Establishment of a lesion paradigm of the adult zebrafish statoacoustic ganglion (SAG). (A) Scheme of the zebrafish head with the brain inside showing the injection position in relation to the brain and the SAG. A 20-gauge needle was used to open the skull before inserting a Hamilton syringe vertically approximately 1.2- to 1.5-mm deep into the skull along the dorsal–ventral axis. Subsequently, the Hamilton syringe was either removed (sham-treated control) or 5–8 µl of either 0.9% NaCl or 100 µM ouabain solution was slowly injected. (B) Overview of live adult zebrafish before and after injection. Close-ups of the blue dotted boxes show the injection side within the triangular bone (dotted blue line). Lower close-up shows the hole after injection (dotted yellow circle). (C) Immunohistochemistry of a transverse cross section at the height of the anterior SAG following CellTracker Red injection into the right inner ear showed a strong labeling of the SAG on the injected side but no labeling of the unlesioned side (ULS). Scale bar: 200 µm.

Cell death in the adult zebrafish SAG upon lesion. (A) Antibody staining of the adult zebrafish SAG labeling ubiq:secAnnexinV-mVenus-positive cells and HuC/D-positive neurons. In contrast to unlesioned sides (ULS), a strong ubiq:secAnnexinV-mVenus-signal was present in the SAG following sham treatment or NaCl and Ouabain injections at 12 h post lesion (hpl). In the medial part of the SAG, which harbors the neurogenic niche and the neuronal cell bodies, only very few HuC/D-positive cells showed a weak ubiq:secAnnexinV-mVenus staining in all samples (arrows in b and f). In the distal part of the SAG, where the neurites of the sensory neurons innervate the sensory patch, a strong ubiq:secAnnexinV-mVenus staining was observed in sham-treated as well as NaCl- and ouabain-injected SAGs, but almost no signal was detected in unlesioned control sides (arrowheads in c, e, and g). Magenta dotted line and asterisk in ouabain-injected SAG indicates a massive hemorrhage in this injected site. (B) Colabeling of mVenus with calretinin revealed the presence of ubiq:secAnnexinV-mVenus in neurites (arrowhead in c) of sham-treated SAGs but not in unlesioned control sides. (C) Antibody staining labeling neurons (elavl3:GFP and calretinin) combined with TUNEL assay. TUNEL assay revealed presence of apoptotic neurons in the SAG of adult elavl3:GFP zebrafish in all three lesion types at 12 hpl, whereas no TUNEL-positive neurons were found in the unlesioned control side. Scale bars: (A) 200 μm; (B): 50 µm.

Accumulation of leukocytes in the adult zebrafish SAG upon lesion. Antibody staining against the pan-leukocyte marker L-plastin and GFP in the SAG of transgenic neurod:GFP zebrafish at 1, 2, 4, and 16 days post lesion (dpl). Shown are representative images of the SAG following sham-treatment, or NaCl or ouabain injections as well as the unlesioned side (ULS) of either sham-treated, or NaCl- or ouabain-injected animal. Boxes show close-ups of the respective neurod:GFP-positive neurogenic niche. In the unlesioned side (ULS, upper left panel in each time point), a few ramified L-plastin-positive cells are distributed evenly throughout the entire SAG. Some round, strong L-plastin-positive cells are localized in close contact to the neurogenic niche on the ventral side of the SAG (arrows). In contrast, accumulation of ramified L-plastin-positive cells (arrowheads) in the SAG is observed close to the neurogenic niche at all time points within sham-treated, or NaCl- or ouabain-injected SAGs (upper right and lower panels in each time point). In addition, the density of L-plastin-positive cells is usually the highest at the distal part of the SAG, especially when the sensory patch is massively destroyed as, for example, seen in 1 dpl/NaCl and ouabain, 2 dpl/ouabain, and 16 dpl/ouabain, irrespective of the lesion type (yellow asterisks). Scale bars: 200 µm.

