Rapid assay development for low input targeted proteomics using a versatile linear ion trap

Abstract

Advances in proteomics and mass spectrometry enable the study of limited cell populations, where high-mass accuracy instruments are typically required. While triple quadrupoles offer fast and sensitive low-mass specificity measurements, these instruments are effectively restricted to targeted proteomics. Linear ion traps (LITs) offer a versatile, cost-effective alternative capable of both targeted and global proteomics. Here, we describe a workflow using a hybrid quadrupole-LIT instrument that rapidly develops targeted proteomics assays from global data-independent acquisition (DIA) measurements without high-mass accuracy. Using an automated software approach for scheduling parallel reaction monitoring assays (PRM), we show consistent quantification across three orders of magnitude in a matched-matrix background. We demonstrate measuring low-level proteins such as transcription factors and cytokines with quantitative linearity below two orders of magnitude in a 1 ng background proteome without requiring stable isotope-labeled standards. From a 1 ng sample, we found clear consistency between proteins in subsets of CD4⁺ and CD8⁺ T cells measured using high dimensional flow cytometry and LIT-based proteomics. Based on these results, we believe hybrid quadrupole-LIT instruments represent a valuable solution to expanding mass spectrometry in a wide variety of laboratory settings.

Fig. 1. Schematic and performance metrics of the Q-LIT instrument.

a The instrument schematic of the Q-LIT MS. Ions enter the first QR5 Plus Segmented Quadrupole Mass Filter with Hyperbolic surface (Q₁) before entering into the Ion Concentrating Routing Multipole (IRM). The IRM behaves as the collision and storage cell. Ions are then moved to the high-pressure cell of the dual-pressure LIT and eventually to the low-pressure cell for mass analysis. b The number of HeLa proteins and peptides detected from 1 to 500 ng inputs analyzed with an Orbitrap (Exploris 480) and LIT (Stellar). Each input level was collected in duplicate. c The intensities of PRTC peptides in a HeLa background diluted in water over >5 orders of magnitude. Intensities were normalized to 500 ng and scaled to 1. The gray dashed line represents a 1 to 1 fitting between the amount analyzed and the intensity acquired.

Fig. 2. Schematic for generating DDA and DIA-based libraries and detection results.

a A schematic for using GPF-DIA to build targeted PRM assays. A variety of library sources are first mined for library entries, then either re-searched using EncyclopeDIA or inferred using direct DIA search engines, constructing a “translation” library of potential PRM targets and their chemical characteristics. Peptides marked for exclusion are removed from the library, while the remaining peptides are sorted according to their experimental relevancy (using the 3rd highest fragment). For this work, we had the software build PRM assays designed for 1, 10, and 100 ng input levels using 10, 20, and 50 peptides per cycle, respectively. b Libraries were generated using either the translation library approach or a more standard DDA method coupled with offline high-pH reverse phase (HPRP) fractionation. c The overlap of potential PRM candidates in the spectral library using HPRP-DDA and the translation library filtered to a 1% peptide-level FDR.

Fig. 3. The figures of merit for PRM curves at 1, 10 and 100 ng of material.

a, b The quantitative accuracy of matrix-matched curves on an ion trap of pooled IL-2 and IL-15 peptides in a background of dimethyl-labeled peptides. We generated three curves loading 100 ng, 10 ng, and 1 ng of material on-column. Box plots show the spread of measured values where the whiskers indicate 5% and 95% points, and the bold line indicates the median measurement. d Each dilution is a different color where colored dashed lines indicate the expected fold change. e, f The distribution of the Figures of Merit for the 1, 10, and 100 ng injections using PRM on the Q-LIT. All boxplots (a–c, e, f) are represented as median value. The box maxima extents to the 1st interquartile range (25th percentile), while the minima extends to the 3rd inner quartile range (75th percentile).

Fig. 4. Three representative peptides that were quantifiable below 1 ng.

a Each row displays a peptide chromatogram at each dilution within the 1 ng curve. Each peptide contains three representative transitions. The first peptide from Granzyme B had the best estimated LoQ at 0.043:1, while the third peptide from IL-2 receptor subunit alpha had an estimated LoQ at 0.132:1 at 1 ng. b LoQ and LoD were estimated on a peptide-by-peptide basis using EncyclopeDIA’s curve fitting algorithm. First, the algorithm determines a maximum line through the noise of the calibration curve and then fits the linear dynamic range. The intersection of both lines is the LoD (shown with a gray shaded, empty circle), while the LoQ (shown at the intersection of the dotted red lines) is three standard deviations of the noise above the LoD. The error associated with the lines fitted through the noise and linear, dynamic range are shown in yellow, and represent 3 standard deviations above and below the median signal for each line.

Fig. 5. A summary of the cell populations in IL-2 and IL-15 stimulated T cells determined by the flow cytometry panel described in Supplementary Table 2.

a The gating procedure was used to determine the relative percentage of each cell type in the IL-2 and IL-15 samples (more details in Supplementary Fig. 7). b The estimated cell populations based on back calculations of the gating results.

Fig. 6. Quantifying immune cell biological replicates at 1 ng.

a Quantitative ratios for the panel proteins assayed in the 10 peptide/cycle PRM. The assay was collected in technical triplicate injections of Day 10 IL-2 and IL-15-stimulated T cell proteomes. The selected panel of proteins is associated with T cell activation, differentiation, or cytokine signaling. No LoQ was determined for CD4 with the 1 ng calibration curve, indicated by a red 5-point star. In the IL-2 stimulated sample, IL2RB was measured below the LoQ determined by the 1 ng calibration curve, indicated by a 6-point star. The 9 data points of each protein were extracted from 3 technical replicate PRM injections for the IL-2 and IL-15 stimulated proteomes by calculating the log2 fold change in all possible combinations using technical replicates. The boxplot centers are represented as median values. For each box, the maxima extents to the 1st interquartile range (25th percentile) and the minima extends to the 3rd inner quartile range (75th percentile. b Coefficient of technical variation (% CV) plots for all peptides quantified in the 1 ng assay. The red dotted line indicates 20% CV on each plot.