You are on page 1of 11

Gene profiling reveals unknown enhancing and

suppressive actions of glucocorticoids


on immune cells
ME GALON,*,,1,2 DENIS FRANCHIMONT,,,1 NAOKI HIROI,, GREGORY FREY,
JERO
ANTJE BOETTNER, MONIKA EHRHART-BORNSTEIN,, JOHN J. OSHEA,*
GEORGE P. CHROUSOS, AND STEFAN R. BORNSTEIN,
*Lymphocyte Cell Biology Section, NIAMS, and Pediatric and Reproductive Endocrinology Branch,
NICHD, National Institutes of Health, Bethesda, Maryland 20892, USA; INSERM U255, Curie
Institute, Paris, France; Department of Gastroenterology, Erasme Hospital, Brussels, Belgium;
and Department of Pediatrics, University of Leipzig, 04103 Leipzig, and Department of
Endocrinology, University of Dusseldorf, 40225 Dusseldorf, Germany
Glucocorticoids continue to be the major immunomodulatory agents used in clinical medicine
today. However, their actions as anti-inflammatory and
immunosuppressive drugs are both beneficial and deleterious. We analyzed the effect of glucocorticoids on
the gene expression profile of peripheral blood mononuclear cells from healthy donors. DNA microarray
analysis combined with quantitative TaqMan PCR and
flow cytometry revealed that glucocorticoids induced
the expression of chemokine, cytokine, and complement family members as well as of newly discovered
innate immune-related genes, including scavenger and
Toll-like receptors. In contrast, glucocorticoids repressed the expression of adaptive immune-related
genes. Simultaneous inhibitory and stimulatory effects
of glucocorticoids were found on inflammatory T
helper subsets and apoptosis-related gene clusters. In
cells activated by T cell receptor cross-linking, glucocorticoids down-regulated the expression of specific genes
that were previously up-regulated in resting cells, suggesting a potential new mechanism by which they exert
positive and negative effects. Considering the broad
and continuously renewed interest in glucocorticoid
therapy, the profiles we describe here will be useful in
designing more specific and efficient treatment strategies.Galon, J., Franchimont, D., Hiroi, N., Frey, G.,
Boettner, A., Ehrhart-Bornstein, M., OShea, J. J.,
Chrousos, G. P., Bornstein, S. R. Gene profiling reveals
unknown enhancing and suppressive actions of glucocorticoids on immune cells. FASEB J. 16, 6171
(2002)

ABSTRACT

Key Words: DNA microarray clinical medicine immune


system

Inflammation is the physiological homeostatic response elicited to shield our tissues from injurious
agents. The regulation of the immense array of cellular
and molecular mediators involved in inflammation is
harmonized by the immune and neuroendocrine sys0892-6638/02/0016-0061 FASEB

tems (1, 2). Adrenal-derived glucocorticoids (GC) assist


the innate and adaptive immunity to eliminate toxins,
tumor cells, and foreign pathogens to limit tissue
damage (1, 3, 4). The broad and triumphant clinical
use of synthetic GC has improved the life of millions of
patients with inflammatory/autoimmune diseases and
graft rejection during the past 50 years (5). Renewed
interest is surfacing within the field of GC therapy for
patients with severe sepsis, septic shock, and the acute
respiratory distress syndrome (ARDS), based on evidence from several clinical trials where prolonged GC
treatment resulted in significant improvement and a
decreased mortality (6 9). However, GC may also
worsen the course of such diseases and prime deleterious inflammatory cascades, resulting in death (10).
A great deal of effort has been focused on trying to
understand the cellular and molecular mechanisms of
action of GC (5, 11). In vivo and in vitro studies have
suggested both enhancing and suppressive effects of
GC on the immune and inflammatory response, and
even more discrepancies have emerged (3, 12, 13). A
comprehensive analysis of gene regulation by GC has
not yet been performed. Therefore, we analyzed the
effect of GC on the gene expression profile of human
peripheral blood mononuclear cells (PBMC) from
healthy donors treated with dexamethasone (Dex). To
accomplish this, we used the novel technique of DNA
microarray chip analysis in combination with online
PCR and flow cytometry. First, our study showed that
GC regulated numerous newly discovered genes that
play critical roles in innate and adaptive immune
responses. Second, the data revealed a role for GC not
only as immunosuppressants, but also as major immunopermissive and immuno-enhancing agents. Finally,
we showed that GC could regulate some genes in
1

These authors contributed equally to the work.


Correspondence: Laboratoire dImmunlogie Cellulaire et
Clinique, Unite INSERM U255, Centre de Recherches Biomedicales des Cordeliers, 15 Rue de lEcole de Medecine, 75270 Paris
Cedex 06 France. E-mail: jerome.galon@u255.bhdc.jussieu.fr
2

61

opposite directions, depending on the state of activation of the target cells. This study supports a major and
highly complex regulatory role of GC in human immune functions.

were isolated by Ficoll-Paque gradient centrifugation (Pharmacia, Piscataway, NJ), activated or not with coated anti-CD3/
anti-CD28 antibodies at 10 g/ml, and stimulated with dexamethasone (107 M) for 18 h in RPMI 1640 containing 1%
FCS. TaqMan primers are described in the online Table 1
supplemental.

