Clonal transcriptomics identifies mechanisms of chemoresistance and empowers rational design of combination therapies
Curation statements for this article:-
Curated by eLife
eLife Assessment:
This manuscript describes a highly novel barcoding strategy for forward genetic lineage tracing of tumor cells in vitro and in the in vivo environment. The technique, coined WILDseq, can be used to track cells present in vitro which are enriched or depleted in the in vivo environment. Treatment further contributes to clonal expansion and retraction and emergence of populations with sensitivity to alternate agents. The studies are rigorously conducted and are highly impactful.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1 and Reviewer #2 agreed to share their names with the authors.)
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
- Cancer Biology (eLife)
Abstract
Tumour heterogeneity is thought to be a major barrier to successful cancer treatment due to the presence of drug resistant clonal lineages. However, identifying the characteristics of such lineages that underpin resistance to therapy has remained challenging. Here, we utilise clonal transcriptomics with WILD-seq; W holistic I nterrogation of L ineage D ynamics by seq uencing, in mouse models of triple-negative breast cancer (TNBC) to understand response and resistance to therapy, including BET bromodomain inhibition and taxane-based chemotherapy. These analyses revealed oxidative stress protection by NRF2 as a major mechanism of taxane resistance and led to the discovery that our tumour models are collaterally sensitive to asparagine deprivation therapy using the clinical stage drug L-asparaginase after frontline treatment with docetaxel. In summary, clonal transcriptomics with WILD-seq identifies mechanisms of resistance to chemotherapy that are also operative in patients and pin points asparagine bioavailability as a druggable vulnerability of taxane-resistant lineages.
Article activity feed
-
-
Author Response
Reviewer #2 (Public Review):
Wild and colleagues develop a barcoding approach, termed WILD-seq, that combines tumor cell barcoding with single cell transcriptional analysis to concurrently examine clonal tumor cell dynamics and cell state changes during drug treatment. They examine two triple-negative breast cancer (TNBC) cell lines in vivo in response to JQ1 and taxanes. Results from these experiments yield several meaningful conclusions. First, they demonstrate that clonal dynamics are fundamentally distinct depend ending on context and microenvironment, with significant differences observable between cell culture, NSG and immunocompetent mice. Second, they show that bulk expression in treatment refractory tumors represents clonal outgrowth of subpopulations in pretreatment tumors that bear gene expression patterns …
Author Response
Reviewer #2 (Public Review):
Wild and colleagues develop a barcoding approach, termed WILD-seq, that combines tumor cell barcoding with single cell transcriptional analysis to concurrently examine clonal tumor cell dynamics and cell state changes during drug treatment. They examine two triple-negative breast cancer (TNBC) cell lines in vivo in response to JQ1 and taxanes. Results from these experiments yield several meaningful conclusions. First, they demonstrate that clonal dynamics are fundamentally distinct depend ending on context and microenvironment, with significant differences observable between cell culture, NSG and immunocompetent mice. Second, they show that bulk expression in treatment refractory tumors represents clonal outgrowth of subpopulations in pretreatment tumors that bear gene expression patterns similar to the tumor relapsed. Finally, they identify mechanisms of in vivo taxman resistance, including EMT and high NRF2 expression - the latter yielding tumors that show collateral sensitivity to L-asparaginase and subsequent resistance mediated by high levels of asparagine synthetase.
This study is a technical tour de force. The authors deeply engage the complexity of cell barcoding, bottle necking, Hamming analysis, single cell expression analysis and microenvironmental cell analysis. The idea that bulk tumor expression states demarcate drug resistant clonal populations in pre-treatment tumors, while not a new concept, finds critical validation using this approach. Moreover, the use of this approach to examine collateral sensitivity and to identify new strategies to target taxane resistance is compelling.
I support this work but might suggest some comparisons of primary and relapse tumors, as well as the nature of the taxane collateral sensitivity, be further extended.
Major comments:
- The authors suggest that the bulk expression analysis in relapsed tumors mirrors clonal populations in pretreatment tumors (which, while requiring barcoding to validate, somewhat obviates the need for barcoding to identify mechanisms of drug resistance). In cases like EMT, it has been argued that mesenchymal tumor cells survive therapy, but then undergo MET in the relapsed state. Thus, in the long term, tumors may revert to pre-treatment clonal states. It would be interesting to see whether that is the case here - and whether the informative nature of bulk gene expression in the drug resistant tumor is lost over time.
