After discussing the role of allostasis and allostatic load in the cardiovascular and immune systems, we come back to the fact that allostatic load is, in many respects, the product of brain function and is influenced by behavior. The brain is the master controller of systems involved in allostasis through its regulation of stress hormone secretion and the autonomic nervous system, and through these pathways it regulates cardiovascular function, food intake and metabolism, and the immune system.

The brain is also the organ that learns and remembers; it harbors and expresses fears and emotions as well as keeping track of daily events, remembering birthdays, shopping lists and numerous other types of transient or periodic information. All this would be complex enough were it not for the fact that the brain is also a target of stress and possesses its own allostatic mechanisms in adapting to challenges. Thus, the brain is also subject to allostatic load, which can cause memory failure and events that lead to permanent damage to brain cells. This conference addresses this important topic as a way of rounding out this initial, detailed discussion of how stress can affect health in different systems of the body.

Stress and the Brain

The physiological stress response of the sympathetic nervous system and the secretion of ACTH and glucocorticoids are initiated, maintained and shut off through the activity of neurons within the hypothalamus and other parts of the brain. For ACTH secretion, neurons in the hypothalamus produce corticotrophin releasing factor (CRF) and secrete it into the portal blood of the median eminence, where it travels to the pituitary gland and triggers the secretion of ACTH. ACTH travels to the adrenal cortex and stimulates the synthesis and secretion of glucocorticoid steroids, which mainly consists of cortisol in the human and infrahuman primate and corticosterone in the rat and some other species.

Glucocorticoids have many actions throughout the body related to the production of energy in the form of glucose (hence the name "glucocorticoids") but they also have actions that contain the producton of inflammatory cytokines and shift the balance of the immune response from cellular towards humoral immunity. Chronic glucocorticoid elevation is associated with catabolic conditions involving muscle wasting and loss of bone mass, as well as suppression of immune function, and it is also associated with the development of insulin resistance and Type II diabetes. These are the consequences of chronic glucocorticoid secretion and, we will see below, what chronic exposure does in the brain. The bottom line is that acute stress is good and promotes adaptation and memory formation but chronic stress is bad and impairs memory and leads eventually to some brain damage.

Acutely, glucocorticoid secretion in stress and the diurnal rhythm of sleep and waking maintains the balance of many systems without leading to the dire consequences described above. That is, the ebb and flow of glucocorticoids around the clock (elevated in the morning; low in the evening hours) helps to coordinate sleep and waking activities such as relegating sharp cognitive function to the waking hours and coordinating metabolism with the ingestion of food.(33) In addition, the ability of adrenal steroid secretion to counter-regulate many systems, like suppressing the production of inflammatory cytokines, attenuating the effects of insulin, reducing edema, reducing the activation of the brain's noradrenergic system, means that the acute consequences of stressful challenges, including infections and trauma, are moderated and the body is less likely to damage itself by over-reacting to the stressor.(39)

Acutely, glucocorticoids participate in the feedback mechanism that determines shut-off of the ACTH stress response, acting somewhat like a thermostat. There are three types of glucocorticoid feedback:

1. rate sensitive rapid regulation, involving rapid actions of rising levels of steroids at the cell membrane level in brain and pituitary gland;
2. level sensitive, intermediate feedback involving delayed actions of steroids, over minutes to hours, to regulate CRF and ACTH release via genomic mechanisms; and
3. level-sensitive, long-term feedback, seen as the ability of glucocorticoids to keep the levels of ACTH and CRF within certain limits without preventing other feedback mechanism from operating.(17)

In addition, the nervous system has the ability to shut down ACTH release independently of adrenal steroid production, and, yet, this mechanism is also influenced by glucocorticoids secreted in the diurnal cycle. A well-defined diurnal glucocorticoid cycle results in crisp turn on and shut off of the stress response, whereas a poorly defined or non existent diurnal cycle, resulting from work, overwork and poor sleep, means a sluggish turn on and shut off of the ACTH stress response.(3),(15)