MATERIALS AND METHODS

Microarrays

Antibodies and cells

Peripheral blood mononuclear cells were prepared as described. RNA extraction was performed (RNAgents, Promega,
Madison, WI) and mRNA was purified with oligoTex mRNA
isolation columns (Qiagen, Valencia, CA). The cDNA made
from 600 ng of purified mRNA from each sample was labeled
with the fluorescent dyes Cy5 and Cy3. The two cDNA probes
were mixed and simultaneously hybridized to a microarray
containing 9182 human expressed genes (GEM microarray;

Monoclonal mouse anti-human CD3 and anti-CD28, FITC- or


PE-conjugated monoclonal anti-CD68, anti-TNFR1, antiCD36, anti-CD163, anti-CD27, anti-CD49d, anti-CD39, antiHLA-DR, anti-CD74, anti-TCR-, anti-TCR-, anti-CD127,
and isotype-matched IgG controls were purchased from
PharMingen (San Diego, CA). PBMC from healthy donors

TABLE 1 Supplemental. Primer sequences, size and Genebank accession numbera


Name

Primer sequences (5-3)

Indoramine-pyrrole
TLR4
CD44
CD163
Integrine alphal
MARCO
THBS1
STAT1
IRF4
Diubiquitin
PAI2
CD127
IL10
CCR2

S
AS
TM
S
AS
TM
S
AS
TM
S
AS
TM
S
AS
TM
S
AS
TM
S
AS
TM
S
AS
TM
S
AS
TM
S
AS
TM
S
AS
TM
S
AS
TM
S
AS
TM
S
AS
TM

AGTCCGTGAGTTTGTCCTTTCAA
TTTCACACAGGCGTCATAAGCT
CCCGCAGGCCAGCATCACCT
AGAGTTTCCTGCAATGGATCAAG
TTATCTGAAGGTGTTGCACATTCC
TTCGTTCAACTTCCACCAAGAGCTGCCT
AAAAATGGTCGCTACAGCATCTC
GGTGCTATTGAAAGCCTTGCA
AGGTCAGCGGCCTCCGTCCG
GGTCGCTCATCCCGTCAGT
CGAAAATGGCCAACAGAACA
CCCAAGGATCCCGACTGCAATAAAGGAT
TCTGGTTTTAAGTAGCAGCAATCAA
GCAGCATTAACAGCAACAATCC
TGCCCGGTAGCCCATCTTTGGATATTT
GACAGCCCGTCCTTCTCCTT
GACTTGCAGCCTCGATGCA
AGCACACCCTGGAGAACACCTGGCT
CATCCGCAAAGTGACTGAAGAG
GTACTGAACTCCGTTGTGATAGCATAG
CCAATGAGCTGAGGCGGCCTCC
GTGGAAAGACAGCCCTGCAT
ACTGGACCCCTGTCTTCAAGAC
CAACGCACCCTCAGAGGCCGC
GCCCAGGTTCACAACTACATGAT
TTTCCGGGTGTGGCTGAT
ACCGAAGCTGGAGGGACTACGTCCC
ATGGCTCCCAATGCTTCCT
TGGCATCAAAGGTCATTAAATCC
CCTCTGTGTGCATGTCCGTTCCGA
GCATGGAGGACGCCTTCA
ACAGGTCATTCCTCTCCGACAT
CAAGGGACGGGCCAATTTCTCAGG
GCATGTGACGCCCCTATTCT
GGCCCATTCTTGCCACTCT
TCCTCTTCCAGGTCCCTAGACTGCAGG
ACGGCGCTGTCATCGATT
TGGAGCTTATTAAAGGCATTCTTCA
TGTGAAAACAAGAGCAAGGCCGTGGA
TGAGTAACTGTGAAAGCACCAGTCA
GCAGTGAGTCATCCCAAGAGTCT
CTGGACCAAGCCACGCAGGTGAC

Primer size

967989
10331012
9911010
19271949
20081985
19561983
140162
205185
164183
32133231
32823263
32333260
32613285
33913370
33053331
268287
345327
297321
10021023
10861060
10371058
11511170
12171196
11731193
612634
690673
646670
1937
9169
3962
10551072
11201099
10741097
11481167
12151197
11691195
399416
479455
424499
901925
974952
927949

Genebank no.

NM_002164
NM_003266
NM_000610
NM_004244
X68742
NM_006770
NM_003246
NM_007315
NM_002460
NM_006398
NM_002575
XM_004013
M57627
NM_000648

a
S: sense primer; AS: antisense primer; TM: TaqMan probe; Primer size: Base pairs; TLR4: Toll-like receptor 4; MARCO: macrophage
receptor with collagenouse structure; THBS1: thrombospondin 1; STAT1: signal transducer and activator of transcription 1; IRF4: interferon
regulatory factor 4; PAI2: plasminogen activator inhibitor type2; CD127: interleukin 7 receptor; IL10: interleukin 10; CCR2: chemokine receptor 2.

62

Vol. 16

January 2002

The FASEB Journal

GALON ET AL.

