All the experimental procedures were performed following the European Community guidelines (63/2010), the Spanish Royal Decree 53/2013 and the Order ECC/566/2015. Protocols were previously approved by the Bioethics Committee of the University of Salamanca. Mice, with a maximum of 5 animals per cage, were housed in SPF Animal Facility of the University of Salamanca. They were fed ad libitum in a 12 h-light/dark cycle, kept at constant temperature of 20–22°C and a relative humidity of 55–65%. Only male mice were used throughout the study. The TrkA-P782S (KI) mice were previously described (Yu et al., 2011, 2014) and wild type littermates were used as controls. TrkA-Cre mice (B6; 129S4-Ntrk1^tm1(cre)Lfr/Mmucd, MMRRC #015500-UCD) were crossed with RCL-tdT mice (B6.Cg-Gt (ROSA)26Sor^{tm9(CAG-tdTomato)Hze}/J, JAX007909) (Madisen et al., 2010) to generate TrkA-Cre; RCL-tdT mice. In these mice, we identified neurons that have expressed or actually express TrkA.

Mice were transcardially perfused with cold 0.1 M phosphate buffer (PB) pH 7.4, followed by 4% paraformaldehyde (PFA) in 0.1 M PB. Brains were extracted, post-fixed overnight in 4% PFA in 0.1 M PB at 4°C and then incubated in PBS with 30% (w/v) sucrose at 4°C until they sank. Each brain was coronally sectioned at 40 μm and floating sections were stored in cold 0.1 M PB containing 0.02% NaN₃.

For 3,3′-diaminobenzidine-based immunostaining, sections were incubated in 0.3% hydrogen peroxide for 10 min at room temperature (RT) to reduce activity of endogenous peroxidase, then washed with PBS with 0.2% Triton X-100 and blocked in PBS with 5% normal goat serum (NGS), 1% bovine serum albumin (BSA) and 0.2% Triton X-100 for at least 2 h at RT. Antibody incubations were performed in PBS containing 2% NGS, 1% BSA and 0.2% Triton X-100 at 4°C with gentle agitation for 48–72 h. The following antibodies were used: rabbit ChAT (1,500, Ab143, Millipore), rabbit TrkA (1:500; gift from L.F.Reichardt, University of California, San Francisco, CA; Clary et al., 1994), rabbit p75^NTR (9992; 1:500; gift from M.V.Chao, Skirball Institute, NYU Medical Center, NY; Benedetti et al., 1993). Biotin-conjugated secondary antibodies (1:500, Cat. 111-065-003, The Jackson Laboratory) were incubated at RT for 1 h 30 min. The staining was developed using Vectastain ABC system (Vector Laboratories) and sections were mounted on gelatin-coated slides, dehydrated and cover-slipped with Entellan (Merck).

For immunofluorescences (IF), floating sections were incubated with primary antibodies as described above overnight at 4°C and Alexa-labeled secondary antibodies (1:750, Cat. A11008 and A 21424, Invitrogen) were utilized for 1 h 30 min at RT, counterstained with Hoechst 33342 (1:13000) and mounted on gelatin-coated slides with Vectashield (Vector Laboratories). The following antibodies were used for IF: rabbit TrkA (1:500; gift from L.F.Reichardt, University of California, San Francisco, CA; Clary et al., 1994) and Nedd4-2 (15 μg/mL; Arevalo et al., 2006). Images were collected using an Olympus AX70 microscope equipped with an Olympus DP70 camera for inmunohistochemistry stainings or with a Leica STELLARIS DMI8 confocal microscope for inmunofluorescences.

To quantify cholinergic neurons in MS and DB, coronal brain sections from WT and TrkAP782S mice spanning from bregma 0.20 to 1.50 mm were immunostained for ChAT, TrkA or p75^NTR. Positive cells for each staining were counted in the whole MS and vertical and horizontal limb of the DB. These regions were defined according to the mouse brain atlas (Paxinos and Franklin, 2001). A section every 280 μm was counted, for a total of 3–4 sections throughout the rostrocaudal extent of the MS and the DB. The cell counting was carried out using ImageJ software by two independent investigators blind to the genotyping and the results presented are the average of both researchers.

The density of cholinergic nerve terminals in the stratum oriens of the CA1 region of the hippocampus was evaluated from sections stained for p75^NTR spanning approximately from bregma-1,2 to-2.5 mm. Nerve terminal density was obtained using the software Photoshop measuring the integrated density of p75^NTR staining above threshold.

Mice were killed by cervical dislocation and brains were quickly dissected. Specific areas of the brain were snap-frozen in liquid nitrogen and homogenized in lysis buffer containing 10 mM Tris pH 7.4, 150 mM NaCl, 2 mM EDTA, 1% NP-40, 0.1% SDS, 1 mM PMSF, 1 μg/mL aprotinin, 2 μg/mL leupeptine, 1 mM vanadate, 10 mM NaF, and 20 mM β-glycerophosphate at 4°C and passed through a 26G needle to dissociate. Lysates were centrifuged at 14,000 rpm for 15 min at 4°C to eliminate the debris.

