Retinal ganglion cell defects cause decision shifts in visually evoked defense responses

Abstract

A variety of visual cues can trigger defensive reactions in mice and other species. In mice, looming stimuli that mimic an approaching aerial predator elicit flight or freezing reactions, while sweeping stimuli that mimic an aerial predator flying parallel to the ground typically elicit freezing. The retinal ganglion cell (RGC) types involved in these circuits are largely unknown. We previously discovered that loss of RGC subpopulations in Brn3b knockout mice results in distinct visual response deficits. Here, we report that retinal or global loss of Brn3b selectively ablates the fleeing response to looming stimuli while leaving the freeze response intact. In contrast, freezing responses to sweeping stimuli are significantly affected. Genetic manipulations removing three RGC subpopulations (Brn3a⁺ betta RGCs, Opn4⁺Brn3b⁺, and Brn3c⁺Brn3b⁺ RGCs) result in milder phenocopies of Brn3b knockout response deficits. These findings show that flight and freezing responses to distinct visual cues are mediated by circuits that can already be separated at the level of the retina, potentially by enlisting dedicated RGC types.NEW & NOTEWORTHY Flight and freezing response choices evoked by visual stimuli are controlled by brain stem and thalamic circuits. Genetically modified mice with loss of specific retinal ganglion cell (RGC) subpopulations have altered flight versus freezing choices in response to some but not other visual stimuli. This finding suggests that "threatening" visual stimuli may be computed already at the level of the retina and communicated via dedicated pathways (RGCs) to the brain.

Keywords: Brn3b/Pou4f2 transcription factor; defensive behavior; looming stimuli; retinal ganglion cells; sweep stimuli.

Fig. 1.

Characterization of flight and freeze responses to looming stimuli in Brn3b^KO/KO mice. A: ethogram of log10 (mouse speed) across trials for Brn3b^WT/WT (top) and Brn3b^KO/KO (bottom) mice. Basic looming stimulus schematic: black disk in center of screen expanding from 0 cm (0°) diameter to 31.5 cm (49.3°) diameter over 250 ms (197°/s), remaining onscreen for 500 ms after expansion at contrast 0.55. Stimulus expands once per trial. Each horizontal track represents 1 trial. For each mouse, 3–4 trials (encompassed by black brackets on the left) are shown in chronological order from top to bottom (i.e., first trial at the top). Time (x-axis) shows 10 s before and after the stimulus (stimulus, t = 0, duration 0.75 s, light green lines). Heat map to the right represents the mouse speed. Black denotes periods of nest entry and white periods during which mouse position was undetermined. For speed ethograms and behavior ethograms for all stimulus conditions, see Supplemental Fig. S4. High speed in these ethograms does not necessarily signify flight toward nest. Behavior assessments of flight computed below also include a directionality toward the nest component (materials and methods). B: cumulative probability plots of fleeing (i, iv, vii, and x), entries into nest (ii, v, viii, and xi), and freezing (iii, vi, ix, and xii) for Brn3b^WT/WT (blue, n = 9 mice) and Brn3b^KO/KO (red, n = 12) mice, 3 trials/mouse. Freeze and flee behaviors are normalized to total number of viable trials in time bins of 0.1 s; nest entries are normalized to total number of trials at each time point. i–iii: Traces of responses at maximal contrast (0.55). iv-xii: Traces of responses under various contrast conditions (0.375 = iv–vi; 0.29 = vii–ix; 0.2 = x–xii) for Brn3b^WT/WT (blue, n = 3 mice) and Brn3b^KO/KO (red, n = 4 mice), 3–4 trials/mouse/stimulus. Gray vertical columns indicate the stimulus duration. C: behavior frequency during stimulus presentation (First 0.75 s, left), or within 3 s (First 3 s, right) of stimulus presentation, for Brn3b^WT/WT (blue) and Brn3b^KO/KO (red) mice. Data are normalized to total number of trials for each genotype. Contrast levels are indicated under the plots. χ² test was used to determine significance. For all graphs: *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 2.