Incyte, St. Louis, MO). The microarray was scanned and the
intensity of the fluorescence at each array element was
proportional to the expression level of that gene in the
sample. The ratio of the two fluorescence intensities provided
a quantitative measurement of the relative gene expression
level in the two cell samples. The detectable dynamic range of
individual mRNA in a sample range from 2 to 2000 pg and
reproducibility was assessed on a validation study (www.incyte.
com/incyte science/technology/gem/reproducibility.
shtml). Using specially designed DNA control elements and a
known RNA set added to each probe labeling reaction, all
GEM microarray hybridizations are assessed for labeling,
hybridization efficiency, and sensitivity. A 96-well plate is
arrayed on each GEM microarray that contains a variety of
control elements arrayed in quadruplicate. The elements are
composed of DNA fragments derived from inter-ORF regions
in Saccharomyces cerevisiae. These fragments are all 1000 bp in
length and have minimal homology to any known gene, to
ensure that no cross-hybridization occurs with experimental
samples. A control set of 14 RNA molecules is generated from
these fragments using PCR with composite primers that
include a T7 promoter and a 30-base oligo-dT sequence. A
simple transcription reaction then generates artificial mRNA
fragments that hybridize specifically to individual control
elements. These fragments are added to the two probelabeling reactions in known amounts, together with the
experimental samples. A set of four fragments is added in
increasing concentrations in both the Cy3 and Cy5 reactions
to control for signal sensitivity. Six additional are added in
known ratios in the two fluorescent labeling reactions to
control for expected ratios. Four RNA fragments previously
converted to fluorescent DNA are either arrayed onto the
control plate or added to the experimental sample to control
for overall scanning and hybridization signal, independent of
either labeling reaction. Additional elements composed of
either housekeeping genes or mixed cDNA libraries are
arrayed in the last two rows of the control plate. These
elements are hybridized by probe from the experimental
sample, and their signal intensity is used to measure experimental RNA quality. All signal intensities in the control plate
as well as total GEM microarray hybridization signals are
collected. The values are then compared with a set of
minimum intensity criteria for each particular control. Because the controls differentiate between processes in the
labeling and hybridization reaction, the quality of experimental RNA can be measured independent of labeling or hybridization efficiency. An absolute balanced differential expression 1.4 was considered significant based on three
parameters: reproducibility of differential expression based
on a statistical study encompassing 70 differential hybridization, known genes regulated by GC under these conditions,
and TaqMan PCR confirmation. Differential expression accuracy and precision were determined by hybridizing samples
against themselves in replicates of 10. The expected differential expression ratio for each gene being 1, experimental
differential expression (calculated from 100,000 data points)
ranged from 0.99977 to 0.99998. Known and unknown GCinducible genes were confirmed for their up- or downregulation by GC using quantitative online PCR or flow
cytometry. GemTools software (version 2.5.0) was used to
analyze the gene distribution, and genes were clustered based
on GemTools clustering software and published literature.
Online TaqMan PCR
RT-PCR experiments were carried out according to the
Thermoscript RT-PCR system kit for RT-PCR instructions
provided by the manufacturer (Life Technologies, Gaithersburg, MD). Total RNA (1 g) of human lymphocyte were

reverse transcribed to complementary DNA (cDNA) by a


reaction containing 2 mM deoxynucleotide mix, 100 mM
DTT, 40 units RNase inhibitor, 50 ng random primer, and 15
units Thermoscript reverse transcriptase. The reaction was
run 25C for 10 min and 50C for 50 min, heated at 85C for
5 min, then cooled to 4C.
To quantify expression of IDO, TLR-4, CD44, CD163,
integrin alpha1, MARCO, TSP-1, STAT1, PLA2, IRF-4,
CD127, IL-10, CCR2, and diubiquitin, we applied the TaqMan
PCR using the 7700 Sequence Detector (Perkin Elmer Applied Biosystems, Foster, CA). Reactions contained 1 TaqMan Universal PCR Master Mix, 900 nM of forward and
reverse primers and 200 nM for the TaqMan-probes (Table 1,
supplemental). 18 S primers and probes were added at 50 nM.
Thermal cycling proceeded with 50 cycles of 95C for 15 s and
60 C for 1 min. Input RNA amounts were calculated with
relative standard curves for both the mRNAs of interest and 18S.
Statistical analysis
Microarrays
The genes were classified into three categories: not regulated,
up-regulated, and down-regulated. The statistical significance
of the correlation between the two DNA microarray experiments was determined using correlation matrix algorithm,
Kendall correlation algorithm, or Spearman correlation algorithm. Similar results were found with all algorithms. When
no fluorescent intensities were detected in one or both
experiments, the corresponding genes were removed from
the statistical analysis. Correlation analyses were performed
by plotting differential gene expression of one experimentagainst the value from the other experiment. The correlation
coefficient (r) and the statistical significance (P) were determined.
Taqman PCR
Input RNA amounts were calculated with relative standard
curves for all mRNA of interest and 18S RNA. Division by
the amount of 18S RNA in each sample corrected the
amounts of input mRNAs. Specific mRNA transcript levels
after Dex stimulation were expressed as percentage of
increase vs. unstimulated cells. Data are presented as
mean se and analyzed by ANOVA T test for each gene
from individual healthy donors (n6). The statistical significance (P) was determined and is represented for each
histogram.
Flow cytometry
Flow cytometry (FACS) was previously described (14). Cells
were washed and incubated with conjugated antibodies and
isotype-matched IgG-PE/and FITC-IgG as controls for 30 min
at 4C. Cells were washed three times with phosphate-buffered saline containing 0.5% bovine serum albumin and
analyzed by flow cytometry on a FACSCalibur flow cytometer
(Becton Dickinson, San Jose, CA).

RESULTS
Global analysis of gene expression profile
We analyzed PBMC gene expression changes mediated
through GC (Fig. 1). In a DNA microarray containing

GENE EXPRESSION PROFILES OF GLUCOCORTICOIDS ON IMMUNE CELLS

63

Figure 1. Global gene expression analysis. Untreated and


Dex-treated PBMC were analyzed for gene expression using
DNA microarray. a) Clustering analysis. Total number of
genes up- or down-regulated among 6 major clusters. b)
Subclustering analysis. The immune cluster was divided into
subclusters and subcategories; percentage of GC-regulated
genes within the immune cluster is presented. c) Correlation
analysis was performed by plotting differential gene expression of one experiment against the value from the other
experiment. Dot plot represents the correlation between data
points from the two experiments and histograms represent
the frequency of differential expression. Correlation coefficient (r0.85) between experiments (P0.0001).

