Primary tumor cells migrating to secondary sites induce fibroblasts to generate an extracellular matrix component needed to support metastatic colonization, scientists report. In vitro and in vivo studies by an Ecole Polytechnique Fédérale de Lausanne-led team found that production of periostin (POSTN) is required to maintain cancer stem cells (CSCs) that infiltrate distal sites. They say that blocking its function prevented metastases in a well-established mouse model.
Describing their results in Nature, Joerg Huelsken, M.D., and colleagues suggest that targeting the formation of CSC-supporting niches may represent a novel strategy against metastastic disease. Their paper is titled “Interactions between cancer stem cells and their niche govern metastatic colonization.”
The development of a metastasis is a multistage process that involves a number of rate-limiting steps, and while the dissemination of tumor cells appears to be an early and frequent event, the successful initiation of metastatic growth, or metastatic colonization, is only achieved by a minority of cancer cells that reach distal sites.
The fact that cancers metastasize more regularly to some distal sites than others suggests there may be factors lacking in some distant tissues that hamper metastatic colonization, the authors write. To investigate which limiting factors might play a role in metastatic success, they turned to the MMTV-PyMT mouse breast cancer model, which spontaneously metastasizes to the lungs.
The team postulated that CSCs might be important for the development of metastatic disease, so they first measured the relative CSC population from primary MMTV-PyMT tumors and their pulmonary metastases using the previously established markers CD90 and CD24. They found that the CSC subset accounted for about 3% of all tumor cells in both primary tumors and metastases. But when either specifically CSC or non-CSC cells isolated from GFP+ tumors were injected into the tail veins of mice, only the CSC population generated lung metastases. Moreover, CD90+CD24+ cells isolated from these pulmonary metastases were the only tumor cell population that efficiently initiated secondary metastases.
CSCs also showed marked differences to non-CSCs in terms of proliferation as the metastasis progressed, the investigators continue. When mice were injected with unfractionated tumor cells the total number of tumor cells in the lung declined rapidly within days of seeding, but the relative numbers of CSCs increased within the first and second weeks, to over 20%.
The CSCs effectively expanded during the early stage of metastatic colonization, whereas non-CSCs showed reduced proliferation over the first few days after seeding and failed to generate CSCs. “Consequently, only CSCs are able to form metastatic colonies and non-CSCs remain as single cells,” the team states. This supports the notion that it is the selective expansion of CSCs that is responsible for initiating metastases.
Even so, not all CSCs will form metastatic nodules, suggesting that additional factors restrict metastatic colonization, the researchers continue. Prior studies have demonstrated that stem cells rely on signals contained in the surrounding stroma, and the Swiss team focused their attention on a fibroblast-produced extracellular matrix component, POSTN, which they identified as a stromal factor of normal stem cell niches and the metastatic niche.
The Postn gene is normally downregulated in adults, except in niches that directly contact with stem cells in various tissues. However, the authors found that in tumors, POSTN was produced by αSMA+VIM+ fibroblasts and was also induced in lung stroma by infiltrating cancer cells. POSTN wasn’t, however, found in the lung tissue of tumor-bearing mice without metastases. Interestingly, human breast cancer patients also demonstrated stromal POSTN expression in some three-quarters of lymph node metastases.
Significantly, in their experimental breast cancer mice, knocking out POSTN (MMTV-PyMTPostn-/- animals) resulted in a marked decrease in the number and size of pulmonary metastases to less than 10% of those in control animals, even though primary tumor size and morphology were unaffected.
Further analyses focused on identifying tumor-derived factors that might induce stromal POSTN production suggested that primary lung fibroblasts upregulate POSTN in response to TGF-β3 and TGF-β2. The team subsequently found through co-culture experiments that tumor cells can trigger POSTN production and that both CSCs and non-CSCs produce high levels of TGFβ3. inhibiting TGF-β3 activity in tumor cells, meanwhile, blocked POSTN expression and prevented metastasis formation in the experimental mice.
The requirement for POSTN by CSCs for metastases was demonstrated by further in vitro tests. The investigators found that CSC cultures couldn’t be established from POST-deficient tumors, whereas adding the protein to mutant tumor cells rescued the formation of sphere formation. Treating wild-type cancer cells with a POSTN-targeting antibody also prevented cancer stem cell maintenance, and CSCs failed to proliferate when co-cultured with POSTN-mutant pulmonary fibroblasts.
In lung metastases CD90+ CSCs tended to localize next to stromal niches, whereas lung metastases in POSTN-deficient animals had reduced CSC populations. “Thus, POSTN is an essential niche factor that supports CSC growth during metastatic colonization,” they remark.
The team then carried out tandem affinity purification (TAP)-tag enrichment and tandem mass spectrometry to identify the pathways with which POSTN interacts. The results indicated that POSTN binds to Wnt ligands Wnt1 and Wnt3A and boosts Wnt signaling activity. Wnt is known to control stem cell maintenance in a range of tissues, including tumors, and the team showed that Wnt signaling was concentrated in the CSC population in vivo. In vitro, the addition of WNT3A was sufficient to rescue stem cell expansion in the absence of POSTN.
“We suggest that the education of stromal cells by infiltrating tumor cells is an important step in metastatic colonization and that preventing de novo niche formation may be a novel strategy for the treatment of metastatic disease,” the authors conclude. “Targeting this metastatic niche promises to be less sensitive to rapid genetic changes in cancer cells and may not only prevent metastatic colonization but may also interfere with the survival of disseminated, dormant cancer cells.”