| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of
*
Pediatric Immunology and
Neonatology, Wilhelmina Children’s Hospital of the University Medical Center, and
Rudolf Magnus Institute, Utrecht, The Netherlands
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
0.01) and incidence (p
0.01) of the inflammatory autoimmune disease experimental autoimmune
encephalomyelitis in adult life. In search of possible mechanisms
responsible for the increased susceptibility to experimental autoimmune
encephalomyelitis, we investigated the reactivity of the
hypothalamo-pituitary-adrenal axis and of immune cells in adult rats
after neonatal glucocorticoid treatment. We observed that neonatal
glucocorticoid treatment reduces the corticosterone response after an
LPS challenge in adult rats (p 0.001).
Interestingly, LPS-stimulated macrophages of glucocorticoid-treated
rats produce less TNF-
and IL-1
in adult life than control rats
(p < 0.05). In addition, splenocytes obtained from
adult rats express increased mRNA levels of the proinflammatory
cytokines IFN-
(p < 0.01) and TNF-
(p < 0.05) after neonatal glucocorticoid
treatment. Apparently, neonatal glucocorticoid treatment has permanent
programming effects on endocrine as well as immune functioning in adult
life. In view of the frequent clinical application of glucocorticoids
to preterm infants, our data demonstrate that neonatal glucocorticoid
treatment may be a risk factor for the development of (auto)immune
disease in man. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To investigate possible effects of neonatal GC treatment, we used a new animal model for early GC treatment. It has been suggested that the development of the human fetus during the last trimester of pregnancy is comparable to the development of the rat in the first weeks of neonatal life (5). In the hospital neonatal GC therapy is mainly applied in the 26th to 33rd weeks, which would normally be in the last trimester of pregnancy. Therefore, the rat can be used as an animal model to study the effects of early neonatal GC therapy in the premature infant. Furthermore, the dose and time scheme of GC treatment were used according to the method of Cummings et al. (1) and current clinical protocols for corticosteroid treatment. Finally, outbred rats were chosen to increase the clinical relevance of this model.
Experimental autoimmune encephalomyelitis (EAE) is an experimental autoimmune inflammatory disease that serves as an animal model for multiple sclerosis (6). The hypothalamo-pituitary-adrenal (HPA) axis is believed to play a major role in determining susceptibility to EAE (7). High plasma corticosterone (CORT) levels during disease are thought to suppress the inflammation (7). Indeed, Sternberg et al. (8) found that Lewis rats, which are highly susceptible to EAE, have reduced plasma CORT levels. Also in other rat strains an inverse correlation was found between disease susceptibility and plasma CORT response (8, 9).
GCs play a major role in the development of the immune system (10), implying that the immune system may be a potential target for the detrimental effects of GC in early life. GC inhibit cytokine expression by macrophages and T cells at both transcriptional and post-transcriptional level (reviewed in Ref. 11). Lymphocytes of transgenic mice with overall reduced GC receptor (GR) expression have increased proliferative responses (12), suggesting that these lymphocytes have an altered cytokine production. On the other hand, the expression of an antisense GR transgene in immature thymocytes, which specifically interferes with thymocyte maturation, decreases susceptibility to autoimmune disease (13). However, very few data, if any, exist that show the effects of neonatal GC exposure on cytokine production and susceptibility to autoimmune disease in later life.
In the current study we investigated the effects of neonatal GC treatment on the susceptibility to and severity of EAE. In addition, we examined possible mechanisms underlying the permanent adverse effects of neonatal GC treatment on immune functioning.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Ten-day-pregnant Wistar rats (250–280 g; Harlan-CPB, Zeist, The Netherlands) were housed individually. They were kept under conventional conditions (dark phase, 1900–0700 h) with free access to commercial rat food and water. Pups were born on days 22–23 of gestation. On the day of birth (designated day 0), all pups were removed from the nests, and eight healthy pups (four females and four males) were randomly placed back with each dam. Pups were weaned at 21 days of age and housed two or three per cage.