9182 human expressed genes, 9% were considered


down-regulated (84541 genes) and 12% up-regulated
(1125135 genes) (Fig. 1a). The sensitivity and reproducibility of DNA microarrays are highlighted by the
fact that appropriate changes of known GC-regulated
genes were detected in separate DNA microarray experiments. Statistical significance of the correlation
between the two DNA microarray experiments was
determined. When no fluorescent intensities were detected in one or both experiments, the corresponding
genes were removed from the statistical analysis. One
experiment showed a greater difference among gene
64

Vol. 16

January 2002

expression values. This is represented by the histograms


(Fig. 1c). However, even with a broader differential
expression, the genes were coregulated in the same
manner in both experiments. Correlation analyses were
performed by plotting differential gene expression of
one experiment against the value from the other experiment. As shown in Fig. 1c, on average, we found a
correlation between data points from the two experiments. The correlation coefficient (r0.85) confirmed
the good correlation between experiments and the
correlation appears to be statistically significant
(P0.0001).
Using independent experimental samples (n6),
online PCR results and/or FACS analysis confirmed
in all cases the direction and magnitude of the gene
expression changes observed (panels a and b in
Tables 13; data not shown). Global gene expression
analysis, based on GemTools clustering software and
published literature, enabled us to quantitate the
relative contribution of GC to changes in gene
cluster expression. Of six major clusters analyzed,
immune response-related genes and unknown genes
(ESTs) were mostly regulated. The immune cluster
was divided into subclusters and subcategories (Fig.
1b). The first group of immune genes regulated by
GC included key anti-inflammatory and proinflammatory factors (Table 1). Global analysis revealed
symmetrical up- or down-regulation within the inflammation subcluster. In contrast, the subclustering
analysis revealed a shift toward innate compared with
adaptive recognition (Fig. 1b). Within the immune
system, a second group involved genes related to the
innate immune response that constitutes the first line
of defense during an inflammatory reaction (Table
2). Finally, we identified a third clusters of genes
involved the adaptive immune response (Table 3).
This group of genes specifically directs the appropriate type of cellular or humoral immune reaction
according to the nature of the injurious agent.
GC as anti-inflammatory agents
Inflammation is an important homeostatic mechanism that limits the effects of injurious agents, but
has the potential for inducing host tissue damage
and must be controlled by the organism. GC suppress
inflammation by down-regulating the expression of
proinflammatory cytokines such as IL-1, lymphotoxin-, IL-1, IL-8, IFN-, and IFN- (11). Indeed,
the expression of these cytokines was decreased by
GC (Table 1, panel a). A balance between the effects
of pro- and anti-inflammatory cytokines may determine the outcome of the inflammatory response.
Cytokines such as transforming growth factor (TGF)13 and IL-10 suppress the production of proinflammatory mediators. We found that the expression
of TGF-3, IL-10, and IL-10R was up-regulated by GC,
supporting their anti-inflammatory action (Table 1).
An important group of proinflammatory genes is that
of chemokines, small peptides that facilitate chemo-

The FASEB Journal

GALON ET AL.

TABLE 1.

taxis of leukocytes from the circulation to the tissues.


The prototypical chemoattractant IL-8 causes neutrophils to degranulate and cause tissue damage. The
expression of many chemokines, including IL-8, Mip1, Mip-3, Mcp-2, Mcp-3, Mcp-4, TARC, and
eotaxin, was down-regulated by dexamethasone treat-

ment (15, 16). In contrast, we found that I-309, IP-10,


and fractalkine were up-regulated by GC (Table 2).
Finally, IL-1RII, a decoy receptor limiting the deleterious effects of IL-1, was among the genes most
strongly up-regulated by GC (17).
Thus, many proinflammatory ligands were down-

GENE EXPRESSION PROFILES OF GLUCOCORTICOIDS ON IMMUNE CELLS

65

TABLE 2.

regulated whereas anti-inflammatory soluble mediators


were mostly up-regulated, by GC, highlighting their
immunosuppressive and protective role against inflammation.
GC induce proinflammatory mediators
A different pattern of regulation was observed for
immune-related membrane receptors and intracellular
mediators. Among genes most strongly up-regulated by
GC were some proinflammatory receptors IL-1RI, the
gp130 subunit of the IL-6 type receptors, IL-8R, IFNR,
CSFRI, CFSRII, IFNRI, IFNRII, and TNFR family
members (Table 1). Some of the intracellular mediators from the IL-1/Toll signaling pathway, such as
IRAK, were up-regulated by GC. In contrast, the antiinflammatory mediator, IL-1Ra, a soluble receptor antagonist released during inflammation, was among the
genes most strongly down-regulated by GC (18). An66

Vol. 16

January 2002

other hallmark of inflammation is the influx of immune cells, mediated by chemokines and chemotactic
factors of the complement system (19). Chemokine
receptors or associated molecules, including Dez, an
orphan G receptor-coupled protein, CCR1, CCR2, and
CCR2-like, were up-regulated by GC. The activation of
complement forms the first line of defense and causes
the release of potent chemotactic complement factors
C3a, C4a, and C5a. The proinflammatory components
of the classical pathway (C1q, C3, C5) and receptors
C3aR1, CR2, and C5aR1 were up-regulated. Our analysis also revealed that C8, a terminal complement
component involved in the lytic activity of complement,
was up-regulated by GC. In contrast, the two inhibitors
tightly controlling the complement cascade, C1-INH
and C4bp, were down-regulated.
The release of matrix metalloproteinases (MMPs)
allows leukocytes to extravasate and penetrate tissues, a
key event in inflammatory disease (20). They also act as

The FASEB Journal

GALON ET AL.