Global and/or retinal-specific loss of Brn3b abolishes flight but not freeze responses to looming stimuli. A: ethogram of log (mouse speed) across looming response trials for Rax:Cre; Brn3b^CKOAP/WT (left), and Rax:Cre; Brn3b^CKOAP/CKOAP (right) mice. Stimulus definition and ethogram description are shown. For speed ethograms and behavior ethograms for all other genotypes in this figure, see Supplemental Fig. S5. B: Cumulative probability plots of fleeing (left), nest entry (middle), and freezing (right), for Brn3b^WT/WT (blue, 3 trials/mouse, n = 9 mice), Brn3b^KO/KO (red, 3 trials/mouse, n = 12 mice), and Brn3b^KO/WT (green, 4 trials/mouse, n = 10 mice). Data for Brn3b^WT/WT and Brn3b^KO/KO mice from i–iii are included here again for comparison purposes. C: cumulative probability plots of fleeing (left), nest entry (middle), and freezing (right) for Rax:Cre; Brn3b^CKOAP/WT (orange, 3–4 trials/mice, n = 14 mice), Rax:Cre; Brn3b^CKOAP/CKOAP (black, 4 trials/mice, n = 12 mice), and Rax:Cre; Brn3b^CKOAP/KO (purple, 3–4 trials/mice, n = 10 mice). D: behavior frequency during stimulus (First 0.75 s, left), or within 3 s (First 3 s, right) of stimulus presentation, for the 6 Brn3b genotypes, color coded according to A. Data are normalized to total number of trials for each genotype. E and F: box-whisker plots for flee (E) and freeze (F) reaction times for trials where reaction was observed within 3 s of basic loom stimulus onset. Box indicates the interquartile interval, dots the individual datapoints, red line the median, and whisker the full range of observations. Trials with freezing/fleeing episode starting before stimulus onset and lasting into stimulus period excluded. For E and F, KS2 test was used, and for D, χ² test was used to determine significance. For all graphs: *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 3.

Looming-evoked flight to freeze shifts in retinal Brn3a knockout (KO) mice. A: ethogram of log (mouse speed) across looming response trials for Rax:Cre; Brn3a^CKOAP/WT (left) and Rax:Cre; Brn3a^CKOAP/KO (right) mice. Stimulus definition and ethogram description as in Fig. 2A. For behavior ethograms, see Supplemental Fig. S5. B: cumulative probability plots of fleeing (left), nest entry (middle), and freezing (right) for Rax:Cre; Brn3a^CKOAP/WT (blue, 3–4 trials/mouse, n = 11 mice) and Rax:Cre; Brn3a^CKOAP/KO (orange, 3–4 trials/mouse, n = 10 mice). C: behavior frequency during stimulus (First 0.75 s, left), or within 3 s (First 3 s, right) of stimulus presentation, for Rax:Cre; Brn3a^CKOAP/WT (blue) and Rax:Cre; Brn3a^CKOAP/KO (orange) mice. Data are normalized to total number of trials for each genotype. D: box-whisker plots for flee (left) and freeze (right) reaction times for trials where reaction was observed within 3 s of basic loom stimulus onset. Significance was determined using χ² test for C and KS2 test for D. For all graphs: *P < 0.05, **P < 0.01.

Fig. 4.

Diminished looming-evoked flight in Opn4^Cre/WT; Brn3b^cDTA/WT mice. A: ethogram of log (mouse speed) across looming response trials for Brn3b^cDTA/WT (left), Opn4^Cre/WT (middle), and Opn4^Cre/WT; Brn3b^cDTA/WT (right) mice. Stimulus definition and ethogram description as in Fig. 2A. For behavior ethograms, see Supplemental Fig. S6. B: cumulative probability plots of fleeing (i), nest entry (ii), and freezing (iii) for Brn3b^cDTA/WT (blue, 4 trials/mouse, n = 9 mice), Opn4^Cre/WT (green, 4 trials/mouse, n = 9 mice), and Opn4^Cre/WT; Brn3b^cDTA/WT (red, 4 trials/mouse, n = 11 mice). C: behavior frequency during stimulus (First 0.75 s, left), or within 3 s (First 3 s, right) of stimulus presentation, for Brn3b^cDTA/WT (blue), Opn4^Cre/WT (green), and Opn4^Cre/WT; Brn3b^cDTA/WT (red). Data are normalized to total number of trials for each genotype. D: box-whisker plots for flee (left) and freeze (right) reaction times for trials where reaction was observed within 3 s of basic loom stimulus onset. Significance was determined using χ² test for C and KS2 test for D. For all graphs: *P < 0.05.

Fig. 5.

