Stress-induced rise of body temperature in rats is the same in warm and cool environments

Prenatal exposure to nicotine attenuates stress-induced hyperthermia in 7- to 8-week-old rats upon exposure to a novel environment

Given that approximately 25% of women in the United States continue to smoke cigarettes during pregnancy, it is important to know if exposure to nicotine during development alters the physiological response of the adult to the "stressors" of everyday life. Our current experiments were carried out to determine if prenatal exposure to nicotine alters "stress-induced hyperthermia" in adult rats upon exposure to a novel environment such as a simulated open field. Forty-eight rats (23 males and 25 females) were exposed to a simulated open field or left in their home cage at 7 to 8 weeks of postnatal life (i.e., adulthood as defined by the ability to reproduce) following prenatal exposure to vehicle or nicotine (6 mg of nicotine tartrate per kilogram of maternal body weight per day) via a maternally implanted osmotic minipump from Day 6 or 7 of gestation. The simulated open field consisted of a 30(W)x60(L)x24(H)-in. white acrylic finish box illuminated by two hanging fluorescent lights and core temperature was measured by telemetry. Exposure to a simulated open field following prenatal exposure to vehicle elicited an increase in core temperature in male and female rats with a magnitude of approximately 1.2 degrees C and a duration of greater than 170 min. Prenatal exposure to nicotine significantly attenuated the thermogenic response of both genders; this was not only evident in the latency, magnitude and duration of the core temperature response but also in the core temperature index determined from the 3-h period following exposure to a simulated open field. Thus, our data provide evidence that prenatal exposure to nicotine attenuates stress-induced hyperthermia in male and female 7- to 8-week-old rats upon exposure to a "stressor" of everyday life (i.e., a novel environment).

Neuroprotective effects of mild hyperthermia prior to focal ischemia in conscious rats

Hyperthermia during or after stroke is known to worsen neuronal damage. Paradoxically, when hyperthermia precedes stroke, it can protect against a subsequent ischemic insult. Other stressors including restraint also have a similar pre-conditioning effect. In the present study, we report the unanticipated finding that conscious rats, restrained for the purpose of intravenous infusion, had markedly reduced neuronal and functional deficits after middle cerebral artery occlusion compared with unrestrained rats. Restrained rats had significantly higher body temperature prior to stroke than unrestrained rats. The findings suggest restraint leading to mild hyperthermia may be sufficient to induce adaptive processes which protect against subsequent ischemia.

Mechanisms and mediators of psychological stress-induced rise in core temperature

OBJECTIVE

Despite numerous case reports on "psychogenic fever," it remains uncertain how psychological stress raises core temperature and whether the rise in core temperature is a real fever or a hyperthermia. This article reviews studies on the psychological stress-induced rise in core temperature (PSRCT) in animals with the aim to facilitate studies on the mechanisms of so-called psychogenic fever in humans.

METHODS

To address this question, we reviewed the mechanisms and mediators of the PSRCT and classic conditioning of the fever response in animals.

RESULTS

The PSRCT is not due to the increased locomotor activity during stress, and the magnitude of the PSRCT is the same in warm and cold environments, indicating that it is a centrally regulated rise in temperature due to an elevated thermoregulatory "set point." The PSRCT caused by conventional psychological stress models, such as open-field stress, is attenuated by cyclooxygenase inhibitors, which block prostaglandin synthesis. On the other hand, the PSRCT elicited by an "anticipatory anxiety stress" is not inhibited by cyclooxygenase inhibitors but by benzodiazepines and serotonin Type 1A receptor agonists. The febrile response can be conditioned to neutral stimuli after paired presentation with unconditioned stimuli such as injection of lipopolysaccharide, a typical pyrogen.

CONCLUSIONS

Most findings indicate that the PSRCT is a fever, a rise in the thermoregulatory set point. The PSRCT may occur through prostaglandin E2-dependent mechanisms and prostaglandin E2-independent, 5-HT-mediated mechanisms. The febrile response can be conditioned. Thus, these mechanisms might be involved in psychogenic fever in humans.

Effect of immobilization and concurrent exposure to a pulse-modulated microwave field on core body temperature, plasma ACTH and corticosteroid, and brain ornithine decarboxylase, Fos and Jun mRNA

Exposure of humans and rodents to radiofrequency (RF) cell phone fields has been reported to alter a number of stress- related parameters. To study this potential relationship in more detail, tube-restrained immobilized Fischer 344 rats were exposed in the near field in a dose-dependent manner to pulse-modulated (11 packets/s) digital cell phone microwave fields at 1.6 GHz in accordance with the Iridium protocol. Core body temperatures, plasma levels of the stress-induced hormones adrenocorticotrophic hormone (ACTH) and corticosterone, and brain levels of ornithine decarboxylase (Odc), Fos and Jun mRNAs were measured as potential markers of stress responses mediated by RF radiation. We tested the effects of the loose-tube immobilization with and without prior conditioning throughout a 2-h period (required for near-field head exposure to RF fields), on core body temperature, plasma ACTH and corticosteroids. Core body temperature increased transiently (+/-0.3 degrees C) during the initial 30 min of loose-tube restraint in conditioned animals. When conditioned/tube-trained animals were followed as a function of time after immobilization, both the ACTH and corticosterone levels were increased by nearly 10-fold. For example, within 2-3 min, ACTH increased to 83.2 +/- 31.0 pg/dl, compared to 28.1 +/- 7.7 pg/dl for cage controls, reaching a maximum at 15-30 min (254.6 +/- 46.8 pg/dl) before returning to near resting levels by 120 min (31.2 +/- 10.2 pg/dl). However, when non-tube-trained animals were submitted to loose-tube immobilization, these animals demonstrated significantly higher (3-10-fold greater) hormone levels at 120 min than their tube-trained counterparts (313.5 +/- 54.8 compared to 31.2 +/- 10.2 pg/dl; corticosterone, 12.2 +/- 6.2 microg/dl compared to 37.1 +/- 6.4 microg/dl). Hormone levels in exposed animals were also compared to those in swim-stressed animals. Swimming stress also resulted in marked elevation in both ACTH and corticosterone levels, which were 10-20 fold higher (541.8 compared to 27.2-59.1 pg/dl for ACTH) and 2-5 fold higher (45.7 compared to 8.4- 20.0 microg/dl for corticosteroids) than the cage control animals. Three time-averaged brain SAR levels of 0.16, 1.6 and 5 W/ kg were tested in a single 2-h RF-field exposure to the Iridium cell phone field. When RF-exposed and sham-exposed (immobilized) animals were compared, no differences were seen in core body temperature, corticosterone or ACTH that could be attributed to near-field RF radiation. Levels of Odc, Fos and Jun mRNA were also monitored in brains of animals exposed to the RF field for 2 h, and they showed no differences from sham-exposed (loose-tube immobilized) animals that were due to RF-field exposure. These data suggest that a significant stress response, indicated by a transient increase in core body temperature, ACTH and corticosterone, occurred in animals placed in even the mild loose-tube immobilization required for near-field RF exposure employed here and in our other studies. Failure to adequately characterize and control this immobilization response with appropriate cage control animals, as described previously, could significantly mask any potential effects mediated by the RF field on these and other stress-related parameters. We conclude that the pulse-modulated digital Iridium RF field at SARs up to 5 W/kg is incapable of altering these stress-related responses. This conclusion is further supported by our use of an RF-field exposure apparatus that minimized immobilization stress; the use of conditioned/tube-trained animals and the measurement of hormonal and molecular markers after 2 h RF-field exposure when the stress-mediated effects were complete further support our conclusion.