TABLE 3.

regulatory molecules by functioning in enzyme cascades and processing matrix proteins such as cytokines,
growth factors, and adhesion molecules to generate
fragments with biological effects. With the exception of
MMP-9, metalloproteinases were up-regulated by GC
(Table 2). Other molecules promoting chemotaxis of
leukocytes to inflammatory sites, such as the thrombospondins (TSPs) and their APC receptor (CD36), are
abundantly expressed in chronically inflamed tissues.
Thrombospondins are structurally related to MMPs and

regulate their functions (21, 22). The expression of TSP-1,


2, 4 and CD36 was strongly up-regulated by GC (Table 1,
panels a, b). In contrast, TSPs membrane receptor on T
cells (CD47) was down-regulated by GC (Table 2).
Therefore, proinflammatory mediators, such as
complement, or cytokine and chemokine receptors
were up-regulated by GC whereas soluble inhibitors
of these pathways were down-regulated, supporting
the positive, enhancing role of GC in the inflammatory response.

GENE EXPRESSION PROFILES OF GLUCOCORTICOIDS ON IMMUNE CELLS

67

GC support innate immunity and regulate specific


genes in opposite directions depending on the
activation state of the cells
A key aspect of the innate immune response is the
ability to discriminate between self and non-self molecules. This is achieved through pattern recognition
receptors, which directly recognize molecular epitopes
expressed by microbes. Scavenger receptors (SRs) are
also involved in the innate immune response for the
removal of damaged tissue and debris, and recognize a
wide variety of pathogens. CD163 expression is downregulated by proinflammatory and up-regulated by
anti-inflammatory mediators and GC (23, 24). Here we
show that CD163, as well as eight other scavenger
receptors, were strongly up-regulated by GC (Table 2,
panels a, b).
Finally, the IL-1R/Toll-like receptor (TLR) superfamily has emerged as an expanding family of receptors
whose function is to respond rapidly to infection and
injury (25). TLR-4 and TLR-2 mediate the host response to gram-negative and gram-positive bacteria,
respectively, and the finding that TLR-4 is important
for responses to LPS may allow for novel means to
intervene therapeutically during sepsis (25, 26). Three
members of the Toll family were regulated by GC;
TLR-4 and TLR-2 were up-regulated, and TLR-3 downregulated (Table 2). That GC regulate this superfamily
could be critical for many aspects of inflammation and
host defense, since TLRs control innate immune responses in vivo.
Besides the positive and negative role of GC on
immune gene clusters, we analyzed the potential differential action of GC at the single gene level. The
regulation of genes implicated in innate and adaptive
immune response in cells at different activation stages
was of particular interest. To mimic an antigenic response, we analyzed the action of GC on human
immune cells activated by T cell receptor cross-linking
with anti-CD3 and anti-CD28 antibodies. Among 20
genes tested by online PCR under these different
conditions, 25% revealed a bidirectional action of GC.
For instance, GC induced TLR-4 expression in the
resting condition, yet after T cell receptor activation,
GC decreased TLR-4 expression (n7) (Fig. 2a, b).
Indoleamine 2,3-dioxygenase (IDO), which participates in innate and adaptive immune responses, is a
novel antioxidant defense molecule. It prevents oxidative tissue damage and suppresses T cell-mediated
immune responses during inflammation (27, 28). The
expression of IDO was oppositely regulated by GC in
resting and activated conditions (n7) (Fig. 2c, d).
These results illustrate the complex bi-directional
regulation of a single gene by GC, depending on the
state of activation of the cells. This study reveals a major
positive role of GC on the innate immune system, and
supports differential actions of GC on the innate and
adaptive immune response during an antigen-mediated
immune reaction.
68

Vol. 16

January 2002

Figure 2. Quantitation of TLR-4 (a, b) and IDO (c, d) mRNA


expression by TaqMan PCR in resting (a, c) and CD3/CD28activated (b, d) PBMC stimulated or not with the indicated
concentrations of Dex. Data represent the mean se (n7).
*P 0.05; **P 0.01; ***P 0.001.

GC inhibit adaptive immunity


Innate immunity is the first line of defense and functions to protect the host from infection while slower
adaptive (acquired) responses are developing. Costimulatory molecules such as CD40-ligand (CD40L)/
CD40 are central regulators of the adaptive immune
response, and we found that GC down-regulated CD40
(Table 3). CD40 regulates immune responses through
its ability to enhance the expression of B-7 costimulatory molecules and through the increased production
of IL-12, a major Th1 cytokine, by monocytes and
dendritic cells. Thus, GC appeared to have an important role in down-regulating CD4 T cell-dependent
Th1 immune responses (29, 30). Furthermore, interferon regulatory factors (IRF) -1, -3, and -7 induced by
Th1 cytokines were down-regulated by GC. Conversely,
the expression of c-maf, the transcriptional regulatory
protein involved in Th2 differentiation and IL-4 transcription, was up-regulated by GC. Thus, corticosteroids
appear to bias the immune response toward Th2 (31
33). Another T cell type that produces only transforming growth factor- (TGF-), but not IL-4 or IFN-, has
been termed a Th3 cell. This type of cell has been
suggested to be important in mediating immune suppression in autoimmune disease. Six members of the
TGF family [TGF-, TGF-3, TGF--induced (68 kDa)
protein, TRAP-1, TGF-RI, TGF-RII, and TGF-RIII]
were up-regulated by GC (Table 1). Glucocorticoids are
potent immunosuppressants and induce apoptosis. At
the same time, they up-regulate 50% of antiapoptoticand proliferation-related molecules (Table 3 and data
not shown).

The FASEB Journal

GALON ET AL.

