Sphingolipids control dermal fibroblast heterogeneity

Science, Journal Year: 2022, Volume and Issue: 376(6590)

Published: April 14, 2022

Human cells produce thousands of lipids that change during cell differentiation and can vary across individual the same type. However, we are only starting to characterize function these cell-to-cell differences in lipid composition. Here, measured lipidomes transcriptomes human dermal fibroblasts by coupling high-resolution mass spectrometry imaging with single-cell transcriptomics. We found variations specific metabolic pathways contribute establishment states involved organization skin architecture. Sphingolipid composition is shown define fibroblast subpopulations, sphingolipid rewiring driving cell-state transitions. Therefore, heterogeneity affects determination states, adding a new regulatory component self-organization multicellular systems.

Language: Английский

---

Variability within rare cell states enables multiple paths toward drug resistance

Nature Biotechnology, Journal Year: 2021, Volume and Issue: 39(7), P. 865 - 876

Published: Feb. 22, 2021

Language: Английский

---

Diverse clonal fates emerge upon drug treatment of homogeneous cancer cells

Nature, Journal Year: 2023, Volume and Issue: 620(7974), P. 651 - 659

Published: July 19, 2023

Language: Английский

---

What shapes eukaryotic transcriptional bursting?

Molecular BioSystems, Journal Year: 2017, Volume and Issue: 13(7), P. 1280 - 1290

Published: Jan. 1, 2017

Isogenic cells in a common environment present large degree of heterogeneity gene expression. Part this variability is attributed to transcriptional bursting: the stochastic activation and inactivation promoters that leads discontinuous production mRNA. The diversity bursting patterns displayed by different genes suggests existence connection between regulation. Experimental strategies such as single-molecule RNA FISH, MS2-GFP or short-lived protein reporters allow quantification comparison kinetics conditions, allowing therefore identification molecular mechanisms modulating bursting. In review we recapitulate impact on aspects transcription chromatin environment, nucleosome occupancy, histone modifications, number affinity regulatory elements, DNA looping factor availability. More specifically, examine their role tuning burst size frequency. While some involved marks can affect every aspect bursting, others predominantly influence (e.g. cis-regulatory elements) frequency availability).

Language: Английский

---

Rare Cell Detection by Single-Cell RNA Sequencing as Guided by Single-Molecule RNA FISH

Cell Systems, Journal Year: 2018, Volume and Issue: 6(2), P. 171 - 179.e5

Published: Feb. 1, 2018

Language: Английский

---

Mammalian gene expression variability is explained by underlying cell state

Molecular Systems Biology, Journal Year: 2020, Volume and Issue: 16(2)