MS from P15 WT and KI littermates were collected in 0.9% sodium chloride (Vitulia, Cat. #999791.5). After 20 min in sodium chloride solution at RT, we performed NGF treatment with mouse NGF 2.5S (Cat. N-100; Alomone labs) at 25 ng/mL or 100 ng/mL for 15 min at RT with gentle agitation. Finally, NGF was removed and tissue was properly washed. Protein lysates preparation was performed as described above and WB were performed.

Protein concentration was determined using the Bradford protein assay (Bio-Rad). Laemmli buffer was added to each sample, boiled for 5 min to denature proteins that were resolved by SDS-PAGE and immunoblotted with specific antibodies. The following antibodies were used: rabbit p42/44 MAPK (ERK1/2) (1:1000, #9102, Cell Signaling Technology), mouse phospho-p42/44 MAPK (ERK1/2) (1:1000, #9106, Cell Signaling Technology), ChAT (1,1,000, Ab144, Millipore), β-actin and β-tubulin III (1:5000 and 1:10000, #A-4700, #T2200 respectively, Sigma-Aldrich), TrkA (RTA; 1:500; gift from L. Reichardt USCF, San Francisco, CA; Clary et al., 1994), rabbit Nedd4-2 (1: 3,000; Arevalo et al., 2006) and rabbit p75^NTR (9992; 1:1000; gift from M.V.Chao, Skirball Institute, NYU Medical Center, NY; Benedetti et al., 1993). Horseradish peroxidase-conjugated secondary antibodies were used at a dilution 1:10000. Blots were then processed by the ECL chemiluminescence method and the intensity bands were quantified as reported (Cañada-García and Arévalo, 2023).

All tests were conducted during the light cycle by an experimenter blind to the genotype. Mice were habituated to handling for at least 2 days before behavioral tests began. Mice were given 1 h to habituate to the behavioral room before any test was conducted. Behaviors were videotaped and analyzed using ANY-Maze software.

The open field test was performed in an arena made of plexiglas (50 × 40 cm) surrounded by walls (30 cm high). Animals were placed next to a corner and allowed to freely explore for 10 min. The center area of the chamber was designated as 30 × 20 cm. The total distance covered and the time spent at the center was quantified.

The elevated plus maze test was performed in a dark plexiglas elevated maze with four arms (30 cm long and 5 cm wide) extending from a central platform (5 × 5 cm). While two facing arms were closed by dark walls (15 cm high), the other two arms were open. Animals were placed in the center of the apparatus and allowed to explore it for 10 min. The time spent in open and closed arms was analyzed.

Mice were allowed to explore the open field arena described above for 10 min. Next day, during the sample phase, mice were exposed to a pair of identical objects placed in the back left and right corners of the arena for 5 min. Object recognition memory (ORM) was tested 1 h after this trial by exposure to a previously familiar object and to a new one for 5 min. Familiar and novel objects were counterbalanced. The snouts of the mice were tracked, and object interactions were measured as time spent with snout within 2 cm of the object. The discrimination ratio was b/(a + b) where a is the exploration time of the familiar object and b is the exploration time of the novel object. ORM was measured as the increased time spent investigating the novel object.

Motor coordination task was performed using an accelerating rotarod. The test assesses fore-and hind-limb balance coordination as well as a learning component for this motor task (Deacon, 2013). The rotation rate of the cylinder increased from 4 to 40 rpm over a 5 min period. The latency of each mouse to fall was automatically recorded by the device. The test was performed for four trials a day for 4 consecutive days.

For neuronal counts and biochemical analysis, experiments were performed on a minimum of three animals for each genotype and age. Data are presented as mean ± SEM, unless otherwise mentioned in the figure legend. Statistical significance was analyzed using GraphPad Prism 6 (GraphPad software). Comparisons between means were assessed by different statistical test described at the figure legend after having checked normality of data. For those data with small numbers or non-normal distribution, comparisons were done with a non-parametric test.