Given the essential function of MHC-II molecules in


T cell activation, the molecular mechanisms regulating
their complex expression profile obviously represent
an important parameter in the normal physiological
control of the immune response (34). Transcription of
MHC-II, CD74 (the invariant chain of the class II MHC
molecules, Ii), and MHC-II-DM genes is up-regulated
by the major CIITA complex, which is induced by
intracellular signals (such as STAT1) generated by
IFN-. Products of the MHC-II-DM genes are essential
for the peptide loading step and are modulated by
MHC-II-DO, which enhances stringent MHC/peptide
association through antigen receptor recognition (35).
Among the adaptive immunity clusters, 17 genes encoding MHC molecules, Ii, MHC-II-DM and -DO, TCR-,
TCR-, granzyme A and B, CD86, and STAT1 and
STAT5a were down-regulated by GC (Table 3, see
panels a, b). A remarkably consistent and coherent
picture emerged, where GC down-regulated positive
acting elements of the MHC-II expression and all
functional units necessary for optimal antigen recognition. Therefore, these data support a role of GC in
inhibiting adaptive immunity.

DISCUSSION
The ubiquitous distribution of the glucocorticoid receptor renders cell differentiation crucial in discriminating the effects of GC on different physiological
systems. Our analysis was carried out on human PBMC
and represents a large panel of a specific set of genes
that participate in the immune and inflammatory response. This panoramic study of gene regulation reveals that GC regulate numerous newly discovered
genes and control the expression of gene families, such
as SRs, TLRs, TSPs, complement, chemokine, TNF,
TGF-, and MHC families. It also elucidates new mechanisms of GC action by unveiling unexpected regulated
genes or gene (super) families and shows how GC may
facilitate one biological signal while counteracting its
inhibitor. This is well documented with the selective
modulation of the complement cascade activation and
inhibition and the regulation of cytokines/chemokines
and their receptors. Finally, it is noteworthy that within
the same gene family, some genes are stimulated and
others repressed, increasing the complexity of GC
action. Potential sources of discordance between RNA
and protein levels include translational control and
altered protein stability, and protein analysis may uncover post-translational modifications that are not predictable at the RNA level. Therefore, we analyzed
protein expression by FACS analysis; for all membrane
receptors tested, a perfect correlation between RNA
and protein expression was found. Thus, these results
illustrate the broad and coherent regulation of the
immune system by GC.
Glucocorticoids clearly have immunosuppressive actions. In fact, anti-inflammatory genes such as IL-10,
TGF-, and IL-1Ra, whose deletion in mice results in

increased susceptibility to inflammatory disease (36,


37), are strongly increased by GC. However, a more
comprehensive genetic dissection of the inflammatory
and immune response shows a more selective and
complex picture of the actions of GC. First, at the gene
cluster level, our study shows a bidirectional action of
GC, which are both immunostimulatory and immunosuppressive at the same time even for the inflammation
cluster (12). They seemed to prime and enhance the
innate immune response while repressing part of the
adaptive immune response in a resting state. This
suggests that GC help clear antigens by stimulating cell
trafficking, scavenger systems, and matrix metalloproteinases while they halt cellular immune responses by
inhibiting antigen presentation and T cell activation.
Second, at the single gene level, we found an opposite
regulation of molecules implicated in the inflammatory
and immune responses by GC, depending on the state
of activation of the cells. This effect may underline that
the signaling through the glucocorticoid receptor system may be mediated by a different set of transcription
factors, coactivators, and corepressors, allowing a switch
in promoter activity depending on the state of cellular
activation (38). Therefore, GC may drive the inflammatory and immune response in a context-dependent
fashion, which could be related to the state of disease
(13, 39). Indeed, the complete understanding of the
complex GC-mediated gene regulation may shed light
on the variable responses to GC, in patients with
GC-sensitive or -resistant inflammatory and autoimmune diseases (33, 40).
Glucocorticoids increase susceptibility to intracellular and opportunistic infections, which appears to
result from the profound immune suppression of antigen cell presentation and adaptive immune response
induced by these hormones. At the same time, GC
appear highly beneficial in the presence of an overwhelming systemic inflammation when administered in
a sustained fashion throughout the course of the disease. This is observed in sepsis, septic shock, ARDS,
and, less frequently, in meningococcal meningitis (6, 9,
41, 42). However, their delayed and time-limited use
can be devoid of beneficial effects and even disastrous
(10). This is clearly documented by the enhanced
TLR-4 expression by GC, which may be progressively
lost in the later phase of the disease, reflecting the clear
permissive action of GC early on the innate immune
response. Depending on the state of activation of the
cells, the differential GC actions underscore that the
timing of GC administration is crucial and may explain
some of the harmful effects of GC.
Therefore, clinical experience, together with a global
molecular approach of gene regulation, may reveal the
efficient timing of GC treatment and will allow a better
selection of patient groups that are likely to produce
the most benefit. Although it is possible that peripheral
blood mononuclear cells from patients with different
diseases may respond differently from healthy donors,
microarray screening techniques in combination with
quantitative RNA and protein assays will enable testing

GENE EXPRESSION PROFILES OF GLUCOCORTICOIDS ON IMMUNE CELLS

69

the profile of new steroids in the search for discriminating between desired and adverse effects. Thus,
synthetic steroids could be designed to act more specifically on certain components of the immune response, selectively enhancing or suppressing such a
response.

18.

REFERENCES
1.
2.
3.
4.

5.

6.

7.

8.

9.
10.

11.
12.
13.

14.

15.

16.