Alarm pheromone enhances stress-induced hyperthermia in rats

Behavioral and physiological effects of alarm pheromones emanating from stressed conspecific animals were investigated. Experimentally naive male Wistar rats were exposed to the test chambers containing alarm pheromones, which had been released by other rats receiving foot shocks in the same chamber beforehand. Along with behavioral analysis, the heart rate (HR) and core body temperature (cBT) were measured simultaneously using a biotelemetory system. Exposure to the alarm pheromones increased freezing, sniffing and walking and decreased resting as compared with rats exposed to control odors. In addition, these pheromone-exposed animals showed consistent increases in body temperature, i.e., stress-induced hyperthermia. After exposure to the alarm substances, immunoreactivity to nuclear Fos protein in the mitral cell layer in the accessory olfactory bulb (AOB) also increased compared with the reaction to control odors. These results suggest that an alarm pheromone enhances stress responses of conspecific animals both behaviorally and physiologically, and that these effects are mediated via activation of the AOB.

Fever and motor activity in rats following day and night injections of Staphylococcus aureus cell walls

Body temperature and physical activity are affected by both circadian cycles and pyrogens. We injected intraperitoneally 2.5 x 10(9) cell walls of the gram-positive organism Staphylococcus aureus or sterile saline at three different times in the circadian temperature and activity rhythm of Sprague-Dawley rats. Irrespective of whether pyrogen injections were made when the rats were inactive (injection at 0900), just before the nighttime rise in activity and body temperature (1630), or during high activity (2100), the peak body temperature attained and the time to reach peak temperature were indistinguishable. The fever response, as measured by the thermal-response index, was greatest, however, when body temperature and activity were in the lowest phase. Physical activity was inhibited by night but not day injection of S. aureus. Our results provide the first description of experimental fever resulting from a gram-positive pyrogen in rats and the first time an aspect of sickness behavior (suppressed motor activity) has been associated with fever resulting from simulated gram-positive bacterial infection.

Stress fever magnitude in laboratory-maintained California ground squirrels varies with season

A previous study demonstrated that California ground squirrels (Spermophilus beecheyi) living in the natural environment had, independent of season, a significantly higher mean diurnal body temperature (T(b)) (39.6 degrees C) than either summer (37.5 degrees C) or winter (36.5 degrees C) laboratory maintained animals. Based upon the previous study it has been suggested that California ground squirrels living in the natural environment may have an elevated set-point for body temperature in a manner analogous to a stress fever response. The present study was conducted to determine if season and/or duration of laboratory open-field exposure influenced the magnitude of laboratory open-field stress fever. If stress fever was involved to some extent in the higher body temperature observed in animals from the natural environment, laboratory maintained animals should exhibit a lower magnitude stress fever during the summer months and a higher magnitude stress fever during the winter months. It was hypothesized that laboratory maintained animals would exhibit the same set-point for stress fever T(b) independent of season, and that the duration of open-field exposure would not influence the magnitude of stress fever. Adult California ground squirrels were acclimated to an ambient temperature of 20+/-1.0 degrees C under either LD 14:10 (summer) or LD 10:14 (winter) photoperiod conditions and individuals from both photoperiod conditions were exposed for periods of 2, 4, and 6 h to an open-field arena. An analysis of the data with a two-factor ANOVA demonstrated that season (photoperiod) significantly influenced the magnitude of the stress fever response (1.1+/-0.1 degrees C for summer animals; 2.1+/-0.2 degrees C for winter animals) while there was no significant influence of open-field exposure duration on stress fever magnitude. These results demonstrate that although the set-point for body temperature in unstressed laboratory maintained California ground squirrels varies with season, the set-point for body temperature in open-field stressed animals does not vary with season. These data lend support to the hypothesis that something like stress fever may play some role in the higher body temperature observed in California ground squirrels living in the natural environment.

Pre-existing bacterial infections, not stress fever, influenced previous studies which labeled Gerrhosaurus major an afebrile lizard species

Early studies indicated that the Sudan plated lizard, Gerrhosaurus major, did not exhibit a febrile response when challenged with bacterial pyrogen. More recent results indicated that a 14-day antibiotic treatment regime produced a significant decrease (0.5 +/- 0.1 degree C) in the mean selected body temperature (MSBT) for this species (31.3 +/- 0.2-30.8 +/- 0.2 degrees C). The antibiotic treatment results suggested that G. major already had a fever caused by a pre-existing bacterial infection. The current study was designed first to determine if a sub-population of G. major, with a higher mean pre-antibiotic treatment MSBT would exhibit a greater decrease in MSBT after antibiotic treatment. A 14-day antibiotic treatment regime for G. major (N = 7) with MSBTs > or = 31.9 degrees C (mean 32.4 +/- 0.2 degrees C) produced a significant decrease of 1.7 +/- 0.4 degrees C in MSBT to 30.7 +/- 0.3 degrees C. Analysis of the combined antibiotic treatment MSBT data from [13] and the current study demonstrated that the magnitude of the change in MSBT after antibiotic treatment was dependent upon the pre-antibiotic treatment MSBT. These data imply that animals with a greater pre-treatment MSBT and greater MSBT change had a greater magnitude fever. In the second portion of this study the MSBT for individual lizards was measured during separate experiments using both indwelling cloacal thermocouples taped to the tail of the lizard and telemeters implanted into the peritoneal cavity of the lizard. This second study was designed to determine if measurement of Tb using thermocouples induced a stress fever which may have masked a portion of the pyrogen-induced fever. The MSBT measured using indwelling cloacal thermocouples (30.5 +/- 0.3 degrees C) was not significantly greater (T > 0.05) than the MSBT increased using telemeters (31.0 +/- 0.2 degrees C). The results of the experiments from this study demonstrate that the previously reported afebrile state for G. major was the result of animals having pre-existing bacterial infections. G. major does exhibit a febrile response similar to other lizard species.

