CoChat
Sign Up
Cited by 129Single-cell and spatial transcriptomicsRNA Research and SplicingGene Regulatory Network Analysis

The DOI matches its Crossref record; 40 of the 40 references it cites were found in a registry or an index.

A combined approach for single-cell mRNA and intracellular protein expression analysis

Open access
Reimegård, Johan, Tarbier, Marcel, Danielsson, Marcus, Schuster, Jens, Baskaran, Sathishkumar, Panagiotou, Styliani, Dahl, Niklas, Friedländer, Marc R., Gallant, Caroline J.·2021·Communications Biology
DOI 10.1038/s42003-021-02142-wPMC PMC8149646
Open this paper in CoChat — free accountOpen PDF

Abstract

Combined measurements of mRNA and protein expression in single cells enable in-depth analysis of cellular states. We present SPARC, an approach that combines single-cell RNA-sequencing with proximity extension essays to simultaneously measure global mRNA and 89 intracellular proteins in individual cells. We show that mRNA expression fails to accurately reflect protein abundance at the time of measurement, although the direction of changes is in agreement during neuronal differentiation. Moreover, protein levels of transcription factors better predict their downstream effects than do their corresponding transcripts. Finally, we highlight that protein expression variation is overall lower than mRNA variation, but relative protein variation does not reflect the mRNA level. Our results demonstrate that mRNA and protein measurements in single cells provide different and complementary information regarding cell states. SPARC presents a state-of-the-art co-profiling method that overcomes current limitations in throughput and protein localization, including removing the need for cell fixation.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s42003-021-02142-w

Cites

40 of 40 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 SPARC profiles RNA and intra-cellular protein from the same single cells. a The SPARC procedure. Single cells are isolated and lysed in the presence of oligo-dT conjugated magnetic beads. Following oligo-dT mRNA hybridization, the protein-containing supernatant is removed for subsequent multiplex proximity extension analysis (PEA) and the mRNA is processed using a modified Smart-seq2 approach. (Protein clip art from Servier Medical Art is used under CC BY 3.0 France.) b Overview of the cellular model analyzed using SPARC. Human embryonic stem cells were analyzed in culture (0 h) and following directed neural induction (24 h and 48 h). c Example mRNA (blue) and protein (red) expression in single cells (violin plots) or replicate 100 cell population control (black dots) measured at 0, 24, and 48 h post neural induction.
Fig. 1 SPARC profiles RNA and intra-cellular protein from the same single cells. a The SPARC procedure. Single cells are isolated and lysed in the presence of oligo-dT conjugated magnetic beads. Following oligo-dT mRNA hybridization, the protein-containing supernatant is removed for subsequent multiplex proximity extension analysis (PEA) and the mRNA is processed using a modified Smart-seq2 approach. (Protein clip art from Servier Medical Art is used under CC BY 3.0 France.) b Overview of the cellular model analyzed using SPARC. Human embryonic stem cells were analyzed in culture (0 h) and following directed neural induction (24 h and 48 h). c Example mRNA (blue) and protein (red) expression in single cells (violin plots) or replicate 100 cell population control (black dots) measured at 0, 24, and 48 h post neural induction.CC BY · PubMed Central
Fig. 2 mRNA abundance only partially reflects and protein abundance in single cells. tSNE plot and pseudotime analysis of the a normalized mRNA and b protein expression data. c Comparison of the cell pseudotime-ordering if calculated based on either mRNA or protein expression data. d Example mRNA and protein expression data plotted for SOX2, EPCAM, and POU5F1. The data is shown either as RNA expression vs. protein expression scatter plot with respective density plots included on each axis, or as cells ordered along RNA pseudotime with either RNA and protein expression plotted on the y-axis. Both the RNA and protein data are plotted as log2. Color corresponds to time point in all plots 0 h (green), 24 h (orange), and 48 h (blue).
Fig. 2 mRNA abundance only partially reflects and protein abundance in single cells. tSNE plot and pseudotime analysis of the a normalized mRNA and b protein expression data. c Comparison of the cell pseudotime-ordering if calculated based on either mRNA or protein expression data. d Example mRNA and protein expression data plotted for SOX2, EPCAM, and POU5F1. The data is shown either as RNA expression vs. protein expression scatter plot with respective density plots included on each axis, or as cells ordered along RNA pseudotime with either RNA and protein expression plotted on the y-axis. Both the RNA and protein data are plotted as log2. Color corresponds to time point in all plots 0 h (green), 24 h (orange), and 48 h (blue).CC BY · PubMed Central
Fig. 3 Transcription factor protein but not mRNA correlates with target mRNA. a Scatter plot of POU5F1 expression in 0 h single cells (steady state), both at the level of RNA and protein compared to the RNA expression of POU5F1 regulatory target genes TDGF1 (upper panels) and OTX2 (lower panels). Blue line represents linear model between the two expression patterns. b Correlation between POU5F1 protein and RNA levels with the RNA expression of POU5F1 regulatory targets in steady state (0 h) or dynamic (0, 24, and 48 h) conditions. Darker shading indicates stronger correlation. Bold outlining indicates in which of the analysis groups the correlation was considered to be significant (FDR < 0.05). c Heatmap with expression levels of POU5F1 at the protein and RNA level and the RNA expression of POU5F1 regulatory target candidate genes. The color scale relates to high (red) and low (blue) expression. The columns in the heatmap are ordered based on the protein expression of POU5F1. d Red lines represent POU5F1 (protein)—target (RNA) ROC curves and blue lines represent POU5F1 (RNA)—target (RNA) ROC curves based on regulatory link weights. Solid line represents dynamic conditions, dotted lines indicate steady-state conditions. Filled squares represent the Youden index of the different ROC curves. e Quantification of total area under the curve for ROC (see Fig. 3d) including a permuted control of 200 permutations each AUC represented as a dot.
Fig. 3 Transcription factor protein but not mRNA correlates with target mRNA. a Scatter plot of POU5F1 expression in 0 h single cells (steady state), both at the level of RNA and protein compared to the RNA expression of POU5F1 regulatory target genes TDGF1 (upper panels) and OTX2 (lower panels). Blue line represents linear model between the two expression patterns. b Correlation between POU5F1 protein and RNA levels with the RNA expression of POU5F1 regulatory targets in steady state (0 h) or dynamic (0, 24, and 48 h) conditions. Darker shading indicates stronger correlation. Bold outlining indicates in which of the analysis groups the correlation was considered to be significant (FDR < 0.05). c Heatmap with expression levels of POU5F1 at the protein and RNA level and the RNA expression of POU5F1 regulatory target candidate genes. The color scale relates to high (red) and low (blue) expression. The columns in the heatmap are ordered based on the protein expression of POU5F1. d Red lines represent POU5F1 (protein)—target (RNA) ROC curves and blue lines represent POU5F1 (RNA)—target (RNA) ROC curves based on regulatory link weights. Solid line represents dynamic conditions, dotted lines indicate steady-state conditions. Filled squares represent the Youden index of the different ROC curves. e Quantification of total area under the curve for ROC (see Fig. 3d) including a permuted control of 200 permutations each AUC represented as a dot.CC BY · PubMed Central
Fig. 4 RNA variation is not predictive for protein variation. Patterns of gene expression variation at steady-state in S-phase cells. a Relation between mean RNA expression mean (RPKM) and RNA expression variation (CV2). Colors indicate variably (red) or stably (blue) expressed genes. b Similar to a, but for protein. c Normalized gene expression variation (residuals) at the RNA and protein levels correlate moderately (Spearman’s rho = 0.28). Genes of interest are indicated. d A simple additive model of RNA variation and estimated translation rate effectively predicts protein expression variation (Spearman’s rho = 0.50). A subset of genes from Fig. 3c are included here, as public data were not available to estimate translation rate for all genes of interest.
Fig. 4 RNA variation is not predictive for protein variation. Patterns of gene expression variation at steady-state in S-phase cells. a Relation between mean RNA expression mean (RPKM) and RNA expression variation (CV2). Colors indicate variably (red) or stably (blue) expressed genes. b Similar to a, but for protein. c Normalized gene expression variation (residuals) at the RNA and protein levels correlate moderately (Spearman’s rho = 0.28). Genes of interest are indicated. d A simple additive model of RNA variation and estimated translation rate effectively predicts protein expression variation (Spearman’s rho = 0.50). A subset of genes from Fig. 3c are included here, as public data were not available to estimate translation rate for all genes of interest.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Cited by

