Skip to main content
. 2021 Feb 2;10:e64370. doi: 10.7554/eLife.64370

Figure 2. SATB2 leads to invadopodia formation and increased migration potential in vitro and in vivo.

(A) Phalloidin staining for F-Actin in primary zebrafish melanoma cell culture, reveals the presence of F-actin-positive foci in MCR:SATB2 (45–3) cells, which are not present in MCR:EGFP (zmel1) cells. Scale bar is 20 μm. Primary tumor immunohistochemistry shows MCR:SATB2 tumors abundantly express Cortactin. Scale bar 50 μm. (B) MCR:SATB2 (63–4) cells show increased Oregon green 488-conjungated gelatin degradation, compared to MCR (CK5) cells 24–25 hr post-seeding. Scale bar is 50 μm. (C) Percentage of cells with degraded gelatin after SATB2 induction in iSATB2 human melanoma cell lines A375, SKMEL2, and SKMEL28 transduced with pInducer20-SATB2. Cells were seeded on gelatin in media +/- doxycycline after 48 hr +/- doxycycline induction. (D) Upon SATB2 induction in human melanoma cell line SKMEL2, cells form invadopodia and show increased matrix degradation Scale bar is 20 μm. (E) Orthotropic allograft migration assay in transparent Casper zebrafish. A total of 300,000 primary pigmented primary melanoma cells were transplanted into the dorsum of irradiated Casper recipients, which are monitored for the formation of pigmented distant metastasis that have spread past the anatomical midline. Metastases are represented as black circles. (F) Pooled recipient data at the experimental end point at 3.5 weeks post-transplantation, 59.4 ± 2.3% (SEM; n = 76 total recipients, grafted from seven individual donor tumors) of MCR:SATB2 transplants formed distant metastasis, compared to 21.8±4.5% (SEM; n = 37 total recipients grafted from five individual donor tumors) of EGFP-control transplants (p<0.0001). MCR:EGFP transplants developed an average 1.1 ± 0.4 (SD) distant metastasis per fish, versus 4 ± 4.2 (SD) in MCR:SATB2, where a maximum of 20 metastases per fish was observed. (G) MCR:SATB2 recipients more frequently developed bilateral (28.9%), and internal metastases (22.4%) compared to MCR:EGFP donors (5.4%), showing spreading along the neural tube (14.5% versus 2.7%). (H) At 3.5 weeks-post transplantation, compared to MCR:EGFP donor transplants, MCR:SATB2 transplants develop distant and internal metastases. Histopathology of MCR:SATB2 Casper recipients showing a (I–I) hypodermal metastasis, and (I–II) internal metastasis with spreading along neural tube.

Figure 2.

Figure 2—figure supplement 1. SATB2 overexpression induces EMT, increased migration and invadopodia formation in zebrafish melanoma.

Figure 2—figure supplement 1.

(A) Scratch assays on cultured MCR tumors show increased migration in MCR:SATB2 tumors. (B) Confocal analysis of immunohistochemistry of cultured MCR:SATB2 tumor shows co-localization of Cortactin (green) and F-actin (phalloidin, red), suggesting the formation of invadopodia. DAPI nuclear stain is shown in blue. (C) MCR:SATB2 (45–3) cells show increased Oregon green 488-conjungated gelatin degradation 24–25 hr post-seeding. Scale bar is 50 μm. (D) Percentage of zebrafish melanoma cells (n) with degraded 488-conjungated gelatin 24 hr post-seeding, observed in MCR (CK5, n = 93), MCR:SATB2 (45–3, n = 126), and MCR:SATB2 (63–4, n = 311) cell lines. (E) Immunofluorescence showing colocalization of F-actin and Cortactin in SKMEL2 overexpressing SATB2 in the presence of doxycycline (48 hr induction).
Figure 2—figure supplement 2. MCR:SATB2 primary tumors and long-term cultures have increase invasion potential in vivo.

Figure 2—figure supplement 2.

(A) Western blot of SATB2 showing retained expression in long term in vitro cultures derived from MCR:SATB2 tumors. Actin loading control is shown. (B) Overall survival and (C) histology of irradiated Casper recipients transplanted with MCR:EGFP Zmel1 (n = 31) or MCR:SATB2 45–3 (n = 31) zebrafish long term in vitro cultures. (D) Weekly prospective imaging of primary MCR:SATB2 individual tumor donor allograft in irradiated Casper recipient. Arrows indicate distant metastasis from injection site. Related to Figure 2H.
Figure 2—figure supplement 3. Analysis of SATB2 expression, amplification, and correlation with survival in human melanoma.

Figure 2—figure supplement 3.

(A) Analysis of SATB2 expression across tumor types in TCGA PanCancer dataset using cBio portal (Gao et al., 2013). Melanoma (highlighted in red) SATB2 mRNA levels are comparable to other solid tumors where SATB2 has been established to play a role (i.e. RCC, Sarcoma, CRC). (B) Western blot of SATB2 and SATB1 with a panel of antibodies showing protein expression in a panel of human melanoma cell lines and zebrafish MCR:EGFP, MCR:SATB2, and MCR:SATB1 control tumors. Actin loading control is shown. (C) Frequency of SATB2 amplification in red across multiple available melanoma genomic studies (ICGC/TCGA Pan-Cancer Analysis of Whole Genomes Consortium, 2020; Hugo et al., 2015; Snyder et al., 2014; Van Allen et al., 2014). (D) Correlation between SATB2 expression and melanoma metastasis-related overall survival in two independent publicly available dataset (GSE8401 and GSE22153) of melanoma followed for metastatic progression risk. Continuous expression Z-score optimized plots (mRNA expression Z-score value separating high vs. low expression that results in maximal statistical separation of the two curves) from the default output of the TIDE Harvard Portal gene query http://tide.dfci.harvard.edu/ (Fu et al., 2020).