1. Developmental Biology
  2. Stem Cells and Regenerative Medicine

Distinct progenitor populations mediate regeneration in the zebrafish lateral line

  1. Eric D Thomas
  2. David W Raible  Is a corresponding author
  1. University of Washington, United States
https://doi.org/10.7554/eLife.43736
Share this article
Cite this article
  1. Eric D Thomas
  2. David W Raible
(2019)
Distinct progenitor populations mediate regeneration in the zebrafish lateral line
eLife 8:e43736.
https://doi.org/10.7554/eLife.43736
  2. Download BibTeX
  3. Download .RIS
12 figures, 1 table and 1 additional file

Figures

Figure 1
Download asset Open asset
Hair cell progenitors are replenished via proliferation of other support cells.

(A, F) Timelines of single-ablation (A) and double-ablation (F) proliferation experiments. (B) Maximum projections of mock- (Mock) and neomycin-treated (Neo) neuromasts. EdU-positive cells are shown in magenta, anti-Parvalbumin-stained hair cells are shown in green, and DAPI-stained nuclei are shown in blue. Arrowheads indicate EdU-positive support cells. Scale bar = 10 μm. (C) Percentage of hair cells per neuromast labeled by EdU. Mock: 6.11 ± 8.69, n = 50 neuromasts (10 fish); Neo: 78.24 ± 20.69, n = 45 neuromasts (nine fish); mean ± SD; Mann Whitney U test, p<0.0001. (D) Total EdU-positive cells per neuromast. Mock: 2.78 ± 1.84, n = 50 neuromasts (10 fish); Neo: 8.18 ± 3.07, n = 45 neuromasts (nine fish); mean ± SD; Mann Whitney U test, p<0.0001. (E) Percentage of EdU-positive cells per neuromast that are either hair cells or support cells. Mock: 29.73% hair cells, 70.27% support cells, n = 50 neuromasts (10 fish); Neo: 72.02% hair cells, 27.98% support cells, n = 45 neuromasts (nine fish); mean ± SD. (G–N) Individual slices of a neuromast following two regenerations at two different planes: apical hair cell layer (G–J) and basal support cell layer (K–N). EdU (visualized by a Click-iT reaction) is labeled in magenta, BrdU (anti-BrdU) is labeled in cyan, and myo6b:GFP hair cells are labeled in green. Arrowheads indicate EdU/BrdU-positive hair cells, and asterisks indicate EdU-positive support cells. Scale bar = 10 μm.

https://doi.org/10.7554/eLife.43736.003
Figure 1—source data 1

Hair cell progenitors are replenished via proliferation of other support cells

https://doi.org/10.7554/eLife.43736.004
Figure 2 with 3 supplements
Download asset Open asset
Genetic labeling of distinct support cell populations.

(A, C, E) Maximum projections of neuromasts from sfrp1a:nlsEos (Peripheral, A), tnfsf10l3:nlsEos (AP, C), and sost:nlsEos (DV, E) fish. Converted nlsEos-positive cells are shown in magenta, and brn3c:GFP-positive hair cells are shown in green. Scale bar = 10 μm. (B, D, F) Percentage of hair cells per neuromast labeled by Peripheral (B), AP (D), and DV cells (F) at 5 and 8 dpf. (B) 5 dpf: 19.04 ± 19.86, n = 50 neuromasts (10 fish); 8 dpf: 19.46 ± 19.44, n = 50 neuromasts (10 fish); mean ± SD; Mann Whitney U test, p=0.7047. (D) 5 dpf: 5.71 ± 8.22, n = 50 neuromasts (10 fish); 8 dpf: 6.36 ± 9.57, n = 50 neuromasts (10 fish); mean ± SD; Mann Whitney U test, p=0.9668. (F) 5 dpf: 38.93 ± 13.46, n = 50 neuromasts (10 fish); 8 dpf: 55.78 ± 14.13, n = 50 neuromasts (10 fish); mean ± SD; Mann Whitney U test, p<0.0001.

https://doi.org/10.7554/eLife.43736.005
Figure 2—source data 1

Genetic labeling of distinct support cell populations

https://doi.org/10.7554/eLife.43736.010
Figure 2—figure supplement 1
Download asset Open asset
Asymmetry of support cell transgene expression in secondary neuromasts is orthogonal to primary neuromasts.

(A–F) Maximum projections of lateral views of trunks (taken at 20x zoom) (A,D) or of individual neuromasts (B–C, E–F) from tnfsf10l3:nlsEos (A–C), and sost:nlsEos fish (D–F). Unconverted nlsEos-positive cells are shown in green. In Low Mag images (A,D), primary neuromasts are labeled with 1°, and secondary neuromasts are labeled with 2°. Scale bar = 10 μm.

https://doi.org/10.7554/eLife.43736.006
Figure 2—figure supplement 2
Download asset Open asset
Support cell transgenes are not expressed in hair cells.

