Dorsal Striatal CINs Receive Monosynaptic Inputs From the CeA and BNST

A, Schematic illustrating the infusion of helper viruses (AAV-DIO-TVA-mCherry and AAV-DIO-RG) and rabies-GFP (RV-GFP) into the dorsal striatum of ChAT-Cre;D1-tdTomato mice. The rabies-GFP was infused 3 weeks after the helper virus infusion, and animals were euthanized one week after the rabies infusion. B, Model for the retrograde transsynaptic labeling of CeA and BNST neurons projecting to dorsal striatal CINs. TVA permits selective infection by the rabies-GFP virus, while RG mediates the retrograde transsynaptic jump from postsynaptic CINs to presynaptic terminals of CeA or BNST neurons. C, Sample image showing CeA neurons projecting to dorsal striatal CINs, as indicated by rabies-GFP expression. AP: -1.50 mm from Bregma. TS, tail of the striatum. Scale bar: 0.5 mm, 50 μm for insert. D, Sample image demonstrating that BNST neurons project to dorsal striatal CINs, as indicated by rabies-GFP expression. Note that the injection site (inj. site) is also displayed in the dorsal striatum. AP: -0.26 mm from Bregma. DS, dorsal striatum; ac, anterior commissure. Scale bar: 0.5 mm, 50 μm for insert. E, The CeA and BNST both send projections to the dorsal striatum, with the BNST providing more inputs. ***p < 0.001 by Mann-Whitney test. n = 13 sections from 4 mice (13/4) for the CeA and 18/4 for the BNST.

CRF-Positive Neurons Are Abundant in the CeA and BNST but Absent in the Dorsal Striatum

A, B, Confocal images showing dense populations of tdTomato-labeled CRF+ neurons in the CeA (A) and BNST (B) of CRF-Cre;tdTomato rats. AP: -1.5 mm from Bregma (A) and -0.26 mm (B). Scale bars: 1 mm (left) and 50 μm (right). DS, dorsal striatum. C, Representative image of the dorsal striatum showing the presence of CRF+ axonal fibers but the absence of CRF+ cell bodies. AP: -0.26 mm from Bregma. Scale bar: 100 μm. D, The CeA and BNST contain more CRF+ neurons than the dorsal striatum. *p < 0.05 by Kruskal-Wallis with Dunn’s Method. n = 20 slices from 7 rats (20/7) for the CeA, 29/7 for the striatum, and 15/7 for the BNST.

CINs Receive Monosynaptic Input From CeA and BNST CRF+ Neurons

A, Schematic showing virus injection (AAVretro-DIO-GFP) into the dorsal striatum of CRF-Cre;tdTomato rats. B, Representative image showing the injection site of AAVretro-DIO-GFP in the posterior dorsal striatum of CRF-Cre;tdTomato rats. Overlapping tdTomato+ (CRF+) and dorsal striatum-derived Cre-driven GFP+ expression in the BNST (depicted in yellow) indicates that these CRF+ neurons project to the dorsal striatum. ac, anterior commissure; LV, left ventricle; tdT, tdTomato; DS-proj, dorsal-striatum-projecting. Scale bar: 0.5 mm (left), 10 μm (right). C, Images showing the overlap (in yellow) of tdTomato+ (CRF+) and dorsal striatum-derived Cre-driven GFP+ expression in the CeA, confirming that these CRF+ neurons also project to the dorsal striatum. TS, the tail of the striatum. Scale bar: 1 mm (left), 10 μm (right). D, The CeA contains more CRF+ neurons than the BNST. ***p < 0.001 by unpaired t test, n = 9 sections from 5 rats (9/5) for the CeA and 8/5 for the BNST. E, The proportion of CRF+ neurons projecting to the dorsal striatum is higher in the BNST compared to the CeA. ***p < 0.001 by unpaired t test, n = 9/5 (CeA) and 8/5 (BNST). F, The total number of dorsal striatum-projecting CRF+ neurons does not differ between these two regions. n.s., not significant, p > 0.05. n = 9/5 (CeA) and 8/5 (BNST). G, Schematic showing the injection of AAV-FLEX-Chrimson-tdTomato into the CeA and BNST of CRF-Cre;ChAT-eGFP mice and the subsequent recording of green striatal CINs during blue light stimulation of surrounding Chrimson-containing fibers from CRF+ neurons. Fibers were stimulated at a wavelength of 590 nm for 2 ms. H, Sample traces showing CIN responses to blue light stimulation of CRF fibers, which were abolished by TTX, recovered with TTX+4-AP and further eliminated by bicuculline (Bic). I, Summary of oIPSC data showing the disappearance of oIPSCs with TTX and reappearance with TTX and 4-AP. *p < 0.05 by one-way ANOVA. n = 4 slices.

