1
Genome-wide parallel quantification of mRNA and protein levels
and turnover in mammalian cells
Björn Schwanhäusser
1
, Dorothea Busse
1
, Na Li
1
, Gunnar Dittmar
1
, Johannes
Schuchhardt
2
, Jana Wolf
1
, Wei Chen
1
, Matthias Selbach
1
1 Max Delbrück Centrum for Molecular Medicine, Robert-Rössle-Str. 10, D-13092 Berlin, Germany
2 MicroDiscovery GmbH, Marienburger Str. 1, D-10405 Berlin, Germany
Correspondence:
Jana Wolf, Max Delbrück Centrum for Molecular Medicine, Robert-Rössle-Str. 10, D-
13092 Berlin, Germany, Tel.: +49 30 9406 2641, Fax.: +49 30 9406 2394, email:
jana.wolf@mdc-berlin.de
Wei Chen, Max Delbrück Centrum for Molecular Medicine, Robert-Rössle-Str. 10, D-
13092 Berlin, Germany, Tel.: +49 30 9406 2995, Fax.: +49 30 9406 3068, email:
wei.chen@mdc-berlin.de
Matthias Selbach, Max Delbrück Centrum for Molecular Medicine, Robert-Rössle-Str.
10, D-13092 Berlin, Germany, Tel.: +49 30 9406 3574, Fax.: +49 30 9406 2394, email:
matthias.selbach@mdc-berlin.de
Running title: mRNA and protein levels and half-lives
2
Summary
Gene expression is a multistep process that involves transcription, translation and
turnover of mRNAs and proteins. Although it is one of the most fundamental processes
of life, the entire cascade has never been quantified on a genome-wide scale. Here, we
simultaneously measured mRNA and protein abundance and turnover by parallel
metabolic pulse labeling for more than 5,000 genes in mammalian cells. While mRNA
and protein levels correlated better than previously thought, corresponding half-lives
showed no correlation. Employing a quantitative model we obtain the first genome-scale
prediction of synthesis rates of mRNAs and proteins. We find that the cellular abundance
of proteins is predominantly controlled at the level of translation. Genes with similar
combinations of mRNA and protein stabilities shared functional properties, suggesting
that half-lives evolved under energetic and dynamic constraints. Quantitative information
about all stages of gene expression obtained in this study provides a rich resource and
helps understanding the underlying design principles.
Gelöscht: different
3
Introduction
The four fundamental cellular processes involved in gene expression are transcription,
mRNA degradation, translation and protein degradation. It is now clear that each step of
this cascade is controlled by gene-regulatory events
1-3
. While each individual process
has been intensively studied
4-7
, little is known about how the combined effect of all
regulatory events shapes gene expression. The fundamental question of how genomic
information is processed at different levels to obtain a specific cellular proteome has
therefore remained unanswered. Genome-wide quantitative data about the flux of
information from genes to proteins is not available for any organism.
Towards a quantitative description of gene expression numerous previous studies
compared steady-state mRNA and protein levels and arrived at the conclusion that the
correlation is poor
8
. However, the available data suffers from several limitations. First,
most studies are limited to a few hundred genes, mainly due to the technical challenges
involved in large scale protein identification and quantification. For example, the largest
mammalian protein copy number dataset comprises only 512 genes
9
. Second, protein
levels measured in one experiment are typically compared to mRNA levels determined
in a different experiment performed at a different time in a different lab, making it difficult
to interpret why the correlation is low. Third, mRNA levels are measured using
microarrays which are less accurate than recent mRNA sequencing methods
10
. Fourth,
many studies were performed in bacteria or yeast and thus do not represent regulatory
mechanisms specific for higher eukaryotes. Finally, mRNA and protein levels result from
coupled processes of synthesis and degradation. Therefore, analysis of mRNA and
protein levels alone cannot provide sufficient information to understand gene expression
comprehensively. mRNA and protein turnover can be measured with drugs to inhibit
4
transcription or translation
11
, but this has severe side-effects. Studies based on artificial
fusion proteins are also problematic since tagging can affect protein stability
12
.
To overcome these limitations we sought to quantify cellular mRNA and protein
expression levels and turnover in parallel in a population of unperturbed mammalian
cells. Pulse labeling with radioactive nucleosides or amino acids is regarded as the gold
standard method to determine mRNA and protein half-lives
13
. Recently, variants of this
approach based on non-radioactive tracers have been established
14-16
. In stable isotope
labeling by amino acids in cell culture (SILAC) cells are cultivated in a medium
containing heavy stable-isotope versions of essential amino acids
17
. When non-labeled
(i.e. light) cells are transferred to heavy SILAC growth medium, newly synthesized
proteins incorporate the heavy label while pre-existing proteins remain in the light form.
This strategy can be used to measure protein turnover
18
or relative changes in protein
translation
19
. Similarly, newly synthesized RNA can be labeled with the nucleoside
analog 4-thiouridine (4sU). 4sU containing mRNA can be biotinylated and affinity
purified. Comparing the newly synthesized and pre-existing fraction allows for global
quantification of mRNA half-lives
16,20
.
Parallel pulse labeling of proteins and mRNAs
We used parallel metabolic pulse labeling with amino acids and 4sU to simultaneously
measure protein and mRNA turnover in a population of exponentially growing non-
synchronized mouse fibroblasts (Fig. 1 A). Protein samples were harvested at three time
points and analyzed by liquid chromatography and online tandem mass spectrometry
(LC-MS/MS) on a high performance instrument (LTQ-Orbitrap-Velos). We identified and
quantified proteins with the MaxQuant software package
21
. During five days of data
acquisition we measured 1,471,375 fragment spectra that resulted in 229,985 peptide
5
identifications (84,924 unique peptide sequences, false discovery rate (FDR) < 1%, see
Supplementary Methods on ´Processing of mass spectrometry data´). These peptides
were assigned to 6,445 unique proteins (FDR < 1%). 5,279 of these proteins were
quantified by at least three heavy to light (H/L) peptide ratios (Fig. 1 B). Tissue-specific
amino acid precursor pools and recycling rates, a pervasive problem for in vivo pulse
labeling experiments
15,22
, did not appreciably affect our results (Fig. S1). We also tested
if protein synthesis rates are uniform over time. In case of constant incorporation rates
the logarithm of H/L ratios should increase linearly with time (Fig. 1 C).
93 % showed
excellent linear correlation indicated by a variability of the linear regression slope smaller
than 1 % (two and three time point measurements, Fig. 1 D).
Thus, our data does not seem to be affected by non-uniform incorporation rates or by
recycling. Also, protein abundance did not influence H/L ratio measurements (Fig. S 2).
In total, we obtained a confident set of 5,028 protein half-lives calculated from the slope
of the regression line (see Supplementary Methods). Cycloheximide-chase experiments
for selected proteins spanning a representative range of half-lives agreed well with half-
lives determined by pulsed labeling and mass spectrometry in all cases (Fig. 1 E).
In parallel, we pulse labeled newly synthesized RNA for 2 h with 4sU. RNA samples
were fractionated into the newly synthesized and pre-existing fractions. Both fractions
and the total unfractionated RNA sample were analyzed by mRNA sequencing on an
Illumina Genome Analyzer. In total, we obtained 80,709,361 sequencing reads in all
three samples, 55,046,553 (68%) of which could be mapped to the mouse genome. In
all three samples, transcripts were quantified based on the number of reads mapped on
their exonic region divided by transcript length and the total number of reads obtained
10
.
We calculated mRNA half-lives based on the ratios of newly synthesized RNA/total RNA
ratio and the preexisting RNA/total RNA using the previously published approach
16
.