Am J Physiol Regul Integr Comp Physiol 301: R1648 –R1657, 2011.
First published September 14, 2011; doi:10.1152/ajpregu.00301.2011.
Integrative and Translational Physiology: Inflammation and
Immunity in Organ System Physiology
CALL FOR PAPERS
Time-dependent mediators of HPA axis activation following live
Escherichia coli
Z. R. Zimomra, V. M. Porterfield, R. M. Camp, and J. D. Johnson
Kent State University, Department of Biological Sciences, Kent, Ohio
Submitted 9 June 2011; accepted in final form 8 September 2011
Zimomra ZR, Porterfield VM, Camp RM, Johnson JD. Time-dependent mediators of HPA axis activation following live Escherichia coli . Am
J Physiol Regul Integr Comp Physiol 301: R1648–R1657, 2011. First
published September 14, 2011; doi:10.1152/ajpregu.00301.2011.—The hypothalamus-pituitary-adrenal (HPA) axis is activated during an immune challenge to liberate energy and modulate immune responses
via feedback and regulatory mechanisms. Inflammatory cytokines and
prostaglandins are known contributors to HPA activation; however,
most previous studies only looked at specific time points following
LPS administration. Since whole bacteria have different immune
stimulatory properties compared with LPS, the aim of the present
studies was to determine whether different immune products contribute to HPA activation at different times following live Escherichia
coli challenge. Sprague-Dawley rats were injected intraperitoneally
with E. coli (2.5 ⫻ 107 CFU) and a time course of circulating
corticosterone, ACTH, inflammatory cytokines, and PGE2 was developed. Plasma corticosterone peaked 0.5 h after E. coli and steadily
returned to baseline by 4 h. Plasma PGE2 correlated with the early rise
in plasma corticosterone, whereas inflammatory cytokines were not
detected until 2 h. Pretreatment with indomethacin, a nonselective
cyclooxygenase inhibitor, completely blocked the early rise in plasma
corticosterone, but not at 2 h, whereas pretreatment with IL-6 antibodies had no effect on the early rise in corticosterone but attenuated
corticosterone at 2 h. Interestingly, indomethacin pretreatment did not
completely block the early rise in corticosterone following a higher
concentration of E. coli (2.5 ⫻ 108 CFU). Further studies revealed that
only animals receiving indomethacin prior to E. coli displayed elevated plasma and liver cytokines at early time points (0.5 and 1 h),
suggesting prostaglandins suppress early inflammatory cytokine production. Overall, these data indicate prostaglandins largely mediate
the early rise in plasma corticosterone, while inflammatory cytokines
contribute to maintaining levels of corticosterone at later time points.
corticosterone; prostaglandin; IL-6; indomethacin
ACTIVATION OF THE hypothalamus-pituitary-adrenal (HPA) axis
is one of the critical brain-mediated sickness responses that
enhance survival of an organism during an immune challenge.
The resulting elevation in circulating glucocorticoids, mainly
cortisol in humans and corticosterone in mice and rats, liberate
energy necessary to mount a fever and have numerous immunomodulatory effects that reduce the risk of septic shock and
increase the chances of survival during infection (52, 53). For
example, glucocorticoids suppress proinflammatory cytokines
Address for reprint requests and other correspondence: J. D. Johnson,
Kent State Univ., Biological Sciences Dept., Kent, Ohio (e-mail:
jjohns72@kent.edu).
R1648
(2, 17, 34) and prostaglandin production (27, 38), stimulate
anti-inflammatory cytokine production (16, 20), upregulate Fc
receptors and major histocompatibility complex class II molecules on phagocytes (24), increase cell adhesion molecules on
endothelial cells (57), enhance acute phase changes in the liver
(3), and mobilize immune cells to sites of infection (10, 40).
However, it remains unclear how and when various immune
factors contribute to the initiation and maintenance of HPA
activation during an immune challenge.
Numerous studies have investigated the role inflammatory
cytokines, TNF-␣, IL-1, and IL-6 play in HPA activation (11,
55) since these are the first cytokines released following
immune activation (typically within 60 –90 min). Studies generally find that inflammatory cytokines are sufficient, but not
necessary, to activate the HPA axis during immune challenge.
Administration of any one of the inflammatory cytokines
potently stimulates the HPA axis and results in elevation of
circulating corticosterone (11, 55). However, blockade of any
individual inflammatory cytokine via administration of a receptor antagonist (12, 13, 50), neutralizing antibodies (46, 60)
or use of selective cytokine or receptor knockout animals (31,
37, 58) fails to block HPA activation following LPS challenge.
IL-6 appears to play the largest role since administration of
IL-6 neutralizing antibodies (60), and IL-6 deficient mice (58)
show some attenuation of corticosterone levels following LPS
administration; however, a robust HPA response is still observed in these animals. The overall conclusion has been that
all the inflammatory cytokines contribute to HPA activation
and blockade of any individual cytokine fails to prevent HPA
stimulation due to duplicity in their actions.
More recently, focus has turned to the role prostaglandins
play in activation of the HPA axis during an immune challenge. In contrast to the soluble protein cytokines, prostaglandins are lipid mediators that can be rapidly (within minutes)
released following immune stimulation via the conversion of
arachidonic acid to PGH2 by constitutively expressed cyclooxygenase-1 (COX-1) and then converted to PGE2 by PGE
synthase; furthermore, later production of PGE2 can be synthesized by the inducible COX-2 (39). Administration of PGE2
is sufficient to stimulate the HPA axis (43, 61, 62), but in
contrast to cytokines, a nonselective COX inhibitor completely
blocks the early (0.5–1 h) corticosterone response after LPS
challenge (14, 23). Similar to inflammatory cytokines, prostaglandins are released by innate immune cells in local tissue and
have been shown to signal the brain via activation of peripheral
nerves such as vagal afferents in the liver (36) or released
0363-6119/11 Copyright © 2011 the American Physiological Society
http://www.ajpregu.org
TIME-DEPENDENT MEDIATORS OF THE HPA AXIS FOLLOWING E. COLI
directly by endothelial cells composing brain microvasculature
(23, 49, 51). The rapid production of prostaglandins compared
with soluble proteins following an immune challenge led to the
hypothesis that PGE2 mediates the initial activation of the HPA
axis, while inflammatory cytokines help maintain HPA activation at later time points.
We present a series of studies that examined the contribution
of both lipid (i.e., prostaglandin) and protein (i.e., IL-6) immune products across time in mediating activation of the HPA
axis following intraperitoneal injection of live Escherichia
coli. Previous studies used purified LPS derived from the cell
membrane of gram-negative bacteria to stimulate immune
responses, but LPS is only one component in gram-negative
bacteria cell membrane that selectively targets toll-like receptor-4, while whole bacteria possesses multiple pathogen-associated molecular patterns capable of activating several toll-like
receptors and is more potent at activating other immune responses, such as complement proteins (5). Here, studies examined corticosterone levels at multiple time points (30, 60, 120,
240 min) following two concentrations of E. coli, and used
indomethacin (a nonselective COX inhibitor) and IL-6 neutralizing antibodies to examine the contribution of prostaglandin
and IL-6 in activation of the HPA.
