Results
CP11 treatments inhibit AEP activity at nerve injury sites – Asparaginyl endopeptidase activity was assayed at the site of sciatic nerve transection and repair at three and seven days after the start of treatments. Measures of activity at the two post-injury times from five nerves from WT mice treated with CP11 and five nerves from WT mice treated with vehicle, at two different times after injury, were compared to the activity measured in five intact nerves from control mice (Fig. 1). Results of a one-way ANOVA of these data were significant (F4, 20 = 9.362, p=0.0002). Based on the results ofpost hoc paired testing, AEP activity was increased significantly after nerve injury, relative to intact nerves, as was anticipated from the results of immunoblotting reported previously (English et al. , 2021). At both post-injury times studied, treatments with CP11 significantly reduced the magnitude of this increase in activity. After seven days of daily CP11 treatments, the reduction in AEP enzymatic activity was to a level that was not significantly different from that measured in the nerves of intact mice (Fig. 1).
CP11 treatments facilitate recovery of neuromuscular function – Compound muscle action potentials (M responses) were evoked in the LG and TA muscles by sciatic nerve stimulation four weeks after sciatic nerve transection and repair in mice treated for two weeks with either CP11 or vehicle (Fig. 2A). Examples of maximum M responses recorded from a vehicle-treated mouse and from a CP11-treated mouse four weeks after sciatic nerve transection and repair are shown in Figure 2B. Two types of treatments were evaluated: daily intraperitoneal (i.p.) injections and daily administration by oral gavage. For the mice receiving i.p injections, average Mmax amplitudes recorded in TA and LG from males and females were compared using a two-way (sex and treatment) analysis of variance (ANOVA). In these analyses, the effect of treatment was significant in both TA (F1,20=6.630, p<0.0203) and LG (F1,20=12.99, p=0.0018), but neither for sex (TA: F1,20=0.186, p=0.671; LG: F1,20=3.639, p=0.071), nor for the interaction of sex and treatment (TA: F1,20=0.058, p=0.812; LG: F1,20=3.114, p=0.0929). Data from males and females were combined for further analysis. We assumed that a similar lack of sex difference was present in the mice treated orally with CP11. We then evaluated the significance of differences between the two groups of CP11-treated mice and vehicle-treated mice for each muscle using a one-way ANOVA and post hoc paired tests, as appropriate. Results of the ANOVA were significant for both LG (F3, 37 = 3.98, p=0.0149) and TA (F3, 37 = 4.567, p=0.0081). For both muscles, both i.p. and oral administration of CP11 resulted in a significant increase in Mmax amplitude relative to identically administered vehicle treatments (Fig. 2C). Differences in Mmax amplitude between animals treated with CP11 via different routes of administration were not significant for both muscles. Four weeks after nerve injury, restoration of neuromuscular activity was improved more than twofold in mice treated with CP11, either orally or via i.p. injection, relative to vehicle-treated controls.