Experimental design
Newborn rats were injected with dexamethasone 21-phosphate (DEX) on neonatal day 1 (0.5 µg/g body weight i.p.), day 2 (0.3 µg/g), and day 3 (0.1 µg/g). A second group of rats was injected with equal volumes of sterile pyrogen-free saline (SAL). A third group received no treatment (UNTR) and was left undisturbed. Only female offspring were used for the experiments. The various studies were performed using different groups of rats.
EAE induction
Starting 4 days before the administration of myelin basic protein (MBP), rats were handled each day. At 8 wk of age rats were injected s.c. in one hind footpad with 100 µl of an emulsion containing 1500 µg of MBP (isolated from guinea pig brain) in 1 ml of PBS mixed with 1 ml of IFA (Difco, Detroit, MI), to which 10 mg of Mycobacterium tuberculosis H37Ra was added (Difco). This emulsion is referred to as MBP/CFA in the following sections. Injections were performed under halothane anesthesia. From day 7 postimmunization (p.i.) onward, rats were examined daily (weight and clinical disease) until day 20 p.i. Neurologic aberrations were graded from 0 to 5: 0, no EAE; 0.5, loss of tip tail tonus; 1, loss of tail tonus; 2, partial tail paralysis; 3, complete tail paralysis; 4, hind limb paresis; and 5, hind limb paralysis. Disease severity was scored by observers in a blinded fashion.
HPA axis activation
Before the start of the experiment, 9-wk-old female rats were moved to a separate room and handled each day for 4 days. On the day of the experiment (0900 h), rats were injected with LPS (2.5 µg/kg, 2.5 µg/ml, i.p.). A control group was injected with equal volumes of saline. At 120 min after injection all rats were decapitated.
CORT assay
Trunk blood was collected in EDTA-containing tubes, centrifuged (3000 rpm, 10 min, 4°C), and stored (-20°C) until assay. Plasma CORT levels were determined by RIA (ICN, Costa Mesa, CA). The sensitivity was 1.5 ng/ml plasma. The intra- and interassay variations were 7.0 and 8.0%, respectively. Plasma CORT concentrations are expressed as nanograms per milliliter of plasma.
Macrophage stimulation
Both 9- to 11-wk-old rats (Expt. 1) and 20- to 25-wk-old rats
(Expt. 2) were used. After decapitation, 20 ml of ice-cold RPMI 1640
was injected into the peritoneum, and peritoneal macrophages (M
)
were harvested after gentle massage of the peritoneum (5 min). Cells
were seeded in 24-well plates (5 x
105/well) in culture medium containing FBS. After
1 h, nonadherent cells were removed by dispensing the culture
medium. M were cultured in the presence of varying concentrations of
LPS. Culture supernatants were collected after 24 h and stored at
-20°C until assay.
TNF-, IL-1, IL-12, and NO production
The TNF- content in supernatant was determined by ELISA
(U-Cytech, Utrecht, The Netherlands). The IL-1 content in
supernatant was also determined by ELISA (Steve Poole, National
Institute of Biological Standards and Control, Pottersbar, U.K.) in
collaboration with Dr. A. van Dam (Department of Pharmacology, Vrije
Universiteit, Amsterdam, The Netherlands). The IL-12p40/p70 content in
the supernatant was measured by ELISA (BioSource, Camarillo, CA). OD
was read at 450 nm.
In a separate set of experiments nitrite/nitrate was measured in supernatant of cells cultured in Iscove’s medium (24 h, 37°C, 5% CO2) supplemented with Nutridoma SP (Roche, Indianapolis, IN). Nitrate was converted to nitrite by the action of nitrate reductase from Asperigillus niger (Sigma, St. Louis, MO) (14). Briefly, supernatants were incubated with 40 µM NADPH (to initiate the reaction) and 14 mU of enzyme in a final volume of 50 µl of 20 mM Tris, pH 7.6. Reaction was terminated by addition of 50 µl of H2O. The metabolic product nitrite in the supernatant was quantified using Griess reagent (Promega, Madison, WI). OD was read at 540 nm.
