Article Figures & Data
Figures
- Figure 1.
Theoretical probabilistic models reveal potential for error in a common paradigm for testing cell autonomous gene function. A, Cell autonomous function of genes is frequently investigated using a cell-specific Cre-loxP system where expression of reporter protein from a stop-floxed reporter construct is assumed to reliably indicate target gene recombination within a single cell. Here, a generic schematic of the genetics used and assumptions employed in this common paradigm are shown. B, The probabilities of true and false reporter signals are shown for three hypothetical scenarios with 100 target cells for visualization. All three scenarios assume an experiment where gene recombination occurs only in the presence of Cre, reporter protein expression is observed in 50% of target cells [P(Rr) = 0.5], and reporter recombination is greater than or equal to target gene recombination [P(Rr) ≥ P(Tr)]. The numbers in red boxes represent given values that are then used to determine the remaining probabilities. The “theoretical ideal” represents an ideal scenario where reporter gene recombination and target gene recombination overlap perfectly. The “different efficiency, max concordance” scenario shows a case where target gene recombines with less efficiency than the reporter, but shows the maximum possible concordance with reporter recombination given that constraint. The “different efficiency, < max concordance” scenario shows a case where target gene recombination is mildly less efficient than reporter gene recombination and concordance is moderately suboptimal, with 75% of target recombined cells also showing reporter recombination. In the hypothetical deviations from ideal, the probability of reporter expression accurately predicting target gene recombination [P(Tr|Rr)] is 0.5 and 0.6, respectively. These examples demonstrate that even mild deviation from ideal can introduce possible substantial error in using reporter gene recombination as an indicator of target gene recombination in the same cell. See also Extended Data Figure 1-1. Figure Contributions: Elizabeth Diana Kirby developed theoretical models and made figures.
- Figure 2.
Recombination-dependent fluorescent reporter gene expression is correlated over the whole SGZ, but frequently fails to co-localize. A, A schematic of the transgenic mouse model employed in our experiments where we combined NestinCreERT2 mice Rosa-stop-floxed-EYFP and Rosa-CAG-stop-floxed-tdTomato mice. Mice were then submitted to three TAM administration conditions, 3D short (B), 3D long (C), and 5D (D; top, representative timeline of TAM injections and recovery; bottom, immunostaining of EYFP and tdTomato in SGZ; scale bars: 100 μm). E, EYFP+ and tdTomato+ DG percent area was compared by two-way ANOVA in 3D short, 3D long, and 5D mice. Tukey’s post hoc comparison within reporter type; *p < 0.05, **p < 0.01, ***p < 0.001 versus 5D. F, EYFP+ and tdTomato+ DG percent area were correlated in all mice. R,p Pearson’s correlation. G, Representative orthogonal images: immunostaining and imaging of EYFP+, tdTomato+, and EYFP+tdTomato+ cells in the SGZ. Scale bar: 20 μm. H, EYFP+ colocalization in tdTomato+ area (H) and tdTomato+ colocalization in EYFP+ area (I) were compared with theoretical 100% colocalization; n = 3 mice per group. Data are shown as mean ± SEM; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Figure Contributions: Tyler Joseph Dause ran experiments and analyzed data. Tyler Joseph Dause and Elizabeth Diana Kirby made figure.
- Figure 3.
Recombination-dependent fluorescent reporter gene expression within RGLs and IPCs frequently fails to co-localize in the same cells. A, SGZ RGLs were identified by immunostaining of EYFP, tdTomato, and GFAP. Scale bars: 20 μm. Arrow head = EYFP+/tdTomato– RGL. Arrow = EYFP–/tdTomato+ RGL. Chevron = EYFP+/tdTomato+ RGL. B, SGZ IPCs were identified by immunostaining of EYFP, tdTomato and Ki-67. Scale bars: 20 μm. Arrow head = EYFP+/tdTomato– IPC. Arrow = EYFP–/tdTomato+ IPC. Chevron = EYFP+/tdTomato+ IPC. Density (C) and percentage (D) of SGZ GFAP+ RGLs coexpressing EYFP only(EYFP), tdTomato only(tdTom), EYFP+ and tdTomato+ (both), or neither are shown. Density (E) and percentage (F) of SGZ Ki67+ IPCs coexpressing EYFP, tdTom, both, or neither are shown; n = 3 mice per group. Data are shown as mean ± SEM. Density represented as cells per area, scale × 10−4. See also Extended Data Figure 3-1. Figure Contributions: Tyler Joseph Dause ran experiments and analyzed data. Tyler Joseph Dause and Elizabeth Diana Kirby made figure.
- Figure 4.
