Tonge D , Zhu N , Lynham S , Leclere P , Snape A , Brewer A , Schlomann U , Ferdous T , Tennyson C , Bartsch JW , Ward M , Pizzey J .

Biotechnology and Biological Sciences Research Council

	Figure 1. Regenerating axons grow towards epithelial cells migrating from skin explants in matrigel. Dark field images show cell migration from a skin explant and axonal outgrowth from the end of a peripheral nerve (PN) after (A) 1 day, (B) 2 days and (C) 3 days in culture. (D) Calcein-labelled image at 3 days. Axonal outgrowth increased rapidly after 2 days, and by 3 days had reached the migrating epithelial cells (arrows). C and D are of a slightly different field to A and B, to include axons from an adjacent PN-DRG growing towards the skin explant. Regenerating axons converged on the leading fronts of migrating cells from explants of cornea (E) and skin (F). Sympathetic ganglia (SG) axons also showed striking convergent axonal growth towards a skin explant (G). Axonal growth was attracted towards skin explants even in the presence of NT-4. Initial axon outgrowth from a PN-DRG preparation was radial, but, approximately 500 μm from the skin explant, axons began to grow towards a small area of migrating cells (H). Axonal growth towards skin explants also depends on gel composition. After 3 days culture in a fibrin gel, a PN-DRG showed some cell migration but no axonal outgrowth towards a skin explant (I), although abundant axonal outgrowth (arrows) occurred 1 day after addition of BDNF (J). In lamininidogen complex gels, directed axonal outgrowth towards skin explants occurred within 3 days (K). Bar: 500 μm in A and H; 1000 μm in G. To quantify the effects of skin and cornea on axonal growth from PN-DRGs, calcein-AM fluorescence images were first inverted to facilitate visualization of individual axons (L). Lines were then drawn on either side and parallel to the end of the cut peripheral nerve, and the proportions of axons extending 200 μm, 400 μm and 600 μm from the end of the nerve were determined as percentages of the numbers of axons initially growing from the end of the nerve on both sides. The lower panels show percentages of axons (as proportions of the numbers initially growing from the cut end of the peripheral nerve) extending laterally to different distances on the side next to the skin (M) or cornea (N) as compared with the opposite side (three experiments). In M, the differences are significant at 200 μm, 400 μm and 600 μm (P < 0.001, P < 0.01 and P < 0.001, respectively), and in N they are significant at 200 μm and 400 μm (P < 0.005 and P < 0.02, respectively).
	Figure 2. Two-dimensional gel autoradiographs of 35S-labelled proteins in media conditioned by cornea and skin explants after culture in matrigel for 4 days in the absence (A and C) or presence (B and D) of ActD. The silver-stained gels corresponding to C and D are also shown (E and F, respectively). Many spots in A and C are absent in B and D, indicating that synthesis and release of these proteins depend on RNA synthesis during culture. These include a charge train of proteins of approximately 50 kDa, present in A and C (boxed areas) and clearly visible with silver staining in E, but absent in B, D and F. Two other proteins of approximately 25 kDa (arrows) are present in A and C (although not visible with silver staining in E), but are absent in B and D.
	Figure 3. (A) Silver-stained two-dimensional gel of proteins in media conditioned by Xenopus froglet skin explants cultured in matrigel for 4 days, showing the spots from the approximately 50-kDa charge train picked for analysis of PTMs by LC-MS/MS after initial identification of spots 4, 5, 8, 9, 10, 11 and 12 as pro-MMP-18. (B) LC-MS/MS sequence coverage for pro-MMP-18. The derived sequence is shown in plain type, and peptides not covered by LC-MS/MS are underlined. The O-linked glycosylation domain is shown boxed, and possible residues for glycosylation within this are shown in bold type.
	Figure 4. Migrating epithelial cells express MMP-18. Tadpole skin explants were cultured for 3 days in matrigel and viewed under fluorescence for MMP-18 immunoreactivity (A) or stained with 4′,6-diamidino-2-phenylindole to label cell nuclei (B and D). MMP-18 labelling was marked at the leading edge of the cells, where they had migrated furthest (arrows). (C) Background fluorescence in a preparation in which the primary antibody was omitted. Bar: 100 μm. (E) Western blot showing pro-MMP-18 (approximately 49 kDa) and a much fainter band corresponding to mature MMP-18 (arrow) in media conditioned by tadpole skin explants cultured in matrigel for 4 days.
	Figure 5. Directed axonal growth requires MMP activity. In the presence of 1 μm BB-94 BB-94 (A) or 25 μm GM6001 (C), axonal outgrowth from the ends of peripheral nerves occurred but was not directed towards skin explants (left side of the panels), although this did occur in control cultures (E) containing vehicle [0.1% dimethylsulphoxide (DMSO)]. The mean proportions (standard error of the mean) of axons (expressed as percentages of the numbers initially growing from the cut end of the peripheral nerve in three independent experiments) extending to different distances on the side adjacent to skin as compared with the contralateral side are also shown. In the presence of BB-94 (B) and GM6001 (D), the differences were not significant. By contrast, in the presence of DMSO (F), the differences were significant at 200 μm, 400 μm and 600 μm (P < 0.005, P < 0.001 and P < 0.005, respectively).
