Transglutaminase 2 (TG2) is a key epidermal squamous cell carcinoma cancer cell survival protein. However, how TG2 maintains the aggressive cancer phenotype is not well understood. The present studies show that TG2, which is highly expressed in epidermal cancer stem–like cells (ECS cells), maintains hepatocyte growth factor (HGF) signaling to drive an aggressive ECS cell cancer phenotype. Inhibiting TG2 reduces MET tyrosine kinase receptor expression and activity and attenuates the cancer cell phenotype. Moreover, inhibition of TG2 or HGF/MET function reduces downstream MEK1/2 and ERK1/2 activity, and this is associated with reduced cancer cell spheroid formation, invasion, and migration, and reduced stem and EMT marker expression. Treatment of TG2 knockdown cells with HGF partially restores the aggressive cancer phenotype, confirming that MET signaling is downstream of TG2. MET knockout reduces ERK1/2 signaling, doubles the time to initial tumor appearance, and reduces overall tumor growth. These findings suggest that TG2 maintains HGF/MET and MAPK (MEK1/2 and ERK1/2) signaling to drive the aggressive ECS cell cancer phenotype and tumor formation, and that TG2-dependent MET signaling may be a useful anti-cancer target.

Implications:

TG2 is an important epidermal squamous cell carcinoma stem cell survival protein. We show that TG2 activates an HGF/MET, MEK1/2 ERK1/2 signaling cascade that maintains the aggressive cancer phenotype.

Epidermal squamous cell carcinoma (SCC) is a common form of surface epithelial cancer that is caused by exposure to ultraviolet (UV) light (1). The worldwide incidence of SCC is increasing because of population aging and increased exposure to UV light (2). SCC can be treated by surgical excision, but aggressive chemotherapy-resistant cancer recurs in 10% to 30% of patients (1). Abundant evidence suggests that cancer stem cells play an important role in tumor formation, recurrence, and metastasis (3–8). We recently characterized epidermal cancer stem–like (ECS) cells that express stem cell markers characteristic of normal epidermal and embryonic stem cells (9), as well as pluripotent stem cell markers that are expressed in other epithelial cancer types (6, 10, 11). ECS cells are extremely aggressive cells that generate large, aggressive and highly vascularized tumors in immune-compromised mice following subcutaneous injection of as few as 100 cells (9).

As part of a search for cancer cell survival proteins, we identified transglutaminase type 2 (TG2) as highly elevated in ECS cells as compared with non-stem cancer cells. TG2 has been reported to maintain the cancer stem cell phenotype in a host of tissues (12–14). TG2 has enzymatic and scaffold functions and is involved in inflammation, tissue repair, and cancer progression (15, 16). It displays transamidase, GTP binding, protein disulfide isomerase, and serine/threonine kinase activities (17) and serves as a scaffold protein (18). The TG2 transamidase and GTP binding activities are the most important and well characterized.

In the epidermis, TG2 is expressed in the basal epidermal layer where it is presumed to have a survival role (19, 20). We have shown that TG2 is markedly elevated in ECS cells and maintains the ECS cell phenotype (21–23). Indeed, knockdown and inhibitor studies show that TG2 is required for ECS cell survival, spheroid formation, migration, and invasion (22). Moreover, inhibition of TG2 activates ECS cell apoptosis. Mutagenesis studies indicate that the TG2 GTP binding/G-protein activity is required for ECS cell survival, but that TG2 transamidase activity is not (22). A major goal of our laboratory is to characterize the downstream signaling pathways that mediate TG2 action. We have shown that TG2 activates VEGF (24), neuropilin-1 (25), GIPC1/SYX/RhoA/p38 (25), α6/β4-integrin (21, 22), and YAP1/TAZ (21) signaling. Hepatocyte growth factor (HGF) ligand interacts with MET, its tyrosine kinase transmembrane receptor, to activate downstream events. We now show that TG2 maintains HGF/MET signaling to drive an aggressive skin cancer phenotype.

Antibodies and reagents

DMEM (11960-077), 0.25% trypsin-EDTA (25200-056), and L-glutamine (25030-164) were purchased from Gibco. Heat-inactivated fetal calf serum (FCS, F4135) and anti-β-actin (A5441) were purchased from Sigma. Anti-TG2 (MAB3839) and the SGX-523 MET inhibitor (448106) were purchased from EMD Millipore Corp. Cell lysis buffer (9803) and MET (8198S), MET-P (3126S), ERK1/2 (9102), ERK1/2-P (4376), MEK1/2 (4694S), MEK1/2-P (9145S), HGF (5244S), CD31 (3528), vimentin (5741S), cleaved PARP (9541), Snail (3895S), CD31 (3528S), and E-cadherin (3195S) antibodies were purchased from Cell Signaling Technology. The CD44v6 (ab78960), Slug (ab27568), Twist (ab49254), and Sox2 (ab15830-100) antibodies were purchased from Abcam. Anti-N-cadherin (610920), anti-Oct4 (611203), and anti-fibronectin (610077) were purchased from BD Biosciences. Anti-PARP (556494) was purchased from BD Pharmingen. Peroxidase-conjugated anti-mouse IgG (NXA931) and anti-rabbit IgG (NA934V) were obtained from GE Healthcare. DAPI (D9542) was from Sigma Chemicals. Human MET siRNA (SMARTpool, M-003156-02-0005) and human HGF-siRNA (SMARTpool, M-006650-01-0005) were purchased from Dharmacon. Matrigel (354234) and BD BioCoat Millicell inserts (d = 1 cm; 8 μm pore size, #353097) were purchased from BD Biosciences. MEK1/2 inhibitor U0126 (V112A) was purchased from Promega. ERK1/2 inhibitor LY3214996 (S853401) was purchased from Selleckchem. Recombinant human HGF protein (294-HG) was obtained from R&D Systems and the TG2 inhibitor NC9 was produced by Dr. Jeffrey Keillor as previously described (21, 22, 26). Cell mycoplasma testing is performed when new cell stocks are thawed for use, after which the cells are cultured for six months to one year. The Student t test was used for comparison of treatment groups.

