Animal experiments were conducted in accordance with the European Communities Council Directive of 24th Nov. 1996 and were approved by the local governmental administrations for animal health (Animal Care Use Committee of the University of California, San Diego, CA, United States and Government of Lower Frankonia, Würzburg, Germany, “TS–9/11”). A53T-tg mice overexpress human α-syn carrying the A53T mutation under the regulatory control of the PDGFβ promoter (Hashimoto et al., 2003). A53T-tg mice were kept group-housed with non transgenic littermates (non-tg) under a 12 h light dark cycle with free access to food and water. Newborn neurons were labeled with the respective retrovirus at 4 months of age (n = 4 animals per group) when survival and proliferation of newborn neurons is impaired (Kohl et al., 2012). For the analysis of the effects of cell-intrinsic overexpression of A53T-mutant α-syn, newborn neurons were labeled in C57Bl/6 mice at the young adult age of 6 weeks (n = 4 animals per group).

A Moloney murine leukemia retrovirus-based CAG-GFP plasmid was used as described earlier (Zhao et al., 2006; Winner et al., 2012; Regensburger et al., 2018). To selectively overexpress A53T-mutant α-syn within newborn neurons, the retrovirus CAG-A53T-GFP was cloned where A53T-mutant α-syn is C-terminally fused with GFP. To this end, GFP cDNA was ligated to the 3′ end of A53T-mutant α-syn cDNA in a Bluescript cloning vector (Invitrogen) and subsequently cloned into the CAG-GFP vector. Correct structure of the construct was confirmed by sequencing. A concentrated viral solution (10⁸ pfu/ml) was prepared with human embryonic kidney 293T packaging cells (Tashiro et al., 2006). Mice were anesthetized using a weight-adjusted i.p. dose of Xylazine/Ketamine. A stereotaxic frame (Kopf Instruments) was used for sequential bilateral infusion into the dentate gyrus (AP −2.00 mm, ML +/− 1.6 mm from bregma, DV −2.3 mm from skull). A total volume of 1.5 μl was slowly infused (0.3μl/min) followed by wound closure and a survival period of 31 days.

Animals were sacrificed 31 days after stereotactic surgery. Euthanasia with Xylazine/Ketamine i.p. was followed by transcardial perfusion with PBS and 4% paraformaldehyde for tissue fixation. Brains were dissected, postfixed for 6 h in 4% paraformaldehyde and stored in 30% sucrose in 0.1 M phosphate buffer at 4°C. 40 μm coronal brain sections were stored in cryoprotectant solution (25% ethylene glycol, 25% glycerol in 0.1 M phosphate buffer) at −20°C.

The following primary antibodies were used: rat-anti-α-synuclein (15G7, Enzo Life Science, Germany, 1:50), ch-anti-GFP (Abcam, 1:500), rb-anti-SERT (ImmunoStar, Hudson, WI, United States; 1:2,000), gt-anti-DCX (C18, Santa Cruz, TX, United States). Secondary antibodies were donkey-derived and conjugated with Alexa-488 (1:1,000; Life Technologies, Carlsbad, CA, United States), Alexa-647, or Rhodamine Red-X (1:1,000; Dianova). Immunofluorescence was conducted as described previously (Winner et al., 2012; Deusser et al., 2015). Sections were blocked in 3% donkey serum/0.1% TritonX100 in TBS, and incubated with primary antibodies at 4°C, with secondary antibodies at room temperature, and washed again in TBS. Nuclei were counterstained with DAPI (Thermo Fisher Scientific, final concentration 1:2,000) and mounted on object glasses (Superfrost Slides, Menzel).

Microscopy and dendrite analyses were performed as reported previously (Regensburger et al., 2018). Recordings were performed on a fluorescence microscope (Observer.Z1, Zeiss, for dendrite growth analysis) and on a confocal laser scanning microscope (LSM710, Zeiss for spine analysis and coexpression studies) using ZEN software. For dendrite growth analyses, on average 6 GFP-positive newborn neurons in the dentate gyrus of each animal were imaged resulting in a cell number of 24 per group. For each neuron, z-series of antibody-enhanced GFP-signal at 1.5 μm were acquired spanning the whole extent of the neuron within the section. Maximum intensity projections were then analyzed with the ImageJ plugin “Simple Neurite Tracer.” Primary, secondary, tertiary and quaternary dendrites were labeled semiautomatically. As readouts, the plugin determined total dendritic length, number of branching points and number of each suborder of dendrites. On a thresholded image of the rendered path of each neuron, dendrite complexity was determined using the “Sholl analysis” plugin, with a step radius of 12.5 μm from soma and a maximum radius of 250 μm.

Spine recordings were performed on dendritic sections localized within the molecular layer, using unstained mounted sections and the 60x object lens. Six newborn neurons were analyzed per group. The field of view was placed on the molecular layer and all dendritic segments of one positive neuron localized within this area were imaged. Upon morphological appearance, spines were classified into thin/stubby spines and the small subpopulation of mushroom spines. The estimated surface area of each spine was calculated as 0.785 × D_major × D_minor with D_major as the biggest diameter and D_minor as the smallest diameter of the respective spine. Mushroom spines exhibit a surface area of at least 0.4 μm² which was measured for each (suspected) mushroom spine (Zhao et al., 2014).

