Download LC-MRM, a rapid tool for high throughput quantification

LC-MRM, a rapid tool for high throughput quantification of protein target expression after mRNA administration
Claudia Escher1, Jeffrey C. Silva2, Jesse J. Chen3
Biognosys AG, Wagistrasse 25, CH-8952 Schlieren
2
Biognosys Inc., One Kendall Square, Building 200, Suite 2203, Cambridge, MA 02139, USA
3
Moderna Therapeutics, Inc., 200 Technology Square, Cambridge MA 20139, USA
1
Summary
Moderna and Biognosys have worked together on several projects with the purpose of using targeted proteomics for the high throughput
quantification of protein target expression after mRNA administration. In this study, LC-MRM assays for 114 target proteins were generated
for the quantification in 480 samples of HeLa cells in several batches. Amongst other targets, we quantified the expression of GTP
cyclohydrolase 1 (GCH1) protein in human HeLa cell lines and identified the mRNA version yielding the highest protein amount. Also,
highly similar cat and dog erythropoietin were quantified in HeLa cells after mRNA administration, showing the high specificity of LC-MRM
for quantifying proteins with high sequence similarities.
Introduction
Messenger RNA (mRNA) is a polyribonucleic acid used by organisms as a template for protein expression and may also serve as a tool for
selected protein expression in cells. Despite decades of interest of the scientific community to engineer nucleic acids for therapeutic use,
mRNA has only recently been recognized as a revolutionary human therapeutic agent (Thess et al. 2015).
mRNA is a carrier of genetic information which recruits the endogenous protein translational machinery within the cell to produce active
therapeutic proteins. Thus, the quantification of the target protein after the administration of mRNA therapeutic is a critical step in efficacy
evaluation of any treatment. Traditionally, this has been done using antibody-based methods such as ELISA or Western blots (Matsui et al.
2015, DeRosa et al 2016). These sensitive methods have several limitations. An antibody may not always be available for a particular mRNA
target protein, especially for non-human or non-mouse species, and method development for fully validated antibody reagents often can’t
be achieved in a time-frame acceptable to pre-clinical R&D. These issues are further exacerbated when it is necessary to examine multiple
variants of a single protein target (point mutants, N- or C-truncated isoforms) for optimal biological activity.
These limitations can be addressed by applying mass spectrometry-based quantitative proteomics, a powerful high-throughput platform
for monitoring protein expression that is a common technology platform for biomedical and clinical research (Pan et al 2009). Targeted
proteomics methods such as multiple reaction monitoring (MRM) are highly specific, sensitive and efficient assays for the quantification
of hundreds of protein targets (Picotti et al. 2010). Protein targets from uncommon species or minor sequence differences can be resolved
due to the high specificity of the mass spectrometer (Shi et al. 2012).
Moderna and Biognosys have worked together on several projects with the purpose of using targeted proteomics for the high throughput
quantification of protein target expression after mRNA administration. Since targeted proteomics overcomes the limitations of antibodybased, it was used in a series of experiments focused on quantifying protein expression in a large number of samples. Here, we present
examples for LC-MRM protein quantification of highly similar non-human proteins and overexpression of human proteins and demonstrate
a rapid assay generation workflow for proteins where no antibody assay is available.
Figure 1: MRM assay
development and protein
quantification workflow. A)
Five modified mRNAs were
administrated to HeLa cells.
B) Proteins were extracted
from HeLa cell lysates and
digested into peptides with
trypsin. C) LC-MS/MS was
used to identify detectable
proteins in the sample
and three peptides per
target protein with the best
response were selected for
LC-MRM assay generation.
D) Target proteins were
measured in LC-MRM for
relative quantification of
the proteins of interest and
analysed using Biognosys’
SpectroDive software.
Methods
Different variants of mRNA encoded for a wide range of human and non-human proteins were applied to HeLa cells. Cells were lysed in a
strongly denaturing buffer and proteins were digested to peptides using trypsin. LC-MS/MS shotgun mass spectrometry was carried out
for target protein detection. The three peptides representing the target protein giving the best response were selected for LC-MRM assay
generation. Up to three peptides for the target protein and up to five peptides representing selected housekeeping proteins were measured
in LC-MRM for relative quantification of the proteins of interest. Data was analysed using SpectroDive, Biognosys’ software for the rapid
analysis of multiplexed LC-MRM data.
