PreprintPDF Available

Slice-PASEF: fragmenting all ions for maximum sensitivity in proteomics

Authors:
Preprints and early-stage research may not have been peer reviewed yet.

Abstract and Figures

We present Slice-PASEF, a novel mass spectrometry technology based on trapped ion mobility separation of ions. Slice-PASEF allows to achieve the theoretical maximum of MS/MS sensitivity and boosts proteomics of low sample amounts. Leveraging Slice-PASEF, we show, for the first time, that comprehensive profiling of single cell-level peptide amounts is possible using ultra-fast microflow chromatography and a general-purpose mass spectrometer, allowing quantification of 1417 proteins from 200 picograms of a HeLa cell peptide standard on an Evosep One LC system coupled to a timsTOF Pro 2, at a 200 samples per day throughput. We implemented a Slice-PASEF module in our DIA-NN data processing software, to make it readily available for the proteomics community.
Content may be subject to copyright.
Slice-PASEF: fragmenting all ions for
maximum sensitivity in proteomics
Lukasz Szyrwiel1,, Ludwig Sinn1,, Markus Ralser1,2, and Vadim Demichev1,
1Department of Biochemistry, Charité – Universitätsmedizin Berlin, Berlin, Germany
2The Wellcome Centre for Human Genetics, Nuffield Department of Medicine, University of Oxford, UK
These authors contributed equally
Correspondence to: markus.ralser@charite.de and vadim.demichev@charite.de
Abstract
We present Slice-PASEF, a novel mass spectrometry technology based on trapped ion mobility
separation of ions. Slice-PASEF allows to achieve the theoretical maximum of MS/MS sensitivity and
boosts proteomics of low sample amounts. Leveraging Slice-PASEF, we show, for the first time, that
comprehensive profiling of single cell-level peptide amounts is possible using ultra-fast microflow
chromatography and a general-purpose mass spectrometer, allowing quantification of 1417 proteins
from 200 picograms of a HeLa cell peptide standard on an Evosep One LC system coupled to a
timsTOF Pro 2, at a 200 samples per day throughput. We implemented a Slice-PASEF module in our
DIA-NN data processing software, to make it readily available for the proteomics community.
.CC-BY 4.0 International licenseperpetuity. It is made available under a
preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
The copyright holder for thisthis version posted October 31, 2022. ; https://doi.org/10.1101/2022.10.31.514544doi: bioRxiv preprint
Introduction
Mass spectrometry-based proteomics has seen a number of technological innovations in recent years,
both at the level of instrumentation and data processing algorithms. These have enabled the
emergence of novel applications with improved proteomic depth, robustness, quantitative accuracy,
acquisition throughput and sensitivity1,2. Large-scale proteomic screens for systems biology
experiments and biomarker discovery in patient cohorts are now streamlined3–12. Furthermore, recent
high-sensitivity applications, such as single-cell proteomics and spatial tissue proteomics, have been
rapidly gaining traction in the field13–18. High sensitivity applications in proteomics are currently
hampered by two key limitations. First, the proteomic depth and quantification precision achieved
from profiling low sample amounts is fold-change lower than that obtained from analysing bulk
samples. Further gains here would thus provide a substantial boost to the ability to derive valuable
biological conclusions or clinical implications. Second, low-sample-amount LC-MS methods benefit
greatly from nanoflow liquid chromatography which facilitates the high sensitivity but comes at the
cost of throughput and higher batch-variability.
Separation of analytes by ion mobility before they enter the quadrupole 1 (Q1) on a trapped ion
mobility time-of-flight (TIMS-TOF) mass spectrometer19 offers both enhanced selectivity and
sensitivity of MS/MS acquisition, in a technology termed parallel accumulation and serial
fragmentation (PASEF), and is one of the approaches that have improved throughput or sensitivity for
a range of novel proteomic applications. The data-dependent acquisition (DDA) method termed
PASEF19,20, as well as a data-independent acquisition (DIA) method termed dia-PASEF21, have been
established for TIMS-capable mass spectrometers. We further increased sensitivity of dia-PASEF by
developing a neural network-enabled data processing workflow for TIMS based on FragPipe and
DIA-NN2, yielding one of the most sensitive discovery proteomics methods to date. Soon after, this
workflow was also demonstrated to enable comprehensive label-free single-cell proteomics using the
timsTOF SCP mass spectrometer22.
In this work, we introduce a novel family of data-independent TIMS-TOF methods, termed Slice-
PASEF. Slice-PASEF is able to maximise the sensitivity through exploiting MS/MS duty cycles up
100%. To achieve this, Slice-PASEF employs continuous slicing of the precursor ion space with
fragmentation of all ions in each slice. We show that the new method significantly increases the
proteomic depth when analysing low sample amounts, compared to dia-PASEF. For instance, in the
analysis of 10ng of a K562 peptide standard, Slice-PASEF increases precursor ion identification by
85% and increases the number of precisely quantified proteins 3.4-fold. Benchmarking the new
method using a microflow LC system (Evosep One23) operated at 200 SPD (samples per day)
throughput, we identify and quantify 1417 proteins from 200 picograms of a HeLa peptide
preparation, using a timsTOF Pro 2 mass spectrometer, paving the way for robust and accessible
large-scale single-cell proteomics. Slice-PASEF can be readily deployed using the standard
instrument control software. We also integrated a Slice-PASEF module in our easy to use automated
DIA-NN software suite24 for streamlined application in any proteomics laboratory.