Axons of more motoneurons regenerate successfully after CP11 treatments – Following injection of the retrograde tracers WGA 555 into GAST and WGA 488 into TA, motoneurons were identified in horizontal sections of the lumbar spinal cord containing these fluorescent markers (Fig. 3A), indicating that their motor axons had regenerated and successfully reinnervated those muscles. A small number of motoneurons contained both retrograde tracers (Fig. 3A: yellow arrows), suggesting that their regenerating axons may have branched and were exposed to both tracers. As above, two different routes of administration of CP11 were used in different groups of mice. In the mice receiving CP11 or vehicle via i.p injections, we conducted a two-way (sex and treatment) ANOVA to evaluate if a sex difference existed. Results for both GAST and TA indicated a significant difference among treatment groups with regard to number of labeled motoneurons (TA: F1,14=10.7, p<0.0075; GAST:F1,14=136.00, p<0.0001), but not for sex (TA: F1,14=0.059, p=0.813; GAST: F1,14=0.045, p=0.835), or the interaction of sex and treatment (TA: F1,14=4.766, p=0.0516; GAST: F1,14=0.0018, p=0.967). We then combined the data from males and females and evaluated the significance of differences in numbers of retrogradely labeled motoneurons reinnervating TA, GAST or containing both tracers between CP11-treated (both oral and i.p injected) and vehicle-treated mice using a one-way ANOVA. Results were significant for GAST (F3, 26 = 35.45, p<0.0001) and TA (F3, 26 = 4.724, p=0.0140), but not for Both (F3, 24 = 0.6285, p=0.6037). Usingpost hoc paired testing, the number of retrogradely labeled motoneurons encountered was significantly greater in mice treated with CP11 than in mice treated with vehicle for both GAST and TA. The number of labeled motoneurons in mice treated with CP11 via intraperitoneal injection was not significantly different from the number in mice treated via oral gavage (Fig. 3B). Both oral and injectable treatments with CP11 nearly doubled the number of motoneurons whose axons regenerated and successfully reinnervated the TA and GAST muscles while not changing the number of motoneurons whose axons were exposed to both retrograde tracers.
CP11 treatments enhance muscle sensory axon regeneration – The same protocol was used to evaluate the effects of CP11 treatments on motor axon regeneration and study the regeneration of sensory axons reinnervating the TA and GAST muscles (Fig. 4A). The numbers of retrogradely labeled neurons in L4 DRGs were compared in mice four weeks after transection and repair of the sciatic nerve and two weeks of treatment with either CP11 or vehicle. Mean (+ SEM) numbers of cells labeled from tracer injections into the different muscles and with different treatments are shown for mice treated with CP11 via intraperitoneal injection and from mice treated orally (Figure 4B). For each data set, a one-way ANOVA was conducted on these data and proved significant for TA and GAST, but not Both (TA: F3,20= 7.588, p<0.001; GAST: F3, 20 = 11.89, p=0.0001; Both: F3, 20 = 1.445, p<0.2596). Applying post hoc paired testing, we found significantly more sensory neurons whose axons had regenerated and reinnervated the TA and GAST muscles in mice treated with CP11 than in mice treated with vehicle. Among the relatively small number of sensory neurons containing both retrograde labels, no significant difference was found between the vehicle-treated and CP11-treated animals.
CP11 treatments promote neurite elongation in cultured DRG cells – We cultured adult DRG neurons for 24 hours and then exposed the cells to different treatments for another 24 hours. To evaluate the effects of these treatments on neurite outgrowth, we measured the lengths of the longest neurites in each of at least 50 neurons per culture, identified by immunoreactivity to β3-tubulin, in cultures from six WT mice. In each well of a four-well plate, half of the initial plating media was removed and replaced with fresh media containing 7,8-DHF (500 nM), CP11 (5 uM), both 7,8-DHF and CP11, or no drug, termed Media. In the Media group, 0.02% DMSO was added, as this was the final concentration of DMSO in media containing diluted drug stocks. An example of a DRG neuron and the neurite measured from it is shown in Figure 5A. Cumulative frequency distributions of neurite lengths were constructed for each run in each of the four treatment groups and the data were fitted with a non-linear (Gaussian percentages) function. Fitted distributions for the six runs (cultures from different mice) were then averaged for each treatment group. These average frequency distributions are shown for the four groups by the solid lines in Figure 5B. The curves fitted to the frequency distributions for the cultures treated with CP11, 7,8-DHF, and the combined treatment are clearly shifted to the right of controls (Media), indicating that neurite lengths were increased by these treatments. Average median neurite lengths in each of these groups (Fig. 5C) were significantly longer than controls (ANOVA F3, 20 = 28.04, p<0.0001). Using post hoc (Tukey) paired testing, average median neurite lengths were significantly longer in the cultures treated with 7,8-DHF, CP11, or 7,8-DHF and CP11 (Both) than those in the control cultures, but no significant difference in median length was found between neurites in the three groups of treated cultures. Treatments with either 7,8-DHF or CP11 enhance neurite length by the same amount. Combining the two treatments did not increase this enhancement.
