Absolute Scaling of Single-Cell Transcriptomes Reveals Pervasive Hypertranscription in Adult Stem and Progenitor Cells

Yun-Kyo Kim, Miguel Ramalho-Santos

Preprint posted on 13 December 2021 https://www.biorxiv.org/content/10.1101/2021.12.13.472426v1

Quantifying the absolute RNA content of single cells is possible for existing data sets and biologically relevant.

Selected by Vanessa Luzak

Categories: biochemistry, bioinformatics, cell biology, developmental biology, molecular biology

Introduction

Single cell RNA-seq (scRNA-seq) is a powerful tool to measure gene expression in individual cells. While relative gene expression differences are currently analyzed with great interest in single cell data, differences in total transcript levels were mainly considered technical artefacts and removed during normalization, assuming that single cells contain similar amounts of total RNA.

In this preprint, Kim et al. aimed to investigate whether current scRNA-seq protocols can be used to measure total transcript levels of single cells. By using a well characterized biological example of cells that increase their total transcript amounts during development (referred to as “hypertranscription”), the authors were able to verify that differences in total transcript levels between cells are not mainly due to technical artefacts. Instead, such differences can be quantified in existing scRNA-seq data sets using UMIs and/ or ERCC spike ins and have a biological meaning. When the authors applied their tool to scRNAseq data from adult mouse organs, they were able to show that hypertranscription is not a rare program restricted to specific cell types and periods during development. Instead, they found that it is a rather general biological process, which occurs in cells with a high demand for biomass.

Definition of the term “hypertranscription”:

Hypertranscription has been characterized previously by various groups for specific cell types during development. Transcription rates and transcript levels are globally increased in hypertranscribing cells, which is accompanied by a global opening of chromatin, high activity of chromatin remodellers and general transcription amplifiers, such as members of the Myc protein family. As a hallmark, genes summarized by the GO terms “chromatin remodelling”, “DNA repair”, “ribosome biogenesis” and “translation” show especially high transcript levels during hypertranscription.

Summary of findings

For the study, no new scRNAseq data was generated. Instead, previously published scRNAseq datasets were used.

For this summary, selected key experiments are described. More supporting data can be found in the preprint.

Step 1: Absolute scaling of RNA levels is possible using ERCC spike in or UMIs (in vitro setup)

Question: Are defined differences of RNA levels in a dilution series detected using UMIs or ERCC spike in?

UMIs: First, the authors analysed data from the previously published “scRNAseq mixology” experiment. In this experiment, extracted mRNA was prepared as a dilution series with a known amount of mRNA in each sample. Absolute scaling to UMI counts did correctly detect those defined differences in total mRNAbetween the samples of the dilution series.

ERCC spike in: The dilution series of mRNA in the “scRNAseq mixology” experiment was also successfully detected by absolute scaling to an ERCC spike in, which was used in the experiment in addition to UMIs.

In general, scaling to ERCC spike ins was even more precise than scaling to UMI counts. This was true for the in vitro control experiment, as well in experiments described below using scRNAseq data from biological samples. Nevertheless, both approaches worked well.

Answer: Yes, defined differences in RNA levels can be detected by absolute scaling to UMIs or ERCC spike in.

Step 2: Absolute scaling identifies a previously characterized case of hypertranscription in biological samples

Question: Can differences in total RNA levels be detected in biological scRNAseq samples?

Experiment 1: Embryonic mouse stem cells (mESCs) were grown in two different conditions before cells were mixed and analyzed by 10x Genomics. Condition one (Serum/LIF) induces hypertranscription, whereas condition two (dual GSK/MEK inhibition) induces hypotranscription. Using UMI counts, absolute scaling could successfully detect higher total transcript levels in Serum/LIF grown cells that undergo hypertranscription.

Experiment 2: 10x Genomics data from mouse gonads during development (from E11.5 to P5) was analyzed. Primordial germ cells (PGCs) were previously described to undergo hypertranscription during the period of mitotic expansion around E12.5 and E14.5. Indeed, absolute scaling to UMI counts could specifically detect hypertranscription in PGC cells (but not other cell types) in the indicated time period.

Answer: Yes, biological differences in total RNA levels can be detected by absolute scaling.

Step 3: Absolute scaling identifies hypertranscription as a general biological process in adult Organisms

Question: Is hypertranscription restricted to certain cell types and periods during development?

Experiment 1: The Tabula Muris atlas contains scRNAseq data of 20 different mouse organs. Most organs have been analyzed by two different scRNAseq approaches, namely 10x Genomics including UMIs and SmartSeq2 including ERCC Spike ins. Using the absolute scaling approach to analyze the Tabula Muris data sets, the authors could show that

Low transcript levels were mainly detected in differentiated cell types.
High transcript levels were detected in proliferative stem and progenitor cells, in polyploid cells and in secretory cells.