.CC-BY 4.0 International licenseperpetuity. It is made available under a
preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
The copyright holder for thisthis version posted October 31, 2022. ; https://doi.org/10.1101/2022.10.31.514544doi: bioRxiv preprint
Results
Slice-PASEF: sliced fragmentation of precursor ions
A trapped ion mobility spectrometry (TIMS) system accumulates incoming ions and subsequently
releases them - gradually and depending on their mobility in a gas (represented by the ‘1/K0’ value).
The released ions can be analysed directly (MS1 acquisition) or first be subjected to fragmentation. In
the latter case, the acquisition is referred to as PASEF19 (Parallel Accumulation SErial
Fragmentation). Peptide ions with z = 2 or 3 and a similar ion mobility have a limited spread in the
m/z dimension, with an interquartile m/z range of approximately 50 m/z. This allows tuning the Q1
quadrupole to match with the release of precursor ions from the TIMS device, such that the Q1
isolation window, used to select precursor ions for fragmentation, changes depending on their ion
mobility. The set of Q1 isolation windows corresponding to the release of all accumulated precursors
from the TIMS device is referred to as ‘frame’, and different frames can feature different Q1
windows, constituting an acquisition scheme which can either be data-dependent or data-independent.
Previously, the dia-PASEF technology has been introduced, wherein each frame features a low
number (typically 2 to 5) of predefined, non-overlapping Q1 windows21.
Conversely, in Slice-PASEF, which consists of a family of related methods (Figure 1), the precursor
ions are sampled by splitting ‘diagonally’ the m/z * 1/K0 space into in general continuous slices and
‘scanning’ each of these slices using a high number of Q1 windows (10 - 15), independent of each
other in the m/z dimension, with each window corresponding to a narrow 1/K0 range (0.03 - 0.045).
The method is different in nature from the Scanning SWATH technology we introduced previously6
as well as from the recently proposed implementation of the scanning quadrupole concept on TIMS-
TOF mass spectrometers25, since in the case of Slice-PASEF each fragment ion signal is captured with
a single Q1 isolation window.
In this work, we describe and benchmark three types of Slice-PASEF schemes (Figure 1). First, a ‘1-
frame’ (1F) scheme, wherein a single frame is used to fragment all individual peptides with charge
state 2 or 3, achieving 100% MS/MS duty cycle and theoretically maximal MS/MS sensitivity.
Second, a ‘2-frame’ (2F) family of methods, wherein each cycle is formed by a pair of frames, with
the m/z boundary between these being varied across cycles. 2F methods achieve a 50% duty cycle.
Third, a multi-frame (MF) family of methods, wherein the precursor ion space is sliced into multiple
frames, with m/z boundaries between these being varied across cycles. Herein we benchmark a ‘4-
frame’ (4F) MF method, which has a 25% duty cycle.
.CC-BY 4.0 International licenseperpetuity. It is made available under a
preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
The copyright holder for thisthis version posted October 31, 2022. ; https://doi.org/10.1101/2022.10.31.514544doi: bioRxiv preprint
Fig. 1. Visualisation of Slice-PASEF methods. a. 1-frame (1F). b. 2-frame (2F). c. 4-frame (4F). Each
individual plot indicates a method cycle. Each cycle is preceded by an MS1 scan. Within each cycle, the
precursor ion space is fragmented in ‘diagonal’ slices, with each slice corresponding to a single PASEF frame
and consisting of a number of isolation windows, which can overlap in the m/z dimension. Different slices are
highlighted with different colours.
Slice-PASEF maximises sensitivity for the analysis of low sample amounts
To comprehensively benchmark Slice-PASEF against dia-PASEF, we assessed its performance on a
dilution series of a commercial tryptic digest standard produced from the K562 human myelogenous
leukaemia cell line, acquired in triplicates. We chose an analytical flow (500 μl/min) platform (1290
Infinity II LC, Agilent) coupled to a first-generation timsTOF Pro instrument (Bruker) and operated
with a 5-minute chromatographic gradient. While analytical flow rate chromatography is not the
system of choice for sensitive proteomics due to the high sample dilution, it is a convenient choice for
conducting comparative benchmarks of acquisition methods and has the advantage of highly
reproducible chromatography and high ion spray stability4. We compared Slice-PASEF to an 8-frame
dia-PASEF scheme featuring 25 Da isolation windows and 12.5% MS/MS duty cycle, which we have
optimised for this analytical flow platform. We note that this scheme is similar in its characteristics to
the scheme (8-frame, 25 Da, 12.5% duty cycle) proposed by Brunner et al for single-cell dia-
PASEF22.
We observed that Slice-PASEF yields a substantial increase in protein identification numbers
specifically with low injection amounts, particularly when using the high duty cycle schemes (Figure
2). For example, analysing 10ng K562 digest on the analytical flow setup, precursor numbers were
increased by 85% and protein numbers by 52% using the 1F Slice-PASEF scheme, in comparison to
dia-PASEF. As expected for a high-sensitivity method, the better coverage was achieved via the
.CC-BY 4.0 International licenseperpetuity. It is made available under a
preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
The copyright holder for thisthis version posted October 31, 2022. ; https://doi.org/10.1101/2022.10.31.514544doi: bioRxiv preprint
identification of low-abundant peptides, missed by dia-PASEF (Figure 2c, left). For the jointly
detected peptides, 1F recorded a higher signal (Figure 2c, right), with an average 5.7-times signal
boost.
