Down-regulation of proliferation-inhibiting factor EGR1 in brain metastatic cancer cells on a soft matrix

Abstract

Metastasis of cancer cells to the brain leads to a poor prognosis in patients with cancer. The brain environment is characterized by cell types, extracellular matrices (ECMs), and mechanical properties that differ from those of the primary tumors. A previous study using human melanoma cells (WM266.4 cells) and its highly brain-metastatic subline cells (WM266.4-BrM3 cells) revealed that WM266.4-BrM3 cells showed enhanced proliferation in brain tissues after cardiac injection in mice compared with WM266.4 cells. However, the effects of mechanical properties such as ECM stiffness on growth and gene expression in WM266.4-BrM3 cells remain to be clarified. In this study, we cultured these cells on ECMs of different stiffnesses. On a soft ECM, WM266.4-BrM3 cells showed significantly higher proliferation and lower expression of early growth response 1 (EGR1) and TP53 than WM266.4 cells. In contrast, on a stiff ECM, the proliferation and EGR1 expression of WM266.4 and WM266.4-BrM3 cells were not significantly different. Additionally, EGR1 knockdown by siRNA transfection in WM266.4 cells results in promoted cell proliferation and downregulated TP53 on a soft ECM. These results suggest that brain metastatic WM266.4 cells decrease EGR1 expression, thereby promoting cell proliferation via TP53 downregulation on a soft ECM.

Key words: EGR1, ECM stiffness, metastasis, cancer, growth

Graphical Abstract

Introduction

Brain metastases (BM) are the most common intracranial tumor. The incidence of BM is higher than that of primary brain tumors such as glioblastoma (Maher et al., 2009). BM leads to a poor prognosis in patients with cancer by inducing various neurological complications. Owing to the inefficient understanding of the molecular and cellular characteristics of BM and the complicated interaction with metastasized tumor cells and the brain microenvironment, BM treatment is challenging (Lu et al., 2025). Approximately 20% of all cancer patients develop BM. Specifically, lung cancers (20–56% of patients), breast cancers (5–20%), and melanoma (7–16%) are frequently metastasized to the brain (Achrol et al., 2019). There are many differences between the brain and primary tumor microenvironments. The brain environment is characterized by specific cell types, extracellular matrices (ECMs), mechanical properties, and the blood–brain barrier (Quail and Joyce, 2017). The brain microenvironment contains unique tissue-resident cells, such as microglia, astrocytes, and neurons, in addition to common tissue-resident cells, including pericytes, endothelial cells, fibroblasts, and immune cells (Quail and Joyce, 2017). A previous study using a human melanoma cell line (WM266.4) and its highly brain-metastatic cell line (WM266.4-BrM3) (Hirata et al., 2020) revealed that WM266.4-BrM3 cells interact with astrocytes and show enhanced proliferation in brain tissues after cardiac injection in mice compared with WM266.4 cells (Ishibashi et al., 2024). The brain microenvironment is one of the softest tissues of the body (Piersma et al., 2020). Unique ECM components, including proteoglycans, glycoproteins, and glycosaminoglycans, such as heparan sulfate proteoglycans and hyaluronic acid (Quail and Joyce, 2017) may contribute to brain softness. However, the effects of softness in the brain microenvironment on growth and gene expression in metastatic brain cancer cells are not well understood.

Tissue stiffening is a key characteristic of solid tumors. Previous studies have reported that cancer tissues are stiffer than the corresponding normal tissues in various organs, including the breasts, lungs, and pancreas (Ishihara and Haga, 2022). Tissue stiffening in cancer is caused by the secretion of ECMs by cancer and stromal cells in the tumor tissue (Tao et al., 2020), remodeling of ECMs by contraction of surrounding cells (Calvo et al., 2013), and ECM crosslinking regulated by crosslinkers such as lysyl oxidases (Levental et al., 2009). Cancer and stromal cells in tumors respond to ECM stiffness and regulate cancer progression by altering their phenotypes with processes, including proliferation, invasion, metastasis, drug resistance, angiogenesis, immune responses, and stemness (Ishihara and Haga, 2022). Particularly, cancer cell proliferation, which is essential for tumor formation and growth, is enhanced by a stiff ECM in various cancers (Ishihara and Haga, 2022). Thus, molecules contributing to the responses of cancer cells to tissue stiffening and cell proliferation are potential therapeutic targets.