Role of AVP in mediating the altered core temperature response to a simulated open field in pregnant rats

Near the term of pregnancy, rats have an attenuated core temperature response on exposure to a novel environment (e.g., a simulated open field) compared with that observed early in pregnancy or in nonpregnant rats. The present experiments were carried out on 26 nonpregnant and 26 pregnant rats to test the hypothesis that arginine vasopressin, functioning as an endogenous antipyretic substance in the central nervous system, mediates this attenuated core temperature response. Exposure to a simulated open field after intracerebroventricular (ICV) vehicle produced a significant increase in core temperature in both nonpregnant and pregnant animals, the magnitude and duration of which were greater in the nonpregnant rats. In nonpregnant rats, exposure to a simulated open field after ICV vasopressin V(1)-receptor antagonist altered the pattern of the core temperature response but not the core temperature index compared with that observed on exposure to a simulated open field after ICV vehicle. In pregnant animals, ICV vasopressin V(1)-receptor antagonist did not alter the core temperature response to a simulated open field compared with that observed after ICV vehicle. Thus our data do not support the hypothesis that a pregnancy-related activation of arginine vasopressin attenuates the core temperature response to a simulated open field in rats near the term of pregnancy.

Gender influences the core temperature response to a simulated open field in adult guinea pigs

The induction of psychological stress is often accompanied by a transient increase in core temperature, commonly referred to as stress induced hyperthermia. Although stress-induced hyperthermia occurs when rats, mice, and pigs are exposed to a novel stimulus (e.g., a simulated open field, restraint, etc.), whether or not it occurs in guinea pigs has not been investigated. The present experiments were therefore carried out to investigate the thermoregulatory responses of both male (n = 7) and female (n = 7) adult guinea pigs when they were exposed to a simulated open field. Unexpectedly, neither the male nor female guinea pigs developed stress-induced hyperthermia. To the contrary, female but not male guinea pigs significantly decreased their core temperature during an open field experiment. The mechanism of the gender-specific thermoregulatory response of the adult guinea pig to psychological stress is presently unknown.

Coordination of autonomic and behavioral thermoregulatory responses during exposure to a novel stimulus in rats

The induction of psychological stress in rats is accompanied by an elevation of core temperature. Our experiments were carried out to determine whether the latency, duration, magnitude, or effector mechanisms of the core temperature response to psychological stress would be altered when rats were allowed to use behavioral as well as autonomic thermoregulation. Core temperature, oxygen consumption, and ambient temperature were measured in adult rats before and after handling and a sham intraperitoneal injection. Seven rats were studied in a thermocline (gradient of 7 to 42 degrees C) and eight rats were studied in a metabolic chamber (25 degrees C). The rats studied in the thermocline selected a warm ambient temperature following the sham intraperitoneal injection and exhibited an increase in core temperature of shorter latency, greater magnitude, and greater duration than those studied in the metabolic chamber. The rats studied in the metabolic chamber exhibited an oxygen consumption response of greater magnitude and duration than the animals studied in the thermocline. Thus the characteristics in addition to the effector mechanisms of the core temperature response to psychological stress are altered when rats are allowed to use behavioral as well as autonomic thermoregulatory effectors.

Testosterone modifies response to chronic heat exposure in rats

Eight weeks of heat exposure (34 +/- 0.5 degrees C) in sham-orchiectomized rats leads to an increase of body temperature, slowing of body growth rate, and decrease of serum corticosterone level, as compared with animals maintained at 21 +/- 2 degrees C. Orchiectomy decreases body temperature, slows growth rate, and increases plasma corticosterone concentration both in control and heat exposed animals. Testosterone administration reverts these parameters to initial values. We conclude that testosterone plays a role in the regulation of heat balance in male rats.

California ground squirrel body temperature regulation patterns measured in the laboratory and in the natural environment

Body temperature (Tb) was measured by telemetry in both laboratory maintained and natural environment California ground squirrels, Spermophilus beecheyi. Laboratory animals had a mean diurnal Tb of 37.5 degrees C under conditions of LD 14:10, 20 degrees C and 36.5 degrees C under conditions of LD 10:14, 20 degrees C (P < 0.01). Nocturnal mean Tbs were 37.1 and 35.2 degrees C, respectively (P < 0.05). Mean diurnal Tbs for each animal in the natural environment ranged from 39.3 to 40.1 degrees C (mean = 39.6 degrees C) during both study seasons which included the hot season months of March through August and the cool season months of December through February. Natural environment hot season mean Tb was not significantly different from cool season mean Tb but both mean Tbs were significantly different from the diurnal mean Tbs measured in the laboratory (P < 0.05). California ground squirrels exhibit an open-field stress induced hyperthermia in the laboratory which can be extended for periods up to 6 h. The hyperthermic response is blocked by L-propranolol at a dosage of 15 mg kg-1. Laboratory animals do not habituate to repeated open-field exposures over a five consecutive day period. It is suggested that stress hyperthermia might be a normal component of thermoregulation in some free-living ground squirrels because of the openness of the habitat in which they exist.

Normal body temperature of rats: the setpoint controversy

Emotional hyperthermia, circadian variations and the rise of body temperature related to exercise, have all been attributed to setpoint temperature shifts. The accepted theory holds that core temperature is regulated by corrective thermoregulatory responses opposing the core temperature deviations from the setpoint level. However, in fever and anapyrexia the thermoregulatory responses appear to be not corrective but helping, that is in the same direction as the core temperature deviation. A supplementary ad hoc hypothesis that setpoint level shifts explains why the thermoregulatory responses still could be considered "corrective" in spite of being in the same direction as the core temperature deviation. But supplementary ad hoc hypotheses immunize a theory to experimental challenges and therefore can no longer be considered a scientific theory. The present work shows that most of the arguments adduced to explain almost every biothermal phenomenon as being due to setpoint shifts cannot withstand a critical analysis.

