The General Adaptation Syndrome Psychology Essay

Published Date: 23 Mar 2015

Stress is inescapably present in the lives of every living organism. While some stress can be beneficial, too much of it is almost always disadvantageous. Frequently the word "stress" is used to describe the imposition of an undesirable psychological or physical state through uncontrollable or overwhelming circumstances. We tend to think of stress as a burden that we carry around with us. One individual might perceive that they are under an enormous amount of stress because of multiple impending deadlines, while another might feel stressed out because their house is not clean and unexpected guests are on their way. Despite the predominantly negative description above, stress itself is not necessarily bad. In fact, stress is an adaptive process without which our survival would be seriously compromised. Moderate stress can even have beneficial effects, such as improving cognitive function to facilitate meeting multiple deadlines and increasing energy supply to muscles to quickly clean the entire house in a matter of minutes. However, too much or prolonged stress can result in a variety of detrimental effects on both physiological function and cognitive processes.

Definition

In order to properly understand the effects of stress on cognition, as measured by performance in a variety of mazes, we must first come to an understanding of what stress is. At its core, stress is can be defined as the response to a threat (or perceived threat) against homeostasis, the tendency of a system to maintain a stable internal environment. Living organisms, including humans and rodents, are open systems which work to maintain homeostasis despite being constantly bombarded by numerous external environmental insults.

Homeostasis

The concept of homeostasis was initially conceived by the French physiologist Claude Bernard in1854. He used the term milieu intérierur to describe the ability of the internal environment (primarily the blood in Bernard's time) of the body to compensate for and re-equilibrate in response to the external environment [1,2]. However, it was Walter Cannon who developed the concept of homeostasis. He proposed that the body maintained steady state conditions through multiple cooperative self-regulating mechanisms. One phenomenon observed by Cannon was that organisms respond to a threat by releasing epinephrine (aka adrenaline) from the adrenal medulla thereby increasing the body's heart rate, respiration, and blood pressure while mobilizing glucose stores and inhibiting non-essential functions such as digestion and reproduction. This phenomenon is the prototypical sympathetic nervous system (a branch of the autonomic nervous system) response, for which Cannon coined the term "fight-or-flight" response [3]. Simply put, when confronted with a threat, an organism, through the narrowing of bodily functions and mobilization of available energy, is primed to respond actively, whether through fleeing or fighting. Thus, the primary contribution of Walter Cannon to the definition of stress was that in response to an environmental insult threatening the internal steady stated, the body initiates the sympathetic response to maintain homeostasis and ultimately increase survival.

General Adaptation Syndrome

The second major contributor to stress research was endocrinologist Hans Selye. Selye was working to discover a new hormone by injecting ovarian extracts in to rats and observing the physiological effects. He noted several changes after injection of this extract, including the 1) enlargement of the adrenal gland, 2) the atrophy of the thymus and lymph nodes and 3) the presence of gastric ulcers [4]. While he initially thought these effects were direct effects of a novel hormone, through additional control experiments he discovered that the injection of numerous agents, physical injury or even excessive exercise produced the same reliable effects [5]. Emphasizing the non-specificity of this response, Selye called these effects the General Adaptation Syndrome initially and later used the term stress to describe it. He further defined stressors as the factors or agents that triggered the stress response. In addition, Selye established the role of glucocorticoids, steroid hormones excreted from the adrenal cortex, in the stress response. Furthermore, Selye described that prolonged exposure to stressors (and the stress response) can lead to illness or disease.

Summary

The combined work of Walter Cannon and Hans Selye formed the basis of stress research. Both agreed that when confronted with disruptive environmental factors, the body generated an adaptive response aimed at re-establishing homeostasis within the internal environment. Cannon focused on the sympathetic branch of this response, while Selye focused on the hormonal (i.e., endocrine) branch of this response. Modern thinking has modified the conceptualization of stress, as it turned out to be not quite as simple as originally thought. Importantly it is no longer thought that stress is non-specific as the magnitude and more subtle characteristics of the stress response are varied based on the type of stressor, the individual's perception of the stress and ability to cope. The following basic features of stress can be gleaned from this historical work:

Stressors are any event, experience or environmental insult that threatens or is perceived to threaten homeostasis

Stress is an adaptive response to re-establish and maintain homeostasis

Stress is ultimately mediated by two branches:

The sympathetic nervous system via release of epinephrine from the adrenal medulla

The endocrine system via release of glucocorticoids from the adrenal cortex

These definitions and characteristics are more conducive to scientific study than the layman use of the term stress which does not distinguish between the trigger and the response, as described in the opening of this section. Thus, the definitions above will be utilized in the following sections. This chapter will further develop these basic concepts and present 1) an overview of the stress response, 2) descriptions of the different types of stressors, 3) the effects of both acute and chronic stress on learning and memory and 4) practical details on how to deal with the nuances of stress in behavioral testing.

THE PHYSIOLOGY OF THE ACUTE STRESS RESPONSE

When a stressor is encountered, the brain triggers a physiological response, aimed at coping with the stressor and restoring homeostasis [6]. This response is governed by the hypothalamic-pituitary-adrenal (HPA) axis and the sympathetic nervous system (SNS). Both of these systems are always actively engaged in maintaining homeostasis, but when posed with a disruption to an organism's internal environment, the HPA axis and SNS go into overdrive. When stressors are encountered several central nervous system (CNS) structures are activated to initiate the stress response, including the preganglionic sympathetic neurons of the spinal cord in addition to several brainstem and limbic forebrain structures [7]. These structures then recruit neural and neuroendocrine systems to initiate a stress response, culminating with the secretion of adrenal glucocorticoids (GCs) and epinephrine, the principle mediators of this adaptive response.

The Sympathetic Nervous System

Two primary pathways are activated by stressors within the SNS, the brain norepinephrine neurons in the locus coeruleus (LC) and the sympathetic adrenomedullary circuitry. Many, but not all stressors result in norepinephrine release from the LC, which ultimately contributes to the majority of circulating NE levels as well as numerous adaptive behaviors [8,9]. Activation of the sympathetic adrenomedullary circuitry via preganglionic neurons results in the release of the excitatory neurotransmitter acetylcholine (ACh) onto postganglionic neurons and the adrenal medulla. Sympathetic postganglionic neurons primarily release of norepinephrine directly onto various target organs. On the other hand, activation of the adrenal medulla results in the synthesis and release of epinephrine into the circulation and indirectly onto target organs. Target organs activated by these catecholamines neurotransmitters correspond directly to the symptoms of the fight-or-flight response, including increased heart rate, blood pressure and respiration, pupil dilation, and inhibition of digestion, liver, kidney and gall bladder function.