Early growth response 1 (EGR1) is expressed in various types of cells and involves physiological processes such as proliferation, differentiation, invasion, and apoptosis (Wang et al., 2021). Previous studies have reported the cell-type-dependent cancer-promoting and antitumor effects of EGR1. For instance, in prostate and gastric tumors, EGR1 expression positively correlates with tumor size, invasion, cancer stage, and prognosis (Gabriel et al., 2016; Ma et al., 2020; Myung et al., 2013). In contrast, antitumor effects have been reported in glioma and melanocytoma, as EGR1 induces apoptosis of tumor cells by upregulating the tumor suppressor phosphatase and tensin homolog (PTEN) (Virolle et al., 2001). Additionally, EGR1 induces the expression of the tumor suppressor factor TP53, which plays an important role in DNA replication, followed by the induction of apoptosis (Yu et al., 2006; Nair et al., 1997). A previous study has reported that cancer cells cultured on soft ECMs expressed higher levels of EGR1 and showed lower proliferative ability than cancer cells cultured on stiff ECMs (Ishihara et al., 2025). However, the role of EGR1 in the proliferation of metastatic brain cancer cells has not been investigated.

In this study, we found that brain metastatic (WM266.4-BrM3) cells cultured on soft ECMs showed higher proliferative ability and lower EGR1 and TP53 expression than parental WM266.4 cells. Furthermore, EGR1 knockdown enhanced proliferation and suppressed TP53 expression in WM266.4 cells on soft ECMs. These results suggest that the suppression of EGR1 is critical for the proliferation of brain metastatic cancer cells on soft ECMs.

Materials and Methods

Cell culture

The luc-mEGFP-WM266.4 human melanoma cells and WM266.4-BrM3 human melanoma brain metastatic cells were maintained in Dulbecco’s Modified Eagle Medium with high glucose (D5796, Sigma-Aldrich, St. Louis, MO, USA) with 10% fetal bovine serum (F2442, Sigma-Aldrich) and 1% antibiotic/antimycotic solution (A5955, Sigma-Aldrich). An incubator at 37°C, 100% humidity, and 5% CO₂ was used for cell culture. The mycoplasma contamination was tested using the Mycoalert Assay Control Set (LT07-518; Lonza Bioscience, Walkersville, MD, USA). The cells were not authenticated.

Substrates

To prepare collagen gel substrates, 500 μL of neutralized collagen solution (1.6 mg/mL, Cellmatrix Type I-P, Nitta Gelatin, Yao, Japan) was poured in 35 mm plastic dishes and incubated at 37°C for 30 min for gelation. For coating with collagen, 35 mm plastic dishes were filled with 500 μL of a collagen solution (0.3 mg/mL Cellmatrix Type I-C (Nitta Gelatin) in HCl solution (pH = 3)) and incubated at 37°C for 30 min. The collagen solution was then removed, and the dishes were washed with the culture medium. As previously reported (Onishi et al., 2023), the 0.4 kPa polyacrylamide (PA) gels for cell culture were prepared using the following reagents (0.05% N,N'-methylenebisacrylamide (BIS) and 5.0% acrylamide).

RNA extraction

On collagen gels or collagen-coated plastic dishes, 1.0 × 10⁵ WM266.4 or WM266.4-BrM3 cells were seeded. After 48 or 72 h, RNA was extracted with TriPure Isolation reagent (11667165001, Roche, Basel, Switzerland) and purified using the FastGene RNA Basic Kit (FG-80250, NIPPON Genetics). On 576 mm² PA gels in 35 mm plastic dishes, 4.0 × 10⁵ WM266.4 or WM266.4-BrM3 cells were seeded. After 48 h, RNA was extracted using the FastGene RNA Basic Kit (FG-80250, Nippon Genetics, Tokyo, Japan).

Quantitative polymerase chain reaction (qPCR)

cDNA was synthesized using the ReverTra Ace qPCR RT Master Mix (FSQ-201, Toyobo, Osaka, Japan). qPCR was performed using the KAPA Fast qPCR Kit (KK4602, Nippon Genetics) and StepOnePlus (Thermo Scientific, Waltham, MA, USA). Relative mRNA expression was determined by normalizing the expression of the target mRNA to that of the ribosomal protein S18 (S18). The following primers were used: (5' to 3'): S18 (Forward: AAGGGTGTGGGCCGAAGATATG, Reverse: GTTCCACCTCATCCTCAGTGAGTTC), EGR1 (Forward: GCCCACCATGGACAACTACC, Reverse: TCAGGAAAAGACTCTGCGGTC), TP53 (Forward: TGACACGCTTCCCTGGATTG, Reverse: AGGGGGACAGAACGTTGTTTTCAG), ATF3 (Forward: TGCCATCCAGAACAAGCACC, Reverse: ATCTTCTTCAGGGGCTACCTCG), BAX (Forward: TGGAGCTGCAGAGGATGATTG, Reverse: CCAGTTGAAGTTGCCGTCAG), TGFB1 (Forward: TTGAGCCGTGGAGGGGAAAT, Reverse: AGAAGCAGGAAAGGCCGGTT), and PTEN1 (Forward: GGACCAGAGACAAAAAGGGAGTAAC, Reverse: GATTGCAAGTTCCGCCACTG).

