Physiology & Behavior, Vol. 61, No. 3, pp.

447453, 1997
Copyright q 1997 Elsevier Science Inc.
Printed in the USA. All rights reserved
0031-9384/97 $17.00 / .00

PII S0031-9384(96)00459-3

The Effects of Voluntary Exercise and

Immobilization on Humoral Immunity and
Endocrine Responses in Rats
DEBORA R. BALDWIN, 1 ZACHARY C. WILCOX AND GUANPING ZHENG

Department of Psychology, University of Tennessee, Knoxville, TN 37996 USA

Received 25 August 1995; accepted 28 August 1996

BALDWIN, D. R., Z. C. WILCOX AND G. ZHENG. The effects of voluntary exercise and immobilization on humoral immunity
and endocrine responses in rats. PHYSIOL BEHAV 61(3) 447453, 1997.This research examined the effect of type of stressor
(physical vs. psychological) on humoral immunity and neuroendocrine responses in male and female rats. Eighty adult Sprague
Dawley rats were assigned to one of the four stress conditions (n 20 animals/group): 1. Voluntary running (high physical/low
psychological stress); 2. immobilization (low physical/high psychological stress); 3. mixed stress (running and immobilization);
and 4. cage control group. The experimental manipulations were conducted over a 6-week period for 4 h/day. Five weeks after
the start of the study, all animals were immunized with 1 ml of a 10% suspension of sheep red blood cells (SRBC) in saline and
sacrificed 1 week later. Data analyses revealed no main effect of stress on any of the immune or endocrine parameters. However,
strong gender differences emerged within the stress conditions on these physiological parameters. The stressed female rats displayed
an enhanced antibody response to SRBC and a higher percentage of peripheral blood lymphocytes than their male counterparts.
However, there were no significant differences between the male and female control animals with respect to these variables. Female
rats consistently displayed elevated levels of plasma corticosterone and adrenal norepinephrine across all conditions. In addition,
female rats displayed heavier relative organ weights (adrenal and spleen). Taken together, the notion of differential immunity
with respect to physical or psychological stress is not supported by the present study. q 1997 Elsevier Science Inc.

Stress Exercise Corticosterone Catecholamine Humoral immunity

A CONSIDERABLE amount of data has already been gathered
showing the influence of laboratory and naturalistic stressors on
immunocompetence in man and animals (1). In addition, there
is evidence showing that the specific characteristics of the stress-
ful event (e.g., type of stressor, duration of stressor, predictability
of the stressor) can produce immunosuppression and immu-
noenhancement (25). For example, Rinner and colleagues (42)
found that a single immobilization session (150 min) produced
a marked loss in peripheral blood lymphocyte proliferative re-
sponses to mitogens. However, repeated exposure to the same
stressor for 6 consecutive days yielded no significant difference
in lymphocyte function compared to that in the control animals.
Given the specificity of the immunological response to stress,
conflicting results tend to create ambiguity with regard to this
phenomenon.
The ambiguity that permeates the stress literature is due, in
part, to the lack of parametric studies (24). For example, one
aspect of the stress experience that has not been adequately in-
vestigated is stressor type. More specifically, research is scant
concerning the comparative aspects of physical and psychologi-
cal stress on immunocompetence. Both have been found to alter
the immune system (1,23,46). However, is there a difference in
the degree of immunological modulation with respect to either
stressor? In other words, does exposure to a stressor that is more
psychological in nature generate a greater immunological deficit
than a stressor that is primarily physical in nature?
It is interesting to note that aerobic exercise has been docu-
mented to enhance physical fitness and reduce stress levels in
humans (41,43). Furthermore, there is a widely held belief that
an increase in physical activity will result in greater resistance to
infectious diseases (14). Recent evidence is mounting showing
that moderate exercise tends to enhance, and intense exercise
tends to suppress, immune responses in man and animals
(31,39). For example, McCarthy and Dale (37) reported that a
30-min exercise bout (jogging, swimming, or playing squash)
produced a significant increase in peripheral blood leukocytes 2.5
h postexercise in healthy men. Tharps and Barnes (50) reported
salivary IgA concentration to be significantly reduced in com-
petitive swimmers after a strenuous workout. However, similar
to emotional stress, the clinical significance of these immunolog-
ical changes with exercise have yet to be determined.
It is well documented that the immune system does not work
autonomously. Numerous studies have shown the interaction be-
tween the neuroendocrine and immune systems (8,25). For ex-