RNase protection assay (RPA)
After dissection, splenic cells were minced through a 100-µm pore size sterile filter. Cells (3 x 106/ml) were stimulated for 48 h with Con A (1.0 µg/ml). RNA was isolated using RNAzol B (Cinna/Tel-Test, Friendswood, TX). Briefly, cultured cells were lysed by addition of RNAzol B. After addition of chloroform and centrifugation (12,000 x g, 4°C), isopropanolol was added to the aqueous phase. After storage (15 min, 4°C) and centrifugation (15 min, 12,000 x g) the RNA precipitate was washed with ethanol and air-dried. The RNA pellet was dissolved in hybridization buffer (PharMingen, San Diego, CA) and stored (-80°C) until use. For determination of rat cytokine mRNA expression, the multiprobe template set for rat cytokines (PharMingen, San Diego, CA) was used. Probe synthesis, hybridization, RNase treatments, and gel electrophoresis were performed according to the manufacturer’s instructions.
Statistics
Data from EAE experiments were evaluated with Friedman’s
two-way ANOVA, Kruskal-Wallis one-way ANOVA, and
2 test. Data from LPS experiments and RPA
experiments were analyzed by one-way ANOVA or Student’s t
test, in the case of the LPS experiments followed by Fisher’s least
significance difference test. Statistical analysis was performed using
SPSS for Windows, version 6.1 (SPSS, Chicago, IL).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Neonatal DEX treatment reduced the body weights of the rats at 8 wk of age compared with those in SAL and UNTR groups (UNTR, 166.4 ± 2.51 g (n = 22); SAL, 167.3 ± 2.91 g (n = 23); DEX, 154.3 ± 3.70 g (n = 26); p < 0.05, by ANOVA).
Neonatal DEX treatment increases the severity and incidence of EAE
At 8 wk of age, rats were immunized with MBP/CFA. Seven days
later, the first clinical signs of EAE appeared. In Fig. 1
the mean severity of disease is shown
from day 7 until day 20 p.i. No differences were noted in the day
of onset among the three experimental groups. Comparison of the mean
cumulative score shows that DEX treatment results in increased severity
of EAE compared with the controls (UNTR, 5.48 ± 0.92; SAL,
7.80 ± 1.58; DEX, 11.69 ± 1.23; DEX vs SAL,
p < 0.01, by ANOVA). The mean cumulative score for EAE
in the SAL group was not different from the score in the UNTR group (by
ANOVA). Neonatal DEX treatment also led to a higher incidence of
disease, defined as the percentage of animals with a disease score of 2
or more, compared with the controls (DEX group, 73.1%
(n = 26); SAL group, 34.7% (n = 23);
p < 0.01, by 2 test). No
effects of neonatal SAL treatment on the incidence of disease were seen
compared with nontreatment (n = 22).
|
Enhanced susceptibility to autoimmune disease is correlated with a
reduction in the CORT response (7). To examine the CORT
response to an immune challenge, adult rats (9 wk of age) were injected
i.p. with 2.5 µg/kg LPS and decapitated after 120 min, i.e., at the
peak of the CORT response (15). In Fig. 2 it is shown that in control rats
injection of LPS induced an 8-fold increase in the plasma CORT
concentration compared with vehicle injection. In the group of rats
that had been subjected to neonatal DEX treatment, a decreased
LPS-induced CORT response was observed compared with that in the SAL
group (p < 0.001, by ANOVA). No effects of
neonatal SAL treatment on LPS-induced increases in plasma CORT levels
were seen compared with the UNTR group (by ANOVA). We have no
indication that the CORT response in DEX-treated animals followed a
time course different from that in controls (data not shown).
