Protein extraction from the sample of interest is the initial phase of any mass spectrometry-based proteomics experiment. Thought should be given to any planned downstream assays, specific needs of proteolysis (LiP-MS, post translational modification enrichments, enzymatic reactions, glycan purification or hydrogen-deuterium exchange experiments) long term project goals (reproducibility, multiple sample types, low abundance samples) as well as to the initial experimental question (coverage of a specific protein, subcellular proteomics, global proteomics, protein-protein interactions or immune or affinity enrichment of a specific classes of modifications.) The 2009 version of Methods in Enzymology: guide to Protein Purification [@isbn:978-0-12-374536-1] serves as a deep dive into how molecular biologists and biochemists traditionally thought about protein extraction. The Protein Protocols handbook [@isbn:978-1-59745-198-7] and the excellent review by Linn [@PMID:19892162] are good sources of general proteomics protocols for a scientist new to the field. Any change in extraction conditions should be expected to create potential changes in downstream results. Be sure to think about and optimize the protein extraction step first and stick with a protocol that works for your needs. If a collaborator is attempting to reproduce your results, make sure they begin with the same extraction protocols.
A common question to proteomics core facilities is, “What is the best buffer for protein extraction?” Unfortunately, there is no one correct answer. For global proteomics experiments where maximizing the number of protein or peptide identifications is a goal, a buffer of neutral pH (50-100 mM PBS, Tris, HEPES, Ammonium Bicarbonate, triethanolamine bicarbonate; pH 7.5-8.5) is used in conjunction with a chaotrope or surfactant to denature and solubilize proteins (e.g., 8 M urea, 6 M guanidine, 5% SDS) [@PMID:16152629; @PMID:20722421]. Often other salts like 50-150 mM NaCl are also added. Complete denaturation of the proteins in the sample in a timely fashion is an advantage as it generally prevents changes to protein status by endogenous proteases, kinases, phosphotases, and other enzymes. If intact protein separations are planned (based on size or isoelectric point) choose a denaturant compatible with those methods, such as SDS[@PMID:31249407]. Compatibility with protease (typically trypsin) and peptide cleanup steps will need to be considered. 8 M urea must be diluted to 2 M or less for trypsin and chymotrypsin digestions, while guanidine and SDS should be removed either through protein precipitation, through filter-assisted sample preparation (FASP), or similar solid phase digestion techniques. Note that some buffers can potentially introduce modifications onto proteins such as carbamylation from urea at high temperatures [@PMID:24161613].
Denaturing conditions will efficiently extract proteins – but they will denature/disrupt most protein-protein interactions. If you are working on an immune- or affinity purification of a specific protein and expect to analyze enzymatic activity, structural features, and/or protein-protein interactions, a non-denaturing lysis buffer should be utilized [@PMID:21364760; @PMID:10504710]. Check the calculated pI and hydrophobicity (the Expasy.org resource ProtParam is useful fo this) for a good idea of starting pH/conductivity, but you may need to perform a stability screen. In general, a good starting point for the buffer will still be close to neutral pH with 50-250 mM NaCl, but specific proteins may require pH as low as 2 or as high as 9 for stable extraction. A low percent of mass spec compatible detergent may also be used. Newer mass spectrometry compatible detergents are also useful for protein extraction and ease of downstream processing – including Rapigest® (Waters), N-octyl-β-glucopyranoside, Azo [@PMID:33232116], PPS silent surfactant [@PMID:21280217], sodium laurate [@PMID:23555778], and sodium deoxycholate[@PMID:17022626]. AVOID the use of tween-20, triton-X, NP-40, and PEGs as these compounds are challenging to remove after digestion [@PMID:29726681].
For non-denaturing buffer conditions, which preserve tertiary and quaternary protein structures, additional additives may not be necessary for successful extraction and to prevent proteolysis or PTM modifications throughout the extraction process. Protease, phosphatase and deubiquitinase inhibitors are optional additives in less denaturing conditions or in experiments focused on specific post-translational modifications. Keep in mind that protease inhibitors may impact digestion conditions and will need to be diluted or removed prior to trypsin addition. For extraction of DNA or RNA binding proteins, addition of a small amount of nuclease or benzonase might be useful for degradation of any bound nucleic acids and result in a more consistent digestion [@PMID:23792921].