Further, all the Slice-PASEF methods demonstrated more precise quantification, at all injection
amounts. For example, the 1F method was able to precisely quantify (coefficient of variation (CV) <
10%) 3.4-times more proteins than dia-PASEF from 10ng acquisitions.
Fig. 2. Slice-PASEF increases sensitivity and quantitative precision in proteomic experiments with low
sample amounts. a. Numbers of protein groups (upper panel) and precursors (lower panel) identified and
quantified from different injection amounts of a K562 tryptic digest analysed in triplicates with a 5-minute 500
μl/min analytical flow gradient on Agilent 1290 II coupled to Bruker timsTOF Pro. b. Quantification precision,
expressed as coefficient of variation (CV) distributions for precursor quantities in the 10ng and 100ng
acquisitions. Median values for the 10ng acquisitions are indicated. c. Left: the distribution of log2-transformed
precursor intensities in 1F 10ng acquisitions, with identifications unique to 1F and shared with dia-PASEF
highlighted. Right: the distribution of log2-transformed intensity ratios for the shared precursors.
Finally, to validate the precursor quantities obtained for ultra low injection amounts of the peptide
preparations (10ng), we plotted the respective log2-transformed quantities against the quantities of the
same precursors obtained from the 100ng acquisitions used as a reference (Figure 3). In addition to a
significantly higher number of precursors detected from 10ng of the standard, the 1F method also
showed higher correlation between 10ng and 100ng sample quantities, indicative of better accuracy.
In fact, even the correlation between 1F 10ng and dia-PASEF 100ng was higher than between 10ng
and 100ng both acquired in dia-PASEF mode.
.CC-BY 4.0 International licenseperpetuity. It is made available under a
preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
The copyright holder for thisthis version posted October 31, 2022. ; https://doi.org/10.1101/2022.10.31.514544doi: bioRxiv preprint
Fig. 3. Quantitative similarity between methods and injection amounts. Normalised log2-transformed
intensities were plotted for precursor identifications shared between method and injection amount combinations,
recorded for a K562 tryptic digest with a 5-minute 500 μl/min analytical flow gradient on Agilent 1290 II
coupled to Bruker timsTOF Pro. In each case, a median was taken across three technical replicates. a.
Comparison of 1F 100 ng and 1F 10ng acquisitions. b. Comparison of dia-PASEF 100ng and dia-PASEF 10ng
acquisitions. c. Comparison of dia-PASEF 100ng and 1F 10ng acquisitions.
Slice-PASEF facilitates high-throughput microflow analysis of single cell-level amounts
Recent progress in mass spectrometry instrumentation and data processing methods has led to a
rapidly increasing interest in novel technologies and applications that involve proteomic profiling of
ultra-low peptide amounts. These facilitate, for example, single-cell proteomics or spatial proteomic
profiling of tissues, which can offer unique biological and biomedical insights13–18,26–29. Still, so far
these methods have inevitably required peptide separation using nanoflow gradients to achieve the
required sensitivity, including, in some cases, flow rates below 100 nl/min. While highly sensitive,
such setups have a number of limitations. First, they tend to have limited throughput. Second, in
comparison to microflow setups, nanoflow setups are more prone to column clogging and emitter
damage, and in general are more difficult to achieve reproducible chromatography on while requiring
sophisticated operation and maintenance. Aiming at scaling up the throughput of sensitive proteomics
applications, we speculated that the sensitivity of Slice-PASEF can allow to benefit from the greater
.CC-BY 4.0 International licenseperpetuity. It is made available under a
preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
The copyright holder for thisthis version posted October 31, 2022. ; https://doi.org/10.1101/2022.10.31.514544doi: bioRxiv preprint
robustness and throughput of microflow chromatography30,31, while still achieving comprehensive
identification and quantification performance when analysing ultra-low peptide amounts.
Previously, acquisition on timsTOF series mass spectrometers coupled to data processing with DIA-
NN has been shown to be suitable for both label-free22 and multiplexed27 single-cell proteomics.
Brunner et al22 demonstrated the capability of the Evosep One preformed gradient system, operated in
a low-nanoflow (100 nl/min) mode, for deep proteomic profiling of single cells on a timsTOF SCP
mass spectrometer, which is specifically optimised for the measurement of low sample amounts and
has significantly higher sensitivity in this setting than timsTOF Pro. The same Evosep One system can
also run microflow gradients at the throughput of hundreds of samples per day, and here we aimed to
explore this capability for the measurement of ultra-low peptide amounts using Slice-PASEF.
Specifically, we tested the 200 SPD (200 samples per day) Evosep method (2 μl/min flow) using the
general-purpose timsTOF Pro 2 mass spectrometer.
To benchmark this setup, we analysed 0.2ng and 1ng of a HeLa cell line tryptic digest (Figure 4).
With the 0.2ng injections, which roughly correspond to a peptide amount per single HeLa cell32, we
quantified, on average, 4840 precursors and 1417 proteins. The respective median CV was 13.8% on
the protein level, from four replicate injections. Further, the quantities obtained from 0.2ng were
similar to those obtained from 1ng, for jointly identified precursors, with the Pearson correlation
between the log2-transformed quantities being 0.95.