1
To whom requests for reprints should be addressed.

447

/ av1 eh0a 2551 Mp 447 Friday Feb 14 04:43 PM ELPHB (v. 61, no. 3) 2551
448 BALDWIN, WILCOX AND ZHENG

ample, cortisol has been found to suppress T-cell function and have been found to induce immune and endocrine changes in
NK cell activity (35,38). Likewise, infusion of epinephrine has animals (17,27,40). All animals were weighed weekly.
been reported to inhibit lymphocyte proliferation to mitogen in
vitro (10). More importantly, plasma cortisol and catechol- Voluntary Running Condition
amines are indices of stress and play a critical role with regard
Animals in this condition were placed in a stainless-steel and
to this response (49). Therefore, it was of interest to further
Plexiglas activity wheel (lafayette Instrument, Lafayette, IN) for
examine the neuroendocrine response to physical and psycholog-
4 h/day. Each wheel was a 35.56 cm in diameter wire mesh drum
ical stress as it relates to immunomodulation.
that rotated on a ball-bearing axle attached to a hub. Each activity
The purpose of the present study was to systematically eval-
wheel contained a mechanical counter that recorded the number
uate the effects of a physical and psychological stressor on hu-
of revolutions in either direction when the rats would run or walk.
moral immunity in particular, and the endocrine correlates as-
Distance was recorded before and after each session. Initially, to
sociated with immunoregulation in general. Voluntary running
enhance running performance, animals were subjected to a slight
served as the purer physical stressor because animals are not
reduction of ad lib food intake (8%) for 1 week. At the end of
forced to run via electrical shock that may precipitate an en-
each session, the activity wheels were cleaned for the next day.
hanced psychological threat.
Furthermore, repeated voluntary wheel running, in conjunc-
Immobilization Condition
tion with an adequate food supply, has been shown to produce
anorexia (15) and decrease antibody titers in rats (21). Likewise, Animals in this stress condition were immobilized in venti-
immobilization served as the purer psychological stressor due to lated polypropylene rodent restraining tubes (Baxter Scientific
the general restriction of large muscle movement. It is worth Products, McGaw Park, IL) for 4 h/day for 6 weeks. The animals
noting that initial exposure to immobilization stress tended to were restrained and then placed in a wire group cage separate
precipitate the behavior of struggling. However, this activity from their home cage. After the stress session, each animal was
quickly diminishes within minutes (preliminary observation). removed from the tube and placed back in their home cage. All
A secondary purpose of this investigation was to examine restraining tubes were cleaned in preparation for the next session.
possible gender differences in immunity as it relates to stress
exposure. Research shows that females tend to display enhanced Mixed Stress and Cage Control Conditions
humoral and cell-mediated immunity compared with their male
Animals in the mixed stress condition were exposed to 2 h of
counterparts (2). However, there is limited information regard-
voluntary running and immobilization stress using a counterbal-
ing stress-induced immunological reactivity as a function of gen-
anced procedure (e.g., day 1, run followed by restraint; day 2,
der. Recently, Matthews and colleagues (36) reported immuno-
restraint followed by run). Animals in the cage control condition
logical alterations in women who displayed the greater
served as age-matched controls. They were handled during rou-
sympathetic reactivity to brief psychological stress. The findings
tine husbandry and data collection periods (e.g., body weight).
of gender differences with regard to sympathetic reactivity to
stress (18) may play a role in determining the immunological
Immunization
responses to stress. It was hypothesized that psychological stress
alone (immobilization) would produce the greatest compromise One week before the day of sacrifice, all animals were injected
in humoral immunity, and female rats would display the greatest IP with 1 ml of a 10% suspension of SRBC in saline. The SRBC
physiological reactivity to stress. were obtained from Lampire Biological Laboratory (Pipersville,
PA) and washed 3 times with saline prior to injection. The in-
jections were administered in the afternoon.
METHODS

