The insular cortex (IC) plays key roles in emotional and regulatory brain functions and is affected across psychiatric diseases. However, the brain-wide connections of the mouse IC have not been comprehensively mapped. Here we traced the whole-brain inputs and outputs of the mouse IC across its rostro-caudal extent. We employed cell-type specific monosynaptic rabies virus tracings to characterize afferent connections onto either excitatory or inhibitory IC neurons, and adeno-associated viral tracings to label excitatory efferent axons. While the connectivity between the IC and other cortical regions was highly bidirectional, the IC connectivity with subcortical structures was often unidirectional, revealing prominent cortical-to-subcortical or subcortical-to-cortical pathways. The posterior and medial IC exhibited resembling connectivity patterns, while the anterior IC connectivity was distinct, suggesting two major functional compartments. Our results provide insights into the anatomical architecture of the mouse IC and thus a structural basis to guide investigations into its complex functions.

How the body regulates fear Although fear is important for survival, it is maladaptive if it is either too strong, as in anxiety disorders, or too weak, as in exaggerated risk taking. Working in mice, Klein et al . observed that the insular cortex has an unparalleled dual role in either enhancing or weakening the extinction of fear, depending on the internal fear state of the animal (see the Perspective by Christianson). This insula function helps to maintain fear within a homeostatic range and depends on bodily feedback signals: Fear-induced freezing behavior is associated with a slowed heart rate, which in turn dampens fear-evoked activity of the insular cortex. Two opposite signals, prediction of threat by fear-associated cues and negative feedback signals from the body, are thus integrated within the insular cortex. —PRS

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...

No abstract