70

Chrousos, G. P. (1995) The hypothalamic-pituitary-adrenal axis


and immune-mediated inflammation. N. Engl. J. Med. 332,
13511362
Wilder, R. L. (1995) Neuroendocrine-immune system interactions and autoimmunity. Annu. Rev. Immunol. 13, 307338
Dhabhar, F. S., and McEwen, B. S. (1999) Enhancing versus
suppressive effects of stress hormones on skin immune function.
Proc. Natl. Acad. Sci. USA 96, 1059 1064
Sapolsky, R. M., Romero, L. M., and Munck, A. U. (2000) How
do glucocorticoids influence stress responses? Integrating permissive, suppressive, stimulatory, and preparative actions. Endocr. Rev. 21, 55 89
Boumpas, D. T., Chrousos, G. P., Wilder, R. L., Cupps, T. R., and
Balow, J. E. (1993) Glucocorticoid therapy for immune-mediated diseases: basic and clinical correlates. Ann. Intern. Med. 119,
1198 1208
Bollaert, P. E., Charpentier, C., Levy, B., Debouverie, M.,
Audibert, G., and Larcan, A. (1998) Reversal of late septic shock
with supraphysiologic doses of hydrocortisone. Crit. Care Med.
26, 645 650
Bronicki, R. A., Backer, C. L., Baden, H. P., Mavroudis, C.,
Crawford, S. E., and Green, T. P. (2000) Dexamethasone
reduces the inflammatory response to cardiopulmonary bypass
in children. Ann. Thorac. Surg. 69, 1490 1495
Meduri, G. U., Headley, A. S., Golden, E., Carson, S. J.,
Umberger, R. A., Kelso, T., and Tolley, E. A. (1998) Effect of
prolonged methylprednisolone therapy in unresolving acute
respiratory distress syndrome: a randomized controlled trial.
J. Am. Med. Assoc. 280, 159 165
Meduri, G. U., and Kanangat, S. (1998) Glucocorticoid treatment of sepsis and acute respiratory distress syndrome: time for
a critical reappraisal. Crit. Care Med. 26, 630 633
Cronin, L., Cook, D. J., Carlet, J., Heyland, D. K., King, D.,
Lansang, M. A., and Fisher, C. J. (1995) Corticosteroid treatment for sepsis: a critical appraisal and meta-analysis of the
literature. Crit. Care Med. 23, 1430 1439
Ashwell, J. D., Lu, F. W., and Vacchio, M. S. (2000) Glucocorticoids in T cell development and function. Annu. Rev. Immunol.18, 309 345
Wilckens, T., and De Rijk, R. (1997) Glucocorticoids and
immune function: unknown dimensions and new frontiers.
Immunol. Today 18, 418 424
Franchimont, D., Martens, H., Hagelstein, M. T., Louis, E.,
Dewe, W., Chrousos, G. P., Belaiche, J., and Geenen, V. (1999)
Tumor necrosis factor alpha decreases, and interleukin-10 increases, the sensitivity of human monocytes to dexamethasone:
potential regulation of the glucocorticoid receptor. J. Clin.
Endocrinol. Metab. 84, 2834 2839
McDermott, M. F., Aksentijevich, I., Galon, J., McDermott,
E. M., Ogunkolade, B. W., Centola, M., Mansfield, E., Gadina,
M., Karenko, L., Pettersson, T., McCarthy, J., Frucht, D. M.,
Aringer, M., Torosyan, Y., Teppo, A. M., Wilson, M., Karaarslan,
H. M., Wan, Y., Todd, I., Wood, G., Schlimgen, R., Kumarajeewa, T. R., Cooper, S. M., Vella, J. P., Kastner, D. L., et al.
(1999) Germline mutations in the extracellular domains of the
55 kDa TNF receptor, TNFR1, define a family of dominantly
inherited autoinflammatory syndromes. Cell 97, 133144
Jahnsen, F. L., Haye, R., Gran, E., Brandtzaeg, P., and Johansen,
F. E. (1999) Glucocorticosteroids inhibit mRNA expression for
eotaxin, eotaxin-2, and monocyte-chemotactic protein-4 in human airway inflammation with eosinophilia. J. Immunol. 163,
15451551
Pype, J. L., Dupont, L. J., Menten, P., Van Coillie, E., Opdenakker, G., Van Damme, J., Chung, K. F., Demedts, M. G., and

Vol. 16

January 2002

17.

19.
20.
21.

22.
23.

24.

25.
26.
27.
28.

29.

30.

31.

32.

33.

34.
35.
36.