(A–C) Maximum projections of neuromasts from Tg[sfrp1a:GFP]w222 (Peripheral, A), Tg[tnfsf10l3:GFP]w223 (AP, B), and sost:NTR-GFP (DV, C) fish. GFP-positive cells are shown in green, and hair cells are shown in magenta via myo6:mKate2. In all three populations, there is no GFP expression in hair cells. Scale bar = 10 μm.

https://doi.org/10.7554/eLife.43736.007
Figure 2—figure supplement 3
Download asset Open asset
Expression of support cell transgenes does not alter hair cell development or regeneration.

(A) Total number of FM 1-43FX-labeled hair cells per neuromast at 5 dpf or following hair cell regeneration (72 hpt) in sfrp1a:nlsEos fish (nlsEos) and non-transgenic siblings (Sib). 5 dpf: 13.28 ± 1.32 (Sib) vs. 12.72 ± 1.93 (nlsEos), n = 25 neuromasts each (5 fish each); Regeneration: 6.76 ± 1.55 (Sib) vs. 7.52 ± 1.56 (nlsEos), n = 25 neuromasts each (5 fish each); mean ± SD; Mann Whitney U test, p=0.1414 (5 dpf), p=0.0658 (Regeneration). (B) Total number of FM 1-43FX-labeled hair cells per neuromast at 5 dpf or following hair cell regeneration (72 hpt) in tnfsf10l3:nlsEos fish (nlsEos) and non-transgenic siblings (Sib). 5 dpf: 12.88 ± 1.56 (Sib) vs. 12.64 ± 2.25 (nlsEos), n = 25 neuromasts each (five fish each); Regeneration: 9.04 ± 1.49 (Sib) vs. 8.72 ± 1.31 (nlsEos), n = 25 neuromasts each (five fish each); mean ± SD; Mann Whitney U test, p=0.9338 (5 dpf), p=0.2722 (Regeneration). (C) Total number of FM 1-43FX-labeled hair cells per neuromast at 5 dpf or following hair cell regeneration (72 hpt) in sost:nlsEos fish (nlsEos) and non-transgenic siblings (Sib). 5 dpf: 12.52 ± 1.76 (Sib) vs. 12.32 ± 1.46 (nlsEos), n = 25 neuromasts each (five fish each); Regeneration: 8.80 ± 1.41 (Sib) vs. 8.64 ± 1.35 (nlsEos), n = 25 neuromasts each (five fish each); mean ± SD; Mann Whitney U test, p=0.4527 (5 dpf), p=0.5396 (Regeneration).

https://doi.org/10.7554/eLife.43736.008
Figure 2—figure supplement 3—source data 1

Expression of support cell transgenes does not alter hair cell development or regeneration

https://doi.org/10.7554/eLife.43736.009
Figure 3
Download asset Open asset
Distinct support cell populations have different regenerative capacities.

(A) Timeline of nlsEos fate mapping experiment. Fish were photoconverted at 5 dpf, treated with neomycin, then fixed and imaged 72 hr post treatment (8 dpf). (B, C, D) Maximum projections of neuromasts from sfrp1a:nlsEos (Peripheral, B), tnfsf10l3:nlsEos (AP, C), and sost:nlsEos (DV, D) fish following photoconversion and hair cell regeneration. Converted nlsEos-positive cells are shown in magenta, and brn3c:GFP-positive hair cells are shown in green. Arrowheads indicate nlsEos-positive hair cells. Scale bar = 10 μm. (E) Percentage of hair cells per neuromast labeled by nlsEos following regeneration. Sfrp1a:nlsEos (Peripheral): 3.59 ± 8.87, n = 50 neuromasts (10 fish); tnfsf10l3:nlsEos (AP): 20.28 ± 20.58, n = 50 neuromasts (10 fish); sost:nlsEos (DV): 60.87 ± 12.37, n = 50 neuromasts (10 fish); mean ± SD; Kruskal-Wallis test, Dunn’s post-test, p=0.003 (Peripheral vs. AP), p<0.0001 (Peripheral vs. DV, AP vs. DV). (F) Total nlsEos-positive support cells per neuromast prior to hair cell ablation. Sfrp1a:nlsEos (Peripheral): 14.30 ± 4.17, n = 50 neuromasts (10 fish); tnfsf10l3:nlsEos (AP): 22.8 ± 4.40, n = 50 neuromasts (10 fish); sost:nlsEos (DV): 23.86 ± 4.45, n = 50 neuromasts (10 fish); mean ± SD; Kruskal-Wallis test, Dunn’s post-test, p<0.0001 (Peripheral vs. AP, Peripheral vs. DV), p>0.9999 (AP vs. DV).

https://doi.org/10.7554/eLife.43736.011
Figure 3—source data 1

Distinct support cell populations have different regenerative capacities

https://doi.org/10.7554/eLife.43736.012
Figure 4
Download asset Open asset
Notch signaling differentially regulates support cell populations.