Fig. 4. Slice-PASEF combined with fast microflow chromatography. a. Numbers of protein groups (upper
panel) and precursors (lower panel) identified and quantified from 200pg of a Hela tryptic standard analysed in
4 replicates using a 200 samples per day 2 μl/min method on Evosep One coupled to Bruker timsTOF Pro 2. b.
CV distributions for protein quantities, median values are indicated. c. Normalised log2-transformed intensities
plotted for precursor identifications shared between 200pg and 1ng acquisitions. In each case, a median was
taken across four technical replicates.
.CC-BY 4.0 International licenseperpetuity. It is made available under a
preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
The copyright holder for thisthis version posted October 31, 2022. ; https://doi.org/10.1101/2022.10.31.514544doi: bioRxiv preprint
Discussion
We describe Slice-PASEF, a family of data-independent proteomic methods utilising trapped ion
mobility separation. In Slice-PASEF, precursor ion space is split into slices for fragmentation. Slice-
PASEF allows for high flexibility in method design. First, Slice-PASEF allows for arbitrarily high or
low MS/MS duty cycles, thus balancing sensitivity and selectivity. This includes enabling 100%
MS/MS duty cycle by fragmenting almost all individual charge 2 or 3 peptide ions released by the
TIMS device, i.e. achieving the theoretical maximum of MS/MS sensitivity. Second, Slice-PASEF
allows to tailor the Q1 isolation window boundaries to achieve optimal separation of peptides in each
1/K0 range bin. Third, Slice-PASEF allows to vary the isolation window boundaries in different DIA
cycles in an arbitrary way, benefiting mass deconvolution of MS/MS signals in the DIA-NN software,
improving selectivity. Fourth, any of the Slice-PASEF frames can be repeated a number of times (up
to 5 times, in the methods described herein), with the signals from these repeats being merged by the
DIA-NN software. This can be done to either boost sensitivity or to fit MS cycle time to a specific LC
gradient.
We report that Slice-PASEF significantly outperforms dia-PASEF for proteome profiling of low
sample amounts, both in terms of identification and quantification performance. To our knowledge,
Slice-PASEF is hence currently the most sensitive approach for discovery proteomics. For instance,
Slice-PASEF facilitated proteomics of single cell-level peptide amounts using high-throughput
microflow chromatography, paving the way for large-scale single-cell experiments and other high-
sensitivity applications, such as spatial profiling of tissues. Slice-PASEF works using the production
acquisition software. Importantly, we have incorporated a Slice-PASEF module into DIA-NN24, a
universal DIA data processing software suite, to make this technology broadly accessible to the
proteomics community.
In this work, we tested Slice-PASEF on a readily available LC-MS platform - the Evosep One 200
SPD setup coupled to a timsTOF Pro 2. Naturally, we would expect a further gain in sensitivity if
using dedicated instruments, such as timsTOF SCP22. Further, the Evosep One system also features
even faster methods, such as 300 SPD, which can likewise be coupled to Slice-PASEF. In addition,
the Slice-PASEF module in DIA-NN is compatible with plexDIA, and hence throughput can be
further tripled by using multiplexing with mTRAQ labels, as we have described previously for regular
DIA27. So far, we have benchmarked only some of the potential Slice-PASEF methods. The concept
of Slice-PASEF is highly flexible, i.e. in each DIA cycle a different slicing approach can be used to
fragment precursor ions. For example, one can devise a scheme which combines 1-frame
fragmentation for even cycles and multi-frame fragmentation for odd cycles, thus achieving both, a
>50% overall MS/MS duty cycle, as well as the selectivity in the m/z dimension afforded by the
multi-frame methods.
.CC-BY 4.0 International licenseperpetuity. It is made available under a
preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
The copyright holder for thisthis version posted October 31, 2022. ; https://doi.org/10.1101/2022.10.31.514544doi: bioRxiv preprint
Methods
LC-MS
Both analysed samples, a HeLa tryptic digest (prepared as described previously2) and a K562 tryptic
digest (V6951, Promega) were diluted in 0.1% FA.
For analytical flow proteomics we used the Agilent 1290 Infinity II liquid chromatography system
coupled to the Bruker timsTOF Pro mass spectrometer equipped with the VIP-HESI source (3000 V
of Capillary voltage, 10.0 l/min of Dry gas and temperature 280 °C, probe gas flow 4.8 l/min and
temperature 450 °C). The peptide separation was performed on a Luna Omega 1.6 μm C18 100 Å 30
x 2.1 mm column at 60°C using a linear gradient ramping from 3% B to 36% B in 5 minutes (Buffer
A: 0.1% FA; Buffer B: ACN/0.1% FA) with a flow rate of 500 μl/min. The column was washed using
an increase to 80% B in 0.5 min and a flow rate of 850 μl/min, maintained for another 0.2 min. In the
next 0.1 min, the B proportion was changed to 3% and flow was reduced to 600 μl/min after 1.2 min
and 500 μl/min after 0.3 min. For the dia-PASEF method, the MS/MS precursor mass range was m/z
401 to 1226 and 1/ 0 0.72 to 1.29, with 33 x 25 Th windows with ramp and accumulation time 72 ms𝐾
and cell cycle estimate 0.7 s. For the 1F, 2F and 4F Slice-PASEF methods the ramp and accumulation
time were 100 ms and the windows setup was chosen as presented on Figure 1 (method definition
files are available at https://osf.io/t2ymc/?view_only=7462fffb20e648fc83afc75d8c67e9f8). The m/z
range was 400 to 1000 and the 1/ 0 range 0.75 to 1.2. All methods were used in the high sensitivity𝐾
mode of the mass spectrometer.
