Homeostasis and the hypothalamus

Goals

• To complete our discussion of sleep with a consideration of REM sleep and dreaming.
• To discuss the blood brain barrier and circumventricular organs
  • To discuss vomiting as an example of neural sensing of toxins
• To discuss the anatomy of the hypothalamus
• To discuss the optogenetic method
• To discuss fluid balance
• To discuss hunger and satiety

Topic slide

Claude Bernard (1813 – 1878) was a renowned French physiologist who developed the concept of homeostasis. Bernard was also devoted to improvements in the scientific method. Bernard was a vigorous proponent of vivisection, which led to an estrangement from his family and public controversy. You can read about Bernard and his many scientific accomplishments here.

Karl Deisseroth (b. 1971) received his MD and Ph.D. degrees from Stanford University, where he is currently Professor of Bioengineering and of Psychiatry and Behavioral Sciences. He is widely acknowledged as the principal creator of the optogenetics method.

Blood brain barrier

Homeostatic mechanisms maintain essential physiological processes in the body, such as energy and fluid balances, within working limits. While much of the heavy lifting of homeostatic control is out-sourced to the other organs of the body (e.g., kidneys, liver, etc), the brain maintains overall control. As we will see in this lecture, the hypothalamus plays an outsized role in neural control of homeostasis.

To what signals does the brain respond to initiate control? And how are those signals conveyed to neural tissue? There are two possibilities:

• neural signals conveyed by specialized receptors
• the blood supply

Since the glucose and oxygen needed to support brain metabolism is conveyed by the blood supply, one might expect that blood borne signals might carry information about energy and fluid balance. One peculiar aspect of the circulation of the brain is that capillaries have tight junctions around them that prevent the movement of large molecules (such as peptides) across. This is known as the blood-brain barrier, and it is essential for defending the brain against blood borne pathogens and toxins. So, many signaling molecules are denied entry into the brain.

It is notable that many pharmaceutical agents also cannot pass the blood-brain barrier, and so clever strategies for getting some drugs into the brain must be devised. One such strategy is focused ultrasound, which can be used to temporarily open the blood-brain barrier in particular brain regions.

Circumventricular organs

There are several regions in the brain where the blood supply can be considered outside of the blood-brain barrier. These regions are called circumventricular organs, as they appear at the floor of particular ventricles. Here, the capillaries are fenestrated, meaning that there are holes (or windows) where large molecules can pass. Here are three circumventricular organs that were discussed in class.

• Subfornical organ (SFO) – below the fornix
• Area posterma (AP) – at the floor of the third ventricle
• Organum Vasculosum of the lamina terminalis (OVLT)
  • sounds like something Harry Potter would say

The circumventricular organs are strongly connected to the hypothalamus, and to each other. They permit neurons to sense signaling molecules circulating in the blood directly.

Vomiting

The circumventricular organs are also sensitive to circulating toxins carried in the blood. I used the initiation of vomiting as an example.

Two signal sources that originate in the gut converge on the chemoreceptive trigger zone (CTZ), a brain area that serves as the central pattern generator for vomiting (a veritable ballet of muscular contractions).

• Vagus nerve (Xth cranial nerve) input from the gut terminating in the nucleus of the solitary tract.
• Neurons in the area postrema receiving input from the blood.

Output from the AP triggers the CTZ and vomiting ensues.

However, we get nauseous from signals that originate from the environment. Functional MRI studies have shown that the amygdala and insula (two areas we discussed previously with respect to disgust) also converge on the AP and can cause vomiting. It is likely that the conscious experience of nausea also involves the insula.

We also get nauseous and vomit when experiencing motion sickness. This is an evolutionary curiosity – how is motion sickness connected to toxins we ingest? The best explanation seems to be that many poisons (such as alcohol) affect the vestibular system and make us dizzy or unsteady. It is thought that vestibular signals, thus, trigger the vomiting reflex and attendant nausea.

I showed several fMRI studies to illustrate the points above.

Anatomy of the hypothalamus

The hypothalamus is part of the diencephalon that sits below the thalamus and along the third ventricle. There are 11 major nuclei of the hypothalamus that are usually broken into groups. You can read a good summary of the hypothalamus that includes nice animations here.

One interesting aspect of the hypothalamus is that it is strongly and reciprocally connected to the amygdala (via the stria terminalis) and the hippocampus (via the fimbria/fornix pathway). We will focus on these connections in an upcoming lecture on stress.

