Disrupting cellular memory to overcome drug resistance

Guillaume Harmange, Raúl A. Reyes Hueros, Dylan Schaff, Benjamin Emert, Michael Saint-Antoine, Shivani Nellore, Mitchell E. Fane, Gretchen M. Alicea, Ashani T. Weeraratna, Abhyudai Singh, Sydney M. Shaffer

Posted on: 18 September 2022 , updated on: 18 March 2024

Preprint posted on 5 August 2022

Article now published in Nature Communications at http://dx.doi.org/10.1038/s41467-023-41811-8

Predicting the future of drug naïve tumor cells by looking at their past.

Selected by Nate Mullin

Categories: bioinformatics, cell biology, genetics, genomics, molecular biology

Updated 18 March 2024 with a postLight by Nate Mullin

A version of this study by Harmange and colleagues is now published in Nature Communications, a bit over a year after the preprint was initially posted on bioRxiv. The initial work was quite robust, and both introduced and validated a new technique, scMemorySeq, and demonstrated its utility in a clinically relevant model of drug resistance in cancer cells.

The main changes to the paper during the peer review process appear to revolve around the organization and logical presentation of experiments. In the preprint, Figure 1 both introduced scMemorySeq and showed data from a drug treatment experiment demonstrating that cell state switching prior to treatment could alter the survival dynamics of a population. In the Nature Communications paper, this experiment has been moved to the last primary figure (Figure 6) and is presented with a couple of additional conditions. Figure 6 now shows how cell states can be modulated in either direction by drugging pathways identified in previous figures.

The order of Figures 2 and 3 has also been flipped. I believe these changes help to create a new logic to the paper in which cell states (primed vs drug-susceptible) are first observed using scMemorySeq, then validated through other tumor models, followed by molecular dissection of their differences and leveraging these differences to show state modulation in the final figure.

The peer review file from Nature Communications is also available for this paper and contains an informative discussion between authors and reviewers on both the points I have discussed here and on analytical methods that were used (and shown in supplemental figures) to strengthen the conclusions of the paper.

Background

Is there a way to understand, or even predict, the seemingly random behavior of certain cells? If so, how could such an understanding be leveraged to combat insidious cellular behaviors such as chemotherapy resistance?

In normal development and in disease states like cancer, the character of individual cells is malleable. Two descriptions of what a cell may be able to do at any particular time are cell state and cell type. While the exact distinction between these categories remains open to debate, cell type is generally regarded as a permanent description based on developmental lineage and function. Cell state exists on the opposite end of the permanence spectrum and can be thought of as a transient set of characteristics taken up by a cell, perhaps in response to an external stimulus or perhaps due only to random chance.

The preprint by Harmange et al. focuses on a type of cellular identity that sits in between these two extremes: cell memory. Shaffer and colleagues previously described cell memory as a heritable (across cell division), yet impermanent feature of a cell. For example, gene expression memory would describe a state of gene expression present in a cell for a certain amount of time that might even be passed into daughter cells yet is not part of the core program that defines the cell’s identity (i.e., once the transient time period ends, the cell would not lose its identity nor cease to be that type).

In the context of cancer, gene expression memory states have important clinical implications. Cell memory may confer drug resistance to certain lineages of cells, leading to growth of drug-resistant tumors after chemotherapy. Because of the transient nature of gene expression memory, these states are not genetically encoded, and may be gained or lost in particular lineages across time. Harmange et al. used the well-described melanoma model of drug resistance to BRAF and MEK inhibitors to define memory states of cancer cells primed to resist or succumb to drug treatment. They then went on to experimentally identify the pathway that allows state switching in these cells prior to drug treatment.

Key findings

Single-cell memory sequencing (scMemorySeq) combines cellular lineage tracing with gene expression profiling through transient cell state transitions.

In order to understand the transcriptional landscape of cells in different states (i.e. primed for drug resistance or susceptible to treatment) and to track lineages of cells that transition between states, the authors developed scMemorySeq (Figure 1). This technique combines lineage tracing via lentiviral barcoding with phenotyping and single-cell gene expression analysis. ScMemorySeq allows the past behavior of a cell to be linked to its current cell state.