For the 200 SPD method, the Evosep One system was coupled with the Bruker timsTOF Pro 2 mass
spectrometer equipped with the Bruker Captive Spray source. The Endurance Column 4 cm x 150 µm
ID, 1.9 µm beads (EV1107, Evosep) was connected to a Captive Spray emitter (ZDV) with a diameter
20 µm (1865710, Bruker). The source parameters were kept as in standard methods offered by Bruker
(Capillary voltage 1400 V, Dry Gas 3.0 l/min and Dry Temp 180 °C). The Evotips were loaded and
maintained following the protocol by the manufacturer.
The Slice-PASEF and dia-PASEF methods were set up in the Bruker timsControl software (v3.0.0,
analytical flow setup, and v1.1.19 68, Evosep One), by importing a text table method definition file
containing the isolation window specification.
Raw data processing
The data were processed using DIA-NN 1.8.2 beta 11, which is available, along with the DIA-NN
pipeline that specifies all the settings used to process the data sets described in this manuscript, at
https://osf.io/t2ymc/?view_only=7462fffb20e648fc83afc75d8c67e9f8. Briefly, the mass accuracies
were fixed to 15ppm (both MS1 and MS2), and the scan window was set to 6 (analytical flow) or 7
(Evosep) analyses. Protein inference was disabled, to use the protein grouping already present in the
spectral library. The spectral library2 that was used to analyse the HeLa 0.2ng and 1ng acquisitions on
the Evosep One system was first refined using an analysis of 5ng HeLa acquisitions, with the Library
generation strategy set to IDs, RT & IM profiling. The --tims-scan option was supplied to DIA-NN
for the analysis of Slice-PASEF acquisitions. Acquisitions corresponding to each combination of the
method and the injection amount were analysed separately, and for protein-level benchmarks, the
.CC-BY 4.0 International licenseperpetuity. It is made available under a
preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
The copyright holder for thisthis version posted October 31, 2022. ; https://doi.org/10.1101/2022.10.31.514544doi: bioRxiv preprint
output was filtered using 1% global protein q-value. For direct comparison of precursor quantities
between different methods and injection amounts, joint analyses of the respective acquisitions were
performed. In the case of the analytical flow acquisitions, DIA-NN was then supplied with the --
restrict-fr and --no-fr-selection commands, which ensured that it used the same fragments to quantify
precursor ions in different runs.
Data availability
Slice-PASEF and dia-PASEF mass spectrometry acquisitions have been deposited to the OSF
repository https://osf.io/t2ymc/?view_only=7462fffb20e648fc83afc75d8c67e9f8. The Slice-PASEF
methods, the DIA-NN 1.8.2 beta 11 setup file, the DIA-NN output reports, the DIA-NN pipeline file
and the spectral libraries have likewise been deposited by the same link.
Acknowledgements
This work is funded by the German Ministry of Education and Research (BMBF), as part of the
National Research Node “Mass spectrometry in Systems Medicine” (MSCoreSys), under grant
agreements 161L0221 (to V.D.) and 031L0220 (to M.R.).
References
1. Ludwig, C. et al. Data-independent acquisition-based SWATH-MS for quantitative proteomics: a
tutorial. Mol. Syst. Biol. 14, e8126 (2018).
2. Demichev, V. et al. dia-PASEF data analysis using FragPipe and DIA-NN for deep proteomics
of low sample amounts. Nat. Commun. 13, 3944 (2022).
3. Bruderer, R. et al. Analysis of 1508 Plasma Samples by Capillary-Flow Data-Independent
Acquisition Profiles Proteomics of Weight Loss and Maintenance. Mol. Cell. Proteomics 18,
1242–1254 (2019).
4. Messner, C. B. et al. Ultra-High-Throughput Clinical Proteomics Reveals Classifiers of COVID-
19 Infection. Cell Syst 11, 11–24.e4 (2020).
5. Demichev, V. et al. A time-resolved proteomic and prognostic map of COVID-19. Cell Syst 12,
780–794.e7 (2021).
6. Messner, C. B. et al. Ultra-fast proteomics with Scanning SWATH. Nat. Biotechnol. 1–9 (2021).
.CC-BY 4.0 International licenseperpetuity. It is made available under a
preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
The copyright holder for thisthis version posted October 31, 2022. ; https://doi.org/10.1101/2022.10.31.514544doi: bioRxiv preprint
7. Slavov, N. Increasing proteomics throughput. Nature biotechnology vol. 39 809–810 (2021).
8. Geyer, P. E. et al. High-resolution serum proteome trajectories in COVID-19 reveal patient-
specific seroconversion. EMBO Mol. Med. 13, e14167 (2021).
