We investigated the effect of homogenization strategy and protein precipitation on downstream protein quantitation using multiple reaction monitoring mass spectrometry (MRM-MS). Our objective was to develop a workflow capable of processing disparate tissue types with high throughput, minimal variability, and maximum purity. Similar abundances of endogenous proteins were measured in nine different mouse tissues regardless of homogenization method, however protein precipitation had strong positive effects on several targets. Best throughput was achieved by lyophilizing tissues to dryness followed by homogenization via bead-beating without sample buffer. Finally, the effect of tissue perfusion prior to dissection and collection was explored in 20 mouse tissues. MRM-MS showed decreased abundances of blood-related proteins in perfused tissues, however complete removal was not achieved. Concentrations of non-blood proteins were largely unchanged, although significantly higher variances were observed for proteins from perfused lung indicating perfusion may not be suitable for this organ. We present a simple yet effective tissue processing workflow consisting of harvest of fresh non-perfused tissue, novel lyophilization and homogenization by bead-beating, and protein precipitation. This workflow can be applied to a range of mouse tissues with the advantages of simplicity, minimal manual manipulation of samples, use of commonly available equipment, and high sample quality.

To optimize the tissue preparation workflow, nine mouse tissues were homogenized using three methods. Method B combined the tissue sample with 3x 3.2 mm stainless steel beads and 1x phosphate buffered saline (PBS) for bead-beating homogenization using cycles of 1 min shaking followed by 1 min rest on ice. Method L lyophilized tissues to dryness overnight using a freeze drier prior to bead-beating as described in method B. Afterwards 1x PBS was added to resuspend the finely ground tissue. For method M, the sample was ground in liquid nitrogen using a mortar and pestle, and subsequently diluted with 1x PBS. Protein precipitation was performed on one half of the resulting homogenate for each workflow. Perfused and non-perfused samples of twenty mouse tissues were processed by lyophilization and homogenization via bead-beating, followed by protein precipitation. All samples were digested with trypsin and stable isotope-labelled peptides were spiked into digests prior to desalting and concentration. Samples were separated on an Agilent Zorbax Eclipse Plus C18 RRHD column (2.1 x 150 mm, 1.8 μm particles) and MRM-MS was performed using an Agilent 6490/6495 series Triple Quadrupole instrument.

To compare the relative abundances of endogenous peptides across workflow optimization samples, 112 heavy-labelled peptides were spiked into samples prepared by each preparation method: B – bead-beating of frozen tissue in sample buffer; L - lyophilization and bead-beating without buffer; M – grinding with mortar and pestle in liquid nitrogen; protein precipitation using acetone was (+) or was not (-) performed for each sample. Protein concentrations in perfused and non-perfused samples were measured using tissue-specific panels of assays developed following Clinical Proteomic Tumor Analysis Consortium (CPTAC) guidelines. Quantitation was performed for each tissue using an external calibration curve constructed with synthetic light peptides (ranging in concentration from 1 to 1000x assay LLOQ) in digested BSA as a surrogate matrix. Heavy-labelled peptides were added to all samples and standards at 100x LLOQ as the normalizer.