Hehr CL et al. (2005), Matrix metalloproteinases are required for reti...

XB-ART-1715

Development 2005 Aug 01;13215:3371-9. doi: 10.1242/dev.01908.

Show Gene links Show Anatomy links

Matrix metalloproteinases are required for retinal ganglion cell axon guidance at select decision points.

Hehr CL , Hocking JC , McFarlane S .

???displayArticle.abstract???
Axons receive guidance information from extrinsic cues in their environment in order to reach their targets. In the frog Xenopus laevis, retinal ganglion cell (RGC) axons make three key guidance decisions en route through the brain. First, they cross to the contralateral side of the brain at the optic chiasm. Second, they turn caudally in the mid-diencephalon. Finally, they must recognize the optic tectum as their target. The matrix metalloproteinase (MMP) and a disintegrin and metalloproteinase (ADAM) families are zinc (Zn)-dependent proteolytic enzymes. The latter functions in axon guidance, but a similar role has not yet been identified for the MMP family. Our previous work implicated metalloproteinases in the guidance decisions made by Xenopus RGC axons. To test specifically the importance of MMPs, we used two different in vivo exposed brain preparations in which RGC axons were exposed to an MMP-specific pharmacological inhibitor (SB-3CT), either as they reached the optic chiasm or as they extended through the diencephalon en route to the optic tectum. Interestingly, SB-3CT affected only two of the guidance decisions, with misrouting defects at the optic chiasm and tectum. Only at higher concentrations was RGC axon extension also impaired. These data implicate MMPs in the guidance of vertebrate axons, and suggest that different metalloproteinases function to regulate axon behaviour at distinct choice points: an MMP is important in guidance at the optic chiasm and the target, while either a different MMP or an ADAM is required for axons to make the turn in the mid-diencephalon.

???displayArticle.pubmedLink??? 15975939
???displayArticle.link??? Development

Species referenced: Xenopus laevis
Genes referenced: adam10 adam28.2 adm fgf2 isl1 mmp11 mmp13 mmp2 mmp9.2 ntn1 slit1 slit2 tec
???displayArticle.antibodies??? GABA Ab2 Glia Ab2 Isl1/2 Ab1 Neuronal Ab1

???attribute.lit??? ???displayArticles.show???