Along differentiation trajectories, hypertranscription was progressively lost.
Specific hallmarks of hypertranscription (see definition in Introduction) were detected in cells with a high demand for biomass, such as stem and progenitor cells and secretory cells.

Experiment 2: Using two scRNAseq data sets that were recorded during organ regeneration after injury, the authors could show that hypertranscription plays a role in this dynamic process. In the first data set, hypertranscription was activated in satellite stem cells shortly after cardiotoxin- induced skeletal muscle damage, which play a key role in skeletal muscle regeneration. In the second data set, hypertranscription was activated in Clu+ revival stem cells after irradiation-induced damage of the intestine. Taken together, both data sets suggest that hypertranscription plays an important role during organ regeneration.

Answer: No, hypertranscription is not restricted to development. It occurs during development as well as in adult organisms as a general biological process in cells with a high demand for biomass.

Key learnings

Technical: Quantification of total transcript levels in single cells is possible.

Absolute scaling of scRNAseq data is possible in data sets using UMIs and/ or ERCC spike ins.

Biological: Quantification of total transcript levels in single cells is biologically relevant.

Measuring absolute transcript levels of single cells reveals an overlooked, but fundamental transcriptional program in adult organs: hypertranscription is identified in cells with a high demand for biomass, such as multipotent proliferative cells during organ maintenance and regeneration, as well as differentiated secretory cells.

Personal opinion – why this work is important

The study by Kim et al. provides a potential biological meaning to an overlooked, but common aspect of scRNAseq data sets: that for individual cells quite different amounts of total transcripts can be detected. For the first time, it was shown that such differences in total RNA levels are not purely technical artefacts. In my opinion, their absolute scaling approach should routinely be used for future analysis of scRNAseq data sets. Very conveniently, the existing experimental setup does not require any changes to do so and even existing data can easily be reanalyzed.

I was amazed how much new knowledge was generated in this study by looking more closely and a bit differently at existing data sets. Hypertranscription was identified as a general cellular program in adult organs and can potentially be used as a marker during differentiation processes. Further, as a researcher studying single-celled eukaryotes, I am very curious to see if hypertrancription occurs and is relevant in single cell organisms as well?

Posted on: 29 March 2022

doi: https://doi.org/10.1242/prelights.31707

Read preprint

(No Ratings Yet)

Questions to the authors

shared

Q1: Are there any practical guidelines for the setup of future scRNAseq experiments when total transcript levels should be measured?

One of the primary goals in our study was to assess whether information about total transcript content can be obtained from existing scRNA-seq datasets, and the answer is that, generally, yes, if ERCCs or UMIs are used. Both approaches work, though ERCC spike-ins reproduce ground-truth mRNA transcript levels somewhat better and are more robust against sequencing depth than UMIs. Follow-up experiments can include cell-number-normalized methods that we and others have previously described (Percharde et al., Developmental Cell 2017) and that can performed in cell populations of interest (e.g., ERCC-normalized bulk RNA-seq or analyses of 5-EU incorporation).

Q2: How can one distinguish if transcript levels vary due to technical reasons or due to biological reasons in a previously uncharacterized biological system? Were you able to determine a range in which transcript levels vary due to technical reasons only and could this be used as a kind of threshold?

We did not identify a particular transcript range that was characterized predominantly by technical variance, although it should be noted that absolute transcript content tends to differ slightly by dataset and experimental setup. One valuable characteristic to consider is that elevated transcript content does not occur in isolation within hypertranscribing populations. Rather, our present and previous studies have indicated that various molecular players, including double-strand break repair factors, ribosomal proteins, and chromatin remodeling proteins tend to be differentially expressed to facilitate hypertranscription. Furthermore, the presence of global upregulation at the level of nascent transcription is a strong indicator of hypertranscription at play. Thus, assuming all other quality control is completed sufficiently, we would encourage researchers to refer to these related biological cues when assessing the presence of hypertranscription in their systems.

Q3: Is there any proof that observed hypertranscription profiles are exclusively produced by increased global transcription? Or is it possible that certain enriched groups of RNAs are further stabilized by reduced RNA degradation, e.g. by a post-transcriptional modification?

While we see a clear basis for hypertranscription being driven by a global up-regulation of nascent transcription, it is expected that post-transcriptional processes also play important roles in hypertranscription. Indeed we have shown that translational output is required for hypertranscription in mouse embryonic stem cells, because key regulators of nascent transcription are unstable proteins (Bulut-Karslioglu et al., Cell Stem Cell 2018). Some of the research ongoing in the Santos lab further supports the idea that the regulation of global transcriptional output integrates nascent transcription, mRNA stability and translation. We believe that our study demonstrates the potential widespread relevance of hypertranscriptive processes and hope that it stimulates other labs to also explore these and other mechanistic questions in multiple contexts.

