice models of cachexia (C26 and BaF) and control mice were used to verify the distribution of peptides belonging to some proteins of interest, identified by a label free proteomic study done on muscle and found with a significantly increased in the groups of cachectic mice (C26) compared to control. This verification was done using a specific multiple reaction monitoring experiment using an instrumental system combining one dimension nanoACQUITY M-Class (Waters) on line with a triple quadrupole: Xevo TQ-S-Mass Spectrometer (Waters) equipped with nanoESI source.

In brief, the plasma samples collected for each animal model of cachexia and relevant controls: C26 model (C26 n=5, CT n= 8) and BaF3 model (BaF3 n=7, Ct n=7), were used to evaluate the relative distribution in plasma of selected proteins considered as potentially secreted. Twenty µg of total protein extract of each mouse plasma sample were digested using (Trypsin/Lys-C Mix, Mass Spec Grade, V5072, Promega, Madison, WI, USA) according to the manufacturer’s instructions. Then, 3.5 µg of the resulting peptide mixtures were purified on ZipTip C18 (Thermo Fisher Scientific), dried in a vacuum centrifuge and stored at -20 C°until MRM experiment. A fraction corresponding to 0.675 µg of protein digest was analyzed using a specific MRM methodology targeting the proteins of interest selected.