(A, E, I) Maximum projections of neuromasts expressing sfrp1a:nlsEos (Peripheral, A), tnfsf10l3:nlsEos (AP, E), and sost:nlsEos (DV, I) following Notch-inhibited hair cell regeneration. Converted nlsEos-positive cells are shown in magenta, and brn3c:GFP-positive hair cells are shown in green. Scale bar = 10 μm. (B) Total number of hair cells per neuromast in sfrp1a:nlsEos fish following hair cell regeneration. Neo: 10.28 ± 1.88, n = 50 neuromasts (10 fish); Neo/LY: 19.07 ± 6.79, n = 46 neuromasts (10 fish); mean ± SD; Mann Whitney U test, p<0.0001. (C) Sfrp1a:nlsEos-positive hair cells per neuromast following hair cell regeneration. Neo: 0.62 ± 1.28, n = 50 neuromasts (10 fish); Neo/LY: 1.15 ± 2.16, n = 46 neuromasts (10 fish); mean ± SD; Mann Whitney U test, p=0.2481. (D) Percentage of sfrp1a:nlsEos-labeled hair cells per neuromast following hair cell regeneration. Neo: 6.31 ± 13.83, n = 50 neuromasts (10 fish); Neo/LY: 4.95 ± 8.82, n = 46 neuromasts (10 fish); mean ± SD; Mann Whitney U test, p=0.5148. (F) Total number of hair cells per neuromast in tnfsf10l3:nlsEos fish following hair cell regeneration. Neo: 8.84 ± 1.75, n = 50 neuromasts (10 fish); Neo/LY: 22.93 ± 5.45, n = 40 neuromasts (eight fish); mean ± SD; Mann Whitney U test, p<0.0001. (G) Tnfsf10l3:nlsEos-positive hair cells per neuromast following hair cell regeneration. Neo: 2.22 ± 1.94, n = 50 neuromasts (10 fish); Neo/LY: 11.38 ± 4.23, n = 40 neuromasts (eight fish); mean ± SD; Mann Whitney U test, p<0.0001. (H) Percentage of tnfsf10l3:nlsEos-labeled hair cells per neuromast following hair cell regeneration. Neo: 25.19 ± 21.72, n = 50 neuromasts (10 fish); Neo/LY: 50.68 ± 19.23, n = 40 neuromasts (eight fish); mean ± SD; Mann Whitney U test, p<0.0001. (J) Total number of hair cells per neuromast in sost:nlsEos fish following hair cell regeneration. Neo: 10.94 ± 2.23, n = 50 neuromasts (10 fish); Neo/LY: 27.06 ± 6.90, n = 48 neuromasts (10 fish); mean ± SD; Mann Whitney U test, p<0.0001. (K) Sost:nlsEos-positive hair cells per neuromast following hair cell regeneration. Neo: 7.40 ± 2.13, n = 50 neuromasts (10 fish); Neo/LY: 15.25 ± 6.36, n = 48 neuromasts (10 fish); mean ± SD; Mann Whitney U test, p<0.0001. (L) Percentage of sost:nlsEos-labeled hair cells per neuromast following hair cell regeneration. Neo: 67.86 ± 14.63, n = 50 neuromasts (10 fish); Neo/LY: 54.69 ± 14.01, n = 48 neuromasts (10 fish); mean ± SD; Mann Whitney U test, p<0.0001.

https://doi.org/10.7554/eLife.43736.013
Figure 4—source data 1

Notch signaling differentially regulates support cell populations

https://doi.org/10.7554/eLife.43736.014
Figure 5
Download asset Open asset
Differences in overlap between sost:NTR-GFP and sost:nlsEos populations.