This is an interesting point. We don’t have any direct evidence of any of the tumour cell lineages in our model undergoing EMT or MET from our work, but it is entirely possible that the tumour cells dynamically transition between states over longer time frames that we haven’t captured in our experiments to date. It is also possible that there are intermediate states that we have not captured by sampling at end-point. WILD-seq presents an excellent method for such studies but these are beyond the scope of the current paper.
For such experiments, it would be essential to use barcoded cells to track clonal lineage, otherwise it is impossible to determine whether changes in the EMT of a tumour cell population was driven by a change in the transcriptome/cell state or a shift in clonal abundance. We have added discussion of these points to the discussion section of the manuscript.
With respect to the necessity of barcoding for identifying treatment resistance mechanisms over bulk approaches, lineage-based analysis serves to prioritise pathways that change in the resistance setting that might otherwise be overlooked as being lower down the list of differential expression in bulk analysis. While not specifically addressed here, being able to differentiate between a pre-existing resistance phenotype or an adaptive mechanism of resistance, may also inform the choice of dosing schedule of agents targeting resistant clones.
- Collateral resistance can either refer to the outgrowth of clones that show enhanced sensitivity to distinct therapies or the therapeutic induction of cell states that respond differently to other drugs. To confirm that L-asparaginase sensitivity results from the specific outgrowth of NRF2 clones, it would be meaningful to show that these clones are lost upon L-asparaginase-only treatment and that pretreatment of L-asparaginase promotes long term efficacy of taxanes.
We agree this is a critical question and one that we had already started to address while the manuscript was under review. The Nrf2-high clones are lowly represented in vehicle treated tumours and on the edge of our detection threshold, thus accurate measurements of their depletion by L-asparaginase-only treatment in tumours derived from our heterogeneous WILD-seq clonal pools is very challenging. To address this question, we have instead chosen to isolate individual resistant clones and directly test their response to L-asparaginase. We were able to isolate two of the Nrf2-high clones (751 and 1240) by growing up clones from single cells. After expansion in vitro, these were implanted as pure monoclonal populations and the resulting tumours treated with L-asparaginase. These new data, presented in Fig 7g, demonstrate that tumours derived from these clones (in contrast to tumours derived from our WILD-seq pools) significantly respond to L-asparaginase-only treatment, suggesting that this cell state is a pre-existing intrinsic property of these clones and not one induced by docetaxel treatment.
-
eLife Assessment:
This manuscript describes a highly novel barcoding strategy for forward genetic lineage tracing of tumor cells in vitro and in the in vivo environment. The technique, coined WILDseq, can be used to track cells present in vitro which are enriched or depleted in the in vivo environment. Treatment further contributes to clonal expansion and retraction and emergence of populations with sensitivity to alternate agents. The studies are rigorously conducted and are highly impactful.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1 and Reviewer #2 agreed to share their names with the authors.)
-
Reviewer #1 (Public Review):
This study by Wild et al. investigates mechanisms of resistance to therapy in triple-negative breast cancer (TNBC) as a result of tumor heterogeneity. They perform clonal transcriptomics in TNBC cell using WILD-seq in tumors undergoing BET inhibition or taxane-based chemotherapy. They identify that increased NRF2 activation as a major mechanism of taxane resistance. Importantly, they observe that as a result of NRF2 activation, the resistant cells become dependent on exogenous asparagine which renders them sensitive to asparagine deprivation therapy. Their observations in the animal models are also validated in patient specimens. This is a very elegant study that identifies mechanisms of resistance in TNBC and a new therapeutic approach to overcome the resistance. The combination of models and approaches …
Reviewer #1 (Public Review):
This study by Wild et al. investigates mechanisms of resistance to therapy in triple-negative breast cancer (TNBC) as a result of tumor heterogeneity. They perform clonal transcriptomics in TNBC cell using WILD-seq in tumors undergoing BET inhibition or taxane-based chemotherapy. They identify that increased NRF2 activation as a major mechanism of taxane resistance. Importantly, they observe that as a result of NRF2 activation, the resistant cells become dependent on exogenous asparagine which renders them sensitive to asparagine deprivation therapy. Their observations in the animal models are also validated in patient specimens. This is a very elegant study that identifies mechanisms of resistance in TNBC and a new therapeutic approach to overcome the resistance. The combination of models and approaches used are highly innovative. The authors describe a new barcoding system that allows them to simultaneously define clonal lineage and determine gene expression changes, termed WILD-seq.
