Cheaper, faster, simpler trypsin digestion for high-throughput targeted protein quantification

Abstract

Introduction: LC-MS-based methods for protein quantification have a stigma of being relatively expensive and low-throughput. This is partly due to the cost and speed of trypsin digestion, which has primarily focused on advancements in research-based biomarker discovery applications that rely on protein/peptide identifications rather than clinical biomarker quantification. However, there is a need for simple, fast, and reproducibly efficient surrogate peptide recovery in clinical biomarker quantification.

Methods: Multiple methodologies were evaluated to enhance tryptic digestion for the analysis of thyroglobulin, a prototypical serum protein biomarker. The main criteria for assessment were the yield and speed of formation of surrogate peptides. Various factors such as different additives, types of trypsin, microwave- and pressure-assisted systems, and enzyme concentration were considered as key variables, in addition to digestion time.

Results: It was observed that digestion additives/denaturants had a significant impact on the speed and yield of digestion for each surrogate peptide. Increasing the concentration of trypsin alone was found to accelerate digestions appreciably for most surrogate peptides, without affecting the yield. However, the use of sequencing-grade trypsins and microwave/pressure-assisted systems did not offer significant advantages over the use of 'standard-grade' TPCK-treated trypsin in combination with a conventional incubator, once digestion time and additive had been optimized.

Conclusion: We have dispelled the notion that trypsin digestion is inherently slow and expensive for targeted quantification of serum proteins. Additionally, we have established a groundwork for experimentation that can pave the way for the creation of efficient trypsin digestion protocols, aiming to optimize yield, speed, and cost. It is our hope that these advancements will promote the wider incorporation of such assays in clinical laboratories.

Keywords: Bottom-up proteomics; Liquid chromatography; Mass spectrometry; Protein quantification; Trypsin.

Fig. 1

The concentration of VIL surrogate peptide was measured during the screening of 9 additives at 2 different concentrations and 3 digestion time points each. The 9 additives tested were urea, guanidine (GnHCl), thiourea, deoxycholate (DOC), CHAPS, trifluoroethanol (TFE), acetonitrile (ACN), methanol (MeOH), and 1-propanol (1-PrpOH). All digestions were performed using TPCK-treated bovine trypsin at an enzyme to protein ratio of 1:200 (w/w). The dotted red line is plotted at 49.1 nmol/L, which represents the cut-off for ‘near-maximum’ yield. This value was defined as 85 % of the maximum yield observed among all conditions for this surrogate peptide. The maximum yield achieved was 57.8 nmol/L, obtained after a digestion time of 20 h with 5 % TFE. It's important to note that this cut-off is not based on the theoretical maximum yield of 160 nmol/L, which is calculated based on the starting amount of protein. Results exceeding the near-maximum cut-off are colored green, indicating success. Results colored orange represent the maximum observed concentration for a specific additive without exceeding the near-maximum cut-off. All results presented are averages from triplicate digestion experiments, and error bars are shown as +/- 1 standard deviation. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 2

The ‘relative yield’ for each surrogate peptide is shown for all tested additives. This was determined by normalizing the highest peptide concentration measured at the three time points for each condition to the maximum concentration yielded across all conditions for that peptide. Results in bold indicate near-maximum yield, defined as > 85 % of the maximum yield for a given peptide. It is important not to confuse this with percent recovery, which was calculated based on a theoretical maximum peptide concentration (160 nmol/L) derived from the initial amount of protein. The maximum percent recovery achieved for each peptide is summarized at the bottom. The speed of digestion for a specific peptide is indicated by superimposed bars. This represents either the minimum time at which near-maximum yield was observed for a given additive concentration or, in cases where near-maximum yield was not attained, the time at which the highest peptide concentration was obtained for the given additive concentration.

Fig. 3

Digestion was conducted using five different trypsins and at eight different time points ranging from 0.5 to 18 h. The same procedure was followed for all digestions, which consisted of 0.2% DOC, an enzyme-to-protein ratio of 1:200 (w/w), and incubation on a Thermomixer. In Figure (A) the concentrations of the VIL, GGA, and EFS surrogate peptides are plotted for each trypsin and time point tested. Each data point represents the mean result of triplicate digestions, with error bars indicating +/- 1SD. The solid line overlaying the data represents the pseudo-first order model for each digestion. In Figure (B), the maximum concentration obtained for each surrogate peptide is derived from the model, along with the digestion time required to achieve maximum yield. The trypsins used in the study were as follows: Trypsin 1 - TPCK-treated bovine trypsin (MilliporeSigma); Trypsin 2 - proteomics-grade dimethylated porcine trypsin (MilliporeSigma); Trypsin 3 - sequencing-grade modified porcine trypsin (Promega); Trypsin 4 - mass spectrometry-grade trypsin gold (Promega); Trypsin 5 - mass spectrometry-grade trypsin/LysC mix (Promega).

Fig. 4

Digestion was conducted in four different ‘enzyme reactors’ at eight different time points. These time points ranged from 0.5 to 18 h for the Thermomixer, 8 min to 4 h for REDS and MARS6, and 4 min to 2 h for the Barocycler. Apart from these variations, all digestions followed the same procedure, which involved 0.2 % DOC and an enzyme:protein ratio of 1:200 (w/w) of TPCK-treated bovine trypsin. (A) The concentrations of the VIL, GGA, and EFS surrogate peptides were measured and plotted for each ‘enzyme reactor’ and time point tested. Each data point represents the mean result of triplicate digestions, with error bars indicating +/-1SD. However, data obtained after > 4 h for the Thermomixer are not shown on the plots to allow for better visualization of the other ‘enzyme reactors’. The solid line overlaid on the data represents the pseudo-first order model of each digestion. (B) The maximum concentration obtained for each surrogate peptide was determined from the model, along with the digestion time required to achieve maximum yield.

Fig. 5

Trypsin digestion was conducted using four different concentrations of trypsin at eight time points ranging from 0.5 to 18 h. All digestions followed the same procedure, which included 0.2% DOC, TPCK-treated bovine trypsin, and incubation on a Thermomixer. (A) Measured concentrations of VIL, GGA, and EFS surrogate peptides are plotted against each trypsin concentration and time point analyzed. Each data point represents the mean value obtained from triplicate digestions, with the error bars indicating +/- 1SD. The solid line displayed in the graph represents the pseudo-first order model applied to each digestion. (B) The maximum concentration achieved for each surrogate peptide, as determined by the model, along with the corresponding digestion duration required to achieve this maximum yield.