Looming-evoked flight-to-freeze shifts in Brn3c^Cre/WT; Brn3b^cDTA/WT mice. A: ethogram of log (mouse speed) across looming response trials for Brn3c^Cre/WT (left) and Brn3c^Cre/WT; Brn3b^cDTA/WT (right) mice. Stimulus definition and ethogram description as in Fig. 2A. For behavior ethograms, see Supplemental Fig. S6. B: cumulative probability plots of fleeing (left), nest entry (middle), and freezing (right) for Brn3c^Cre/WT (orange, 4 trials/mouse, n = 7 mice) and Brn3c^Cre/WT; Brn3b^cDTA/WT (black, 4 trials/mouse, n = 10 mice). C: behavior frequency during stimulus (First 0.75 s, left), or within 3 s (First 3 s, right) of stimulus presentation, for Brn3c^Cre/WT (orange) and Brn3c^Cre/WT; Brn3b^cDTA/WT (black). Data are normalized to total number of trials for each genotype. D: box-whisker plots for flee (left) and freeze (right) reaction times for trials where reaction was observed within 3 s of basic loom stimulus onset. Significance was determined using χ² test for C and KS2 test for D.

Fig. 6.

Characterization of freeze and flight responses to sweeping stimuli in Brn3b^KO/KO mice. A: ethogram of log (mouse speed) across trials for Brn3b^WT/WT (left) and Brn3b^KO/KO (right) mice. Basic sweeping stimulus schematic: 2.6 cm (4.3°) black disk moving diagonally across screen at 21.8°/s over 4 s at contrast 0.55. Each horizontal bar is 1 trial; trials from each animal are marked with brackets and arranged top to bottom in chronological order (1st trial at the top). Stimulus duration is indicated by green outline, and time (x-axis) is centered on stimulus onset. Black areas denote nest entry, while white areas denote mice out of frame. For behavior ethograms, see Supplemental Fig. S7. B: cumulative probability plots of freezing (i, iv, vii, x, and xiii), flight (ii, v, viii, xi, and xiv), and nest entries (iii, vi, ix, xii, and xv). Freeze and flee behaviors are normalized to total number of viable trials in time bins of 0.1 s; nest entries are normalized to total number of trials at each time point. B, i–iii: responses to basic stimulus conditions (speed 21.8 ^o/s, spot diameter = 4.3°) for Brn3b^WT/WT (blue, 3–4 trials/mouse, n = 10 mice) and Brn3b^KO/KO (red, 3–4 trials/mouse, n = 11 mice). B, iv–vi: responses to “Big” stimulus conditions (speed 21.8 ^o/s, spot diameter = 7.67°). B, vii–ix: Responses to “Small” stimulus conditions (speed 21.8 ^o/s, spot diameter = 1.92°). B, x–xii: responses to “Fast” stimulus conditions (speed 43.5 ^o/s, spot diameter = 4.3°). B, xiii–xv: Responses to “Slow” stimulus conditions (speed 10.9 ^o/s, spot diameter = 4.3°). Experiments for each condition in iv–xv were carried out for Brn3b^WT/WT (blue, 2–4 trials, n = 8 mice) and Brn3b^KO/KO mice (red, 2–4 trials, n = 6 mice). C: number of trials where behavior was observed during stimulus period, normalized to total number of trials. In C–E, stimulus variants are indicated under the plots and colors of bars refer to mouse genotypes as indicated in A. D: time spent freezing for each trial, normalized to length of stimulus. Trials where no reactions were observed within stimulus period excluded. No significant differences found between the different genotypes under optimal (basic) stimulus conditions. Asterisks indicate significant effect of stimulus speed on normalized time spent freezing for WT mice. E: freeze reaction times. Trials with freezing episode starting before stimulus onset or after stimulus end excluded. For C, χ² test was used to determine significance. For D and E, KS2 test was used to determine significance within same condition between genotypes, and Kruskal-Wallis test was used to determine whether there was an effect of size/speed of stimulus on reaction time or time spent freezing. For all graphs: *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 7.