CRFR1 is Expressed in Striatal CINs

A, Representative image of a coronal striatal section from a CRFR1-Cre-tdTomato rat showing CRFR1 expression overlapping with anti-ChAT immunoreactivity. The section was stained using an anti-ChAT antibody and imaged using a far-red (647 nm) wavelength (white). Scale bars: 1 mm (left), 100 μm (middle), and 20 μm (right). B, There are significantly fewer CRFR1+ CINs than the total number of CINs in the dorsal striatum. ***p < 0.001 by Mann-Whitney test. n = 8 sections from 2 rats. C, Around 30% of CINs express CRFR1. n = 8 sections from 2 rats.

CRF Enhances CIN Activity and ACh Release in the Dorsal Striatum

A, Sample images of GFP-labeled CINs in the striatum of a ChAT-eGFP mouse. Scale bar: 0.5 mm, 50 μm (inset). B, Bath application of CRF (100 nM) increased the spontaneous firing of dorsal striatal CINs in cell-attached electrophysiological recordings. This effect was prevented by pretreatment with the CRFR1 antagonist NBI 35695 (5 μM). n = 7 cells from 3 mice (7/3) for CRF and 6/3 for CRF plus antagonist recordings. C, Data showing that CRF significantly increases the firing from baseline and the firing frequency in the presence of CRF following pretreatment with CRFR1 antagonist is significantly lower. ***p < 0.001 by paired t-test, ***p < 0.001 by Mann Whitney test. n = 7 cells from 3 mice. D, Sample images of ACh sensor fluorescence in dorsal striatal slices before and during bath application of CRF (100 nM). AAV-GRABACh4m was infused into the dorsal striatum of wildtype mice, and live-tissue confocal imaging was conducted two weeks post-infusion. Scale bar: 10 μm for left and right. E, Sample trace of spontaneous ACh release events (indicated by red arrows, top). Bath application of CRF increased ACh sensor fluorescence (bottom). F, Summary data showing a significant increase in ACh sensor fluorescence following CRF application. *p < 0.05 by Mann-Whitney test. n = 7 slices from 7 mice. G, Schematic of cell-attached electrophysiological recordings from dorsal striatal CINs in CRF-Cre;Ai32;ChAT-eGFP mice, with simultaneous optogenetic stimulation of CRF+ fibers using blue light (470-nm, 2 ms, 50 Hz, 60 sec). H, Sample trace showing the increase in CIN firing frequency during blue light stimulation of CRF+ fibers. I, Data demonstrating a significant and reversible increase in CIN firing frequency during optogenetic stimulation, *p < 0.05, ***p < 0.001 by one-way RM ANOVA. n = 13 neurons from 8 animals.

Acute Alcohol Exposure Attenuates CRF-Induced Enhancement of CIN Activity

A, Time course of spontaneous firing in CINs from dorsal striatal slices of ChAT-eGFP mice, before, during, and after bath application of alcohol (EtOH; 50 mM). B, Summary data demonstrates a significant reduction in CIN firing frequency following alcohol application. **p < 0.01 by Mann-Whitney test. n = 6 cells from 4 mice. C, Schematic of the experimental design in which striatal slices were pretreated with alcohol (50 mM) for 1 h in an incubation chamber and then washed for 15 min in the recording chamber. CINs were selected for cell-attached electrophysiological recordings, measuring firing frequency for 10 min (baseline), followed by CRF (40 nM) and alcohol (50 mM) bath applications for 10 min each. D, E, Data showing spontaneous firing of CINs during baseline (BL), CRF, and alcohol bath application for the control (D) and alcohol pretreated (E) groups. *p < 0.05, **p < 0.01, ***p < 0.001 by one-way RM ANOVA. n = 10 cells from 8 mice (10/8) for the control group and 10/7 for the alcohol pretreated group. F, Data showing that alcohol pretreatment attenuated CRF-induced enhancement of CIN firing frequency. *p < 0.05 by unpaired t-test. n = 10 cells from 8 mice (10/8) for the control group and 10/7 for the pretreated group.