Western blotting

On 576 mm² PA gels in 35 mm plastic dishes, 4.0 × 10⁵ WM266.4 or WM266.4-BrM3 cells were seeded. After 48 h, proteins were extracted from the cells using a sodium dodecyl sulfate (SDS) sample buffer (0.125 M Tris-HCl (pH = 6.8), 2.3% SDS, 10% glycerol, 5% dithiothreitol, and 0.01% bromophenol blue), sonicated, and boiled at 95°C for 5 min. 8% PA gels was used for SDS-polyacrylamide gel electrophoresis (SDS-PAGE) (20 mA/gel). After SDS-PAGE, blotting was performed on polyvinylidene difluoride membranes (86 mA/gel) for 60 min. After the blotting, the membranes were incubated in 5% skim milk (α-tubulin) or 5% bovine serum albumin (EGR1) in Tris-buffered saline plus Tween 20 (TBS-T) for 30 min at room temperature. After washing the membranes three times with TBS-T for 5 min, the membranes were incubated with primary antibodies (1:30,000 anti-α-tubulin (T9026, Sigma-Aldrich) in TBS-T or 1:30,000 anti-EGR1 (4153S, Cell Signaling Technology, Danvers, MA, USA) in Can Get Signal Solution 1 (NKB-101, Toyobo)) at 4°C overnight. After washing with TBS-T three times, the membranes were incubated with secondary antibodies (1:10,000 anti-mouse IgG, HRP-linked antibody (7076S, Cell Signaling Technology) in TBS-T for α-tubulin or 1:10,000 anti-rabbit IgG, HRP-linked antibody (7074S, Cell Signaling Technology) in Can Get Signal Solution 2 (NKB-101, Toyobo) for EGR1) for 1 h at room temperature. After washing the membrane three times with TBS-T, signals were detected using Immobilon Western Chemiluminescent HRP substrate (WBKLS0500, Millipore, Burlington, MA, USA) and a ChemiDoc Touch Imaging System (Bio-Rad, Hercules, CA, USA). The relative intensity of EGR1 was determined by normalizing the intensity of EGR1 with the intensity of α-tubulin analyzed via Image Lab software (Bio-Rad).

siRNA transfection

On 35 mm plastic dishes, 1.0 × 10⁵ WM266.4 cells were seeded and transfected with 1.0 pmol siEGR1 or a negative control (NC) non-targeting RNA using Lipofectamine RNAiMAX Transfection Reagent (13778150, Invitrogen, Waltham, MA, USA). After 72 h, the cells transfected with siRNA were used in subsequent experiments. The siRNA was synthesized using an in vitro Transcription T7 kit (6140, Takara Bio Inc., Kusatsu, Japan) and Klenow Fragment (3'→5' exo-) (M0212L, New England Biolabs, Ipswich, MA, USA). The target sequences (5' to 3') were as follows: non-targeting (anti-sense: AAACTACATGTCACATCACGGCCCTATAG, sense: AACCGTGATGTGACATGTAGTCCCTATAG), EGR1 (anti-sense: AAATTATCTTCACATCAAGAGCCCTATAG, sense: TACTCTTGATGTGAAGATAATCCCTATAG).

Cell growth assay

On 500 μL collagen gels in 35 mm plastic dishes or collagen-coated 35 mm plastic dishes, 1.0 × 10⁵ WM266.4 or WM266.4-BrM3 cells were seeded and cultured for 48 h (WM266.4 vs WM266.4-BrM3 cells; WM266.4 with NC RNA vs WM266.4 with siEGR1 cells) or 72 h (WM266.4 with NC RNA vs WM266.4 with siEGR1 cells). For the cells grown on collagen-coated plastic dishes, the culture medium was collected in a centrifuge tube. After washing the cells twice with phosphate-buffered saline (PBS), the PBS was collected in the same tube. Then, the cells were detached by 0.25% trypsin-EDTA (25200-072, Gibco, Watham, MA, USA) treatment for 5 min at 37°C. The suspended cells were collected in the same tube. The mixture was centrifuged at 161 g for 2 min. After supernatant removal, the cell pellet was suspended in 250 μL of fresh culture media. For the cells grown on collagen gels, the culture medium was collected in a centrifuge tube. Then, the cells were treated with 0.1% collagenase (034-22363, Wako, Osaka, Japan) in PBS for 5 min at 37°C. The collagenase suspension was collected in the same tube and centrifuged at 161 g for 2 min. Simultaneously, the dish was treated with trypsin-EDTA and incubated for 5 min at 37°C. After removal of the supernatant from the tube, trypsin-EDTA solution was added to the same tube. The solution was then resuspended and incubated within 3 min. The cell suspension was centrifuged at 161 g rpm for 2 min. After supernatant removal, the cell pellet was suspended in 250 μL of fresh culture medium. Five microliters of cell suspension was mixed with the same volume of Trypan Blue Stain (0.4%) for use with Countes Automated Cell Counter (T10282, Invitrogen). The total number of cells and dead cells was counted using a hemocytometer. The relative cell number was determined as the ratio of the total cell number. The proportion of surviving cells was determined as the number of living cells (trypan blue-negative)/total cell number.

