Mapping the niche of breast cancer metastases in lung and liver

  1. Magdalena K. Sznurkowska
  2. Francesc Castro-Giner
  3. Yongzhan Zhang
  4. Alexander Ring
  5. Fabienne Dominique Schwab
  6. Francesca Albrecht
  7. Kirsten D. Mertz
  8. Karin Strittmatter
  9. Lauren L. Ozimski
  10. Selina Budinjas
  11. Gaël Auray
  12. Claudio Giachino
  13. Verdon Taylor
  14. Ana Gvozdenovic
  15. Ilaria Malanchi
  16. Heike Frauchiger-Heuer
  17. Andreas Wicki
  18. Marcus Vetter
  19. Nicola Aceto

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

Breast cancer progression to visceral organs such as lung and liver is regarded as a dreadful event, unequivocally associated with a poor prognosis. Yet, these vital sites are characterized by highly diverse cellular microenvironments and physiological functions, suggesting that they may influence cancer cells behavior in divergent ways. Unexpectedly, we find that while the liver microenvironment fosters metastasis-boosting properties of its metastatic seeds and favors secondary spread, the lungs impose a drastic roadblock to the same processes. Using patient data and mouse models of breast cancer that spontaneously metastasize to visceral organs, coupled with genetic barcode-mediated metastasis tracing, niche labeling technology and single cell analysis of both tumor cells and their direct microenvironment at each site, we characterize this dichotomy and identify differential microenvironmental factors that contribute to liver vs lung metastasis biology. Mechanistically, we uncover pulmonary cytokines, such as Granulin, whose activity naturally blocks metastatic properties. Conversely, we identify BMP2-producing endothelial cells within the liver metastatic niche, capable to enhance further spread. Together, we reveal an unexpected contrast in the site-specific behavior of visceral metastases in triple-negative breast cancer models, highlighting microenvironmental factors that contribute to its geographical diversity, as well as organ-specific opportunities for intervention.

Article activity feed

  1. Life Science Editors Foundation

    Review coordinated by Life Science Editors Foundation Reviewed by: Dr. Angela Andersen, Life Science Editors Foundation & Life Science Editors. Potential Conflicts of Interest: None.

    Punchline:The liver and lung microenvironments influence various phenotypes of metastasized triple-negative breast cancer (TNBC) cells in different ways. Liver microenvironment accelerates disease progression and negatively impacts patient outcomes. Interesting paper and seems well done, identifies specific players in the different environment – consistent with the concept that the niche affects the phenotype of metastatic outgrowths.

    Why is this an important problem? Metastasis is the primary cause of cancer-related deaths. Understanding how different organ environments influence the behavior of metastatic cancer cells is crucial for developing effective treatments for Stage IV disease. This research specifically focuses on TNBC, an aggressive breast cancer subtype, and its behavior in lung and liver metastatic sites.

    What did we already know?

    • • Breast cancer frequently metastasizes to bone, liver, lungs, and brain.
    • • TNBC, a particularly aggressive subtype, primarily spreads to visceral organs, mainly the lungs and liver.
    • • The presence of bone metastases is associated with a less aggressive disease course compared to other sites.
    • • There are conflicting data on the aggressiveness of disease in the context of lung and liver metastases.

    Results:

