Persistent enhancement of basolateral amygdala-dorsomedial striatum synapses causes compulsive-like behaviors in mice

BLA (Basolateral amygdalar nucleus) MBA:295 projects to CP (Caudoputamen) MBA:672 · Mouse

Anterograde viral tracing from the BLA (AAV-tdTomato / AAV-ChR2-EYFP) produced labelled axon terminals in the dorsomedial striatum; the terminal field formed a patch-like pattern overlapping mu-opioid-receptor-positive striosomes and overlapped the somata of D1-expressing medium spiny neurons much more than those of D2 neurons (657 D1 cells from 5 mice vs 276 D2 cells from 4 mice).

Fig. 2a (bottom), Fig. 2h,i

Made by an AI model reading the paper (claude-opus-5, extract@0.3.0); a second AI model (claude-opus-5) agrees.

BLA (Basolateral amygdalar nucleus) MBA:295 functionally connects to CP (Caudoputamen) MBA:672 · Mouse

Optical stimulation of BLA axons also evoked excitatory postsynaptic currents in putative D2 medium spiny neurons of the dorsomedial striatum (6 cells from 6 mice), but these responses were significantly smaller than in D1 neurons (5 cells from 5 mice) at every light intensity tested.

Fig. 2k

Made by an AI model reading the paper (claude-opus-5, extract@0.3.0); a second AI model (claude-opus-5) agrees.

BLA (Basolateral amygdalar nucleus) MBA:295 projects to ACB (Nucleus accumbens) MBA:56 · Mouse

Two different retrograde dyes injected separately into the dorsomedial striatum and the nucleus accumbens each labelled BLA neurons, but only a few BLA cells were double-labelled, indicating that accumbens-projecting and striatum-projecting BLA populations are largely separate.

Fig. 3e

Made by an AI model reading the paper (claude-opus-5, extract@0.3.0); a second AI model (claude-opus-5) agrees.

BLA (Basolateral amygdalar nucleus) MBA:295 projects to CP (Caudoputamen) MBA:672 · Mouse

Retrograde tracer (Retrobeads/CTB) placed in the dorsomedial striatum labelled neurons in the basolateral amygdala, confirming an amygdalostriatal projection; many of these retrogradely labelled cells, especially in the anterior BLA, also expressed c-Fos after predator stress.

Fig. 2a (top), Fig. 2b,d,e; Fig. 3e

Made by an AI model reading the paper (claude-opus-5, extract@0.3.0); a second AI model (claude-opus-5) agrees.

BLA (Basolateral amygdalar nucleus) MBA:295 functionally connects to CP (Caudoputamen) MBA:672 · Mouse

Light activation of ChR2-expressing BLA axons evoked excitatory postsynaptic currents in D1-expressing medium spiny neurons of the dorsomedial striatum, which were larger than those in D2 neurons across stimulation intensities; the AMPA/NMDA ratio of these BLA-evoked currents in D1 neurons increased after chronic circuit activation (9 cells from 3 control mice vs 5 cells from 4 CNO-treated mice).

Fig. 2j,k; Fig. 6a-c

Made by an AI model reading the paper (claude-opus-5, extract@0.3.0); a second AI model (claude-opus-5) agrees.

BLA (Basolateral amygdalar nucleus) MBA:295 projects to ACB (Nucleus accumbens) MBA:56 · Mouse

Anterograde tracing from the BLA showed labelled fibres in the nucleus accumbens, with a more widespread terminal field there than in the dorsomedial striatum.

Fig. 2a; Results (BLA-NAc section)

Made by an AI model reading the paper (claude-opus-5, extract@0.3.0); a second AI model (claude-opus-5) is unsure. “The text states BLA inputs are more widespread in NAc than DMS referring to Fig. 2a, but the figure caption only describes amygdalostriatal tracing, so the NAc labelling cannot be confirmed from this text.”

BLA (Basolateral amygdalar nucleus) MBA:295 synapses onto CP (Caudoputamen) MBA:672 · Mouse

An anterogradely transsynaptic AAV2/1-Flpo injected into the BLA, combined with a Flp-dependent reporter in the dorsomedial striatum, labelled numerous striatal (D1) neurons, indicating transsynaptic transfer across BLA-to-DMS synapses; restricting inhibitory DREADD expression to these BLA-innervated D1 neurons blocked the behavioural effects of BLA terminal stimulation.

Fig. 3h,i

Made by an AI model reading the paper (claude-opus-5, extract@0.3.0); a second AI model (claude-opus-5) agrees.