**Figure 1 scMemorySeq overview.** Cells were permanently tagged with a heritable barcode that could be read via polyA-enriched RNA sequencing. Cell state (i.e., primed for resistance) was measured and cells were allowed to expand. Rare state-switching events that occur during expansion are shown in the schematic. Finally, the transcriptome of individual cells was read out, and lineages were reconstructed based on embedded barcodes. *Adapted from Harmange et al. 2022.*

Lineages that switch between susceptible and resistance-primed states are transcriptomically distinct from those that do not.

Related cells (i.e. with a common progenitor) can change state to either gain or lose the transcriptional profile associated with drug susceptibility. By comparing lineages of cells that make this switch to lineages that do not, key genes involved in state conversion can be identified (Figure 2). These can be aggregated into a gene expression module (“crossing lineage score”) that individual cells can be scored on. The authors identified epithelial mesenchymal transition (EMT)-related genes as enriched in state-switching lineages.

**Figure 2 State-switching characterization by scMemorySeq**. Reconstructed lineages identified cells that had switched from primed to susceptible or vice versa during the pooled expansion. Comparing the gene expression profiles of switched lineages allowed for the identification of state switching-associated genes. *Adapted from Harmange et al. 2022.*

Modulation of drug-response behavior through targeting of transition-mediating pathways.

The authors targeted the pathways involved in state switching identified by scMemorySeq using agonists and antagonists of the TGFB and PI3K pathways. Inhibition of PI3K in a mixed population of primed and susceptible cells led to a decrease in the proportion of primed cells, whereas activation of TGFB signaling increased the proportion of primed cells (Figure 3). Transcriptionally, newly primed or susceptible cells resembled those without treatment (Figure 3F).

**Figure 3 Targeting pathways involved in state-switching can modulate the proportion of cells in different states**. Primed and drug-susceptible states could be identified based on gene expression profile. TGFB1 activation increased the proportion of cells in the primed state, while PI3K inhibition pushed cells toward drug susceptibility. Notably, the cells in either state following treatment with TGFB or PI3K-targetting compounds resembled untreated cells of the same state transcriptionally. *Adapted from Harmange et al. 2022.*

PI3K modulation increases the proportion of drug-susceptible cells through state-switching.

State-switching behavior was modeled using a simulation taking into account three parameters: the rate of state switching, the rate of cell death, and the rate of cell division. The authors found that an increased rate of state switching best explained their experimental data. This finding supports the hypothesis that drug treatment alters population dynamics through increased switching, not by killing off certain populations. The authors could finally show that pre-treatment with state-switching drugs does in fact increase the effectiveness of BRAFi/MEKi treatment against drug-naïve melanoma cells in vitro.

What I like about this work

In recent years, single-cell genomic and transcriptomic techniques have put the spotlight on cellular heterogeneity in the context of disease and normal physiology. However, such are mostly restricted to a single point in time. This means that while we can observe sub-populations of cells within a tissue, we cannot know how they have in the past or would in the future react to particular insults or unique environments. Such knowledge is essential for understanding how disease develops (e.g. why are some, but not all cells of a certain type affected by a genetic variant) and how transient subpopulations could best be targeted in the treatment of disease

In this preprint, the authors have demonstrated how scMemorySeq can be used to overcome this temporal hurdle. They showed that even within validated clusters, pre-fated trajectories are coded into individual cells. As a reader and researcher interested in genetic disease, the concept that these fate trajectories are knowable and detectable with current technology shifts how I think about cellular heterogeneity.

Questions for authors

Is state-switching itself a lineage-specific phenomenon? Do certain lineages switch states more frequently than others?
Similarly, what do you think changes a lineage’s ability to switch states? Could this be genetically encoded, or are all cells genetically identical?
Are transient gene expression memory states important outside of cancer? Do you think that these underly cell fate decisions in normal development?

doi: https://doi.org/10.1242/prelights.32390

Read preprint

(No Ratings Yet)

Author's response

Guillaume Harmange shared

Is state-switching itself a lineage-specific phenomenon? Do certain lineages switch states more frequently than others?

We have not specifically tested whether all lineages undergo state-switching, or if some lineages switch states more frequently than others, however, our observations in the lab suggest that all lineages have roughly the same state-switching rate. One piece of evidence for consistent state-switching rates between lineages is that when we grow single cell clones from these cell lines, the percentage of primed cells in each clone is very consistent, suggesting similar rates of state-switching in each clonal population.