The Hypothalamic-Pituitary-Adrenal Axis (Figure 1: HPA axis from Lupien 2009 Review)

The present chapter will focus on the endocrine branch of the stress response. While the activation of the sympathetic nervous system is a key part of stress, many of the effects of stress on learning and memory have been shown to be dependent on GCs. The secretion of adrenal GC hormones is under the stimulatory drive of the medial parvocellular neurons in the paraventricular nucleus (PVN) of the hypothalamus. In response to stressors, brainstem and limbic forebrain regions activate these neurons to secrete corticotropin releasing hormone (CRH) and vasopressin into the hypophysial circulation to stimulate the anterior pituitary to release adrenocorticotrophic hormone (ACTH) into the peripheral circulation which ultimately results in the release of GCs from the adrenal gland [10]. GCs are carried to every organ via the circulation to allow for a coordinated adaptive response between the brain and bodily functions. GCs mobilize energy, suppress immune and inflammatory responses, inhibit bone and muscle growth and reproductive function as well as increase attention and impact learning and memory processes [11]. GCs also regulate further CRH and ACTH release via negative feedback loops by binding receptors in the pituitary, PVN, hippocampus (HPC) and prefrontal cortex (PFC) to inhibit further GC release and to return the homeostatic balance of the HPA axis [10] (Fig. 1). While these are the primary targets of GCs, virtually every neuron in the brain has receptors for these hormones. Thus, GCs are an important enabler of normal brain function in addition to their role in HPA axis function.

Corticosterone (CORT), the primary GC in the rat, is the ligand for the glucocorticoid receptor (GR) and mineralocorticoid receptor (MR), steroid hormone receptors that function as transcription factors and regulate neuronal gene transcription. GR and MR have different affinities for CORT in the brain. GRs are expressed ubiquitously in the brain, and are most concentrated in the CRH neurons of the PVN and ACTH releasing neurons in the anterior pituitary and also in the HPC [10]. The highest expression of MRs in the brain is in the HPC, though they are expressed in other forebrain regions [10]. MRs are saturated by modest levels of CORT, such as those generated during circadian oscillations, and thus mediate daily changes in HPA axis activity. On the other hand, GRs are saturated by more extreme levels of CORT, such as those seen following a stressful experience, and mediate negative feedback to restore homeostasis [12,10].

Time Course of the Acute Stress Response (Figure 2: rise and fall of CORT)

The typical stress response begins within a very short period of time. The SNS response occurs more rapidly than the HPA axis response, due to direct neurotransmitter release on target tissues vs. indirect hormonal release into the circulation, respectively. Within seconds, the post-ganglionic neurons of the SNS release norepinephrine and the adrenal medulla releases epinephrine into circulation. Also within seconds, CRH is released from the PVN of the hypothalamus, followed quickly by the release of ACTH into peripheral circulation from the anterior pituitary [13]. Within 3-5 minutes, the adrenal cortex releases CORT into the circulation, by which time peak plasma ACTH levels are reached. While peak levels of GCs vary according to stressor type and duration, circulating CORT levels typically reach peak levels15-30 minutes after stressor initiation [14]. At this time ACTH levels have also returned to baseline [14]. However, the effects of GCs on target tissues do not occur until about an hour after stressor initiation [13]. After peak CORT levels are reached, negative feedback mechanisms shut down further CORT secretion so that circulating CORT levels recover to near baseline levels within 60-120 minutes after stressor initiation [15,14]. The time required to reach baseline levels also varies depending on stressor type and duration. Circulating levels of norepinephrine and epinephrine follow a similar trajectory, but with more rapid increases and decreases [16].

Biological Rhythms

Plasma levels of CORT are not static, but follow a circadian rhythm that is closely aligned with the sleep-wake cycles. In both humans and rodents, plasma GCs rise as the sleep cycle (inactive phase) ends and peaks upon waking up. From this peak, GC levels fall during the active phase and eventually reach the circadian trough or nadir (the lowest point in the cycle) [17,18]. However, as nocturnal animals, the active phase for rodents is during the dark period and the inactive phase occurs during the light period. Thus, in laboratory animals, peak CORT levels are found just after the lights are turned off and nadir CORT levels are measured when the lights are turned on. In addition, within this circadian rhythm GCs are released in an ultradian (cycles repeated within the 24 hour period) pulsatile pattern which rises and falls according to the circadian rhythm [19]. Basal levels of CORT measured in rats can typically range from 0-200 ng/mL in adult male rats or 100-500 ng/mL in adult female rats [18]. In response to acute stressors, peak stress levels of CORT generally range from 200-600 ng/mL in adult male rats and 500-1000 ng/mL in adult female rats. Both basal and peak stress levels of CORT can vary depending on the sampling time within the circadian rhythm.

Sex, Age and Strain Differences

As noted above, sex differences in HPA axis activity and responses to acute stressors are well documented. Female rodents exhibit higher basal and stress induced levels of both ACTH and CORT [20,21]. Sex differences in CORT secretion are primarily mediated by estrogen, with higher levels of ACTH and CORT found during proestrus, when circulating levels of estrogen are high [20,18]. Much of the research on gonadal steroid regulation of the HPA axis suggests that estrogen has excitatory effects while androgens have inhibitory effects [22-26]. In addition, females have higher levels of corticosteroid binding globulin (CBG), which partially buffers the higher levels of CORT [27,24]. CBG binds circulating CORT and renders it biologically inactive [28,29]. However, CBG levels take several hours to increase after exposure to a stressor, resulting in higher levels of biologically active CORT in females in response to stressors [30].