Optogenetics

Optogenetics is a method that improves upon electrical brain stimulation. One can place an anode and cathode into a brain region and stimulate the neurons in-between. The greatest current densities will occur at the electrodes themselves, with the path and density of the current between the anode and cathode determined by the relative impedances of the tissues and the voltage drop between the anode and cathode.

All of the surrounding neurons will be affected by the electrical stimulation. But what if the neurons represent a heterogenous mixture of neurons – for example, some that excite a particular behavior, and some that inhibit that behavior.

Optogenetics makes use of phenomena with which we are already familiar – the transduction of light into neuronal excitation or inhibition. Recall how rhodopsin in the photopigment of rods in the retina responds to light by initiating a cascade of (g-protein coupled) events that lead to the opening of ionic channels. In a sense, a photon of light acts like a neurotransmitter opening a receptor in a neuron’s membrane.

By clever techniques borrowed from gene therapy, genes that encode for different light-activated processes are borrowed from algae and inserted into the DNA of neurons using viruses. Channel rhodopsin, for example, is coded by a gene from algae. Once expressed in the neuron, it creates a channel in the membrane that responds to light – essentially, an ionotropic channel. Other types of genes can have different effects on the neurons – some excitatory and some inhibitory.

Once inserted, stimulating the neuron(s) by laser light of particular frequencies can activate these different opsins and excite or inhibit neurons. This can then activate pathways and behaviors associated with those neurons.

The key advance that differentiates optogenetics from electrical stimulation is that the genes can be inserted into different groups of neurons that are intermingled in the brain regions of interest. Thus, the light stimulus may only activate neurons that are engaged in one process (eating) and not another process (satiety).

These are several strategies involved in targeting the appropriate neurons – some of which take advantage of the specificity of the laser light (i.e., it is delivered only to a small region of a nucleus, or even to a single neuron), or over the virus vector used to insert the gene (e.g., the viral coat may allow insertion only into axons rather than neuronal cell bodies). The molecular tricks for targeting genes is well beyond the scope of this course (and your instructor), but the degree of specificity for targeting has been remarkable.

Optogenetics is being used to revolutionize the study of homeostatic mechanisms in the brain, and so that is why it is introduced here. As it is an invasive technique that uses viral vectors, the primary work is being done in rodents. Happily, with regard to homeostasis, the human brain has a lot in common with the mouse brain.

The video from Nature embedded below provides a nice summary of the optogenetics method.

Fluid balance: Drinking

There are two components to fluid balance:

• Maintain osmolarity between inside and outside of cells (isotonicity)
  • You don't want your cells shrinking or swelling
• Maintain blood volume within a tight balance
• Hypervolemia
  • Electrolyte balance is off
• Hypovolemia
  • 20% reduction results in shock

The hypothalamus receives signals about two processes:

• sodium concentration in the blood
• blood volume
  • Kidneys detect hypovolemia and activate the renin angiotensin system.
  • Angiotensin II travels in the blood to subfornical organ (SFO).
  • Initiates drinking behavior and salt preference.
  • Paraventricular n. of hypothalamus releases vasopressin (AVP) to posterior pituitary which causes kidneys to reabsorb water

I discussed in detail a paper by Oka and colleagues (Nature, 2015) that illustrated how the SFO contains two distinct populations of neurons: one of which stimulates immediate drinking, and one that causes drinking to immediately cease. The Oka et al. Nature 2015 paper can be found here.

I also discussed a paper by Zimmerman and colleagues (Nature, 2016) that discusses the anticipatory control of drinking.

From Zimmerman et al. Nature 2016

Deviations of blood volume or osmolarity from their set points are detected by specialized neurons within the circumventricular organs (CVOs) of the brain… Activation of these neurons generates thirst, which motivates animals to find and consume water and thereby restore fluid balance.

Zimmerman et al (2016) notes that most drinking is anticipatory .

…regulation precedes rather than responds to changes in the blood… For example, drinking quenches thirst tens of minutes before ingested water reaches the circulation and alters the composition of the blood, indicating that thirst is sated before homeostasis is restored. Yet animals somehow calibrate their water consumption to match their physiological need precisely… most drinking occurs during meals, as a result of prandial thirst that develops long before the ingested food has been absorbed and altered the blood tonicity.. These observations indicate that much normal drinking behaviour is anticipatory in nature..

