Zach Chia, George J. Augustine, Gilad Silberberg
(A) Illustration of the experimental paradigm. Briefly, ChR2 was injected using an anterograde viral vector into the insula (100 nl, 200 nL/min). PV-tdTomato transgenic mice were used to enable unambiguous identification of the claustrum. Mice were sacrificed for experiments after at least 20 days of recovery. Whole-cell patch clamp recordings of claustrum cells and whole field illumination was made to test for postsynaptic response to photostimulation.
(B1) Confocal image of a 250 μm thick slice, showing injection into layer 5 of insula without spillage into claustrum core and shell regions (31 out of 103 injected mice had acceptable injections for these experiments). Close-up into the claustrum identified by PV-rich a region shows YFP-expressing fibers in ipsilateral claustrum (Right) but not contralateral claustrum (Left).
(B2) Image of injection into layer 2/3 of the insula shows that projections are confined within layer 2/3 and do not reach into the claustrum.
(C) Postsynaptic responses were recorded in the ipsilateral claustrum (65/161 recordings, 0/24 in contralateral claustrum) from layer 5 insula projections only (0/24 responses in claustrum after layer 2/3 insula injection).
(D) A subset of recordings was further tested for verifying the existence monosynaptic connections (n = 11). Synaptic responses to photostimulation were abolished following bath application of TTX and recovered upon additional 4AP bath application. Another subset of recordings was tested for excitatory connections (n = 12) via the use NBQX and APV which abolished the postsynaptic response to photostimulation.
(E) Both PV and non-PV neurons received input from the insula. Top: voltage responses to suprathreshold step current injections. Bottom: examples of synaptic responses of simultaneously recorded PV and non-PV neurons to photostimulation of synaptic terminals.
(F–I) Properties of synaptic responses in PV and non-PV neurons to photostimulation of terminals from insula (Top). The same properties were extracted from simultaneous (paired) recordings PV and non-PV neurons (Bottom). Whiskers represent 25-75 percentile ± SD
(F) Latency of response from photostimulation shows that PV cells respond with shorter latency compared to non-PV neurons (p < 0.01, t test), this difference in onset latency is observed also in paired recordings (p < 0.01, paired t test).
(G) EPSP size from photostimulation shows that PV cells have larger EPSP size compared to non-PV neurons (p < 0.01, t test); This same difference in EPSP size is also observed in paired recordings (p < 0.05, paired t test).
(H) Input resistance size of PV cells is smaller than non-PV neurons (p < 0.01, t test); this difference in input resistance is observed also in paired recordings (p < 0.05, paired t test).
(I) The rise time of the response neurons is faster in PV cells compared to non-PV neurons (p < 0.05, t test), this difference in rise time is observed also in paired recordings (p < 0.05, paired t test). Response rates of PV and non-PV neurons in ipsilateral claustrum are comparable (p > 0.05, chi-square test, not shown).