MATERIALS AND METHODS
Animals
Adult male Sprague-Dawley rats (175–199 g; Harlan) were used
for all studies in this investigation. Animals were single housed in
Plexiglas cages with food and water available ad libidum and a
12;12-h light-dark cycle beginning at 07:00. Animals were given 1 wk
to adjust to their new environment before any study was performed.
Animals were handled once a day in the morning for 4 days before
any study was performed to help minimize their stress response during
the studies. All studies began in the morning (⬃09:00 –10:00 h) to
minimize the effects of the natural rise of corticosterone in the early
evening. Care and use of the animals was in accordance with protocols
approved by the Kent State University Institutional Care and Use
Committee. All studies were designed to minimize pain and the
number of animals used.
Drug Concentration and Administration
Indomethacin (Sigma-Aldrich), a nonselective COX inhibitor, was
injected at a concentration of 5 mg/kg in a volume of 100 ␮l of sterile
33% dimethyl sulfoxide. Rat IL-6 antibody (cat. no. AF506; R&D
Systems, Minneapolis, MN) was injected at a concentration of 4.2
␮g/kg in a volume of 250 ␮l of sterile saline. All drugs were injected
intraperitoneally.
Bacterial Growth
E. coli (0111:B4; ATCC 15746; American Type Culture Collection) was rehydrated and grown overnight in 40 ml of brain-heart
infusion broth (DIFCO Laboratories) at 37°C at 5% CO2. Bacterial
cultures were then aliquoted into 1 ml brain-heart infusion broth with
10% glycerol and frozen at ⫺20°C. All studies used bacteria from
these stock cultures. One day prior to experimentation, stock cultures
were thawed and cultured overnight in 40 ml of brain-heart infusion
broth at 37°C and 5% CO2. Bacteria were quantified by extrapolating
from previously determined growth curves. Cultures were centrifuged
for 20 min at 3,000 rpm, and supernatants were discarded. Bacteria
were resuspended in sterile 0.9% saline at a concentration of 1 ⫻ 108
CFU/ml for studies 1, 2, and 3, and at a concentration of 1 ⫻ 109
CFU/ml for studies 4 and 5. Rats were injected intraperitoneally with
AJP-Regul Integr Comp Physiol • VOL
R1649
250 ␮l of E. coli (2.5 ⫻ 107 CFU for studies 1, 2, and 3 and 2.5 ⫻ 108
CFU for studies 4 and 5).
Blood Collection
For studies 1 and 5, rats were euthanized by decapitation and truck
blood was collected in 10 ml EDTA-coated vacutainer tubes (BD
PharMingen) and stored on ice until the completion of the study.
Blood was centrifuged for 10 min at 4,000 rpm at 4°C, and plasma
was collected and stored at ⫺80°C until the time of assay. For studies
2, 3, and 4 tail vein blood was collected at designated time points. Rats
were gently wrapped in a towel and lightly secured with a Velcro strap
prior to tail blood collection. A lateral tail vein was nicked at the
posterior end of the tail using a scalpel blade, and the tail was stroked
from anterior to posterior to facilitate the movement of blood. Approximately 200 –300 ␮l of blood was collected in an Eppendorf
snap-cap tube within ⬃60 –90 s of removing the animal from the cage.
Blood was centrifuged for 10 min at 10,000 rpm, and plasma was
collected and stored at ⫺80°C until time of assay.
Tissue Collection
For studies 1 and 5, a small specimen of liver was quickly
harvested from the rats and frozen in liquid nitrogen, and then stored
at ⫺80°C until time of use. Each liver specimen was sonicated in
Iscove’s Dulbecco Medium with 2% aprotinin for ⬃10 s. Specimens
were centrifuged at 10,000 rpm for 10 min at 4°C and the supernatant
was collected for measurement of cytokines (IL-1␤, IL-6, and TNF-␣)
by ELISA. Cytokine measurements were corrected for protein by
Bradford assay.
Measurement of Peripheral Mediators
Cytokines were measured using available ELISAs for IL-1, IL-6,
and TNF-␣ (R&D Systems). Plasma ACTH also was measured by
ELISA (ALPCO). PGE2 and corticosterone were measured by enzyme immunoassay (Cayman Chemical and Assay Designs, respectively). All ELISAs and enzyme immunoassays were performed
according to manufacturer’s recommendations. PGE2 samples were
diluted 1:20 in the manufacturer’s enzyme immunoassay buffer, and
corticosterone samples were diluted 1:50 in dH2O and heated for 1 h
in 70°C water bath.
Experimental Design
Study 1. Animals (n ⫽ 6 – 8/group) were injected intraperitoneally
with either saline or E. coli (2.5 ⫻ 107 CFU) and euthanized by
decapitation 0.5, 1, 2, 4, or 24 h later. One or two saline-control
animals were euthanized at each time and presented as a 0-h time
point. Trunk blood was collected for measurement of cytokines,
ACTH, corticosterone, and PGE2. A small specimen of liver was
quickly harvested, and cytokines was measured by ELISA.
Study 2. A baseline sample of tail vein blood was collected prior to
drug and E. coli administration. Animals (n ⫽ 6/group) were injected
intraperitoneally with either vehicle or indomethacin (5 mg/kg) 30
min prior to intraperitoneal E. coli administration (2.5 ⫻ 107 CFU).
Tail vein blood was then collected at 1, 2, and 4 h after E. coli
challenge for measurement of corticosterone. PGE2 was also measured in blood samples to verify blockade of PGE2 formation by
indomethacin.
Study 3. A baseline sample of tail vein blood was collected prior to
drug and E. coli administration. Animals (6 –9/group) were injected
intraperitoneally with either saline or IL-6 antibody (4.2 ␮g/kg) 30
min prior to intraperitoneal E. coli administration (2.5 ⫻ 107 CFU).
Tail vein blood was collected at 1, 2, and 4 h after E. coli challenge
for measurement of corticosterone.
Study 4. A baseline sample of tail vein blood was collected prior to
drug and E. coli administration. Animals (6 –9/group) were injected
with either DMSO or indomethacin 30 min prior to intraperitoneal E.
301 • DECEMBER 2011 •
www.ajpregu.org
R1650
TIME-DEPENDENT MEDIATORS OF THE HPA AXIS FOLLOWING E. COLI
coli challenge (2.5 ⫻ 108 CFU). Tail vein blood was collected at 1, 2,
and 4 h after E. coli challenge for measurement of corticosterone.
Study 5. Animals (6 –9/group) were injected with either DMSO or
indomethacin 30 min prior to intraperitoneal saline or E. coli (2.5 ⫻
108 CFU) administration. Animals were killed by decapitation at 0.5
and 1 h, and trunk blood was collected for measurement of cytokines,
corticosterone, and PGE2 by ELISA or enzyme immunoassay. A
small specimen of liver was quickly harvested for measurement of
tissue cytokines.