Animals and Materials Blood and Organ Collection

Twenty-four hours after their last treatment exposure, all an-
Forty male and female adult SpragueDawley rats 55 days
imals were sacrificed by rapid decapitation on day 7 of postim-
old were obtained from a commercial breeder (Harlan Labora-
munization. This procedure was conducted during the morning
tories, Indianapolis, IN). All animals were housed in same sex
hours to avoid possible circadian influences. Truncal blood was
groups (2 per cage) in 48.26 cm 1 25.40 cm polypropylene cages
collected into centrifuge tubes and allowed to clot overnight at
that contained sawdust bedding. A standard 12-h light-dark cycle
47C. The blood samples were subsequently centrifuged at 700 1
was used, and all animals were allowed to acclimate to the AAA-
g for 15 min. The serum was drawn off and stored at 02007C.
LAC-approved facility. Tap water and standard rat chow were
At the time of sacrifice, the total white blood cell count (WBC)
available ad lib, except during the stress sessions.
was determined in 10 ml of whole blood using a Sysmex Micro-
After adjusting to the vivarium conditions (1 week), animals
cell counter y (Baxter). In addition, blood smears were stained
were weighed (initial weight for males 245 { 20 g; initial weight
(Diff Quick Hematological Stainy, Baxter) and used to deter-
for females 185 { 20 g) and randomly assigned to one of the 4
mine percent peripheral blood lymphocytes per 100 leukocytes.
following conditions (n 20 animals per condition): 1. Vol-
The spleen, thymus, and left adrenal gland were harvested,
untary running (high physical-low psychological stressor); 2.
cleaned, and wet weight taken. The adrenals were stored at
immobilization (low physical-high psychological stressor); 3.
07007C for subsequent analysis of catecholamine content.
mixed stress (running and restraint); and 4. cage control group.
An equal number of males and females were assigned to each
Serological Analysis
condition.
At 62 days of age, the animals were exposed to the experi- Antibody response to SRBC was measured by an antibody
mental manipulations, which were maintained for 6 consecutive titer as previously described by Baldwin et al. (6). The highest
weeks. All experimental manipulations were performed in the dilution giving hemoagglutination was considered to be the titer.
morning between 0800 and 1200 h to reduce diurnal fluctuation Plasma corticosterone was assayed in duplicate using a commer-
in endocrine levels. In addition, approximately 4-h stress sessions cial radioimmunoassay kit (ICN Biochemical, Irvine, CA). Cor-

/ av1 eh0a 2551 Mp 448 Friday Feb 14 04:44 PM ELPHB (v. 61, no. 3) 2551
STRESS AND IMMUNITY
FIG. 1. Mean antibody titer (log2 ) for the stress and control groups as a
function of gender. Each bar represents the mean { SEM. The females
were significantly different from the males across conditions, * p 0.05.
ticosterone levels were determined from a standard curve and
expressed in ng/ml.
Catecholamine Analysis
Each adrenal gland was placed in a 150-ml epidorf tube con-
taining 0.5 ml of 0.07 M perchloric acid (PCA) with 3,4-dihy-
droxybenzylamine hydrobromide (DHBA), which served as an
internal standard. The adrenal gland was sonicated with a micro-
probe in an ice-water bath until homogenized. Chloroform/n-
propanol (0.25 ml) and 0.5 ml of PCA/DHBA were added to
each sample and then sonicated. The samples were centrifuged
for 20 min at 10,000 rpm. The supernatant was diluted to 10%
by taking 25 ml of the original supernatant into 475 ml of 0.07
PCA/DHBA. Fifty microliters of the diluted supernatant was in-
jected onto the reverse-phase high-performance liquid chroma-
tography (HPLC) column (Phenomenex, Torrance, CA).
Adrenal catecholamine content was measured using electrochem-
ical detection with a mobile phase consisting of 40 ml of n-pro-
panol, 5 ml of 10% sodium octane sulfate, 3 ml of 10% Na2
EDTA, and 100 ml of 1.0N NH4H2PO4 .
RESULTS
Dependent measures were analyzed with a 4 (experimental
and control groups) 1 2 (male and female) between-groups anal-
ysis of variance (ANOVA). Any significant F value was further
analyzed with the Scheffe post hoc test or student t-test. To assess
the relationship between the immune and endocrine variables
with regard to stress, separate correlational analyses were per-
formed for males and females. The significance level was set at
p 0.05, and graphic illustrations were generated by Jandel Sci-
entific (Sausalito, CA).
Immune Measures
Figure 1 illustrates the antibody response to SRBC for male
and female rats across the stress and cage control conditions.
Although no main effect of stress was found (p 0.05), strong
gender differences emerged with respect to this immune measure
F(1,72) 14.51, p 0.001. The female rats displayed higher
antibody titers than the male rats. There was no significant inter-
action between the stress and gender variables (p 0.05). How-
ever, to assess whether or not this gender difference was stress-
induced, 2 separate t-tests were performed between the male and
female rats in the stressed (voluntary running, immobilization,
/ av1 eh0a 2551 Mp 449 Friday Feb 14 04:44 PM ELPHB (v. 61, no. 3) 2551
449
FIG. 2. Mean percentage of peripheral blood lymphocytes for the stress
and control groups as a function of gender. Each bar represents the mean
{ SEM. The females were significantly different from the males across
the conditions, * p 0.05.
and mixed) and cage control groups. Post hoc analysis indicated
that stressed female rats displayed a greater antibody response to
the antigen (SRBC) than their male counterparts in the same
conditions t(29) 03.77, p 0.001. However, there was no
significant gender difference in the cage control condition (p
0.10) on this immunological measure.
With regard to the percentage of peripheral blood lympho-
cytes, there was no significant main effect of stress (p 0.10).
However, a main effect of gender was significant F (1,72)
26.03, p 0.001. These data are presented in Fig. 2. Although
the interaction was not significant (p 0.10), 2 separate t-tests
were performed between the genders for the stressed and cage
control conditions. Similar to the antibody results, stressed fe-
male rats displayed a higher percentage of peripheral blood lym-
phocytes than their male counterparts t (29) 04.26, p 0.01.
There was no significant gender difference on this parameter with
the cage control group (p 0.10). The ANOVA revealed no
significant stress or gender difference in total WBC counts (p
0.05).
Endocrine Measures
Figure 3 illustrates the plasma corticosterone data. Although
no main effect of stress was found for this dependent measure,
FIG. 3. Mean serum corticosterone (ng/ml) concentration for the stress
and control groups. Each bar represents the mean { SEM. The females
were significantly different from the males across the conditions, * p
0.05.
450 BALDWIN, WILCOX AND ZHENG

