MAG and OMgp synergize with Nogo-A to restrict axonal growth and neurological recovery after spinal cord trauma

Abstract

Functional recovery after adult CNS damage is limited in part by myelin inhibitors of axonal regrowth. Three molecules, Nogo-A, MAG, and OMgp, are produced by oligodendrocytes and share neuronal receptor mechanisms through NgR1 and PirB. While each has an axon-inhibitory role in vitro, their in vivo interactions and relative potencies have not been defined. Here, we compared mice singly, doubly, or triply mutant for these three myelin inhibitor proteins. The myelin extracted from Nogo-A mutant mice is less inhibitory for axons than is that from wild-type mice, but myelin lacking MAG and OMgp is indistinguishable from control. However, myelin lacking all three inhibitors is less inhibitory than Nogo-A-deficient myelin, uncovering a redundant and synergistic role for all three proteins in axonal growth inhibition. Spinal cord injury studies revealed an identical in vivo hierarchy of these three myelin proteins. Loss of Nogo-A allows corticospinal and raphespinal axon growth above and below the injury, as well as greater behavioral recovery than in wild-type or heterozygous mutant mice. In contrast, deletion of MAG and OMgp stimulates neither axonal growth nor enhanced locomotion. The triple-mutant mice exhibit greater axonal growth and improved locomotion, consistent with a principal role for Nogo-A and synergistic actions for MAG and OMgp, presumably through shared receptors. These data support the hypothesis that targeting all three myelin ligands, as with NgR1 decoy receptor, provides the optimal chance for overcoming myelin inhibition and improving neurological function.

Figure 1.

nogoab^trap/trapmag ^−/− omgp ^−/− mice have normal brain anatomy. A, Immunoblot of cortical brain lysates from adult wild type, nogoab^trap/trap, mag ^−/− omgp ^−/−, and nogoab^trap/trapmag ^−/− omgp ^−/− detecting Nogo-A, MAG, OMgp, Nogo receptor-1, MBP, and GAPDH. Photomicrographs illustrate comparable myelin staining (white and green) in coronal sections of brain and cervical spinal cord in wild-type (B, D, E, F), nogoab^trap/trap (G, I, J, K), mag ^−/− omgp ^−/− (L, N, O, P), and nogoab^trap/trapmag ^−/− omgp ^−/− (Q, S, T, U) mice. Higher-power photomicrographs of motor cortex reveal comparable neuronal (NeuN, red and white) density and architecture in wild-type (C, D) nogoab^trap/trap (H, I), mag ^−/− omgp ^−/− (M, N), and nogoab^trap/trapmag ^−/− omgp ^−/− (R, S) mice. Scale bars: B (for B, G, L, Q), 2 mm; C (for C, D, H, I, M, N, R, S), 200 μm; E (for E, F, J, K, O, P, T, U), 500 μm.

Figure 2.

Extracted myelin from nogoab^trap/trapmag ^−/− omgp ^−/− mice is a less potent inhibitor of DRG neurite outgrowth. A–E, Photomicrographs of adult wild-type DRG cells immunostained for βIII tubulin after being cultured for 18 h on substrates coated with vehicle (A) or myelin extracted from wild-type (B), mag ^−/− omgp ^−/− (C), nogoab^trap/trap (D), and nogoab^trap/trapmag ^−/− omgp ^−/− (E) mice. Quantification of neurite outgrowth (F, data are presented as mean neurite length in micrometers per neuron ± SEM) shows that DRG cells grown on a vehicle substrate had significantly longer average neurite extension than cells grown on myelin extracted from wild-type (B), mag ^−/− omgp ^−/− (C), and nogoab^trap/trap (*p < 0.01, ANOVA) mice. Myelin from nogoab^trap/trap mice is significantly less inhibitory than wild-type and mag ^−/− omgp ^−/− myelin (**p < 0.01, ANOVA). Myelin extracted from nogoab^trap/trapmag ^−/− omgp ^−/− mice was significantly less inhibitory to neurite outgrowth than wild-type, mag ^−/− omgp ^−/−, and nogoab^trap/trap myelin (^# p < 0.01, ANOVA), but insignificantly different from neurite extension observed on vehicle substrates. Scale bar, 100 μm.

Figure 3.