	Fig. 5. Optic pathway apparently unaffected by MMP inhibition. Embryos were exposed at stage 31 to either a control solution (A,C,E,G,I,K) or 25 μM SB-3CT (B,D,F,H,J,L). At stage 37/38, embryos were fixed. (A-D) Embryos were processed as wholemounts for in situ hybridization for Xenopus slit mRNA. Vibratome sections (50 μm) were cut. The ventral diencephalon at the level of the optic chiasm is represented in A,B, and eye sections are shown in C,D. (E-H) Immunoreactivity for the glial cell marker 3CB2. (G,H) Higher power views of E,F, showing the optic nerves and the optic chiasm in the ventral diencephalon. (I,J) Islet 1 immunoreactivity showing label in the ventral region of the diencephalon adjacent to the optic chiasm, and labelling of RGCs and cells in the inner nuclear layer in the retina. (K,L) GABA immunoreactivity showing labelling of amacrine cells in the inner nuclear layer, and a population of neurons in the ventral diencephalon. L, lens; nc, notochord; ap, anterior lobe of the pituitary body; nr, neural retina; PE, pigment epithelium; oc, optic chiasm; on, optic nerve, RGCL, RGC layer; Am, amacrine cells; INL, inner nuclear layer; ONL, outer nuclear layer; Di, diencephalon. D, dorsal; V, ventral. Orientation in A is for A,B,E,L, and orientation in C is for C,D. Scale bars: in A, 50 μm for A,B; 25 μm for C,D; in E, 50 μm for E,F,I,L; 25 μm for G,H.
	Fig. 6. MMP2 is expressed in tissues that abut the developing visual system. Embryos at stages 32 (A) and 35/36 (B) were processed as wholemounts for in situ hybridization with a digoxigenin-labelled (blue) XMMP2 antisense riboprobe. Vibratome transverse sections (50 μm) through the embryo at the level of the diencephalon. Pigmented cells are shown in black. Broken line represents the ventral diencephalon, with the optic chiasm found at the midline. The arrowheads point to Xmmp2 mRNA-expressing cells that neighbour the optic tectum. R, retina; oc, optic chiasm; Tec, tectum; Ve, ventricle; D, dorsal, V, ventral. Scale bar: 50 μm.
	Fig. 1. The MMP inhibitor SB-3CT has differential effects on two guidance decisions made by RGC axons. Lateral views of the brains and HRP-labelled optic projections of stage 40 Xenopus embryos exposed at stage 33/34 to either control media or to the MMP-specific inhibitor IV (SB-3CT). (A) Control exposed brain. (B-D) Brains exposed to SB-3CT. At concentrations of 10 μM (B) and 25 μM (C) axons extend, make the turn in the mid-diencephalon (star), but are misguided at their target, the optic tectum. A demonstration of the turning angle at the mid-diencephalon is shown in A and C. By contrast, in GM6001-treated brains we found that axons failed to make the turn, and instead grew towards the pineal gland (Webber et al., 2002). (D) With a higher concentration (50 μM) of the inhibitor, axon extension defects are sometime seen. (E,F) A cyclic peptide MMP inhibitor, CTT, produces a similar mistargeting phenotype (F), while a highly related inactive peptide, STT, has no effect on the optic projection (E). The white dots indicate the approximate anterior of the optic tectum (see Materials and methods). D, dorsal; A, anterior; Tec, tectum; Tel, telencephalon; Di, diencephalon; Pi, pineal gland; Hb, hindbrain. Scale bar: 50 μm.
	Fig. 3. RGC axons that cross the optic chiasm in the presence of metalloproteinase inhibitors fail to extend into the contralateral diencephalon. The optic chiasm was exposed at stage 30-31, by removing the cement gland and the mesenchyme underlying the ventral forebrain. Embryos were bathed in control media (A) or in media to which either the broad-spectrum metalloproteinase inhibitor GM6001 (5-10 μM; D), SB-3CT (25 μM; C) or the GM6001 negative control (5-10 μM; B) was added. At stage 40, RGC axons were anterogradely labelled with HRP. In GM6001 and SB-3CT, the optic projection (arrowhead) failed to enter the contralateral diencephalon. Tec, tectum. Scale bar: 50μ m.
	Fig. 4. Axon guidance defects caused by application of metalloproteinase inhibitors as RGC axons cross the optic chiasm. Vibratome (50 μm) sections through the diencephalon and eyes of embryos exposed at stage 31 to either control (A,C), SB-3CT (B,D,E) or GM6001 (H,I) solutions. The HRP-labelled optic projections are visible. (A,C) In control, RGC axons cross the midline at the optic chiasm and grow dorsally through the diencephalon to innervate the optic tectum. C is a higher power view of the optic chiasm shown in A. (B,D,E) In the 25 μM SB-3CT-treated brains, RGC axons often failed to enter the contralateral diencephalon (star), or grew aberrantly into the contralateral optic nerve (arrows). E is a higher power view of the optic projection shown in B and D is another example of a 25 μM SB-3CT-treated brain. (F,G) Fluorescent micrographs of the HRP-labelled optic projections in 12 μm transverse cryostat sections through the optic chiasm in the ventral diencephalon. Control axons shown in F cross the midline (arrow), whereas SB-3CT treated axons (G) stalled at the midline (arrow). (G) A more weakly labelled optic projection was chosen for visualization purposes. (H,I) Similar axonal phenotypes are observed in a 10 μM GM6001-treated brain. (H) Axons in the optic chiasm are seen to defasciculate (arrowheads). In a more caudal section (I), axons are seen within the contralateral eye (arrowhead). Tec, tectum; D, dorsal; V, ventral; oc, optic chiasm; R, retina. Scale bar: 50 μm for A,B,H,I; 25 μm for C-G.
	Fig. 2. Quantitation of the effects of SB-3CT on RGC axon outgrowth. Camera lucida representations of the brains and optic projections of control and treated embryos were digitized and analyzed for two measures of optic projection development: the mean angle through which RGC axons turned at the mid- diencephalon (A); and the length of the optic projection (B). (A) Dose-response curve for SB-3CT effects on the mean turning angle. The inset shows a schematic of how the turning angle was measured for a control and a SB-3CT-treated brain. Briefly, a standard reference line (1) was drawn between the optic chiasm and the midbrain-hindbrain boundary. A second line (2) was drawn at a 60° angle to line 1, and the angle through which the RGC axons turn in the mid-diencephalon was measured as the angle between line 2 and a third line (3) drawn through the middle of the optic projection just after the turn. A demonstration of the turning angle at the mid- diencephalon is shown in Fig. 1A,C. (B) Dose-response curve for SB-3CT effects on the optic projection length. Error bars are s.e.m., and numbers in brackets represent the number of optic projections analyzed. (***P<0.001; one-way ANOVA, Student-Newmann Keuls post-hoc comparisons test).