Upon NGF binding, TrkA is ubiquitinated by different E3 ubiquitin ligases (Geetha et al., 2005; Makkerh et al., 2005; Arevalo et al., 2006). One of those E3 ubiquitin ligases is Nedd4-2. Nedd4-2 binds to the TrkA PPXY motif and modification of this motif prevents Nedd4-2 binding and renders a gain-of-function TrkA receptor as in the case of TrkAP782S (Arevalo et al., 2006). To study in vivo the NGF/TrkA signaling, we generated a mutant mouse (TrkAP782S, hereinafter TrkA KI) unable to bind Nedd4-2, yielding an overactivated TrkA receptor in the PNS (Yu et al., 2011, 2014). To assess whether increased NGF/TrkA signaling has any effect on CNS, we performed different experiments. But first, we checked Nedd4-2 expression in the basal forebrain, particularly in the MS region where TrkA is expressed, with several approaches: (1) using lysates from the basal forebrain of 2 months old mice, we detected Nedd4-2 expression as well as TrkA and p75^NTR in both WT and TrkA KI samples (Figure 1A); (2) we stained with Nedd4-2 antibodies brain slices from the TrkA-Cre;RCL-tdT reporter mice where TrkA-positive cells are labeled with the fluorescent protein Tomato (Figure 1B); and (3) we co-stained with TrkA and Nedd4-2 antibodies in brain slices from control and TrkA KI mice (Figure 1C). The results indicated that Nedd4-2 is co-expressed with TrkA-positive neurons in the MS. To address whether TrkAP782S receptor is more active than the WT in the MS, brain slices were stimulated with different amounts of NGF for 15 min and lysates were obtained. We observed that activation of ERK1/2, which is downstream of NGF/TrkA signaling, was increased in TrkA KI samples (Figure 1D). Therefore, TrkAP782S protein is more active than the WT in cholinergic neurons from the MS. Altogether these data indicate that Nedd4-2 modulates TrkA activation in MS.

To assess whether enhanced NGF/TrkA signaling has any effect on cholinergic neurons, we used different approaches using TrkA KI mice and WT littermates as controls. First, we assessed ChAT protein levels in lysates from MS at different ages (Figure 2A) and detected a significant increase exclusively at P15 in mutant mice (Figure 2B). To address whether this enhancement was due to a difference in the number of ChAT-positive neurons, we stained brain sections from WT and TrkA KI mice (Figure 2C). We found that in TrkA KI mice ChAT-positive neuron number was significantly increased at P15 (Figure 2D), which correlates with the increase amount of ChAT at this age. However, at P360 there is a reduced number of ChAT-positive neurons in TrkA KI mice that it is not maintained at P540. Since cholinergic neurons in the MS express TrkA (Sobreviela et al., 1994), we assessed TrkA protein levels and also performed immunohistochemistry using WT and TrkA KI mice from P15 up to P540. We did not observe differences at the protein level (Supplementary Figures S1A,B) nor at the number of TrkA-positive neurons (Supplementary Figures S1C,D). Therefore, TrkA-enhanced signaling increases ChAT protein levels and the number of cholinergic neurons during early postnatal development.

It is known that cholinergic neurons from the MS project into the dorsal hippocampus, into CA1 region (Amaral and Kurz, 1985; Yoshida and Oka, 1995; Müller and Remy, 2018), so we decided to test these projections. Knowing that p75^NTR is expressed in these cholinergic neurons of the MS (Sobreviela et al., 1994), we stained sections with an antibody against p75^NTR. First of all, we quantified the number of p75^NTR-positive neurons in the MS of TrkA WT vs. KI mice and we did not find differences between genotypes at any ages (Figures 3A,B). Then, we quantified the projections into the CA1 region of the hippocampus and found no differences between WT and KI mice (Figures 3C,D). Thus, a mutant TrkA leading to overactivation in cholinergic neurons does not alter the number of p75^NTR-positive neurons nor innervation of CA1 region.

To assess general locomotion, we performed the open field test in TrkA WT and KI mice. Mice from both genotypes covered similar total distance at P60, P180, and P360 (Figure 4A), with a slight decrease due to aging in both genotypes, and spent the same amount of time in the center of the arena (Figure 4B). We also carried out experiments using the elevated plus maze to test their anxiety-like behavior and we did not observe differences in the time spent in the open or closed arms between genotypes at the aforementioned ages (Figures 4C,D). Therefore, TrkAP782S mice do not show alterations in general locomotion nor anxiety.

To assess motor skill coordination, we performed rotarod accelerated task. TrkA WT and KI mice show similar locomotor coordination at P90, P180, and P360 (Figures 5A–C), but the mutant mice at P90 did not improve their motor coordination after 4 sessions of training (4 trials per session) as the TrkA WT did (Figure 5A). Therefore, TrkA KI mice show a transient deficit in locomotor learning at P90.

Cholinergic neurons from MS have been implicated in different behaviors such as attention and memory (Ananth et al., 2023). To assess the increase of NGF/TrkA signaling in MS in the working memory and the attention system, we performed the NOR test (Figure 6A). We measured the discrimination ratio between the novel and familiar object. At P60, TrkA WT and KI mice performed similarly, whereas at P180 the KI ones showed a significant deficit in recognizing the novel object (p = 0.005, KI vs. WT mice) and a tendency at P360 (p = 0.0736, KI vs. WT mice) (Figure 6B). In addition, the percentage of TrkA KI mice that discriminate the novel object vs. the familiar one above 50% was reduced at every age compared. At P60 (77% vs. 92%, KI vs. WT), at P180 (61% vs. 86%, KI vs. WT mice) and at P360 (41% vs. 64%, KI vs. WT mice) (Figure 6C). These deficits were not due to lack of exploration since KI mice explored as much as WT, both novel and familiar objects, in every experimental age (Figure 6D). Therefore, an enhanced NGF/TrkA signaling in cholinergic neurons leads to an impairment in memory and attention in mice.