Have your say Cancel reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Sign up to customise the site to your preferences and to receive alerts

Also in the biochemistry category:

Mechano-dependent sorbitol accumulation supports biomolecular condensate

Stephanie Torrino, William M Oldham, Andrés R Tejedor, et al.

Selected by Teodora Piskova

Proteasome condensate formation is driven by multivalent interactions with shuttle factors and K48-linked ubiquitin chains

Kenrick A. Waite, Gabrielle Vontz, Stella Y. Lee, et al.

Selected by Aniruddha Das

Conserved cardiolipin-mitochondrial ADP/ATP carrier interactions assume distinct structural and functional roles that are clinically relevant

Nanami Senoo, Dinesh K. Chinthapalli, Matthew G. Baile, et al.

Selected by Barbora Knotkova

Also in the bioinformatics category:

The genome of the colonial hydroid Hydractinia reveals their stem cells utilize a toolkit of evolutionarily shared genes with all animals

Christine E. Schnitzler, E. Sally Chang, Justin Waletich, et al.

Selected by Isabella Cisneros

A Phosphoproteomics Data Resource for Systems-level Modeling of Kinase Signaling Networks

Song Feng, James A. Sanford, Thomas Weber, et al.

Selected by Benjamin Dominik Maier

Similarity metric learning on perturbational datasets improves functional identification of perturbations

Ian Smith, Petr Smirnov, Benjamin Haibe-Kains

Selected by Benjamin Dominik Maier, Anna Foix Romero

Also in the cell biology category:

TLR2 Regulates Hair Follicle Cycle and Regeneration via BMP Signaling

Luyang Xiong, Irina Zhevlakova, Xiaoxia Z. West, et al.

Selected by Marina Schernthanner

Mechano-dependent sorbitol accumulation supports biomolecular condensate

Stephanie Torrino, William M Oldham, Andrés R Tejedor, et al.

Selected by Teodora Piskova

CRISPR-dependent base editing screens identify separation of function mutants of RADX with altered RAD51 regulatory activity

Madison B. Adolph, Atharv S. Garje, Swati Balakrishnan, et al.

Selected by Jessica Chevallier, Pierre Caron

Also in the developmental biology category:

The genome of the colonial hydroid Hydractinia reveals their stem cells utilize a toolkit of evolutionarily shared genes with all animals

Christine E. Schnitzler, E. Sally Chang, Justin Waletich, et al.

Selected by Isabella Cisneros

VEGFA mRNA-LNP promotes biliary epithelial cell-to-hepatocyte conversion in acute and chronic liver diseases and reverses steatosis and fibrosis

Fatima Rizvi, Yu-Ri Lee, Ricardo Diaz-Aragon, et al.

Selected by Organogenesis and Cancer Program

Glial Draper signaling triggers cross-neuron plasticity in bystander neurons after neuronal cell death

Yupu Wang, Ruiling Zhang, Sihao Huang, et al.

Selected by Matthew Davies

Also in the molecular biology category:

Hypoxia blunts angiogenic signaling and upregulates the antioxidant system in elephant seal endothelial cells

Kaitlin N Allen, Julia María Torres-Velarde, Juan Manuel Vazquez, et al.

Selected by Sarah Young-Veenstra

A Phosphoproteomics Data Resource for Systems-level Modeling of Kinase Signaling Networks

Song Feng, James A. Sanford, Thomas Weber, et al.

Selected by Benjamin Dominik Maier

PYCR1 Levels Track with Premature and Chronological Skin Aging

Kortessa Sotiropoulou, Saniye Yumlu, Tomoko Hirano, et al.

Selected by Preethi Krishnaraj

preLists in the biochemistry category:

Preprint Peer Review – Biochemistry Course at UFRJ, Brazil

Communication of scientific knowledge has changed dramatically in recent decades and the public perception of scientific discoveries depends on the peer review process of articles published in scientific journals. Preprints are key vehicles for the dissemination of scientific discoveries, but they are still not properly recognized by the scientific community since peer review is very limited. On the other hand, peer review is very heterogeneous and a fundamental aspect to improve it is to train young scientists on how to think critically and how to evaluate scientific knowledge in a professional way. Thus, this course aims to: i) train students on how to perform peer review of scientific manuscripts in a professional manner; ii) develop students' critical thinking; iii) contribute to the appreciation of preprints as important vehicles for the dissemination of scientific knowledge without restrictions; iv) contribute to the development of students' curricula, as their opinions will be published and indexed on the preLights platform. The evaluations will be based on qualitative analyses of the oral presentations of preprints in the field of biochemistry deposited in the bioRxiv server, of the critical reports written by the students, as well as of the participation of the students during the preprints discussions.