Acute stress responses can also vary across the lifespan. After birth, from day 2 to 14, rats experience a stress hyporesponsive period (SHRP) wherein circulating basal CORT remain very low and neonatal pups fail to show an elevation in CORT in response to stress or ACTH administration[31-33]. Adolescence is the next developmental epoch, which includes three stages: pre-pubescence/early adolescence (21-34 days), mid-adolescence (34-46 days) and late adolescence (46-59 days) [34,35]. During early adolescence, rats exhibit an exaggerated HPA axis response to both acute and chronic stress compared to adults [36,37]. In addition, sex differences in HPA axis function mature over the adolescent period [38,39]. The acute stress response has also been investigated in aging animals, though a great deal of heterogeneity, particularly with respect to the aging model has been found [40]. However, the general consensus, based on earlier studies, is that aged animals exhibit elevated baseline (or basal) CORT and a prolonged stress-induced elevation in CORT release, possibly due to decreased GR-mediated negative feedback [41,42]. It should also be noted that aging is associated with increased variability in the acute stress response [40]. Sex differences in acute HPA axis responses are not frequently examined in aged rats. However, since estrogen decreases with age, and males demonstrate increased basal CORT, it would be expected that sex differences would become less robust [43,44]

Robust strain differences in both basal and peak stress levels of CORT also exist. In general, the in bred Fischer 344 (F344) strain of rats secrete higher levels of both basal and stress induced CORT levels compared to both the inbred Lewis (LEW) strain and the outbred Sprague Dawley (SD) strain (Dhabar 1993). Interestingly, both F344 and SD rats show increased basal CORT levels at the end of the inactive period, while LEW rats did not show a robust circadian rhythm (Dhabar 1993). As mentioned above, strain differences can also interact with age differences to further complicate the picture of a normal HPA axis stress response across the lifespan.

CHRONIC STRESS

Exposure to chronic stress can result in long term elevations in GCs and has been shown to alter the structure and function of the brain regions involved in regulating the HPA axis as well as learning and memory and numerous other behaviors [7]. Two general responses in HPA axis function occur as a result of chronic stress, namely habituation and sensitization. Habituation typically occurs following repeated exposure to the same (homotypic) mild stressor, with the magnitude of the HPA axis response (i.e. CORT secretion) diminishing with each subsequent exposure to the stressor [45,46]. On the other hand, repeated exposure to different (heterotypic) and unpredictable stressors can diminish the habituation of the HPA axis response [47-49]. Interestingly, both chronic homotypic and heterotypic unpredictable stressors cause sensitization of the HPA axis response to a novel stressor, resulting in increased ACTH and CORT secretion [45,48].

The most frequently studied brain regions that are vulnerable to chronic stress are the HPC, amygdala (AMG) and PFC [50]. In both the HPC and PFC, chronic restraint stress results in dendritic atrophy and decreased GR expression, which facilitates decreased HPA axis feedback and impaired memory [51-54,7]. In addition, chronic stress can also lead to altered hippocampal excitability, neurochemistry and neurogenesis (refs from Conrad 2010 review if needed?). On the other hand, chronic stress causes dendritic hypertrophy and increased CRH expression in the AMG as well as dendritic hypertrophy in the dorsolateral striatum (DLS), which facilitate HPA axis excitability and anxiety and habitual behavior, respectively [7,55-58].

Sex, Age and Strain Differences

Sex differences in response to chronic stress exposure are less frequently examined. The above-mentioned effects are well documented in males. However, chronic stress does not appear to have the same effects in females. For example, females appear to be resilient against chronic stress induced dendritic atrophy in the HPC and spatial memory impairments [59-63]. In the PFC, females exhibit dendritic hypertrophy, whereas males demonstrate dendritic atrophy in response to repeated stress exposure [64]. In addition, chronic stress is associated with sex-specific PFC-mediated behavioral effects. Chronically stressed males show impairments in recall of fear extinction, possibly reflecting decreased behavioral flexibility [65-67]. However, chronically stressed females demonstrate impaired memory for the acquisition of fear conditioning, which is more dependent on the AMG [68-70]. Interestingly, males show dendritic hypertrophy and increased activation of the AMG and an associated increase in the acquisition of fear conditioning following chronic stress [67]. Thus, in females, chronic stress causes more moderate morphological and behavioral outcomes that may be related to sex-specific changes in AMG function.

Chronic stress also has differential effects across the lifespan. In general, earlier exposure to chronic stress (i.e. prenatal and neonatal) have protracted and irreversible effects compared to adult chronic stress. Exposing pregnant rat dams to chronic stress increases circulating CORT which crosses through the placenta and reaches the developing fetus to alter brain development and HPA axis activity [71]. The long-term effects of prenatal stress include increased basal and stress induced CORT secretion as well as learning impairments, increased anxiety- and depression-like behaviors and sensitivity to drugs of abuse [72,73]. Postnatal stress is typically induced by disrupting the maternal-pup interactions (i.e., maternal separation or daily handling of pups). The long-term effects of early postnatal stress include altered anxiety-like behavior and stress-induced HPA axis activity in adulthood though the direction of these effects are varied and highly dependent on the age of the pup, as well as the type and duration of the manipulations [72,74,75]. Chronic adolescent stress can result in enduring effects in adulthood, including decreased hippocampal volume, impaired spatial learning, and increased anxiety-like behavior adulthood [76-79]. However, there is less evidence for adolescent stress exposure to result in depressive behaviors in adulthood [80,81]. Studies of long lasting effects of adolescent stress exposure on HPA axis function are mixed, while most report no effect on basal or stress-induced CORT or ACTH secretion, some studies find increased basal and stress induced CORT secretion in adulthood [82,78,76,83,84]. Chronic stress exposure during adolescence can lead to long term behavioral and neuroendocrine effects, depending on the timing of exposure, the sex of the animal and the type of stressors used [82,84,85]. In aged animals, chronic stress has differential effects compared to young or middle-aged adult rats. For example, following 3 weeks of chronic restraint stress in aged males and females, stress-induced sex differences were no longer evident on spatial and non-spatial memory tasks and chronic stress either facilitated or did not impair performance in aged male rats [86,87]. In addition, in aged animals sex-specific effects of chronic stress are reversed [86,87]. Furthermore, chronic stress in aged rats may potentiate stress-induced neurotoxicity, particularly in the HPC [72,88].

