Optogenetic Stimulation of the Basolateral Amygdala Increased Theta-Modulated Gamma Oscillations in the Hippocampus

BLA (basal amygdaloid nucleus) UBERON:0002887 projects to CA1 (Field CA1) UBERON:0003881 · Rat

After AAV5-CaMKII-hChR2-EYFP infusion into the BLA, which labelled cell bodies only in the amygdala, EYFP-positive axons (no labelled somata) were seen in the temporal half of the hippocampus, most notably in the stratum lacunosum-moleculare of CA1.

Results, Histological Verification; Fig. 1D,E

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

BLA (basal amygdaloid nucleus) UBERON:0002887 functionally connects to CA1 (Field CA1) UBERON:0003881 · Rat

Blue-light stimulation of channelrhodopsin-expressing putative glutamatergic BLA neurons in freely moving rats evoked frequency-matched LFP responses in CA1 at 8, 20, 50 and 50/8 Hz, whereas control near-infrared light evoked no response. The CA1 peak followed the first light pulse at 12.7 ms versus 6.67 ms in the BLA (a 6.03 ms lag, interpreted as a monosynaptic BLA-to-CA1 drive); latencies after the final pulse were 14.0, 19.3, 16.0 and 9.33 ms for the 8, 20, 50 and 50/8 Hz conditions. Stimulation at 50 and 50/8 Hz significantly raised CA1 slow gamma power, and only 50/8 Hz stimulation significantly increased CA1 theta-gamma comodulation.

Figs. 2, 3, 4, 5, 6; Results sections on 1-s and 5-s stimulation

Made by an AI model reading the paper (claude-opus-5, extract@0.3.0); a second AI model (claude-opus-5-5) disagrees. “The 12.7 ms value is the CA1 latency to peak from light onset, not a conduction delay; the BLA-to-CA1 lag the paper reports is 6.03 ms.”

BLA (basal amygdaloid nucleus) UBERON:0002887 projects to SUB (Subiculum) UBERON:0002191 · Rat

The same viral injection into the BLA produced fluorophore-labelled fibres in the subiculum of the temporal hippocampus, with labelling confined to axons rather than cell bodies there.

Results, Histological Verification; Fig. 1E

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