Atomic force microscopy

The measurement of surface stiffness of a collagen gel was performed using Nanowizard 4 (JPK Instruments, Berlin, Germany) on a TE300 microscope (Nikon Instech, Tokyo, Japan). A 10 μm diameter bead (F8834, Life Technologies, Carlsbad, CA, USA) tip-bound silicon nitride cantilever (MLCT-E, Veeco, Woodbury, NY, USA) with a spring constant of 0.055 N/m was used. The gel was indented with a calibrated forces of 0.2 nN in a scan area of 1 μm² (4 pixels × 4 lines). The Hertz model of impact was used to determine stiffness (Young’s modulus). A Poisson’s ratio of 0.5 was used to calculate the Young’s modulus.

Statistics

Comparing a value with no variance to another value with variance, we determined significance using a 95% confidence interval. To compare the two values showing variance, we determined whether the two datasets had the same variance using an F-test. For the results of the F-test with p<0.05 (the variance of datasets was significantly different), we performed a two-sided Welch’s t-test. For the results of the F-test with p>0.05 (the variance of datasets was not significantly different), we performed a two-sided Student’s t-test. For data that contained more than three samples, we used the Bonferroni correction. All analyses were performed using Excel software.

Results

Proliferation of WM266.4-BrM3 cells was faster than WM266.4 cells on soft ECMs

First, we studied parental WM266.4 and its highly brain-metastatic WM266.4-BrM3 melanoma cells cultured on collagen gels (soft ECMs, approximately 8 Pa measured by atomic force microscopy) or collagen-coated plastic dishes (stiff ECMs, >1 GPa (Ishihara et al., 2013)). The stiffness of the collagen gels was more appropriate for brain tissues (Young’s modulus ~0.2 kPa) (Hirata et al., 2020) than that of the collagen-coated plastic dishes. On stiff ECMs, both cell types exhibited similar morphology after culturing for 48 h (Fig. 1A). In contrast, WM266.4-BrM3 cells showed a more spherical shape than WM266.4 cells on soft ECMs (Fig. 1A). To investigate cell proliferation, we compared the number of WM266.4 and WM266.4-BrM3 cells on soft or stiff ECMs. On stiff ECMs, both cell types showed similar cell numbers, whereas the number of WM266.4-BrM3 cells was greater than that of WM266.4 cells on soft ECMs after culturing for 48 h (Fig. 1B). Additionally, there was no significant difference in the cell survival rate between WM266.4 and WM266.4-BrM3 cells on both stiff and soft ECMs after culturing for 48 h (Fig. 1C). In Addition, the number of cells did not significantly differ between WM266.4 and WM266.4-BrM3 cells on soft ECMs after culturing for 72 or 96 h, possibly because of the effects of cell confluency and heterogeneity (Supplementary Fig. S1). These results showed that on soft ECMs, WM266.4-BrM3 cells proliferated faster than WM266.4 cells.

Fig. 1

Proliferation of WM266.4-BrM3 cells was faster than WM266.4 cells on a soft matrix

(A) Phase-contrast images of WM266.4 and WM266.4-BrM3 cells cultured for 48 h on collagen gel (soft) or collagen-coated plastic (stiff) substrates. (B) Relative cell number analyzed from (A) samples. (C) Proportion of surviving cells analyzed from (A) samples. N = 3 experiments; scale bar = 100 μm; mean with S.D. and each data point; *, statistical significance determined with 95% confidence interval; n.s, no significance determined with 95% confidence interval; n.s, no significance determined with Welch’s t-test, p>0.05; For multiple comparisons, we analyzed significance using the Bonferroni correction.

EGR1 was downregulated in WM266.4-BrM3 cells compared with WM266.4 cells on soft ECMs

Next, we examined the expression of EGR1, which is highly expressed in pancreatic cancer cells on soft ECMs and suppresses proliferation (Ishihara et al., 2025). By analyzing the RNA-seq results (Hirata et al., 2020), WM266.4-BrM3 cells showed lower expression of EGR1 than WM266.4 cells on non-coated plastic dishes (Fig. 2A). Additionally, qPCR results showed that WM266.4-BrM3 cells expressed a significantly lower level of EGR1 compared with WM266.4 cells on soft ECMs, whereas the expression of EGR1 was not significantly different between WM266.4 and WM266.4-BrM3 cells on stiff ECMs (Fig. 2B). We also prepared soft (0.4 kPa) PA gels and cultured WM266.4 or WM266.4-BrM3 cells on them (Fig. 2C). The qPCR results showed that EGR1 expression in WM266.4-BrM3 cells was lower than that in WM266.4 cells on soft PA gels, in agreement with the results for the cells on collagen gels (Fig. 2B, D). We confirmed that the EGR1 protein expression in WM266.4-BrM3 cells was lower than that in WM266.4 cells on soft PA gels (Fig. 2E, F). Collectively, these results showed that the expression of EGR1 was lower in WM266.4-BrM3 cells than in WM266.4 cells on soft ECMs.