Results and discussion
In this study an LC-MRM platform was used to evaluate the efficacy of several mRNA therapeutics by confirming the protein expression in
the target cells and quantifying the relative expression across multiple mRNA variants.
LC-MRM assays for 114 target proteins were generated for the quantification in 480 samples of HeLa cells in several batches. The time
between sample receipt and results was only one week including the development of specific assays (Figure 1).
The expression of GTP cyclohydrolase 1 (GCH1) protein in human HeLa cell lines was identified in all 5 mRNA-treated samples, but not in
the control sample (Figure 2). Furthermore, sample 4 was identified as the mRNA variant yielding the highest level of GCH1 protein.
The high specificity of the LC-MRM method becomes apparent especially when target proteins share large portions of their amino acid
sequence and antibodies are often not able to resolve a single amino acid difference. The highly similar cat and dog erythropoietin were
quantified in HeLa cells after mRNA administration using specific LC-MRM assays for either species (Figure 3a). Shared peptides between
the cat and dog protein were detected and quantified in both samples (Figure 3b).
This study shows that LC-MRM is a robust and rapid screening tool for protein quantification that can be applied in the development of
therapeutic mRNAs. In addition, the method can be easily multiplexed with no effect on specificity or the time required to carry out the assay,
and multiple peptides per protein can be analysed further enhancing the confidence in results. When absolute quantification of the target
protein is required, stable isotope-labelled reference peptides can incorporated, to configure a multiplex LC-MRM assay for target proteins
at attomole sensitivity.
References
DeRosa F, Guild B, Karve S, et al. Therapeutic efficacy in a hemophilia B model using a biosynthetic mRNA liver depot system. Gene Ther. 2016 Oct;23(10):699-707.
Matsui A, Uchida S, Ishii T, Itaka K, Kataoka K. Messenger RNA-based therapeutics for the treatment of apoptosis-associated diseases. Sci Rep. 2015 Oct 28;5:15810.
Pan S, Aebersold R, Chen R, et al. Mass spectrometry based targeted protein quantification: methods and applications. J Proteome Res. 2009 Feb;8(2):787-97.
Picotti P, Rinner O, Stallmach R, et ak. High-throughput generation of selected reaction-monitoring assays for proteins and proteomes. Nat Methods. 2010 Jan;7(1):43-6.
Shi T, Su D, Liu T, Tet al. Advancing the sensitivity of selected reaction monitoring-based targeted quantitative proteomics. Proteomics. 2012 Apr;12(8):1074-92.
Thess A, et al. Sequence-engineered mRNA Without Chemical Nucleoside Modifications Enables an Effective Protein Therapy in Large Animals. Mol Ther. 2015;23(9):145664.
Figure 2: Quantifying the expression of GTP cyclohydrolase 1 (GCH1) protein and identification of high-yielding mRNAs. Three GCH1 specific
peptides were selected for LC-MRM protein quantification: AASAMQFFTK, VHIGYLPNK and IVEIYSR. A) GCH1 specific peptide AASAMQFFTK
was detected in all mRNA-treated samples, but not in the untreated control sample. B) Quantification using three selected peptides (independent
analytes) identifies sample 4 (*) as the mRNA variant yielding the highest level of GCH1 expression.
Figure 3: High specificity of LC-MRM for
quantifying proteins with high sequence
similarities. Cat and dog erythropoietin
(EPO) mRNAs were administrated to
HeLa cells and their protein expression
was monitored with LC-MRM. A) Cat
and dog erythropoietin sequences are
highly similar. Three different peptides
(marked in red) were selected for LCMRM quantification. Two of them were
species specific, while the third selected
peptide is shared in both sequences. B)
LC-MRM was able to accurately detect
both cat- and dog-specific EPO peptides
in only the species-specific expression
experiment, while the common peptide
was detected in both experiments.