Statistical Analysis
Data from study 1, which examined the effects of E. coli on various
mediators of HPA axis activation, were analyzed using a 1 ⫻ 6
ANOVA across time (0, 0.5, 1, 2, 4, and 24 h). In studies 2, 3, and 4,
tail blood was taken from each animal to examine the effects of drug
administration on corticosterone release across time. A repeatedmeasures ANOVA was used to examine the effects of each treatment
across time (0, 1, 2, and 4 h), and significant interactions were further
analyzed using one-way ANOVAs at each time point. Necessary post
hoc analyses were done using Fisher’s least significant difference test.
P ⬍ 0.05 was used in all cases as a measure of significance to reject
the null hypothesis.
RESULTS
Study 1: Time Course of Mediators Involved in HPA
Activation
To establish a time course of HPA activation following E.
coli administration and to determine the factors that might be
involved in HPA stimulation at each time point, animals were
injected intraperitoneally with 2.5 ⫻ 107 CFU of E. coli or
saline and killed 0.5, 1, 2, 4, and 24 h later. After E. coli
challenge, plasma corticosterone levels peaked at 0.5 h and
decreased in a time-dependent manner, thereafter returning to
baseline levels by 4 h [F(5,44) ⫽ 14.720, P ⬍ 0.001]. Post hoc
analysis revealed that corticosterone was significantly elevated
at 0.5 h (P ⬍ 0.001), 1 h (P ⬍ 0.001), and 2 h (P ⫽ 0.020)
following E. coli challenge compared with saline-injected
controls (0 h) (Fig. 1A). Plasma ACTH also increased across
time [F(5,42) ⫽ 7.655, P ⬍ 0.001] following E. coli challenge.
ACTH began to increase at 0.5 h (P ⫽ 0.072) although levels
did not reach significance until 1 h (P ⬍ 0.001) and returned to
baseline levels by 2 h (Fig. 1B).
Plasma PGE2 showed a similar pattern following E. coli
challenge as plasma corticosterone with peak levels at 0.5 h
and levels slowly decreasing across time [F(5,42) ⫽ 17.318,
P ⬍ 0.001]. Post hoc analysis revealed that plasma PGE2 is
rapidly elevated by 0.5 h (P ⬍ 0.001), continues to be significantly elevated at 1 h (P ⫽ 0.004), and returns to baseline
levels by 2 h. Interestingly, PGE2 significantly decreased at 4
h (P ⫽ 0.015) compared with baseline levels before returning
to basal levels by 24 h (Fig. 1C). Not surprisingly, the inflammatory cytokines took longer to elevate in plasma following
immune challenge. Plasma IL-6 levels increased across time
[F(5,45) ⫽ 53.588, P ⬍ 0.001] with levels significantly elevated at 2 h (P ⬍ 0.001) (Fig. 1D), but no longer significantly
elevated at 4 h (P ⫽ 0.072). Plasma TNF-␣ also increased
across time [F(5,43) ⫽ 11.723, P ⬍ 0.001] with TNF-␣ levels
elevated at 1 h (P ⫽ 0.193) and significantly elevated by 2
h (P ⬍ 0.001), and returned to baseline levels by 4 h (Fig. 1E).
Plasma IL-1␤ was measured, but there were no significant
interactions across time [F(5,42) ⫽ 1.768, P ⫽ 0.141] with
only low levels of IL-1␤ being detected (figure not shown).
AJP-Regul Integr Comp Physiol • VOL
Liver IL-1␤ showed a significant interaction across time
[F(5,45) ⫽ 8.840, P ⬍ 0.001] following E. coli challenge. Post
hoc analysis revealed that liver IL-1␤ levels did not rise until
2 h after E. coli challenge (P ⬍ 0.001). Liver IL-1␤ was still
elevated at 4 and 24 h, but the values were not significant (P ⫽
0.075 and P ⫽ 0.357, respectively) (Fig. 1F). Liver IL-6 and
TNF-␣ levels were not elevated at any time point [F(5,47) ⫽
1.189, P ⫽ 0.331] and [F(5,47) ⫽ 0.318, P ⫽ 0.899], respectively (figures not shown).
Study 2: Effects of Prostaglandins on HPA Axis Activation
Based on study 1, the time course of plasma PGE2 best
matched that of plasma corticosterone as both were shown to
rapidly elevate following E. coli challenge, with peak values
reached within 30 min. To examine the role PGE2 may play on
HPA activation and ultimately corticosterone secretion, animals were injected with either DMSO or indomethacin prior to
E. coli administration, and plasma corticosterone was measured across time. A repeated-measures ANOVA revealed a
significant effect of treatment across time [F(2,17) ⫽
16.354, P ⬍ 0.001]. Administration of E. coli resulted in a
rapid increase in plasma corticosterone with the greatest increase at 1 h and then levels steadily decreasing across time.
Indomethacin alone had no significant effects on plasma corticosterone at any time point, but completely blocked the rise in
corticosterone 1 h after E. coli challenge, but not 2 h following
E. coli challenge (Fig. 2A).To verify that the effects of indomethacin did not wear off by 2 h, plasma PGE2 levels were
measured at 1 and 2 h after E. coli challenge. Indomethacin
completely blocked the rise in plasma PGE2 at both 1 h
[F(2,17) ⫽ 11.305, P ⫽ 0.001] and 2 h [F ⫽ (2,14) ⫽ 15.525,
P ⫽ 0.001] following E. coli challenge (Fig. 2B).
Study 3: Effects of IL-6 on HPA Axis Activation
As demonstrated in study 2, plasma corticosterone was
elevated at 2 h following E. coli challenge, despite continual
blockade of plasma PGE2. Since plasma IL-6 was significantly
elevated at 2 h and IL-6 has previously been shown to contribute to LPS-induced HPA activation, we examined the
contribution of IL-6 on E. coli-induced plasma corticosterone
via immunoneutralization techniques. Administration of rat
IL-6 antibody alone had no significant effects on plasma
corticosterone compared with baseline levels across time; however, administration of IL-6 antibody significantly attenuated
plasma corticosterone levels at 2 h following peripheral E. coli.
A repeated-measures ANOVA revealed a significant effect of
drug treatment across time [F(2,17) ⫽ 30.023, P ⬍ 0.001] with
post hoc analyses revealing that administration of IL-6 antibody only attenuated plasma corticosterone levels 2 h after E.
coli challenge (P ⫽ 0.004) (Fig. 3).
Study 4: Effects of Prostaglandins on HPA Activation After
Higher Concentration of E. coli.
To further investigate the role of PGE2 in mediating HPA
responses to higher concentrations of bacteria, we injected
animals with either DMSO or indomethacin prior to a 10-fold
higher concentration of E. coli (2.5 ⫻ 108 CFU) and measured
plasma corticosterone across time. Administration of E. coli
resulted in high levels of plasma corticosterone at 1 and 2 h
with levels slightly lower at 4 h. Unlike with the lower
301 • DECEMBER 2011 •
www.ajpregu.org
TIME-DEPENDENT MEDIATORS OF THE HPA AXIS FOLLOWING E. COLI
R1651
Fig. 1. Sprague-Dawley rats were injected
intraperitoneally (IP) with 2.5 ⫻ 107 CFU
Escherichia coli and euthanized 0.5, 1, 2, 4, or
24 h later for measurement of plasma corticosterone (A), plasma ACTH (B), plasma PGE2
(C), plasma IL-6 (D), plasma TNF-␣ (E), and
liver IL-1␤ (F). Baseline (BL) levels were
determined from animals injected IP with sterile saline. Data points represent means ⫾ SE
(n ⫽ 6 – 8/group). *Significant differences P ⬍
0.05 compared with baseline values.
concentration of E. coli, indomethacin only attenuated plasma
corticosterone at 1 h (P ⫽ 0.044) compared with animals
injected with E. coli alone as corticosterone levels were significantly elevated compared with baseline levels (Fig. 4).