Immunoblot

Cells and tissue were sonicated and prepared in Laemmli buffer (0.063 mol/L Tris-HCl, pH 7.5, 10% glycerol, 5% SDS, 5% β-mercaptoethanol) and equivalent amounts of protein were loaded on 8%–12% SDS-PAGE gels, electrophoresed under denaturing and reducing conditions, and transferred to nitrocellulose membrane. The membranes were blocked in 5% non-fat milk for 1 hour before primary antibody incubation. Membranes were incubated in 1:500–1:1,000 diluted primary antibodies overnight, washed, incubated with 1:5,000 diluted secondary antibodies for 2 hours, and antibody binding was visualized using chemiluminescence detection.

qRT-PCR

Total RNA was isolated using the Illustra RNAspin mini kit (GE Healthcare), and 1 μg of total RNA was reverse transcribed to cDNA by Superscript III reverse transcriptase (Invitrogen). The Light Cycler 480 SYBR Green I Master mix (Roche Diagnostics) was used to measure the change of mRNA level. The following gene primers were used: MET (forward: 5′-TGC ACA GTT GGT CCT GCC ATG A, reverse: 5′-CAG CCA TAG GAC CGT ATT TCG G), cyclophilin A (forward: 5′-CAT CTG CAC TGC CAA GAC TGA, reverse: 5′-TTC ATG CCT TCT TTC ACT TTGC). Fold change of MET mRNA level was normalized to the cyclophilin A mRNA level.

Cell culture

SCC-13 (27) and HaCaT (28) cells were maintained in growth medium consisting of DMEM, 4.5 mg/mL D-glucose, 2 μmol/L L-glutamine, 100 mmol/L sodium pyruvate, and 10% heat-inactivated FCS (9). SCC-13 cells are aggressive tumor-forming epidermis-derived cells, and HaCaT cells are epidermis-derived immortalized non-tumorigenic cells. The cells were grown in 100-mm culture dishes (Corning) and passaged every four to five days. Cell authenticity was determined by STR profiling at six-month intervals. All experiments used ECS cells, derived by culture of SCC-13 and HaCaT cells as spheroids on ultra-low attachment plates in stem cell medium containing DMEM/F12 (1:1; DMT-10-090-CV; Mediatech Inc), 2% B27 serum-free supplement (17504-044, Invitrogen), 20 ng/mL EGF (E4269, Sigma), 0.4% bovine serum albumin (B4287, Sigma), and 4 μg/mL insulin (#19278, Sigma (9). These cells were then plated in stem cell medium for proliferation, spheroid formation, invasion, and migration experiments.

SCC13-MET-knockout cell clones were produced using CRISPR/Cas9 technology followed by limiting dilution to select stable clones. MET-specific CRISPR guide RNA, forward (5′-caccGTCATACTGCTGACATACAGT) and reverse (5′-aaacACTGTATGTCAGCAGTATGAC), were identified at http://crispr.technology and cloned in the U6-driven pSpCas9(BB)-2A-Puro (PX459) V2.0 vector from Addgene. The vector was then electroporated into cells using the AMAXA electroporator. At 48 hours after electroporation, cells were treated with 2 μg/mL puromycin, and single-cell clones were selected by dilution cloning. For siRNA electroporation, 1 million cells were electroporated with 100 μL nucleofection reagent VPD-1002 mixed with 3 μg of target-specific siRNA using the T-018 setting on the AMAXA electroporator. Electroporated cells were maintained in growth media for 48 hours and collected for second electroporation. Double electroporated cells were cultured in growth medium for 24 hours before use in biological assays (29, 30). Cell spheroid formation, Matrigel invasion, and migration assays were performed as previously described (9, 23). We do not show data comparing treatment with control-siRNA versus mock (no siRNA) treatment, as these values are always identical. Representative migration images are presented and the quantitative results are presented as average percent final open area = (final open area/initial open area) × 100 (n = 3). The Student t test was used for comparison of treatment groups.

Tumor xenograft assay

Single-cell suspensions of spheroid-derived ECS cells were resuspended in PBS containing 30% Matrigel, and 100,000 spheroid-derived ECS cells were subcutaneously injected into the two front flanks in each of five eight-week-old female immune-compromised NSG (NOD/SCID/IL2Rg−/−) mice. Tumor size was monitored weekly by measuring tumor diameter and calculating tumor volume using the formula, volume = 4/3π × (diameter/2)3 (21). Animal studies were reviewed and approved by the Institutional Animal Care and Use Committee and followed standard international practices for treatment of animals.

TG2, HGF/MET signaling, and the ECS cell phenotype

We have shown that TG2 GTP binding activity is required for optimal cancer cell spheroid formation, invasion, migration, and tumor growth (21, 22, 24, 25). Figure 1 confirms these findings by showing that TG2 knockdown (Fig. 1A) reduces SCC-13 cell invasion and migration (Fig. 1B and C). The migration experiment shows that control siRNA-treated cells have completely closed the wound, but 28.5% of the wound remains unclosed in TG2-siRNA–treated cells.

Figure 1.

TG2 maintains HGF/MET signaling to drive the cancer phenotype. AC, TG2 knockdown reduces SCC-13 cell invasion and migration. TG2 siRNA significantly reduces wound closure (n = 3, P ≤ 0.006) as compared with control siRNA-treated cells. D, TG2 knockdown reduces SCC-13 cell MET mRNA level. E–G, SCC-13 and HaCaT cell MET and ERK1/2 activity and cell proliferation are reduced by TG2 inhibitor (NC9) treatment. H and I, TG2 knockdown reduces SCC-13 and HaCaT cell invasion, and the cells display minimal MET and ERK1/2 activity, and HGF treatment enhances ERK1/2 signaling. The single asterisks indicate a significant reduction (n = 3, P ≤ 0.001).

Figure 1.