-
Reviewer #2 (Public Review):
Wild and colleagues develop a barcoding approach, termed WILD-seq, that combines tumor cell barcoding with single cell transcriptional analysis to concurrently examine clonal tumor cell dynamics and cell state changes during drug treatment. They examine two triple-negative breast cancer (TNBC) cell lines in vivo in response to JQ1 and taxanes. Results from these experiments yield several meaningful conclusions. First, they demonstrate that clonal dynamics are fundamentally distinct depend ending on context and microenvironment, with significant differences observable between cell culture, NSG and immunocompetent mice. Second, they show that bulk expression in treatment refractory tumors represents clonal outgrowth of subpopulations in pretreatment tumors that bear gene expression patterns similar to the …
Reviewer #2 (Public Review):
Wild and colleagues develop a barcoding approach, termed WILD-seq, that combines tumor cell barcoding with single cell transcriptional analysis to concurrently examine clonal tumor cell dynamics and cell state changes during drug treatment. They examine two triple-negative breast cancer (TNBC) cell lines in vivo in response to JQ1 and taxanes. Results from these experiments yield several meaningful conclusions. First, they demonstrate that clonal dynamics are fundamentally distinct depend ending on context and microenvironment, with significant differences observable between cell culture, NSG and immunocompetent mice. Second, they show that bulk expression in treatment refractory tumors represents clonal outgrowth of subpopulations in pretreatment tumors that bear gene expression patterns similar to the tumor relapsed. Finally, they identify mechanisms of in vivo taxman resistance, including EMT and high NRF2 expression - the latter yielding tumors that show collateral sensitivity to L-asparaginase and subsequent resistance mediated by high levels of asparagine synthetase.
This study is a technical tour de force. The authors deeply engage the complexity of cell barcoding, bottle necking, Hamming analysis, single cell expression analysis and microenvironmental cell analysis. The idea that bulk tumor expression states demarcate drug resistant clonal populations in pre-treatment tumors, while not a new concept, finds critical validation using this approach. Moreover, the use of this approach to examine collateral sensitivity and to identify new strategies to target taxane resistance is compelling.
I support this work but might suggest some comparisons of primary and relapse tumors, as well as the nature of the taxane collateral sensitivity, be further extended.
Major comments:
1. The authors suggest that the bulk expression analysis in relapsed tumors mirrors clonal populations in pretreatment tumors (which, while requiring barcoding to validate, somewhat obviates the need for barcoding to identify mechanisms of drug resistance). In cases like EMT, it has been argued that mesenchymal tumor cells survive therapy, but then undergo MET in the relapsed state. Thus, in the long term, tumors may revert to pre-treatment clonal states. It would be interesting to see whether that is the case here - and whether the informative nature of bulk gene expression in the drug resistant tumor is lost over time.
2. Collateral resistance can either refer to the outgrowth of clones that show enhanced sensitivity to distinct therapies or the therapeutic induction of cell states that respond differently to other drugs. To confirm that L-asparaginase sensitivity results from the specific outgrowth of NRF2 clones, it would be meaningful to show that these clones are lost upon L-asparaginase-only treatment and that pretreatment of L-asparaginase promotes long term efficacy of taxanes.
-
Reviewer #3 (Public Review):
Wild et al examined several syngeneic breast cancer cell lines using the novel cell tracing technology they have developed. This identified subpopulations of cells that were differentially represented in the in vitro pool and when grown in vivo. This difference extended to immunodeficient as compared to immunocompetent hosts. Treatment altered the dynamics of clonal persistence, allowing the authors to interrogate pathways expressed in cells in response to treatment with a BET inhibitor or docetaxel, revealing the exciting finding that taxane resistant subclones were dependent on NRF2 and emerged with a collateral sensitivity to l-aspariginase, which was demonstrated in vivo, and further observed in human tumor samples.
Overall, this is an exciting, elegant, and rigorously developed strategy, which gives a …
Reviewer #3 (Public Review):
Wild et al examined several syngeneic breast cancer cell lines using the novel cell tracing technology they have developed. This identified subpopulations of cells that were differentially represented in the in vitro pool and when grown in vivo. This difference extended to immunodeficient as compared to immunocompetent hosts. Treatment altered the dynamics of clonal persistence, allowing the authors to interrogate pathways expressed in cells in response to treatment with a BET inhibitor or docetaxel, revealing the exciting finding that taxane resistant subclones were dependent on NRF2 and emerged with a collateral sensitivity to l-aspariginase, which was demonstrated in vivo, and further observed in human tumor samples.
Overall, this is an exciting, elegant, and rigorously developed strategy, which gives a new level of insight into the tumor biology. The authors make an important discovery regarding dynamic intratumoral response to systemic therapy.
-
-