Insular cortex process strong feelings as fear or bodily discomfort

The insular cortex is an important, yet almost hidden part of the cerebral cortex. Here, sensory information, bodily states, feelings and emotions come together.

However, how the insular cortex processes this information and how this affects behavior is largely unknown – knowledge that would help us to better understand the mechanisms involved in depression, anxiety and eating disorders, for example.

Nadine Gogolla and her team at the Max Planck Institute of Neurobiology were now able to show how the insular cortex of mice processes such strong feelings as fear or bodily discomfort, and how this affects their behavior.

Feelings and emotions greatly influence our behavior.

“Which is a good thing!” says Nadine Gogolla, Research Group Leader at the Max Planck Institute of Neurobiology.

“For example, if a mouse smells a fox, the feeling of fear causes it to hesitate from exploring its surroundings, and also stops it from eating.”

Negative bodily states such as nausea have a similar inhibitory influence.

Gogolla and her team have now shown that these very different negative feelings and behavioral adjustments are linked via the posterior insular cortex.

First author Daniel Gehrlach and his colleagues in Gogolla’s group discovered that nerve cells of the posterior insular cortex respond to a large number of different sensory information, emotions and bodily states.

All of the information that is processed here has a direct negative effect or acts as an aversive signal for the animal.

Interestingly, individual nerve cells can react to many different negative stimuli such as bitter taste, fear, pain, thirst, or bodily discomfort.

As soon as the cells detect these negative states, they forward the information to the amygdala or nucleus accumbens via two different pathways.

These two brain regions are known to directly regulate an animal’s behavior. “For the first time, we could now demonstrate the influence of the insular cortex on behavior via these two connections,” says Gogolla.

Activation of the neuronal pathway from the insular cortex to the amygdala primarily causes behavioral adjustments to fear: The mouse reduces its food intake, social contacts and exploration of its surroundings. When the researchers suppressed this pathway’s activity, the animals become less afraid.

Activating the pathway to the nucleus accumbens, on the other hand, had a similar effect as an illness – the mice stopped eating. Conversely, by inactivating this pathway, the animals still ate even when being nauseous. “By being able to directly modify and observe neuronal activity in mice, we were able to show specific mechanistic relationships that are an important step towards a true understanding of anxiety, depression and eating disorders,” explains Gogolla.

Both strong negative emotions and the feeling of being sick should induce humans and animals alike to take care of, and protect themselves.

Anxieties and depression arise when negative emotions become too strong or too frequent.

“It is possible that the insular cortex learns from previous experiences, so that the cells react stronger or faster to the next negative impression,” reasons Gogolla.

“Once we have learned to understand such relationships, we may be able to find a way to reverse or at least contain them.”

The insular cortex is a brain region important for regulating emotions, as well as empathy and social behavior.

It lies deeply embedded in the folds of the cerebral cortex of humans, primates and many other mammals.

Since the cerebral cortex in mice, rats and some other smaller mammals is smooth, the insular cortex is more accessible to study in these animal models.

---

The forebrain representation and integration of the physiological condition of the organs and tissues of the body (interoception) shapes perceptual awareness and underlies the neurobiology of subjective feelings (Craig, 2009; Critchley and Garfinkel, 2018).

The neuronal pathway that encodes interoception from the periphery to the spinal cord, brainstem, thalamus, and cerebral cortex does not primarily end in the classical primary somatosensory cortex but in the dorsal fundus of the insular cortex, or “primary interoceptive cortex” (Craig, 2002).

Functional evidence in humans indicates that the primary interoceptive cortex hosts a rather “objective” topographic representation of physiological changes (e.g., linear innocuous cooling or linear transcutaneous histamine concentration), which translates into a “subjective” representation in the anterior insular cortex (AIC) where brain activity tends to correlate with the perceptual report of a sensation rather than with the actual physiological changes itself (Craig et al., 2000; Drzezga et al., 2001; Craig, 2009).

In fact the activity of the human AIC strongly correlates with a vast range of subjective activities, including subjective bodily sensations (Mutschler et al., 2009), emotional feelings [e.g., (Bartels and Zeki, 2004; Gu et al., 2013; Smith et al., 2015)], empathy (Lamm and Singer, 2010), as well as more complex perceptions, such as, the recognition of oneself in a mirror (Devue and Bredart, 2011) and other visual perceptual tasks (Salomon et al., 2016), temporal discrimination (Pastor et al., 2004), and intention forming (Brass and Haggard, 2010).