One typical lysis buffer is 8 M urea in 100 mM Tris, pH 8.5; the pH based on optimum trypsin activity [@PMID:25664860] Small mammalian cell pellets and exosomes will lyse almost instantly upon addition denaturing buffer. Efficiency of extraction and degradation of nucleic acids can be improved using various sonication methods: 1) probe sonicator with ice; 2) water bath sonicator with ice or cooling; 3) bioruptor® sonication device 4) Adaptive focused acoustics (AFA®) [@PMID:21060726]. Key to these additional lysis techniques are to keep the temperature of the sample from rising significantly which can cause proteins to aggregate or degrade. Some cell types may require additional force for effective lysis (see below). For bacteria with cell wells, lysozyme is often added in the lysis buffer. Any added protein will be present in downstream results, however, so excessive addition of lysozyme is to be avoided unless tagged protein purification will occur.
Although small pieces of soft tissue can often be successfully extracted with the probe and sonication methods described above, larger/harder tissues as well as plants/yeast/fungi are better extracted with some form of additional mechanical force. If proteins are too be extracted from a large amount of sample, such as soil, feces, or other diffuse input, one option is to use a dedicated blender and filter the sample after followed by centrifugation. If samples are smaller, such as tissue, tumors, etc., cryo-homogenization is recommended. The simplest form of this is grinding the sample with liquid nitrogen and a mortar and pestle. Tools such as bead beaters are also used, where the sample is placed in a tube with appropriately sized beads and shaken rapidly. Cryo-mills are chambers where liquid nitrogen is applied around a vessel and large bead or beads. Cryo-fractionators homogenize samples in special bags that are frozen in liquid nitrogen and smashed with various degrees of force [@PMID:34002278]. After homogenization, samples can be sonicated by one of the methods above to fragment DNA and increase solubilization of proteins.
Following protein extraction, samples should be centrifuged (10-14,000 g for 10-30 min depending on sample type) to remove debris prior to calculating protein concentration. The amount of insoluble material remaining should be noted throughout an experiment as a large change may indicate protein extraction issues. Protein concentration can be calculated using a number of assays or tools [@PMID:18429326; @PMID:12703310]; generally UV absorption methods are facile and affordable, such as Bradford or BCA assays. Protein can also be estimated by tryptophan fluorescence, which has the benefit of not consuming sample [@DOI:10.1021/ac504689z]. A nanodrop UV spectrophotometer may be used to measure absorbance at UV280. Consistency in this method is important as each method will have inherent bias and error [@PMID:26342307; @PMID:30234128]. Extraction buffer components will need to be compatible with any assay chosen; alternatively, buffer may be removed (see below) prior to protein concentration calculation.
Typically, disulfide bonds in proteins are reduced and alkylated prior to proteolysis in order to disrupt structures and simplify peptide analysis. This allows better access to all residues during proteolysis and removes the crosslinked peptides created by S-S inter peptide linkages. There are a variety of reagent options for these steps. For reduction, the typical agents used are 5-15 mM concentration of TCEP/DTT/2BME. For the following alkylation step, a slightly higher 10-20mM concentration, alkylating agent such as Chloroacetamide/iodoacetamide, n-ethyl maleimide can be used [PMID:29019370; @PMID:15351294; @PMID:28539326]. It is also possible to alklylate free cysteines with one reagent, reduce di-sulfide bonds (or other Cysteine modifications) and alkylate with a different reagent in order to monitor which residues are linked/modified in a protein [@PMID:32132231; @PMID:28445428; @PMID:23074338]. Reduction reactions are generally carried out in the dark at room temperature to avoid excessive off target alklation.
If extraction must take place in a buffer which is incompatible for efficient proteolysis (check the guidelines for the protease of choice), then protein cleanup should occur prior to digestion. This is generally performed through precipitation of proteins. The most common types are 1) acetone, 2) trichloroacetic acid (TCA), and 3) methanol/chloroform/water [@PMID:14753699; @PMID:19892180]. Proteins are generally insoluble in most pure organic solvents, so cold ethanol or methanol are sometimes used. Pellets should be washed with organic solvent for complete removal especially of detergents. Alternatively, solid phase based digestion methods such as S-trap [@PMID:33750040], FASP [@PMID:19377485; @PMID:30259475], SP3 [@PMID:3117935; @PMID:28948796] and on column/bead can allow for proteins to be applied to a solid phase and buffers removed prior to proteolysis [@PMID: 29754492]. Specialty detergent removal columns exist (Pierce/Thermo Fisher Scientific) but add expense and time consuming steps to the process. Additionally these should be checked for efficiency prior to implementing in a workflow with many samples as avoiding detergent contamination in the LC/MS is very important.
Often you will be given protein extraction conditions from molecular biologists or biochemistry which you will have to make work with downstream mass spectrometry applications. For bottom-up proteomics, the overreaching goal is efficient and consistent extraction and digestion.