Fig 2: Expression of MT1-MMP in the healthy adult mouse retina. a Immunostaining with ab7 on a flatmounted retina revealed that MT1-MMP is expressed by all RGC axons. b-c The axonal labeling was confirmed by a double staining for phosphorylated neurofilament heavy (detected with the RT-97 antibody). Scale bar, 200 μm. d-d’ In addition to this abundant expression by RGC axons, MT1-MMP expression is also seen in the Müller glia with ab7. Of note, glial expression levels of MT1-MMP are lower than in the axons, and image exposure time was lengthened to reveal this more faint MT1-MMP expression. e A double staining with glutamine synthetase (GS), confirmed that – besides expression in the axons (arrow 1) –, ab7 also visualized MT1-MMP expression in Müller glia end feed (arrow 2) and radial processes (arrow 3) and in photoreceptor outer segments (arrow 4) on retinal sections. Scale bar, 20 μm. f Immunostaining with ab8 confirmed MT1-MMP expression in the Müller glia end feet (arrow 2) and radial processes (arrow 3) and in photoreceptor inner segments (arrow 4), but did not reveal axonal expression. g-h MT1-MMP expression in Müller glia was confirmed by a double staining with glutamine synthetase. Scale bar, 20 μm. i Western blotting with antibody ab7 revealed soluble, active MT1-MMP (50–53 kDa), in addition to several MT1-MMP subspecies (40 kDa, 32 kDa, 27 kDa). Western blotting with ab8 revealed membrane-bound, active MT1-MMP (57 kDa), as well as soluble, active MT1-MMP (50 kDa) and 44 kDa, 40 kDa, 32 kDa and 18 kDa fragments.

Fig 3: The H4K16 HAT hMOF is recruited to microglial upregulated genes. BV2 microglia are cultured for 4 h as monocultures or with C6 glioma cells as segregated cocultures and immunoprecipitated with hMOF (A, C, E, G and I) or H4K16ac (B, D, F, H and J) antibodies. hMOF and H4K16ac ChIP signals in 1 kb region downstream of the transcription start sites for the genes Il1β, Il6, Mmp14, Ccl22 and Chil3 were analyzed using qPCR and normalized to the levels of input DNA. BV2 microglia grown as monocultures serves as a control, set as 1. *P < 0.05 and **P < 0.01 (two-tailed Student's t-test). Data are from three (Ccl22 and Chil3) or five (Il1β, Il6 and Mmp14) independent experiments (mean and s.d.).

Fig 4: Glioma‐induced microglia tumor‐supporting phenotype is associated with increased H4K16 acetylation. (A) qPCR analysis of Il1β, Il6, Mmp14, Ccl22, Chil3 and Nos2 mRNA expression in BV2 microglia grown as monocultures or with C6 glioma cells as segregated cocultures (key); results are presented relative to those of each monoculture, set as 1. (B) Immunoblot analysis of MMP14 and β-actin (right margin, molecular size, in kilodaltons (kDa)) with quantification (C) in BV2 microglia cultured for 4 h as a monoculture or with C6 glioma cells as segregated coculture; results are presented to relative to those of BV2 microglia monoculture, set as 1. (D) Quantification of the migration of C6 glioma cells in Transwells with BV2 microglia or C6-cocultured BV2 microglia in the lower compartment; results are presented relative to those of C6 cells alone, set as 1. (E and F) Immunoblot analysis of H4K16ac and β-actin (right margin, molecular size, in kilodaltons (kDa)) (E) with quantification (F) in BV2 microglia control (0 h) or cultured for 2 h, 4 h or 6 h as monoculture or with C6 glioma cells as segregated coculture; results are presented to relative to those of BV2 microglia control (0 h), set as 1. (G) Confocal microscopy of BV2 microglia cultured for 4 h as a monoculture or with C6 glioma cells as segregated coculture, with immunostaining for H4K16ac and Hoechst nuclear counterstain. Scale bar, 10 μm. ns, non-significant, *P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001 (two-tailed Student's t-test). Data are from three independent experiments (A to D) or five independent experiments (E and F) (mean and s.d.).

Fig 5: Mmp14a knockdown minimally affects overall embryonic morphogenesis. A Knockdown of Mmp14a with 2 ng of Mmp14a-ATG MO at 3 dpf results in embryos with edema and a deformed body axis. However, embryos injected with 2 ng of both Mmp14a-ATG and p53 MOs do not show any morphological abnormalities. B Validation of Mmp14a-ATG MO efficiency via Western blotting using antibodies to zebrafish Mmp14a or to mouse MMP14, reveals reduced levels of active Mmp14a protein (60 kDa band) in 30 hpf Mmp14a morphants. Total protein coomassie blue staining was used for loading control. C RT-PCR analysis of Mmp14a-splice MO injected embryos reveals efficient splice blocking, resulting in a dose-dependent increase in aberrantly spliced mmp14a mRNA (345 bp band) and a concomitant decrease in correctly spliced mmp14a mRNA (168 bp band). D Semi-quantitative scoring of gastrulation defects at the tail bud stage (10 hpf) does not reveal a significant delay in gastrulation in Mmp14a morphant (Mmp14aKD) embryos, as compared to control embryos. The black arrowheads indicate the position of the germ cell layer (n = 141 from 3 independent experiments). E Measurements of the OVL, analysed by dividing EED by IED, in 31 hpf embryos, confirms the absence of a developmental delay after Mmp14a knockdown (n = 86 from 2 independent experiments). F Whole mount in situ hybridization for the neural markers dlx2, pax2.1 and shh shows normal development of the brain in control and Mmp14a morphant embryos at 32 hpf (n = 20 from 2 independent experiments). G Combined knockdown of Mmp14a and p53 (Mmp14aKD) only results in minor craniofacial abnormalities, including mild malformations of the Ceratohyal cartilage (white arrow) and the Meckel's cartilage (black arrow) at 5 and 11 dpf. Left and middle panel are ventral views, the right panel is a lateral view. C, control; dpf, days post fertilization; EED, Eye-Ear-Diameter; IED, Inner-Ear-Diameter; hpf, hours post fertilization; MO, morpholino; OVL, Otic Vesicle Length.