A Novel Role of Connective Tissue Growth Factor in the Regulation of the Epithelial Phenotype

Abstract

Background: Epithelial-mesenchymal transition (EMT) is a biological process where epithelial cells lose their adhesive properties and gain invasive, metastatic, and mesenchymal properties. Maintaining the balance between the epithelial and mesenchymal stage is essential for tissue homeostasis. Many of the genes promoting mesenchymal transformation have been identified; however, our understanding of the genes responsible for maintaining the epithelial phenotype is limited. Our objective was to identify the genes responsible for maintaining the epithelial phenotype and inhibiting EMT.

Methods: RNA seq was performed using an vitro model of EMT. CTGF expression was determined via qPCR and Western blot analysis. The knockout of CTGF was completed using the CTGF sgRNA CRISPR/CAS9. The tumorigenic potential was determined using NCG mice.

Results: The knockout of CTGF in epithelial ovarian cancer cells leads to the acquisition of functional characteristics associated with the mesenchymal phenotype such as anoikis resistance, cytoskeleton remodeling, increased cell stiffness, and the acquisition of invasion and tumorigenic capacity.

Conclusions: We identified CTGF is an important regulator of the epithelial phenotype, and its loss is associated with the early cellular modifications required for EMT. We describe a novel role for CTGF, regulating cytoskeleton and the extracellular matrix interactions necessary for the conservation of epithelial structure and function. These findings provide a new window into understanding the early stages of mesenchymal transformation.

Keywords: anoikis resistance; cytoskeleton; epithelial–mesenchymal transition; extracellular matrix remodeling; metastasis; ovarian cancer.

Figure 1

Differentially expressed pathways and biological processes in epithelial ovarian cancer cells and cells in the E/M hybrid state. (A). Model outlining the process of epithelial–mesenchymal plasticity in R182 ovarian cancer cells as described previously in our lab by Tedja et al. [15]. (B). Volcano plot of differentially expressed genes (DEGs). Blue dots represent downregulated DEGs and red dots represent upregulated DEGs. (C). Bar plot of top 10 differentially regulated pathways. (D). Chord diagram of top three differentially regulated pathways and their associated DEGs (Star points to CTGF). (E). Top 20 differentially regulated biological processes (arrows highlight processes associated with ECM reorganization).

Figure 2

CTGF regulates epithelial and mesenchymal markers in ovarian cancer cells. (A). Western blot demonstrating CTGF expression in various OC cell lines. (B). Western blot demonstrating loss of CTGF by CRISPR KO in R182 OC cell lines. (C). Sanger sequencing verifying deletion of CTGF in R182 CTGF KO cell line. (D). Western blot evaluating expression of epithelial and mesenchymal markers in wild-type and CTGF-KO R182 cell lines. Representative figures of three independent experiments. N = 3. The uncropped blots are shown in Supplementary File S1.

Figure 3

CTGF negatively regulates anoikis resistance. R182 WT and R182 CTG-KO cells were cultured in ultra-low attachment conditions. (A) Culture morphology was assessed via microscopy after 24 and 48 h. Scale bar 1000 μm. (B) Cell viability was quantified at designated time points using Celltiter96 assay. Experiments were performed independently and in triplicate. Data are presented as mean ± SEM and an unpaired t-test was used to calculate statistical significance. (C) R182 WT, R182 CTG-KO, and R182 CTG-KO cells treated with 100 ng/mL recombinant CTGF were cultured in 50% Matrigel. Culture morphology was assessed via microscopy at day 6. Scale bar 1000 μm. (D) Quantitation of invasion assay at 160 h. Independent experiments were performed in triplicate. Data are presented as mean ± SEM and an unpaired t-test was used to calculate statistical significance. *** p ≤ 0.001. **** p ≤ 0.0001.

Figure 4

Loss of CTGF reprograms cell adhesion and ECM in OC cells. RNA sequencing was performed in R182 WT and R182 CTG_KO cells. (A) Volcano plot of differentially expressed genes (DEGs). Blue dots represent downregulated DEGs and red dots represent upregulated DEGs. (B) Bar plot of top 10 differentially regulated pathways. (C) Top seven differentially regulated biological processes. (D) Dendogram of top seven differentially regulated biological processes.

Figure 5

CTGF reprograms ECM–receptor interaction. (A) DEGs in the ECM–receptor interaction pathway comparing R81 WT and R182 CTGF-KO cells. (B) Validation of identified DEGs via qPCR. Data are presented as mean + SEM and an unpaired t-test was used to calculate statistical significance. *** p ≤ 0.001; ** p ≤ 0.01 and * p ≤ 0.05. (C) LAMC2 protein expression in cell lysate of WT and CTGF KO cells (arrow). (D) LAMC2 IF in R182 WT, R182 CTGF-KO, and R182CTGF-KO treated with 100 ng/mL rCTGF. Scale bar 10 μm. (E) Secreted LAMC2 protein expression in media of R182 WT and R182 CTGF-KO cells. (F) Addition of conditioned media from R182-CTGF KO cells to R182 WT cells confers anoikis resistance in R182 WT cells. Anoikis resistance was measured as described in the Materials and Methods section. Briefly, R182 cells were plated in either optimum or CTGF KO media and anoikis resistance was measured up to 72 h. Three independent experiments were performed in triplicate. Data are presented as mean ± SEM and an unpaired t-test was used to calculate statistical significance. * denotes p ≤ 0.01. The uncropped blots are shown in Supplementary File S1.

Figure 6

Loss of CTGF promotes extracellular matrix remodeling. (A) Proposed model of the role of CTGF in epithelial to mesenchymal cell transition in OC. (B) F-actin IF in R182 WT and R182 CTGF-KO cells demonstrates the presence of lamellopodia and (C) the reorganization of actin filaments. A representative figure of three independent experiments performed in triplicates.

Figure 7

Cell stiffness of CTGF KO and WT R182 cells. (A) Top rows are Brillouin images. Bottom rows are co-registered bright-field images. Dashed line indicates the cell body. Scale bar: 10 µm. (B) Brillouin shift results. Multiple measurements were taken for each cell line; wt (n = 50); KO (n = 26); and rCTGF (n = 39). * p < 1.6 × 10⁻⁵. ** p = 0.015. *** p = 0.002. (C) Invasion assay performed with R182 WT and R182-CTGF KO with 25% and 50% Matrigel measured at 120 h. Independent experiments were performed in quadruplicate. Data are presented as mean ± SEM and an unpaired t-test was used to calculate statistical significance. Scale bar 1000 μm. (D) Representative images of invasion assay with R182 WT and CTGF KO cells in 25 and 50% Matrigel.

Figure 8

Tumorigenicity. (A) Tumor growth curves of R182 WT and R182 CTGF-KO cells in NCG mice. R182 CTGF-KO cells can form s.c. tumors while no detectable tumors were observed with R182 wt cells. (B) Histology of tumors formed by R182 CTGF-KO cells. Note cells invading the Matrigel (M) and the presence of neovascular process (BV). Representative figures of five independent animals. Scale bar: 100 μm.