9. Niu, L., Thiele, M., Geyer, P. & Rasmussen, D. N. A derivation and validation study of paired
liver and plasma proteomics in 659 individuals reveals circulating biomarkers for alcohol-related
liver disease. J. At. Mol. Phys.
10. Messner, C. B. et al. The Proteomic Landscape of Genome-Wide Genetic Perturbations. bioRxiv
2022.05.17.492318 (2022) doi:10.1101/2022.05.17.492318.
11. Muenzner, J. et al. The natural diversity of the yeast proteome reveals chromosome-wide dosage
compensation in aneuploids. bioRxiv 2022.04.06.487392 (2022) doi:10.1101/2022.04.06.487392.
12. Gonçalves, E. et al. Pan-cancer proteomic map of 949 human cell lines. Cancer Cell 40, 835–
849.e8 (2022).
13. Slavov, N. Driving Single Cell Proteomics Forward with Innovation. J. Proteome Res. 20, 4915–
4918 (2021).
14. Mund, A., Brunner, A.-D. & Mann, M. Unbiased spatial proteomics with single-cell resolution in
tissues. Mol. Cell 82, 2335–2349 (2022).
15. Mund, A. et al. Deep Visual Proteomics defines single-cell identity and heterogeneity. Nat.
Biotechnol. 40, 1231–1240 (2022).
16. Slavov, N. Scaling Up Single-Cell Proteomics. Mol. Cell. Proteomics 21, 100179 (2022).
17. Burnum-Johnson, K. E. et al. New Views of Old Proteins: Clarifying the Enigmatic Proteome.
Mol. Cell. Proteomics 21, 100254 (2022).
18. Nwosu, A. J. et al. In-Depth Mass Spectrometry-Based Proteomics of Formalin-Fixed, Paraffin-
Embedded Tissues with a Spatial Resolution of 50-200 μm. J. Proteome Res. 21, 2237–2245
(2022).
19. Meier, F. et al. Parallel Accumulation–Serial Fragmentation (PASEF): Multiplying Sequencing
Speed and Sensitivity by Synchronized Scans in a Trapped Ion Mobility Device. J. Proteome
Res. 14, 5378–5387 (2015).
20. Meier, F. et al. Online Parallel Accumulation-Serial Fragmentation (PASEF) with a Novel
.CC-BY 4.0 International licenseperpetuity. It is made available under a
preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
The copyright holder for thisthis version posted October 31, 2022. ; https://doi.org/10.1101/2022.10.31.514544doi: bioRxiv preprint
Trapped Ion Mobility Mass Spectrometer. Mol. Cell. Proteomics 17, 2534–2545 (2018).
21. Meier, F. et al. diaPASEF: parallel accumulation–serial fragmentation combined with data-
independent acquisition. Nat. Methods 17, 1229–1236 (2020).
22. Brunner, A.-D. et al. Ultra-high sensitivity mass spectrometry quantifies single-cell proteome
changes upon perturbation. Mol. Syst. Biol. 18, e10798 (2022).
23. Bache, N. et al. A Novel LC System Embeds Analytes in Pre-formed Gradients for Rapid, Ultra-
robust Proteomics. Mol. Cell. Proteomics 17, 2284–2296 (2018).
24. Demichev, V., Messner, C. B., Vernardis, S. I., Lilley, K. S. & Ralser, M. DIA-NN: neural
networks and interference correction enable deep proteome coverage in high throughput. Nat.
Methods 17, 41–44 (2020).
25. Tenzer, S. Method and apparatus for data independent combined ion mobility and mass
spectroscopy analysis. US Patent (2022).
26. Specht, H. et al. Single-cell proteomic and transcriptomic analysis of macrophage heterogeneity
using SCoPE2. Genome Biol. 22, 50 (2021).
27. Derks, J. et al. Increasing the throughput of sensitive proteomics by plexDIA. Nat. Biotechnol.
(2022) doi:10.1038/s41587-022-01389-w.
28. Huffman, R. G. et al. Prioritized single-cell proteomics reveals molecular and functional
polarization across primary macrophages. Preprint at https://doi.org/10.1101/2022.03.16.484655.
29. Webber, K. G. I. et al. Label-Free Profiling of up to 200 Single-Cell Proteomes per Day Using a
Dual-Column Nanoflow Liquid Chromatography Platform. Anal. Chem. 94, 6017–6025 (2022).
30. Vowinckel, J. et al. Cost-effective generation of precise label-free quantitative proteomes in
high-throughput by microLC and data-independent acquisition. Sci. Rep. 8, 4346 (2018).
31. Bian, Y. et al. Robust, reproducible and quantitative analysis of thousands of proteomes by
micro-flow LC–MS/MS. Nat. Commun. 11, 1–12 (2020).
32. Kulak, N. A., Pichler, G., Paron, I., Nagaraj, N. & Mann, M. Minimal, encapsulated proteomic-
sample processing applied to copy-number estimation in eukaryotic cells. Nat. Methods 11, 319–
324 (2014).
.CC-BY 4.0 International licenseperpetuity. It is made available under a
preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in
The copyright holder for thisthis version posted October 31, 2022. ; https://doi.org/10.1101/2022.10.31.514544doi: bioRxiv preprint
... 10ng), which can drastically boost the number of quantified proteins. So far, either diluted bulk cell population digests or samples containing multiple cells have been used for this purpose 3,35,39,40 . ...