The CRF system of the hypothalamus is under negative feedback control of glucocorticoids via the long-term feedback mechanism. The same is true of ACTH production by the pituitary. CRF is produced by other neurons in the brain and is under different forms of control by glucocorticoids.(13) In the hypothalamus, glucocorticoids exert a negative control over CRF production; but, in many brain regions, glucocorticoid influences on CRF levels or CRF messenger RNA levels are negligible.(13) In the amygdala, a brain region associated with fear and anxiety, CRF production is increased and CRF mRNA levels are elevated by chronic stress and by glucocorticoids.(23),(28),(40) This provides a positive feedback loop and there is growing evidence that CRF produced by extra-hypothalamic neurons plays a major role in the brain as a regulator of anxiety and a promoter of anorexia.(42),(49)

The amygdala, a bilateral structure shaped like an almond in the temporal region of the brain, plays an prominent role in the long-term memories associated with fearful or otherwise traumatic memories.(5),(22) Glucocorticoids, along with adrenalin, are both involved in the formation of these memories.(6),(43) The amygdala is also implicated in the sensitization of the startle response in patients suffering from post-traumatic stress disorder and it may play a role in the flashbacks associated with the disorder.(11) The amygdala may also play a large role in the "anticipatory angst" that can lead to chronic elevation of stress hormone systems and lead to allostatic load.(47) In other words, the fears and anxieties that we harbor, whether or not there are real reasons to be worried, can create a condition of chronic stress and persistent elevation of hormone secretion, leading to allostatic load.

Allostatic Load and the Brain

The consequences of allostatic load in the brain include wear and tear on nerve cells in brain regions that are sensitive to stress hormones and are important for learning and memory processes. This is particularly true of the hippocampus, which has high concentrations of adrenal steroid receptors.(31) The hippocampus has a major role in episodic and declarative memories, which are the forms of memory of daily events and the daily information, such as shopping lists and names of people, places and things, that we need to know to keep track of our daily lives. The hippocampus is particularly important for the memory of context, the time and place of events that have a strong emotional bias(9),(21) and for which the amygdala also plays an important role. Glucocorticoids are involved in consolidation of contextual fear conditioning,(41) such as the memory of unpleasant or traumatic events, and in promoting other aspects of memory formation.

Excessive stress hormone levels cause impairment of memory formation and retrieval(26) and so does chronic stress. The impairment of function of the hippocampus decreases the reliability and accuracy of episodic and declarative memories, including those associated with context. This may exacerbate stress by preventing information input needed to decide that a situation is not a threat.(44) The hippocampus is also a regulator of the stress response and acts to inhibit shut-off of the HPA stress response.(12),(16) Thus, impairment of hippocampal function can promote allostatic load by impairing our ability to perceive that something is no longer stressful and by impairing our ability to shut off the stress hormone response.

The mechanism for stress-induced hippocampal dysfunction and memory impairment is two-fold. First, acute stress elevates adrenal steroids, which suppress the mechanisms in the hippocampus and temporal lobe that subserves short-term memory.(20),(34) Stress can impair memory acutely, but, fortunately, these effects are reversible and relatively short-lived.(27) Second, repeated stress causes neurons in the hippocampus to undergo an atrophy of dendrites resulting in a reduction in the connections of these neurons to the rest of the brain; the mechanisms for the atrophy involve the actions of both glucocorticoids and excitatory amino acid neurotransmitters that are released from synapses in the hippocampus during and after stress.(30)

This atrophy of dendrites, which look like the branches of a tree, is reversible if the stress is short-lived, but stress lasting many months or years can kill hippocampal neurons.(45),(50) Although the information about atrophy of nerve cell dendrites in hippocampus has been collected in studies on experimental animals like rats and tree shrews, recent information using MRI has shown that stress-related disorders, such as recurrent depressive illness, post-traumatic stress disorder, and Cushing's syndrome, are associated with atrophy of the human hippocampus; likewise, there is atrophy of the hippocampus in some individuals as they age, accompanied by memory dysfunction which precedes any signs of dementia.(32),(46) We do not know, at this time, which forms of human hippocampal atrophy are reversible and which forms reflect permanent neuron loss. Making this distinction is important for devising strategies and treatments that can either reverse a reversible atrophy or prevent an irreversible neuron loss by intervening early in the particular disorder.

