Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2013 Sep 26;8(9):e74923.
doi: 10.1371/journal.pone.0074923. eCollection 2013.

FHOD1, a Formin Upregulated in Epithelial-Mesenchymal Transition, Participates in Cancer Cell Migration and Invasion

Free PMC article

FHOD1, a Formin Upregulated in Epithelial-Mesenchymal Transition, Participates in Cancer Cell Migration and Invasion

Maria Gardberg et al. PLoS One. .
Free PMC article


Cancer cells can obtain their ability to invade and metastasise by undergoing epithelial-to-mesenchymal transition (EMT). Exploiting this mechanism of cellular plasticity, malignant cells can remodel their actin cytoskeleton and down-regulate proteins needed for cell-cell contacts. The mechanisms of cytoskeletal reorganisation resulting in mesenchymal morphology and increased invasive potential are poorly understood. Actin nucleating formins have been implicated as key players in EMT. Here, we analysed which formins are altered in squamous cell carcinoma related EMT. FHOD1, a poorly studied formin, appeared to be markedly upregulated upon EMT. In human tissues FHOD1 was primarily expressed in mesenchymal cells, with little expression in epithelia. However, specimens from oral squamous cell cancers demonstrated consistent FHOD1 upregulation in mesenchymally transformed cells at the invasive edge. This upregulation was confirmed in an oral squamous carcinoma model, where FHOD1 expression was markedly increased upon EMT in a PI3K signalling dependent manner. In the EMT cells FHOD1 contributed to the spindle-shaped morphology and mesenchymal F-actin organization. Furthermore, functional assays demonstrated that FHOD1 contributes to cell migration and invasion. Finally, FHOD1 depletion reduced the ability of EMT cancer cells to form invadopodia and to degrade extracellular matrix. Our results indicate that FHOD1 participates in cytoskeletal changes in EMT. In addition, we show that FHOD1 upregulation occurs during cancer cell EMT in vivo, which indicates that FHOD1 may contribute to tumour progression.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.


