Strategies by which WWOX-deficient metastatic cancer cells utilize to survive via dodging, compromising, and causing damage to WWOX-positive normal microenvironment

Proapoptotic tumor suppressor WWOX is upregulated in the early stage of cancer initiation, which probably provides limitation to cancer growth and progression. Later, WWOX protein is reduced to enhance cancer cell growth, migration, invasiveness and metastasis. To understand how WWOX works in controlling cancer progression, here we demonstrate that apoptotic stress mediated by ectopic WWOX stimulated cancer cells to secrete basic fibroblast growth factor (bFGF) in order to support capillary microtubule formation. This event may occur in the cancer initiation stage. Later, when WWOX loss occurs in cancer cells, hyaluronidase production is then increased in the cancer cells to facilitate metastasis. We determined that inhibition of membrane hyaluronidase Tyr216-phosphorylated Hyal-2 by antibody suppresses cancer growth in vivo. WWOX-negative (WWOX-) cells dodged WWOX+cells in the microenvironment by migrating individually backward to avoid physical contacts and yet significantly upregulating the redox activity of WWOX+parental cells or other WWOX+cell types for causing apoptosis. Upon detecting the presence of WWOX+cells from a distance, WWOX- cells exhibit activation of MIF, Hyal-2, Eph, and Wnt pathways, which converges to MEK/ERK signaling and enables WWOX- cells to evade WWOX+cells. Inhibition of each pathway by antibody or specific chemicals enables WWOX- cells to merge with WWOX+cells. In addition, exogenous TGF-β assists WWOX- cells to migrate collectively forward and merge with WWOX+cells. Metastatic WWOX- cancer cells frequently secrete high levels of TGF-β, which conceivably assists them to merge with WWOX+cells in target organs and secure a new home base in the WWOX+microenvironment. Together, loss of WWOX allows cancer cells to develop strategies to dodge, compromise and even kill WWOX-positive cells in microenvironment.

Antibody against MIF (1 g/ml) was added to the coculture of MDA-MB-231 cells (left) and WWOX-positive L929 cells (right). MDA-MB-231 migrated forward collectively and merged with L929. Each picture frame was taken per 10 minutes. This data is linked to Figure 3g-i. Video S7. Inhibition of MEK/ERK by U0126 abolishes the retrograde migration of MDA-MB-231 upon facing L929. MEK inhibitor U0126 (10 M) was added to the coculture of MDA-MB-231 cells (left) and WWOX-positive L929 cells (right) for 4 hr, followed by washing and preparing for time-lapse microscopy. MDA-MB-231 migrated forward in a collective manner, and merged with L929. Each picture frame was taken per 10 minutes. No cell death occurred for both sides. This data is linked to Figure S7a  MDA-MB-231 cells (left) were treated with MEK inhibitor U0126 (10 M) for 4 hr, followed by removing the culture-insert, washing and preparing for coculture with L929 cells (right) for timelapse microscopy. MDA-MB-231 migrated forward in a collective manner, and merged with L929. Each picture frame was taken per 10 minutes. No cell death occurred for both sides. The data is linked to figure S7a and c.

Video S9. Knockout Wwox -/-MEF cells dramatically upregulate the redox activity in wild type MEF cells from a remote distance (merged channels).
Wwox wild type and knockout MEF cells were seeded in either side of a culture-insert. Following overnight culture, cells were stained with Redox Sensor Red CC-1 to measure their redox activity, and then subjected to time-lapse microscopy (merged channels from bright field and red fluorescence). When wild type cells detected the presence of migrating knockout cells from a distance of 500 m (and vice versa), wild type cells rapidly exhibited a significantly upregulated redox activity in less than 2 hr. Note that wild type cells underwent apoptosis in 12 hr under the influence of knockout cells. The original video data are linked to Figure 6a to c.

Video S10. Knockout Wwox -/-MEF cells dramatically upregulate the redox activity in wild type MEF cells from a remote distance (red fluorescence only).
The experiment is the same as in Video S9. The red fluorescence is shown only.

Video S11. Wild type versus wild type MEF cells in time-lapse microscopy (merged channels).
The experiment was carried out similarly to that in Videos S9 and S10. When wild type faced wild type cells in either side, little or no induced redox activity and no apoptosis were observed (merged channels from bright field and red fluorescence). The original video data are linked to Figure 6b and c.

Video S12. Wild type versus wild type MEF cells in time-lapse microscopy (red fluorescence only).
The experiment is the same as in Video S11. The red fluorescence is shown only.
Videos S13. MDA-MB-435s versus wild type MEF cells in time-lapse microscopy. When WWOX-deficient breast MDA-MB-435s migrated versus wild type MEF cells, apoptosis occurred mainly in the wild type cells.

Videos S14. MDA-MB-231 cells induce a greater extent of L929 apoptosis under serum-free conditions.
Under serum-free conditions, MDA-MB-231 induced a greater numbers of L929 cells undergoing apoptosis. This video data is linked to Figure S10b.

Video S15. Restoration of WWOX in MDA-MB-231 allows them to fend off WWOXnegative parental cells.
MDA-MB-231 cells were treated with methylation inhibitor 5-aza-2' deoxycytidine (5-aza, 5μM) for 5 days to restore WWOX expression (see Fig. 6f). By time-lapse microscopy, parental MDA-MB-231 underwent retrograde migration (right) upon facing 5-aza-treated cells (left) (see blue triangle). This video is linked to Fig. 6e to h. Video S16. Ectopic expression of the N-terminus of WWOX allows MDA-MB-231 to merge with L929. MDA-MB-231 cells were transiently transfected with an expression construct for EGFP-WW1/2 domains (the N-terminal head and the two WW domains). These cells moved forward and then merged with L929. All cells were stained with Cell Tracker Red. This video data is linked to fig. S10d.