Allostatic load also impacts on the aging process in the brain where neurons are usually not replaced if they die. Long-term stress is known to accelerate the appearance of several biological markers of aging in rats, including the loss of hippocampal pyramidal neurons and increasing the excitability of hippocampal neurons; both of these aspects of aging appear to be related to a loss of the ability to contain levels of calcium ions inside of cells, leading to the generation of free radicals and oxidative damage.(19) Glucocorticoids may mediate these effects by enhancing calcium currents in the hippocampus.(18) This is important and paradoxical, since calcium ions have a key role not only in destructive events, as noted above, but also in plastic processes in hippocampal neurons such as changes in excitability, including long-term potentiation, that are thought to be involved in learning and memory.(7),(29),(38) In other words, the same processes that cause damage are a necessary part of the ability of the hippocampus to adapt, change and help us learn and remember.

Not only is the wear and tear of aging associated with loss of the capacity to contain calcium ion increases, the aging rat hippocampus also shows a persistent release of the excitatory amino acid, glutamate, in the hippocampus after stress; this failure to reduce extracellular glutamate levels after stress is likely to contribute to age-related neuronal damage(25) and may potentiate atrophy and possibly result in neuronal loss.

Life Long Influences of Early Stress on the Brain

It is becoming clear that events early in life can have a life-long impact on how the hippocampus and, possibly, other brain regions, age. Studies almost 40 years ago showed that a simple procedure, called "handling" or separating newborn rat pups from their mothers for 10 minutes per day for the first two weeks of neonatal life, produces a rat that is less emotional and shows lower stress hormone output as an adult when confronted by new and potentially stressful situations.(1),(24) More recent data indicates that "handling" reduces the rate of hippocampal aging and preserves neurons and function of this important brain area in aging rats.(35),(37) The contrasting situation, making rats more emotional and more reactive with stress hormone secretion as adults, has been achieved by unpredictable prenatal stress and also by longer periods of separation of the pups from their mothers after birth.(10),(36),(51) This condition increases the rate of hippocampal aging and results in loss of function earlier as the more reactive rats grow older.(8) Thus, early experience increases the allostatic load, thereby altering the rate of wear and tear that leads to hippocampal aging by setting the level of reactivity of the allostatic systems that produce stress hormones: these systems overreact in animals subject to early unpredictable stress or separation from the mother and they underreact in animals exposed to the neonatal handling procedure.(37) In the former condition, brain aging is accelerated, whereas in the latter, brain aging is reduced.

For humans, we do not know yet what corresponds to handling in rats and what forms of prenatal or neonatal stress have long-term adverse effects on emotionality. Nevertheless, we are just now becoming aware of the long-term effects of childhood abuse on later brain structure and functioning.(4),(14) We also recognize individual differences in human brain aging that correspond somewhat to differences seen in aging of the rat brain. That is, there are individuals in whom there is more rapid loss of cognitive function that depends on the hippocampus and this pathway is associated with progressive elevations in glucocorticoid levels that points to an allostatic load lasting over a number of years.(48) However, much more needs to be learned in order to translate successfully between animal studies and what is seen in human beings.

What Can Be Done?

The brain is very much a biological organ and is subject to the effects of chronic stress and trauma. One lesson is very important -- if we damage our brains and destroy nerve cells, there is no regenerative process presently known that will help to repair it. Therefore, the lessons learned for protecting the immune system and the cardiovascular system, which are summarized in those conferences, also apply to the brain but with an even greater urgency because of the irreversible nature of what could happen.

Among these lessons is the importance of reorganizing work schedules to minimize time pressures and tedium and maximize individual initiative and a sense of control, because tedious jobs with great time pressures and a lack of control lead to allostatic load and cause cardiovascular disease. The venting of emotions and achieving a sense of support from other people are other important ways to reduce allostatic load and promote better resistance of cancer and infections. From what we know of stress effects on the brain, these types of intervention would also be expected to reduce allostatic load in the brain.

Individual differences in the aging of the human body and brain bring us back to a consideration of the contributions of genes and environment. While we have a tendency to attribute bad things, like rapid aging or diseases, to bad genes, that is only part of the story. The 40-60% concordances among identical twins for many diseases, including diabetes, schizophrenia and Alzheimer's, make it abundantly clear that experiences and influences of environmental variables throughout the lifespan can make a difference as to if, when and how many genetic traits are expressed in terms of a disease. Basic research is needed, not only to identify candidate genes, but also to learn how those genes are regulated. Finally, research on the human organism in relation to specific stress-related diseases and in the context of social change and socioeconomic gradients(2) will help put that information into the real life perspective of health and disease.