Endotoxin tolerance does not alter open field-induced fever in rats

Exposure to an open field has been shown to cause a rise in the body temperature of rats. In many respects, this rise in body temperature is similar to fevers caused by endotoxin and other inflammatory stimuli. Rats repeatedly injected with endotoxin develop tolerance to the fever-inducing action of endotoxin. We hypothesized that repeated pretreatment with endotoxin would modify the fever caused by exposure to psychological stress. To test this hypothesis, we compared open field-induced fevers in rats made endotoxin tolerant to those rats not endotoxin tolerant. We found that endotoxin tolerance had no effect on open field fevers.

I. Role of the pineal gland in multiple sclerosis: a hypothesis

Despite intensive research over the past several decades, the etiology and pathogenesis of multiple sclerosis (MS) remain elusive. The last 20 years have seen only meager advances in the treatment of the disease in part because too much attention has been devoted to the process of demyelination and its relationship to the neurologic symptoms and recovery of the disease. A host of biological phenomena associated with the disease involving interactions among genetic, environmental, immunologic, and hormonal factors, cannot be explained on the basis of demyelination and, therefore, require refocusing attention on alternative explanations, one of which implicates the pineal gland as the pivotal mover of the disease. This review summarizes the evidence linking dysfunction of the pineal gland with the epidemiology, pathogenesis, clinical manifestations, and course of the disease. The pineal hypothesis of MS also provided the impetus for the development of a novel and highly effective therapeutic modality, one that involves the transcranial application of AC pulsed electromagnetic fields in the picotesla flux density.

Effects of handling and physical restraint on rectal temperature, cortisol, glucose and leucocyte counts in the silver fox (Vulpes vulpes)

The present paper describes the effects of handling and one hour of physical restraint on rectal temperature (Trec), plasma cortisol, plasma glucose and leucocyte counts in six 8-months old silver fox vixens (Vulpes vulpes). Mean Trec in silver foxes 5 min after capture was 40.1 degrees C and increased during restraint, showing a maximum of 40.8 degrees C at 30 min thereafter. Supplementary, deep body temperature (Tb) was recorded with surgically implanted biotelemetry devices in 6 adult silver fox vixens kept isolated from environmental disturbances in a barn. Mean Tb in these foxes ranged between 38.0-38.4 degrees C, showing a diurnal variation and being at the lowest between 0700-1600 hour. When a person approached a fox and was present for 5 min, Tb increased rapidly. The results indicated that a stress-induced hyperthermia (SIH) was evoked rapidly within the first registration at 5 min after capture, and that this response continued during one hour of physical restraint. Plasma glucose and plasma cortisol levels increased during one hour of physical restraint, whereas numbers of lymphocytes, total white blood cell counts, and total granulocytes decreased. Furthermore, previously reported base levels of plasma cortisol and plasma glucose were exceeded. The results indicate that the hypothalamic-pituitary-adrenal (HPA) axis and the sympathoadrenal medullary (SAM) system were activated within 5 min of handling and restraint. Furthermore, hyperthermia is a promising indicator of acute stress in silver foxes.

Suprachiasmatic nuclei lesions do not eliminate homeostatic thermoregulatory responses in rats

Electrolytic lesions aimed at the suprachiasmatic nuclei (SCN) were made in male Long-Evans rats. Body temperature (Tb), activity, and drinking were monitored continuously in a 12-h light:12-h dark (12:12 LD) cycle at an ambient temperature of 23 degrees C. Large SCN lesions eliminated activity and drinking rhythms and abolished or reduced the circadian rhythm of Tb. The Tb responses of the rats were measured in L after exposure to cold and injection of lipopolysaccharide (LPS), a fever-producing drug, and in both L and D during a 30-min exposure to a novel cage. Rats with SCN lesions (SCNX) maintained their Tb as well as did controls during 2-h exposure to 2 degrees C. They also showed the expected increases in Tb in response to novelty and LPS. Nevertheless, there were differences between SCNX rats and other rats. When measured 9 h after LPS injection, SCNX rats had lower Tb in D than did sham-lesioned or intact rats or rats with lesions that missed the SCN. This is not surprising; the Tb of SCNX rats does not go as high as that of intact rats in D. However, it was surprising that at night SCNX rats increased their Tb in response to novelty (lights on in the test situation), whereas normal rats did not. For some reason, light inhibits the Tb rise to novelty in normal rats but does not do so in rats with SCN lesions.

Pregnancy alters body-core temperature response to a simulated open field in rats

Exposure of a rat to a novel environment (e.g., a simulated open field) induces a transient increase in body-core temperature, which is often called stress-induced hyperthermia. Although pregnancy is known to influence thermoregulatory control, its effect on stress-induced hyperthermia is unknown. Therefore, 24 Sprague-Dawley rats (8 nonpregnant and 16 pregnant) were studied to test the hypothesis that pregnancy would alter the development of stress-induced hyperthermia after exposure to a simulated open field. Body-core temperature index increased significantly after exposure to a simulated open field in nonpregnant and gestation day-10 rats but not in gestation day-15 and day-20 rats. Thus our data provide evidence that pregnancy influences the body-core temperature response of rats exposed to a simulated open field in a gestation-dependent fashion. The functional consequences as well as the mechanisms involved remain to be determined.

Cyclophosphamide cystitis as a model of visceral pain in rats: minor effects at mesodiencephalic levels as revealed by the expression of c-fos, with a note on Krox-24