|
and
IL-1 after neonatal DEX treatment
M mediate the LPS-induced activation of the HPA axis
(15). To investigate whether reduced production of TNF-
and IL-1 by M of neonatally DEX-treated rats plays a role in the
reduced CORT response in these animals, M from adult rats were
isolated and stimulated with LPS. In Fig. 3 it is shown that LPS increases TNF-
and IL-1 secretion dose-dependently, with a maximum at 3–10 ng/ml
LPS. Neonatal DEX treatment reduces the capacity of M to produce
TNF- after LPS stimulation compared with the controls (Fig. 3A;
treatment effect, p < 0.05, by ANOVA). Apart from a
decrease in TNF- production in rats treated neonatally with DEX,
LPS-induced IL-1 production by M was also decreased compared with
that in controls (Fig. 3B; treatment effect, p < 0.05,
by ANOVA). No significant effects of neonatal DEX treatment on
LPS-induced NO production by M were seen (data not shown).
Interestingly, in 20- to 25-wk-old rats both TNF- (DEX, 65.2 ±
5.1 U/ml; SAL, 112.6 ± 11.6 U/ml; p = 0.01, by
ANOVA) and IL-1 (DEX, 156.8 ± 42.1 pg/ml; SAL, 374.3 ±
79 pg/ml; p = 0.03, by ANOVA) production by
LPS-stimulated (10 ng/ml) M were decreased in DEX-treated rats
compared with control rats.
|
with LPS (Fig. 4). A dose-dependent
increase in the production of IL-12 (p40/p70) was observed, with a
maximum at 100 ng/ml LPS. However, M obtained from both SAL- and
DEX-treated rats were equally capable of producing IL-12 upon LPS
stimulation (p > 0.1, by t
test).
|
An increase in EAE susceptibility in DEX-treated rats could be
caused by an increased ability to produce proinflammatory cytokines
(17). To investigate this possibility, spleen cells
obtained from naive 8-wk-old DEX-treated rats and control rats were
cultured in the presence of Con A. mRNA expression was detected after
48 h of stimulation. Fig. 5 shows
the mRNA expression of a representative experiment as a percentage of a
household gene (L32) in both DEX-treated and SAL-treated rat spleen
cells. Stimulation of rat spleen cells resulted in a relatively high
mRNA expression of TNF-, IL-2, IFN-, and TNF-. Furthermore,
significant production of IL-6 and IL-10 was found. Interestingly,
neonatal DEX treatment increased the expression of the cytokines
IFN- (p < 0.01, by t test), and
TNF- (p < 0.05, by t test).
Although the mean mRNA expression of IL-2 and of TNF- were also
increased in the neonatally DEX-treated rats, this was not
statistically significant (0.05 < p < 0.1, by
t test). No effects were seen on the expression of IL-6 and
of IL-10 (Fig. 5). Furthermore, there was no effect of the injection
procedure on cytokine mRNA expression (data not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and IL-1
production by M is affected over the long term by neonatal DEX
treatment. Finally, the mRNA expression of Th1-type cytokines in the
spleen was profoundly increased in adult rats after neonatal DEX
treatment. Apparently, exposure to GC in early neonatal life has
permanent programming effects on both immunocompetence and
neuroendocrine functioning in later life.
Our data, which show decreased reactivity of the HPA axis after
neonatal GC exposure, are supported by other studies
(18, 19, 20). Furthermore, we show that the HPA response in
neonatally DEX-treated rats is decreased after an immune challenge. The
latter suggests that during the induction and course of EAE, the HPA
response is reduced in treated animals. The endotoxin-induced
activation of the HPA axis is mediated via LPS-induced IL-1,
TNF-, and IL-6 production by M (21, 22, 23). Our
data, which show decreased IL-1 and TNF- production by M
after in vitro stimulation with LPS, suggest that the diminished M
response is responsible for the reduced HPA reactivity in adult life
after neonatal DEX treatment. The latter conclusion does not exclude
the possibility that neonatal DEX exposure also affects one or more
components of the HPA axis itself. The preliminary observation that
neonatally DEX-treated rats have a decreased HPA response after
exposure to a novel environment supports this idea (data not
shown).