TG2 maintains HGF/MET signaling to drive the cancer phenotype. AC, TG2 knockdown reduces SCC-13 cell invasion and migration. TG2 siRNA significantly reduces wound closure (n = 3, P ≤ 0.006) as compared with control siRNA-treated cells. D, TG2 knockdown reduces SCC-13 cell MET mRNA level. E–G, SCC-13 and HaCaT cell MET and ERK1/2 activity and cell proliferation are reduced by TG2 inhibitor (NC9) treatment. H and I, TG2 knockdown reduces SCC-13 and HaCaT cell invasion, and the cells display minimal MET and ERK1/2 activity, and HGF treatment enhances ERK1/2 signaling. The single asterisks indicate a significant reduction (n = 3, P ≤ 0.001).

Close modal

To gain new insights regarding additional target proteins that may mediate the response to TG2, we performed RNA-seq transcriptomic gene-expression analysis of wild-type versus TG2 knockout cells. This analysis revealed that HGF signaling was among the most reduced pathways following TG2 knockout. Although this transcriptomics analysis was performed using mesothelioma cancer cells, we surmised that the results may provide insights into other cancer types. Indeed, qRT-PCR analysis (Fig. 1D) shows that MET mRNA level is markedly reduced following TG2 knockdown. To further assess if HGF/MET signaling is maintained by TG2, we treated cells with NC9, a TG2-specific inhibitor (12, 21–23, 26), and monitored the impact on MET activity and biological responses. Figure 1E and F shows that NC9 treatment reduces SCC-13 and HaCaT cell proliferation, and Fig. 1G shows that this is associated with reduced MET and ERK1/2 activities. We next determined if reduced invasion observed in TG2 knockdown ECS cells could be reversed by HGF treatment. Figure 1H and I shows that TG2 knockdown reduces SCC-13 and HaCaT cell invasion, and MET, MEK1/2, and ERK1/2 activity, and that these changes are reversed by HGF treatment. These findings suggest that HGF/MET is a TG2 downstream target.

Impact of inhibiting HGF/MET in SCC-13 and HaCaT cells

Having implicated HGF/MET signaling as a TG2 target, we next sought to directly demonstrate a role for HGF/MET in maintaining ECS cell function. Figure 2 shows that HGF and MET knockdown (Fig. 2A) reduces spheroid formation, invasion, and migration (Fig. 2BD). Moreover, treatment with SGX-523, a MET inhibitor, reduces MET activity (Fig. 2E) and suppresses invasion and migration (Fig. 2F and G). We also show that HGF treatment stimulates MET activity (Fig. 2H) leading to increased invasion and migration (Fig. 2I and J) and that HGF stimulation of MET activity can be attenuated by MET siRNA (Fig. 2H) leading to reduced invasion and migration (Fig. 2I and J). We previously reported that loss of TG2 activity is associated with reduced stem cell and epithelial–mesenchymal transition (EMT) marker expression (23). Consistent with a role of HGF/MET as a downstream TG2 target, knockdown of MET, or treatment with MET inhibitor (SGX-523), results in reduced expression of stem and EMT markers (Fig. 2K and L).

Figure 2.

HGF/MET signaling maintains an aggressive SCC-13 cell cancer phenotype. A–D, HGF and MET knockdown reduce SCC-13 cell spheroid formation, invasion, and migration. HGF and MET siRNA treatment significantly reduces SCC-13 cell wound closure (n = 3, P ≤ 0.006). EG, MET inhibitor (SGX-523) treatment reduces SCC-13 cell MET activity and suppresses cell invasion and migration. SGX-523 reduces wound closure (n = 3, P ≤ 0.001). HJ, HGF stimulates SCC-13 cell MET activity and invasion and migration, and these responses are reversed by treatment with MET siRNA. HGF increases wound closure (n = 3, P ≤ 0.003), and this effect is reversed by MET siRNA treatment (n = 3, P ≤ 0.002). K and L, MET knockdown and MET inhibitor (SGX-523) treatment reduce SCC-13 cell stem and EMT marker levels. For plotted data, a single asterisk indicates a significant decrease (n = 3, P ≤ 0.001) and double asterisks indicate a significant increase (n = 3, P ≤ 0.001).

Figure 2.

HGF/MET signaling maintains an aggressive SCC-13 cell cancer phenotype. A–D, HGF and MET knockdown reduce SCC-13 cell spheroid formation, invasion, and migration. HGF and MET siRNA treatment significantly reduces SCC-13 cell wound closure (n = 3, P ≤ 0.006). EG, MET inhibitor (SGX-523) treatment reduces SCC-13 cell MET activity and suppresses cell invasion and migration. SGX-523 reduces wound closure (n = 3, P ≤ 0.001). HJ, HGF stimulates SCC-13 cell MET activity and invasion and migration, and these responses are reversed by treatment with MET siRNA. HGF increases wound closure (n = 3, P ≤ 0.003), and this effect is reversed by MET siRNA treatment (n = 3, P ≤ 0.002). K and L, MET knockdown and MET inhibitor (SGX-523) treatment reduce SCC-13 cell stem and EMT marker levels. For plotted data, a single asterisk indicates a significant decrease (n = 3, P ≤ 0.001) and double asterisks indicate a significant increase (n = 3, P ≤ 0.001).

Close modal

To confirm these findings in another cell line, we monitored the role of HGF/MET in HaCaT cells which are immortalized nontumorigenic cells derived from epidermis (28). MET knockdown reduces HaCaT cell MET level and activity (Fig. 3A), and this is associated with reduced spheroid formation (Fig. 3B), invasion (Fig. 3C), and migration (Fig. 3D). Cell invasion and migration (Fig. 3C and D) are also reduced by HGF knockdown. Figure 3E shows that SGX-523 treatment reduces HaCaT cell MET activity, and this is associated with reduced invasion (Fig. 3F) and migration (Fig. 3G). Moreover, HGF treatment stimulates MET activity (Fig. 3H), leading to increased invasion (Fig. 3I) and migration (Fig. 3J), and these responses are antagonized by MET knockdown (Fig. 3HJ). In addition, MET knockdown and SGX-253 treatment reduces the level of many, but not all, stem and EMT markers (Fig. 3K and L).