Absolute Scaling of Single-Cell Transcriptomes Reveals Pervasive Hypertranscription in Adult Stem and Progenitor Cells

Share this:

Have your say Cancel reply

Sign up to customise the site to your preferences and to receive alerts

Also in the biochemistry category:

Mechano-dependent sorbitol accumulation supports biomolecular condensate

Proteasome condensate formation is driven by multivalent interactions with shuttle factors and K48-linked ubiquitin chains

Conserved cardiolipin-mitochondrial ADP/ATP carrier interactions assume distinct structural and functional roles that are clinically relevant

Also in the bioinformatics category:

The genome of the colonial hydroid Hydractinia reveals their stem cells utilize a toolkit of evolutionarily shared genes with all animals

A Phosphoproteomics Data Resource for Systems-level Modeling of Kinase Signaling Networks

Similarity metric learning on perturbational datasets improves functional identification of perturbations

Also in the cell biology category:

TLR2 Regulates Hair Follicle Cycle and Regeneration via BMP Signaling

Mechano-dependent sorbitol accumulation supports biomolecular condensate

CRISPR-dependent base editing screens identify separation of function mutants of RADX with altered RAD51 regulatory activity

Also in the developmental biology category:

The genome of the colonial hydroid Hydractinia reveals their stem cells utilize a toolkit of evolutionarily shared genes with all animals

VEGFA mRNA-LNP promotes biliary epithelial cell-to-hepatocyte conversion in acute and chronic liver diseases and reverses steatosis and fibrosis

Glial Draper signaling triggers cross-neuron plasticity in bystander neurons after neuronal cell death

Also in the molecular biology category:

Hypoxia blunts angiogenic signaling and upregulates the antioxidant system in elephant seal endothelial cells

A Phosphoproteomics Data Resource for Systems-level Modeling of Kinase Signaling Networks

PYCR1 Levels Track with Premature and Chronological Skin Aging

preLists in the biochemistry category:

Preprint Peer Review – Biochemistry Course at UFRJ, Brazil

CellBio 2022 – An ASCB/EMBO Meeting

20th “Genetics Workshops in Hungary”, Szeged (25th, September)

Fibroblasts

ASCB EMBO Annual Meeting 2019

EMBL Seeing is Believing – Imaging the Molecular Processes of Life

Cellular metabolism

MitoList

Also in the bioinformatics category:

9th International Symposium on the Biology of Vertebrate Sex Determination

Alumni picks – preLights 5th Birthday

Fibroblasts

Single Cell Biology 2020

Antimicrobials: Discovery, clinical use, and development of resistance

Also in the cell biology category:

The Society for Developmental Biology 82nd Annual Meeting

CSHL 87th Symposium: Stem Cells

Journal of Cell Science meeting ‘Imaging Cell Dynamics’

9th International Symposium on the Biology of Vertebrate Sex Determination

Alumni picks – preLights 5th Birthday

CellBio 2022 – An ASCB/EMBO Meeting

Fibroblasts

EMBL Synthetic Morphogenesis: From Gene Circuits to Tissue Architecture (2021)

FENS 2020

Planar Cell Polarity – PCP

BioMalPar XVI: Biology and Pathology of the Malaria Parasite

Cell Polarity

TAGC 2020

3D Gastruloids

ECFG15 – Fungal biology

ASCB EMBO Annual Meeting 2019

EMBL Seeing is Believing – Imaging the Molecular Processes of Life

Autophagy

Lung Disease and Regeneration

Cellular metabolism

BSCB/BSDB Annual Meeting 2019

MitoList

Biophysical Society Annual Meeting 2019

ASCB/EMBO Annual Meeting 2018

Also in the developmental biology category:

The Society for Developmental Biology 82nd Annual Meeting

CSHL 87th Symposium: Stem Cells

Journal of Cell Science meeting ‘Imaging Cell Dynamics’

9th International Symposium on the Biology of Vertebrate Sex Determination

Alumni picks – preLights 5th Birthday

CellBio 2022 – An ASCB/EMBO Meeting

2nd Conference of the Visegrád Group Society for Developmental Biology

Fibroblasts

EMBL Synthetic Morphogenesis: From Gene Circuits to Tissue Architecture (2021)

EMBL Conference: From functional genomics to systems biology

Single Cell Biology 2020

Society for Developmental Biology 79th Annual Meeting

FENS 2020

Planar Cell Polarity – PCP

Cell Polarity