The interaction between tropomyosin-related kinase B receptors and presynaptic muscarinic receptors modulates transmitter release in adult rodent motor nerve terminals

Abstract

The neurotrophin brain-derived neurotrophic factor (BDNF), neurotrophin-4 (NT-4) and the receptors tropomyosin-related kinase B (trkB) and p75(NTR) are present in the nerve terminals on the neuromuscular junctions (NMJs) of the levator auris longus muscle of the adult mouse. Exogenously added BDNF or NT-4 increased evoked ACh release after 3 h. This presynaptic effect (the size of the spontaneous potentials is not affected) is specific because it is not produced by neurotrophin-3 (NT-3) and is prevented by preincubation with trkB-IgG chimera or by pharmacological block of trkB [K-252a (C₂₇H₂₁N₃O₅)] or p75(NTR) [Pep5 (C₈₆H₁₁₁N₂₅O₁₉S₂] signaling. The effect of BDNF depends on the M₁ and M₂ muscarinic acetylcholine autoreceptors (mAChRs) because it is prevented by atropine, pirenzepine and methoctramine. We found that K-252a incubation reduces ACh release (~50%) in a short time (1 h), but the p75(NTR) signaling inhibitor Pep5 does not have this effect. The specificity of the K-252a blocking effect on trkB was confirmed with the anti-trkB antibody 47/trkB, which reduces evoked ACh release, like K-252a, whereas the nonpermeant tyrosine kinase blocker K-252b does not. Neither does incubation with the fusion protein trkB-IgG (to chelate endogenous BDNF/NT-4), anti-BDNF or anti-NT-4 change ACh release. Thus, the trkB receptor normally seems to be coupled to ACh release when there is no short-term local effect of neurotrophins at the NMJ. The normal function of the mAChR mechanism is a permissive prerequisite for the trkB pathway to couple to ACh release. Reciprocally, the normal function of trkB modulates M₁- and M₂-subtype muscarinic pathways.

Figure 1.

Immunolocalization of BDNF, NT-4, trkB, and p75^NTR at the neuromuscular synapses. A–D, Triple labeling of the neurotrophins BDNF (A) and NT-4 (B) and their receptors trkB (C) and p75^NTR (D) (green fluorescence) with syntaxin (blue fluorescence) and AChR-α bungarotoxin (red fluorescence). We used plastic-embedded semithin (0.5 to 0.7 μm) cross-sections (STS) as a tool for high-resolution, triple-labeling immunofluorescence analysis. E, Example of triple labeling of syntaxin (green fluorescence), S-100 (blue fluorescence), and nAChR-α bungarotoxin (red fluorescence), which proves that the molecular markers of the three synaptic cells are separated well. F, High-resolution images of synaptic boutons with the terminal axon labeled for one of the proteins in green in the nerve terminal area between the blue Schwann cell and the red nAChR cluster. Scale bars: A–E, 10 μm; F, 1 μm.

Figure 2.

Effect of exogenous BDNF and NT-4 on neuromuscular transmission: dose–response and time course. A, Dose–response study showing that incubation with BDNF or NT-4 for 1 h does not change EPP size. B, The time course study showing that after 3 h of incubation, both BDNF and NT-4 significantly increased EPP amplitude. Preincubation (1 h) with trkB-IgG completely prevents the potentiating effect (at 3 h) of the neurotrophins. Five muscles and a minimum of 15 fibers per muscle are included for each point. Values are expressed as mean ± SEM. *p < 0.05 with respect to initial values.

Figure 3.

BDNF and NT-4 on neuromuscular transmission. A, After trkB block (K-252a) or p75^NTR block (PEP 5), incubation with 10 nm BDNF or 2 nm NT-4 does not change EPP size after 3 h. B, Spontaneous release (MEPPs) at 1 and 3 h of incubation. These concentrations of BDNF and NT-4 do not affect the muscle membrane potential or the size of the MEPPs, although the MEPP frequency does increase after 3 h in the presence of BDNF. C, Representative raw data traces show how BDNF affects MEPP frequency (the top and bottom traces were recorded from different fibers). For each column (A, B), five muscles and a minimum of 15 fibers per muscle are included. Values are expressed as mean ± SEM. *p < 0.05 with respect to initial values.

Figure 4.