(A–B) Maximum projections of neuromasts from sost:NTR-GFP; sost:nlsEos fish at 3 dpf (A) and 5 dpf (B). Sost:NTR-GFP cells are shown in green and sost:nlsEos cells are shown in magenta. Arrowheads indicate cells expressing sost:nlsEos but not sost:NTR-GFP. Scale bar = 10 μm. (C) Support cells per neuromast expressing either NTR-GFP and nlsEos (green) or nlsEos only (magenta) at 3 dpf and 5 dpf. NTR-GFP/nlsEos: 9.04 ± 2.39 (3 dpf) vs. 8.47 ± 2.27 (5 dpf), n = 49 neuromasts each (10 fish each); nlsEos only: 6.10 ± 2.27 (3 dpf) vs. 10.86 ± 2.72 (5 dpf), n = 49 neuromasts each (10 fish each); mean ± SD; Kruskal-Wallis test, Dunn’s post-test, p>0.9999 (NTR-GFP/nlsEos 3 dpf vs. 5 dpf), p<0.0001 (nlsEos only 3 dpf vs. 5 dpf). (D–F) Maximum projections of neuromasts from sost:NTR-GFP; sost:nlsEos fish following mock treatment (D; Mock), Mtz at 5 dpf (E; Mtz5), and Mtz at 3 dpf and 5 dpf (F; Mtz3/5). Sost:NTR-GFP cells are shown in green and sost:nlsEos cells are shown in magenta. Scale bar = 10 μm. (G) Support cells per neuromast solely expressing sost:nlsEos following Mtz treatment. Mock: 11.18 ± 2.04, n = 50 neuromasts (10 fish); Mtz5: 9.72 ± 2.03, n = 50 neuromasts (10 fish); Mtz3/5: 6.76 ± 2.12, n = 50 neuromasts (10 fish); mean ±SD; Kruskal-Wallis test, Dunn’s post-test, p=0.0288 (Mock vs. Mtz5), p<0.0001 (Mock vs. Mtz3/5, Mtz5 vs. Mtz3/5).

https://doi.org/10.7554/eLife.43736.015
Figure 5—source data 1

Differences in overlap between sost:NTR-GFP and sost:nlsEos populations

https://doi.org/10.7554/eLife.43736.016
Figure 6 with 2 supplements
Download asset Open asset
Ablation of DV cells decreases number of regenerated hair cells.

(A) Timeline of DV cell-ablation experiment. Larvae were treated with Mtz at 3 dpf, photoconverted, then treated with neomycin, then treated with Mtz again at 5 dpf, and fixed and immunostained at 72 hpt (8 dpf). (B–D) Maximum projections of neuromasts from sost:NTR-GFP; sost:nlsEos fish following neomycin (B; Neo), neomycin and Mtz (C; Neo/Mtz5), and Mtz, neomycin, and Mtz treatments (D; Mtz3/Neo/Mtz5). Sost:NTR-GFP cells are shown in green, sost:nlsEos cells are shown in magenta, and anti-Parvalbumin-stained hair cells are shown in cyan. Scale bar = 10 μm. (E) Total hair cells per neuromast following regeneration. Neo: 11.73 ± 2.10, n = 49 neuromasts (10 fish); Neo/Mtz5: 9.33 ± 1.88, n = 39 neuromasts (8 fish); Mtz3/Neo/Mtz5: 7.52 ± 1.74, n = 50 neuromasts (10 fish); mean ± SD; Kruskal-Wallis test, Dunn’s post-test, p=0.0001 (Neo vs. Neo/Mtz5), p<0.0001 (Neo vs. Mtz3/Neo/Mtz5), p=0.0016 (Neo/Mtz5 vs. Mtz3/Neo/Mtz5). (F) Sost:nlsEos-positive hair cells per neuromast following regeneration. Neo: 7.78 ± 2.36, n = 49 neuromasts (10 fish); Neo/Mtz5: 4.90 ± 2.02, n = 39 neuromasts (eight fish); Mtz3/Neo/Mtz5: 1.16 ± 1.46, n = 50 neuromasts (10 fish); mean ± SD; Kruskal-Wallis test, Dunn’s post-test, p=0.0003 (Neo vs. Neo/Mtz5), p<0.0001 (Neo vs. Mtz3/Neo/Mtz5, Neo/Mtz5 vs. Mtz3/Neo/Mtz5). (G) Percentage of hair cells per neuromast labeled by sost:nlsEos following regeneration. Neo: 65.81 ± 14.89, n = 49 neuromasts (10 fish); Neo/Mtz5: 51.40 ± 17.17, n = 39 neuromasts (8 fish); Mtz3/Neo/Mtz5: 14.29 ± 18.10, n = 50 neuromasts (10 fish); mean ± SD; Kruskal-Wallis test, Dunn’s post-test, p=0.0147 (Neo vs. Neo/Mtz5), p<0.0001 (Neo vs. Mtz3/Neo/Mtz5, Neo/Mtz5 vs. Mtz3/Neo/Mtz5).

https://doi.org/10.7554/eLife.43736.017
Figure 6—source data 1

Ablation of DV cells decreases number of regenerated hair cells

https://doi.org/10.7554/eLife.43736.022
Figure 6—figure supplement 1
Download asset Open asset
Mtz treatment does not inherently impact hair cell regeneration.