The reduced cytokine response of M from DEX-treated rats to an
endotoxin challenge suggests that these M are less capable of
mounting an adequate response to invading pathogens. This could have
important consequences for the susceptibility to pathogens of infants
who have been treated with GCs in early life. Interestingly, in one
study increased reports of infectious diseases in infants were recorded
after antenatal GC treatment (3). Our data strongly
suggest that the increased incidence may be due to a relatively lower
activity of M in these infants.
IL-12 has been shown to be a key cytokine in the regulation of EAE
(16, 24). Moreover, GCs can directly inhibit the
production of IL-12 by monocytes and dendritic cells
(25, 26, 27). However, we did not find differences in
LPS-induced IL-12 production by M obtained from adult rats after
neonatal SAL or DEX treatment. This suggests that increased
susceptibility to EAE in DEX-treated rats cannot be explained by
increased production of IL-12 in these rats.
It is important to realize that not every neonatal manipulation exerts the same effect on immune responsiveness in later life. We found increased susceptibility to EAE after neonatal GC treatment. However, decreased susceptibility to autoimmunity in later life has been reported when rat pups were subjected to maternal deprivation from days 1–28 of life (28). Maternal deprivation is regarded as a stressor for the pups. It may be possible that exposure in neonatal life to GC, on the one hand, or to stress, on the other hand, leads to opposite effects with regard to disease susceptibility in adult life. Interestingly, in humans certain perinatal environmental stress factors (e.g., neonatal anesthesia and surgery) have been suggested to increase the susceptibility to develop allergy (29). It may well be that some manipulations (such as early life stress factors) increase susceptibility to allergy and asthma, whereas others (such as administration of GC in early life) increase susceptibility to autoimmune disease. Thus, the nature of the stimulus may be a decisive factor for the outcome.
Could the increased susceptibility to EAE in the DEX-treated animals be caused by the relatively low reactivity of the HPA axis in these rats? It is known that endogenous plasma CORT levels rise during the clinical phase of EAE (30). These enhanced plasma CORT levels are thought to prevent the immune response from overshooting. For example, Lewis rats with relatively reduced HPA reactivity compared with Fischer rats are more susceptible to autoimmune disease than Fischer rats (8, 9). Administration of GC inhibits autoimmune disease in Lewis rats, whereas administration of the GR antagonist RU486 increases the severity of disease in these animals (8, 31). In addition, enhanced plasma CORT levels may favor the production of Th2 cytokines over Th1 cytokines (25, 32, 33), thus reducing the severity of EAE (34). Because of their lower HPA reactivity, DEX-treated animals may therefore be less capable of suppressing the inflammation. Moreover, their Th1/Th2 balance may be shifted more toward Th1, resulting in a more severe EAE in the DEX-treated animals.
The profound increased mRNA expression of the Th1 cytokines IFN- and
TNF- is in line with the hypothesis that neonatal DEX treatment
shifts the Th1/Th2 balance toward Th1. Moreover, neonatal DEX treatment
may have altered not only the HPA axis but also neuroendocrine
circuitry at a higher brain level. Kurosawa et al. reported that
neonatal GC treatment increased the norepinephrine, dopamine, and
serotonin contents of several brain regions, including the hypothalamus
(35). This is interesting because high serotonin levels
promote cell-mediated immune responses (36). Dopamine may
also play a pivotal role in this phenomenon. In an earlier publication
we described that the sensitivity for a dopaminergic agonist is
associated with Th1/Th2 balance and induction of autoimmunity, i.e.,
EAE (9). We demonstrated that a low sensitivity to the
dopaminergic agonist apomorphine is associated with a shift toward Th1
cytokines, resulting in relatively more IFN- mRNA expression.
Moreover, Rots et al. reported that a low sensitivity to dopamine is
functionally associated with a relatively low activity of the HPA axis
(37). Therefore, we propose that neonatal DEX treatment
has altered the Th1/Th2 balance on the basis of a permanent shift in
neuroendocrine circuitry involving dopamine and serotonin and
consequently changes in HPA axis function.