    • • Patients with liver metastases, as their first metastatic event, showed significantly shorter overall survival and time to the next metastatic event compared to those with lung metastases. Focused on a subset of patients with "mono-metastasis", meaning that a single metastasis was present in either the lung (12/318 patients with metastases) or liver (10/318 patients with metastases) at the initial diagnosis of metastasis. Kaplan-Meier plots of the progression and survival differences between patients with lung versus liver mono-metastases support the idea that liver metastases are associated with more aggressive disease progression compared to lung metastases in breast cancer patients.
    • • Barcode labeling and metastasis clonality assessment in a mouse model of TNBC: mouse TNBC cells (MVT1) were tagged with unique genetic barcodes before being injected into the mammary fat pad of mice. The barcode composition of cells in the circulation (circulating tumor cells) and from each site (liver, lung) was analyzed. A greater number of unique barcodes were found in the lung metastases compared to the liver. Principal component analysis revealed that lung metastases were a distinct population compared to liver metastases and circulating tumor cells.
    • • After mammary fat pad injection & metastasis, mouse MVT1 or 4T1 cancer cells were isolated from either the lungs or liver, injected into the tail veins of healthy mice. Liver-resident TNBC cells have an enhanced ability to establish secondary metastases at lung and liver when compared to lung-resident cells.
    • • MVT1 mouse TNBC cells expressed GFP and a soluble mCherry protein that is taken up by neighboring niche cells. Identified and analyzed both tumor and niche cell populations.
    • • Single-cell RNA sequencing of cancer cells: MTV1 metastatic TNBC cells have distinct molecular profiles depending on where they reside: energy generation pathways in liver-resident cells (might contribute to their increased metastatic activity) and stress and detoxification pathways in lung-resident cells (suggests a less favorable environment).
    • • Single-cell RNA sequencing of niche cells: Endothelial cells in the liver metastatic niche, unlike the primary tumor niche and lung niche. The unique abundance of endothelial cells in the liver niche suggests their potential involvement in shaping the metastatic behavior of liver-resident TNBC cells. Specific ligand-receptor pairs suggest a distinct communication network between cancer cells and their niche in the liver and lungs, whereby the liver niche is enriched in endothelial cells secreting Bmp2 and Bmp6 cytokines, while the lung niche is enriched in macrophages secreting Grn and Ssp1.
    • • Specific cytokine-receptor interactions between cancer cells and their niche were identified, with BMP2/6 secreted by liver endothelial cells and Granulin secreted by lung macrophages.
    • • In vivo CRISPR-Cas9 screen used to investigate the roles of the identified cytokine-receptor pairs in metastasis formation. MTV1 TNBC cells were engineered with loss-of-function of the receptors for BMP2/6 (BMPR2, BMPR1A, and ACVR1) implicated in liver metastasis and Granulin (TNFRSF1A/B) implicated in lung metastasis. The engineered cells were injected into the tail vein and allowed to metastasize, isolated from lung and liver, and the abundance of different gene knockouts was analyzed. Additionally, they were reinjected into the tail vein to detect secondary metastasis.
    • • Using this approach the organ of primary metastasis influenced the secondary metastasis (lung went to lung, liver to liver) - in contrast to when the experiment was done with WT cells injected into the mammary fat pad and isolated from liver and lung (when liver-derived TNBC showed higher secondary metastasis by tail vein to lung and liver, Fig 1I). This disconnect is a bit confusing.
    • • There was no difference between the knockouts. This lack of organ specific knockouts between lung- and liver-resident cells could be due to the pleiotropic role of many of these receptors and the existence of interactions with additional cytokines. Notably, the expression of these receptors in cancer cells at the RNA level was similar in liver and lung metastases.
    • • Kaplan-Meier curve compares the survival of mice re-injected with either lung- or liver-resident cancer cells. Liver metastases were associated with decreased survival when compared to lung metastasis. Consistent with liver metastases associated with more aggressive disease progression compared to lung metastases.
    • • To investigate the effect of the niche on TNBC cells, TNBC cells were treated with either BMP2 or Granulin in vitro before being injected into the tail vein. This simulates the effects of the liver and lung microenvironment, respectively. BMP2 pre-treatment of TNBC cells enhanced metastasis formation in the lung, whereas Granulin treatment suppressed it. Supports the model that the liver niche boosts metastasis through BMP2, while the lung niche inhibits it through Granulin. This is a bit confusing with the apparent tropism of liver to liver in figure 5 but more consistent with figure 1.

    What's new? Sheds light on how the liver and lung microenvironments (endothelial vs macrophage, respectively) distinctly influence the behavior of TNBC cells and why liver metastasis is associated with poor survival. Offers potential therapeutic targets for liver vs lung metastases. The discovery of the contrasting roles of BMP2 and Granulin, and their cellular sources within the respective metastatic niches, is interesting.

    Potential impact

    • • Treatment strategies for TNBC patients with liver or lung metastases could potentially be tailored based on the identified niche-specific vulnerabilities.
    • • Targeting BMP2 signaling in liver metastases could potentially reduce secondary spread.
    • • Stimulating Granulin activity could offer a new approach to inhibiting TNBC metastasis.

    Limitations:

    • • The study primarily focused on TNBC, and further research is needed to determine if these findings apply to other cancer types metastasizing to the liver and lungs.
    • • I would have liked to see preclinical models with xenografts. Therapeutic potential not shown.
    • • The CRISPR-Cas9 screen did not identify organ-specific knockouts, likely due to the pleiotropic roles of the targeted receptors and the complex interplay of various cytokines within the niche.

    Future work:

    • • Investigate the applicability of these findings to other cancer types with similar metastatic patterns.
    • • Further explore the complex interplay of cytokines and signaling pathways within the liver and lung metastatic niches.
    • • Develop therapeutic strategies to target BMP2 signaling in liver metastases and stimulate Granulin activity in lung metastases.
    • • Validate these findings in a larger cohort of patients to determine their clinical relevance.
  2. Version published to 10.1101/2024.10.04.616491v1 on bioRxiv