Figure 3.

HGF/MET signaling maintains the HaCaT cell cancer phenotype. AD, MET siRNA knockdown reduces HaCaT cell spheroid formation, invasion, and migration. HGF and MET siRNA treatment significantly reduces HaCaT cell wound closure (n = 3, P ≤ 0.005). EG, MET inhibitor (SGX-523) treatment inhibits HaCaT cell MET activity and reduces invasion and migration. SGX-523 treatment significantly reduces wound closure (n = 3, P ≤ 0.001). HJ, HGF stimulates MET activity and cell invasion and migration, and these responses are attenuated by treatment with MET siRNA. HGF treatment increases wound closure (n = 3, P ≤ 0.001), and this increase is suppressed by MET siRNA treatment (n = 3, P ≤ 0.001). K and L, MET knockdown or treatment with MET inhibitor reduces MET activity and suppresses stem cell and EMT marker expression. For plotted data, a single asterisk indicates a significant decrease (n = 3, P ≤ 0.001) and double asterisks indicate a significant increase (n = 3, P ≤ 0.001).

Figure 3.

HGF/MET signaling maintains the HaCaT cell cancer phenotype. AD, MET siRNA knockdown reduces HaCaT cell spheroid formation, invasion, and migration. HGF and MET siRNA treatment significantly reduces HaCaT cell wound closure (n = 3, P ≤ 0.005). EG, MET inhibitor (SGX-523) treatment inhibits HaCaT cell MET activity and reduces invasion and migration. SGX-523 treatment significantly reduces wound closure (n = 3, P ≤ 0.001). HJ, HGF stimulates MET activity and cell invasion and migration, and these responses are attenuated by treatment with MET siRNA. HGF treatment increases wound closure (n = 3, P ≤ 0.001), and this increase is suppressed by MET siRNA treatment (n = 3, P ≤ 0.001). K and L, MET knockdown or treatment with MET inhibitor reduces MET activity and suppresses stem cell and EMT marker expression. For plotted data, a single asterisk indicates a significant decrease (n = 3, P ≤ 0.001) and double asterisks indicate a significant increase (n = 3, P ≤ 0.001).

Close modal

Characterization of MET knockout cells

To further characterize the role of HGF/MET, we used CRISPR/Cas 9 to create clonal SCC-13 MET knockout cell lines. As shown in Fig. 4A, these cells lack MET signaling, and this is associated with reduced expression of the Sox2 and CD44v6 stem cell markers, increased E-cadherin expression, and reduced Twist expression. These findings are consistent with reduced stemness and EMT. These cells also display reduced spheroid formation and invasion (Fig. 4B and C). We next measured the response of these cells to HGF treatment. As expected, HGF treatment enhances MET activity in wild-type SCC-13 cells, but does not stimulate MET activity in the MET knockout cells (Fig. 4D). Moreover, HGF treatment does not stimulate MET knockout cell spheroid formation (Fig. 4E), invasion (Fig. 4F), or migration (Fig. 4G). These findings are consistent with HGF acting via MET to enhance the cancer phenotype.

Figure 4.

MET knockout attenuates the cancer phenotype. AC, MET knockout reduces MET activity, suppresses stem marker (Sox2 and CD44v6) levels, and suppresses EMT (reduces Twist and increases E-cadherin), and this is associated with reduced spheroid formation and invasion. DF, SCC-13 cell spheroid formation and invasion are stimulated by HGF treatment, and MET knockout attenuates HGF stimulation of these responses. G, HGF treatment enhances wound closure in SCC-13 cells (n = 3, P ≤ 0.05), but HGF stimulation of wound closure is attenuated in the MET knockout cell lines (n = 3, P ≤ 0.009). For plotted data, a single asterisk indicates a significant decrease and double asterisks indicate a significant increase (n = 3, P ≤ 0.005).

Figure 4.

MET knockout attenuates the cancer phenotype. AC, MET knockout reduces MET activity, suppresses stem marker (Sox2 and CD44v6) levels, and suppresses EMT (reduces Twist and increases E-cadherin), and this is associated with reduced spheroid formation and invasion. DF, SCC-13 cell spheroid formation and invasion are stimulated by HGF treatment, and MET knockout attenuates HGF stimulation of these responses. G, HGF treatment enhances wound closure in SCC-13 cells (n = 3, P ≤ 0.05), but HGF stimulation of wound closure is attenuated in the MET knockout cell lines (n = 3, P ≤ 0.009). For plotted data, a single asterisk indicates a significant decrease and double asterisks indicate a significant increase (n = 3, P ≤ 0.005).

Close modal

ERK1/2 is a downstream target

We next worked to identify HGF/MET downstream signaling events. HGF/MET activates a wide range of signaling responses including PKC, Src/FAK, STAT3, Ras/Raf, PI3K/Akt, and MAPK in a cell type–selective manner (31, 32). To identify MET receptor signaling targets, we screened a host of candidate kinases (not shown) and found that only ERK1/2 activity was consistently reduced in the MET knockout clones (Fig. 5A). MEK1/2 controls ERK1/2 activity (33), and so we explored the impact of MEK1/2 and ERK1/2 inhibition on cell function. Treatment with MEK1/2 inhibitor (U0126) reduced ERK1/2 activity as evidenced by reduced phosphorylation of the ERK1/2 activation loop (ERK1/2-P; Fig. 5B). Treatment with the ERK1/2 inhibitor (LY3214996) inhibits ATP binding to the ERK1/2 catalytic site to inhibit ERK1/2 modification of downstream targets (34) and LY3214996 treatment can result in a compensatory increase in ERK1/2 activation loop phosphorylation even though ERK1/2 activity is inhibited. We observe a compensatory increase in ERK1/2 activation loop phosphorylation (ERK1/2-P) in LY3214996-treated cells (Fig. 5B). We next monitored the biological impact of these inhibitors. As expected, SCC13-MEK-KOc4-3 cells display reduced ERK1/2 activity (Fig. 5A), which is associated with attenuated invasion and migration (Fig. 5C and D). Moreover, the knockout cells are minimally affected by inhibitor treatment (Fig. 5C and D). These findings suggest that MEK1/2 and ERK1/2 are important MET targets.

