A whole-brain connectivity map of mouse insular cortex

Abstract

1The insular cortex (IC) plays key roles in emotional and regulatory brain functions and is 2 affected across psychiatric diseases. However, the brain-wide connections of the mouse IC have 3 not been comprehensively mapped. Here we traced the whole-brain inputs and outputs of the 4 mouse IC across its rostro-caudal extent. We employed cell-type specific monosynaptic rabies 5 virus tracings to characterize afferent connections onto either excitatory or inhibitory IC neurons, 6 and adeno-associated viral tracings to label excitatory efferent axons. While the connectivity 7 between the IC and other cortical regions was highly reciprocal, the IC connectivity with 8 subcortical structures was often unidirectional, revealing prominent top-down and bottom-up 9pathways. The posterior and medial IC exhibited resembling connectivity patterns, while the 10 anterior IC connectivity was distinct, suggesting two major functional compartments. Our results 11 provide insights into the anatomical architecture of the mouse IC and thus a structural basis to 12 guide investigations into its complex functions. 13 50 subdivisions for the two major neuronal subclasses that is excitatory pyramidal neurons and 51 inhibitory interneurons. 52 Results 53 4 Viral tracing approach to reveal the input-output connectivity of the mouse 54 IC 55To map the connectivity of the entire mouse IC, we injected viral tracers into three evenly spaced 56 locations along the rostro-caudal axis with the aim of comprehensively tracing from its entire 57 extent and to assess possible parcellation of the mouse IC into connectivity-based subdomains. 58The most anterior region, aIC ranged from +2.45 mm to +1.20 mm from Bregma; the medial 59 part, mIC, from +1.20 mm to +0.01 mm from Bregma, and the posterior part, pIC, from +0.01 60 mm to -1.22 mm from Bregma (see also Fig. 1c). 61 In order to trace the monosynaptic inputs to the IC we utilized a modified SAD∆G-eGFP(EnvA) 62 rabies virus (RV), which has been shown to label monosynaptic inputs to selected starter cells 63 with high specificity (Wall, Wickersham, Cetin, De La Parra, & Callaway, 2010; Wickersham, 64 Finke, Conzelmann, & Callaway, 2007). This virus lacks the genes coding for the rabies virus 65 glycoprotein (G) and is pseudotyped with the avian viral envelope EnvA. This restricts its 66 infection to neurons expressing the avian TVA receptor and to monosynaptic retrograde infection 67of afferents ( Fig. 1a). We infected the IC of CamKIIα-Cre and GAD2-Cre expressing mouse 68 lines to specifically target TVA and rabies virus glycoprotein expression to excitatory pyramidal 69 or inhibitory interneurons, respectively (see Fig. 1a and Methods). 70 In order to trace and quantify the axonal projections (outputs) of the IC, we injected Cre-71 dependent adeno-associated virus (AAV2/5-DIO-eYFP) into CamKIIα-Cre and GAD2-Cre 72 transgenic mice (see Fig 1b and Methods). We did not observe long-range projections from IC 73 GAD2-Cre tracings (data not shown). Therefore, we here only present outputs from exc...

Cellular Anatomy of the Mouse Primary Motor Cortex

“…16 , second row), ipsilateral thalamic VAL, and claustrum. Altogether, these data suggest that input sources to Cre- and target-defined MOp-ul neuron populations are similar, consistent with other recent findings that global input patterns are independent of starter cell type 47–49 . Nevertheless, axonal inputs to the MOp-ul arising from different cortical and thalamic regions display clear laminar preferences ( Extended data Fig.…”

Brain‐wide mapping of presynaptic inputs to basolateral amygdala neurons

“…We then analyzed the distribution of input cells, which were found in various brain regions: the medial orbital cortex (MO), the medial prefrontal cortex (mPFC), the insular cortex (Ai), the claustrum (Cl), the piriform cortex (Pir), the basal forebrain (BF), the cortex-amygdala transition (CxA), the midline thalamus (MT), the cortical amygdala (CoA), the entorhinal cortex (Ent), the perirhinal cortex (PRh), the ectorhinal cortex (Ect), the auditory cortex (AuC), the auditory thalamus (AuT), the ventral hippocampus (vHPC), the amygdala piriform transition (APir) and the ventral periaqueductal gray (vPAG) (Figure 5). Notably, RVdG virus infection occurs even in the absence of Cre recombinase activity due to leaky expression of TVA (Gehrlach et al, 2020;Miyamichi et al, 2013). Therefore, as a control experiment, we carried out the same experiments without injection of the retroAAV2-Cre virus to estimate the number of Cre-independent RVdG virus infections.…”

The Posterior Insular Cortex is Necessary for the Consolidation of Tone Fear Conditioning

“…This selective role in memory consolidation of TFC is also supported by anatomical studies showing that the pIC is a candidate for the integration of viscerosensory inputs (Allen, 2020;Livneh et al, 2020) as well as for the convergence of both conditioned stimulus (CS, i.e., tone) and unconditioned Stimulus (US, i.e., footshock), resulting from multisensory afferences of thalamic nuclei, as well as from the lateral and basolateral nuclei of the amygdala (Gehrlach et al, 2020;McDonald & Jackson, 1987;Shi & Cassell, 1998a;. Although there seems to be a consensus in the literature that the IC is not necessary for the acquisition of fear conditioning -since manipulations before training have shown no effect (Brunzell & Kim, 2001;Gehrlach et al, 2019;Lanuza et al, 2004) -it is possible that the pIC is preferably recruited to consolidate and maintain the association between a discrete stimulus (i.e., tone) and the US, which would explain the selective effect observed with post-training inactivation of the pIC in our study.…”

“…Traditionally, two major IC subdivisions -an anterior (rostral, aIC) and a posterior (caudal, pIC) -have been described based on their thalamic and amygdalar connections (McDonald, 1998;McDonald et al, 1999;Nieuwenhuys, 2012;Shi & Cassell, 1998b, 1998avan der Kooy et al, 1984). Recently, a comprehensive viral tracing study in mice has also highlighted the existence of two IC clusters characterized by a markedly distinct connectivity profile along the antero-posterior IC axis (Gehrlach et al, 2020). The different cytoarchitectural and connectivity patterns of the aIC and pIC warrant further studies on a possible functional dissociation at the IC subregion level (Gehrlach et al, 2020;Gogolla, 2017).…”

“…Recently, a comprehensive viral tracing study in mice has also highlighted the existence of two IC clusters characterized by a markedly distinct connectivity profile along the antero-posterior IC axis (Gehrlach et al, 2020). The different cytoarchitectural and connectivity patterns of the aIC and pIC warrant further studies on a possible functional dissociation at the IC subregion level (Gehrlach et al, 2020;Gogolla, 2017). Some evidences suggest a functional heterogeneity of the IC regarding anxiety (Méndez-Ruette et al, 2019) and some memory processes, such as the acquisition of conditioned taste aversion and spatial memory (Nerad et al, 1996).…”