Strain differences are also noted in the effects of chronic stress. Interestingly, and in line with acute stress strain differences, the inbred F344 strain do not demonstrate habituation of the plasma CORT response to chronic stress with a repeated homotypic stressor, while SD and LEW rat strains do [46]. F344, but not SD and LEW, rats also exhibit adrenal gland hypertrophy after 3 sessions of restraint over 10 days [46]. F344 rats also exhibit a lack of HPA axis habituation following 14 days of restraint compared to SD rats [89]. Consistent with these findings, robust differences between SD and LEW rats were not found in either male or female rats in HPA axis response to an acute stressor after chronic exposure to unpredictable heteroptyic stressors [90].

TYPES OF STRESSORS

Acute Stressors (Figure 3: Equipment/Stressors)

In addition to being a threat against homeostasis, stressors can be described as any event or environment that is novel, unpredictable, poses a threat to well-being or ego, or creates a sense of loss of control (Lupien 2007). Acute stressors are stressors which occur on a single occasion with varying durations. Typically acute stressors last anywhere from 5 minutes to upwards of 6 hours, but in general for a period of time less than 24 hours. Acute stressors of longer durations are sometimes referred to as single prolonged stressors. Essentially the purpose of an acute stressor is for the experimental subject to mount a single stress response, or activation of the HPA axis. There are many different types of stressors and categorization of these stressors is attempted by many, though not all stress researchers agree. Stressors can vary in the type of perceived stress such as physical or psychological (or some combination of the two) as well as intensity and duration. Depending on the type of stressor used, differential effects on the HPA axis and SNS responses will occur.

Physical stressors are stressors that are predominantly identified or perceived through the senses or involuntary (autonomic) mechanisms. Examples of physical stressors include extreme temperatures, loud noise, chemical exposure, extreme activity, metabolic or immune system disruptions, hemorrhage and pain. The most extreme physical stressors are necessarily limited to short durations due to discomfort. Many physical stressors can be administered in multiple ways or in combination with other stressors. To manipulate temperature, rodent subjects can be place in a standard cold room (~4-6°C) or a hot room (~30-40°C) or in cold (~18-20°C) or hot water (~25-35°C) as part of a forced swim stressor (forced exercise) [91-93]. Loud noises stressors are most effective at a range of ~90-110 dB [94]. Chemical stressors include exposure to ether vapors, hypoxic conditions, and injection of formalin or other noxious substances [93,95]. Stressors involving extreme activity include forced swim and on the opposite end of the spectrum, immobilization, wherein the subject is splayed out on a board with all its limbs and head secured to prevent movement. Metabolic disruptions can include fasting or food restriction, injections of insulin to induce hypoglycemia or lipopolysaccharide to induce an immune reaction. Hemorrhage or hypovolemia is induced by removing a percentage of blood volume with a moderate hemorrhage consisting of removal of ~20% estimated blood volume [96]. Finally, pain can be induced via numerous methods. Frequently used stressors include tail pinch with clothespins or hemostats. Another common stressor associated with pain is mild (<0.5mA) or moderate foot shock (0.5-1.0 mA), though this stressor also has psychological characteristics due to the rapid development of fear and anticipatory stress [91,97].

Psychological stressors are perceived threats against homeostasis or well-being. Important qualities of psychological stressors include loss of control, unpredictability and novelty. Identification of psychological stressors can also be experience or species dependent, in that the anticipation of the event itself can be a stressor, even in the absence of the actual event. Psychological stressors may also induce fear, anxiety or other behavioral changes. The effectiveness of psychological stressors is primarily under the control of limbic brain regions such as the HPC, AMG and PFC [7]. Several types of psychological stressors are used, including those that activate innate and species-specific behaviors and fear, inescapable/uncontrollable stressors, and social stressors. Activation of innate fear (based on a threat to safety/survival) in rodents can be accomplished with exposure to a predator (i.e. a cat or snake) or to a novel or unfamiliar environments particularly those with are open and brightly lit (i.e. open field or elevated plus maze). Inescapable and uncontrollable stressors include inescapable foot or tail shock, forced swim and restraint. Rodent restraint is the most frequently used and well-documented stressor. It can also be combined with other physical stressors such as being placed in a rotating platform or in combination with tail pinch. Interestingly, even animal handling can be categorized as this type a stressor.

Social stressors make up their own entire subcategory. Many researchers believe that social stressors are the most translationally relevant because the primary source of stress in humans comes from social interactions, expectations and even trauma. Rats are by nature social animals, but too much or too little can elicit a stress response. Two of the least effortful social stressors are isolation housing and high density housing (i.e. crowding). Interestingly males show a greater stress response to crowding while females show a greater stress response to single (i.e. isolation) housing [98]. A popular social stressor is social defeat, utilizing the resident intruder paradigm. This paradigm consists of using aggressive male rats (i.e. retired breeders or reared in isolation without handling) as the "resident" rat. The experimental rat (i.e. the one which will mount a stress response) is the "intruder" and is placed in the home cage of the "resident" rat. Typically the rats are allowed contact with each other until the "intruder" demonstrates defeat behaviors (i.e. submissive supine posturing). In one modification of this paradigm, the "intruder" remains in the home cage of the "resident" rat but is separated, and therefore protected from physical harm, by a wire mesh enclosure [99,100]. A novel socially based stressor that is less labor intensive is to place a male rat in the dirty cage of another male rat. Thus the male rat is in an inescapable territory of another male rat, without physical contact [101].

Chronic Stress Paradigms

Chronic stress is prolonged stress that typically occurs for at least 5 days, typically 1-3 weeks and sometimes up to 6 weeks or more. Many different types of chronic stress paradigms can be used, depending on the desired outcome measures. An additional variable to consider with chronic stress paradigms is the likelihood of habituation of the stress response over time, as mentioned above. Examples of chronic stress paradigms include chronic restraint, chronic mild stress, chronic unpredictable stress, chronic variable stress, chronic social stress, and chronic intermittent stress.