Bilateral sprouting of BDA+ CST fibers in cervical spinal cord after dorsal hemisection at T8 in nogoab^trap/trapmag ^{−[supi]/ −} omgp ^{−[supi]/ −} mice. Photomicrographs A–P illustrate BDA+ CST axons in cervical spinal cord from intact (A, B) and lesioned (I, J) wild-type mice, intact (C, D) and lesioned (K, L) mag ^−/− omgp ^−/− mice, intact (E, F) and lesioned (M, N) nogoab^trap/trap mice, and intact (G, H) and lesioned (O, P) nogoab^trap/trapmag ^−/− omgp ^−/− mice. BDA+ fibers can be seen exiting the contralateral dorsal column and arborizing in both dorsal and ventral gray matter, and remaining mostly unilateral in all genotypes. There was no significant difference in the number of BDA+ CST axons that crossed the midline in the cervical cord after sham lesion in wild-type, nogoab^trap/trap, mag ^−/− omgp ^−/−, and nogoab^trap/trapmag ^−/− omgp ^−/− mice. Assessment of the number of BDA+ fibers that cross the midline after DhX (Q) revealed that significantly more CST fibers sprouted ipsilaterally in nogoab^trap/trap (^## p < 0.01, ANOVA) than in sham-lesioned nogoab^trap/trap mice and lesioned wild-type and mag ^−/− omgp ^−/− mice (^# p < 0.01, ANOVA). Lesioned nogoab^trap/trapmag ^−/− omgp ^−/− mice revealed significantly more CST fibers crossing the midline than sham-lesioned nogoab^trap/trapmag ^−/− omgp ^−/− mice (**p < 0.01, ANOVA) and lesioned wild-type, nogoab^trap/trap, and mag ^−/− omgp ^−/− mice (*p < 0.01, ANOVA). Scale bars: A (for A, C, E, G, I, K, M, O), 500 μm; B (for B, D, F, H, J, L, N, P), 100 μm.

Figure 4.

Regeneration of BDA+ CST axons in nogoab^trap/trapmag ^−/− omgp ^−/− mice. A, C, E, G, I, K, M, O, Low-power photomicrographs of BDA+ CST axons in sagittal sections of thoracic spinal cord after dorsal hemisection in wild-type (A), nogoab^WT/trapmag ^+/− omgp ^+/− (C), nogoab^trap/trap (E), mag ^−/− omgp ^−/− (G), and nogoab^trap/trapmag ^−/− omgp ^−/− (I, K, M, O) mice, with finer detail shown in higher-power insets (i and ii) for each image. Camera lucida of 10 BDA+ sagittal sections from each mouse were reconstructed and projected in a single image (B, D, F, H, J, L, N, P) to illustrate the extent of CST axon growth. Q, Quantification of the number of BDA+ CST axons 3 and 1 mm rostral to the lesion and 0, 1, 2, and 3 mm caudal to the lesion. No significant difference in number of axons was observed between all genotypes of mice rostral to the lesion and between wild-type, nogoab^WT/trapmag ^+/− omgp ^+/−, and mag ^−/− omgp ^−/− mice caudal to the lesion. Nogoab^trap/trap displayed significantly more CST axons 0, 1, and 2 mm and caudal to the lesion site than wild-type, nogoab^WT/trapmag ^+/− omgp ^+/−, and mag ^−/− omgp ^−/− mice (*p < 0.01, ANOVA). The nogoab^trap/trapmag ^−/− omgp ^−/− mice displayed significantly more BDA+ CST axons at 0, 1, 2, and 3 mm caudal to the lesion site than wild-type, nogoab^WT/trapmag ^+/− omgp ^+/−, mag ^−/− omgp ^−/−, and nogoab^trap/trap, mice (**p < 0.001, ANOVA). Scale bars: A (for A–P), 2 mm; i (for i, ii), 100 μm.

Figure 5.

Regenerating CST axons preferentially grow in white matter in nogoab^trap/trapmag ^−/− omgp ^−/− mice. Low-power photomicrographs (A, C, E, G) illustrate composite projection images of four serial sagittal sections of spinal cord representing four zones, central (A), medial (C), mediolateral (E), and lateral (G), in one nogoab^trap/trapmag ^−/− omgp ^−/− mouse. Each composite photomicrograph is recreated in camera lucida with an individual color depicting axons from each original sagittal section for the central (B), medial (D), mediolateral (F), and lateral (H) zones. I, Entire sagittal spinal camera lucida reconstruction. J, A composite projection image of four sections of spinal cord (1 from each zone) stained with GFAP-IR, with the entire camera lucida reconstruction layered on top. High-power photomicrographs show many axons growing around and through the lesion site in the central (K), medial (L), mediolateral (M), and lateral (N) zones. O, Higher-power image of the entire camera lucida reconstruction merged with anti-GFAP staining (from J) shows the extent of the number of axons crossing into the caudal spinal cord. Scale bars: A (for A–J), 2 mm; K (for K–O), 100 μm.

Figure 6.