Fig. 2

EGR1 was downregulated in WM266.4-BrM3 cells compared with WM266.4 cells on a soft matrix

(A) EGR1 expression of WM266.4 and WM266.4-BrM3 cells. RNA sequencing results from Hirata et al., iScience. 2020. (B) qPCR of EGR1 in WM266.4 and WM266.4-BrM3 cells on collagen gel (soft) or collagen-coated plastic (stiff) substrates. (C) Phase-contrast images of WM266.4 and WM266.4-BrM3 cells on 0.4 kPa PA gels. (D) qPCR of EGR1 in WM266.4 and WM266.4-BrM3 cells on 0.4 kPa PA gels. (E) Western blot of EGR1 and α-tubulin in WM266.4 and WM266.4-BrM3 cells on 0.4 kPa PA gels. (F) Relative intensity of EGR1 to α-tubulin in (E). N = 3 experiments; scale bar = 100 μm ; mean with S.D and each data point; *, statistical significance determined with 95% confidence interval; n.s, no significance determined with 95% confidence interval; n.s, no significance determine with Welch’s t-test, p>0.05; ★, statistical significance determined with 95% confidence interval; For multiple comparisons, we analyzed significance using the Bonferroni correction; S18 was used as an internal control of qPCR.

EGR1 knockdown enhanced proliferation of WM266.4 cells on soft ECMs

Next, we investigated whether EGR1 suppresses the proliferation of melanoma cells. WM266.4 cells were transfected with siEGR1 and cultured for 48 or 72 h on collagen gels (Fig. 3A and Supplementary Fig. S2A). The number of EGR1-knockdowned WM266.4 cells increased by approximately 1.5 times compared with WM266.4 cells transfected with control RNA (Fig. 3B, C and Supplementary Fig. S2B, 2C). In contrast, cell survival between control and siEGR1-transfected WM266.4 cells was not remarkably different (Supplementary Fig. S2D), although the differences were significant in the cells cultured for 72 h (Fig. 3D). Thus, EGR1 was critical for the inhibition of proliferation in WM266.4 cells on soft ECMs.

Fig. 3

EGR1 knockdown enhanced the proliferation of WM266.4 cells cultured for 72 h on a soft matrix

(A) qPCR of EGR1 in WM266.4 cells transfected with negative control (NC) RNA or siEGR1 on collagen gel substrates. (B) Phase-contrast images of WM266.4 and WM266.4-BrM3 cells on collagen gel substrates. (C) Relative cell number analyzed from (B) samples. (D) Proportion of surviving cells analyzed from (B) samples. N = 3 experiments; scale bar = 100 μm; mean with S.D. and each data point; ★, statistical significance determined with 95% confidence interval; S18 was used as an internal control of qPCR.

TP53 was downregulated in WM266.4-BrM3 and EGR1-knockdowned WM266.4 cells on soft ECMs

Next, we investigated the expression of EGR1-target genes (TP53, ATF3, BAX, PTEN1, and TGFB1), that regulate cell proliferation (Wang et al., 2021). Compared with WM266.4 cells on collagen gels, WM266.4-BrM3 cells expressed significantly lower levels of TP53 and BAX and higher levels of PTEN1 (Fig. 4A). Additionally, EGR1-knockdowned WM266.4 cells on collagen gels expressed significantly lower levels of TP53 and ATF3 than control WM266.4 cells (Fig. 4B). Collectively, these results indicate that TP53 is suppressed by the downregulation of EGR1 in WM266.4-BrM3 cells on soft ECMs.

Fig. 4

TP53 expression was downregulated in WM266.4-BrM3 and EGR1-knockdowned WM266.4 cells compared with control WM266.4 cells on a soft matrix

(A) qPCR of TP53, ATF3, BAX, PTEN1, and TGFB1 in WM266.4 and WM266.4-BrM3 cells on collagen gel substrates. (B) qPCR of TP53, ATF3, BAX, PTEN1, and TGFB1 in WM266.4 cells transfected with negative control (NC) RNA or siEGR1 on collagen gel substrates. N = 3 experiments.; mean with S.D. and each data point; ★, statistical significance determined with 95% confidence interval; S18 was used as an internal control of qPCR.