Figure 5.

MEK1/2 and ERK1/2 are MET downstream targets in the TG2 regulatory pathway. A, ERK1/2 activity is reduced in MET knockout ECS cells. BD, Treatment of wild-type cells with MEK1/2 (U0126) and ERK1/2 inhibitor (LY3214996) decreases ERK1/2 activity and reduces cell invasion and migration, but these inhibitors minimally impact MET knockout cell invasion or migration. U0126 and LY3214996 treatment reduces SCC-13 cell wound closure compared with untreated SCC-13 cells (n = 3, P ≤ 0.002). MET knockout cell lines display reduced wound closure compared with wild-type cells (n = 3, P ≤ 0.002), and are less responsive to the MEK1/2 or ERK1/2 inhibitor. EG, Treatment with MEK1/2 or ERK1/2 inhibitor inhibits HGF stimulation of HaCaT cell invasion and migration. HGF treatment increases wound closure (n = 3, P ≤ 0.0004), and HGF-stimulated wound closure is suppressed by inhibitor treatment (n = 3, P ≤ 0.001). For plotted data, a single asterisk indicates a significant decrease and double asterisks indicate a significant increase (n = 3, P ≤ 0.005).

Figure 5.

MEK1/2 and ERK1/2 are MET downstream targets in the TG2 regulatory pathway. A, ERK1/2 activity is reduced in MET knockout ECS cells. BD, Treatment of wild-type cells with MEK1/2 (U0126) and ERK1/2 inhibitor (LY3214996) decreases ERK1/2 activity and reduces cell invasion and migration, but these inhibitors minimally impact MET knockout cell invasion or migration. U0126 and LY3214996 treatment reduces SCC-13 cell wound closure compared with untreated SCC-13 cells (n = 3, P ≤ 0.002). MET knockout cell lines display reduced wound closure compared with wild-type cells (n = 3, P ≤ 0.002), and are less responsive to the MEK1/2 or ERK1/2 inhibitor. EG, Treatment with MEK1/2 or ERK1/2 inhibitor inhibits HGF stimulation of HaCaT cell invasion and migration. HGF treatment increases wound closure (n = 3, P ≤ 0.0004), and HGF-stimulated wound closure is suppressed by inhibitor treatment (n = 3, P ≤ 0.001). For plotted data, a single asterisk indicates a significant decrease and double asterisks indicate a significant increase (n = 3, P ≤ 0.005).

Close modal

We next determined if MEK1/2 or ERK1/2 inhibitor treatment suppresses HaCaT cell function. As shown in Fig. 5E, treatment with HGF activates MET, and this is associated with reduced MET protein level, suggesting that HGF treatment may enhance HaCaT cell MET turnover. In addition, HGF treatment is associated with increased ERK1/2 activity (ERK1/2-P). Treatment with U0126 reduced ERK1/2-P, but as expected LY3214996 produced a compensatory increase in ERK1/2-P. However, treatment with U0126 and LY3214996 suppressed the HGF-stimulated HaCaT cell invasion (Fig. 5F) and migration (Fig. 5G). These findings suggest that a HGF, MET, MEK1/2, ERK1/2 survival pathway is important in HaCaT cells.

Role of MET in tumor formation

Our previous studies show that inhibition of TG2 reduces tumor formation (21). To determine if MET inactivation phenocopies this response, we injected spheroid-derived SCC-13 and SCC13-MET-KOc4-3 cells into NSG mice and monitored tumor formation. Tumor growth was monitored weekly from one through eight weeks after cancer cell injection. SCC-13 tumors are first detected at three weeks after injection whereas SCC13-MET-KOc4-3 cell tumors are not detected until six weeks (Fig. 6A). Thus, MET knockout doubles the time to initial tumor appearance. Because of the large tumor size, it was necessary to sacrifice the SCC-13 tumor animals at four weeks. The tumor images in Fig. 6A show that four-week SCC-13 tumors are larger than the eight-week SCC13-MET-KOc4-3 cell tumors.

Figure 6.

MET knockout reduces tumor formation. A and B, Spheroid-derived ECS cells were injected into each front flank of NSG mice (n = 10 tumors/treatment group) and tumor formation was monitored weekly for eight weeks. SCC-13 cell tumors were harvested at four weeks due to the large size, and SCC13-MET-KOc4–3 cell tumors were harvested at eight weeks. Tumor extracts were prepared to monitor MET and ERK1/2 signaling, and EMT marker and PARP levels. For plotted data, double asterisks indicate a significant increase compared with controls (n = 10, P < 0.001). C and D, MET knockout tumors display reduced vascularization as evidenced by a 3.5-fold reduction in CD31 levels (n = 3, P ≤ 0.005) in TG2 knockout tumors as determined by immunostaining and confirmed by immunoblot. E, Schematic depiction of the TG2–HGF/MET–MEK1/2–ERK1/2 pathway that is proposed to maintain the aggressive cancer phenotype.

Figure 6.

MET knockout reduces tumor formation. A and B, Spheroid-derived ECS cells were injected into each front flank of NSG mice (n = 10 tumors/treatment group) and tumor formation was monitored weekly for eight weeks. SCC-13 cell tumors were harvested at four weeks due to the large size, and SCC13-MET-KOc4–3 cell tumors were harvested at eight weeks. Tumor extracts were prepared to monitor MET and ERK1/2 signaling, and EMT marker and PARP levels. For plotted data, double asterisks indicate a significant increase compared with controls (n = 10, P < 0.001). C and D, MET knockout tumors display reduced vascularization as evidenced by a 3.5-fold reduction in CD31 levels (n = 3, P ≤ 0.005) in TG2 knockout tumors as determined by immunostaining and confirmed by immunoblot. E, Schematic depiction of the TG2–HGF/MET–MEK1/2–ERK1/2 pathway that is proposed to maintain the aggressive cancer phenotype.

