Mixed representations of choice direction and outcome by GABA/glutamate cotransmitting neurons in the entopeduncular nucleus
- Department of Anatomy and Neurobiology, Boston University Chobanian & Avedisian School of Medicine, United States
- Howard Hughes Medical Institute, Department of Neurobiology, Harvard Medical School, United States
Figures
Figure 1 with 2 supplements
Mice alter choices to changing reward probabilities in a probabilistic switching task.
(A) Illustration of the animal movements and epochs (Trial Start, Choice, and Evaluation) of a single trial in the probabilistic two-port choice task. Yellow color of port(s) indicates when LED is …
Figure 1—figure supplement 1
Behavioral data showing individual animals.
p(switch) across different trial histories, and additional behavioral metrics. (A) The probability of choosing the highly rewarded port p(high port, 90/10 rew. prob.) around a block transition …
Figure 1—figure supplement 2
Behavioral modeling using a recursively formulated logistic regression (RFLR).
(A) The RFLR model, which calculates the log odds of the mouse’s next choice (Ψt+1) given its most recent choice (ct) and a series of prior choices and rewards. ct represents choice, rt represents …
Figure 2 with 2 supplements
Neural activity in EPSst+ neurons encodes both choice and value.
(A) Viral injection location for specific infection of EPSst+ neurons with GCaMP6f in a Sst-Flp mouse line and fiber implant location for photometry recording. (B) Fiber photometry recording of EPSst…
Figure 2—figure supplement 1
Alignment and sorting of photometry signals from EPSst+ neurons to different behavioral events.
(A) Fiber photometry recording of EPSst+ neurons for individual trials during a behavioral session. Trials are aligned to center port entry (CE) and red dots indicated side port entry (SE). Trials …
Figure 2—figure supplement 2
Examining the influence of reward history on photometry signals from EPSst+ neurons.
(A) (left) Ipsilateral rewarded trial-averaged (± SEM) photometry signals across reward probabilities aligned to side exit (SX) plotted to examine if reward expectation impacts photometry signals. …
Figure 3 with 2 supplements
Generalized linear model of EPSst neural activity during behavior.
(A) GLM workflow: behavioral variables are convolved with their kernels. Each time shift in the kernel consists of an independent β coefficient fit jointly by minimizing a cost function. The …
Figure 3—figure supplement 1
Generalized linear model (GLM) reconstructions of photometry signal with a partial model.
(A) Average original (black) and reconstructed (green, using model where direction and reward were omitted) photometry signals across trials aligned to behavioral events used in the full model …
Figure 3—figure supplement 2
Generalized linear model (GLM) reconstructions of photometry signal with choice and reward variables omitted.
(A) Average original (black) and reconstructed (green, using model where direction and reward were omitted) photometry signals across trials aligned to behavioral events used in the partial model …
Figure 4 with 3 supplements
Effects of permanent genetic silencing of synaptic release from EPSst+ neurons on continued performance of a two-port choice probabilistic switching task.
(A) Viral injection location resulting in Cre-dependent expression of GFP (control) or tetanus toxin in EPSst+ neurons (green = Tettx GFP, gray = DAPI). (B) The probability of choosing the highly …
Figure 4—figure supplement 1
Additional behavioral performance metrics before and after viral injection and electrophysiological validation of Tettx effects on GABA/glutamate cotransmission from EPSst+ neurons.
(A) Trial duration (normalized to mean before adeno-associated virus (AAV) injection day- 5 - day-1, red dashed line) across behavioral sessions (days) before and after AAV injection of GFP (black) …
Figure 4—figure supplement 2
Body weight and total number of trials per session before and after viral injection.
(A) Body weight (normalized to mean before adeno-associated virus, AAV injection) across behavioral sessions (days) before and after AAV injection of GFP (black) or Tettx (gray) (n=6 GFP, n=6 Tettx …
Figure 4—figure supplement 3
Psychometric curves of port choice before and after adeno-associated virus (AAV) injection.
(A) The probability that an animal chooses the left port (p(left port chosen)) compared to log-odds (Ψt+1) of the animal’s next choice calculated by the recursively formulated logistic regression …
Figure 5 with 3 supplements
Effects of CRISPR Cas9 deletion of synaptic glutamate release from EPSst+ neurons on continued performance of a two-port choice probabilistic switching task.