The evoked expression of the immediate-early gene-encoded proteins c-Fos and Krox-24 was used to study activation of mesodiencephalic structures as a function of the development of cyclophosphamide (CP) cystitis in behaving rats. This article is the third of a series and completes previously published data obtained at both spinal and hindbrain levels. CP-injected animals received a single dose of 100 mg/kg i.p. under transient volatile anesthesia and survived for 1-4 h in order to cover the entire postinjection period during which the disease develops. Survival times longer than 4 h were not used owing to ethical considerations. Results from CP-injected groups are compared with those from either noninjected controls or saline-injected animals having survived for the same times as CP-injected ones. Quantitative results come from c-fos expression. At mesodiencephalic levels a high and widespread basal c-fos expression was observed in control animals; maximum staining was observed at the midthalamic level. Four groups of nuclei were identified with regard to the density of staining. The first group included nuclei showing clustered, intensely labeled cells; these areas were restricted in extent and related to the maintenance of circadian rythms (intergeniculate leaf, suprachiasmatic nucleus, dorsal parts of either paraventricular thalamic nuclei or central gray), sleep-arousal cycle (supramamillary nucleus), or changes in arterial pressure (laterodorsal tegmental nucleus). The second group included nuclei showing scattered, moderately labeled cells; these areas were widespread at all rostrocaudal levels and related to either autonomic/neuroendocrine regulations (central gray, lateral habenula, hypothalamus) or motor behavior, orienting reflex and oculomotor coordination (unspecific subdivisions of both colliculi and their adjoining mesencephalic regions, zona incerta dorsal). The third group included nuclei with evenly distributed, faintly labeled cells; these areas, which, with few exceptions, covered almost the entire diencephalon, mainly concerned nuclei of multisensory convergence having functions in either discriminative tasks (laterodorsal and lateroposterior thalamic nuclei) or emotional responses (intralaminar and midline thalamic nuclei). The fourth group included nuclei free of labeling; these were areas that received the bulk of unimodal sensory/motor inputs (central inferior colliculus, pretectal optic nuclei, ventral medial geniculate nucleus, ventral anterior pretectal nucleus, dorsal lateral geniculate nucleus, ventrobasal complex; zona incerta ventral, parafascicular thalamic nucleus) and are thus the most discriminative regarding specific modalities. Variations in staining were of the same magnitude in both saline- and CP-injected animals. A sequential study spanning every postinjection hour revealed maximum staining at 1 h postinjection, which was followed by a progressive, time-related decrease. Increases in the number of labeled cells 1 h postinjection were significant in only a restricted number of nuclei showing low basal expression (Edinger-Westphal nucleus and paraventricular, supraoptic, and lateral hypothalamic nuclei); time-related reductions in staining that were correlated to sleep or quiescence behaviors finally resulted in staining equal to or below that seen in control animals. No structures showed significantly increased staining in relation to the full development of cystitis, i.e., with the increase of visceronociceptive inputs. Comparing the present results with those previously obtained at more caudal levels, it appears that subtelencephalic levels primarily driven by visceronociceptive inputs, i.e., those that increase and/or maintain their activity in parallel with the degree of nociception, are confined to brainstem-spinal cord junction levels and only comprise certain subdivisions of the nucleus of the solitary tract (nucleus medialis, nucleus commissuralis, and ventralmost part of area po

Central mechanism of neural activation with cold acclimation of rats using Fos immunohistochemistry

The expression of Fos protein in the rat diencephalon, brain stem, cerebellum, and spinal cord was investigated using immunohistochemistry during chronic cold exposure, in order to clarify the neural regions involved in the thermoregulatory responses and the central mechanism of neural activation with cold acclimation. Numerous Fos-positive cells were observed in many brain regions after cold exposure and changes in the number of Fos-positive cells were analyzed quantitatively. Fos-positive regions were classified into three groups on the basis of the expression period of Fos protein. The first group was where a significant number of Fos-positive cells were seen 3 h and 24 h after cold exposure, but not observed 14 days after exposure; the regions included the lateral septal nucleus (LS), parvocellular paraventricular hypothalamic nucleus (pPVN), posterior hypothalamic area (PH), supramammillary nucleus (SuM), locus coeruleus (LC), dorsal tegmental nucleus (DTg), vestibular nucleus (Ves), and nucleus of solitary tract (Sol). The second group was where a significant number of Fos-positive cells were found 3 h, 24 h and 14 days after cold exposure; the regions included the preoptic hypothalamic area (POA), paraventricular thalamic nucleus (PV), lateral preoptic area (LPO), zona incerta (ZI), subparafascicular thalamic nucleus (SPF), lateral dorsal central grey (CGLD), lateral ventral central grey (CGLV), microcellular tegmental nucleus (MiTg), lateral lemniscus nucleus (LL), dorsal parabrachial nucleus (DPB), and the cerebellum. The third group was where Fos-positive cells were more numerous 14 days after cold exposure than they were after 3 h and 24 h of exposure; these regions included the ventromedial hypothalamic nucleus (VMH) and the spinal cord. These results demonstrate that the numbers and regions of Fos-positive cells in the rat brains changed during chronic cold exposure, and such changes may reveal the cellular adaptation of the thermogenic responsive neurons in the rat brain to cold acclimation.

Alterations in sensitivity to intracerebral vasopressin and the effects of a V1a receptor antagonist on cellular, autonomic and endocrine responses to repeated stress

We have shown previously that repeated restraint stress results in differential adaptation at both macrophysiological and cellular levels. Chronic stress accentuates vasopressinergic control of adrenocorticotropic hormone secretion in the pituitary. The present work determined whether endogenous vasopressin plays a role in response to repeated restraint. The first experiment explored changes in the response of repeatedly stressed animals to intracerebral vasopressin infusions. The second determined the effect of pretreating rats with a vasopressin V1a receptor antagonist on the way that they adapted to repeated restraint. Experiment 1: rats were subjected either to daily 60-min restraint for 10 days or transferred to the testing room where restraint sessions took place (controls). On the 11th day, they were infused with either artificial cerebrospinal fluid or 250 pmol vasopressin. The behavioural response to vasopressin was unaltered by previous stress. Plasma corticosterone was lowered in vasopressin-treated rats only after previous stress. Sixty minutes after vasopressin infusion, the central amygdala, locus coeruleus, the nucleus of the solitary tract and the dorsal vagal nucleus expressed increased levels of c-fos, and there were significant two-way interactions between stress and infusion for dorsal paraventricular nucleus, locus coeruleus and dorsal vagal nucleus. One-way analysis suggested that previous stress also reduced the c-fos response to vasopressin in the nucleus of the solitary tract. These results show that previous stress causes differential alterations in behavioural, endocrine and cellular responses to vasopressin. Experiment 2: rats were implanted with a transmitter which monitored heart rate and core temperature and a lateral cerebroventricular cannula. For 10 days, either artificial cerebrospinal fluid or 2500 pmol V1a antagonist, [d(CH2)1(5)-O-Me-Tyr2-Arg8]-vasopressin were infused i.c.v. 10 min prior to a 60-min restraint session. On the 11th day, no infusions were carried out, but rats received the usual period of restraint. The vasopressin antagonist was followed by motor responses (freezing, grooming and burrowing), more evident during the third and fifth days of stress. Core temperature responses were altered by the antagonist: stress-induced hypothermia was greatly reduced. Reduced baseline core temperatures, observed in controls as successive stress proceeded, were absent in antagonist-treated rats. By contrast, there were no significant effects of vasopressin antagonism on stress-induced tachycardia, nor in the way that this adapted to repeated restraint. On the 11th day (no i.c.v. infusions), hypothermic responses were no different in rats previously receiving either antagonist or control vehicle, but secondary hyperthermia was greater in the first group. Corticosterone levels were not altered by previous i.c.v. infusions.(ABSTRACT TRUNCATED AT 400 WORDS)