There were no other significant differences between the two
treatments at any other time point.
Study 5: Effects of Indomethacin on Cytokines Production
Following a Higher Concentration of E. coli.
Because indomethacin only attenuated the early HPA response after a high concentration of E. coli, we further investigated possible mediators that could be responsible for early
activation of the HPA axis following a high concentration of E.
coli. Animals were either injected with DMSO or indomethacin 30 min prior to intraperitoneal saline or E. coli adminisAJP-Regul Integr Comp Physiol • VOL
tration, and animals were killed 0.5 h or 1 h later. Again, E. coli
administration significantly elevated plasma corticosterone
compared with DMSO-injected controls at both the 0.5 h (P ⫽
0.001) and 1 h (P ⬍ 0.001) time points. There was a significant
interaction between the treatments at both time points [F(2,15) ⫽
8.351 P ⫽ 0.005 and F(2,14) ⫽ 22.812, P ⬍ 0.001, respectively] with post hoc analysis revealing animals injected with
indomethacin prior to E. coli challenge had significantly attenuated plasma corticosterone levels at both 0.5 h (P ⫽ 0.024)
and 1 h (P ⫽ 0.001) compared with E. coli-injected animals
(Fig. 5A). E. coli administration also significantly elevated
plasma ACTH compared with DMSO-injected controls at both
the 0.5 h (P ⬍ 0.001) and 1 h (P ⬍ 0.001) time points. There
was a significant interaction between the treatments at both
time points [F(2,16) ⫽ 15.799, P ⬍ 0.001 and F(2,19) ⫽
301 • DECEMBER 2011 •
www.ajpregu.org
R1652
TIME-DEPENDENT MEDIATORS OF THE HPA AXIS FOLLOWING E. COLI
Fig. 3. Sprague-Dawley rats were injected IP with goat anti-rat polyclonal IL-6
antibodies 30 min prior to 2.5 ⫻ 107 CFU E. coli and blood samples collected
from a tail vein 1, 2, and 4 h later for measurement of circulating plasma
corticosterone. Baseline blood samples were collected immediately before
saline or antibody (Ab) administration. Data points represent means ⫾ SE
(n ⫽ 6 –9/group). *Significant differences P ⬍ 0.05 compared with baseline.
#Significant differences P ⬍ 0.05 compared with E. coli-injected animals.
Fig. 2. Sprague-Dawley rats were injected IP with either DMSO or indomethacin 30 min prior to 2.5 ⫻ 107 CFU E. coli and blood samples collected from
a tail vein 1, 2, 4, and 24 h later for measurement of circulating plasma
corticosterone (A). Circulating plasma PGE2 1 h and 2 h after E. coli
administration is also presented (B). Baseline blood samples were collected
immediately before saline or indomethacin administration. Data points represent means ⫾ SE (n ⫽ 6/group). *Significant differences P ⬍ 0.05 compared
with baseline values (or indomethacin controls in Fig. 2B). #Significant
differences P ⬍ 0.05 compared with E. coli-injected animals.
14.721, P ⬍ 0.001, respectively] with post hoc analysis revealing that indomethacin administration prior to E. coli challenge
significantly attenuated plasma ACTH levels at both 0.5 h (P ⬍
0.001) and 1 h (P ⬍ 0.001) compared with E. coli-injected
animals (Fig. 5B).
Plasma PGE2 also showed a significant interaction between
treatments after E. coli challenge at 0.5 h [F(2,17) ⫽ 7.926
P ⫽ 0.004] and 1 h [F(2,17) ⫽ 6.142, P ⫽ 0.010]. Post hoc
analysis revealed animals injected with E. coli had significant
increases in plasma PGE2 compared with control animals at
both 0.5 h (P ⫽ 0.008) and 1 h (P ⫽ 0.011), and indomethacin
completely blocked the elevations in PGE2 at both time points
(Fig. 5C). Similar to that observed following the low concentration of E. coli presented above in Fig. 1, plasma TNF-␣,
IL-1␤, and IL-6 were not significantly elevated at 0.5 h or 1 h
following the high concentration of E. coli. Interestingly, a
significant effect of treatment was observed for plasma IL-1␤
at 1 h [F(2,17) ⫽ 4.066, P ⫽ 0.035], plasma IL-6 at both 0.5
h [F(2,17) ⫽ 4.057, P ⫽ 0.039] and 1 h [F(2,17) ⫽ 10.007,
P ⫽ 0.001], and plasma TNF-␣ at both 0.5 h [F(2,17) ⫽ 6.816,
P ⫽ 0.008] and 1 h [F(2,17) ⫽ 8.537, P ⫽ 0.004]. In all cases,
AJP-Regul Integr Comp Physiol • VOL
animals injected with indomethacin prior to E. coli challenge
had significantly elevated plasma cytokines compared with
control animals and animals injected with E. coli in the absence
of indomethacin (Fig. 5, D–F).
Liver IL-␤ showed a similar pattern with a significant effect
of treatment at 0.5 h [F(2,14) ⫽ 6.080, P ⫽ 0.015] and 1 h
[F(2,14) ⫽ 14.814, P ⬍ 0.001]. Indomethacin treatment prior
to E. coli significantly increased liver IL-1␤ at 0.5 and 1 h
compared with controls (P ⫽ 0.005 at 0.5 h and P ⬍ 0.001 at
1 h) and animals receiving E. coli in the absence of indomethacin (P ⫽ 0.047 at 0.5 h and P ⫽ 0.011 at 1 h) (Fig. 6A). Liver
Fig. 4. Sprague-Dawley rats were injected IP with either DMSO or indomethacin 30 min prior to 2.5 ⫻ 108 CFU E. coli and blood samples collected from
a tail vein 1, 2, and 4 h later for measurement of circulating plasma corticosterone (A). Baseline blood samples were collected immediately before saline
or indomethacin (Indo) administration. Data points represent means ⫾ SE
(n ⫽ 6/group). *Significant differences P ⬍ 0.05 compared with baseline
values. #Significant differences P ⬍ 0.05 compared with E. coli-injected animals.
301 • DECEMBER 2011 •
www.ajpregu.org
TIME-DEPENDENT MEDIATORS OF THE HPA AXIS FOLLOWING E. COLI
R1653
Fig. 5. Sprague-Dawley rats were injected IP
with either DMSO or indomethacin 30 min
prior to saline or 2.5 ⫻ 108 CFU E. coli and
animals euthanized 0.5 h or 1 h later for measurement of plasma corticosterone (A), plasma
ACTH (B), plasma PGE2 (C), plasma IL-1␤
(D), plasma IL-6 (E), plasma TNF-␣ (F). Data
points represent means ⫾ SE (n ⫽ 6 –9/group).