TABLE 1
MEAN ADRENAL EPINEPHRINE VALUES (NG/GLAND) FOR MALE AND FEMALE RATS
ACROSS STRESS CONDITIONS (N 20/GROUP)

Sex Run Immobilization Mix Control

Male 11,170 (998) 18,632 (7348) 24,537 (1374) 9779 (1330)

Female 16,835 (1981) 15,849 (1677) 15,732 (1766) 15,568 (1669)

SEM is in parentheses.

there was a significant main effect of gender F(1,72) 21.98,
p 0.001. Female rats had significantly higher levels of basal
plasma corticosterone than the male rats. There was no significant
interaction with regard to this measure. Similar to the immuno-
logical data, 2 separate t-tests were performed between the sexes
within the stressed and cage control conditions. The analyses
indicated that the stressed female rats had higher levels of plasma
corticosterone than their male counterparts t (29) 04.35, p
0.01. However, there was a nonsignificant trend for the female
rats in the cage control condition to have higher basal plasma
corticosterone levels than their male counterparts (p 0.08).
The ANOVA revealed no significant main effect of stress or
gender on adrenal epinephrine content (see Table 1). However,
Fig. 4 illustrates the significant gender effect for adrenal norepi-
nephrine content F (1,72) 9.35, p 0.003 across all condi-
tions. Post hoc analysis, similar to the previous data, indicated
that female rats in the stressed conditions displayed elevated lev-
els of norepinephrine compared to that in the male rats in the
same conditions t (29) 02.58, p 0.04. There was a nonsig-
nificant trend for the female rats in the cage control condition to
display elevated levels of adrenal norepinephrine compared to
that in their male counterparts (p 0.08). The interaction be-
tween the stress and gender variables was not significant for any
of the catecholamine measures (p 0.05).
Weight and Activity Measures
Figure 5 displays the average body weight changes for each
of the stress groups as a function of gender. Body weight change
was determined by subtracting the initial weight from the final
weight. The ANOVA revealed a main effect of stress, F(3,72)
6.40, p 0.001, and gender, F (1,72) 205.43, p 0.001.
The Scheffes analysis indicated that animals in the control con-
dition gained significantly more weight than the animals in the
other conditions. Furthermore, the male rats gained considerable
amounts of weight compared with the female rats. However,
there was no significant interaction for this dependent variable
(p 0.05).
Table 2 displays the average relative organ weight data for
each stress group as a function of gender. A main effect of stress
was significant for relative spleen weight only F(3,72) 4.73,
p 0.01. Post hoc analysis indicated that animals in the run
condition displayed smaller spleens compared with animals in
the immobilization conditions. No other significant effect of
stress emerged. However, significant gender differences were
found for relative adrenal F(1,72) 95.01, p 0.001, spleen
F(1,72) 114.51, p 0.001, and thymus F (1,72) 14.53, p
0.001 weights. Female rats tended to display heavier organ
weights than the male rats across all conditions.
Correlational Measures
To determine the relationships between the immune and en-
docrine measures, correlational analyses were performed for each
gender collapsed across the stress conditions. Control animals
were not included in this process. With regard to the female an-
imals subjected to stress, plasma corticosterone (r 00.36, p
0.05), adrenal norepinephrine (r 00.44, p 0.05) and epi-
nephrine (r 00.57, p 0.01) were negatively related to total
WBC counts. Surprisingly, only one endocrine measure was sig-
nificantly correlated with an immune outcome for male rats.
Plasma corticosterone levels were positively correlated with per-
cent peripheral blood lymphocytes (r 0.40, p 0.05).
DISCUSSION
There have been few systematic studies that have attempted
to clarify differences between potentially beneficial and detri-

FIG. 4. Mean adrenal norepinephrine (ng/gland) for the stress and con- FIG. 5. Mean body weight gain (g) for the stress and control groups.
trol groups. Each bar represents the mean { SEM. The females were Each bar represents the mean { SEM. The male rats consistently gained
significantly different from the males across the conditions, * p 0.05. more weight than the female rats. * p 0.05.

