Ciliary neurotrophic factor is not required for terminal sprouting and compensatory reinnervation of neuromuscular synapses: re-evaluation of CNTF null mice

Abstract

Loss of synaptic activity or innervation induces sprouting of intact motor nerve terminals that adds or restores nerve-muscle connectivity. Ciliary neurotrophic factor (CNTF) and terminal Schwann cells (tSCs) have been implicated as molecular and cellular mediators of the compensatory process. We wondered if the previously reported lack of terminal sprouting in CNTF null mice was due to abnormal reactivity of tSCs. To this end, we examined nerve terminal and tSC responses in CNTF null mice using experimental systems that elicited extensive sprouting in wildtype mice. Contrary to the previous report, we found that motor nerve terminals in the null mice sprout extensively in response to major sprouting-stimuli such as exogenously applied CNTF per se, botulinum toxin-elicited paralysis, and partial denervation by L4 spinal root transection. In addition, the number, length and growth patterns of terminal sprouts, and the extent of reinnervation by terminal or nodal sprouts, were similar in wildtype and null mice. tSCs in the null mice were also reactive to the sprouting-stimuli, elaborating cellular processes that accompanied terminal sprouts or guided reinnervation of denervated muscle fibers. Lastly, CNTF was absent in quiescent tSCs in intact, wildtype muscles and little if any was detected in reactive tSCs in denervated muscles. Thus, CNTF is not required for induction of nerve terminal sprouting, for reactivation of tSCs, and for compensatory reinnervation after nerve injury. We interpret these results to support the notion that compensatory sprouting in adult muscles is induced primarily by contact-mediated mechanisms, rather than by diffusible factors.

Figure 1

Nerve terminals and tSCs in CNTF−/− mice react normally to exogenous CNTF. High magnification confocal images of junctions from wildtype or CNTF−/− LAL muscles treated with BSA or CNTF for 14 days. Preterminal and terminal axons were immunolabeled for neurofilaments and a synaptic vesicle protein, SV2 (Green), Schwann cells were labeled for S100 (Blue), and muscle AChRs were labeled for α-bungarotoxin (Red). Insets of Figures 1A-C”’ show z-y optical sections through the horizontal axis indicated by the dotted line in the AChR panels. (A-A”’), A control junction treated with BSA. Terminal branches are covered by tSCs and tSC processes, which are precisely apposed to AChR clusters. Stars point to cell bodies of tSCs, as also in Fig. 1B', C', D'. (B-B”’), A typical junction from a CNTF+/+ muscle treated with CNTF displaying no extrasynaptic growth of terminal sprouts. Bundles of intraterminal sprouts (e.g., white arrows in Fig. 1B) extend along parental branches of axon terminals and tSCs. Extrasynaptic, but not intrasynaptic (e.g., blue arrows in Fig. 1'), growth of new tSC processes is also rare. (C-C”’) and (D-D”’), Representative junctions from CNTF−/− muscles treated with CNTF displaying vigorous intrasynaptic growth of terminal sprouts and tSC processes. As in wildtype junctions treated with CNTF, extraterminal sprouts/tSC processes are rare or absent. An arrowhead in Fig. 1D' points to a tSC process that extends in association with a terminal sprout (arrow in Fig. 1D). (E), Percent distribution of junctions with terminal sprouts. In muscles of both genotypes, nearly 90% of nerve terminals sprout in response to exogenous CNTF. Substantial numbers of junctions responded with only intrasynaptic growth of terminal sprouts. (F), Percent distribution of junctions with extraterminal sprouts of different length. The extent of terminal sprouting, measured by the length of extraterminal sprouts, does not differ in CNTF+/+ and −/− muscles. Data were collected from 206 and 249 en face junctions taken from 3 LAL muscles each for CNTF+/+ and −/− mice, respectively. extraT, extraterminal sprouts; intraT, intraterminal sprouts. Data are presented as mean ± SEM. Scale bar, 10 μm