Reactive proliferation of neurod:GFP-positive progenitors in the adult zebrafish SAG upon lesion. Immunohistochemistry against the proliferation marker proliferating cell nuclear antigen (PCNA) and GFP in the SAG of neurod:GFP transgenic animals. Shown are representative images of the neurogenic niche of the SAG following sham treatment, or NaCl or ouabain injection as well as the unlesioned side (ULS) of either sham-treated, or NaCl- or ouabain-injected animals at 2, 4, 8, 16, and 57 days post lesion (dpl). Boxes show close-ups to visualize costaining. Proliferating but marker-negative cells close to neurod:GFP-positive cells (arrows) were found frequently in unlesioned sides, for example, at 2 dpl and in sham-treated SAGs at 57 dpl. However, proliferating neurod:GFP/PCNA double-positive cells (arrowheads) were also present from 2 dpl onward in ouabain-injected animals (right column) and frequently from 4 dpl onward in sham-treated and NaCl-injected animals (medial columns). Proliferating neurod:GFP/PCNA double-positive cells continued to be present at 8 and 16 dpl before returning to homeostatic levels at 57 dpl when the majority of sham-treated, or NaCl- or ouabain-injected animals did not show proliferating neuronal progenitors in the lesioned SAG. In SAGs of unlesioned sides, PCNA/neurod:GFP double-positive cells are found occasionally, in particular, in the unlesioned sides of ouabain-injected fish at 8 dpl. Scale bar: 50 µm.

Quantification of reactive proliferation of neurod:GFP-positive progenitors. (A) Quantification of all PCNA-positive cells. Compared with the respective SAGs of unlesioned sides (ULS sham, ULS NaCl, ULS Ouabain), the number of PCNA-positive cells per section is increased following sham treatment, or NaCl or ouabain injections in the first 8 days post lesion (dpl). An even significant increase compared with the unlesioned side is present at 2 dpl following NaCl injection, at 4 dpl following ouabain injection and at 8 dpl following sham treatment, respectively. At 57 dpl, the proliferation rate in the lesioned SAG was significantly decreased to the levels of the unlesioned side. (B) Quantification of neurod:GFP/PCNA double-positive cells. Compared with the respective SAGs of unlesioned sides, the number of neurod:GFP-positive cells per section that re-enter the cell cycle is increased following sham treatment, or NaCl or ouabain injections. The peak of proliferation of active cycling neuronal progenitors is observed at 8 dpl with a significantly increased number of five to six neurod:GFP/PCNA double-positive cells per section in sham-treated and ouabain-injected SAGs. The number of proliferating neurod:GFP neuronal progenitors is significantly lower in NaCl-injected SAGs compared with sham-treated and ouabain-injected SAGs at 8 dpl. However, changes in the overall proliferation rate in NaCl-injected SAGs follow a similar trend as seen in sham-treated and ouabain-injected SAGs. The number of neurod:GFP/PCNA double-positive cells per section has significantly decreased at 16 and 57 dpl in sham-treated and ouabain-injected SAGs, respectively, although a small number of lesioned animals still showed low numbers of neurod:GFP/PCNA double-positive cells in each lesion type. Notably, neurod:GFP/PCNA double-positive cells were infrequently also observed in the unlesioned side, in particular, in ouabain-injected animals, in which an average of 0.5–1 neurod:GFP/PCNA double-positive cells per section was seen. Samples size n = 3 (n = fish with three sections/SAG; exception: 4 dpl/NaCl: n = 2); data are presented as mean ± SEM. Two-way ANOVA with Tukey’s multiple comparisons test; ***p ≤ 0.001; **p ≤ 0.01; *p ≤ 0.05.