Global and/or retinal specific Brn3b ablation differentially affects freeze and flight responses to sweeping stimuli. A: ethogram of log (mouse speed) across trials for Rax:Cre; Brn3b^CKOAP/WT (left) and Rax:Cre; Brn3b^CKOAP/CKOAP (right) mice. For Figs. 7–10, basic sweep stimulus was used: a 4.3° black disk moving diagonally across screen at 21.8°/s over 2.76 s at contrast 0.55. Ethogram description is as in Fig. 6A. For speed ethograms and behavior ethograms for all other genotypes in this figure see Supplemental Fig. S8. B: cumulative probability plots of freezing (left), fleeing (middle), and nest entry (right) for Brn3b^WT/WT (blue, 3–4 trials/mouse, n = 10 mice), Brn3b^KO/KO (red, 3–4 trials/mouse, n = 11 mice), and Brn3b^KO/WT (green, 3–4 trials/mouse, n = 10 mice). Data for Brn3b^WT/WT and Brn3b^KO/KO mice from B are included here again for comparison purposes. C: cumulative probability plots of freezing (left), fleeing (middle), and nest entry (right) for Rax:Cre; Brn3b^CKOAP/WT (orange, 3–4 trials/mice, n = 14 mice), Rax:Cre; Brn3b^CKOAP/CKOAP (black, 4 trials/mice, n = 12 mice), and Rax:Cre; Brn3b^CKOAP/KO (purple, 3–4 trials/mice, n = 10 mice). D: box-whisker plots for freeze reaction times. E: behavior frequency during stimulus. Data are normalized to total number of trials for each genotype. Data points in D and bars in E indicate the 6 Brn3b genotypes, color coded according to A. For D, KS2 test was used, and for E, χ² test was used to determine significance. For all graphs: *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 8.

Sweep-evoked freeze and flight responses in retinal Brn3a knockout (KO) mice. A: ethogram of log (mouse speed) across sweep response trials for Rax:Cre; Brn3a^CKOAP/WT (left) and Rax:Cre; Brn3a^CKOAP/KO (right) mice. Stimulus definition and ethogram description as in Fig. 7A. For behavior ethograms, see Supplemental Fig. S8. B: cumulative probability plots of freezing (left), fleeing (middle), and nest entry (right) for Rax:Cre; Brn3a^CKOAP/WT (blue, 3–4 trials/mouse, n = 11 mice) and Rax:Cre; Brn3a^CKOAP/KO (orange, 3–4 trials/mouse, n = 10 mice). C: behavior frequency during stimulus for Rax:Cre; Brn3a^CKOAP/WT (blue) and Rax:Cre; Brn3a^CKOAP/KO (orange) mice. Data are normalized to total number of trials for each genotype. D: box-whisker plots for freezing reaction times. For D, KS2 test was used, and for C, χ² test was used to determine significance. For all graphs: *P < 0.05, **P < 0.01.

Fig. 9.

Flight and freeze reactions to sweeping stimuli are not affected in Opn4^Cre/WT; Brn3b^cDTA/WT mice. A: ethogram of log (mouse speed) across sweeping response trials for Brn3b^cDTA/WT (left), Opn4^Cre/WT (middle), and Opn4^Cre/WT; Brn3b^cDTA/WT (right) mice. Stimulus definition and ethogram description as in Fig. 7A. For behavior ethograms, see Supplemental Fig. S9. B: cumulative probability plots of freezing (left), fleeing (middle), and nest entry (right) for Brn3b^cDTA/WT (blue, 4 trials/mouse, n = 9 mice), Opn4^Cre/WT (green, 4 trials/mouse, n = 9 mice), and Opn4^Cre/WT; Brn3b^cDTA/WT (red, 4 trials/mouse, n = 11 mice). C: box-whisker plots for freeze reaction times. D: behavior frequency during stimulus, for Brn3b^cDTA/WT (blue), Opn4^Cre/WT (green), and Opn4^Cre/WT; Brn3b^cDTA/WT (red). Data are normalized to total number of trials for each genotype. For C, KS2 test was used, and for D, χ² test was used to determine significance. For all graphs: *P < 0.05.

Fig. 10.

Modest shifts in sweeping-evoked freeze and flight responses in Brn3c^Cre/WT; Brn3b^cDTA/WT mice. A: ethogram of log (mouse speed) across looming response trials for Brn3c^Cre/WT (left) and Brn3c^Cre/WT; Brn3b^cDTA/WT (right) mice. Stimulus definition and ethogram description as in Fig. 8A. For behavior ethograms, see Supplemental Fig. S9. B: cumulative probability plots of freezing (left), fleeing (middle), and nest entry (right), for Brn3c^Cre/WT (orange, 4 trials/mouse, n = 7 mice) and Brn3c^Cre/WT; Brn3b^cDTA/WT (black, 4 trials/mouse, n = 10 mice). C: behavior frequency during stimulus for Brn3c^Cre/WT (orange) and Brn3c^Cre/WT; Brn3b^cDTA/WT (black) mice. Data are normalized to total number of trials for each genotype. D: box-whisker plots for freeze reaction times. For D, KS2 test was used, and for C, χ² test was used to determine significance. For all graphs: *P < 0.05.