Chronic restraint stress and chronic mild stress have historically been the most frequently used chronic stress paradigms. Interestingly, these paradigms have different phenotypic outcomes. Chronic restraint stress is frequently carried out with daily 6 hour restraint sessions for 3 weeks (Conrad XXXX). However, because chronic restraint stress is repeated exposure to a single homotypic stressor, habituation of the stress response after the first few days likely occurs as evidenced by blunted CORT and ACTH release in response to restraint following completion of the chronic restraint paradigm [102,103]. This paradigm is well characterized with regard to spatial learning and memory deficits as well as altered dendritic morphology, particularly dendritic retraction in the hippocampus [104]. Interestingly, spontaneous recovery of both altered dendritic morphology and spatial learning and memory deficits has been observed [105]. Chronic restraint stress also alters anxiety- and depressive-like behaviors, impairs response inhibition, facilitates fear learning and reduces motivation for food reward [106-108,53]

Chronic mild stress (CMS) is typically used to induce an anhedonic (i.e. depressive-like) phenotype [109]. This paradigm typically involves exposure of the subjects to one mild stressor a day, typically for 6-8 weeks [109,110]. Stressors are administered on a set weekly schedule and include wet bedding, cage tilt, mild footshock, alterations in the light-dark cycle, food/water deprivation and cage changes [109,110]. Anhedonic effects of CMS, which can last up to 3 months, include decreased sucrose consumption, increased threshold for intracranial self-stimulation and decreased locomotor activity in the absence an anxiety-like phenotype [109]. Chronic unpredictable stress (CUS) is essentially an iteration of CMS, which lasts for a shorter period of time (~2 weeks) and presents mild to moderate stressors once a day in a more randomized order and time of day [111,112]. Depressive-like phenotypes are seen after CUS exposure in addition to anxiety-like phenotypes and cognitive deficits [112,91]. Chronic variable stress (CVS) is yet another more severe iteration of a repeated administration of heterotypic stressors. More recently, CVS paradigms have differentiated themselves from CMS paradigms by the use of more moderate to severe stressors [113] occurring at least twice per day in a variable order and time for 1-2 weeks [113,82]. Stressors included in CVS paradigms include physical, psychological and social stressors such as: tail pinch, foot shock, shaking, forced swim (at hot and cold temperatures), hypoxia, social defeat, social crowding or isolation, cage exchange [93,82,73,114]. Typical effects of CVS paradigms still include some depressive- and anxiety-like phenotypes as well as increased fear reactivity [115,116]. However, one must take care when interpreting the type of stressor paradigm. In some instances, what are termed CUS and CVS paradigms might actually more closely resemble CMS paradigms.

The most frequently used chronic social stressor is repeated exposure, ~1-5 weeks, to the resident-intruder social defeat paradigm described above [117,99]. After the first day of defeat, "intruder" rats can either be placed into the enclosure within the "residents" cage or subjected to additional physical defeat episodes [99,118]. An additional adult social stressor includes exposure to unstable social environments, such as mixed-sex colonies or repeated changing of cage mates, which has been used in conjunction with an additional psycho-social stress, predator exposure [119-121] . In addition, the visible burrow system is a model of social stress which creates a naturalistic environment and allows complex social hierarchies to be formed [122,123]. In this paradigm, both subordinate and dominant rats exhibit alterations in HPA axis function and responsiveness. In opposition to the above social stresses involving aggressive or unstable group conditions, social isolation is also used but appears to affect females more than males [98]. In general, chronic social stress results in decreased locomotor activity, increased anxiety, anhedonia and submissive behavior, decreased aggression and sexual behavior, as well as spatial learning deficits [124-126,117,127].

Determination of Stressor Effectiveness

Determination of the effectiveness of acute stressors is quite simply done by measuring HPA axis or SNS output. The gold standard is to observe the characteristic rise in plasma CORT and ACTH following stressor initiation [128]. Effectiveness of an acute stressor on the SNS response is indicated by plasma increases in epinephrine or NE. In addition, physiological parameters such as increased heart rate or blood pressure can be used to measure the SNS response. Determination of chronic stress effectiveness is more complicated and variable. The most frequently used measure is that of attenuated body weight gain throughout the duration of stressor administration. In addition, many (but not all) chronic stress paradigms result in adrenal hypertrophy (increased adrenal weights) and thymus atrophy (decreased thymus weights) such as originally noted by Selye (for example [129,128]). One measure to determine chronic stressor effectiveness is to determine long term changes in HPA axis function [128]. Many chronic stress paradigms result in increased HPA axis activity, either at baseline or in response to a novel acute stressor [48,115,128]. However, depending on animal strain, individual stressor intensity, stress duration, and recovery time following cessation of the stress paradigm, hyperactivation of the HPA axis is not always found (for example [115]). In general, most investigators use body and adrenal/thymus weights to confirm stressor effectiveness as this is both labor and cost effective.

STRESS, MAZES AND LEARNING AND MEMORY (Figure 4: Mazes)

Spatial learning and memory is the most commonly examined outcome on mazes. Stress effects on spatial memory are consistently assessed by various iterations of land (i.e. Y-maze, T-Maze, radial arm maze) and water maze tasks (i.e. Morris water maze, radial arm water maze). The hippocampus serves to integrate salient information about the environment or context (i.e., spatial cues) and ultimately guide behavior and is thus a key neural correlate and mediator of spatial memory. Interestingly, the receptors for GCs, (MR and GR) are highly expressed in the hippocampus [10].Thus hippocampal-dependent learning and memory are highly susceptible to the effects of both acute and chronic stress [104,130]. Stress effects on non-spatial learning and memory can also be investigated with mazes, such as stimulus-response or striatal mediated learning and memory [131].

