FIGURE SUMMARY
Title

Matrix metalloproteinases as promising regulators of axonal regrowth in the injured adult zebrafish retinotectal system

Authors
Lemmens, K., Bollaerts, I., Bhumika, S., De Groef, L., Van Houcke, J., Darras, V.M., Van Hove, I., Moons, L.
Source
Full text @ J. Comp. Neurol.

Mapping of RGC axonal regrowth after ONC in adult zebrafish. A,B: Western blotting (WB) for Gap-43 (20-25 kDa) on retinal extracts harvested at different timepoints after optic nerve crush (ONC) reveals a peak in Gap-43 expression at 4 days postinjury (dpi). For the quantitative analysis, values from ONC samples were plotted relative to sham-operated values, which were put at 100%. Data were collected from at least three retinal samples per timepoint and are shown as mean ± SEM (*P < 0.05). C-F: Immunostainings for Gap-43 on retinal sections at various stages postinjury show Gap-43 upregulation in retinal ganglion cell (RGC) somata at 4 dpi and increasing Gap-43 signal in RGC axons from 4-7 dpi. Gap-43 expression reapproaches control levels at 14 dpi. DAPI (blue) was used as nuclear counterstain. G,H: Microscopic images of biocytin-labeled brain sections of uncrushed control fish (UCC) and 7 dpi fish visualizing innervation of the contralateral optic tectum (OT) by RGC axons. Regenerating RGC axons entering the OT at 1 week postinjury are indicated by arrows. I: Quantification of the area covered by RGC axons in the OT at 7 dpi, relative to UCC, confirms that axonal regeneration is fully ongoing. Data represent mean ± SEM (n = 5 animals per group, *P < 0.05). ONC, optic nerve crush; dpi, days postinjury; UCC, uncrushed control; NFL, nerve fiber layer; RGCL, retinal ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer; PRL, photoreceptor layer. Scale bars = 20 µm in C-F; 200 µm in G,H.

Retinal broad-spectrum MMP inhibition after ONC reduces OT reinnervation. A-D: Representative images depicting reinnervation of the OT by RGC axons at 7 dpi after PBS, DMSO, or GM6001 treatment, reveal a clearly diminished reinnervated OT area after broad-spectrum matrix metalloproteinase (MMP) inhibition (see arrows) as opposed to vehicle control and UCC fish. E: Quantification of the area covered by RGC axons in the OT of the various treatment conditions, relative to UCC fish. Both DMSO- and PBS-treated fish show 60% reinnervation of the OT relative to the completely innervated OT in the UCC condition. A significant reduction in tectal reinnervation was noted in GM6001-treated fish compared to the vehicle controls. Data represent mean ± SEM and 8-15 animals were used per condition over three independent experiments (*P < 0.05; **P < 0.005; ***P < 0.001). UCC, uncrushed control; PBS, phosphate-buffered saline; DMSO, dimethyl sulfoxide. Scale bars = 200 µm.

Lowering general MMP activity after ONC does not influence RGC survival, nor reduces retinal thickness. A,B: Representative confocal images of retinal sections immunostained for activated caspase-3 reveal no apoptotic cells in GM6001-treated eyes at 2 dpi (A) and only a limited number of apoptotic cells in the RGCL at 7 dpi (see arrows) (B). C-F: Magnifications of an activated caspase-3 immunostaining on retinal sections of PBS-, DMSO-, and GM6001-treated fish indicate a similar amount of apoptotic RGCs at 7 dpi in the three conditions, while apoptosis is absent in UCC fish. Moreover, no obvious disruption of retinal structure, nor retinal thinning can be observed in GM6001-injected fish. DAPI (blue) was used as nuclear counterstain. G: Quantitative analysis of the ratio of activated caspase-3+ cells/DAPI+ cells in the RGCL reveals no difference between PBS-, DMSO-, and GM6001-treated fish at 7 dpi. Data represent five fish per group and are shown as mean ± SEM. H: Likewise, the number of DAPI+ cells per 100 µm of RGCL was similar in all conditions, both at 2 and 7 dpi, confirming that no significant cell loss is induced after ONC/treatment. Data represent five fish per group and are shown as mean ± SEM. I: A quantitative analysis on retinal sections did not reveal any significant differences in retinal thickness of uninjected, PBS-, DMSO-, and GM6001-treated fish at 7 dpi. Six retinal sections were analyzed per fish, using five fish per condition. Data are shown as mean ± SEM. Dpi, days postinjury; PBS, phosphate-buffered saline; DMSO, dimethyl sulfoxide; NFL, nerve fiber layer; RGCL, retinal ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer; PRL, photoreceptor layer. Scale bars = 50 µm in A,B; 20 µm in C-F.

Retinal spatiotemporal expression pattern of Mmp-13a protein after ONC in adult zebrafish. A,B: Representative picture and bar graph revealing western blotting analysis data for Mmp-13a on retinal extracts after ONC, showing significantly increased active Mmp-13a (45 kDa) levels at 7 dpi, represented as a relative percentage to sham-operated values. No differences in pro Mmp-13a (80 kDa) were observed. Data were collected from at least three retinal samples per timepoint and are shown as mean ± SEM (**P < 0.005). C-E: Immunohistochemical stainings for Mmp-13a on retinal sections of sham-operated fish and fish with ONC, show an elevated Mmp-13a signal in the RGCL at 7 dpi, during axonal regrowth. DAPI (blue) was used as nuclear counterstain. F-H: High-magnification photographs of the RGCs after H&E, DAB, or immunofluorescent staining for Mmp-13a, altogether indicating expression of Mmp-13a in RGC somata and primary dendrites after ONC. DAPI (blue) or hematoxylin was used as nuclear counterstain. ONC, optic nerve crush; dpi, days postinjury; NFL, nerve fiber layer; RGCL, retinal ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer; PRL, photoreceptor layer. Scale bars = 20 µm.