Close modal

We next assessed the impact of MET knockout on expression of EMT markers by comparing MET wild-type and knockout tumors. Figure 6B shows that MET knockout tumors display reduced MET and ERK1/2 activity, and reduced expression of several key EMT markers including fibronectin, vimentin, Slug, and Twist. Moreover, cleaved PARP levels are increased, suggesting increased apoptosis. Thus, MET knockout produces similar changes in tumors as observed in cultured cells. In addition, we observed a 3.5-fold decrease in blood vessel density in TG2 knockout tumors as measured by CD31 immunostaining (Fig. 6C), and this was also reflected in a reduction in CD31 level as detected by immunoblot (Fig. 6D).

TG2 maintains HGF/MET signaling and the cancer phenotype

TG2 is an important pro-cancer regulator that acts to enhance the malignant phenotype in a wide range of cancers (12, 13, 35). TG2 operates by activating a host of key pro-cancer signaling pathways in a cell type–specific manner (12, 13). We have focused on the role of TG2 in epidermal SCC and shown that TG2 maintains VEGF (24), NRP1 (25), GIPC1/SYX/RhoA/p38 (25), α6/β4-integrin (21, 22), and YAP1/TAZ (21) signaling. To identify additional TG2 targets, we performed RNA-seq transcriptomic analysis of gene expression in mesothelioma cells, which revealed a marked reduction in HGF mRNA following TG2 knockout. Based on this finding, we reasoned that TG2 may also target HGF/MET signaling in epidermal SCC. This was confirmed by showing that TG2 knockdown or treatment with TG2 inhibitor reduces cancer cell HGF/MET activity. Moreover, the fact that MET activity and the aggressive cancer phenotype can be partially restored by HGF treatment of TG2-null cells suggests that HGF/MET signaling is a downstream target of TG2 (31, 32).

Role of HGF/MET in regulating the cancer cell phenotype

Because TG2 stimulates multiple pro-cancer signaling pathways in skin cancer (13), it is important to prove that HGF/MET signaling is a biologically important TG2 target. To do this, we show that HGF or MET knockdown or treatment with MET inhibitor reduces MET activity, and that this is associated with reduced cancer cell spheroid formation, invasion, and migration. In addition, we observe a reduction in expression of selected cancer stem and EMT markers. These responses phenocopy changes observed following TG2 knockdown (9). Moreover, these responses require HGF interaction with MET, as inhibiting MET action attenuates the response to HGF. HGF/MET is known to activate a host of downstream signaling cascades, including PKC, Src/FAK, STAT3, Ras/Raf, PI3K/Akt, and MAPK, which are activated in a cell type–specific manner (31, 32). We show that inhibiting TG2 or HGF/MET reduces MEK1/2 and ERK1/2 signaling. Moreover, treatment with MEK1/2 or ERK1/2 inhibitor also reduced invasion and migration, which directly confirms a role for MEK1/2 and ERK1/2.

These findings strongly suggest that TG2 maintains HGF/MET signaling to drive the aggressive cancer phenotype. However, to provide parallel in vivo evidence supporting this hypothesis, we measured wild-type and MET knockout ECS cell tumor growth in NSG immune-compromised mice. These experiments show that control cell tumors first appear at three weeks and grow rapidly. In contrast, MET knockout cell tumors first appear at six weeks and grow slowly. Thus, MET loss doubles the time to tumor first appearance and reduces overall tumor growth. Consistent with the cell culture findings, the MET knockout tumors display reduced MET, MEK1/2, and ERK1/2 activity and reduced EMT marker levels. Moreover, the tumors appear to be undergoing apoptosis as evidenced by increased levels of active PARP. Because HGF/MET signaling has been reported to stimulate VEGF production (36), we also monitored expression of the CD31 angiogenesis marker and found that CD31 level is reduced in the MET knockout tumors, suggesting reduced angiogenesis. This is consistent with our previous finding showing that VEGF signaling is an important TG2-actived target in SCC (24).

TG2 and HGF/MET signaling

Previous studies implicate HGF/MET signaling in various squamous cancers. HGF activates Stat3 signaling in head and neck cancer, and this is inhibited by dominant-negative Stat3 (37). High-level MET expression is observed in oral SCC (38), which is associated with enhanced invasion (39). In metastatic Ly-1 and Ly-2 cells, which were derived from Pam 212 mouse keratinocytes, HGF stimulates cell scattering, VEGF production, and tumor formation (36). MET also participates in cross-talk with other cascades. For example, MET signaling leads to EGFR activation in epidermal keratinocytes to initiate squamous carcinogenesis (40), ERK2 regulates cross-talk between MET and EGFR in squamous cell cancer of the tongue (41), and inactivation of Smad2 increases MET level in mouse epidermis to drive cancer (42).

Summary

The present studies point to a new role for TG2 in maintaining HGF/MET signaling to drive an aggressive skin cancer phenotype. We propose that TG2 maintains HGF and MET level and activity, which stimulates MEK1/2–ERK1/2 signaling to drive cancer stemness, EMT, invasion, migration, and tumor formation (Fig. 6E). An interesting general feature of the TG2 maintenance of the cancer phenotype is that it maintains/activates membrane-localized proteins (31, 32) that are known to act in a cooperative manner to drive the cancer phenotype. The present studies add HGF/MET signaling to the range of pro-cancer pathways already reported to be activated by TG2 (12, 13) and suggest that TG2 may be a useful therapy target for the treatment of epidermal SCC.

No disclosures were reported.

X. Chen: Investigation. G. Adhikary: Data curation, investigation, and methodology. S. Shrestha: Validation, investigation. W. Xu: Investigation. J.W. Keillor: Resources. W. Naselsky: Investigation. R.L. Eckert: Conceptualization, resources, supervision, funding acquisition, investigation, writing–original draft.