Preprint
Full-text available
Single-cell resolution analysis of complex biological tissues is fundamental to capture cell-state heterogeneity and distinct cellular signaling patterns that remain obscured with population-based techniques. The limited amount of material encapsulated in a single cell however, raises significant technical challenges to molecular profiling. Due to extensive optimization efforts, mass spectrometry-based single-cell proteomics (scp-MS) has emerged as a powerful tool to facilitate proteome profiling from ultra-low amounts of input, although further development is needed to realize its full potential. To this end, we carried out comprehensive analysis of orbitrap-based data independent acquisition (DIA) for limited material proteomics. Notably, we found a fundamental difference between optimal DIA methods for high- and low-load samples. We further improved our low-input DIA method by relying on high-resolution MS1 quantification, thus more efficiently utilizing available mass analyzer time. With our ultra-low input tailored DIA method, we were able to accommodate long injection times and high resolution, while keeping the scan cycle time low enough to ensure robust quantification. Finally, we establish a complete experimental scp-MS workflow, combining DIA with accessible single-cell sample preparation and the latest chromatographic and computational advances and showcase our developments by profiling real single cells.
Article
Full-text available
The proteome provides unique insights into disease biology beyond the genome and transcriptome. A lack of large proteomic datasets has restricted the identification of new cancer biomarkers. Here, proteomes of 949 cancer cell lines across 28 tissue types are analyzed by mass spectrometry. Deploying a workflow to quantify 8,498 proteins, these data capture evidence of cell-type and post-transcriptional modifications. Integrating multi-omics, drug response, and CRISPR-Cas9 gene essentiality screens with a deep learning-based pipeline reveals thousands of protein biomarkers of cancer vulnerabilities that are not significant at the transcript level. The power of the proteome to predict drug response is very similar to that of the transcriptome. Further, random downsampling to only 1,500 proteins has limited impact on predictive power, consistent with protein networks being highly connected and co-regulated. This pan-cancer proteomic map (ProCan-DepMapSanger) is a comprehensive resource available at https://cellmodelpassports.sanger.ac.uk.
Article
Full-text available
Current mass spectrometry methods enable high-throughput proteomics of large sample amounts, but proteomics of low sample amounts remains limited in depth and throughput. To increase the throughput of sensitive proteomics, we developed an experimental and computational framework, called plexDIA, for simultaneously multiplexing the analysis of peptides and samples. Multiplexed analysis with plexDIA increases throughput multiplicatively with the number of labels without reducing proteome coverage or quantitative accuracy. By using three-plex non-isobaric mass tags, plexDIA enables quantification of threefold more protein ratios among nanogram-level samples. Using 1-hour active gradients, plexDIA quantified ~8,000 proteins in each sample of labeled three-plex sets and increased data completeness, reducing missing data more than twofold across samples. Applied to single human cells, plexDIA quantified ~1,000 proteins per cell and achieved 98% data completeness within a plexDIA set while using ~5 minutes of active chromatography per cell. These results establish a general framework for increasing the throughput of sensitive and quantitative protein analysis. Proteomics of small sample sizes using data-independent acquisition methods achieves higher throughput with multiplexing.
Article
Full-text available
The dia-PASEF technology uses ion mobility separation to reduce signal interferences and increase sensitivity in proteomic experiments. Here we present a two-dimensional peak-picking algorithm and generation of optimized spectral libraries, as well as take advantage of neural network-based processing of dia-PASEF data. Our computational platform boosts proteomic depth by up to 83% compared to previous work, and is specifically beneficial for fast proteomic experiments and those with low sample amounts. It quantifies over 5300 proteins in single injections recorded at 200 samples per day throughput using Evosep One chromatography system on a timsTOF Pro mass spectrometer and almost 9000 proteins in single injections recorded with a 93-min nanoflow gradient on timsTOF Pro 2, from 200 ng of HeLa peptides. A user-friendly implementation is provided through the incorporation of the algorithms in the DIA-NN software and by the FragPipe workflow for spectral library generation.
Article
Full-text available
All human diseases involve proteins, yet our current tools to characterize and quantify them are limited. To better elucidate proteins across space, time, and molecular composition, we provide a >10 year projection for technologies to meet the challenges that protein biology presents. With a broad perspective, we discuss grand opportunities to transition the science of proteomics into a more propulsive enterprise. Extrapolating recent trends, we describe a next generation of approaches to define, quantify and visualize the multiple dimensions of the proteome, thereby transforming our understanding and interactions with human disease in the coming decade.
Article
Full-text available
Despite the availabilty of imaging-based and mass-spectrometry-based methods for spatial proteomics, a key challenge remains connecting images with single-cell-resolution protein abundance measurements. Here, we introduce Deep Visual Proteomics (DVP), which combines artificial-intelligence-driven image analysis of cellular phenotypes with automated single-cell or single-nucleus laser microdissection and ultra-high-sensitivity mass spectrometry. DVP links protein abundance to complex cellular or subcellular phenotypes while preserving spatial context. By individually excising nuclei from cell culture, we classified distinct cell states with proteomic profiles defined by known and uncharacterized proteins. In an archived primary melanoma tissue, DVP identified spatially resolved proteome changes as normal melanocytes transition to fully invasive melanoma, revealing pathways that change in a spatial manner as cancer progresses, such as mRNA splicing dysregulation in metastatic vertical growth that coincides with reduced interferon signaling and antigen presentation. The ability of DVP to retain precise spatial proteomic information in the tissue context has implications for the molecular profiling of clinical samples.