Figure 5. Retinal spatiotemporal expression pattern of Mmp-2 protein after ONC in adult zebrafish. A: Representative picture showing the presence of active Mmp-2 in adult zebrafish retinal extracts, as confirmed by labeling of both pro (72 kDa) and active (64 kDa) Mmp-2 in zebrafish embryo lysates (left). A,B: Western blotting for Mmp-2 on retinal extracts at different timepoints postinjury indicates a peak in active Mmp-2 expression at 10 dpi during retinotectal regeneration, as compared to sham-operated controls (put at 100%). Data represent at least three retinal samples per timepoint and are shown as mean ± SEM (*P < 0.05). C-F: Immunostainings for Mmp-2 on retinal sections at 4 dpi disclose a decreased expression of Mmp-2 in RGC axons compared to controls, but an increased presence in RGC somata. During axonal regrowth, Mmp-2 expression becomes upregulated in RGC somata and axons, whereafter its expression again confines to RGC axons. G: An additional immunostaining on zebrafish retinal sections at 0 dpi, using a different MMP-2 antibody (EMD Millipore, AB19167), confirms Mmp-2 expression in the nerve fiber layer (NFL). For all immunostainings, DAPI (blue) was used as a nuclear counterstain. ONC, optic nerve crush; dpi, days postinjury; NFL, nerve fiber layer; RGCL, retinal ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer; PRL, photoreceptor layer. Scale bars = 20 µm.

Overlap of α-tubulin and Mmp-2 expression on retinal sections localizes Mmp-2 in/on axons of the NFL. A-D: A double immunostaining for α-tubulin and Mmp-2 on retinal sections shows a similar expression of both proteins in the NFL and thus indicates expression of Mmp-2 in/on RGC axons. B-D: High-magnification pictures of double (B) and single (C,D) stainings. DAPI (blue) was used as a nuclear counterstain. ONC, optic nerve crush; dpi, days postinjury; NFL, nerve fiber layer; RGCL, retinal ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer; PRL, photoreceptor layer. Scale bars = 20 µm in A; 5 µm in B-D.

Retinal spatiotemporal expression pattern of Mmp-14 protein after ONC in adult zebrafish. A,B: Representative picture and bar graph revealing western blotting analysis data for Mmp-14 on retinal extracts after ONC, respectively showing significantly increased mature Mmp-14 (55-60 kDa) levels at 6 hpi and decreased levels at 10 dpi. Values from ONC samples were plotted relative to sham-operated values, which were put at 100%. Data represent at least three retinal samples per timepoint and are shown as mean ± SEM (*P < 0.05). C-F: Immunostainings for Mmp-14 on retinal sections harvested at varying days after ONC equally show an upregulation of Mmp-14 in the nerve fiber layer (NFL) and inner plexiform layer (IPL) at 6 hours postinjury (hpi) and a decrease herein during axonal regrowth. G-J: Double staining for α-tubulin and Mmp-14 indicates expression of Mmp-14 in/on RGC axons in retinal sections of control sham-operated fish. H-J: High- magnification pictures of double (H) and single (I,J) stainings. K-N: Double labeling of retinal sections for MAP2 and Mmp-14 reveals Mmp-14 expression in IPL neurites at 6 hpi. L-N: High- magnification pictures of double (L) and single (M,N) stainings. O-P: Additional immunostainings on zebrafish retinal sections at 0 dpi, using two alternative antibodies for MMP-14, RP2-MMP-14, and RP5-MMP-14 (Triple Point Biologics, ATK-MMP-14) confirm Mmp-14 expression in the NFL and IPL neurites. * indicates the ON head. For all immunostainings, DAPI (blue) was used as a nuclear counterstain. ONC, optic nerve crush; hpi, hours postinjury; dpi, days postinjury; NFL, nerve fiber layer; RGCL, retinal ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer; PRL, photoreceptor layer. Scale bars = 20 µm in C-G,K; 5 µm in H-J,L-N; 50 µm in O-P.

Figure 8. Retinal spatiotemporal expression pattern of Mmp-9 protein after ONC in adult zebrafish. A: Representative picture showing presence of active Mmp-9 in zebrafish retinal extracts, as confirmed by labeling of both pro (92 kDa) and active (76 kDa) Mmp-9 in zebrafish embryo lysates (left). A,B: Western blotting for Mmp-9 on retinal extracts at different timepoints postinjury indicates a peak in active Mmp-9 expression at 14 dpi compared to sham-operated controls. Data represent at least three retinal samples per timepoint and are shown as mean ± SEM (*P < 0.05). C,D: Immunostainings for Mmp-9 on retinal sections show an increased Mmp-9 signal in three prominent IPL strata at 14 dpi. E-H: Double staining with SV2 confirms expression of Mmp-9 in a subset of IPL synapses. F-H: High-magnification pictures of double (F) and single (G,H) stainings. For all immunostainings, DAPI (blue) was used as a nuclear counterstain. ONC, optic nerve crush; dpi, days postinjury; NFL, nerve fiber layer; RGCL, retinal ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer; PRL, photoreceptor layer. Scale bars = 20 µm in C-E; 5 µm in F-H.

Acknowledgments
This image is the copyrighted work of the attributed author or publisher, and ZFIN has permission only to display this image to its users. Additional permissions should be obtained from the applicable author or publisher of the image. Full text @ J. Comp. Neurol.