*Significant differences P ⬍ 0.05 compared
with saline-injected animals. #Significant differences P ⬍ 0.05 compared with E. coliinjected animals.
IL-6 displayed no significant interactions between treatments at
either time point (Fig. 6B). No elevations in liver TNF-␣ were
reported at the 0.5-h time point, but there was a significant
difference between treatments at 1 h [F(2,17) ⫽ 4.743, P ⫽
0.022]. Post hoc analysis again revealed that only animals
injected with indomethacin prior to E. coli challenge had
elevations in liver TNF-␣ at 1 h, and these elevations were
significant compared with both control animals (P ⫽ 0.015)
and animals injected with E. coli alone (P ⫽ 0.017) (Fig. 6C).
DISCUSSION
The present data demonstrate a shift across time in the
factors that mediate HPA activation following bacterial challenge. HPA responses following peripheral E. coli challenge
are extremely rapid, within 30 min, and occur prior to detection
of circulating or liver inflammatory cytokines. Data presented
here demonstrate the initial activation of the HPA axis is
largely mediated by COX-induced prostanoid synthesis. Not
only did the early rise in circulating corticosterone correlate with the rise in circulating PGE2, but the administration
of indomethacin, a nonselective COX inhibitor, completely
AJP-Regul Integr Comp Physiol • VOL
blocked the early rise in plasma corticosterone following a low
concentration of peripheral E. coli and attenuated the early rise
in plasma corticosterone following a high concentration of E.
coli. Interestingly, corticosterone levels peaked long before
peak circulating ACTH, suggesting prostaglandins may stimulate corticosterone release through an ACTH-independent
mechanism. This would support previous work demonstrating
PGE2 directly mediate corticosterone release from cultured rat
adrenals (41) and cortisol release from human adrenal H295R
cells (56).
Even while prostaglandin production was completely inhibited, corticosterone levels increased 2 h following E. coli
challenge, indicating other factors stimulate HPA responses at
this time. The later rise in corticosterone correlated with the
rise of inflammatory cytokines in plasma (IL-6 and TNF-␣)
and in liver (IL-1␤). Administration of rat IL-6 antibodies
attenuated the later rise in corticosterone 2 h following E. coli
challenge. It should be noted that the magnitude of the corticosterone response in the original time course study was less
compared with subsequent studies for unknown reasons. One
possibility is that some bacteria were lost during the decanting
301 • DECEMBER 2011 •
www.ajpregu.org
R1654
TIME-DEPENDENT MEDIATORS OF THE HPA AXIS FOLLOWING E. COLI
Fig. 6. Sprague-Dawley rats were injected IP with either DMSO or indomethacin 30 min prior to saline or 2.5 ⫻ 108 CFU E. coli and animals euthanized
0.5 h or 1 h later for measurement of liver IL-1␤ (A), liver IL-6 (B), and liver
TNF-␣ (C). Data points represent means ⫾ SE s (n ⫽ 6 –9/group). *Significant
differences P ⬍ 0.05 compared with saline-injected animals. #Significant
differences P ⬍ 0.05 compared with E. coli-injected animals.
process for the original study, which would have reduced the
number of actual E. coli injected. However, this does not
change the interpretation of the individual studies, that following bacterial challenge prostaglandins largely initiate HPA
responses, while inflammatory cytokines, at least IL-6, are
likely mediators that maintain HPA activation at later time
points.
Most studies in the literature have utilized purified LPS, a
component in the cell membrane of gram-negative bacteria, to
stimulate immune responses and examine HPA activation.
AJP-Regul Integr Comp Physiol • VOL
Here, we chose to use live E. coli to more closely mimic a true
bacterial challenge as whole bacteria have multiple antigenic
epitopes that can stimulate a wider range of immune responses
(5). In fact, the time course of HPA activation presented here
following E. coli is different than that commonly reported
following LPS challenge. Administration of E. coli resulted in
a very rapid rise in circulating corticosterone that peaked ⬃30
min after E. coli challenge and decrease thereafter, while
studies that administer LPS report a slower rise in circulating
corticosterone that peaks ⬃2 h after challenge (regardless of
dose and route of administration) (13, 23, 26, 32, 35, 60).
While there appears to be a rather dramatic shift in the kinetics
of HPA activation between E. coli and LPS, the mechanisms
mediating HPA responses appear to be very similar. By use of
genetic and pharmacological techniques, recent studies have
demonstrated inhibition of COX-1 results in a significant
reduction in corticosterone levels in the first hour after LPS
challenge but has little effect on corticosterone thereafter (14,
15, 23). Our data demonstrate COX-mediated prostanoid synthesis is also critical for the initial induction of corticosterone
following E. coli challenge, particularly following lower concentrations of bacteria. However, because of the kinetic differences between E. coli and LPS, this study reveals prostaglandins may play a larger role in HPA activation during a bacterial
infection than suggested by LPS studies. In studies using LPS,
COX inhibitors only block a very small fraction of the overall
corticosterone response because very little corticosterone is
released in the first hour following LPS challenge (14, 23),
while following E. coli the bulk of the corticosterone response
is observed within the first hour, especially following lower
concentrations of bacteria.
At later time points, inflammatory cytokines likely help
maintain HPA responses. Many studies have examined the role
of inflammatory cytokines and found that genetic or pharmacological blockade of individual inflammatory cytokines have
little effect on attenuating HPA responses to LPS (12, 13, 31,
37, 46, 50, 58), likely due to duplicity in inflammatory cytokine
signaling. One exception has been that the pretreatment with
IL-6 antibodies often attenuates ACTH (46) and corticosterone
(60) responses 2– 4 h following LPS but not at early time points
(18, 60). Here, we replicate this finding following E. coli
challenge by demonstrating administration of goat anti-rat
polyclonal IL-6 antibodies attenuate corticosterone levels 2 h
following bacterial challenge. Together, these findings suggest
prostaglandins mediate early HPA activation and IL-6 (along
with potentially other inflammatory cytokines) help maintain
HPA activation as needed.