This work was supported by NIH R01 grants CA184027 and CA211909 awarded to R.L. Eckert. W. Naselsky was supported by the Cancer Biology T32 Training Grant (T32 CA154274) and an International Lung Cancer Foundation Young Investigator Award. The facilities of the Greenebaum Comprehensive Cancer Center were used in these studies (P30 CA134274).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Lansbury
L
,
Bath-Hextall
F
,
Perkins
W
,
Stanton
W
,
Leonardi-Bee
J
. 
Interventions for non-metastatic squamous cell carcinoma of the skin: systematic review and pooled analysis of observational studies
.
BMJ
2013
;
347
:
f6153
.
2.
Lomas
A
,
Leonardi-Bee
J
,
Bath-Hextall
F
. 
A systematic review of worldwide incidence of nonmelanoma skin cancer
.
Br J Dermatol
2012
;
166
:
1069
80
.
3.
Thenappan
A
,
Li
Y
,
Shetty
K
,
Johnson
L
,
Reddy
EP
,
Mishra
L
. 
New therapeutics targeting colon cancer stem cells
.
Curr Colorectal Cancer Rep
2009
;
5
:
209
.
4.
Carpentino
JE
,
Hynes
MJ
,
Appelman
HD
,
Zheng
T
,
Steindler
DA
,
Scott
EW
, et al
Aldehyde dehydrogenase-expressing colon stem cells contribute to tumorigenesis in the transition from colitis to cancer
.
Cancer Res
2009
;
69
:
8208
15
.
5.
Al-Hajj
M
,
Wicha
MS
,
Benito-Hernandez
A
,
Morrison
SJ
,
Clarke
MF
. 
Prospective identification of tumorigenic breast cancer cells
.
Proc Natl Acad Sci U S A
2003
;
100
:
3983
8
.
6.
Li
C
,
Yan
Y
,
Ji
W
,
Bao
L
,
Qian
H
,
Chen
L
, et al
OCT4 positively regulates Survivin expression to promote cancer cell proliferation and leads to poor prognosis in esophageal squamous cell carcinoma
.
PLoS One
2012
;
7
:
e49693
.
7.
Lee
CJ
,
Dosch
J
,
Simeone
DM
. 
Pancreatic cancer stem cells
.
J Clin Oncol
2008
;
26
:
2806
12
.
8.
Saini
V
,
Shoemaker
RH
. 
Potential for therapeutic targeting of tumor stem cells
.
Cancer Sci
2010
;
101
:
16
21
.
9.
Adhikary
G
,
Grun
D
,
Kerr
C
,
Balasubramanian
S
,
Rorke
EA
,
Vemuri
M
, et al
Identification of a population of epidermal squamous cell carcinoma cells with enhanced potential for tumor formation
.
PLoS One
2013
;
8
:
e84324
.
10.
Tsai
LL
,
Hu
FW
,
Lee
SS
,
Yu
CH
,
Yu
CC
,
Chang
YC
. 
Oct4 mediates tumor initiating properties in oral squamous cell carcinomas through the regulation of epithelial-mesenchymal transition
.
PLoS One
2014
;
9
:
e87207
.
11.
Huang
CF
,
Xu
XR
,
Wu
TF
,
Sun
ZJ
,
Zhang
WF
. 
Correlation of ALDH1, CD44, OCT4 and SOX2 in tongue squamous cell carcinoma and their association with disease progression and prognosis
.
J Oral Pathol Med
2014
;
43
:
492
8
.
12.
Eckert
RL
,
Fisher
ML
,
Grun
D
,
Adhikary
G
,
Xu
W
,
Kerr
C
. 
Transglutaminase is a tumor cell and cancer stem cell survival factor
.
Mol Carcinog
2015
;
54
:
947
58
.
13.
Eckert
RL
. 
Transglutaminase 2 takes center stage as a cancer cell survival factor and therapy target
.
Mol Carcinog
2019
;
58
:
837
53
.
14.
Eckert
RL
,
Kaartinen
MT
,
Nurminskaya
M
,
Belkin
AM
,
Colak
G
,
Johnson
GV
, et al
Transglutaminase regulation of cell function
.
Physiol Rev
2014
;
94
:
383
417
.
15.
Kim
SY
. 
Transglutaminase 2 in inflammation
.
Front Biosci
2006
;
11
:
3026
35
.
16.
Siegel
M
,
Khosla
C
. 
Transglutaminase 2 inhibitors and their therapeutic role in disease states
.
Pharmacol Ther
2007
;
115
:
232
45
.
17.
Gundemir
S
,
Colak
G
,
Tucholski
J
,
Johnson
JGV
. 
Transglutaminase 2: a molecular Swiss army knife
.
Biochim Biophys Acta
2012
;
1823
:
406
19
.
18.
Park
D
,
Choi
SS
,
Ha
KS
. 
Transglutaminase 2: a multi-functional protein in multiple subcellular compartments
.
Amino Acids
2010
;
39
:
619
31
.
19.
Pitolli
C
,
Pietroni
V
,
Marekov
L
,
Terrinoni
A
,
Yamanishi
K
,
Mazzanti
C
, et al
Characterization of TG2 and TG1-TG2 double knock-out mouse epidermis
.
Amino Acids
2017
;
49
:
635
42
.
20.
Fok
JY
,
Ekmekcioglu
S
,
Mehta
K
. 
Implications of tissue transglutaminase expression in malignant melanoma
.
Mol Cancer Ther
2006
;
5
:
1493
503
.
21.
Fisher
ML
,
Kerr
C
,
Adhikary
G
,
Grun
D
,
Xu
W
,
Keillor
JW
, et al
Transglutaminase interaction with α6/β4-integrin to stimulates YAP1-dependent ΔNp63α stabilization and leads to enhanced cancer stem cell survival and tumor formation
.