/ av1 eh0a 2551 Mp 450 Friday Feb 14 04:44 PM ELPHB (v. 61, no. 3) 2551
STRESS AND IMMUNITY 451

TABLE 2
MEAN RELATIVE ORGAN WEIGHTS FOR MALE AND FEMALE RATS ACROSS STRESS CONDITIONS
(N 20/GROUP)

Variable Run Immobilization Mix Control

Spleen
Male 0.184 (0.004) 0.216 (0.005) 0.197 (0.004) 0.199 (0.005)
Female 0.242 (0.008)* 0.268 (0.007) 0.255 (0.012)* 0.243 (0.005)*
Thymus
Male 0.187 (0.005) 0.098 (0.006) 0.074 (0.002) 0.090 (0.006)
Female 0.102 (0.006) 0.094 (0.006) 0.106 (0.007)* 0.110 (0.004)*
Adrenal
Male 0.004 (0.000) 0.007 (0.000) 0.005 (0.000) 0.004 (0.000)
Female 0.012 (0.000)* 0.013 (0.000)* 0.013 (0.000)* 0.012 (0.000)*

Relative organ weights are expressed g/g of body weight times 100. *Females significantly different
from males at p .05; Immobilization stress increased spleen weight compared to the other groups,
p .001; SEM is in parentheses.

mental effects of stress. Previous investigations usually incor-
porated 1 type of stressor and measured its effect on the immune
system via peripheral blood. According to Smith (46), hemato-
logical changes may not accurately reflect changes within the site
of infection, lymphoid tissue. In this study, we compared the
effects of physical (voluntary running) and psychological (im-
mobilization) stress on the immune system via hematological and
organ tissue examination. In this study, we found no significant
effect of stressor type on the examined immune and endocrine
parameters. However, very strong gender differences did emerge
with respect to these physiological systems. More specifically,
female rats displayed a higher level of immune and endocrine
activity compared with the male rats. More importantly, we re-
port that the gender differences found within the immune system,
in particular, may be stress-induced.
With respect to physical stress, voluntary exercise in the form
of treadmill running has emerged as an alternative method for
studying aerobic fitness in rats. According to Russell and col-
leagues (44), prolonged running by rats can be achieved without
the application of electrical shock or physical prodding. In an
attempt to reduce the possible negative psychological conse-
quences of forced exercise, rats were subjected to voluntary run-
ning. On average, the rats ran 1.4 { 0.90 (SD) km per week
during the morning hours. Although some researchers report av-
erage run distance per week to be approximately 35 km, this
distance is often based upon 24-h access to the wheel (28,44).
With regard to stressor intensity, animals in the running con-
dition gained significantly less weight than the control animals.
This finding is consistent with previous exercise studies (19,28).
More importantly, our findings are consistent with those of Cole-
man and Rager (9), who reported the primary antibody response
to Keyhole Limpet Hemocyanin (KLH) to be unaffected by 8
weeks of nocturnal voluntary running in rats. Likewise, forced
exercise has been found to have no demonstrated affect on the
primary antibody response to antigens in animals (7,13). How-
ever, Liu and Wang (29) reported 23 days of forced treadmill
running to enhance secondary antibody response to Salmonella
typhi in mice. Taken together, our findings suggest that light to
moderate voluntary exercise may have no demonstrated affect on
the primary antibody response.
Physical restraint or immobilization is a widely used experi-
mental stressor that has been found to enhance tumor prolifera-
tion (48), and suppress stimulation of peripheral blood lympho-
cytes to mitogens (42). More specifically, Feng and colleagues
(16) subjected mice to immobilization stress for 2 weeks (16 h/
day) after inoculating them with the influenza virus. Immobili-
zation stress had no affect on the magnitude of the antibody re-
sponse (IgM, IgG, and IgA), but did delay the seroconversion
in the IgG and IgA antibody isotypes when compared to that in
the control animals. In the present study, the kinetic aspect of
humoral immunity was not measured. However, we found no
significant difference in the magnitude of the in vitro antibody
response to SRBC as a function of immobilization. Likewise,
Taylor and Ross (49) reported that the antibody response to the
Streptococcus antigen in adult SD rats was unaffected by re-
peated exposure to immobilization stress. However, the in vitro
antibody titers were suppressed in rats by neonatal immobiliza-
tion. Taken together, our findings do not support the premise of
differential effects of stressor type with regard to adult primary
humoral immunity.
The release of catecholamines and glucocorticoids into the
blood stream with stress is well documented (34,47). From an
immunological perspective, these hormones have been reported
to suppress antibody formation, lymphocyte proliferation to mi-
togens, and lymphocyte migration to the site of tissue damage
(8,11,22,38). In addition, elevated levels of circulating cate-
cholamines have been shown to increase NK cell activity (33).
However, we report no significant effect of stress on plasma cor-
ticosterone and adrenal catecholamine levels. One could argue
that the stress manipulation used in the present study was insuf-
ficient to evoke a heightened neuroendocrine response and
thereby alter humoral immunity.
Although repeated measures of these variables were not per-
formed, we argue that physiological adaptation may have oc-
curred. Within the field of psychoneuroimmunology, a wealth of
data exist concerning the acute effects of stress on immunity
(11,25). However, stressor duration is an important considera-
tion with regard to clarifying the influence of a given hormone
on a specific immunological outcome (23). With chronic expo-
sure to stress, the neuroendocrine responses tend to decline
(4,12). This decline may dampen immunological reactivity over
time. Our data may reflect such an occurrence.
In the present study, we found gender differences with regard
to most of the immune and endocrine measures. The female
stressed rats displayed a greater antibody response to the SRBC
antigen and a higher percentage of peripheral blood lymphocytes
compared to their male counterparts. Female rats, in general, dis-
played larger relative lymph organ weights (spleen and thymus).