Indomethacin failed to completely block the early rise in
corticosterone following higher concentrations of E. coli, suggesting additional mediators may be involved in HPA activation as bacterial challenge increases. We examined whether a
higher concentration of E. coli would stimulate a more rapid
increase in inflammatory cytokines; but this was not the case as
there was no significant increase in any circulating or liver
inflammatory cytokines at either 30 or 60 min following the
higher dose of E. coli. However, indomethacin administration
resulted in exaggerated inflammatory cytokine responses to E.
coli challenge as demonstrated by significant increases in liver
IL-1␤ at 0.5 and 1 h, plasma IL-6 at 0.5 and 1 h, and increases
in plasma TNF-␣ at 0.5 and 1 h and liver TNF-␣ at 1 h. This
suggests inflammatory cytokines may elevate in an active
301 • DECEMBER 2011 •
www.ajpregu.org
TIME-DEPENDENT MEDIATORS OF THE HPA AXIS FOLLOWING E. COLI
compensatory fashion to stimulate corticosterone production,
thereby, explaining why indomethacin only attenuates corticosterone responses at 0.5 and 1 h following a high concentration
of E. coli. However, the cytokine levels at 0.5 h are very low,
and it is unclear whether such values would be sufficient to
activate the HPA axis. If these low cytokine levels are sufficient to stimulate corticosterone production, then one would
have expected a further increase in corticosterone production
as cytokine levels increased from 0.5 to 1 h, but this was not
observed. One would also have expected a similar enhancement of inflammatory cytokines and attenuation of corticosterone responses in indomethacin-treated animals challenged
with the lower concentration of E. coli, yet indomethacin
completely blocked the elevation in corticosterone when administered following the lower concentration of E. coli. Our
interpretation of these data is that additional factors contribute
to HPA activation at higher bacterial concentrations. One
possibility is activation of complement. Complement proteins
are constitutively produced and continuously circulate in the
blood, and following contact with the surface of bacteria
complement proteins can be rapidly activated, thus available to
rapidly activate HPA responses. Complement products C3a
and C5a have been shown to elevate ACTH (21, 22) and
activate paraventricular hypothalamic neurons (8) as well as
act directly on the adrenal gland (21). Complement products
can also trigger prostaglandin synthesis and release by Kupffer
cells (45, 47) adding to HPA activation.
The fact that indomethacin pretreatment resulted in exaggerate inflammatory cytokine responses following E. coli challenge indicates that prostaglandin signaling inhibits inflammatory cytokine production. Prostaglandins can directly inhibit
inflammatory cytokines, particularly TNF-␣, as demonstrated
in cultured Kupffer cells (19, 36, 54), peritoneal (1, 63),
alveolar (48), and bone-derived macrophage (30, 59). Additionally, the dampened HPA response in indomethacin-treated
animals likely reduced glucocorticoid inhibition of inflammatory cytokine production resulting in widespread increases in
inflammatory cytokine production. These data are of particular
interest because nonsteroidal anti-inflammatory drugs are commonly used for the relief of pain and inflammation, yet chronic
use is associated with an increased risk of atherosclerosis and
thrombosis (6, 25, 28). It is possible that COX inhibitors create
a prothrombotic state by elevating proinflammatory cytokines
responses, which upregulate adhesion molecules on endothelial
cells promoting leukocyte adhesion and aggregation to vessel
walls and increase the production of tissue factor while simultaneously preventing the formation of PGI2 that typically
inhibit proinflammatory cytokine production (44, 64) as well as
platelet activation (7) and aggregation (4, 9, 29, 33, 42).
Perspectives and Significance
Our studies demonstrate for the first time a complete time
course of the possible mediators involved in HPA activation
following peripheral E. coli challenge. The data indicate that
early PGE2 synthesis is not only responsible for early corticosterone responses following an immune challenge but is also
responsible for controlling inflammatory cytokine responses
that ultimately themselves fine tune the HPA response. It
should not be surprising that extremely rapid mechanisms,
even ACTH-independent mechanisms, have evolved to stimuAJP-Regul Integr Comp Physiol • VOL
R1655
late glucocorticoid production during bacterial infection given
their importance in reducing the risk of septic shock and
modulating immune responses to increase the chances of survival during infection.
GRANTS
This work was supported by National Institute of Mental Health Grant
R21MH-77004 and the Ohio Board of Regents Research Initiative.
DISCLOSURES
No conflicts of interest, financial or otherwise, are declared by the
author(s).
AUTHOR CONTRIBUTIONS
Author contributions: Z.R.Z. and J.D.J. conception and design of research; Z.R.Z., V.M.P., R.M.C., and J.D.J. performed experiments; Z.R.Z.,
V.M.P., and J.D.J. analyzed data; Z.R.Z., V.M.P., and J.D.J. interpreted
results of experiments; Z.R.Z. prepared figures; Z.R.Z. drafted manuscript;
J.D.J. edited and revised manuscript; J.D.J. approved final version of
manuscript.
REFERENCES
1. Akaogi J, Yamada H, Kuroda Y, Nacionales DC, Reeves WH, Satoh
M. Prostaglandin E2 receptors EP2 and EP4 are up-regulated in peritoneal
macrophages and joints of pristane-treated mice and modulate TNF-␣ and
IL-6 production. J Leukoc Biol 76: 227–236, 2004.
2. Amano Y, Lee SW, Allison AC. Inhibition by glucocorticoids of the
formation of interleukin-1␣, interleukin-1␤, and interleukin-6: mediation
by decreased mRNA stability. Mol Pharmacol 43: 176 –182, 1993.
3. Baumann H, Gauldie J. The acute phase response. Immunol Today 15:
74 –80, 1994.
4. Belch JJ, McLaren M, Lau CS, Mackay IR, Bancroft A, McEwen J,
Thompson JM. Cicaprost, an orally active prostacyclin analogue: its
effects on platelet aggregation and skin blood flow in normal volunteers.
Br J Clin Pharmacol 35: 643–647, 1993.
5. Bjerre A, Brusletto B, Mollnes TE, Fritzsonn E, Rosenqvist E,
Wedege E, Namork E, Kierulf P, Brandtzaeg P. Complement activation
induced by purified Neisseria meningitidis lipopolysaccharide (LPS),
outer membrane vesicles, whole bacteria, and an LPS-free mutant. J Infect
Dis 185: 220 –228, 2002.
6. Bunimov N, Laneuville O. Cyclooxygenase inhibitors: instrumental
drugs to understand cardiovascular homeostasis and arterial thrombosis.
Cardiovasc Hematol Disord Drug Targets 8: 268 –277, 2008.
7. Cheng Y, Austin SC, Rocca B, Koller BH, Coffman TM, Grosser T,
Lawson JA, FitzGerald GA. Role of prostacyclin in the cardiovascular
response to thromboxane A2. Science 296: 539 –541, 2002.
8. Crane JW, Buller KM. Systemic blockade of complement C5a receptors
reduces lipopolysacharride-induced responses in the paraventricular nucleus and the central amygdala. Neurosci Lett 424: 10 –15, 2007.
9. Demolis JL, Robert A, Mouren M, Funck-Brentano C, Jaillon P.
Pharmacokinetics and platelet antiaggregating effects of beraprost, an oral
stable prostacyclin analogue, in healthy volunteers. J Cardiovasc Pharmacol 22: 711–716, 1993.
10. Dhabhar FS, Miller AH, McEwen BS, Spencer RL. Stress-induced
changes in blood leukocyte distribution. Role of adrenal steroid hormones.
J Immunol 157: 1638 –1644, 1996.
11. Dunn AJ. Cytokine activation of the HPA axis. Ann NY Acad Sci 917:
608 –617, 2000.
12. Dunn AJ. Effects of the IL-1 receptor antagonist on the IL-1- and
endotoxin-induced activation of the HPA axis and cerebral biogenic
amines in mice. Neuroimmunomodulation 7: 36 –45, 2000.
13. Dunn AJ. The role of interleukin-1 and tumor necrosis factor-␣ in the
neurochemical and neuroendocrine responses to endotoxin. Brain Res Bull
29: 807–812, 1992.