Cancer Res
2016
;
76
:
7265
76
.
22.
Fisher
ML
,
Keillor
JW
,
Xu
W
,
Eckert
RL
,
Kerr
C
. 
Transglutaminase is required for epidermal squamous cell carcinoma stem cell survival
.
Mol Cancer Res
2015
;
13
:
1083
94
.
23.
Fisher
ML
,
Adhikary
G
,
Xu
W
,
Kerr
C
,
Keillor
JW
,
Eckert
RL
. 
Type II transglutaminase stimulates epidermal cancer stem cell epithelial-mesenchymal transition
.
Oncotarget
2015
;
6
:
20525
39
.
24.
Grun
D
,
Adhikary
G
,
Eckert
RL
. 
VEGF-A acts via neuropilin-1 to enhance epidermal cancer stem cell survival and formation of aggressive and highly vascularized tumors
.
Oncogene
2016
;
35
:
4379
87
.
25.
Grun
D
,
Adhikary
G
,
Eckert
RL
. 
NRP-1 interacts with GIPC1 and SYX to activate p38 MAPK signaling and cancer stem cell survival
.
Mol Carcinog
2019
;
58
:
488
99
.
26.
Keillor
JW
,
Chica
RA
,
Chabot
N
,
Vinci
V
,
Pardin
C
,
Fortin
E
, et al
The bioorganic chemistry of transglutaminase – from mechanism to inhibition and engineering
.
Can J Chem
2008
;
86
:
271
6
.
27.
Rheinwald
JG
,
Beckett
MA
. 
Tumorigenic keratinocyte lines requiring anchorage and fibroblast support cultures from human squamous cell carcinomas
.
Cancer Res
1981
;
41
:
1657
63
.
28.
Boukamp
P
,
Petrussevska
RT
,
Breitkreutz
D
,
Hornung
J
,
Markham
A
,
Fusenig
NE
. 
Normal keratinization in a spontaneously immortalized aneuploid human keratinocyte cell line
.
J Cell Biol
1988
;
106
:
761
71
.
29.
Adhikary
G
,
Chew
YC
,
Reece
EA
,
Eckert
RL
. 
PKC-delta and -eta, MEKK-1, MEK-6, MEK-3, and p38-delta are essential mediators of the response of normal human epidermal keratinocytes to differentiating agents
.
J Invest Dermatol
2010
;
130
:
2017
30
.
30.
Chew
YC
,
Adhikary
G
,
Wilson
GM
,
Reece
EA
,
Eckert
RL
. 
PKCdelta suppresses keratinocyte proliferation by increasing p21CIP1 level by a KLF4-dependent mechanism
.
J Biol Chem
2011
;
286
:
28771
82
.
31.
Organ
SL
,
Tsao
MS
. 
An overview of the c-MET signaling pathway
.
Ther Adv Med Oncol
2011
;
3
:
S7
S19
.
32.
Maroun
CR
,
Rowlands
T
. 
The Met receptor tyrosine kinase: a key player in oncogenesis and drug resistance
.
Pharmacol Ther
2014
;
142
:
316
38
.
33.
Cobb
MH
. 
MAP kinase pathways
.
Prog Biophys Mol Biol
1999
;
71
:
479
500
.
34.
Bhagwat
SV
,
McMillen
WT
,
Cai
S
,
Zhao
B
,
Whitesell
M
,
Shen
W
, et al
ERK inhibitor LY3214996 targets ERK pathway-driven cancers: a therapeutic approach toward precision medicine
.
Mol Cancer Ther
2020
;
19
:
325
36
.
35.
Mehta
K
,
Kumar
A
,
Kim
HI
. 
Transglutaminase 2: a multi-tasking protein in the complex circuitry of inflammation and cancer
.
Biochem Pharmacol
2010
;
80
:
1921
9
.
36.
Dong
G
,
Lee
TL
,
Yeh
NT
,
Geoghegan
J
,
Van
WC
,
Chen
Z
. 
Metastatic squamous cell carcinoma cells that overexpress c-Met exhibit enhanced angiogenesis factor expression, scattering and metastasis in response to hepatocyte growth factor
.
Oncogene
2004
;
23
:
6199
208
.
37.
Syed
ZA
,
Yin
W
,
Hughes
K
,
Gill
JN
,
Shi
R
,
Clifford
JL
. 
HGF/c-met/Stat3 signaling during skin tumor cell invasion: indications for a positive feedback loop
.
BMC Cancer
2011
;
11
:
180
.
38.
Morello
S
,
Olivero
M
,
Aimetti
M
,
Bernardi
M
,
Berrone
S
,
Di Renzo
MF
, et al
MET receptor is overexpressed but not mutated in oral squamous cell carcinomas
.
J Cell Physiol
2001
;
189
:
285
90
.
39.
Uchida
D
,
Kawamata
H
,
Omotehara
F
,
Nakashiro
K
,
Kimura-Yanagawa
T
,
Hino
S
, et al
Role of HGF/c-met system in invasion and metastasis of oral squamous cell carcinoma cells in vitro and its clinical significance
.
Int J Cancer
2001
;
93
:
489
96
.
40.
Cataisson
C
,
Michalowski
AM
,
Shibuya
K
,
Ryscavage
A
,
Klosterman
M
,
Wright
L
, et al
MET signaling in keratinocytes activates EGFR and initiates squamous carcinogenesis
.
Sci Signal
2016
;
9
:
ra62
.
41.
Gusenbauer
S
,
Zanucco
E
,
Knyazev
P
,
Ullrich
A
. 
Erk2 but not Erk1 regulates crosstalk between Met and EGFR in squamous cell carcinoma cell lines
.
Mol Cancer
2015
;
14
:
54
.
42.
Hoot
KE
,
Oka
M
,
Han
G
,
Bottinger
E
,
Zhang
Q
,
Wang
XJ
. 
HGF upregulation contributes to angiogenesis in mice with keratinocyte-specific Smad2 deletion
.
J Clin Invest
2010
;
120
:
3606
16
.