Total number of hair cells per neuromast following regular hair cell regeneration (Neo) or DV cell-ablated regeneration (Mtz3/Neo/Mtz5) in non-transgenic siblings of sost:NTR-GFP fish. Neo: 9.5 ± 1.50, n = 50 neuromasts (10 fish); Mtz3/Neo/Mtz5: 9.98 ± 1.51, n = 40 neuromasts (eight fish); mean ±SD; Mann Whitney U test, p=0.2317.

https://doi.org/10.7554/eLife.43736.018
Figure 6—figure supplement 1—source data 1

Mtz treatment does not inherently impact hair cell regeneration

https://doi.org/10.7554/eLife.43736.019
Figure 6—figure supplement 2
Download asset Open asset
Hair cell development and regeneration is unaffected in fish expressing both sost:NTR-GFP and sost:nlsEos.

Total number of FM 1-43FX-labeled hair cells per neuromast at 5 dpf or following hair cell regeneration (72 hpt) in sost:NTR-GFP fish (NTR-GFP), sost:nlsEos fish (nlsEos), sost:NTR-GFP; sost:nlsEos fish (NTR-GFP/nlsEos), and non-transgenic siblings (Sib). 5 dpf: 13.00 ± 1.41 (Sib) vs. 13.56 ± 2.26 (NTR-GFP) vs. 13.00 ± 1.26 (nlsEos) vs. 13.36 ± 1.98 (NTR-GFP/nlsEos), n = 25 neuromasts each (5 fish each); Regeneration: 8.72 ± 1.65 (Sib) vs. 8.32 ± 1.70 (NTR-GFP) vs. 8.00 ± 0.91 (nlsEos) vs. 8.52 ± 1.23 (NTR-GFP/nlsEos), n = 25 neuromasts each (5 fish each); mean ± SD; Kruskal-Wallis test, Dunn’s post-test; 5 dpf: p > 0.9999 (all comparisons); Regeneration: p=0.4448 (Sib vs. nlsEos), p>0.9999 (all other comparisons).

https://doi.org/10.7554/eLife.43736.020
Figure 6—figure supplement 2—source data 1

Hair cell development and regeneration is unaffected in fish expressing bothsost:NTR-GFP and sost:nlsEos

https://doi.org/10.7554/eLife.43736.021
Figure 7
Download asset Open asset
DV cell-ablation reduces the number of supernumerary hair cells formed during Notch-inhibited hair cell regeneration.

(A) Timeline of dual DV cell-ablation, Notch-inhibition experiment. Sost:NTR-GFP larvae were treated with Mtz at 3 dpf, treated with neomycin at 5dpf, then co-treated with Mtz and LY411575 for 8 hr, then washed out and treated with LY411575 for 16 additional hours (24 hr total LY). (B–E) Maximum projections of sost:NTR-GFP neuromasts following normal hair cell regeneration (B; Neo), Notch-inhibited hair cell regeneration (C; Neo/LY), DV cell-ablated hair cell regeneration (D; Mtz3/Neo/Mtz5), and DV cell-ablated and Notch-inhibited hair cell regeneration (E; Mtz3/Neo/Mtz5/LY). Sost:NTR-GFP cells are shown in green, and anti-Parvalbumin immunostained hair cells are shown in magenta. Scale bar = 10 μm. (F) Total number of hair cells per neuromast following hair cell regeneration. Neo: 9.42 ± 1.85, n = 50 neuromasts (10 fish); Neo/LY: 21.08 ± 4.42, n = 50 neuromasts (10 fish); Mtz3/Neo/Mtz5: 6.86 ± 1.76, n = 50 neuromasts (10 fish); Mtz3/Neo/Mtz5/LY: 15.06 ± 3.51, n = 50 neuromasts (10 fish); mean ± SD; Kruskal-Wallis test, Dunn’s post-test, p<0.0001 (Neo vs. Neo/LY; Mtz3/Neo/Mtz5 vs. Mtz3/Neo/Mtz5/LY), p=0.0058 (Neo vs. Mtz3/Neo/Mtz5), p=0.0029 (Neo/LY vs. Mtz3/Neo/Mtz5/LY).

https://doi.org/10.7554/eLife.43736.023
Figure 7—source data 1

DV cell-ablation reduces the number of supernumerary hair cells formed during Notch-inhibited hair cell regeneration

https://doi.org/10.7554/eLife.43736.024
Figure 8
Download asset Open asset
AP cells and DV cells define separate progenitor populations.