Acute Stress and Spatial Memory (Figure 5: Inverted U)

Timing, Intensity and Difficulty

The effects of acute stress on spatial learning and memory are dependent on several interacting factors including 1) the stage of memory examined, 2) the timing of stressor administration, 3) stressor intensity, and 4) the aversiveness or difficulty of the task [130,132,133]. The first two factors are intimately related, however the stages of memory including acquisition, consolidation and retrieval, will be defined first. Acquisition refers to the initial learning or training stage of a maze task, wherein the subject accumulates and integrates information, including cues within the maze itself (intra-maze) and cues in the surrounding room (extra-maze) (see Chapter XX). Consolidation refers to the period of time, immediately after to several hours after learning/training, wherein recent learning is strengthened and stored. Retrieval is typically the final stage used in maze learning and memory tasks and can also be called retention or long-term memory. At its core, retrieval is the recalling or access of previously learned and stored information. Thus acute stress can have effects on various stages of maze learning and memory depending on the timing of administration. In general, pre-training stress impacts acquisition, stress associated with the performance of task can also affect acquisition, post-training stress generally affects consolidation and pre-retrieval impacts retrieval of established memory. Stressor intensity can determine the circulating levels of CORT during the memory stage investigated. Finally, it should be noted that acute stress effects on all types of memory depend on task aversiveness (i.e. learning an association between an intense footshock and a tone) and difficulty (i.e., integration of a complex or reversible cues or rules). Highly aversive or complex tasks are affected by acute stress differently than non-aversive or simple tasks [133,130]. Typically acute stress effects on simple tasks follow an inverted U shaped curve (Figure 3), such that very low and very high levels of stress are associated with poor performance or memory and moderate levels of stress are associated with improved or optimal performance and memory. However, in highly aversive or complex tasks, the impact of stress on learning and memory can be described by a positive linear relationship such that increased stress levels results in improved performance [133,130]. The following sections will discuss the effects of acute stress on spatial learning and memory in light of these contributing factors, with a focus on the three stages of memory.

Pre-training Stress

Acquisition

Manipulations that are administered prior to training can potentially affect acquisition of learning and memory. Surprisingly few studies have effects of found pre-training stress or manipulations of the HPA axis effects on performance during training, or learning. For example, rats exposed to high intensity stressors such as inescapable shock or a predator did not show altered learning performance on the MWM [134-136]. However, when the task itself is stressful (i.e. increases circulating CORT), such as a water maze with cold water (19ËšC), improved learning performance is observed compared to training in a warm (25ËšC) water maze [137]. Furthermore, when rats tested in warm water are given a low dose of exogenous CORT (10 mg/kg), spatial learning improves, but not when given a higher dose of CORT (25 mg/kg) [137]. Similarly in a more extreme cold water maze (12ËšC), spatial learning is impaired [138]. Together these studies indicate that spatial learning performance is resilient to acute stressors, unless they are intrinsic to the task itself, in which case CORT effects on spatial learning follows an inverted U-shaped curve [132].

Acquisition and Consolidation

Interestingly, acute stress, as well as pharmacological or surgical manipulations of the HPA axis system, result in consistent effects on spatial memory, which follow the inverted U relationship. Manipulations that result in zero to low circulating CORT or inhibited GR receptors such as ADX, GR antagonists and metyrapone (inhibits CORT synthesis) result in impaired spatial memory on the Y-maze or MWM [139-142]. Manipulations resulting in moderate levels of circulating CORT or occupation of GC receptors such as ADX + MR agonists, administration of MR antagonists in intact animals result in normal to optimal spatial memory on the Y-maze or MWM [139,140]. Finally, increasing CORT to high or stress levels prior to training leads to impaired spatial memory during testing on the MWM. These manipulations have included exposure to an intense stressor (inescapable tail shock or predator exposure [134,136]), though not in every instance [135]). Because the impact of these pre-training manipulations was not evident until later memory tests, the consolidation of learning during training was likely impacted. To tease apart the effects of acute stress on acquisition vs. consolidation, stressors or HPA axis manipulations can be conducted immediately after training. For example, if both pre- and post-training stress impair spatial memory on the same task, and pre-training stress does not impact learning, it is likely that stress impairs spatial memory as a result of disrupting consolidation processes.

Post-training stress

Consolidation

Very few studies have examined the effects of post-training acute stress/ HPA axis manipulations on spatial memory. However, it appears that in non-aversive maze tasks, stress impacts the consolidation of spatial memory in a pattern consistent with the inverted U-shaped curve. Manipulations that result in low circulating CORT or GR occupation, such as administration of GR antagonists impairs spatial memory on the MWM [141]. Selective occupation of MR (i.e. ADX + aldosterone administration, similar to non-stress conditions) immediately after training results in normal spatial memory on the Y-maze [140]. However, selective occupation of GR (i.e. ADX + GR agonists, similar to stress conditions) results in impaired spatial memory on the Y-maze [140]. It should be noted that in aversive learning paradigms, including testing in aversive (cold water) mazes, increasing levels of CORT or GR occupation results in improved spatial memory, thus following a positive linear relationship [130,132].

Delay and Pre-retention stress

Consolidation and Retrieval

Some maze tasks utilize a training period followed by a brief delay period (i.e. 30 minutes to 4 hours) before testing memory. Stress manipulations during this delay period can thus impact both consolidation and retrieval of spatial memory. Exposure to a predator or a novel environment during this delay period impaired spatial memory on the land or water versions of the radial arm water maze [143-146]. In addition, several studies have examined the selective effects of acute stress on retrieval of spatial memory with pre-retention test manipulations. Together these studies reveal that stress impacts the retrieval of spatial memory in a pattern consistent with the inverted U-shaped curve. Administration of metyrapone, which results in very low levels of circulating CORT, prior to the retention test on the MWM resulted in impaired spatial memory retrieval [142]. Exposure to a predator before a 24 hour memory test impaired spatial memory on the radial arm water maze [147]. In addition, administration of footshock or injection of CORT 30 minutes prior to the retention test on the MWM resulted in impaired retrieval of spatial memory [148]. These manipulations correspond to the time course of the HPA axis response, specifically the time at which peak CORT levels (30 minutes post-stress) are reached and begin the descent back to basal levels (90-120 minutes post-stress) (Fig X). Interestingly, if footshock was administered 24 hours or 2 minutes prior to retention testing, no effects were observed. Thus elevated circulating CORT is required to impair spatial memory recall during the retention test. Manipulations resulting in moderate GC receptor occupation, (i.e., administration of MR and GR agonists to ADX rats) prior to the retention test on the Y-maze enhances or normalizes retrieval of spatial memory on the Y-maze [140].