(A) Viral injection location resulting in Cre-dependent expression of oChief-tdTom +SaCas9-sgRNA for ROSA26 (control) or Slc17a6 (vGluT2) in EPSst+ neurons (red = tdTomato, gray = DAPI). (B) The …
Figure 5—figure supplement 1<bold>.</bold>
Additional behavioral performance metrics before and after viral injection and electrophysiological validation of CRISPR-SaCas9 mediated deletion of Slc17a6 (vGluT2) on GABA/glutamate cotransmission from EPSst+ neurons.
(A) Trial duration normalized to mean before adeno-associated virus (AAV) injection (day-5- day-1), red dashed line across behavioral sessions (days) before and after AAV injection of sgROSA26 …
Figure 5—figure supplement 2
Total number of trials per session and animal body weight changes before and after viral injection.
(A) Body weight (normalized to mean before adeno-associated virus, AAV injection) across behavioral sessions (days) before and after AAV injection of sgROSA26 (black) or sgSlc17a6 (gray) (n=10 …
Figure 5—figure supplement 3
Psychometric curves of port choice before and after adeno-associated virus (AAV) injection.
(A) The probability that an animal chooses the left port (p(left port chosen)) compared to log-odds (Ψt+1) of the animal’s next choice calculated by the recursively formulated logistic regression …
Tables
Key resources table
| Reagent type (species) or resource | Designation | Source or reference | Identifiers | Additional information |
|---|---|---|---|---|
| Strain (Mus musculus) | C57Bl/6 J | The Jackson Laboratory | Cat# JAX:000664 RRID:IMSR_JAX:000664 | |
| Genetic reagent (Mus musculus) | Sst-IRES-Cre | The Jackson Laboratory | Cat# JAX: 013044 RRID:IMSR_JAX:013044 | |
| Genetic reagent (Mus musculus) | Sst-IRES-Flpo | The Jackson Laboratory | Cat# JAX: 031629 RRID:IMSR_JAX:031629 | |
| Genetic reagent (Mus musculus) | Pvalb-2A-Flp | The Jackson Laboratory | Cat# JAX: 022730 RRID:IMSR_JAX:022730 | |
| Genetic reagent (AAV) | AAV1-Syn-FLEX-GCaMP6f | Addgene | Addgene# 100833 RRID:Addgene_100833 | |
| Genetic reagent (AAV) | AAV8-CAG-FLEX-tdTomato | Addgene | Addgene# 51503 RRID:Addgene_51503 | |
| Genetic reagent (AAV) | AAVrg-Ef1a-fDIO-Cre | Addgene | Addgene# 121675 RRID:Addgene_121675 | |
| Genetic reagent (AAV) | AAV8-Syn-FLEX-TeLC-P2A-GFP | Addgene | Addgene# 135391 RRID:Addgene_135391 | |
| Genetic reagent (AAV) | AAV8-Syn-DIO-EGFP | Addgene | Addgene# 50457 RRID:Addgene_50457 | |
| Genetic reagent (AAV) | AAV1-CMV-FLEX-SaCas9-sgSlc17a6 | Addgene | Addgene# 124847 RRID:Addgene_124847 | |
| Genetic reagent (AAV) | AAV1- CMV-FLEX-SaCas9-sgROSA26 | Addgene | Addgene# 159914 RRID:Addgene_159914 | |
| Genetic reagent (AAV) | AAV8-Ef1a-DIO-oChief-tdTomato | Addgene | Addgene# 51094 RRID:Addgene_51094 | |
| Antibody | anti-GFP (chicken polyclonal) | Invitrogen | Cat# A10262 RRID:AB_2534023 | IF(1:1000) |
| Antibody | anti-mcherry (rabbit polyclonal) | Abcam | Cat# Ab167453 RRID:AB_2571870 | IF(1:500) |
| Software, algorithm | MATLAB (R2015a) | MathWorks | RRID:SCR_001622 | |
| Software, algorithm | Fiji | PMID:22743772 | RRID:SCR_002285 | https://imagej.net/Fiji |
| Software, algorithm | Python | https://www.python.org/ | RRID:SCR_008394 | https://www.python.org/ |