(A–C) Maximum projections of neuromasts from tnfsf10l3:nlsEos (AP, (A), sost:nlsEos (DV, (B), and tnfsf10l3:nlsEos/sost:nlsEos fish (AP + DV, (C) following photoconversion and regeneration. Converted nlsEos-positive cells are shown in magenta, and brn3c:GFP-positive hair cells are shown in green. Arrowheads indicate nlsEos-positive hair cells. Scale bar = 10 μm. (D) Number of nlsEos-positive hair cells per neuromast in each of the nlsEos lines following regeneration. Tnfsf10l3:nlsEos (AP): 2.4 ± 1.84, n = 50 neuromasts (10 fish); sost:nlsEos (DV): 6.34 ± 1.87, n = 50 neuromasts (10 fish); tnfsf10l3:nlsEos/sost:nlsEos (AP + DV): 8.24 ± 1.99, n = 50 neuromasts (10 fish); mean ±SD; Kruskal-Wallis test, Dunn’s post-test, p<0.0001 (AP vs. DV, AP vs. AP + DV), p=0.0031 (DV vs. AP + DV). (E) Percentage of hair cells per neuromast labeled by nlsEos lines following regeneration. AP: 27.59 ± 20.21, n = 50 neuromasts (10 fish); DV: 65.16 ± 13.89, n = 50 neuromasts (10 fish); AP + DV: 87.57 ± 13.02, n = 50 neuromasts (10 fish); mean ± SD; Kruskal-Wallis test, Dunn’s post-test, p<0.0001 (all comparisons).

https://doi.org/10.7554/eLife.43736.025
Figure 8—source data 1

AP cells and DV cells define separate progenitor populations

https://doi.org/10.7554/eLife.43736.026
Figure 9
Download asset Open asset
AP population doesn’t compensate for the loss of the DV population during hair cell regeneration.

(A–B) Maximum projections of tnfsf10l3:nlsEos; sost:NTR-GFP neuromasts following normal hair cell regeneration (A; Neo) or DV cell-ablated hair cell regeneration (B; Mtz3/Neo/Mtz5). Sost:NTR-GFP cells are shown in green, tnfsf10l3:nlsEos cells are shown in magenta, and anti-Parvalbumin-stained hair cells are shown in cyan. Arrowheads indicate nlsEos-positive hair cells. Scale bar = 10 μm. (C) Total number of hair cells per neuromast following hair cell regeneration. Neo: 10.36 ± 1.60, n = 50 neuromasts (10 fish); Mtz3/Neo/Mtz5: 7.98 ± 1.74, n = 50 neuromasts (10 fish); mean ± SD; Mann Whitney U test, p<0.0001. (D) Number of nlsEos-positive hair cells per neuromast following hair cell regeneration. Neo: 2.88 ± 1.83, n = 50 neuromasts (10 fish); Mtz3/Neo/Mtz5: 3.14 ± 1.43, n = 50 neuromasts (10 fish); mean ± SD; Mann Whitney U test, p=0.3855. (E) Percentage of hair cells per neuromast labeled by nlsEos following hair cell regeneration. Neo: 27.26 ± 16.00, n = 50 neuromasts (10 fish); Mtz3/Neo/Mtz5: 40.43 ± 19.44, n = 50 neuromasts (10 fish); mean ± SD; Mann Whitney U test, p=0.0002.

https://doi.org/10.7554/eLife.43736.027
Figure 9—source data 1

AP population doesn’t compensate for the loss of the DV population during hair cell regeneration

https://doi.org/10.7554/eLife.43736.028
Figure 10
Download asset Open asset
DV population regenerates via proliferation.

(A–B) Maximum projections of neuromasts from sost:NTR-GFP fish either untreated (A; Mock) or treated with 10 mM Mtz (B; Mtz). Sost:NTR-GFP cells are shown in green and EdU-positive cells are shown in magenta. Arrowheads indicate EdU-positive sost:NTR-GFP cells. Scale bar = 10 μm. (C) Total number of sost:NTR-GFP cells per neuromast following DV cell regeneration. Mock: 8.94 ± 1.62, n = 50 neuromasts (10 fish); Mtz: 5.34 ± 2.14, n = 50 neuromasts (10 fish); mean ± SD; Mann Whitney U test, p<0.0001. (D) Percentage of sost:NTR-GFP cells per neuromast labeled by EdU following DV cell regeneration. Mock: 14.47 ± 17.95, n = 50 neuromasts (10 fish); Mtz: 57.49 ± 32.34, n = 50 neuromasts (10 fish); mean ± SD; Mann Whitney U test, p<0.0001.

https://doi.org/10.7554/eLife.43736.029
Figure 10—source data 1

DV population regenerates via proliferation

https://doi.org/10.7554/eLife.43736.030
Figure 11 with 2 supplements
Download asset Open asset
DV cells are replenished by other support cell populations.