To determine the density of SERT-positive fibers, stacked images of the molecular layer were scanned, spanning a z-axis of 11 μm at 1 μm distance. On maximum intensity projection images, SERT-positive fibers were manually traced within the molecular layer using the multipoint line tool in ImageJ. Density was calculated by division of the total SERT length by analyzed area and 11 μm.

All data are shown as mean ± standard deviation except for Sholl analyses (mean ± SEM). Statistical analyses were performed using Graph Pad Prism (GraphPad Software, La Jolla, CA, United States). Statistical significance was indicated by ^∗P < 0.05, ^∗∗P < 0.01, and ^∗∗∗P < 0.001, as determined by unpaired, two-sided t-tests.

It was previously shown that at the age of 4 months, proliferation and survival of newborn neurons are impaired in A53T-tg mice (Kohl et al., 2012). Thus, we asked whether the morphology of newborn neurons within the A53T-tg α-syn overexpressing microenvironment was changed. To this end, newborn neurons were labeled by stereotactic injection of a GFP overexpressing retrovirus into the dentate gyrus of 4 months old mice (Figures 1A,B). After a survival period of 1 month, newborn neurons were visualized by fluorescence microscopy for GFP (Zhao et al., 2006). Using an antibody specific for human α-syn, we found that transgenic A53T α-syn was widely expressed in the dentate gyrus, including GFP positive newborn neurons (Figure 1A).

We observed no significant differences in the total dendritic length of newborn neurons, comparing A53T-tg and non-tg (Figures 1C,E and Table 1). Similarly, there were also no differences in dendritic complexity, as measured by the number of branching points (Figure 1F) and the subquantification of the numbers of primary, secondary, tertiary and quaternary dendrites (Figure 1G). Sholl analysis showed a significant reduction of dendrite complexity at 25.0 μm distance from soma, but was otherwise unchanged (Figure 1H).

As an indicator of postsynaptic integration of newborn neurons, we next visualized dendritic spines of GFP labeled newborn neurons in the molecular layer of A53T-tg and non-tg (Figure 1D). Upon quantification, there was a significant, 24% decrease of the density of spines per dendrite length in A53T-tg as compared to non-tg (Figure 1I), indicating an integration deficit of adult newborn neurons in A53T-tg. Since there are only few filopodia and stubby spines on adult newborn granule cells, dendritic spines can be subdivided into thin spines and a smaller proportion of mushroom spines, based upon their morphology (Zhao et al., 2006). Indeed, also the density of mushroom spines was significantly reduced by 48% in A53T-tg (Figure 1J).

In summary, adult newborn neurons were unchanged in A53T-tg regarding dendrite outgrowth, but postsynaptic integration was impaired as marked by decreased densities of overall spine density and mushroom spine density.

Serotonergic degeneration is a widespread feature in adult A53T-tg mice, including specific layers of the prefrontal cortex and the hilus of the dentate gyrus (Deusser et al., 2015; Wihan et al., 2019). In light of the reduced spine density of newborn neurons, we investigated alterations of presynaptic serotonergic fibers within the molecular layer of adult A53T-tg mice. In accordance with previously published data in a C57Bl/6 based reporter model (Migliarini et al., 2013), there was a dense innervation of the molecular layer both in non-tg mice and in A53T-tg mice (Figure 2A). Upon quantification of the density of serotonergic fibers per volume, no significant differences between A53T-tg and non-tg were present (Figure 2B). This suggests that there is no degeneration of serotonergic fibers within the molecular layer of A53T-tg at this age.

In order to dissect the role of cell-intrinsic presence of tg A53T α-syn vs. an effect of the transgene within the microenvironment, we next overexpressed A53T α-syn specifically within newly generated neurons. To this end, human A53T α-syn cDNA was cloned into the CAG-GFP retroviral construct. Stereotactic injections of either CAG-GFP or CAG-A53T-GFP were performed in adult wildtype mice, targeting the dentate gyrus of the hippocampus to label newly generated cells (Figure 3A). Perfusion was performed after 31 days. Using a human specific α-syn antibody, we observed that the A53T α-syn GFP fusion protein was localized within the somal, axonal and dendritic compartments of newborn neurons (Figure 3B).

Following the strategy described above for the analysis of A53T-tg mice, we first investigated dendritic morphology (Figure 3C). Cell-autonomous overexpression of A53T-GFP caused no significant differences in dendrite outgrowth, the number of branching points and the number of primary, secondary, and tertiary dendrites when compared to GFP only (Figures 3E–G). Sholl analysis showed that dendrite complexity was unchanged between groups (Figure 3H). This indicates that neurite outgrowth is not impaired by cell-intrinsic presence of A53T α-syn. However, analyzing spine density in CAG-A53T-GFP, we found a significant reduction by 31% as compared to CAG-GFP (Figures 3D,I) which was comparable to the A53T-tg model. Interestingly, mushroom spine density was not significantly reduced upon cell-autonomous A53T-GFP overexpression, although there was a trend (Figure 3J). To sum up, retrovirus based cell-autonomous overexpression of A53T α-syn within the dentate gyrus resulted in decreased overall spine density, but mushroom spine density was not significantly altered.