The effect of long-term castration on the neuronal and physiological responses to acute or repeated restraint stress: interactions with opioids and prostaglandins

Sixty min supine restraint stress induced a marked, but transient, hypothermic response in intact male rats, which tended to recover towards pre-stress levels or slightly overshoot during the later stages of restraint. Castration reduced the initial hypothermia but increased overshoots. Baseline (pre-stress) core temperature was also higher in castrated than intact rats, but the reduction in stress-induced hypothermia was still present even when this difference had been taken into account. The hypothermic response was not altered during the course of 10 sessions of daily repeated restraint in either intact or castrated rats. Castration did not alter cardiac responses to restraint. Both intact and castrated rats showed marked tachycardia during the initial 12 min of restraint, followed by a gradual fall towards baseline values. Repeated restraint accentuated the second phase of the cardiac response, without modifying the initial tachycardia, in both intact and castrated animals. The response of blood corticosterone levels to the first period of restraint was unaltered by castration but the attenuation observed after 10 sessions of stress was more complete in castrated rats. The neuronal c-fos response 60 min after the last of the series of repeated restraints was less in the hypothalamic paraventricular nucleus, medial amygdala, and locus coeruleus compared with that following the first session, but not in the lateral septum or the bed nucleus of the stria terminalis. Castration did not change the c-fos profile following the same number of restraint sessions. Castration depleted completely the vasopressinergic innervation in the lateral septum, diagonal band of Broca and medial amygdala.(ABSTRACT TRUNCATED AT 250 WORDS)

Regional changes in c-fos expression in the basal forebrain and brainstem during adaptation to repeated stress: correlations with cardiovascular, hypothermic and endocrine responses

Acute stress is known to evoke a discrete pattern of c-fos expression in the brain. The work reported here shows that this pattern is modified in regionally specific ways following repeated stress, and that this can be correlated with changes in telemetered heart rate, core temperature and corticosterone output that occur during adaptation. Intact male rats were restrained for 60 min daily for one or 10 days. Stress-induced tachycardia was maximal 10 min following the onset of restraint, and decreased thereafter. The peak value was not altered by repeated restraint, but levels fell towards baseline values more rapidly with increasing bouts of stress. Core temperature showed marked reduction during the first 10 min of the initial stress, followed by a minor (and not very consistent) overshoot during the remainder of the stress period. In contrast to heart rate, stress-induced hypothermia did not alter during repeated restraint. Corticosterone was raised dramatically immediately following the first 60-min session of restraint, and this was attenuated by repeated stress. Sixty minutes after the end of the first stress session, there was pronounced c-fos expression in the lateral septum, lateral preoptic area, lateral hypothalamic area, all divisions of the hypothalamic paraventricular nucleus, the medial (but not central) amygdala, the locus ceruleus and a brainstem structure (thought to be Barrington's nucleus), compared to rats transferred to the testing room but not restrained. Sixty minutes after the 10th stress session, c-fos expression was markedly decreased in some of these areas compared with the pattern observed after the first stress, especially in the paraventricular nucleus (dorsal and medial parvicellular regions) and in medial amygdala. However, all other areas measured demonstrated a sustained response even after repeated stress. There were no significant differences in c-fos expression in rats repeatedly transferred to the testing room (but not stressed) compared to singly transferred counterparts. These results show that both neuronal and physiological responses adapt to a repeated stress, but that in both cases this has highly specific components. It seems likely that adaptive changes in c-fos expression are associated with those in some features of autonomic and endocrine reactions. It is noteworthy that there is evidence that the lateral septum, in which c-fos expression did not diminish after repeated stress, may be involved in temperature control, whereas the paraventricular nucleus, in which c-fos did alter, has been linked with both cardiac and corticoid regulation.

Differences in Fos expression in the rat brains between cold and warm ambient exposures

Fos expression in the rat diencephalon, brain stem, cerebellum, and spinal cord was examined after warm (33 degrees C) and cold (10 degrees C) ambient exposures. Fos expression was examined with use of immunohistochemical method and the number of Fos-positive neurons in each nucleus was quantitatively analyzed. When rats were exposed to cold ambient, significant number of Fos-positive neurons was found in the lateral septal nucleus (LS), preoptic hypothalamic area (POA), parvocellular paraventricular hypothalamic nucleus (pPVN), lateral preoptic area (LPO), zona incerta (ZI), paraventricular thalamic nucleus (PV), ventromedial hypothalamic nucleus (VMH), subparafascicular thalamic nucleus (SPF), posterior hypothalamic area (PH), supramammillary nucleus (SuM), microcellular tegmental nucleus (MiTg), lateral lemniscus nucleus (LL), lateral dorsal central grey (CGLD), lateral ventral central grey (CGLV), dorsal parabrachial nucleus (DPB), locus coeruleus (LC), dorsal tegmental nucleus (DTg), vestibular nucleus (Ves), nucleus of solitary tract (Sol), spinal cord, and cerebellum. When animals were exposed to warm ambient, the numbers of Fos-positive neurons in the LS, POA, PV, LPO, and SuM were significantly increased to be equal to those of cold ambient. However, after warm ambient exposure the numbers of Fos-positive neurons in the DPB and spinal cord were increased but less than those of cold ambient, and those in the pPVN, VMH, ZI, SPF, PH, CGLD, CGLV, MiTg, LL, LC, DTg, Ves, Sol, and cerebellum were not significantly increased as compared with those of control or cold ambient. Abdominal temperature was not changed during cold ambient exposure, but the temperature was significantly increased during warm ambient exposure.(ABSTRACT TRUNCATED AT 250 WORDS)

24-hour control of body temperature in the rat: II. Diisopropyl fluorophosphate-induced hypothermia and hyperthermia