(A–B, D–E, G–H) Maximum projections of neuromasts expressing sost:NTR-GFP and sost:nlsEos (A–B), sfrp1a:nlsEos (D–E), and tnfsf10l3:nlsEos (G–H) in the absence of (A, D, G; Mock) or following Mtz-induced DV cell ablation (B, E, H; Mtz). Sost:NTR-GFP cells are shown in green and nlsEos-positive cells are shown in magenta. Arrowheads indicate nlsEos-positive sost:NTR-GFP cells. Scale bar = 10 μm. (C) Percentage of sost:NTR-GFP cells per neuromast labeled by sost:nlsEos following DV cell regeneration. Mock: 97.39 ± 7.14, n = 50 neuromasts (10 fish); Mtz: 56.09 ± 33.72, n = 50 neuromasts (10 fish); mean ± SD; Mann Whitney U test, p<0.0001. (F) Percentage of sost:NTR-GFP cells per neuromast labeled by sfrp1a:nlsEos following DV cell regeneration. Mock: 6.15 ± 11.14, n = 50 neuromasts (10 fish); Mtz: 31.27 ± 24.41, n = 50 neuromasts (10 fish); mean ± SD; Mann Whitney U test, p<0.0001. (I) Percentage of sost:NTR-GFP cells per neuromast labeled by tnfsf10l3:nlsEos following DV cell regeneration. Mock: 7.31 ± 9.55, n = 50 neuromasts (10 fish); Mtz: 21.11 ± 22.51, n = 50 neuromasts (10 fish); mean ± SD; Mann Whitney U test, p=0.0004.

https://doi.org/10.7554/eLife.43736.031
Figure 11—source data 1

DV cells are replenished by other support cell populations

https://doi.org/10.7554/eLife.43736.036
Figure 11—figure supplement 1
Download asset Open asset
Ablation of DV cells does not decrease the number of other support cell populations.

(A) Total number of sost:nlsEos-positive support cells per neuromast following DV cell regeneration. Mock: 21.42 ± 3.29, n = 50 neuromasts (10 fish); Mtz: 12.70 ± 3.35, n = 50 neuromasts (10 fish); mean ± SD; Mann Whitney U test, p<0.0001. (B) Total number of sfrp1a:nlsEos-positive support cells per neuromast following DV cell regeneration. Mock: 12.00 ± 2.54, n = 50 neuromasts (10 fish); Mtz: 11.32 ± 2.71, n = 50 neuromasts (10 fish); mean ± SD; Mann Whitney U test, p=0.2413. (C) Total number of tnfsf10l3:nlsEos-positive support cells per neuromast following DV cell regeneration. Mock: 18.40 ± 2.44, n = 50 neuromasts (10 fish); Mtz: 17.46 ± 2.49, n = 50 neuromasts (10 fish); mean ± SD; Mann Whitney U test, p=0.0699.

https://doi.org/10.7554/eLife.43736.032
Figure 11—figure supplement 1—source data 1

Ablation of DV cells does not decrease the number of other support cell populations

https://doi.org/10.7554/eLife.43736.033
Figure 11—figure supplement 2
Download asset Open asset
Expression of nlsEos transgenes does not alter number of sost:NTR-GFP cells.

Total number of sost:NTR-GFP cells per neuromast in sost:nlsEos (sost), sfrp1a:nlsEos (sfrp1a), and tnfsf10l3:nlsEos (tnfsf10l3) fish. Sost: 7.08 ± 1.59, n = 50 neuromasts (10 fish); Sfrp1a: 6.80 ± 1.92, n = 50 neuromasts (10 fish); Tnfsf10l3: 6.96 ± 1.70, n = 50 neuromasts (10 fish); mean ± SD; Kruskal-Wallis test, Dunn’s post-test, p>0.9999 (all comparisons).

https://doi.org/10.7554/eLife.43736.034
Figure 11—figure supplement 2—source data 1

Expression of nlsEos transgenes does not alter number of sost:NTR-GFP cells

https://doi.org/10.7554/eLife.43736.035
Figure 12
Download asset Open asset
Model of neuromast progenitor identity.

Sost:nlsEos-positive cells, located in the dorsoventral (DV) region of the neuromast, contain immature hair cell progenitors (shown in light pink) and mature hair cell progenitors (shown in magenta). Immature hair cell progenitors do not directly generate new hair cells (outlined in dark green) during regeneration, but do become mature hair cell progenitors, which comprise the majority of hair cell progenitors (see magenta-filled hair cells following regeneration). Tnfsf10l3:nlsEos-positive cells (shown in gold), located in the anteroposterior (AP) region of the neuromast, also serve as hair cell progenitors (see gold-filled hair cells following regeneration). Both of these populations are regulated by Notch signaling, and both can replenish immature hair cell progenitors. Finally, sfrp1a:nlsEos-positive cells (shown in light green), located in the periphery, do not serve as hair cell progenitors, nor are they regulated by Notch signaling. However, they are capable of replenishing immature hair cell progenitors, and can thus be classified as an upstream progenitor.