Discussion

In this study, the human melanoma cell line WM266.4 and its brain metastatic cell line WM266.4-BrM3 were cultured on soft or stiff ECMs. A previous study has reported that WM266.4-BrM3 cells have a higher proliferative ability than WM266.4 cells via interaction with glial cells on soft ECMs (Ishibashi et al., 2024). However, the role of ECM stiffness in the proliferation of WM266.4 and WM266.4-BrM3 cells has not been investigated. Herein, we showed that there was no significant difference in the proliferation of WM266.4 and WM266.4-BrM3 cells on stiff ECMs, whereas the proliferation of WM266.4-BrM3 cells was higher than that of WM266.4 cells on soft ECMs. These results indicate that brain metastatic melanoma cells can proliferate on both stiff and soft ECMs, whereas parental melanoma cells on soft ECMs are less proliferative than those on stiff ECMs. Additionally, EGR1 mRNA and protein levels were significantly lower in WM266.4-BrM3 cells than in WM266.4 cells. Moreover, the proliferation of WM266.4 cells on soft ECMs was significantly promoted by siEGR1 transfection. These results suggest that EGR1 plays an antitumor role by suppressing cell proliferation in WM266.4 melanoma cells, similar to KP4 pancreatic cancer cells (Ishihara et al., 2025). We also showed that cultured on soft ECMs, WM266.4-BrM3 cells and EGR1-knockdowned WM266.4 cells suppressed TP53 expression compared with WM266.4 and control cells, respectively, suggesting that TP53 is suppressed by downregulating EGR1 in brain metastatic melanoma cells on soft ECMs.

The metastatic cell line WM266.4-BrM3 expressed lower EGR1 and higher PTEN levels than WM266.4 cells in this study. Additionally, cell survival did not significantly differ between WM266.4 and WM266.4-BrM3 cells. These results contrast those of previous studies reporting that EGR1 promotes apoptosis in tumor cells (Pan et al., 2010) through upregulating the tumor suppressor PTEN via binding to the PTEN promoter (Virolle et al., 2001). The researchers in these studies used embryonic kidney-derived 293T cells and the gastric cancer cell line HGC-27; thus, the discrepancy between our findings and those in previous studies may be attributable to differences in the tumor or cell types used. However, the detailed mechanisms through which PTEN expression is regulated by EGR1 remain unclear.

On soft ECMs, differences in proliferation and EGR1 expression between WM266.4-BrM3 and WM266.4 cells were observed, whereas on stiff ECMs, there were no significant differences. These results indicate that ECM stiffness, which mimics in vivo tissues or organs, could be critical for in vitro analysis of cellular phenomena or gene expression in specific tissues or organs, such as the brain. We previously showed that activating transcription factor 5 (ATF5) is activated by a stiff ECM that mimics the primary tumor microenvironment and downregulates EGR1 expression in pancreatic cancer cells. Integrin β1 is a possible ECM receptor that regulates activation of ATF5 in primary pancreatic tumors (Ishihara et al., 2025). This integrin binds to several types of primary tumor ECMs, such as collagen and fibronectin (Grzesiak et al., 2007). The brain environment is not rich in these ECMs, although it mainly contains proteoglycans, glycoproteins, and glycosaminoglycans (Quail and Joyce, 2017). The pathways by which brain metastatic melanoma cells suppress EGR1 expression on soft ECMs have not been identified; however, their mechanisms may be independent of the integrin β1-ATF5 pathway.

In this study, we found that EGR1 was downregulated in the brain metastatic melanoma cell line WM266.4-BrM3 and suppressed proliferation in the parental melanoma cell line WM266.4. These results suggest that induction of EGR1 expression is a potential therapeutic strategy for BM. Previous studies have shown that EGR1 plays a tumor-suppressive role in glioma (Virolle et al., 2001). Thus, EGR1 induction may be a potential therapeutic strategy for brain tumors, including primary gliomas and metastatic melanomas.

Author Declaration Statements

Funding

This work was supported by Japan Agency for Medical Research and Development (AMED) Grant Number JP16gm0810007 to HH, JP17gm0810011 to HH, JP22ym0126814 to HH; JSPS KAKENHI Grant Numbers JP17K07150 to HH, JP21K07142 to HH, JP18K15232 to SI, JP21K07141 to SI, JP23KK0143 to SI, JP24H01917 to SI, JP24K10302 to SI; the cancer research grant of SGH Foundation to SI; Grants-in-Aid for Regional R&D Proposal-Based Program from Northern Advancement Center for Science & Technology of Hokkaido Japan to SI; the research grant of Astellas Foundation for Research on Metabolic Disorders to SI; the research grant of The Uehara Memorial Foundation to SI; the research grant of The Yasuda Medical Foundation to SI; the research grant of The Akiyama Life Science Foundation to SI; Extramural Collaborative Research Grant of Cancer Research Institute, Kanazawa University to SI; Next Generation Leader Training Program of Hokkaido University to SI; Next Generation Life Science Collaborative Research Projects of Hokkaido University to SI; recurring donations from M.D. Mariko Takamura to HH; Co-Creation Core for Soft Materials Aspiring Research & Translation (C3-SMART) and Japan Society for the Promotion of Science (JSPS) Program for Forming Japan’s Peak Research Universities (J-PEAKS) “Soft Materials Platform Aspiring the Unique Properties from Natural Polymers” and “Analysis of cancer specific mechanoresponse for drug discovery.”

