Context Dependent Effects of the Transforming Growth Factor-beta Signaling and Role Played by WNT4 in the Activation of Fibroblasts
Abstract
Transforming growth factor-β (TGF-β) superfamily of cytokines comprises of several members, which can broadly be sub-divided into three classes [TGF-βs, Activin/Nodal, and Bone morphogenetic proteins (BMPs)]. Most members of this family play critical roles during embryo development differentiation and regulation of homeostasis. In mammals there are three TGF-β isoforms, TGF-β1, 2 and 3. All the three TGF-β isoforms have important roles in embryo development as revealed by mouse knock-out models. TGF-β has also been associated with several pathological conditions such as inflammation, Fibrosis, and cancer.
In cancers, TGF-β plays both tumor suppressive and tumor promoting roles depending upon the context. TGF-β has growth inhibitory effect on epithelial cells which is essential to maintain tissue homeostasis. TGF-β induces the expression of several cyclin dependent kinase inhibitors such as p21Cip1, p15Ink4b while down-regulating the expression of cMYC in the epithelial cells. In lieu of its tumor suppressive role, several cancers harbor mutations in the components of the TGF-β signaling axis such as receptors and effector molecules called SMADs. Interestingly various cancers also show hyper activation of TGF-β signaling. It has been suggested that cancer cells become unresponsive to the growth inhibitory effects of TGF-β by losing the expression of p21Cip1, and p15INK4b. Oncogenic transformation of cancer cells can override the growth inhibitory effects of TGF-β. While the loss of growth inhibitory effects by TGF-β
are seen in the tumor cells, several tumor promoting actions are also observed in these cells such as induction of EMT. TGF-β activates mesenchymal cells leading to the formation of a reactive stroma in tumors and TGF-β suppresses almost all types of cells of the immune system causing a local immune-suppressive environment. TGF-β also recruits mesenchymal stem cells into the stroma which secrete several cytokines. The sum total of all these effects is pro-angiogenic, pro-infiltrative and pro-metastatic.
In the canonical TGF-β signaling pathway, ligands bind to the hetero-tetrameric receptor complex of TGFβR1 and TGFβR2 leading to activation of the TGFβR1 by TGFβR2.
Activated TGFβR1 then phosphorylates and activates R-SMAD molecules (SMAD2, SMAD3) which complexes with the co-SMAD (SMAD4) and translocate into the nucleus to effect transcriptional changes. Non-canonical TGF-β signals are many and almost all the known signaling pathways like MAPK, WNT, PI3K-AKT, NOTCH, Integrin, Hedgehog, Hippo etc. have been shown to be activated by TGF-β in different contexts.
The canonical TGF-β/SMAD pathway has been shown to be essential for both tumor suppressive and tumor promoting actions of TGF-β. Although the non-canonical signalling pathways have been shown to be context dependent, the exact mechanisms have not been elucidated. In previous studies, we have shown the importance of non-canonical TGF-β signaling in normal vs. carcinoma cells. However, there has been no study that addressed the differential effects of TGF-β on cells of connective tissue origin. To throw light on such questions we have undertaken this study with the following objectives:
1) Whole genome expression profiling of TGF-β targets in normal fibroblasts, transformed fibroblasts and sarcoma cells
2) Elucidation of non-canonical signaling pathways differentially regulated by TGF-β
3) Identification and characterization of novel TGF-β targets
The cell-lines chosen for the study are: 1) hFhTERT (human foreskin fibroblasts
immortalized with human terminal telomerase); 2) hFhTERT-LTgRAS (hFhTERT
transformed with SV40 large T antigen and activated RAS); and 3) HT1080 (fibrosarcoma).
We performed whole genome expression profiling using 4×44K Agilent Human Whole Genome Oligonucleotide Arrays. Analysis of the microarray results revealed that TGF-β regulated a large number of genes in all the three cell-lines but few targets were found to be commonly regulated between any two or all the three cell-lines. 5291 genes were differentially regulated by TGF-β between hFhTERT and hFhTERT-LTgRAS and 2274 genes were differentially regulated by TGF-β between hFhTERT and HT1080 cells. Gene set enrichment analysis (GSEA) of these two gene lists revealed enrichment of similar gene sets in the HT1080 and hFhTERT-LTgRAS cells compared to the hFhTERT cells. MAPK signaling pathway components were enriched in the hFhTERT cells. Closer inspection revealed that several upstream regulators of the MAPK pathway were in fact down-regulated by TGF-β in these cells compared to both hFhTERT-LTgRAS and HT1080 cells suggesting a depression of the MAPK pathway by TGF-β in the hFhTERT cells. Assessment of the phosphorylation status of ERK1/2 and p38 MAPK proteins after TGF-β treatment showed that both ERK1/2 and p38 MAPK pathways were not activated in response to TGF-β in the hFhTERT cells. On the other hand in hFhTERT-LTgRAS and HT1080 cells, both ERK1/2 and p38 MAPK were activated post TGF-β treatment. Activity of the AP1 and SMAD responsive p3TP-lux reporter plasmid was dependent on only the SMAD pathway in hFhTERT cells while in the hFhTERT-LTgRAS and HT1080 cells both MAPK and SMAD pathway were found to regulate the expression of the p3TP-lux
reporter. This suggests activation of MAPK and SMAD pathways in transformed and tumor cells while there is no activation of MAPK in normal cells of mesenchymal origin. Components of the WNT signaling pathway such as WNT ligands WNT4, and WNT11, frizzled receptors, FZD4, FZD8 and FZD9, regulators like SFRP1, SFRP2, AXIN2 and several targets of the WNT-β-catenin pathway were regulated by TGF-β in the hFhTERT cells but not in the hFhTERT-LTgRAS and HT1080 cells suggesting a positive regulation of the pathway by TGF-β in the hFhTERT cells. Indeed, TGF-β induced the activity of the WNT responsive reporter, pTOP-FLASH in the hFhTERT cells but not in the hFhTERTLTgRAS and HT1080 cells.
WNT4 and WNT11 were two of the novel targets of TGF-β identified in hFhTERT cells. Further experiments suggested that TGF-β conferred regulation of these genes was specific to the fibroblast cells since induction of these genes by TGF-β was not observed in any of the cancer cell lines or in HaCaT cells. Some recent studies have demonstrated remodelling of cytoskeleton in epithelial cells by the non-canonical WNT ligands such as WNT5a, WNT4 and WNT11. WNT4 has also been shown to be required for the maintenance of α-SMA levels in smooth muscle cells. In this study we have shown that WNT4 can induce α-SMA in the hFhTERT cells leading to their activation. TGF-β conferred activation of these cells was also found to be dependent on the presence of WNT4. In brief, our study identified differentially activated pathways by TGF-β in immortal and transformed fibroblasts. WNT4 was identified as a crucial molecule required for the TGF-β conferred activation of fibroblasts.