Acute Stress and Non-spatial Memory

In maze tasks where hippocampal and non-hippocampal memory can be dissociated (i.e., dual solution water or land plus mazes, T-mazes, or radial arm mazes) acute stress can shift the strategy used to solve the maze to non-hippocampal processes. For example, in a dual solution T-maze where in rats are trained to find a food reward (land) or platform (water), either spatial cues, discrete visual cues or automatic (stimulus-response) motoric strategies can be used. Discrete visual cues (i.e. a visible platform) and automatic motoric responses are mediated by the dorsal striatum. Acute stress, including restraint, restraint plus footshock and predator odor, all facilitate automatic memory (dorsal striatum-dependent) at the expense of spatial memory (hippocampal-dependent) [134,149,131]. Thus, depending on the outcomes of interest, intra- and extra-maze cues as well as randomization of motoric patterns must be carefully considered.

Sex Differences

The majority of studies on the effects of acute stress on maze memory have been conducted in male rats. Endogenous levels of ovarian hormones, which fluctuate during the estrous cycle, can significantly interact with the effects of acute stress. Of the few studies that have included females, acute stress generally has the opposite effects. For example, when rats are tested in a cold (19˚C) water maze, male rats and female rats in estrous (low circulating estrogen levels) show facilitated performance, while female rats in proestrus exhibit impaired performance [150]. Similarly, acute restraint stress prior to Y-maze training resulted in impaired spatial memory in males but facilitated memory in intact females regardless of estrous cycle stage, despite greater stress-induced increases in CORT in proestrus females [151]. However, more intense stressors such as predator exposure may overcome these sex-differences, resulting in sex independent impairment of spatial memory in the RAWM [152]. Age differences? I didn't do an exhaustive search, but was unable to find much at all on acute stress effects across the lifespan. The focus tends to be more on chronic stress effects, which will be covered in the next section…

Chronic Stress and Spatial Memory

Duration, Motivation and Recovery

The well-established link between chronic stress and hippocampal dysfunction has led to numerous investigations of chronic stress effects on spatial memory. As described above, repeated or long-term exposure to stressors results in significant alterations in hippocampal structure and function, which are associated with impairments in spatial learning and memory. Important factors that influence these effects include 1) the duration of the chronic stress paradigm, 3) the type of stress paradigm, 2) the motivational aspects of the maze (i.e. appetitive/land vs. aversive/water), and 3) the period of recovery prior to the initiation of training. Depending on the intensity of the stress paradigm, effects on spatial memory are more robust after at least 1 week of chronic stress [104]. Surprisingly, the nature of the chronic stress paradigm is less of a contributing factor. In general, chronic stress impairs spatial learning (acquisition) in tasks that are less aversive (i.e. land mazes) but has minimal effects on more aversive mazes such as the RAWM. However, chronic stress impairs spatial memory or retention in at least one domain of memory (i.e., reference memory vs. working memory; see Chapter XX), regardless of maze type [104]. Finally, it has been shown that spatial learning and memory impairments can recovery with the passage of time [105]. Thus, longer delays between chronic stress exposure and maze training are associated with diminished spatial learning and memory deficits. The following section will discuss the effects of chronic stress on spatial learning and memory with a focus on the motivational characteristics of the maze task investigated.

Chronic Stress and Land Mazes

Appetitive mazes are land mazes, including the Y-maze, T-maze and RAM, that use food or water to train rats to make desired responses, such as learning the association of a spatial location with a reward (see Chapter XX). To motivate rats, food or water restriction is implemented depending on the type of reward. It should be noted that food restriction is an established metabolic stressor and can further interact with the effects of chronic stress. Thus, the nature (i.e. time limited free feeding vs. a set number of pellets) or the extent (i.e., target body weight percentage) of the food restriction can be adjusted to minimize the potential impact on chronic stress related outcomes. However, anhedonic rats, such as rats that have been exposed to a CMS paradigm may not be appropriately motivated even in the face of food/water restriction. Perhaps as a result of these concerns, the overwhelming majority of relevant studies have utilized chronic restraint stress paradigms [104]. Despite these potential confounding factors, the majority of studies indicate that chronic stress impairs both acquisition and retention of spatial learning on these tasks [104]. When chronic stress (primarily 6h/d of restraint but also including restraint + tail shock and shock-escape stress, and cold water exposure) occurred for longer than 20 days, rats demonstrated impaired performance during training and retention trials [104]. However, if chronic stress occurred for shorter periods of time, such as 3, 7 or 13 days, or occurs after training, acquisition was not impaired. Importantly, on the RAM, where spatial reference and working memory can be assessed separately, chronic stress consistently impairs reference memory with less robust effects on working memory. In non-appetitive land maze tasks, such as spatial recognition in the Y-maze, the motivation is the natural tendency of the rat to investigate novelty. On these tasks, chronic stress of adequate duration also tends to impair retention of spatial memory. Together, these studies indicate that chronic stress significantly impairs both spatial acquisition and spatial memory on appetitive and non-appetitive land mazes.

Chronic Stress and Water Mazes

The two primary types of water mazes used to investigate the effects of chronic stress on spatial learning and memory include various iterations of the MWM and the RAWM (see Chapter XX). The effects of chronic stress on spatial acquisition on the MWM are convoluted and dependent on the intensity of the chronic stress paradigm. Overall, it appears that exposure to moderate to severe chronic stress paradigms (i.e. chronic restraint stress, 1 month CUS and 5 months of unstable housing) results in impaired spatial learning on the MWM. This general pattern can be further influenced by testing conditions. For instance, rats exposed to chronic restraint stress show spatial learning deficits when trained in cold (19ËšC) but not warm (24ËšC) water. On the other hand, less intense chronic stress exposure (i.e., 10 days CUS or 21 days constant light) tends to facilitate spatial learning. An additional important variable is whether distance and/or velocity to escape is included as a measure to control for whether the impairments are due to spatial learning or motivational deficits. The effects of chronic stress on spatial memory or retention in the MWM are even more equivocal. In tasks that utilized a probe trial, just over half of the studies report spatial memory impairments, while the others report no impairments or even facilitation of spatial memory. However, use of different training schedules and the potential for use of non-spatial strategies as a result of over-training or chronic stress exposure ultimately leave this literature open to interpretation. The RAWM is similar to the MWM in that it uses aversive motivation (escape) yet utilizes the same brief training and testing procedures as the land version, the RAM. Benefits of the RAWM are the increased potential to control for spatial vs. non-spatial strategy use and the ability to differentiate spatial reference and working memory impairments. The majority of studies using the RAWM employ chronic social stress (i.e., predator exposure and unstable housing conditions) though chronic restraint stress has been employed more recently. As with the land version of the task, the overall effects of chronic stress are to impair reference memory. Interestingly, chronic stress appears to have little impact on the acquisition of the RAWM task. Due to the potential stress-sensitive variables inherent in MWM testing, the RAWM may be a more accurate reflection of the effects of chronic stress on spatial learning and memory.