The AIC, together with the anterior cingulate (ACC), is also the cortical region with the highest concentration of the specialized von Economo (VEN) and Fork (FN) neurons (Allman et al., 2010). Both neurons are selectively depleted in the early stage of the behavioral variant of the frontotemporal dementia (bvFTD), which is characterized by a subtle loss of self-conscious feelings (Kim et al., 2011; Santillo and Englund, 2014; Nana et al., 2019).

Unsurprisingly, the human AIC and ACC are the most common targets of psychiatric disorders (Nagai et al., 2007; Seeley, 2008; Goodkind et al., 2015). Furthermore, a single-case clinical report indicated that microstimulation of the AIC/claustrum region can immediately alter states of consciousness (Koubeissi et al., 2014) and its functional connectivity with the arousal centers of the brainstem has been shown to be massively disrupted in coma (Fischer et al., 2016).

More recently, the AIC was also shown to be temporally the last commonly active brain region in response to functionally distinct sensory stimuli (words, touch, and pain) during the anesthetic induction of a complete loss of behavioral response (Warnaby et al., 2016), suggesting a role in gating salient information in their access to conscious behavior.

Although its exact role still has to be elucidated, the insula, and in particular its anterior part, are likely to have a crucial role in functions that directly relate to awareness. In order to unravel fundamental organizational principles, our laboratory recently initiated a vast series of neuroanatomical, functional, and molecular examinations of the macaque monkey insula. The present review incorporates findings from our and other laboratories into a novel model of the anatomical and functional organization of the insular cortex, upon a backdrop of refined architectonic parcellation.

Macroscopic Organization

In anthropoid primates, the insular cortex constitutes a separate cortical lobe, located on the lateral aspect of the forebrain, in the depth of the Sylvian or lateral fissure (LF) (Figure 1; Reil, 1809; Retzius, 1902; Naidich et al., 2004; Afif and Mertens, 2010; Gonzalez-Arnay et al., 2017). (See also Naidich in this special issue).

It is adjoined anteriorly by the orbital prefrontal cortex, and it is covered dorsally by the frontoparietal operculum and ventrally by the temporal operculum (Figure 1A,B).

The excision of the two opercula and part of the orbital prefrontal cortex reveals the insula proper, delimited by the anterior, superior, and inferior peri-insular (or limiting or circular) sulci (Figure 1A’).

The human insula is gyrencephalic (i.e., convoluted), like the rest of the human neocortex (Figure 1A’).

It is divided into posterior and anterior lobules by the central sulcus of the insula (CS).

There are usually two long (longus) gyri in the posterior lobule (l1 and l2), three short (brevis) anterior gyri in the anterior lobule (s1, s2, and s3), and one accessory gyrus (ac), which is continuous with the ventral transverse gyrus (t).

The orientation and size of the gyri as well as the exact number of short gyri varies across individuals and hemispheres (Wysiadecki et al., 2018).

The accessory gyrus varies in volume and demarcation (Ture et al., 1999; Wysiadecki et al., 2018). When large, it is separated from the anterior short gyrus by a distinct folding (named here “accessory APS”, aAPS) that bifurcates from the actual APS (Ture et al., 1999).

Notably, the localization of the accessory gyrus and transverse gyrus corresponds approximately with the localization of the frontoinsula (FI) (Von Economo and Koskinas, 1925) which was recently defined by the presence of a high concentration of VENs and FNs (von Economo, 1926; Allman et al., 2010).

The macaque insula is almost entirely lissencephalic (i.e., smooth), which complicates establishing homologies solely based on macroscopic criteria (Figure 1B”).

The macaque insula has an incipient horizontal gyrus (or “mound”) ventrally, and, in many cases, a shallow but distinct vertical ridge anteriorly.

The cortical region anterior to this vertical ridge contains a high concentration of VENs and FNs (Evrard et al., 2012), suggesting a partial structural homology of this region with the human FI, and of the vertical ridge per se with the human accessory APS. Unlike in humans, beyond this shallow dimple, there is no distinct APS separating the insula from the orbital prefrontal cortex; they are rather continuous.

---

More information: Daniel A. Gehrlach et al. Aversive state processing in the posterior insular cortex, Nature Neuroscience (2019). DOI: 10.1038/s41593-019-0469-1

Journal information: Current Biology , Nature Neuroscience
Provided by Max Planck Society