Similarly, what do you think changes a lineage’s ability to switch states? Could this be genetically encoded, or are all cells genetically identical?

The cells used in this study are clonal cell lines, therefore the cells in the population should be very similar to each other genetically. We have done experiments in which we isolate individual clones and found both states regardless of the clone. These experiments show that it is not a single genetic clone that is capable of state-switching. Therefore, we believe genetics is not the main determinant for the ability of these cells to switch states. The root cause of state-switching is still an open question. Based on the findings of this paper, what we can say is that cells seem to switch into the primed state if the TGFB signaling pathway is activated, and can switch back to the drug-susceptible state by stopping PI3K signaling. However, the endogenous triggers (or the rate-limiting factors) for these signaling pathways remain to be fully characterized.

Are transient gene expression memory states important outside of cancer? Do you think that these underlie cell fate decisions in normal development?

This is an interesting question that we do not think has been fully explored yet. Developmental systems could be an interesting application for scMemory-seq and can help determine if there are memory states important in the development process. To our knowledge, there are no studies specifically using these techniques to quantify memory of gene expression in development. However, Weinreb et al. is an example where they used cellular barcoding in the context of development to demonstrate that cell lineage is an important factor in determining cell fate (https://www.science.org/doi/10.1126/science.aaw3381).

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 bioinformatics category:

Deep learning-based predictions of gene perturbation effects do not yet outperform simple linear methods

Constantin Ahlmann-Eltze, Wolfgang Huber, Simon Anders

Selected by 11 November 2024

Benjamin Dominik Maier

Functional Diversity of Memory CD8 T Cells is Spatiotemporally Imprinted

Miguel Reina-Campos, Alexander Monell, Amir Ferry, et al.

Selected by 22 August 2024

Marina Schernthanner

Enhancer-driven cell type comparison reveals similarities between the mammalian and bird pallium

Nikolai Hecker , Niklas Kempynck , David Mauduit, et al.

Selected by 02 July 2024

Rodrigo Senovilla-Ganzo

Also in the cell biology category:

Motor Clustering Enhances Kinesin-driven Vesicle Transport

Rui Jiang, Qingzhou Feng, Daguan Nong, et al.

Selected by 16 November 2024

Sharvari Pitke

Cellular signalling protrusions enable dynamic distant contacts in spinal cord neurogenesis

Joshua Hawley, Robert Lea, Veronica Biga, et al.

Selected by 15 November 2024

Ankita Walvekar

Green synthesized silver nanoparticles from Moringa: Potential for preventative treatment of SARS-CoV-2 contaminated water

Adebayo J. Bello, Omorilewa B. Ebunoluwa, Rukayat O. Ayorinde, et al.

Selected by 14 November 2024

Safieh Shah, Benjamin Dominik Maier

Discussion

Also in the genetics category:

Intracellular diffusion in the cytoplasm increases with cell size in fission yeast

Catherine Tan, Michael C. Lanz, Matthew Swaffer, et al.

Selected by 18 October 2024

Leeba Ann Chacko, Sameer Thukral

Discussion

HIF1A contributes to the survival of aneuploid and mosaic pre-implantation embryos

Estefania Sanchez-Vasquez, Marianne E. Bronner, Magdalena Zernicka-Goetz

Selected by 11 October 2024

Anchel De Jaime Soguero

Significantly reduced, but balanced, rates of mitochondrial fission and fusion are sufficient to maintain the integrity of yeast mitochondrial DNA

Brett T. Wisniewski, Laura L. Lackner

Selected by 30 August 2024

Leeba Ann Chacko

Discussion

Also in the genomics category:

A fine kinetic balance of interactions directs transcription factor hubs to genes

Apratim Mukherjee, Samantha Fallacaro, Puttachai Ratchasanmuang, et al.

Selected by 23 July 2024

Deevitha Balasubramanian

Enhancer-driven cell type comparison reveals similarities between the mammalian and bird pallium

Nikolai Hecker , Niklas Kempynck , David Mauduit, et al.

