Tips and Tricks for Proteomic Sample Preparation

Sample preparation for proteomics involves balancing analyte recovery against the target’s stability and the requirements of the analysis platform, which more often than not is high performance liquid chromatography (HPLC) followed by mass spectrometry.

To help manage that balancing act, Emma Lind, Global Product Manager at Cytiva, offers the following general tips:

• Use procedures that are as gentle as possible, avoiding vigorous tissue disruption
• Perform isolation steps quickly, at sub-ambient temperatures
• Use a buffer that maintains pH/ionic strength and stabilizes the sample
• Add protease inhibitors and/or quenching agents to protect the analyte from ongoing metabolic processes

Cell lysis

For bacterial cultures, sonication is the most popular way to disrupt cell walls from two liters or less of cell culture. “The main challenge here is controlling temperature by keeping the cellular suspension on ice and sonicating in short pulses of between 5 and 10 seconds, with 10 to 30 second pauses,” Lind says. Homogenizers also work but care must be taken to avoid high temperatures and foaming, which inactivate proteins. Enzymatic lysis, which digests the peptidoglycan layer of the bacterial cell wall, works with many bacterial species but Gram-negative bacteria (e.g., E. coli) require a detergent for complete permeabilization of their peptidoglycan layer.

“A lot of DNA is released during lysis, which makes the preparation viscous. DNase digests DNA but can get expensive for large preps,” Lind adds. “You might consider sonication in these circumstances.”

Homogenization also works for yeast, but the homogenizer must operate at much higher pressures than for bacterial preps. Glass bead vortexing is also widely used for yeast, especially for volumes of up to 15 mL. To avoid overheating, Lind recommends agitating cells in several cycles for between 30 and 60 seconds, with cooling between agitations.

“Enzymatic lysis of yeast cells is based on the digestion of the cell wall by an enzyme, with Zymolyase™ and lyticase the most widely used reagents,” Lind adds. “This works at any scale but, as noted, large-scale preparations can get expensive.”

Insect cells lack cell walls, so they lyze rapidly. The most common disruption method uses low-concentration detergent using a tissue grinder, for example a Dounce homogenizer. “If the volume of your cell suspension is larger than 25 mL, consider using an electric homogenizer,” Lind says. “Alternatively, commercial protein extraction kits containing mild, non-ionic detergents work well with small volumes.”

For mammalian cells and secreted proteins, centrifugation or filtration yields a clarified sample suitable for the first chromatography step. However, when the target protein is expressed intracellularly, Lind recommends using a homogenizer (e.g., a French press) for lysis. “Alternatively, you can use commercial protein extraction kits for small volumes of cell suspension.”

DNA/RNA removal and clarification

Nucleic acids released during lysis must be removed to avoid high viscosity and/or interference with subsequent chromatographic steps. Methods include enzymatic digestion with DNase on ice for 10 to 15 minutes, mechanical shearing during sonication (with DNase addition if using a French press), or precipitation with polyethyleneimine or protamine sulfate followed by centrifugation.

Lind recommends standard centrifugation or filtration to clarify samples. “To ensure they’re clear and free of particulates, pass samples through a 0.22 µm or 0.45 µm filter and/or centrifuge immediately before applying them to the column.”

Centrifugation should always be conducted in the cold. For cell lysates, centrifuge at 40,000 to 50,000 × g for 30 minutes, which may be shortened to 10 to 15 minutes if needed. For small sample volumes, centrifuge at the highest available g-force, such as 15,000 × g for 15 minutes in a benchtop centrifuge.

“Filtration is less time-consuming than centrifugation and you can often use syringe filters for relatively large volumes up to 50 to 100 mL,” Lind tells Biocompare. “Use a syringe filter with low protein binding, such as regenerated cellulose, PES, PVDF, or cellulose acetate.

Needles and haystacks

Proteomic sample preparation is confounded by the sheer number of proteins expressed in a typical sample, the proteome’s huge concentration dynamic range (seven to eight orders of magnitude), and expression level differences among various cells and tissues.

Plasma, the starting material for many proteomic workflows, contains thousands of proteins but a few high-abundance species make up 99% of the mass, according to Daniel Lopez-Ferrer, Ph.D., Director of Proteomics and Translational Research at Thermo Fisher Scientific. “Some proteins are expressed in in one cell type, but not another. Skin, for example, expresses structural proteins while heart tissue expresses proteins related to cardiac function.”

The more steps in a prep the greater the opportunity for something to go wrong, especially with manual workflows, where the effects of timing discrepancies and variability in technique are compounded. “You may not notice the consequences until much later on,” Lopez-Ferrer says, “particularly when the experiment relies on LC-MS.” Incomplete removal of salts or detergents, for example, can interfere with the detection stage. “Proteomic sample preparation is a puzzle in which all pieces must fit together perfectly.”

Target enrichment and interference depletion can pick up some of the slack, but which strategy is used depends on the analyte and the matrix. Note that these processes are easy automatable, through pipetting or liquid handling robots, as long as the sample is completely solubilized.

Legacy solubilization methods include the use of urea, detergents, and organic solvents, each with advantages and disadvantages. These reagents should be selected not to affect proteins adversely, be easily removed, and be compatible with LC-MS. Proteomic sample preparation kits, like the Thermo Scientific EasyPep™ MS Sample Prep Kit, reduce the complexity by providing all reagents and a standard protocol. “We try to provide a universal chemistry that extracts proteins from a wide array of samples,” Lopez-Ferrer explains. “Results are consistent because the processes are consistent and the inputs for the LC-MS analysis are consistent.”

Industrialization

The pressure to repurpose bioanalytics to support medical diagnostics, personalized medicine in particular, has been strong but serious challenges exist.

“Although I don’t see pharmacy chains installing MS instruments anytime soon, I think we will get there eventually,” Lopez-Ferrer says. “A 2021 paper from Prof. Akhilesh Pandey demonstrates how clinical laboratories can use our instruments in diagnostic assays. The challenge will be in method standardization and the ability to scale up these assays. We are actually quite close to this idea, as we’ve demonstrated the rapid, automated, simultaneous preparation of hundreds of samples. But, we still have a lot to learn about getting samples into and out of the analysis platform, and how to store, retrieve, and learn from the large quantities of data they generate. Plus, there are always instances in which a simpler assay will suffice. So however, this happens, the drivers will be cost and the democratization or modernization of analysis technology. If MS is involved, the instrumentation will not require the bells and whistles of research instruments, but will be purpose-driven or dedicated to one particular set of assays.”