	Figure 6. Fluorescence and dark field (upper and middle panels, respectively) images of a nodose ganglion and skin explant co-cultured in matrigel containing DQ gelatin, showing onset of axonal growth in relation to MMP activity. Fluorescence around the edges of the skin explants increased in intensity and extent from day 1 (A, D) to day 3 (C, F). No specific fluorescence was seen when DQ gelatin was omitted from the matrigel (data not shown). Plots of changes in fluorescence intensity (in arbitrary units) along a line between the skin and ganglion (dots) are shown in the lower panels. Axonal growth was already predominantly oriented (arrow in E) towards the skin explant on day 2, although the zone of MMP activity (B) had not yet extended to the tips of the growing axons. By day 3, axons had continued to extend through the area of MMP activity to make contact with migrating cells from the skin explants (F). Bar: 200 μm.
	Figure 7. The orientation of neurite outgrowth from dissociated DRG neurons is influenced by diffusible factor(s). After 3 days in culture, the orientation of neurite outgrowth of neurons on a layer of collagen was orthogonal to adjacent skin explants (towards the bottom of each panel) in an underlying layer of matrigel (A) or lamininidogen gel (C), but generally not at distances > 500 μm (B). In the presence of batimastat, the orientation of neurite outgrowth was not influenced by skin explants, and neurite lengths were shorter (D). Scale bar: 100 μm in A, B and D; 250 μm in C.
	Figure 8. Forced expression of MMP-18 in COS-7 cells promotes axonal growth. (A) Western blot showing bands corresponding to the molecular mass of pro-MMP-18 in CM from COS-7 cells stably transfected with pro-MMP-18, concentrated eight-fold (lanes 2 and 3), but scarcely detectable in unconcentrated medium (lanes 4 and 5). These two pairs of lanes represent CM from replicate cultures. By contrast, pro-MMP-18 was readily detectable in unconcentrated skin explant CM (lane 6), indicating that its concentration was higher than in media conditioned by COS-7 cells. MMP-18 was not detected in concentrated media conditioned by COS-7 cells stably transfected with the empty vector (lane 1). (B and C) Co-cultures of PN-DRGs with COS-7 cells (target). In co-cultures with COS-7 cells transfected with empty vector, longer axonal outgrowth was seen (visualized by calcein-AM fluorescence and image inversion) from the end of the peripheral nerve (PN) adjacent to aggregates of the COS-7 cells than from the opposite side (B). However, the difference in axonal lengths was much more marked in co-cultures with COS-7 cells expressing MMP-18 (C). Bar: 200 μm. Lower panels: percentages of axons (as proportions of the numbers of axons initially growing from the cut end of the peripheral nerve) extending laterally to different distances on the target side (next to control COS-7 cells transfected with empty vector or COS-7 cells expressing pro-MMP-18) as compared with the contralateral side. Results from three independent experiments were pooled, and differences in percentages between target and contralateral sides at different distances from the COS-7 cells were compared by paired t-test. In cultures with the control COS-7 cells (D), significantly more axons extended towards the COS cells at 200 μm and 400 μm (P < 0.01 and P < 0.02, respectively), but not at 600 μm. By contrast, significantly more axons extended towards the COS cells expressing pro-MMP-18 (E) at all three distances measured (P < 0.001), indicating that these cells had a greater stimulatory effect on axonal outgrowth than the control COS-7 cells. The differences in gaps between the cut ends of the peripheral nerves and COS-7 cell aggregates transfected with pro-MMP-18 cDNA and empty vector (879 94 μm and 846 85 μm, respectively) were not significant.
	Figure 9. (A) Western blots demonstrating nidogen cleavage by MMPs released from cultured skin and cornea. In lane 1, where freshly diluted matrigel (100 ng) was applied, there are several high molecular mass bands (> 80 kDa), but no smaller breakdown products are visible. By contrast, in lanes 2, containing media conditioned by polymerized matrigel without (lanes 2) or with (lanes 5) Xenopus skin explants for 3 days, there are several lower molecular mass bands (280 kDa), and fragments of 50 kDa and 90 kDa (arrows) are observed in samples incubated with Xenopus skin explants (lanes 5 and 6), but are absent from BB-94-treated cultures (lane 7). In lanes 2/5 and 3/6, samples are from two independent experiments. (B) Media conditioned by lamininidogen gels containing skin explants (lanes 1 and 2) or cornea explants (lane 3). Note the absence of the 50-kDa band (arrow) in CM from a BB-94-treated culture (lane 1). In media conditioned by cornea in lamininidogen complex gels, there is an additional nidogen fragment of 57 kDa, marked by an arrowhead. (C) Diluted lamininidogen complex (100 ng/μL) incubated with cell culture supernatants from COS7 cells transfected with empty vector (lane 1) or an MMP-18 expression vector (lane 2). The arrow marks the position of the cleaved nidogen band (50 kDa).