Published: Feb. 1, 2020

Article11 February 2020Open Access Transparent process Mammalian gene expression variability is explained by underlying cell state Robert Foreman Institute for Quantitative and Computational Biosciences, University of California, Los Angeles, CA, USA Program in Bioinformatics Systems Biology, San Diego, Search more papers this author Roy Wollman Corresponding Author [email protected] orcid.org/0000-0003-3865-2605 Department Integrative Biology Physiology, Chemistry Biochemistry, Information Foreman1,2 *,1,2,3 1Institute 2Program 3Department *Corresponding author. Tel: +1 858 210 0905; E-mail: Molecular (2020)16:e9146https://doi.org/10.15252/msb.20199146 PDFDownload PDF article text main figures. Peer ReviewDownload a summary the editorial decision including letters, reviewer comments responses to feedback. ToolsAdd favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Abstract Gene mammalian systems plays an important role physiological pathophysiological conditions. This can come from differential regulation related (extrinsic) allele-specific transcriptional bursting (intrinsic). Yet, relative contribution these two distinct sources unknown. Here, we exploit qualitative difference patterns covariance between quantify their contributions variance cells. Using multiplexed error robust RNA fluorescent situ hybridization (MERFISH), measured multivariate distribution 150 genes Ca2+ signaling coupled with dynamic response live cells ATP. We show that after controlling cellular phenotypic states such as size, cycle stage, ATP, remaining effectively at Poisson limit most genes. These findings demonstrate majority results differences relatively minimal. Synopsis Single-cell measurements dynamics, cycle, signaling-related not gene-specific. The gene-specific approaches Cell information, i.e. ligand was alongside single-cell using MERFISH. observed be 13 features state. unexplained Introduction ubiquitous all biological systems. In multicellular organisms, heterogeneity different types confers specialized function giving rise complexity whole-system behavior (Raj van Oudenaarden, 2008; Eldar Elowitz, 2010; Symmons Raj, 2016; Suo et al, 2018; Tabula Muris Consortium 2018). Similarly, organisms viruses were shown utilize population level create diverse phenotypes, bet-hedging strategies changing environments (Veening Vega Gore, 2014; Rouzine 2015). While provide useful functional organism or population, it necessarily always beneficial 2016). Unregulated stochastic events, i.e., noise, cells' ability respond accurately introduce have negative overall fitness. Indeed, many mechanisms buffering (Stoeger 2016) feedback loops (Jangi Sharp, Schmiedel 2015) been suggested detrimental effect variability. Quantification cause step toward determining what degree represents uncontrolled "noise" stratification function. Two key contributors are global factors analysis powerful approach decompose into classes. Landmark works used investigate bacterial cells, which laid foundation decomposing (intrinsic) originate affect multiple alleles relate (Elowitz, 2002; Paulsson, 2005). work later extended yeast (Raser O'Shea, 2004) 2006; Sigal Singh 2012). decomposition components simple. Allele-specific noise upstream component propagate downstream (Sigal 2006), whereas temporal fluctuations shared nontrivial consequences on distributions (Paulsson, 2004; Pedraza 2005; Shahrezaei 2008). Finally, use terms "intrinsic" "extrinsic" sometimes ill-defined some models include "coupled intrinsic" mode well, form hence (Rodriguez 2019). Despite confusing nomenclature, distinguish approach. addition covariance-based approaches, relationship mean provides quantitative framework gain insights (Munsky comparison straightforward scales its mean. Three statistical tools commonly describe normalized variance: coefficient variation (CV), CV squared (CV2), Fano factor. CV2 both unitless measures where defined standard deviation divided simply squared, squared. compare scale because nature. third measure, factor, therefore unitless, but has special property being equal one case process. Many processes ratio least so factor define "standard dispersion", result, smaller/bigger than considered under/over-dispersed, respectively. Therefore, simple quantification scaled mechanism generating (Choubey 2015; Hansen 2018a). Multiple studies across bacteria, yeast, over-dispersed distributions. observation interpretations. One interpretation over-dispersion result superposition (Battich other itself (Suter 2011; Dar Corrigan Tantale latter popularized introduction phenomenological model named two-state random telegraph represented existing either "on" "off" (Peccoud Ycart, 1995; Kepler Elston, 2001; Thattai Kaern Friedman Raj Swain, Suter Molina 2013; Sanchez Golding, Fukaya Lenstra More complex also Zoller Nicolas 2018) does change way. suggest transcription should occur bursts transcripts generated when "on". described parameters: burst size frequency control resulting lower larger contributing distribution. Overall, interpretations, state, explain over-dispersion. There mounting evidence genes, variables rather intrinsically noisy Nonetheless, still widely (Larsson 2019; Ochiai 2019) calling systematic investigation. very understand modern world highly measurements. new technologies revealing structure "cell space" occupying large array (Han Rosenberg 2018), (Trapnell, Cheng 2019), fronts (Shoval 2012) reflect stratification. our knowledge manifest heterogeneity, total interpreted arising alone gap understanding hindering progress assigning roles (Dueck To address gap, utilized properties variability: dispersion. dispersion joint individual each cell, measured, well measurement expression. sequential smFISH (MERFISH implementation) (Moffitt allowed us measure ~ 5000 single Since same pathway higher compared functions Stewart-Ornstein 2012), focused network function, ATP epithelial wound healing (Funaki Handly Wollman, 2017). advantage < 15 min, fast timescale precludes any ATP-induced changes transcription. combined dataset, able separate correlated uncorrelated linear regression guided matrix. found removing components, shows little measured. indicate only minor contributor Results assess stems vs variability, took fact signatures. Figure 1 simulated data illustrate how signatures By definition, gene, due will likely similar (Fig 1A). When dominates 1B top) regulatory small contribution, there correlation remain largely unchanged conditioning top right). residual intrinsic greater one. On hand, bottom), reduce At limit, accounted for, zero residuals one, bottom comparable middle), final even middle Conditioning dual therefore, possible whether removed obvious extrinsic conditioned out, confidently interpret under- over-dispersed. 1. Transcriptional trans-acting causes cell-to-cell Cartoon depicting activities trans-factor (TF) molecules effects bursting. Simulated showing three example cases: dominated (top panels), mixture (middle three), (bottom three). diminished levels After factors, potentially significantly smaller (right). Inset Pearson A B (brown) (blue). Download figure PowerPoint situations above requires accurate sufficient number correlate achieve this, developed experimental protocol combines MERFISH, counting (Chen Moffitt rich profiling 2). MCF10A mammary line, often nontransformed nature accessibility imaging (Selimkhanov Qu share function: involvement network, tissue wounding (Minns Trinkaus-Randall, Justet advantages (i) expect high (Stewart-Ornstein (ii) fast, emergent phenotype min 2A), faster (Shamir protocol, rapidly fixed (10–15 stimulation fixation, Appendix Fig S1), unlikely changed agonist. 2. Paired MERFISH calcium Experimental overview—live imaged before spots over several rounds aligned encoded specific series dark bright throughout hybridization. Left, representative trajectories demonstrating heterogeneous stimulation, left. right panel image plot 5000+ successfully paired Cellular volume per shown. marker cycle-related derive g2m score (coloring). Right, left S cell. Correlation (ATP2A2) marks differentiation status (CD44). Table categories complete list (MLR) model. multiplexing scheme transcript identity barcode-based, barcodes During round, dye-labeled oligos hybridized subset species sample imaged, appears diffraction-limited spots; then, dye quenched, repeated until barcode "bits" imaged. linking rounds, decode identifying images spot XYZ coordinate 2B). combinatorial labeling allows exponential scaling rounds. mostly limited built-in correction experiment, 24 (eight hybs × colors) molecule labeled four An 2B. 131 annotated involved (Kanehisa Goto, 2000; Bandara Kanehisa 17 mark stages (Whitfield 2002), sub-differentiated (Qu estimate detection efficiency 95.5% false-positive rate 1% Overall spearman bulk RNAseq 0.84 (Appendix S3). Our based factors. would ideal directly causatively accessible measurements, e.g., causative Given probe signaling, first extracted time cytoplasmic calibrated GCaMP5 biosensor S2A). 2C) showed response, qualitatively quantitatively previous mixed wide range phenotypes (Yao feature-based representation represent anticipate 2G S2). few Specifically, volume, markers 2D–F). As past, strongly 2D) indicating must (das Neves Shalek Battich Padovan-Merhar However, strength complexly S4). stage 2E F) (Buettner Ca2+-related focusing augmented established like expected capture comes heterogeneity. consistent widespread differentiation-related transcriptome 3A scatter plots (ATP2A2 RRM1) plotted against markers, feature. indeed features. amount feature MLR independent 85% median 0.62 3B). feature, looked fraction explanatory power category 3C). power, contribute meaningfully variance. had cases, pure sampling, permutation-based testing here statistically 3D). 3. Decomposition Representative (rows) (columns). percent corner. histogram gene. Stacked bar MLR. significance estimated Z-scoring slope null 1,600 bootstrapped shuffled slopes. significant [adjusted P-value (Bonferroni) 0.05]. Whiskers 1.5 times interquartile range. uniqueness fashion allowing estimation 4A matrix raw counts, counts expected, increase included MLR, gene-to-gene goes down. Interestingly, full includes identical significantly. correlation, principal (PCA; 4B). Without > 40% reduced substantially 10% variance, substantial reduction demonstrates condition away components. Still, completely removed, added another term PCA taking account. sufficiently captured With last "hidden" close values data. captures information required 4. Residual contains less covariation within Gene–gene matrices Explained increasing dotted line shuffled. boxplots. whiskers Dashed expectation technical noise. Scatter dashed line. wanted determine on, saw magnitudes 4C). hidden included, Note do perform noise; so, theoretical. means log–log 4D). proximity levels. super-Poissonian It unclear they

