Automated In-gel Digestion on a Commercial Autosampler Directly Coupled to Nano LC-MS/MS
Poster Aug 01, 2015
Achermann François, Bolliger Reto, Buchs Natasha, Doiron Nicholas, Lagache Braga Sophie, Heller Manfred, Boehm Guenter
We have adapted a manual in-gel digestion protocol, including reduction, alkylation and acid-labile detergent assisted trypsin digestion of proteins, to be performed on a PAL RTC liquid handling – autosampler system (CTC Analytics AG, Switzerland). The system is equipped with park stations for a syringe tool and samples/reagents, a cooled tray stack, an evaporation tool, a heatable incubator with orbital shaking capability, a vortex mixer, a syringe wash station, a centrifuge, and a LC injection valve. The current PAL RTC setup treats batches of 6 gel samples within 7 hours. Peptide extracts in HPLC vials are transferred to a nanoLC-orbitrap XL mass spectrometer (Thermo Scientific, Germany) or can directly be injected by the PAL system. Proteins were identified and quantified with MaxQuant software.
A 12-protein molecular weight standard (Bio-Rad) was separated over the entire width of a 12.5% SDS-PAGE at two concentrations corresponding to 1ng or 5ng of each protein standard present in vertical slices of 1.5mm width cut from the gel. Four batches, including three slices of each protein concentration, were processed by an experienced human operator or on the PAL RTC system during four consecutive days. Protein yields were determined by identification and quantification with MaxQuant software. The low molecular weight proteins (Mr <20 kDa) were not consistently identified from the 1ng samples with both procedures. This is a known problem with the gel-LC-MS/MS approach. Keratin contaminations were very similar with both procedures and protein concentrations, indicating that keratin proteins were introduced into the samples before the in-gel digestion process. One-way Anova test was used to compare MaxQuant derived non-normalized protein intensities of batches and procedures with Tukey's honestly significant difference criterion to evaluate statistical significance (alpha = 0.05). No statistically significant differences in protein yield between batches and between procedures were found. The average coefficients of variation of the 12 standard proteins were 19.4% and 11.8% at 1ng protein, and 5.8% and 6.5% at 5ng for the manual and the automated procedure, respectively. These results indicate that the PAL RTC automated in-gel digestion protocol performs as well as an experienced human operator, with potentially a better reproducibility achieved on the PAL system when dealing with low protein concentrations. The next step is the direct coupling of the PAL system to the nanoLC-MS/MS system, which will enable to operating on a 24/24 hours, 7/7 days schedule.