Enhancement of neurite outgrowth produced by CP11 treatment is TrkB-independent An additional set of cultures from WT mice was treated with CP11 or 7,8-DHF, as above, but some of the treated cultures also were exposed to the specific TrkB inhibitor, ANA-12. The distributions of neurite lengths measured in these cultures are shown in Figure 6A. Because the standard deviations of the groups compared were found to be significantly different, we performed a Brown-Forsythe ANOVA (F5, 302.5 = 117.18, p<0.0001) and Dunnett’s T3 post hoc paired testing to evaluate the significance of differences in neurite lengths in the different culture conditions. Neurite lengths were increased significantly by exposure to CP11, relative to Media controls, whether or not the treatment also included ANA-12. As anticipated, a significant increase in neurite elongation produced by 7,8-DHF treatment alone was blocked completely by administration of ANA-12. Treatment of cultures with ANA-12 alone produced no significant effect on neurite length. Enhancement of neurite elongation produced by CP11 is thus TrkB-independent.
We also investigated whether the effects of CP11 treatments on neurite elongation were effective in cultured DRG neurons expressing or lacking the TrkB receptor. Cultures from wild type mice were treated as above and neurons were identified by immunoreactivity to ß3-tubulin to mark neurites. Cells were also reacted with an antibody to the extracellular domain of the TrkB receptor. For each neurite measured, the cell of origin was scored as TrkB+ or TrkB-. Examples of such cells are shown in Figure 6B. Five sets of cultures were studied. Neurite lengths measured in each culture were scaled to the average median neurite length in the untreated control cultures (Media). Results of these scaled neurite lengths from TrkB+ and TrkB- neurons in control, CP11-treated and 7,8-DHF-treated cultures are shown in Figure 6C as cumulative histograms. Average (+ SEM) median scaled neurite lengths are shown in Figure 6D. Significance of differences between groups was evaluated using ANOVA and post hoc paired testing. Treatments with CP11 or 7,8-DHF were equally effective in promoting outgrowth of neurites from TrkB+ DRG neurons (Fig. 6C: Top), but only treatment with CP11 resulted in longer neurites in TrkB- neurons (Fig. 6C: Bottom).
Neurite outgrowth is enhanced in AEP knockout mice Finally, we compared the elongation of neurites from neurons cultured from wild type (WT) mice to similar treatments in cultures from AEP knockout (AEP KO) mice (Shirahama-Noda et al. , 2003). All of the data from DRG neurons lacking AEP were compared to results of similar measurements from treated and untreated cultures from wild type mice. For analysis, all neurite lengths were scaled to the average median length of neurites measured in untreated cultures derived from WT mice. Results of a one-way ANOVA of these scaled neurite lengths were significant (F5, 27 = 4.663, p=0.0034). Using post hoc paired testing, neurite lengths in untreated DRG cells from AEP KO mice were significantly longer than those in cultures from untreated WT mice (Fig. 7). Lengths of neurites from neurons from AEP KO mice that had been treated with CP11 or 7,8-DHF also were significantly longer than those in the untreated cultures from wild type mice, but not significantly longer than neurites from untreated cells from AEP KO mice (Fig. 7). In the absence of AEP, neurite elongation was increased by about 50% over that observed in cells from WT mice without any treatment. This enhancement was not further increased by addition of the AEP inhibitor, CP11, or by treatment with 7,8-DHF. In addition, the lengths of neurites in all groups of DRG neurons from AEP KO mice were not significantly different from those of neurons derived from WT mice that had been exposed to either 7,8-DHF or CP11. Thus, inhibition of AEP by CP11 treatment resulted in an effect on neurite outgrowth similar to knocking out the AEP gene.