Verleden, G. M. (1999) Expression of monocyte chemotactic


protein (MCP)-1, MCP-2, and MCP-3 by human airway smoothmuscle cells. Modulation by corticosteroids and T-helper 2
cytokines. Am. J. Resp. Cell. Mol. Biol. 21, 528 536
Re, F., Muzio, M., De Rossi, M., Polentarutti, N., Giri, J. G.,
Mantovani, A., and Colotta, F. (1994) The type II receptor as
a decoy target for interleukin 1 in polymorphonuclear leukocytes: characterization of induction by dexamethasone and
ligand binding properties of the released decoy receptor. J. Exp.
Med. 179, 739 743
Arzt, E., Sauer, J., Pollmacher, T., Labeur, M., Holsboer, F.,
Reul, J. M., and Stalla, G. K. (1994) Glucocorticoids suppress
interleukin-1 receptor antagonist synthesis following induction
by endotoxin. Endocrinology 134, 672 677
Carroll, M. C. (1998) The role of complement and complement
receptors in induction and regulation of immunity. Annu. Rev.
Immunol.16, 545568
Opdenakker, G., and Van Damme, J. (1992) Cytokines and
proteases in invasive processes: molecular similarities between
inflammation and cancer. Cytokine 4, 251258
Kuno, K., Kanada, N., Nakashima, E., Fujiki, F., Ichimura, F.,
and Matsushima, K. (1997) Molecular cloning of a gene encoding a new type of metalloproteinase-disintegrin family protein
with thrombospondin motifs as an inflammation associated
gene. J. Biol. Chem. 272, 556 562
Chen, H., Herndon, M. E., and Lawler, J. (2000) The cell
biology of thrombospondin-1. Matrix Biol. 19, 597 614
Hogger, P., Dreier, J., Droste, A., Buck, F., and Sorg, C. (1998)
Identification of the integral membrane protein RM3/1 on
human monocytes as a glucocorticoid-inducible member of the
scavenger receptor cysteine-rich family (CD163). J. Immunol.
161, 18831890
Buechler, C., Ritter, M., Orso, E., Langmann, T., Klucken, J.,
and Schmitz, G. (2000) Regulation of scavenger receptor CD163
expression in human monocytes and macrophages by pro- and
antiinflammatory stimuli. J. Leukoc. Biol. 67, 97103
Aderem, A., and Ulevitch, R. J. (2000) Toll-like receptors in the
induction of the innate immune response. Nature (London) 406,
782787
Krutzik, S. R., Sieling, P. A., and Modlin, R. L. (2001) The role
of Toll-like receptors in host defense against microbial infection. Curr. Opin. Immunol. 13, 104 108
Mellor, A. L., and Munn, D. H. (1999) Tryptophan catabolism
and T-cell tolerance: immunosuppression by starvation? Immunol. Today 20, 469 473
Mellor, A. L., Sivakumar, J., Chandler, P., Smith, K., Molina, H.,
Mao, D., and Munn, D. H. (2001) Prevention of T cell-driven
complement activation and inflammation by tryptophan catabolism during pregnancy. Nat. Immunol. 2, 64 68
Franchimont, D., Galon, J., Gadina, M., Visconti, R., Zhou, Y.,
Aringer, M., Frucht, D. M., Chrousos, G. P., and OShea, J. J.
(2000) Inhibition of Th1 immune response by glucocorticoids:
dexamethasone selectively inhibits IL-12-induced Stat4 phosphorylation in T lymphocytes. J. Immunol. 164, 1768 1774
Blotta, M. H., DeKruyff, R. H., and Umetsu, D. T. (1997)
Corticosteroids inhibit IL-12 production in human monocytes
and enhance their capacity to induce IL-4 synthesis in CD4
lymphocytes. J. Immunol. 158, 5589 5595
Wu, C. Y., Sarfati, M., Heusser, C., Fournier, S., Rubio-Trujillo,
M., Peleman, R., and Delespesse, G. (1991) Glucocorticoids
increase the synthesis of immunoglobulin E by interleukin
4-stimulated human lymphocytes. J. Clin. Invest. 87, 870 877
Dozmorov, I. M., and Miller, R. A. (1998) Generation of
antigen-specific Th2 cells from unprimed mice in vitro: effects
of dexamethasone and anti-IL-10 antibody. J. Immunol. 160,
2700 2705
Lane, S. J., Adcock, I. M., Richards, D., Hawrylowicz, C., Barnes,
P. J., and Lee, T. H. (1998) Corticosteroid-resistant bronchial
asthma is associated with increased c-fos expression in monocytes and T lymphocytes. J. Clin. Invest. 102, 2156 2164
McDevitt, H. O. (2000) Discovering the role of the major
histocompatibility complex in the immune response. Annu. Rev.
Immunol.18, 117
Alfonso, C., and Karlsson, L. (2000) Nonclassical MHC class II
molecules. Annu. Rev. Immunol.18, 113142
Shull, M. M., Ormsby, I., Kier, A. B., Pawlowski, S., Diebold, R. J.,
Yin, M., Allen, R., Sidman, C., Proetzel, G., Calvin, D., et al.

The FASEB Journal

GALON ET AL.

37.
38.
39.

40.

(1992) Targeted disruption of the mouse transforming growth


factor-beta 1 gene results in multifocal inflammatory disease.
Nature (London) 359, 693 699
Kuhn, R., Lohler, J., Rennick, D., Rajewsky, K., and Muller, W.
(1993) Interleukin-10-deficient mice develop chronic enterocolitis. Cell 75, 263274
McKenna, N. J., Lanz, R. B., and OMalley, B. W. (1999) Nuclear
receptor coregulators: cellular and molecular biology. Endocr.
Rev. 20, 321344
Leung, D. Y., Martin, R. J., Szefler, S. J., Sher, E. R., Ying, S., Kay,
A. B., and Hamid, Q. (1995) Dysregulation of interleukin 4,
interleukin 5, and interferon gamma gene expression in steroidresistant asthma. J. Exp. Med. 181, 33 40
Bamberger, C. M., Schulte, H. M., and Chrousos, G. P. (1996)
Molecular determinants of glucocorticoid receptor function

and tissue sensitivity to glucocorticoids. Endocr. Rev. 17,


245261
41. McGowan, J. E., Chesney, P. J., Crossley, K. B., and LaForce,
F. M. (1992) Guidelines for the use of systemic glucocorticosteroids in the management of selected infections. Working Group
on Steroid Use. Antimicrobial Agents Committee, Infectious
Diseases Society of America. J. Infect. Dis. 165, 113
42. Odio, C. M., Faingezicht, I., Paris, M., Nassar, M., Baltodano, A.,
Rogers, J., Saez-Llorens, X., Olsen, K. D., and McCracken, G. H.
(1991) The beneficial effects of early dexamethasone administration in infants and children with bacterial meningitis. N. Engl.
J. Med. 324, 15251531

GENE EXPRESSION PROFILES OF GLUCOCORTICOIDS ON IMMUNE CELLS

Received for publication May 1, 2001.


Revised for publication August 15, 2001.

71