Figure 2

Nerve terminals and tSCs in CNTF−/− mice react normally to acute paralysis. (A-A”) and (B-B”), Confocal images of representative junctions from CNTF−/− LAL muscles paralyzed for 4 days with BoTX. Muscles were labeled as described in Fig. 1. Short terminal sprouts (arrows in A, B) extend extrasynaptically in association with tSC processes that grow in advance of terminal sprouts (arrowheads in A', B') or independently of axons (arrows in A', B'). In addition, CNTF+/+ and −/− muscles show no difference in the percent distribution of junctions with sprouts (C), the average number of extraterminal sprouts extended from those junctions (D), or the average length of extraterminal sprouts (E). Stars in A' and B' point to cell bodies of tSCs. Data were collected from 114 and 121 en face junctions taken from 3 LAL muscles of CNTF+/+ or −/− mice. Scale bar, 10 μm

Figure 3

Motor nerve terminals in CNTF−/− mice sprout extensively in response to prolonged paralysis. (A, A'), Representative junctions of wildtype muscles paralyzed for 14 days displaying lengthy extraterminal sprouts (e.g., arrows). (B, B'), Representative junctions of CNTF−/− muscles paralyzed for 14 days displaying lengthy extraterminal sprouts (e.g., arrows). (C), Percent distribution of junctions with sprouts. Nearly 80% of nerve terminals in both CNTF +/+ and −/− muscles sprout by 14 days after paralysis. (D), Average number of terminal sprouts extended in CNTF+/+ or −/− muscles paralyzed for 7 or 14 days. (E), Percent distribution of junctions with extraterminal sprouts of different length in muscles paralyzed for 14 days. The length of terminal sprouts is similar in CNTF+/+ and CNTF−/− muscles. n=177, 205 junctions for 3 CNTF+/+ or −/− muscles paralyzed for 7 days; n=109, 120 junctions for 3 CNTF+/+ or −/− muscles paralyzed for 14 days. Data are presented as mean values ± SEM. Scale bar, 20 μm

Figure 4

Compensatory reinnervation by terminal sprouts and tSC bridges proceeds normally in CNTF−/− mice. Confocal image of an innervated and a reinnervated endplate in CNTF−/− muscles linked by a terminal sprout (A; white arrow) and a tSC bridge (B; black arrows). Muscles were partially denervated for 14 days and labeled as described in Fig. 1, except that p75 antibody was used together with S100 antibody to label both quiescent and reactive Schwann cells. Note that the terminal sprout that connects the two endplates (C) is associated with a tSC bridge, suggesting normal induction and guidance of terminal sprouts by tSCs in CNTF−/− muscles. Scale bar, 20 μm

Figure 5

Compensatory reinnervation by nodal sprouts proceeds normally in CNTF−/− mice. Muscles were partially denervated for 14 days and labeled as in Fig. 4. A nodal sprout (A; arrowhead) that extended along the preterminal SCs (B, arrow) innervates a previously denervated endplate (C, right endplate). The nodal sprout continues to grow along tSC processes (B, arrowheads), and elaborates an ‘escaped’ fiber (e.g., arrows in A) that extends beyond the synaptic boundary. These features are typical of reinnervation of wildtype muscles after partial denervation. Note that nodal sprouts are easily distinguishable from myelinated axons innervating intact endplates (A, a thick axon innervating the left endplate) both by axon thickness and by the extent of terminal arborization. Scale bar, 20 μm

Figure 6

Quiescent and reactive tSCs express little if any CNTF. Confocal images of intact or denervated junctions of CNTF+/+ (A-A”, C-C”) and CNTF−/− EDL muscles (B-B”). For 4-color labeling of denervated muscles, a transgenic mouse line expressing GFP selectively in Schwann cells was used (see Methods). (A-A”), Representative junction from intact, wildtype muscle displaying CNTF expression only in myelinating Schwann cells (A”, arrows). (B-B”), A junction from intact, CNTF−/− muscle verifying complete elimination of CNTF in the CNTF null mice. (C-C”), A junction from a CNTF+/+ EDL muscle, completely denervated for 10 days, displays faint if any CNTF labeling in tSCs and their processes. Image of axons were omitted. Note also marked reduction of CNTF expression in some of the denervated myelinating Schwann cells (e.g., C”, arrow). Scale bar, 20 μm