Conflict of Interest Statement

The authors declare no competing interests.

Data Availability Statement

RNA seq data in this study were published (https://doi.org/10.1016/j.isci.2020.101480).

Author Contribution Statement

Miki Omukai: Writing – original draft, Methodology, Investigation

Seiichiro Ishihara: Writing – review & editing & original draft, Project administration, Methodology, Investigation, Supervision, Funding acquisition

Eishu Hirata: Writing – review & editing, Resources

Hisashi Haga: Writing – review & editing, Project administration, Supervision, Funding acquisition

Ethics Approval and Consent to Participate

Not applicable.

Patient Consent for Publication

Not applicable.

Acknowledgments

The authors thank Hinami Nakamura for preparing a collagen gel for stiffness measurement and all the present and former members of the Haga laboratory at Hokkaido University for helpful support.

References

•  Achrol,  A.S.,  Rennert,  R.C.,  Anders,  C.,  Soffietti,  R.,  Ahluwalia,  M.S.,  Nayak,  L.,  Peters,  S.,  Arvold,  N.D.,  Harsh,  G.R.,  Steeg,  P.S., and  Chang,  S.D. 2019. Brain metastases. Nat. Rev. Dis. Primers, 5: 5.
•  Calvo,  F.,  Ege,  N.,  Grande-Garcia,  A.,  Hooper,  S.,  Jenkins,  R.P.,  Chaudhry,  S.I.,  Harrington,  K.,  Williamson,  P.,  Moeendarbary,  E.,  Charras,  G., and  Sahai,  E. 2013. Mechanotransduction and YAP-dependent matrix remodelling is required for the generation and maintenance of cancer-associated fibroblasts. Nat. Cell Biol., 15: 637–646.
•  Gabriel,  K.N.,  Jones,  A.C.,  Nguyen,  J.P.T.,  Antillon,  K.S.,  Janos,  S.N.,  Overton,  H.N.,  Jenkins,  S.M.,  Frisch,  E.H.,  Trujillo,  K.A., and  Bisoffi,  M. 2016. Association and regulation of protein factors of field effect in prostate tissues. Int. J. Oncol., 49: 1541–1552.
•  Grzesiak,  J.J.,  Ho,  J.C.,  Moossa,  A.R., and  Bouvet,  M. 2007. The integrin-extracellular matrix axis in pancreatic cancer. Pancreas., 35: 293–301.
•  Hirata,  E.,  Ishibashi,  K.,  Kohsaka,  S.,  Shinjo,  K.,  Kojima,  S.,  Kondo,  Y.,  Mano,  H.,  Yano,  S.,  Kiyokawa,  E., and  Sahai,  E. 2020. The Brain microenvironment induces DNMT1 suppression and indolence of metastatic cancer cells. iScience, 23: 101480.
•  Ishibashi,  K.,  Ichinose,  T.,  Kadokawa,  R.,  Mizutani,  R.,  Iwabuchi,  S.,  Togi,  S.,  Ura,  H.,  Tange,  S.,  Shinjo,  K.,  Nakayama,  J.,  Nanjo,  S.,  Niida,  Y.,  Kondo,  Y.,  Hashimoto,  S.,  Sahai,  E.,  Yano,  S.,  Nakada,  M., and  Hirata,  E. 2024. Astrocyte-induced mGluR1 activates human lung cancer brain metastasis via glutamate-dependent stabilization of EGFR. Dev. Cell, 59: 579–594.
•  Ishihara,  S.,  Enomoto,  A.,  Sakai,  A.,  Iida,  T.,  Tange,  S.,  Kioka,  N.,  Nukuda,  A.,  Nagasato,  A.I.,  Yasuda,  M.,  Tokino,  T., and  Haga,  H. 2025. Stiff extracellular matrix activates the transcription factor ATF5 to promote the proliferation of cancer cells. iScience, 28: 112057.
•  Ishihara,  S. and  Haga,  H. 2022. Matrix stiffness contributes to cancer progression by regulating transcription factors. Cancers, 14: 1049.
•  Ishihara,  S.,  Yasuda,  M.,  Harada,  I.,  Mizutani,  T.,  Kawabata,  K., and  Haga,  H. 2013. Substrate stiffness regulates temporary NF-κB activation via actomyosin contractions. Exp. Cell Res., 319: 2916–2927.