14. Elander L, Engstrom L, Ruud J, Mackerlova L, Jakobsson PJ,
Engblom D, Nilsberth C, Blomqvist A. Inducible prostaglandin E2
synthesis interacts in a temporally supplementary sequence with constitutive prostaglandin-synthesizing enzymes in creating the hypothalamicpituitary-adrenal axis response to immune challenge. J Neurosci 29:
1404 –1413, 2009.
301 • DECEMBER 2011 •
www.ajpregu.org
R1656
TIME-DEPENDENT MEDIATORS OF THE HPA AXIS FOLLOWING E. COLI
15. Elander L, Ruud J, Korotkova M, Jakobsson PJ, Blomqvist A.
Cyclooxygenase-1 mediates the immediate corticosterone response to
peripheral immune challenge induced by lipopolysaccharide. Neurosci
Lett 470: 10 –12, 2010.
16. Elenkov IJ, Chrousos GP. Stress hormones, proinflammatory and antiinflammatory cytokines, and autoimmunity. Ann NY Acad Sci 966: 290 –
303, 2002.
17. Fantuzzi G, Ghezzi P. Glucocorticoids as cytokine inhibitors: role in
neuroendocrine control and therapy of inflammatory diseases. Mediators
Inflamm 2: 263–270, 1993.
18. Fattori E, Cappelletti M, Costa P, Sellitto C, Cantoni L, Carelli M,
Faggioni R, Fantuzzi G, Ghezzi P, Poli V. Defective inflammatory
response in interleukin 6-deficient mice. J Exp Med 180: 1243–1250,
1994.
19. Fennekohl A, Sugimoto Y, Segi E, Maruyama T, Ichikawa A, Puschel
GP. Contribution of the two Gs-coupled PGE2-receptors EP2-receptor
and EP4-receptor to the inhibition by PGE2 of the LPS-induced TNF␣formation in Kupffer cells from EP2-or EP4-receptor-deficient mice.
Pivotal role for the EP4-receptor in wild type Kupffer cells. J Hepatol 36:
328 –334, 2002.
20. Franchimont D. Overview of the actions of glucocorticoids on the
immune response: a good model to characterize new pathways of immunosuppression for new treatment strategies. Ann NY Acad Sci 1024:
124 –137, 2004.
21. Francis K, Lewis BM, Akatsu H, Monk PN, Cain SA, Scanlon MF,
Morgan BP, Ham J, Gasque P. Complement C3a receptors in the
pituitary gland: a novel pathway by which an innate immune molecule
releases hormones involved in the control of inflammation. FASEB J 17:
2266 –2268, 2003.
22. Francis K, Lewis BM, Monk PN, Ham J. Complement C5a receptors in
the pituitary gland: expression and function. J Endocrinol 199: 417–424,
2008.
23. Garcia-Bueno B, Serrats J, Sawchenko PE. Cerebrovascular cyclooxygenase-1 expression, regulation, and role in hypothalamic-pituitaryadrenal axis activation by inflammatory stimuli. J Neurosci 29: 12970 –
12981, 2009.
24. Girard MT, Hjaltadottir S, Fejes-Toth AN, Guyre PM. Glucocorticoids enhance the ␥-interferon augmentation of human monocyte immunoglobulin G Fc receptor expression. J Immunol 138: 3235–3241, 1987.
25. Gitlin JM, Loftin CD. Cyclooxygenase-2 inhibition increases lipopolysaccharide-induced atherosclerosis in mice. Cardiovasc Res 81: 400 –407,
2009.
26. Givalois L, Dornand J, Mekaouche M, Solier MD, Bristow AF, Ixart
G, Siaud P, Assenmacher I, Barbanel G. Temporal cascade of plasma
level surges in ACTH, corticosterone, and cytokines in endotoxin-challenged rats. Am J Physiol Regul Integr Comp Physiol 267: R164 –R170,
1994.
27. Goppelt-Struebe M. Molecular mechanisms involved in the regulation of
prostaglandin biosynthesis by glucocorticoids. Biochem Pharmacol 53:
1389 –1395, 1997.
28. Grosser T, Fries S, FitzGerald GA. Biological basis for the cardiovascular consequences of COX-2 inhibition: therapeutic challenges and opportunities. J Clin Invest 116: 4 –15, 2006.
29. Gryglewski RJ, Bunting S, Moncada S, Flower RJ, Vane JR. Arterial
walls are protected against deposition of platelet thrombi by a substance
(prostaglandin X) which they make from prostaglandin endoperoxides.
Prostaglandins 12: 685–713, 1976.
30. Harizi H, Norbert G. Inhibition of IL-6, TNF-␣, and cyclooxygenase-2
protein expression by prostaglandin E2-induced IL-10 in bone marrowderived dendritic cells. Cell Immunol 228: 99 –109, 2004.
31. Hayley S, Kelly O, Anisman H. Corticosterone changes in response to
stressors, acute and protracted actions of tumor necrosis factor-␣, and
lipopolysaccharide treatments in mice lacking the tumor necrosis factor-␣
p55 receptor gene. Neuroimmunomodulation 11: 241–246, 2004.
32. Johnson JD, O’Connor KA, Deak T, Spencer RL, Watkins LR, Maier
SF. Prior stressor exposure primes the HPA axis. Psychoneuroendocrinology 27: 353–365, 2002.
33. Kozek-Langenecker SA, Spiss CK, Michalek-Sauberer A, Felfernig
M, Zimpfer M. Effect of prostacyclin on platelets, polymorphonuclear
cells, and heterotypic cell aggregation during hemofiltration. Crit Care
Med 31: 864 –868, 2003.
34. Lee SW, Tsou AP, Chan H, Thomas J, Petrie K, Eugui EM, Allison
AC. Glucocorticoids selectively inhibit the transcription of the interleukin
AJP-Regul Integr Comp Physiol • VOL
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
1␤ gene and decrease the stability of interleukin 1␤ mRNA. Proc Natl
Acad Sci USA 85: 1204 –1208, 1988.
Lenczowski MJ, Van Dam AM, Poole S, Larrick JW, Tilders FJ. Role
of circulating endotoxin and interleukin-6 in the ACTH and corticosterone
response to intraperitoneal LPS. Am J Physiol Regul Integr Comp Physiol
273: R1870 –R1877, 1997.
Li Z, Perlik V, Feleder C, Tang Y, Blatteis CM. Kupffer cellgenerated PGE2 triggers the febrile response of guinea pigs to intravenously injected LPS. Am J Physiol Regul Integr Comp Physiol 290:
R1262–R1270, 2006.
Liege S, Moze E, Kelley KW, Parnet P, Neveu PJ. Activation of the
hypothalamic-pituitary-adrenal axis in IL-1␤-converting enzyme-deficient
mice. Neuroimmunomodulation 7: 189 –194, 2000.
Masferrer JL, Reddy ST, Zweifel BS, Seibert K, Needleman P, Gilbert
RS, Herschman HR. In vivo glucocorticoids regulate cyclooxygenase-2
but not cyclooxygenase-1 in peritoneal macrophages. J Pharmacol Exp
Ther 270: 1340 –1344, 1994.