Sorted by citations

See all 129

Related

See all

Metadata

Authors
Reimegård, J., Tarbier, M., Danielsson, M., Schuster, J., Baskaran, S., Panagiotou, S., Dahl, N., Friedländer, M. R., & Gallant, C. J.
Affiliations
  • Uppsala University (SE)
  • Stockholm University (SE)
Corresponding author
Gallant, Caroline J.
ORCID
Venue
Communications Biology
Publisher
Springer Science and Business Media LLC
Published
2021-05-25
Volume
4
Issue
1
Pages
624-624
ISSN
2399-3642 (ISSN-L)
DOI
10.1038/s42003-021-02142-w
PMC
PMC8149646
OpenAlex
W3164706810
Cited by
129
References
39
Citations by year
2026: 25 · 2025: 26 · 2024: 22 · 2023: 23 · 2022: 28 · 2021: 5
FWCI
6.12
Type
journal article
Language
en
Open access
Licence
Locations
  • Communications Biology (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Uppsala University Publications (Uppsala University) (submittedVersion, other-oa)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
Vetenskapsrådet (2017-05229)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Single-cell and spatial transcriptomics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › RNA Research and Splicing
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Gene Regulatory Network Analysis
Keywords
single-cell transcriptomics, single-cell proteomics, proximity extension assay, SPARC, intracellular protein expression, neuronal differentiation, transcription factor activity, mRNA-protein correlation, protein localization
MeSH terms
Humans, Proteins, RNA, Messenger, Transcription, Genetic, Sequence Analysis, RNA, Gene Expression Profiling, Proteomics, Single-Cell Analysis

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review