•  Levental,  K.R.,  Yu,  H.,  Kass,  L.,  Lakins,  J.N.,  Egeblad,  M.,  Erler,  J.T.,  Fong,  S.F.T.,  Csiszar,  K.,  Giaccia,  A.,  Weninger,  W.,  Yamauchi,  M.,  Gasser,  D.L., and  Weaver,  V.M. 2009. Matrix crosslinking forces tumor progression by enhancing integrin signaling. Cell, 139: 891–906.
•  Lu,  Y.,  Huang,  Y.,  Zhu,  C.,  Li,  Z.,  Zhang,  B.,  Sheng,  H.,  Li,  H.,  Liu,  X.,  Xu,  Z.,  Wen,  Y.,  Zhang,  J., and  Zhang,  L. 2025. Cancer brain metastasis: Molecular mechanisms and therapeutic strategies. Mol. Biomed., 6: 12.
•  Ma,  Z.,  Gao,  X.,  Shuai,  Y.,  Wu,  X.,  Yan,  Y.,  Xing,  X., and  Ji,  J. 2020. EGR1-mediated linc01503 promotes cell cycle progression and tumorigenesis in gastric cancer. Cell prolif., 54: e12922.
•  Maher,  E.A.,  Mietz,  J.,  Arteaga,  C.L.,  DePinho,  R.A., and  Mohla,  S. 2009. Brain metastasis: Opportunities in basic and translational research. Cancer Res, 69: 6015–6020.
•  Myung,  E.,  Park,  Y.-L.,  Kim,  N.,  Chung,  C.-Y.,  Park,  H.-B.,  Park,  H.-C.,  Myung,  D.-S.,  Kim,  J.-S.,  Cho,  S.-B.,  Lee,  W.-S., and  Joo,  Y.-E. 2013. Expression of early growth response-1 in human gastric cancer and its relationship with tumor cell behaviors and prognosis. Pathol. Res. Pract., 209. 692–699.
•  Nair,  P.,  Muthukkumar,  S.,  Sells,  S.F.,  Han,  S.S.,  Sukhatme,  V.P., and  Rangnekar,  V.M. 1997. Early growth response-1-dependent apoptosis is mediated by p53. J. Biol. Chem., 272: 20131–20138.
•  Onishi,  K.,  Ishihara,  S.,  Takahashi,  M.,  Sakai,  A.,  Enomoto,  A.,  Suzuki,  K., and  Haga,  H. 2023. Substrate stiffness induces nuclear localization of myosin regulatory light chain to suppress apoptosis. FEBS Lett., 597: 643–656.
•  Pan,  L.,  Matloob,  A.F.,  Du,  J.,  Pan,  H.,  Dong,  Z.,  Zhao,  J.,  Feng,  Y.,  Zhong,  Y.,  Huang,  B., and  Lu,  J. 2010. Vitamin D stimulates apoptosis in gastric cancer cells in synergy with trichostatin A/sodium butyrate-induced and 5-aza-2'-deoxycytidine-induced PTEN upregulation. FEBS J., 277: 989–999.
•  Piersma,  B.,  Hayward,  M.-K., and  Weaver,  V.M. 2020. Fibrosis and cancer: A strained relationship. Biochim. Biophys. Acta Rev. Cancer, 1873: 188356.
•  Quail,  D.F. and  Joyce,  J.A. 2017. The microenvironmental landscape of brain tumors. Cancer Cell, 31: 326–341.
•  Tao,  B.,  Song,  Y.,  Wu,  Y.,  Yang,  X.,  Peng,  T.,  Peng,  L.,  Xia,  K.,  Xia,  X.,  Chen,  L., and  Zhong,  C. 2020. Matrix stiffness promotes glioma cell stemness by activating BCL9L/Wnt/β-catenin signaling. Aging, 13: 5284–5296.
•  Virolle,  T.,  Adamson,  E.D.,  Baron,  V.,  Birle,  D.,  Mercola,  D.,  Mustelin,  T., and  Belle,  I. d. 2001. The Egr-1 transcription factor directly activates PTEN during irradiation-induced signalling. Nat. Cell Biol., 3: 1124–1128.
•  Wang,  B.,  Guo,  H.,  Yu,  H.,  Chen,  Y.,  Xu,  H., and  Zhao,  G. 2021. The role of the transcription factor EGR1 in cancer. Front. Oncol., 11: 642547.
•  Yu,  J.,  Baron,  V.,  Mercola,  D.,  Mustelin,  T. and  Adamson,  E.D. 2006. A network of p73, p53 and Egr1 is required for efficient apoptosis in tumor cells. Cell Death. Differ., 14: 436–446.