/ av1 eh0a 2551 Mp 451 Friday Feb 14 04:44 PM ELPHB (v. 61, no. 3) 2551
452
More specifically, female rodents tend to have the superior cell-
mediated and humoral immunity (2,5). Furthermore, castration
of male rodents tends to enhance the functional capacity of the
immune system and enlarge lymph organs (45). Interestingly,
we found no apparent gender differences in immunity within the
cage control condition. Taken together, it was concluded that
immunological reactivity to a stressor may be influenced by the
gender of the organism investigated.
Research shows that the gender differences in immunity ap-
pear to be mediated by the sex hormones (20). For example,
estrogen tends to enhance the antibody response and testosterone
tends to decrease humoral and cell-mediated immunity in female
rodents via binding to specific receptor sites reportedly found in
lymphatic tissue (45). Not surprisingly, there is evidence that
shows elevation of gonadal hormones with stress. For example,
MacNiven and colleagues (32) found significant increases in
plasma concentration of estrogen, corticosterone, and progester-
one in female rats subjected to 5 days of immobilization stress.
REFERENCES
1. Ader, R.; Felton, D. L.; Cohen, N. Psychoneuroimmunology. New
York: Academic Press; 1991.
2. Ahmed, S. A.; Penhale, W. J.; Talal, N. Sex hormones, immune
responses, and autoimmune diseases: Mechanisms of sex hormone
action. Am. J. Pathol. 121:531551; 1985.
3. Allen, M. T.; Stoney, C. M.; Owen, J. F.; Matthew, K. A. Hemo-
dynamic adjustments to laboratory stress: The influence of gender
and personality. Psychosom. Med. 55:505517; 1993.
4. Armario, A.; Castellanos, J. M.; Balasch, J. Adaptation of anterior
pituitary hormones to chronic noise stress in rats. Behav. Neural
Biol. 41:7176; 1984.
5. Astorquiza, M. I.; Cisternas, C.; Leal, X. Sex-dependent differences
in the IgE response modulated by phytohemagglutinin. Immun. Lett.
16:2730; 1987.
6. Baldwin, D. R.; Wilcox, Z. C.; Baylosis, R. C. Impact of differential
housing on humoral immunity following exposure to an acute stres-
sor in rats. Physiol. Behav. 57:649653; 1995.
7. Baldwin, D. R.; Wilcox, Z. C.; Zheng, G. Stressor duration and
immune and endocrine responses in rats. FASEB J. 9:A5331; 1995.
8. Berciz, I. Pituitary function and immunity. Boca Raton, FL: CRC
Press; 1986.
9. Coleman, J. J.; Rager, D. R. Effects of voluntary exercise on immune
function in rats. Physiol. Behav. 54:771774; 1993.
10. Cray, B.; Borysenko, M.; Sutherland, D. C.; Kutz, I.; Borysenko,
J. Z.; Benson, H. Decrease in mitogen responsiveness of mononu-
clear cells from peripheral blood after epinephrine administration in
humans. J. Immunol. 130:694699; 1983.
11. Dantzer, R.; Kelley, K. W. Stress and immunity: An integrated view
of relationships between the brain and the immune system. Life Sci.
44:19952008; 1989.
12. De Boer, S. F.; Koopmans, S. J.; Slangen, J. L.; Van Der Gugten, J.
Plasma catecholamine, corticosterone and glucose responses to re-
peated stress in rats: Effect of interstressor interval length. Physiol.
Behav. 47:11171124; 1990.
13. Douglass, D. J. The effects of physical training on the immunolog-
ical response in mice. J. Sports Med. 14:4854; 1974.
14. Eichner, E. R. Infection, immunity, and exercise. Physician Sports
Med. 21:125135; 1993.
15. Epling, W. F.; Pierce, W. D. Activity-based anorexia in rats as a