Using expression of one reporter to predict expression of the other results in high false signals similarly across TAM protocols. A, The mean observed recombination frequencies from the 5D TAM group are represented here as reporter expression in 100 hypothetical target cells, either RGLs or IPCs. The probabilities of true and false signals are given as if tdTomato expression is being used to predict EYFP expression. B, The percent of RGLs and IPCs combined (total NSPCs) that show recombination in both reporter genes (true +) or neither (true –) is shown for three TAM groups. C, The total true signal between the three TAM groups is shown; n = 3 mice per group. Data shown are mean ± SEM; *p < 0.05 determined by two-way ANOVA. See also Extended Data Figures 4-1, 4-2. Figure Contributions: Elizabeth Diana Kirby performed statistical analyses and made figures.
Tables
Primary antibody Vendor/ product no. Dilution Secondary Vendor/product no. Dilution Rabbit anti-mCherry Abcam ab167453 1:500 Donkey anti-Rabbit IgG (H+L) highly cross-adsorbed secondary antibody, AlexaFluor 555 ThermoFisher Scientific
A-315721:500 Hoechst Fisher 33342 1:2000 N/A N/A N/A Goat
Anti-GFP antibodyAbcam Ab6673 1:1000 Donkey anti-Goat IgG (H+L) cross-adsorbed secondary antibody, AlexaFluor 488 Fisher A-11055 1:500 Mouse
Anti-glial fibrillary acidic protein, clone GA5 (GFAP)EMD Millipore
MAB3601:1000 Donkey anti-mouse IgG (H+L) highly cross-adsorbed secondary antibody, Alexa Fluor 647 Fisher A-31571 1:500 Rat
anti-Ki-67 monoclonal antibodyInvitrogen
14-5698-821:500 AlexaFluor 647 AffiniPure donkey anti-Rat IgG (H+L) (712-605-153) Jackson ImmunoResearch
712-605-1531:500 EYFP and tdTomato DG percentage area comparison TAM administration n Test Comparison Statistic P Significant 2E 3D Short 3 Two-way ANOVA TAM × reporter F(2,6) = 8.880 0.0161 Yes* 2E 3D Long 3 TAM F(2,6) = 11.32 0.0092 Yes** 2E 5D 3 Reporter F(1,6) = 14.01 0.0096 Yes** Subject F(6,6) = 12.52 0.0036 Yes** EYFP tdTomato Post hoc Comparisons Adjusted P Significant Adjusted P Significant Tukey’s multiple comparisons 3D short vs 3D long ≥0.9999 No 0.9976 No 3D short vs 5D 0.0009 Yes*** 0.0274 Yes* 3D long vs 5D 0.0009 Yes*** 0.0308 Yes* EYFP and tdTomato DG percentage area correlation Figure TAM administration n Test Statistic P Significant 2F All 9 Pearson's Correlation r(8) = 0.9157 0.0005 Yes*** EYFP+ percentage area colocalization with tdTomato+ area TAM administration n Test Statistic P Significant 2H 3D Short 3 One-sample t test w/ comparison to 100% t(2) = 35.87 0.0008 Yes*** 2H 3D Long 3 t(2) = 13.15 0.0057 Yes** 2H 5D 3 t(2) = 15.36 0.0042 Yes** tdTomato+ percentage area colocalization with EYFP+ area Figure TAM administration n Test Statistic P Significant 2I 3D Short 3 One-sample t test w/ comparison to 100% t(2) = 119.7 0.0001 Yes**** 2I 3D Long 3 t(2) = 10.40 0.0091 Yes** 2I 5D 3 t(2) = 11.27 0.0078 Yes** DG percentage of NSPCs with True+ Signal Figure TAM administration n Test Comparison Statistic P Significant 4B 3D Short 3 Two-way ANOVA True signal × Tam F(2,6) = 6.084 0.0360 Yes* 4B 3D Long 3 True signal F(1,6) = 3.215 0.1231 No 4B 5D 3 TAM F(2,6) = 3.201 0.1132 No Subject F(6,6) = 0.02808 0.9998 No Post hoc Comparisons Adjusted P Significant Tukey’s multiple comparisons 3D short vs 3D long 0.8875 No 3D short vs 5D 0.0225 Yes* 3D long vs 5D 0.0516 No DG percentage of NSPCs with True−Signal Figure TAM administration n Test Comparison Statistic P Significant 4B 3D Short 3 Two-way ANOVA True−Signal × TAM F(2,6) = 6.084 0.0360 Yes* 4B 3D Long 3 True−Signal F(1,6) = 3.215 0.1231 No 4B 5D 3 TAM F(2,6) = 3.201 0.1132 No Subject F(6,6) = 0.02808 0.9998 No Post hoc Comparisons Adjusted P Significant Tukey’s multiple comparisons 3D short vs 3D long 0.9389 No 3D short vs 5D 0.0328 Yes* 3D long vs 5D 0.0179 Yes* DG percentage of NSPCs with True ± Signal Figure TAM administration n Test Statistic P Significant 4C 3D Short 3 One-way ANOVA F(2,6) = 3.201 0.1132 No 4C 3D Long 3 F(2,6) = 3.201 0.1132 No 4C 5D 3 F(2,6) = 3.201 0.1132 No *p < 0.05, **p < 0.01,***p < 0.001, ****p < 0.0001.