Diisopropyl fluorophosphate (DFP) and other anticholinesterase (antiChE) agents have been found to induce marked hypothermic responses in laboratory rodents. To characterize the effects of DFP on autonomic and behavioral thermoregulation, rats of the Long-Evans strain were injected with DFP while housed in a temperature gradient. The gradient allowed for the measurement of selected ambient temperature (Ta) and motor activity (MA) over a 6- to 7-day period. Core temperature (Tc) and heart rate (HR) were also monitored simultaneously using radiotelemetry. Injection of the peanut oil vehicle led to transient elevations in Tc, HR, and MA, but no change in selected Ta. The next day animals were injected with 0.25, 1.0, or 1.5 mg/kg DFP. DFP (1.0 AND 1.5 mg/kg) led to a marked reduction in Tc. The decrease in Tc was accompanied by reductions in HR, MA, and selected Ta. During the first night after DFP, selected Ta remained elevated as Tc recovered to its preinjection level. The second 24-h period after 1.0 and 1.5 mg/kg DFP was associated with a significant elevation in the daytime Tc. In conclusion, with the option of using behavioral thermoregulatory responses, the hypothermic effects of acute DFP treatment are mediated by a selection for cooler TaS. An elevation in Tc during recovery from acute DFP corroborates the many incidents of fever in humans exposed to anti-ChE agents.

The relationship between traumatic brain injury-induced changes in brain temperature and behavioral and anatomical outcome

Alteration of brain temperature, experimentally induced or spontaneous, has been shown to affect the symptoms resulting from a variety of cerebral insults. This study examined the effect of traumatic brain injury (TBI) on brain and body temperature in rats and the relationship between TBI-induced temperature changes, neuropathology, and behavioral recovery. Anesthesia, surgery and TBI all caused changes in brain and body temperatures. The level of brain (but not body) temperature at the time of TBI was positively correlated with the severity of hippocampal and thalamic pathology. In contrast, the measured levels of both brain and body temperatures after TBI were not related to behavioral or neuroanatomical outcome. Interestingly, the increase in brain (but not body) temperature from the time of TBI to 5 to 10 minutes after termination of anesthesia was negatively correlated with behavioral and anatomical outcome. Simply stated, the more rapidly brain temperature returned toward normal, the better the rats' behavioral and anatomical outcome. This rate of return toward normal brain temperature is not interpreted as causally related to outcome but rather as an index of the severity of brain injury.

Fever: causes and consequences

The present review distinguishes pathogenic, neurogenic, and psychogenic fever, but focuses largely on pathogenic fever, the hallmark of infectious disease. The data presented show that a complex cascade of events underlies pathogenic fever, which in broad outline - and with frank disregard of contradictory data - can be described as follows. An invading microorganism releases endotoxin that stimulates macrophages to synthesize a variety of pyrogenic compounds called cytokines. Carried in blood, these cytokines reach the perivascular spaces of the organum vasculosum laminae terminalis (OVLT) and other regions near the brain where they promote the synthesis and release of prostaglandin (PGE2). This prostaglandin then penetrates the blood-brain barrier to evoke the autonomic and behavioral responses characteristic of fever. But then once expressed, fever does not continue unchecked; endogenous antipyretics likely act on the septum to limit the rise in body temperature. The present review also examines fever-resistance in neonates, the blunting of fever in the aged, and the behaviorally induced rise in body temperature following infection in ectotherms. And finally it takes up the question of whether fever enhances immune responsiveness, and through such enhancement contributes to host survival.

Interactions between corticotropin-releasing factor and endogenous opiates on the cardioaccelerator, hypothermic, and corticoid responses to restraint in the rat

These experiments investigated the effect of either systemic opiate blockade by naloxone (5 mg/kg) or intracerebroventricular CRF (250 pmol), or the two treatments combined, on physiological and endocrine responses of male rats to two types of stress: restraint by itself (representing a psychological stress), and restraint combined with a tail clip (representing an additional mild physical nociceptive stress). Rats were restrained in a plastic container for 15 min, with or without a tail clip. Heart rate, body temperature, and serum corticosterone were measured. The first experiment showed that restraint induced marked tachycardia, maximal at 5 min, and declining thereafter. There was also a pronounced hypothermia, maximal at 10 min, and serum corticosterone was elevated 10 min after the end of the period of restraint. The presence of a tail clip increased the cardioaccelerator response, but had no effect on hypothermia. Naloxone had no effect on heart rate during restraint or on postrestraint corticosterone, but accentuated hypothermia. The effects of naloxone occurred independently of the presence of a tail clip. A subsidiary experiment showed that rats transferred to an unfamiliar cage showed a marked hyperthermic response, as described by others. The second experiment showed that CRF (250 pmol ICV) did not modify the tachycardiac response to restraint, but reduced hypothermia. This also occurred irrespective of the presence of a tail clip. The third experiment investigated the interaction between naloxone and CRF, and showed that the ameliorative effects of ICV CRF on restraint-induced hypothermia were prevented by systemic naloxone, but that neither tachycardia nor corticosterone responses were altered.(ABSTRACT TRUNCATED AT 250 WORDS)

Cold increases and warmth diminishes stress-induced rise of colonic temperature in rats

It is generally believed that the rise of core temperature of rats induced by handling is due to a shift of set-point temperature as in fever. Changes in core temperature due to set-point shifts should not be affected by changes in the ambient temperature. Nevertheless, when the colonic temperature of rats was taken in a cold environment the usual emotional rise was higher and when the colonic temperature was taken in a warm environment the emotional rise was lower. These results contradict the hypothesis that the emotionally induced rise in temperature of rats is a fever.