function of opportunity to run on an activity wheel. Nutr. Behav.
2:3749; 1984.
16. Feng, N.; Pagniano, R.; Tovar, C. A.; Bonneau, R. H.; Glaser, R.;
Sheridan, J. F. The effect of restraint stress on the kinetics, magni-
tude, and isotype of the humoral immune response to influenza virus
infection. Brain Behav. Immun. 5:370382; 1991.
17. Garcia-Marquez, C.; Armario, A. Chronic stress depresses explora-
tory activity and behavioral performance in the forced swimming
test without altering acth response to a novel acute stressor. Physiol.
Behav. 40:3338; 1987.
/ av1 eh0a 2551 Mp 452 Friday Feb 14 04:44 PM ELPHB (v. 61, no. 3) 2551
BALDWIN, WILCOX AND ZHENG
However, this is generally not the case with testosterone secretion
(30). Our findings tend to support studies showing possible dif-
ferential activation of the pituitary-adrenal and sympathetic-ad-
renal pathways with stress as a function of gender (3,26).
Despite the difficulty in extrapolating results from animal
studies to humans, the findings from this present study suggest
that exposure to physical or psychological stress may yield sim-
ilar immunological results. Further studies are warranted to de-
termine the influence of stressor intensity as a function of stressor
type and gender on humoral immunity.
ACKNOWLEDGEMENTS
The authors thank Kamesha Bowen, Kevin DeFord, Ingrid Edemon,
Joseph Feagins, Matt Jorgensen, Ellen Oliver, Howard Rasey, Marketta
Rhae, and Ruey Wang for their assistanced in the laboratory. This re-
search was supported by Grant 9306919 from the National Science Foun-
dation.
18. Girdler, S. S.; Turner, J. R.; Sherwood, A.; Light, K. Gender differ-
ences in blood pressure control during a variety of behavioral stres-
sors. Psychosom. Med. 52:571591; 1990.
19. Goodrick, C. L. Effects of long-term voluntary wheel exercise on
male and female Wistar rats. Gerontology 26:2233; 1980.
20. Grossman, C. J. Interactions between the gonadal steroids and the
immune system. Science 227:257261; 1985.
21. Hara, C.; Ogawa, N. Influence of maturation on ulcer-development
and immunodeficiency induced by activity-stress in rats. Physiol.
Behav. 30:757761; 1983.
22. Heilig, M.; Irwin, M.; Grewal, I.; Sercarz, E. Sympathetic regulation
of the t-helper cell function. Brain Behav. Immun. 7:154163; 1993.
23. Herbert, T. B.; Cohen, S. Stress and immunity in humans: A meta-
analytic review. Psychosom. Med. 55:364379; 1993.
24. Hobfoll, S. E. Conservation of resources: A new attempt at concep-
tualizing stress. Am. Psychol. 44:513524; 1989.
25. Khansari, D. N.; Murgo, A. J.; Faith, R. E. Effects of stress on the
immune system. Immunol. Today 11:170175; 1990.
26. Kirschbaum, C.; Wust, S.; Hellhammer, D. Consistent sex differ-
ences in cortisol responses to psychological stress. Psychosom. Med.
54:648657; 1992.
27. Komori, T.; Fujiwara, R.; Nomura, J.; Yokoyama, M. M. Effects of
restraint stress on plaque-forming cell response in normal and olfac-
tory bulbectomized mice. Biol. Psychiatr. 29:695698; 1991.
28. Lambert, M. I.; Noakes, T. D. Spontaneous running increases VO 2
max and running performance in rats. J. Appl. Physiol. 68:400403;
1990.
29. Liu, Y. G.; Wang, S. Y. The enhancing effect of exercise on the
production of antibody to Salmonella typhi in mice. Immunol. Lett.
14:117120; 1987.
30. Lopez-Calderon, A.; Ariznavarreta, C.; Gonzalez-Quijano, M. I.;