Figure 1
Figure 1. Transcriptomic analysis of the UT-SCC-43A, UT-SCC-43B and 43A-SNA oral SCC cell lines.
A) Microarray transcriptomic profiling of UT-SCC-43A, UT-SCC-43B and 43A-SNA cells. Transcripts significantly altered in UT-SCC-43B cells in comparison with UT-SCC-43A cells are indicated by the blue circle and transcripts altered in 43A-SNA cells in comparison with UT-SCC-43A are indicated by the green circle. The number of genes altered in both comparisons is shown in the overlapping area. The number of up-regulated genes is shown on the top and down-regulated genes on the bottom. B) mRNA level alterations for selected epithelial (top) and mesenchymal (bottom) genes. Depicted genes encode following proteins: CDH1 =  E-cadherin, CLDN3 =  Claudin 3, CLDN 7 =  Claudin 7, KRT5 =  Keratin 5, KRT6B  =  Keratin 6b, CDH2 =  N-Cadherin, VIM  =  Vimentin. C) mRNA levels for formins with significant alterations in both UT-SCC-43B and 43A-SNA cells that were subsequently confirmed by RT-PCR. D) mRNA levels for formins confirmed by RT-PCR.
Figure 2
Figure 2. FHOD1 is strongly expressed in endothelium and plasma cells but only weakly in epithelial cells.
Paraffin-embedded normal human tissues were stained with FHOD1 antibody. A) In the spleen, endothelial cells lining blood vessels show intensive FHOD1 immunoreactivity. B) Endometrial stroma and glands express little FHOD1. Endothelial cells in blood vessel walls are stained strongly (arrowheads). C) In the ovary, oocytes (o) and follicle cells (asterisk) express little FHOD1. D) In the brain, neither neurones nor glial cells show FHOD1 staining. The endothelial cells in minute capillaries (arrows) stain weakly. E) In the lymph node, most lymphocytes stain weakly. Occasional strongly staining cells have plasma cell morphology (open arrowheads). F) In the bone marrow, erythropoietic cells do not express FHOD1. Myelopoietic cells and megakaryocytes (M) stain weakly. Occasional small cells that express FHOD1 strongly (open arrowheads) are plasma cells. G) In colonic mucosa, epithelial cells show only weak FHOD1 staining. Stromal plasma cells (open arrowheads) express FHOD1 strongly. H) Alveolar epithelium in the lung stains weakly for FHOD1, while alveolar macrophages (MP) show moderate FHOD1 staining. I, J) Double immunofluorescence staining of FHOD1-positive inflammatory cells demonstrates that the same cells express CD138 (arrowheads). The cell morphology and CD138 indicate that they are plasma cells. In A–H scale bar  = 100 μm.
Figure 3
Figure 3. FHOD1 is upregulated in clinical oral SCC, as well as in vitro in SCC cells with EMT features.
A) In normal or non-neoplastic stratified squamous epithelium, no FHOD1 can be detected (top). In invasive squamous cell carcinomas, moderate to strong FHOD1 immunoreactivity is seen in spindle-shaped cells at the invasive front that on morphological grounds have undergone EMT (bottom; details in inset). In the tumour bulk, which consists of cells with an epithelial morphology, only weak immunoreactivity is present. Scale bar: 200 μm. B) Western blot analysis of cell lines show that the epithelial SCC cell line UT-SCC-43A does not express detectable FHOD1, while both the spontaneous EMT cell line UT-SCC-43B and the Snail-induced EMT cell line 43A-SNA express FHOD1. UT-SCC-43A expresses the epithelial marker E-cadherin but not N-cadherin, whereas UT-SCC43-B and 43A-SNA express N-cadherin but not E-cadherin. C) F-actin organization of UT-SCC-43A (upper panel) is typically epithelial, with distinct cell submembraneous filaments and scant stress fibres. UT-SCC-43B shows features of mesenchymal organization (middle panel). Cell-cell contacts are few, the cells are elongated and contain lamellopodia, filopodia and stress fibres. The insert (bottom panel) shows that a fraction of FHOD1 co-localizes with stress fibres in UT-SCC-43B cells (arrowheads). Nuclei are stained with DAPI (blue). D) FHOD1 upregulation in UT-SCC-43B cells is dependent of PI3K signalling. Treatment with MEK 1/2 inhibitor U0126 reduces phosphorylation of ERK 1/2 but does not influence FHOD1 expression. In contrast, PI3K inhibition by LY294992 markedly reduces FHOD1 expression. The reduction of p-Akt indicates that the pathway is efficiently inhibited.
Figure 4
Figure 4. FHOD1 siRNA knockdown changes the morphology of UT-SCC-43B cells and reduces the number of stress fibres.
A) Western blotting shows that FHOD1 siRNA treatment significantly reduces FHOD1 expression. The cells are still E-cadherin negative and the expression of N-cadherin is unaltered. B) Cells treated with non-targeting siRNA (upper panel) express FHOD1 and have a mesenchymal phenotype. Stress fibres are abundant. FHOD1 siRNA abolishes FHOD1 staining (lower panel). siRNA treated cells have less actin stress fibres and are morphologically rounder and flatter. Nuclei are stained with DAPI (blue). C) F-actin staining is significantly reduced in FHOD1 siRNA treated cells. Phalloidin staining intensity is reduced by 49%. Bars indicate standard error of mean. *** p<0.0001. AU  =  arbitrary units.
Figure 5
Figure 5. FHOD1 silencing significantly reduces the motility and invasiveness of UT-SCC-43B cells.
UT-SCC-43B cells were treated with FHOD1 or control siRNA and subjected to wound healing assay and a Matrigel invasion assay. A) The wound of control cells has fully healed by 24 h, whereas the wound in siRNA treated cells is only partially healed. B) A graph showing the healing process at different time points. Bars indicate standard error of mean. Interaction between time and group are statistically significant (*** P<0.0001). The difference is statistically significant at every time point (P<0.002). n = 4 in each group. C) In a 72 hour Matrigel invasion assay, invasion is significantly reduced in FHOD1-deficient cells when compared to control siRNA treated cells. (* P<0.05) n = 3–5 in each group.
Figure 6
Figure 6. FHOD1 silencing inhibits proteolytic activity and invadopodia formation.
A) Images from a zymography assay performed with untreated, nt siRNA and FHOD1 siRNA treated SCC-43B cells. Degradation of Cy 3 labelled gelatin is reduced in FHOD1 siRNA treated cells. B) Quantification of the degraded area/cell. C) Invadopodia formation is visualized by phalloidin staining, which reveals the comet- or ring-shaped actin structures on the ventral cell surface (arrowheads). Invadopodia are present in a larger proportion of untreated and control cells than in FHOD1 siRNA treated cells. D) Quantification of percentage of cells with invadopodia. (* P<0.05). N.S.  =  not significant.

Similar articles

See all similar articles

Cited by 18 articles

See all "Cited by" articles


    1. Kalluri R, Weinberg RA (2009) The basics of epithelial-mesenchymal transition. J Clin Invest 119: 1420–1428. - PMC - PubMed
    1. Iwatsuki M, Mimori K, Yokobori T, Ishi H, Beppu T, et al. (2010) Epithelial-mesenchymal transition in cancer development and its clinical significance. Cancer Sci 101: 293–299. - PubMed
    1. Faix J, Grosse Rm (2006) Staying in shape with formins. Dev Cell 10: 693–706. - PubMed
    1. Schönichen A, Geyer M (2010) Fifteen formins for an actin filament: A molecular view on the regulation of human formins. Bioch Biophys Acta 1803: 152–163. - PubMed
    1. Parri M, Chiarugi P (2010) Rac and Rho GTPases in cancer cell motility control. Cell Commun Signal 8: 23. - PMC - PubMed

Publication types

MeSH terms

Grant support

This research was supported by funding from The Academy of Finland, Finska Läkaresällskapet, Sigrid Juselius Foundation, K Albin Johansson Foundation, The Finnish Cancer Organizations and Turku University Hospital Research Funds. Maria Gardberg is a Ph.D. student supported by the National Graduate School of Clinical Investigation. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.