Chronic Stress and Non-spatial Learning and Memory

Fewer studies have investigated the effect chronic stress in non-spatial mazes tasks in rats. However, evidence suggests that chronic stress can impair object discrimination memory in the Y-maze. More recent studies on the effects of on learning strategy suggest that chronic stress may facilitate cue-based/stimulus-response/automatic learning strategies at the expense of spatial strategies on both the dual solution T-maze or a cued Y-maze [153,131]. This is similar to the effects of acute stress, though additional studies are needed to confirm these findings.

Sex and Age Differences

Many more studies have investigated both sex and age differences in the effects of chronic stress on maze learning and memory compared to acute stress. In opposition to adult male rates, adult female rats tend to show facilitated spatial learning and memory when tested after exposure to chronic stress [87]. Facilitated performance in chronically stressed female rats has been observed on the RAM, Y-maze and MWM and RAWM (?). Chronic stress has differing effects on maze-related spatial memory across the lifespan. Early pre-and postnatal chronic stress can produce long lasting spatial memory impairments, while chronic stress-induced spatial memory impairments can be reverses in adult rats. Few studies have examined the effects of adolescent chronic stress on adult spatial learning and memory. However, two studies suggest that exposure to variable physical stressors or the elevated plus maze in adolescence leads to impaired spatial memory on the MWM in adulthood [84]. Interestingly, these impairments were not present in adolescence, indicating that chronic adolescent stress may alter the developmental trajectory of the hippocampus and its function. In aged rats, exposure to chronic stress or long-term elevations in CORT caused accelerated impairments in spatial ability compared to younger cohorts [154]. In summary, chronic stress has consistently detrimental effects on spatial learning and memory across the life span in male but not female rats. The extent and duration of these effects vary based on the developmental stage of the brain and stress axis.

ELIMINATING UNWANTED STRESS EFFECTS

While the bulk of this chapter has discussed the effects of stress on maze learning and memory, it is equally important to consider measures to avoid unwanted or confounding effects of stress on maze tasks. Importantly, the less aversive and appetitive maze tasks are more susceptible to the effects of experimenter or testing stress. More aversive and complex water mazes are less susceptible to experimenter-induced stress, but are still sensitive to intrinsic potentially stressful testing conditions. The following section will briefly consider factors that can result in unwanted stress in appetitive and aversive maze tasks.

Land Mazes

Appetitive tasks, which are typically land mazes, include various iterations of the Y-maze, T-maze, and RAM. The first consideration is handling prior to testing. Tasks that require repeatedly picking up the rat and placing it into an open maze can be benefitted by moderate handling procedures prior to testing. Handling of rats can itself increase CORT and thus serve as a low intensity stressor. To avoid the unwanted effects of handling the rats during testing, it is advised that the rats be habituated to experimenter handling prior to testing. Likewise, the experimenter should be a familiar individual as the presence of strange or novel experimenters is associated with a stress response. An additional consideration is the intensity of light in the testing room. High intensity light (greater than 200 lux?) can be anxiogenic and thus result in testing-associated stress. Not only that, but in brighter testing environments, rats are less likely to perform, despite being properly motivated. Another sensory related source of unwanted stress can be unexpected noise, which can also generate a stress response and even induce freezing during testing. To avoid this type of stress, appetitive maze testing should be conducted in a relatively soundproof room away from foot traffic and frequently used doors. Alternatively, a source of background noise, such as a fan or white noise machine (<60 dB) can be used. A final confounding variable that can both be a source of stress and interact with chronic stress effects is the use of food restriction, which is a metabolic stressor. While in many instances food restriction cannot be avoided, alternative options exist such as the use of highly motivating food rewards such as sweet condensed milk or sucrose pellets. Additionally, the maze task can be modified to use a water reward, as water restriction can be effective within a much shorter time frame (24 hours) compared to food restriction (one week or more). While it is impossible to eliminate every possible source of stress during appetitive maze testing, the above suggestions can be easily implemented and result in reduced variability due to the unwanted effects of stress.

Water Mazes

Water mazes are less susceptible to unwanted stress associated with handling and the testing environment. This is due to the fact that water mazes are more inherently stressful than land mazes. While rats can swim relatively well, water mazes take advantage of their high motivation to escape the water, because it is aversive, as exemplified by the use of forced swimming as an acute stressor. As described above, water temperature can significantly impact spatial learning and memory. Specifically, cold water (below 20ËšC) increases circulating CORT while lukewarm/warm water (~24-25ËšC) does not. However, warmer temperatures could potentially lead to enhanced physical exhaustion. Thus, to avoid unwanted effects of stress on water maze testing, a conservative range of optimal temperatures would be 24-26ËšC. Finally, training procedures that are more physically taxing could also be a source of stress, as a form of extreme exercise. For example, consecutive training trials in a water maze would lead to physical exhaustion more readily than blocked training trials with a built in delay.

CONCLUSIONS

Stress, a threat against homeostasis, can significantly impact maze learning and memory. The methods used to induce stress are numerous and can themselves lead to more or less robust outcomes depending on the intensity and type of stressor as well as the duration of exposure. In general, spatial memory is impaired by very low or very high levels of the stress hormone, corticosterone, such as are induced by an acute stress response. Chronic stress paradigms of adequate intensity and duration can also significantly impair spatial learning and memory, depending primarily on the motivational nature of the maze. Both acute and chronic stress can also result in the use of automatic or stimulus response learning strategies at the expense of spatial and integrative learning strategies. All aspects of acute and chronic stress responses and their respective effects on maze learning and memory are further modified by sex, age, and strain differences as well as testing conditions. Thus, many variables must be considered in order to either induce or avoid the effects of stress on maze learning and memory.