Finally, I discussed human studies that used fMRI in the study of drinking. Subjects were water restricted and then scanned while images of drinking water (among other stimuli) were displayed. Activation of the mid-cingulate, insula, and amygdala was noted.

In a second study, thirsty subjects were scanned with fMRI while water was delivered. Activity in the mid-cingulate and insula regions evoked by water cues decreased as water was delivered over the course of the experiment.

Energy balance: Hunger and satiety

The control of energy balance through hunger and satiety is more complex than drinking. Our current obesity epidemic demonstrates that there are strong societal influences over food consumption. As we learned in our lecture on sleep, and as we will learn in our lecture on stress, there are strong interrelationships between lack of sleep, chronic stress, and food intake.

Research on hunger and satiety has had a consistent focus upon the hypothalamus. Back in the 1950 – 1960s, Philip Teitelbaum (one of my dissertation committee members) and Eliot Stellar were among those who described two centers in the hypothalamus:

• Lateral Hypothalamus ('Hunger' center)
  • Stimulation – eat
  • Ablation – starve
• Ventromedial Hypothalamus ('Satiety' center)
  • Ablation – obese

It seemed that this issues was solved, except for the details. However, since that time, much has been learned about the complexity of the control of hunger – particularly the conflation of activity and motivation with eating (eating has both appetitive and consummatory components – i.e., we are attracted to food and we eat the food). The multiple roles of hypothalamic nuclei have also been more clearly elucidated. Recall, for example, that the lateral hypothalamus (under the control of the SCN) produces the Orexin neuropeptide. Orexin's name suggests 'appetite', and is thought to play an important role in food regulation. However, it is also involved in activating the 'awake state' in animals, which of course is also necessary for eating.

Signals of hunger

• Hypoglycemia (low blood sugar) is detected by glucose-sensing neurons that have been identified in several regions of the brain, including the ventromedial hypothalamus, lateral hypothalamus, arcuate nucleus of the hypothalamus, amygdala, nucleus accumbens, medial prefrontal cortex, and hippocampus. It is interesting to note that some of these regions play an important role in the brain’s reward system that will be discussed in Lecture 23.
  • As will be discussed further below, the arcuate nucleus of the hypothalamus contains two distinct populations of neurons that are important in both hunger and satiety.
• Ghrelin, a hormone produced in the stomach, is a potent signal of hunger.

Signals of satiety

Satiety is more complicated than hunger, in that there are short term signals to cease eating, and there are longer term signals to manage body weight.

Short term signals:

• The 'full tummy' sensation is conveyed by the vagus nerve to the nucleus of the solitary tract in the medulla. (note, this is the same signaling route we discussed earlier with respect to vomiting).
• Cholecystokinin (CCK) is a peptide released by the gut that signals short term satiety. CCK is detected by the Area Postrema (AP).
• The signals from the AP and NTS converge on the caudal vagal complex.

Long term signals:

• Leptin (hormone) is released by adipose tissue (fat) and detected in the arcuate n. of the hypothalamus.
• Insulin is released by the pancreas and detected in the arcuate n. of the hypothalamus.
• Leptin and insulin reduce food intake.

Leptin was discovered by parabiosis studies, in which the blood supplies of two rodents are joined. In this way, blood borne signals such as hormones generated in one animal affects the other animal. In some studies of this type, rodents who were grossly obese (from diabetes, or who were genetically disposed to obesity, or who had lesions to the ventromedial hypothalamus) were paired with normal weight rodents. The normal weight rodents stopped eating and became grossly underweight, presumably due to a signal circulating in the obese rodent. It turned out that the adipose (fat) tissue in the obese rodent was signaling satiety in the normal rodent, causing that rodent to lose weight. The signaling peptide hormone was leptin.

Some humans have a leptin deficiency which can be remedied by leptin supplements. In an fMRI study, such individuals were tested with and without leptin. When challenged with food cues with low leptin, there was an increase in activation of the insula.

Circuits for hunger and satiety

Several papers in the last several years were published in Nature and other journals regarding the circuitry of hunger and satiety. I found this 2012 Nature paper by Atasoy and colleagues useful as an overview from which much of the following has been abstracted.

I synthesized the information from several articles into the diagram below. The numbers on the diagram refer to the points enumerated below.