https://doi.org/10.7554/eLife.43736.037

Tables

Key resources table
Reagent type
(species) or resource		DesignationSource or referenceIdentifiersAdditional information
Strain, strain
background (Danio rerio)		sfrp1a:nlsEosThis paperTg[sfrp1a:nlsEos]w217;
CRISPR-mediated
transgenesis
Strain, strain
background (Danio rerio)		sfrp1a:GFPThis paperTg[sfrp1a:GFP]w222;
CRISPR-mediated
transgenesis
Strain, strain
background (Danio rerio)		tnfsf10l3:nlsEosThis paperTg[tnfsf10l3:nlsEos]w218;
CRISPR-mediated
transgenesis
Strain, strain
background (Danio rerio)		tnfsf10l3:GFPThis paperTg[tnfsf10l3:GFP]w223;
CRISPR-mediated
transgenesis
Strain, strain
background (Danio rerio)		sost:nlsEosThis paperTg[sost:nlsEos]w215;
CRISPR-mediated
transgenesis
Strain, strain
background (Danio rerio)		sost:NTR-GFPThis paperTg[sost:NTR-GFP]w216;
CRISPR-mediated
transgenesis
Strain, strain
background (Danio rerio)		myo6:mKate2This paperTg[myosin6b:mKate2]w232;
Gateway cloning and
Tol2-mediated transgenesis
Strain, strain
background (Danio rerio)		brn3c:GFPPMID: 15930106ZFIN ID: ZDB-TGCONSTRCT-070117–142Tg[Brn3c:GAP43-GFP]s356t
Strain, strain
background (Danio rerio)		myo6:GFPPMID: 27991862ZFIN ID: ZDB-TGCONSTRCT-170321–2Tg[myosin6b:GFP]w186
Antibodyanti-Parvalbumin
(mouse monoclonal)		Millipore SigmaMAB1572(1:500)
Antibodyanti-GFP
(rabbit monoclonal)		Thermo Fisher
Scientific		ABfinity G10362(1:500)
Antibodyanti-BrdU
(mouse monoclonal)		Developmental
Studies Hybridoma
Bank		DHSB: G3G4; RRID: AB_2314035(1:100)
Recombinant
DNA reagent		pBSK mbait-GFPPMID: 25293390
Recombinant
DNA reagent		mbait-nlsEosThis paperpDEST vector expressing
nuclear-localized Eos following
the mbait-HSP sequence from
pBSK mbait-GFP; see Materials
and methods for more info.
Recombinant
DNA reagent		pBSK mbait-NTR-GFPThis paperModified pBSK mbait-GFP
vector expressing enhanced
potency nitroreductase fused
to GFP; generated via Gibson
assembly
peptide,
recombinant protein		Cas9 protein with
NLS (injection ready)		PNA BioPNA Bio: CP02
Commercial
assay or kit		NucleoSpin PlasmidMachery-Nagel740588.25
Commercial
assay or kit		NucleoSpin Gel and
PCR Clean-up		Machery-Nagel740609.25
Commercial
assay or kit		Plasmid MaxiQiagen12163
Commercial
assay or kit		RNA Clean and
Concentrator		Zymo ResearchRCC-25
Chemical
compound, drug		Gateway BP Clonase IIThermo Fisher
Scientific		11789020
Chemical
compound, drug		Gateway LR Clonase IIThermo Fisher
Scientific		11791020
Chemical
compound, drug		5x Isothermal Reaction
Buffer		PMID: 19363495
Chemical
compound, drug		MetronidazoleSigma-AldrichSigma-Aldrich: M1547
Chemical
compound, drug		LY411575Sigma-AldrichSigma-Aldrich: SML0506
Chemical
compound, drug		NeomycinSigma-AldrichSigma-Aldrich: N1142
Chemical
compound, drug		f-ara-EdUSigma-AldrichSigma-Aldrich: T511293
Chemical
compound, drug		BrdUSigma-AldrichSigma-Aldrich: B5002
Chemical
compound, drug		FM 1-43FXMolecular ProbesMolecular Probes: F35355
Software,
algorithm		GraphPad PrismGraphPad Softwarewww.graphpad.com
Software,
algorithm		SlidebookIntelligent Imaging
Innovations (3i)		www.intellgent-imaging.com
Software,
algorithm		Zen BlackZeisswww.zeiss.com
Software,
algorithm		FIJIPMID: 22743772

Additional files

Transparent reporting form
https://doi.org/10.7554/eLife.43736.038

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Eric D Thomas
  2. David W Raible
(2019)
Distinct progenitor populations mediate regeneration in the zebrafish lateral line
eLife 8:e43736.
https://doi.org/10.7554/eLife.43736