Differential effects of ACTH4-10, DG-AVP, and DG-OXT on heart rate and passive avoidance behavior in rats

A computerized telemetry system was used to monitor heart rate (HR), core temperature (CT), and gross locomotor activity in rats treated with saline or neuropeptides during a passive avoidance behavior task. Rats were exposed to a single mild footshock (0.15 mA, for 3 s). Retention tests were conducted at 24 and 48 h after the learning trial. One h prior to the 24-h retention test, each rat received one of the following treatments (SC): saline (SAL), desglycinamide [Arg8]-vasopressin (DG-AVP), ACTH4-10, or desglycinamide-oxytocin (DG-OXT), at a dose of 3 micrograms/rat for DG-AVP and DG-OXT, and 50 micrograms/rat for ACTH4-10. Rats treated with SAL showed a modest increase in avoidance latency accompanied by bradycardia at both retention tests. Rats receiving DG-AVP retained the highest avoidance latency among the experimental groups at both the 24- and 48-h retention test. These rats showed a decrease in HR of the same magnitude as the SAL-treated animals at both retention tests. Rats treated with ACTH4-10 showed an increase in avoidance latency during the 24-h but not during the 48-h retention test. In addition, following ACTH4-10 treatment, a tachycardiac response was found during the 24-h retention test. DG-OXT induced both behavioral and cardiac responses opposite to those found in rats given DG-AVP. CT gradually increased while the rats remained on the platform, irrespective of the treatment. Changes in HR and CT were not influenced by somatomotor activity, as no difference in gross locomotor activity was found among the groups.(ABSTRACT TRUNCATED AT 250 WORDS)

Handling elevates the colonic temperature of mice

Handling mice for repeatedly measuring their colonic temperature (Tc) resulted in a significant rise in their Tc. When the procedure was repeated day after day, this response diminished by habituation, showing the emotional origin of Tc rise. Salicylate lowered both the maximal Tc reached during handling and the Tc before handling without affecting the difference between the initial and the maximal Tc. During the first Tc measurement in a session, concomitant to the Tc rise, ear pinna temperature decreased. When Tc reached a plateau during the last measurements, ear temperature increased. This vasomotor response suggests that stress Tc rise is a regulated Tc change. However, since, contrary to what was reported in the rat, the salicylate did not diminish the magnitude of the Tc rise, it is doubtful that the emotional Tc rise in mice is a true fever.

Hippocampal and body temperature changes in rats during delayed matching-to-sample performance in a cold environment

In order to study the effects of temperature changes induced by cold stress on working memory, telemetry thermistor probes were implanted into the hippocampal region of the brain and into the peritoneal cavity of rats. Temperatures in these regions were monitored while rats performed on a delayed matching-to-sample (DMTS) task at ambient temperatures of 23 degrees C and 2 degrees C. Matching accuracy was significantly decreased during exposure to 2 degrees C, indicating a marked impairment of short-term or working memory. Temperature in the hippocampus increased 2 degrees C during exposure to 23 degrees C, but only 1 degrees C when the environmental temperature was 2 degrees C. Body temperature showed a similar but less pronounced pattern in that cold exposure attenuated the increase in temperature observed when animals performed the DMTS task. These results suggest that cold-induced impairment of working memory may be associated with subtle temperature changes in the brain.

The effect of physical restraint on IL-1 beta- and LPS-induced fever

The stress of physical restraint has been shown to cause an elevation in the body temperature of rats. In this study, we compared the febrile responses of restrained and unrestrained animals to the injection of interleukin-1 beta (IL-1 beta) and lipopolysaccharide (LPS). We found that, prior to injection, the body temperature of the restrained animals was about 1 degree C higher than that of the unrestrained rats. The restrained rats showed significantly smaller increases in body temperature in response to the injection of either pyrogen than did the unrestrained animals.

Lipopolysaccharide increases ambient temperature preference in C57BL/6J adult mice

The hypothesis was tested that animals exposed to a potentially dangerous endotoxin would attempt to behaviorally elevate their body temperature, perhaps in an effort to engage those immunological mechanisms which would counter the adverse effects of the endotoxin. Lipopolysaccharide (LPS) from Escherichia coli injected subcutaneously (100 micrograms) in adult C57BL/6J mice increased gradient temperature preference by 2.4 degrees C over saline controls. The increase in body temperature of 1.1 degrees C after LPS injection was due to the preference for higher ambient temperatures and was not the result of a systemic reaction to LPS (animals not exposed to the gradient did not differ in body temperature). In summary, our data indicate that adult mice self-induce a febrile response, perhaps as an attempt to compensate for the physiological impact of the endotoxin.

Stress hyperthermia: physiological arguments that it is a fever

The theory that stress (or emotional) rise in central temperature (Tc) in rats is a fever with an upward shift of the set-point temperature was tested with three experiments: 1) Measurement of tail skin temperature and Tc during the emotional Tc rise; 2) Investigation of the effect of ambient temperature on the emotional Tc rise; and 3) The assessment of emotional Tc rise during daytime and nighttime. Skin vasomotor responses helped the increase of Tc toward a higher level and contributed to the regulation of central temperature at this new higher level. The cold environment did not diminish the emotional rise of central temperature as it would be expected in the case of a hyperthermia. However, at night emotional fever reached a higher level than during the daytime, suggesting that prostaglandin rise in Tc is distinct from emotional or stress-induced hyperthermia. In conclusion, the experiments reported here confirm the hypothesis that the rise of Tc induced by handling or disturbance of the rats is regulated, and is due to a shift of the set-point as occurs in fever.

Activation of brown adipose tissue thermogenesis by chemical stimulation of the posterior hypothalamus

The posterior hypothalamus (PH) is involved in the generation of behavioral thermoregulatory responses, but the importance of the PH in the control of autonomic thermoregulatory responses such as heat production in brown adipose tissue (BAT) is not well defined. In the present study, selective stimulation of PH neurons by local application of the excitatory amino acid glutamate (250 nl of 1 M solution, unilaterally) caused a sharp, transient increase in interscapular BAT (IBAT) and core temperature in urethane-anesthetized rats. This effect was blocked by pretreatment with the sympathetic ganglionic blocker, chlorisondamine chloride (2 mg/kg) or the beta-adrenergic receptor blocker, propranolol (2 mg/kg), implicating the involvement of the sympathetic system. The effect of intra-PH injection of glutamate on IBAT and core temperatures could be mimicked by injection of the gamma-aminobutyric acid (GABA) receptor antagonist, bicucullin methiodide (BMI, 50 ng), into the same PH site. This effect of BMI could be blocked by co-injection of the GABA(A) receptor agonist, muscimol (25 ng). Further, BMI co-injection potentiated the effect of intra-PH injection of glutamate on IBAT and core temperatures. Conversely, muscimol co-injection prevented the stimulatory effect of intra-PH injection of glutamate. Taken together, the results indicate that direct chemical stimulation of neurons in the PH can activate an autonomic mechanism controlling heat production in BAT. Further, they suggest the neural mechanism in the PH mediating this effect is tonically inhibited by GABA, as blockade of GABAergic function in the PH produces an effect similar to that observed after direct stimulation of PH neurons with glutamate.(ABSTRACT TRUNCATED AT 250 WORDS)