Generation of new sensory neurons from the neurod:GFP-positive progenitor pool upon lesion. (A) Short-term lineage tracing using the persistence of GFP in the SAG of neurod:GFP transgenic animals. Immunohistochemistry against the neuronal marker HuC/D and GFP label newborn neurons in sham-treated, and NaCl- and ouabain-injected animals as well as the unlesioned side (ULS) of either sham-treated, or NaCl- or ouabain-injected animals at 4, 8, and 57 days post lesion (dpl). Shown are representative overviews of the neurogenic niche of the SAG as well as close-ups to visualize co-staining. At 4 dpl, several neurod:GFP/HuC/D double-positive cells can be found in the lesioned side independent of the type of lesion, whereas neurogenesis is almost not existing in the corresponding unlesioned side (upper row). At 8 dpl, neurogenesis is abundant in the lesioned sides and increased as well in the respective unlesioned sides in sham-treated, and NaCl- and ouabain-injected animals (middle row). At 57 dpl, neurogenesis returned to low levels in the lesioned and unlesioned side with only a few newborn neurons present in the SAG (lower row). (B) Quantification of neurod:GFP/HuC/D double-positive cells per section at 2, 4, 8, 16, and 57 days post lesion. The number of neurod:GFP/HuC/D double-positive cells per section revealed a significant increase in neurogenesis in the lesioned side in comparison with the homeostatic baseline (0.56 neurod:GFP/Huc/D double-positive cells per section) calculated from Schwarzer et al., 2020 (gray box). The number of newborn neurons per section peaked at 4 dpl in ouabain-injected SAGs and at 8 dpl in sham-treated as well as in NaCl-injected SAGs, respectively. At 57 dpl, neurogenesis returned to homeostatic levels in all three lesion conditions. Interestingly, neurogenesis rates in the SAGs of the respective unlesioned sides were also increased over time and slightly above the homeostatic baseline at 8 dpl in sham-treated and NaCl-injected animals as well as at 16 dpl in ouabain-injected animals. Baseline levels in neurogenesis in the respective unlesioned sides was reached again at 16 and 57 dpl, respectively. (C) Time course for bromodeoxyuridine (BrdU) pulse chase experiment. Six-month-old Tg(neurod:GFP) zebrafish were lesioned, followed by two 24-h lasting BrdU treatments at 2 and 4 days post lesion (dpl). Animals were sacrificed at 19 dpl corresponding to 14 days after the second BrdU treatment. (D) Antibody staining against the thymidine analog BrdU, the neuronal marker HuC/D and GFP in the SAG of sham-treated, and NaCl- and ouabain-injected neurod:GFP animals at 19 dpl (14 days post second BrdU pulse). In the SAG of the unlesioned side, BrdU-positive cells can be found in close proximity to neurod:GFP-positive cells (gray arrow). In contrast, neurod:GFP/BrdU double-positive progenitors (arrow), neurod:GFP/BrdU/HuC/D triple-positive newborn neurons (arrowhead), and BrdU/HuC/D double-positive neurons (open arrowhead) can be found in the SAG of sham-treated, and NaCl- and ouabain-injected animals. Scale bars: 50 µm. For quantification sample size n = 3 (n = fish with three sections/SAG; exception: 4 dpl/NaCl: n = 2); data are presented as mean ± SEM. Two-way ANOVA with Tukey’s multiple comparisons test; ***p ≤ 0.001; **p ≤ 0.01; *p ≤ 0.05.

FIGURE 7. Schematic illustrations of the events in the adult zebrafish SAG upon lesion. (A) Scheme depicting the cellular events within the neurogenic niche. During homeostasis, the neurogenic niche of the adult zebrafish SAG is quiescent in regard to proliferation and shows only very rare neurogenesis. Upon lesion, mature sensory neurons as well as other marker-negative cells undergo apoptosis within several hours. Within days, reactive proliferation of marker-negative and neuronal progenitor cells is observed accompanied by reactive neurogenesis from the neuronal progenitor pool. Reactive proliferation and reactive neurogenesis continue over the next weeks until previous homeostatic levels are reached. (B) Scheme of the temporal order of events. During homeostasis, only basal levels of proliferating marker-negative cells, proliferating neuronal progenitors, and leukocytes are present in the neurogenic niche of the SAG. In addition, only basal levels of newborn neurons derived from the neuronal progenitors and no apoptotic cells are found. Upon lesion, a strong increase in the number of apoptotic cells and, in particular, apoptotic sensory neurons is observed, which is also resolved rapidly. Also within hours, a massive accumulation of leukocytes occurs, which persists for several weeks. Within days, the number of proliferating cells increases significantly within the neurogenic niche of the SAG and remains high for a couple of weeks until it returns to lower levels. Following the general gain in proliferation, the number of proliferating neuronal progenitors is also significantly increased for approximately 2–3 weeks until it returns to baseline levels. Onset of proliferation of neuronal progenitors is accompanied by increased production of newborn neurons, which remains high for approximately 2–3 weeks until it returns to homeostatic levels.