1. Ghrelin (a hormone released by the gut when empty), Leptin (a hormone released by adipose tissue) and insulin (a hormone released by the pancreas) (and glucose) are carried through the vasculature and affect two intermingled populations of neurons in the Arcuate nucleus of the hypothalamus.
   • These two populations are named for genes that they express and/or neurotransmitters that they use. The ArRP/NPY neurons express agouti-related protein and uses NPY as a neurotransmitter. These neurons colored in green are involved in feeding. POMC/CART express prio-opiomelanocortin and cocaine- and amphetamine-regulated transcript. These neurons colored in red are involved in satiety.
2. AgRP/NPY neurons increase food intake.
   • Ghrelin (produced in the stomach when empty) stimulates the AgRP/NPY neurons.
   • Stimulation of AgRP/NPY neurons evokes voracious eating, and ablation of these neurons causes aphagia (not eating).
   • Ghrelin also inhibits the POMC/CART neurons. These neurons signal satiety, thus Ghrelin stimulates hunger signals and inhibits satiety signals.
3. AgRP/NPY neurons stimulate the Lateral Hypothalamus and the production of Orexin and MCH (melanin-concentrating hormone). These molecules stimulate both the appetitive and consummatory components of eating.
   • AgRP neurons also increase feeding by inhibiting neurons in the Paraventricular nucleus (PVN) of the hypothalamus.
   • AgRP neurons also inhibit oxytocin neurons in the brain. Oxytocin neurons in the forebrain strongly inhibit eating, and so AgRP's effect on oxytocin is disinhibition.
4. The Lateral Hypothalamus also stimulates the reward system of the brain, including the nucleus accumbens and the amygdala.
   • The amygdala (through its connection to the bed nucleus of the stria terminalis, BNST) exerts top down control upon the Lateral Hypothalamus and can thus stimulate or inhibit eating.
5. Leptin and insulin excite POMC/CART neurons leading to decrease in eating.
   • POMC/CART neurons provide a strong anorexigenic effect—secretion of the POMC and CART neuropeptides from these neurons decreases food intake and body weight.
   • Leptin and insulin inhibit the AgRP/NPY neurons, thus reducing food intake.
6. The POMC/CART neurons excite the Paraventricular n. of the hypothalamus to release Coriticotropin Releasing Hormone (CRH), Thyroitropin Releasing Hormone (TRH) and other hormones which stimulates movement and other activity in the animal, and thus uses energy.
   • PVH lesions lead to hyperphagia and obesity
   • The POMC/CART neurons also stimulate the Ventromedial Hypothalamus (VMH), which acts as a stop center for many behaviors. Lesions to the VMH also lead to obesity.

Lateral hypothalamus area (LHA) redux

We started out discussion by considering the early studies performed by Teitelbaum, Hoebel, Stellar, and others that implicated the LHA of the hypothalamus as a hunger area – stimulate and animals eat, ablate and animals starve. Recent studies by Jennings and colleagues have attempted to distinguish whether the LHA plays a role in the appetitive or consummatory aspects of eating. Using calcium imaging to monitor the firing of individual neurons in the LHA, Jennings and colleagues show that both appetitive and consummatory behaviors are represented in the LHA, but are separated into different neural populations.

Jennings also demonstrated that the amygdala exerts a strong effect upon the LHA.

Studies of hunger and satiety in humans

A large number of fMRI studies have been conducted using food cues (usually photos) as stimuli in healthy and obese individuals, whether fasting or sated. Like we observed for drinking in thirsty individuals, the insula and amygdala were preferentially activated by food cues in the fasting state.

Beyond the hypothalamus

The lateral hypothalamus communicates with the ventral tegmentum (VTA) which provides dopaminergic input to the Nucleus Accumbens (N Acc) and the reward system of the brain.

The amygdala (through the bed nucleus of the stria terminalis, BNST) can both stimulate or inhibit the lateral hypothalamus, leading to eating or satiety.

The satiety hormone Leptin and also Insulin can inhibit the VTA directly, leading to inhibition of the VTA, and less reward from continuing to eat.

The hunger hormone Grehlin can directly influence the hippocampus, leading to an increase in neurogenesis, and to increased learning for social aspects of eating.

Grehlin can also stimulate the VTA, making eating 'more rewarding'.

Videos

Prerecorded videos for 2020

Live videos from 2019

The video embedded below was recorded in Fall, 2019 and contains the second-half of Lecture 20. The first-half of Lecture 20 completed the topic of sleep, and can be found in the notes for Lecture 19.

The video embedded below was recorded in Fall, 2019, and completes the topic of homeostasis begun in the video above.