Regeneration of RST axons in nogoab^trap/trapmag ^−/− omgp ^−/− mice. Photomicrographs of L4 ventral spinal cord illustrate 5-HT-IR (white and green) and NeuN-IR (red) in sham-lesioned nogoab^trap/trapmag ^−/− omgp ^−/− (A, B) and wild-type (C, D), mag ^−/− omgp ^−/− (E, F), nogoab^trap/trap (G, H), and nogoab^trap/trapmag ^−/− omgp ^−/− (I, J) mice after dorsal hemisection. Quantification of 5-HT axon density (K) reveals that DhX results in a significant reduction in serotonergic axon density in wild-type (C, D), mag ^−/− omgp ^−/− (E, F), and nogoab^trap/trap (G, H) mice (^# p < 0.001, ANOVA) in comparison to sham-lesioned nogoab^trap/trapmag ^−/− omgp ^−/− mice (A, B). Significantly higher 5-HT axon density was observed in nogoab^trap/trap (G, H) mice after DhX than in wild-type (C, D) and mag ^−/− omgp ^−/− (E, F) mice (*p < 0.001, ANOVA). Significantly higher 5-HT axon density was observed in nogoab^trap/trapmag ^−/− omgp ^−/− mice after dorsal hemisection than in wild-type (C, D), mag ^−/− omgp ^−/− (E, F), and nogoab^trap/trap (G, H) mice (**p < 0.001, ANOVA). There was no significant difference in 5-HT axon density between intact (A, B) and lesioned (I, J) nogoab^trap/trapmag ^−/− omgp ^−/− mice (K). Scale bar (in B): 50 μm.

Figure 7.

Improved locomotor recovery in mice lacking Nogo-A/B after DhX. Photomicrographs of sagittal sections of thoracic spinal cord immunostained for GFAP in nogoab^trap/trap (A), mag ^−/− omgp ^−/− (B), nogoab^wt/trapmag ^+/− omgp ^+/− (C), and nogoab^trap/trapmag ^−/− omgp ^−/− (D) mice after DhX. Quantification of GFAP-negative spared tissue is equivalent between genotypes after DhX (E). All mice were behaviorally assessed using the BMS (F–J); nogoab^trap/trap (n = 11, blue line) recovered significant hindlimb function in comparison to nogoab^WT/trap mice (n = 9, stippled blue line, repeated-measures ANOVA, p < 0.001, F). There was no significant functional recovery in mag ^−/− omgp ^−/− (n = 12, red line) in comparison to mag ^+/− omgp ^+/− control mice (n = 8, stippled red line, G). nogoab^trap/trapmag ^−/− omgp ^−/− (n = 22, green line) recovered significant function in comparison to nogoab^WT/trapmag ^+/− omgp ^+/− control lesioned mice (green stippled line, repeated-measures ANOVA, p < 0.001, H). There was no significant difference in BMS scores between heterozygote mice at any behavioral time point (I); therefore, these groups were combined for comparison with behavioral scores of single, double, and triple mutants (WT/HT group, J). Repeated-measures ANOVA revealed that there was no significant difference in behavioral recovery between WT/HT (black line) and mag ^−/− omgp ^−/− (red line) mice; however, both nogoab^trap/trap (blue line, p < 0.005) and nogoab^trap/trapmag ^−/− omgp ^−/− (green line, p < 0.001) mice recovered significant hindlimb function in comparison to WT/HT mice. Furthermore nogoab^trap/trapmag ^−/− omgp ^−/− recovered significantly more hindlimb function than nogoab^trap/trap (p < 0.05). Scale bar, 500 μm.

Figure 8.

Improved locomotor recovery in mice lacking Nogo-A/B after subcomplete transection. Photomicrographs of sagittal sections of thoracic spinal cord immunostained for GFAP in nogoab^trap/trapmag ^−/− omgp ^−/− after dorsal hemisection (A) and nogoab^wt/trapmag ^+/− omgp ^+/− (B) and nogoab^trap/trapmag ^−/− omgp ^−/− (C) mice after subcomplete transection (scTx). D, Quantification of GFAP-negative spared tissue is significantly less in nogoab^wt/trapmag ^+/− omgp ^+/− and nogoab^trap/trapmag ^−/− omgp ^−/− mice after scTx than after hemisection (*p < 0.001, ANOVA). E, There was no significant difference between nogoab^wt/trapmag ^+/− omgp ^+/− and nogoab^trap/trapmag ^−/− omgp ^−/− mice after scTx. nogoab^trap/trapmag ^−/− omgp ^−/− mice recovered significant hindlimb function in comparison to nogoab^wt/trapmag ^+/− omgp ^+/− after scTx (repeated-measures ANOVA, p < 0.001). Scale bar, 500 μm.