Taken together, early neonatal GC treatment increases the susceptibility to EAE. This suggests that early neonatal GC treatment may be a risk factor for MS or other proinflammatory autoimmune diseases. The mechanisms involved will be a low reactivity of the HPA axis in adult rats after neonatal DEX treatment and increased Th1 cytokine production. Furthermore, neonatally GC-treated rats have a long term decreased production of cytokines by macrophages. This may lead to increased susceptibility to bacterial infections in later life. Clearly, studies in man are urgently needed that focus on the consequences of neonatal GC therapy in infants for susceptibility to (auto)immune diseases in later life.
|
Acknowledgments |
|---|
in culture supernatants. We also thank
J. C. Frankhuijzen for her expert technical assistance, and Prof.
Dr. V. M. Wiegant for the helpful discussions. |
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Joost M. Bakker, Department of Pediatric Immunology, KC 03.068.0, Wilhelmina Children’s Hospital of the University Medical Center, Lundlaan 6, 3584 E. A. Utrecht, The Netherlands.
3 Abbreviations used in this paper: GC, glucocorticoid; EAE, experimental autoimmune encephalomyelitis; HPA axis, hypothalamo-pituitary-adrenal axis; CORT, corticosterone; GR, GC receptor; DEX, dexamethasone; SAL, saline; UNTR, untreated; MBP, myelin basic protein; p.i., postimmunization; M, macrophage; RPA, RNase protection assay.
Received for publication April 12, 2000. Accepted for publication August 16, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and TNF on the activity of the pituitary-adrenal axis and on food intake of rats. Am. J. Physiol. 268:E551.This article has been cited by other articles:
|
|
 |
|
  R. Karemaker, J. M. Karemaker, A. Kavelaars, M. Tersteeg-Kamperman, W. Baerts, S. Veen, J. F. Samsom, F. van Bel, and C. J. Heijnen Effects of Neonatal Dexamethasone Treatment on the Cardiovascular Stress Response of Children at School Age Pediatrics, November 1, 2008; 122(5): 978 - 987. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Karemaker, A. Kavelaars, M. ter Wolbeek, M. Tersteeg-Kamperman, W. Baerts, S. Veen, J. F. Samsom, G. H. A. Visser, F. van Bel, and C. J. Heijnen Neonatal Dexamethasone Treatment for Chronic Lung Disease of Prematurity Alters the Hypothalamus-Pituitary-Adrenal Axis and Immune System Activity at School Age Pediatrics, April 1, 2008; 121(4): e870 - e878. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Schwartz and J. Kipnis Multiple Sclerosis as a By-Product of the Failure to Sustain Protective Autoimmunity: A Paradigm Shift Neuroscientist, October 1, 2002; 8(5): 405 - 413. [Abstract] [PDF]
|
|
|
|
 |
|
 R R Dietert, J-E Lee, and T L Bunn Developmental immunotoxicology: emerging issues Human and Experimental Toxicology, September 1, 2002; 21(9-10): 479 - 485. [Abstract] [PDF]
|
|
|
|
 |
|
 B. Marchetti, M. C. Morale, J. Brouwer, C. Tirolo, N. Testa, S. Caniglia, N. Barden, S. Amor, P. A. Smith, and C. D. Dijkstra Exposure to a Dysfunctional Glucocorticoid Receptor from Early Embryonic Life Programs the Resistance to Experimental Autoimmune Encephalomyelitis Via Nitric Oxide-Induced Immunosuppression J. Immunol., June 1, 2002; 168(11): 5848 - 5859. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Wang, R. Charboneau, S. Balasubramanian, R. A. Barke, H. H. Loh, and S. Roy The immunosuppressive effects of chronic morphine treatment are partially dependent on corticosterone and mediated by the {micro}-opioid receptor J. Leukoc. Biol., May 1, 2002; 71(5): 782 - 790. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. B. Flagel, D. M. Vazquez, S. J. Watson Jr., and C. R. Neal Jr. Effects of tapering neonatal dexamethasone on rat growth, neurodevelopment, and stress response Am J Physiol Regulatory Integrative Comp Physiol, January 1, 2002; 282(1): R55 - R63. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
