Solid tumor metastasis is the leading cause of cancer-related mortality. Cancer invasion through the confining basement membrane (BM) is the initial step in tumor dissemination and metastasis, and it represents a key diagnostic feature of cancer. Thus, identifying the mechanisms involved in the breaching of cancer cells through the BM is potentially important for developing novel therapeutic approaches. BM is a dense sheet of specialized extracellular matrix proteins that separates tissue compartments. It is also a nanoporous structure, and since the average width of a cell is ~10 µm, invasion requires extensive widening of the BM nanopores. Previous research provided evidence that this expansion of BM nanopores involves protease degradation. However, protease inhibitors have failed to prevent cell invasion and metastasis in clinical trials, suggesting that cells may also breach the BM barrier using physical and mechanical mechanisms. Currently, it is unknown what mechanical mechanisms human tumor cells use to breach the BM. This is in part due to the difficulty of visualizing interactions at the cell-BM interface during cell invasion. Here, we designed and published a 3-dimensional in vitro organoid model of cancer spheroids encapsulated by a basement membrane and embedded in 3D collagen gels to visualize the early events of cancer invasion by confocal microscopy and live-cell imaging.
We first found that human breast cancer cells generated large numbers of basement membrane perforations, or holes, of varying sizes that expanded over time during cell invasion. We used a wide variety of small molecule inhibitors to probe the mechanisms of basement membrane perforation and hole expansion. Protease inhibitor treatment (BB94), led to a 63% decrease in perforation size. After myosin II inhibition (blebbistatin), the basement membrane perforation area decreased by only 15%. These treatments produced correspondingly decreased cellular breaching events. Interestingly, inhibition of actin polymerization dramatically decreased basement membrane perforation by 80% and blocked invasion. Our findings suggest that human cancer cells can primarily use proteolysis and actin polymerization to perforate the BM and to expand perforations for basement membrane breaching with a relatively small contribution from myosin II contractility.
We also found by using live timelapse imaging that cancer cells can send out long, actin-based prehensile protrusions (~30-100 microns in length) through the BM that subsequently grip and pull on the surrounding collagenous matrix to help cells pull themselves through the BM for invasion. These long protrusions are supported by microtubules and pull on the surrounding collagen using actomyosin contractility. We quantified this pulling exerted on collagen by generating kymographs for control and treatment groups and measuring collagen displacement over time. We found that by specifically inhibiting actin polymerization, microtubule formation, or actomyosin contractility, tumor organoids are unable to form these protrusions and fail to pull on the surrounding collagen matrix to enable invasion. Furthermore, by inhibiting the cell surface receptor for collagen, integrin ⍺2ß1, organoids could not form protrusions nor pull on the surrounding matrix, indicating that protrusions use integrin ⍺2ß1 to attach to and pull on the collagen matrix during the initial stages of invasion through the BM. In conclusion, some cancer cells extend long actin-based protrusions to bind to collagen via integrin ⍺2ß1 and use pulling forces driven by actomyosin contractility exerted on the surrounding extracellular matrix to squeeze through perforations in the basement membrane for translocating their cell body across this major tissue barrier to cancer invasion.