Selected by 02 July 2024

Rodrigo Senovilla-Ganzo

Modular control of time and space during vertebrate axis segmentation

Ali Seleit, Ian Brettell, Tomas Fitzgerald, et al.

AND

Natural genetic variation quantitatively regulates heart rate and dimension

Jakob Gierten, Bettina Welz, Tomas Fitzgerald, et al.

Selected by 24 June 2024

Girish Kale, Jennifer Ann Black

Also in the molecular biology category:

Green synthesized silver nanoparticles from Moringa: Potential for preventative treatment of SARS-CoV-2 contaminated water

Adebayo J. Bello, Omorilewa B. Ebunoluwa, Rukayat O. Ayorinde, et al.

Selected by 14 November 2024

Safieh Shah, Benjamin Dominik Maier

Discussion

Non-disruptive inducible labeling of ER-membrane contact sites using the Lamin B Receptor

Laura Downie, Nuria Ferrandiz, Megan Jones, et al.

Selected by 15 October 2024

Jonathan Townson

HIF1A contributes to the survival of aneuploid and mosaic pre-implantation embryos

Estefania Sanchez-Vasquez, Marianne E. Bronner, Magdalena Zernicka-Goetz

Selected by 11 October 2024

Anchel De Jaime Soguero

preLists in the bioinformatics category:

‘In preprints’ from Development 2022-2023

A list of the preprints featured in Development's 'In preprints' articles between 2022-2023

Disrupting cellular memory to overcome drug resistance

Background

Key findings

Single-cell memory sequencing (scMemorySeq) combines cellular lineage tracing with gene expression profiling through transient cell state transitions.

Lineages that switch between susceptible and resistance-primed states are transcriptomically distinct from those that do not.

Modulation of drug-response behavior through targeting of transition-mediating pathways.

PI3K modulation increases the proportion of drug-susceptible cells through state-switching.

What I like about this work

Questions for authors

Share this:

Have your say Cancel reply

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

Also in the bioinformatics category:

Deep learning-based predictions of gene perturbation effects do not yet outperform simple linear methods

Functional Diversity of Memory CD8 T Cells is Spatiotemporally Imprinted

Enhancer-driven cell type comparison reveals similarities between the mammalian and bird pallium

Also in the cell biology category:

Motor Clustering Enhances Kinesin-driven Vesicle Transport

Cellular signalling protrusions enable dynamic distant contacts in spinal cord neurogenesis

Green synthesized silver nanoparticles from Moringa: Potential for preventative treatment of SARS-CoV-2 contaminated water

Also in the genetics category:

Intracellular diffusion in the cytoplasm increases with cell size in fission yeast

HIF1A contributes to the survival of aneuploid and mosaic pre-implantation embryos

Significantly reduced, but balanced, rates of mitochondrial fission and fusion are sufficient to maintain the integrity of yeast mitochondrial DNA

Also in the genomics category:

A fine kinetic balance of interactions directs transcription factor hubs to genes

Enhancer-driven cell type comparison reveals similarities between the mammalian and bird pallium

Modular control of time and space during vertebrate axis segmentation

Natural genetic variation quantitatively regulates heart rate and dimension

Also in the molecular biology category:

Green synthesized silver nanoparticles from Moringa: Potential for preventative treatment of SARS-CoV-2 contaminated water

Non-disruptive inducible labeling of ER-membrane contact sites using the Lamin B Receptor

HIF1A contributes to the survival of aneuploid and mosaic pre-implantation embryos

preLists in the bioinformatics category:

‘In preprints’ from Development 2022-2023

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:

BSCB-Biochemical Society 2024 Cell Migration meeting

‘In preprints’ from Development 2022-2023

preLights peer support – preprints of interest

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 genetics category:

BSDB/GenSoc Spring Meeting 2024

BSCB-Biochemical Society 2024 Cell Migration meeting

9th International Symposium on the Biology of Vertebrate Sex Determination

Alumni picks – preLights 5th Birthday

Semmelweis Symposium 2022: 40th anniversary of international medical education at Semmelweis University

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

2nd Conference of the Visegrád Group Society for Developmental Biology

EMBL Conference: From functional genomics to systems biology

TAGC 2020

ECFG15 – Fungal biology

Autophagy