Fig 2: Hypoxic glia in the retina after OHT. Sagittal retinal sections were labeled with glial markers. (A). GFAP (glial fibrillary acidic protein, astrocyte marker, yellow) showed colocalization (arrows) with tdT (magenta) in GCL. (B). Iba-1 (ionized calcium binding adaptor molecule 1, microglia marker, yellow) showed colocalization (arrows) with tdT in INL. (C). Vim vimentin, Müller glia marker, yellow) showed colocalization (arrows) with tdT. (D). Pimonidazole (pimo, hypoxia marker, green, arrows) labeling was reported in Müller glia, and cells in GCL 6 h, 3 d, and 28 d after OHT. Pimonidazole was shown separately since it was overridden by tdT in previous images. tdT (tdTomato) expression (magenta), DAPI (cell nuclei, blue). Retinal layers were labeled as follows: GCL (ganglion cell layer), INL (inner nuclear layer), and ONL (outer nuclear layer). Scale bar: 25 µm. n = 6 eyes at 6 h, 6 eyes at 3 d, 6 eyes at 14 d, 6 eyes at 28 d, and 10 control eyes. Insets represent a magnification of hypoxic glia.

Fig 3: Tumor formation and metastasis by PyMT-1099 cells. (A) PyMT-1099 cells untreated or treated with TGFß for >20 days (PyMT-1099 LT) were injected orthotopically into mammary fat pads of NSG mice. The graph represents tumor growth in PyMT-1099 and PyMT-1099 LT group of mice. (B) Histological tumor sections from tumors of PyMT-1099 or PyMT-1099 LT cells described in (A) were stained with H&E to assess the morphology of primary tumors. Representative microphotographs are shown from tumors of 2 out of the 6 mice used in the experiment. (C) Immunofluorescence analysis was performed to assess the expression of the EMT markers FN1, VIM, E-CAD and N-CAD in tumors formed by PyMT-1099 or PyMT-1099 LT cells in the experiment described in (A). DAPI was used as a nuclear counterstain. Representative pictures are shown from tumors of 2 out of the 6 mice used in the experiment. Scale bar, 100 µm. (D) The graph represents the number of lung metastases formed in NSG mice orthotopically transplanted with PyMT-1099 or PyMT-1099 LT cells; n = 6. (E) The graph represents the number of lung metastases formed in NSG mice injected with PyMT-1099 or PyMT-1099 LT cells through the tail vein; n = 6. The mice were sacrificed 8 weeks post-injection, and lungs were resected for the analysis of cancer cell colonization/ metastases formation.

Fig 4: VIF nuclear cage impacts nuclear shape and volume. (a) Confocal immunofluorescence images of the VIF network in cells fixed and stained with anti-vimentin (top, green) and nucleus (DAPI, white); and in orthogonal view for the 3D renderings of the nuclei of the same cells (bottom) in vim+/+ and vim-/- mEFs. Images show loss of vimentin changes nuclear morphology. Scale bar is 20 µm. The nuclei of cells lacking vimentin are rounder and less smooth compared with those in vim+/+ mEFs. (b) Vim+/+ and rescued mEFs have more compressed, elongated nuclear shapes compared with the vim-/- mEFs (n = 83–219 per condition from at least n = 2 independent experiments). (c) Vim-/- mEFs have significantly smaller spread areas compared with the vim+/+ or rescued mEFs (n = 62–87 per condition from n = 2 independent experiments). (d) Vim+/+ and vim-/- mEFs are elongated to the same extent as determined by measuring cell aspect ratio (n = 150–200 per condition from n = 4 independent experiments). Data are presented as means ± standard error, as noted in the statistical analysis section. NS, non significant. (e and f) Lack of vimentin decreases nuclear (e; n = 83–219 per condition from at least n = 2 independent experiments) and cell volume (f; n = 62–87 per condition from n = 2 independent experiments). The lower and upper parts of the box represent the 25th and 75th percentiles. The ends of the notch indicate 1.57 times the interquartile range divided by the square root of the number of samples. The whiskers indicate the range of the circular variance up to 1.5 times the interquartile range away from the quartiles.(g) Cell volume linearly correlates with nuclear volume but with different slopes (P < 0.035) for vim+/+ (9.1) and vim-/- mEFs (7.2); the shaded regions correspond to 95% confidence levels based on linear regression analysis. (h) Quantification of nuclear shape by determinations using Fourier amplitudes, u(q) (n = 40–43) following staining for immunofluorescence with anti-nuclear lamin A to to trace nuclear contour in vim+/+ and vim-/- mEFs. The normalized amplitude decreases as q2 with a prefactor based on undulated membrane theory that is inversely proportional to the nuclear stress (n = 25 cells per condition from n = 2 independent experiments). The shaded blue and red regions of h represent ± 1 SEM for vim+/+ and vim-/- cells, respectively. Scale bar is 5 µm.

Fig 5: Loss of vimentin increases nuclear damage but does not alter lamin expression levels. (a–c) Immunofluorescence images showing lamin A and lamin B1/2 on the top and bottom of membranes with 3 µm pores (a); on the membrane bottom, vim-/- mEFs exhibit nuclear abnormalities such as large nuclear blebs (b), and some blebs exhibit dilated lamin A networks (c). (d) Migration through 3-µm pores results in an increased percentage of nuclei with blebs in vim+/+ and vim-/- mEFs, but vim-/- mEFs have a significantly higher percentage of migration-induced blebs (n = 70–115 per condition from n = 2 independent experiments). Data are presented as means ± standard error, as noted in the statistical analysis section. (e and f) Loss of vimentin increases the number of postmigration DNA double-strand break foci, as determined by immunofluorescence with anti-?H2AX in nuclei after mEFs have migrated to the bottom of membranes with 3 µm pores (n = 70–115 per condition from n = 2 independent experiments). (g) Foci appear throughout the nucleus and in protruding blebs. Vim+/+ and vim-/- mEFs express equivalent levels of lamin A, B1/2, and C, as shown by quantitative immunoblotting. Scale bars: a, 10 µm; b, 2 µm; c, 10 µm, 2 µm (magnification); f, 10 µm. NS, non significant.