Supplementary Extended Data Table 2-1
Main Figures Raw Data. Download Table 2-1, DOCX file.
Supplementary Extended Data Table 2-2
Extended Data Figures Raw Data and Statistics. Download Table 2-2, DOCX file.
Extended Data Figure 1-1
Equation derivation for true and false signal probabilities. A, Assuming reporter recombination is greater than or equal to target gene recombination [P(Rr) ≥ P(Tr)], probability of total true signal is derived using standard conditional and unconditional probability formulas. B–D, Application of the equations in A to the scenarios described in Figure 1B are shown in full detail. Figure Contributions: Elizabeth Diana Kirby developed theoretical models. Download Figure 1-1, TIF file.
Extended Data Figure 2-1
Comparison of oil- and TAM-injected adult NestinCreERT2;Rosa(EYFP/tdTom) mice. A, Immunostaining in the adult DG shows that oil administration does not stimulate expression of either reporter gene. B, TAM administration induces robust recombination-dependent expression of both EYFP and tdTomato in the SGZ. Scale bars: 100 μm. Figure Contributions: Tyler Joseph Dause ran TAM experiments and tissue staining. Tyler Joseph Dause and Elizabeth Diana Kirby made figures. Download Figure 2-1, TIF file.
Extended Data Figure 3-1
Cell-specific fluorescent reporter recombination frequency and correlation. A, Percent of EYFP+ or tdTomato+/GFAP+ RGLs were compared in 3D short, 3D long, and 5D mice. B, Correlation of percent of GFAP+ RGLs that express EYFP and tdTomato in all mice. C, Percent of EYFP+ or tdTomato+ IPCs were compared in 3D short, 3D long, and 5D mice. D, Correlation of percent of Ki67+ IPCs that express EYFP and tdTomato in all mice; n = 3 mice per group. Data are shown as mean ± SEM; *p < 0.05, determined by two-way ANOVA (A, C) or Pearson’s correlation (B, D). Figure Contributions: Tyler Joseph Dause ran experiments and analyzed data. Tyler Joseph Dause and Elizabeth Diana Kirby made figures. Download Figure 3-1, TIF file.
Extended Data Figure 4-1
Effects of cell type and TAM protocol on accuracy of using expression of one reporter to predict the other. A, The mean observed recombination frequencies from the 3D short (left) and 3D long (right) groups are represented here as reporter expression in 100 hypothetical target cells, either RGLs or IPCs. The probabilities of true and false signals are given as if tdTomato expression is being used to predict EYFP expression. B, The percent of GFAP+ RGL cells that show recombination in both reporter genes (true +) or neither (true –) is shown for three TAM groups. C, The total true signal (+ and –) in GFAP+ RGL cells for the three TAM groups is shown. D, The percent of Ki67+ IPCs that show recombination in both reporter genes (true +) or neither (true –) is shown for three TAM groups. E, The total true signal (+ and –) in Ki67+ IPCs for the three TAM groups is shown; n = 3 mice per group. Data shown are mean ± SEM; *p < 0.05, determined by two-way ANOVA. Figure Contributions: Tyler Joseph Dause ran experiments and analyzed data. Tyler Joseph Dause and Elizabeth Diana Kirby made figures. Download Figure 4-1, TIF file.
Extended Data Figure 4-2
Fluorescent reporter recombination and colocalization in SVZ NSPCs. A, Immunostaining of EYFP, tdTomato in SVZ NSPCs. Scale bars: 100 μm. Arrowhead = EYFP+/tdTomato– NSPC. Arrow = EYFP–/tdTomato+ NSPC. Chevron = EYFP+/tdTomato+ NSPC. B, Correlation of EYFP+ and tdTomato+ SVZ percent area in all mice. C, Comparison of EYFP+ colocalization in tdTomato+ area to theoretical 100% colocalization. D, Comparison of tdTomato+ colocalization in EYFP+ area to theoretical 100% colocalization; n = 9 mice. Data are shown as mean ± SEM; ****p < 0.0001 determined by Pearson’s correlation (B) or one-sample t test against a theoretical 100% (C, D). Figure Contributions: Tyler Joseph Dause ran experiments and analyzed data. Tyler Joseph Dause and Elizabeth Diana Kirby made figure. Download Figure 4-2, TIF file.
In this issue