Language: Английский

---

Signal Percolation within a Bacterial Community

Cell Systems, Journal Year: 2018, Volume and Issue: 7(2), P. 137 - 145.e3

Published: July 25, 2018

Signal transmission among cells enables long-range coordination in biological systems. However, the scarcity of quantitative measurements hinders development theories that relate signal propagation to cellular heterogeneity and spatial organization. We address this problem a bacterial community employs electrochemical cell-to-cell communication. developed model based on percolation theory, which describes how signals propagate through heterogeneous medium. Our predicts becomes possible when is organized near critical phase transition between disconnected fully connected conduit signaling cells. By measuring population-level with single-cell resolution wild-type genetically modified communities, we confirm distribution at predicted transition. findings suggest point, benefit outweighs level cost. The thus appears be according theoretically promotes efficient transmission.

Language: Английский

---

A Post-Transcriptional Feedback Mechanism for Noise Suppression and Fate Stabilization

Cell, Journal Year: 2018, Volume and Issue: 173(7), P. 1609 - 1621.e15

Published: May 10, 2018

Language: Английский

---

Contribution of RNA Degradation to Intrinsic and Extrinsic Noise in Gene Expression

Cell Reports, Journal Year: 2019, Volume and Issue: 26(13), P. 3752 - 3761.e5

Published: March 1, 2019

Genetically identical cells contain variable numbers of molecules, even if the share same environment. This stochastic variability is prominent when molecules have low abundance, which case for mRNA noise. Most studies focused on how transcription affects noise, and little known about role RNA degradation. To discriminate fluctuations in these processes during decay a pair reporter mRNAs, we quantified uncorrelated intrinsic correlated extrinsic noise using single-molecule FISH. Intrinsic converges to Poisson level decay. mRNAs that short half-life are more susceptible than stable mRNAs. However, Xrn1 exonuclease NMD pathways, degrade rapidly, were found lower fluctuation, mitigates short-lived permits across entire range half-lives.

Language: Английский

---

Gene Networks with Transcriptional Bursting Recapitulate Rare Transient Coordinated High Expression States in Cancer

Cell Systems, Journal Year: 2020, Volume and Issue: 10(4), P. 363 - 378.e12

Published: April 1, 2020

Language: Английский

---