Article
Formalin-fixed, paraffin-embedded (FFPE) tissues are banked in large repositories to cost-effectively preserve valuable specimens for later study. With the rapid growth of spatial proteomics, FFPE tissues can serve as a more accessible alternative to more commonly used frozen tissues. However, extracting proteins from FFPE tissues is challenging due to cross-links formed between proteins and formaldehyde. Here, we have adapted the nanoPOTS sample processing workflow, which was previously applied to single cells and fresh-frozen tissues, to profile protein expression from FFPE tissues. Following the optimization of extraction solvents, times, and temperatures, we identified an average of 1312 and 3184 high-confidence master proteins from 10 μm thick FFPE-preserved mouse liver tissue squares having lateral dimensions of 50 and 200 μm, respectively. The observed proteome coverage for FFPE tissues was on average 88% of that achieved for similar fresh-frozen tissues. We also characterized the performance of our fully automated sample preparation and analysis workflow, termed autoPOTS, for FFPE spatial proteomics. This modified nanodroplet processing in one pot for trace samples (nanoPOTS) and fully automated processing in one pot for trace sample (autoPOTS) workflows provides the greatest coverage reported to date for high-resolution spatial proteomics applied to FFPE tissues. Data are available via ProteomeXchange with identifier PXD029729.
Article
Mass spectrometry (MS)-based proteomics has become a powerful technology to quantify the entire complement of proteins in cells or tissues. Here, we review challenges and recent advances in the LC-MS-based analysis of minute protein amounts, down to the level of single cells. Application of this technology revealed that single-cell transcriptomes are dominated by stochastic noise due to the very low number of transcripts per cell, whereas the single-cell proteome appears to be complete. The spatial organization of cells in tissues can be studied by emerging technologies, including multiplexed imaging and spatial transcriptomics, which can now be combined with ultra-sensitive proteomics. Combined with high-content imaging, artificial intelligence and single-cell laser microdissection, MS-based proteomics provides an unbiased molecular readout close to the functional level. Potential applications range from basic biological questions to precision medicine.
Preprint
Aneuploidy, an imbalance in chromosome copy numbers, causes genetic disorders, and drives cancer progression, drug tolerance, and antimicrobial resistance. While aneuploidy can confer stress resistance, it is not well understood how cells overcome the fitness burden caused by aberrant chromosomal copy numbers. Studies using both systematically generated and natural aneuploid yeasts triggered an intense debate about the role of dosage compensation, concluding that aneuploidy is transmitted to the transcriptome and proteome without significant buffering at the chromosome-wide level, and is, at least in lab strains, associated with significant fitness costs. Conversely, systematic sequencing and phenotyping of large collections of natural isolates revealed that aneuploidy is frequent and has few - if any - fitness costs in nature. To address these discrepant findings at the proteomic level, we developed a platform that yields highly precise proteomic measurements across large numbers of genetically diverse samples, and applied it to natural isolates collected as part of the 1011 genomes project. For 613 of the isolates, we were able to match the proteomes to their corresponding transcriptomes and genomes, subsequently quantifying the effect of aneuploidy on gene expression by comparing 95 aneuploid with 518 euploid strains. We find, as in previous studies, that aneuploid gene dosage is not buffered chromosome-wide at the transcriptome level. Importantly, in the proteome, we detect an attenuation of aneuploidy by about 25% below the aneuploid gene dosage in natural yeast isolates. Furthermore, this chromosome-wide dosage compensation is associated with the ubiquitin-proteasome system (UPS), which is expressed at higher levels and has increased activity across natural aneuploid strains. Thus, through systematic exploration of the species-wide diversity of the yeast proteome, we shed light on a long-standing debate about the biology of aneuploids, revealing that aneuploidy tolerance is mediated through chromosome-wide dosage compensation at the proteome level.
Article
Single-cell proteomics (SCP) has great potential to advance biomedical research and personalized medicine. The sensitivity of such measurements increases with low-flow separations (<100 nL/min) due to improved ionization efficiency, but the time required for sample loading, column washing, and regeneration in these systems can lead to low measurement throughput and inefficient utilization of the mass spectrometer. Herein, we developed a two-column liquid chromatography (LC) system that dramatically increases the throughput of label-free SCP using two parallel subsystems to multiplex sample loading, online desalting, analysis, and column regeneration. The integration of MS1-based feature matching increased proteome coverage when short LC gradients were used. The high-throughput LC system was reproducible between the columns, with a 4% difference in median peptide abundance and a median CV of 18% across 100 replicate analyses of a single-cell-sized peptide standard. An average of 621, 774, 952, and 1622 protein groups were identified with total analysis times of 7, 10, 15, and 30 min, corresponding to a measurement throughput of 206, 144, 96, and 48 samples per day, respectively. When applied to single HeLa cells, we identified nearly 1000 protein groups per cell using 30 min cycles and 660 protein groups per cell for 15 min cycles. We explored the possibility of measuring cancer therapeutic targets with a pilot study comparing the K562 and Jurkat leukemia cell lines. This work demonstrates the feasibility of high-throughput label-free single-cell proteomics.