Tresquerres, J. A.; Calderon, M. D. Stress induced changes in testis
function. J. Steroid Biochem. Mol. Biol. 40:473479; 1991.
31. Mackinnon, L. T. Exercise and immunology. Champaing, IL: Hu-
man Kinetics Publishers, Inc.; 1992.
32. MacNiven, E.; DeCatanzaro, D.; Younglai, E. V. Chronic stress in-
creases estrogen and other steroids in inseminated rats. Physiol. Be-
hav. 52:159162; 1992.
33. Madden, K. S.; Sanders, V. M.; Felten, D. L. Catecholamine influ-
ences and sympathetic neural modulation of immune responsive-
ness. Annu. Rev. Pharmacol. Toxicol. 35:41748; 1995.
34. Manuck, S. B.; Cohen, S.; Rabin, B. S.; Muldoon, M. F.; Bachen,
E. Individual differences in cellular immune response to stress. Psy-
chol. Sci. 2:111115; 1991.
35. Masera, R.; Gatti, G.; Sartori, M. L.; Carignoia, R.; Salvanori, A.;
Magro, E.; Angeli, A. Involvement of Ca 2/ dependent pathways in
the inhibition of human natural killer (NK) cell activity by cortisol.
Immunopharmacology 18:1122; 1989.
STRESS AND IMMUNITY
36. Matthews, K. A.; Caggiula, A. R.; McAllister, C. G.; Berga, S. L.;
Owens, J. F.; Flory, J. D. Miller, A. L. Sympathetic reactivity to
acute stress and immune response in women. Psychosom. Med.
57:564571; 1995.
37. McCarthy, D. A.; Dale, M. M. The leucocytosis of exercise: A re-
view and model. Sports Med. 6:333363; 1988.
38. Munck, A.; Guyre, P. M.; Holbrook, N. J. Physiological functions
of glucocorticoids in stress and their relation to pharmacological
actions. Endocrinol. Rev. 5:2544; 1984.
39. Northoff, H.; Berg, A. Immunologic mediators as parameters of the
reaction to strenuous exercise. Int. J. Sports Med. 12:915; 1991.
40. Odio, M. R.; Maickel, R. P. Comparative biochemical responses of rrats
to different stressful stimuli. Physiol. Behav. 34:595599; 1985.
41. Pate, R.; Pratt, M.; Blair, S. N.; Haskell, W. L., et al. Physical activity
and public health. JAMA 273:402407; 1995.
42. Rinner, I.; Schauenstein, K.; Mangge, H.; Porta, S.; Kvetnansky, R.
Opposite effects of mild and severe stress on in vitro activation of
rat peripheral blood lymphocytes. Brain Behav. Immun. 6:130140;
1992.
/ av1 eh0a 2551 Mp 453 Friday Feb 14 04:44 PM ELPHB (v. 61, no. 3) 2551
453
43. Ross, C. E.; Hayes, D. Exercise and psychological well-being in the
community. Am. J. Epidemic 127:762771; 1988.
44. Russell, J. C.; Epling, W. F.; Pierce, D.; Amy, R. M.; Boer, D. P.
Induction of voluntary prolonged running by rats. J. Appl. Physiol.
63:25492553; 1987.
45. Schuurs, A. H.; Verheul, H. A. Effects of gender and sex steroids
on the immune response. J. Steroid Biochem. 35:157172; 1990.
46. Smith, J. A. Guidelines, standards, and perspectives in exercise im-
munology. Med. Sci. Sports Exerc. 27:497506; 1995.
47. Stanford, S. C.; Salmon, P. Stress: From synapse to syndrome. San
Diego, CA: Academic Press; 1993.
48. Steplewski, Z.; Vogel, W. H.; Ehya, H.; Poropatich, C.; Smith,
J. M. Effects of restraint stress on inoculated tumor growth and im-
mune response in rats. Cancer Res. 45:51285133; 1985.
49. Taylor, C. E.; Ross, L. L. Alteration of antibody response to pneu-
mococcal polysaccharide type III in rats by neonatal immobilization
stress. Brain Behav. Immun. 3:160170; 1989.
50. Tharps, G. D.; Barnes, M. W. Reduction of saliva immunoglobulin
levels by swim training. Eur. J. Appl. Physiol. 60:6164; 1990.