Endogenous BDNF and NT-4. A, B, The dose–response study shows that the trkB blocker K-252a (from 200 nm) reduced EPP size at 1 h of incubation (Bii, raw data; superimposed traces of EPPs) and increased MEPP frequency (Bi, representative raw data). However, the p75^NTR receptor blocker Pep5 (in the range 50 nm to 1 μm) does not significantly affect evoked (A) and spontaneous release parameters (data not show). The specificity of the K-252a blocking effect was confirmed by the incubation of the ex vivo preparation (1 h) with an anti-trkB antibody (47/trkB) that reduces evoked ACh release, as K-252a does (A, only one point at 100 nm; Biv, raw data), and similarly increases MEPP frequency (Biii, raw data). The nonpermeant tyrosine kinase blocker K-252b (500 μm) has no effect on transmitter release in our conditions (A). C, Incubation with the fusion protein trkB-IgG to chelate endogenous BDNF/NT-4 or with anti-BDNF (20 μg/ml) or anti-NT-4 (10 μg/ml) does not change ACh release. For each point (A) and column (C), five muscles and a minimum of 15 fibers per muscle are included. Values are expressed as mean ± SEM. *p < 0.05 with respect to initial values.

Figure 5.

trkB and mAChRs on ACh release. A, In the presence of several blockers of presynaptic mAChR (1 h of preincubation), BDNF does not have a potentiating action. B, Effect (at 1 h of preincubation) of K-252a and the muscarinic blockers on ACh release (white columns) in a normal situation. The M₁ antagonist pirenzepine (10 μm) reduced evoked neurotransmission, whereas the M₂ antagonist methoctramine (1 μm) increased the evoked release. The unselective blockade of both M₁ and M₂ with atropine considerably increased release. When K-252a was present, pirenzepine did not affect release, whereas methoctramine and atropine reduced it. The figure also shows the result of the reciprocal experiments (the first 1 h incubation with an mAChR blocker followed by a second incubation with K-252a to block trkB). For each column in A and B, five muscles and a minimum of 15 fibers per muscle are included. Values are expressed as mean ± SEM. *p < 0.05 with respect to initial values. C, Raw data for the effect of pirenzepine and methoctramine acting after K-252a preincubation.

Figure 6.

trkB expression and phosphorylation. A, Western blotting to see the expression and tyrosine phosphorylation of the trkB protein in different conditions. The quantitation by densitometry in B shows that in control (CTR) muscles trkB is expressed and phosphorylated and that preincubation with BDNF (10 nm; 3 h) increases phosphorylation somewhat compared with the control. The M₂ block with methoctramine (MET; 1 μm), the M₁ block with pirenzepine (PIR; 10 μm), or the full mAChR block with atropine (ATR; 2 μm) strongly increases the expression and the level of tyrosine phosphorylation of the trkB. C, D, Western blotting (C) and quantitation (D) to show that the expression of the BDNF and NT-4 proteins in the muscle does not change after preincubation with the muscarinic blockers. “CTR−” means the negative control incubated without primary antibody. Samples are 50 μg of protein. Values are shown as mean ± SD. *p < 0.05 with respect to the control.

Figure 7.

Involvement of mAChRs and trkB in neurotransmitter release. The diagram shows the main interactions described in this study. The green arrows indicate that the normal coupling of a molecule (trkB, M₁, or M₂) to ACh release allows normal coupling to the release of the molecule (or the pathway) on the arrowhead. The red arrow indicates inhibitory action. When the trkB receptor is blocked, the M₁ release-potentiating pathway (point 1) does not operate because it lacks a hypothesized link (point 3) that enables the M₁ mechanism. The M₂ inhibitory mechanism (point 4) reverses its coupling to transmitter release because it lacks a hypothesized link (point 6) that enables this mechanism. In this case, an M₂-mediated potentiation (point 8) may be unmasked. A similar reversion in the M₂-mediated function occurs in several conditions that share low ACh release. The low release produced by interference with the trkB receptor (point 7) may be the cause of the functional anomaly in the muscarinic pathway. The reciprocal experiments show that total or partial block of the muscarinic mechanism completely prevents the trkB from coupling to ACh release (points 2 and 5 represent the mAChR enabling function on the trkB receptor). In the absence of the muscarinic function, the trkB activation may be coupled to mechanisms (point 9) other than immediate transmitter release.