Integrins Can Act as Suppressors of Ras-Mediated Oncogenesis in the Drosophila Wing Disc Epithelium

Abstract

Cancer is the second leading cause of death worldwide. Key to cancer initiation and progression is the crosstalk between cancer cells and their microenvironment. The extracellular matrix (ECM) is a major component of the tumour microenvironment and integrins, main cell-ECM adhesion receptors, are involved in every step of cancer progression. However, accumulating evidence has shown that integrins can act as tumour promoters but also as tumour suppressor factors, revealing that the biological roles of integrins in cancer are complex. This incites a better understating of integrin function in cancer progression. To achieve this goal, simple model organisms, such as Drosophila, offer great potential to unravel underlying conceptual principles. Here, we find that in the Drosophila wing disc epithelium the βPS integrins act as suppressors of tumours induced by a gain of function of the oncogenic form of Ras, Ras^V12. We show that βPS integrin depletion enhances the growth, delamination and invasive behaviour of Ras^V12 tumour cells, as well as their ability to affect the tumour microenvironment. These results strongly suggest that integrin function as tumour suppressors might be evolutionarily conserved. Drosophila can be used to understand the complex tumour modulating activities conferred by integrins, thus facilitating drug development.

Keywords: Drosophila integrins; cancer; cell growth; invasion.

Figure 1

Integrin knockdown enhances Ras^V12 hyperplasia in the Drosophila wing disc. (A–E,A’’–E’’) Confocal views of third-instar larvae wing discs of the indicated genotypes, stained with anti-GFP (green), Rhodamine Phalloidin (RhPh) to detect F-actin (red in A–E’, white in A”–E”’) and Hoechst for DNA detection (blue). (A’–E’,A’’’–E’’’) Confocal yz sections of wing discs, along the white dotted line shown in (A–E). (F) Violin plots of the percentage of the apterous domain area versus total area per disc. The statistical significance of differences was assessed with a Welch test, ****, **, and * p values <0.0001, <0.01 and <0,05, respectively. Scale bars 50 μm (A–E,A’’–E’’) and 10 μm (A’–E’,A’’’–E’’’).

Figure 2

Inhibition of integrin expression in Ras^V12 tumoural cells does not affect cell polarity or BM deposition. (A) Scheme of apico-basal polarity in the pseudostratified epithelium of the Drosophila wing disc, showing the Dorsal (green) and Ventral (blue) domains, the localization of the apical determinant aPKC (red) and the BM component Perlecan (yellow). (B–I’) Confocal xz sections of third-instar larvae wing discs of the designated genotypes, stained with anti-GFP (green), Hoechst (DNA, blue), aPKC (red in B–E, white in B’–E’) and Perlecan (yellow in F–I, white in F’–I’). Scale bars 50 μm (B–I’).

Figure 3

Integrin expression downregulation increases Ras^V12-dependent cell shape changes and growth in Drosophila wing discs. (A–E) Maximal projection of confocal images of third-instar larvae wing discs of the designated genotypes, stained with anti-GFP (green), RhPh (red in A–E, white in A’–E’’) and Hoechst for DNA detection (blue, A’’’–E’’’). (A’–E’) Apical and (A”–E”) basal surface views of the region specified in the white boxes in (A–E). (A’’’–E’’’) Confocal xz sections along the white dotted lines of wing discs shown in (A–F). The apical side of wing discs is at the top. White brackets indicate cell height. (F–H) Violin plots of the apical cell area (F), basal cell area (G) and cell height (H) of the indicated genotypes. The statistical significance of differences was measured using the welch-test, ****, ***, **, * and ‘ns’ p values <0.0001, <0.001, <0.01, <0.05 and >0.05, respectively. Scale bars 50 μm (A–E), 5 μm (A’–E’,A”–E”) and 10 μm (A’’’–E’’’).

Figure 4

Integrin knockdown enhances the changes in cell cycle progression in Drosophila Ras^V12 wing disc cells. (A) Scheme depicting CycB-RFP and E2F1-GFP nuclear expression during the cell cycle. (B–F) Maximal projection of confocal images of the apterous domains of third-instar larvae wing discs of the indicated genotypes, expressing E2f1-GFP (green) and CycB-RFP (red). (B’–F’) Scatter plots representing the fluorescence intensity of E2f1-GFP (green) and CycB-RFP (red). The statistical significance of differences was measured using the Xi-square test, with all comparisons having a p-value < 0.0001. Scale bars 50 μm (B–F).

Figure 5

Integrin knockdown increases the ability of Ras^V12 to induce JNK activity in wild-type adjacent cells. (A–E’) Maximal projection of confocal views of third-instar wing discs of the specified genotypes, stained with anti-GFP (green), anti-pJNK (red in A–E, white in A’–E’) and Hoechst (DNA, blue). (F) Violin plots of mean fluorescent pJNK intensity at the dorsal-ventral boundary of wing discs of the designated genotypes. The statistical significance of differences was assessed with the welch-test, ****, ** and * p values < 0.0001, <0.01 and <0.05, respectively. Scale bars, 50 μm (A–E’).

Figure 6

Integrins downregulation enhances the invasive behaviour of Ras^V12 cells. (A–E) Confocal views of third-instar larvae wing discs of the designated genotypes, stained with anti-GFP (green in A–E’ and white in A’’–E’’), DCP1 (red in A–E’ and white in A’’’–E’’’) and Hoechst (DNA, blue) (A–E’). (A’–E’’’) Magnifications of the white boxes in (A–E), respectively. (F) Bar plot of the frequency of appearance of non-dead GFP⁺ cells in the ventral domain. Scale bars 50 μm (A–E) and 10 μm (A’–E’’’).

Figure 7

Integrins regulate the invasive capacity of Ras^V12 cells. (A–A’’) Scheme illustrating the region of the wing disc where movies had been taken. (A) Confocal view of a third-instar larvae wing disc, stained with Hoechst (blue), RhPh (red) and anti-GFP (green), and (A’) and its corresponding orthogonal view. (A’’) Zoom-in of the region in the yellow dotted box in (A’), corresponding to the region where movies had been taken. (B–F’’’) Confocal images taken with a 40 min difference of live wing discs of the designated genotypes.