Matsuoka T, Narumiya S. The roles of prostanoids in infection and
sickness behaviors. J Infect Chemother 14: 270 –278, 2008.
McEwen BS, Biron CA, Brunson KW, Bulloch K, Chambers WH,
Dhabhar FS, Goldfarb RH, Kitson RP, Miller AH, Spencer RL, Weiss
JM. The role of adrenocorticoids as modulators of immune function in
health and disease: neural, endocrine and immune interactions. Brain Res
Brain Res Rev 23: 79 –133, 1997.
Mohn CE, Fernandez-Solari J, De Laurentiis A, Prestifilippo JP, de la
Cal C, Funk R, Bornstein SR, McCann SM, Rettori V. The rapid
release of corticosterone from the adrenal induced by ACTH is mediated
by nitric oxide acting by prostaglandin E2. Proc Natl Acad Sci USA 102:
6213–6218, 2005.
Moncada S, Gryglewski RJ, Bunting S, Vane JR. A lipid peroxide
inhibits the enzyme in blood vessel microsomes that generates from
prostaglandin endoperoxides the substance (prostaglandin X) which prevents platelet aggregation. Prostaglandins 12: 715–737, 1976.
Nasushita R, Watanobe H, Takebe K. A comparative study of
adrenocorticotropin-releasing activity of prostaglandins E1, E2, F2␣
and D2 in the rat. Prostaglandins Leukot Essent Fatty Acids 56:
165–168, 1997.
Nishimura T, Yamamoto T, Ueda Y, Yamada K, Hara Y. Stable
prostaglandin I1 analog SM-10906 modulates productions of tumor necrosis factor-␣, interleukin-1 and interleukin-6 in mouse macrophages.
Biol Pharm Bull 23: 708 –711, 2000.
Perlik V, Li Z, Goorha S, Ballou LR, Blatteis CM. LPS-activated
complement, not LPS per se, triggers the early release of PGE2 by
Kupffer cells. Am J Physiol Regul Integr Comp Physiol 289: R332–
R339, 2005.
Perlstein RS, Whitnall MH, Abrams JS, Mougey EH, Neta R. Synergistic roles of interleukin-6, interleukin-1, and tumor necrosis factor in the
adrenocorticotropin response to bacterial lipopolysaccharide in vivo. Endocrinology 132: 946 –952, 1993.
Puschel GP, Hespeling U, Oppermann M, Dieter P. Increase in prostanoid formation in rat liver macrophages (Kupffer cells) by human anaphylatoxin C3a. Hepatology 18: 1516 –1521, 1993.
Ratcliffe MJ, Walding A, Shelton PA, Flaherty A, Dougall IG. Activation of E-prostanoid 4 and E-prostanoid 2 receptors inhibits TNF-␣
release from human alveolar macrophages. Eur Respir J 29: 986 –994,
2007.
Rivest S. Interactions between the immune and neuroendocrine systems.
Prog Brain Res 181: 43–53, 2010.
Schotanus K, Tilders FJ, Berkenbosch F. Human recombinant interleukin-1 receptor antagonist prevents adrenocorticotropin, but not interleukin-6 responses to bacterial endotoxin in rats. Endocrinology 133: 2461–
2468, 1993.
Serrats J, Schiltz JC, Garcia-Bueno B, van Rooijen N, Reyes TM,
Sawchenko PE. Dual roles for perivascular macrophages in immune-tobrain signaling. Neuron 65: 94 –106, 2010.
Sheagren JN. Septic shock and corticosteroids. N Engl J Med 305:
456 –458, 1981.
Spencer R, Kalman B, Bhabhar F. Role of endogenous glucocorticoids
in immune system function: regulation as well as counterregulation. In:
Handbook of Physiology. The Endocrine System. Coping with the Environment: Neural and Endocrine Mechanisms. New York: Oxford University Press, 2001, sect. 7, vol. IV, pt. V, chapt. 18, p. 381–424.
Treffkorn L, Scheibe R, Maruyama T, Dieter P. PGE2 exerts its effect
on the LPS-induced release of TNF-␣, ET-1, IL-1␣, IL-6 and IL-10 via the
301 • DECEMBER 2011 •
www.ajpregu.org
TIME-DEPENDENT MEDIATORS OF THE HPA AXIS FOLLOWING E. COLI
55.
56.
57.
58.
59.
EP2 and EP4 receptor in rat liver macrophages. Prostaglandins Other
Lipid Mediat 74: 113–123, 2004.
Turnbull AV, Rivier CL. Regulation of the hypothalamic-pituitaryadrenal axis by cytokines: actions and mechanisms of action. Physiol Rev
79: 1–71, 1999.
Vakharia K, Hinson JP. Lipopolysaccharide directly stimulates cortisol
secretion by human adrenal cells by a cyclooxygenase-dependent mechanism. Endocrinology 146: 1398 –1402, 2005.
van den Berg TK, van Die I, de Lavalette CR, Dopp EA, Smit LD, van
der Meide PH, Tilders FJ, Crocker PR, Dijkstra CD. Regulation of
sialoadhesin expression on rat macrophages. Induction by glucocorticoids
and enhancement by IFN-␤, IFN-␥, IL-4, and lipopolysaccharide. J
Immunol 157: 3130 –3138, 1996.
van Enckevort FH, Sweep CG, Span PN, Demacker PN, Hermsen CC,
Hermus AR. Reduced adrenal response to bacterial lipopolysaccharide in
interleukin-6-deficient mice. J Endocrinol Invest 24: 786 –795, 2001.
Vassiliou E, Jing H, Ganea D. Prostaglandin E2 inhibits TNF production
in murine bone marrow-derived dendritic cells. Cell Immunol 223: 120 –
132, 2003.
AJP-Regul Integr Comp Physiol • VOL
R1657
60. Wang J, Dunn AJ. The role of interleukin-6 in the activation of the
hypothalamo-pituitary-adrenocortical axis and brain indoleamines by endotoxin and interleukin-1␤. Brain Res 815: 337–348, 1999.
61. Watanabe T, Morimoto A, Morimoto K, Nakamori T, Murakami N.
ACTH release induced in rats by noradrenaline is mediated by prostaglandin E2. J Physiol 443: 431–439, 1991.
62. Watanobe H, Nasushita R, Takebe K. A study on the role of circulating
prostaglandin E2 in the adrenocorticotropin response to intravenous administration of interleukin-1␤ in the rat. Neuroendocrinology 62: 596 –
600, 1995.
63. Yamane H, Sugimoto Y, Tanaka S, Ichikawa A. Prostaglandin E2
receptors, EP2 and EP4, differentially modulate TNF-␣ and IL-6 production induced by lipopolysaccharide in mouse peritoneal neutrophils.
Biochem Biophys Res Commun 278: 224 –228, 2000.
64. Zhou W, Hashimoto K, Goleniewska K, O’Neal JF, Ji S, Blackwell TS,
Fitzgerald GA, Egan KM, Geraci MW, Peebles RS Jr. Prostaglandin I2
analogs inhibit proinflammatory cytokine production and T cell stimulatory function of dendritic cells. J Immunol 178: 702–710, 2007.
301 • DECEMBER 2011 •
www.ajpregu.org