Research seeks to both identify and quantify the complete proteome within a cell, tissue, or organism. As the depth of proteome profiling coverage increases, so does our insight into complex physiological processes and their effect on phenotype. The limitations of complex proteome quantitative profiling are linked to high complexity and a wide dynamic range of proteins that require better separation and detection. Here we demonstrate the latest advances in nanoLC-MS technology and data processing pipelines used to enhance peptides separation for the deepest label-free proteome profiling in HeLa cell protein digest.
The next-generation nano-, capillary- and micro-flow UHPLC system was coupled to Orbitrap Exploris 480 HRAM MS via the EASY-Spray interface. HeLa protein digests were separated on EASY-Spray or double nanoViper columns (75 µm x 75 cm, 2 µm) that were used independently or coupled to generate a 1.5 m long column. The separations were performed with flow rates from 200 to 500 nL/min. The temperature of each column was controlled independently to utilize temperature gradients. The data were processed in Proteome Discoverer 2.5 and spectral library search with the ability to resolve “chimera” MS/MS spectra. Results were filtered for 1% FDR on peptide and protein levels.
The increase in column length results in higher efficiency, lower FWHM, and ultimately to higher peak capacity. The 75 μm × 75 cm column outperforms the 75 μm × 50 cm in a 90-min gradient by 2 sec on average at FWHM. We optimized gradient separation for methods with 90, 120, 180, and 240 min peptide elution window using the direct injection and accelerated sample loading and column equilibration at 1500 bar. While PWHM was increasing from 9 sec to 19 sec (×2.1 times) for gradients from 90 to 240 min (×2.7 times) the separation power was also increasing proportionally. We were able to achieve the proteome depth of > 7000 proteins and > 80K peptides for single-shot experiments. The results were reproducible between columns and between two sites where identical nanoLC-MS setups and methods were used. The reprocessing data with a data processing pipeline that resolves “chimera” spectra led to the identification of more than 9100 proteins and 100K peptides with only 1 μg of injection using 240 min gradient method. Interestingly, more than 8300 proteins and 80K peptides could be identified with only 90 min gradient and the same loading amount.