| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |





From the Laboratoire de Biologie Cellulaire*
and
the Laboratoire dImmunologie,
Faculté
de Médecine Cochin-Port-Royal, Université Paris; the
Pharmacie,
Hôpital Bicêtre; and
the Service de Chirurgie Générale et de Transplantation
Hépatique,§
Hôpital Cochin,
Paris, France
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
50% of patients experience at
least one episode of acute rejection.1,2
This can be
reversed in many cases by high doses of steroids, but 20 to 30% of
patients do not respond and need additional therapy.2-8 Interleukin-1 (IL-1) plays a key role in inflammatory and immune-mediated diseases.9 During allograft rejection, IL-1 production precedes allograft dysfunction and injury.10-12 Inhibition of IL-1 production is an important mechanism by which corticosteroids suppress immune response.13-16
A natural antagonist of the IL-1 receptor (IL-1Ra) has recently been identified.17-21 Soluble IL-1Ra can counteract the proinflammatory properties of IL-1 by competitively binding to the cell-surface receptor without inducing signal transduction.22-25 Its role as an anti-inflammatory protein has been reported both in vitro and in vivo.26-28 IL-1Ra production is also enhanced in stable human kidney graft recipients, and could thus be a key factor in early down-regulation of the allogeneic immune response.29
The mechanisms of steroid resistance after organ transplantation are unknown. We explored the possible roles of a reduced sensitivity of IL-1 production to corticosteroid therapy, or defective IL-1Ra production. Our results indicate that both IL-1 production and its inhibition by corticosteroids are normal in patients whose acute rejection is steroid-resistant whereas, in contrast, IL-1Ra secretion is consistently reduced, pointing to a constitutional defect of IL-1Ra production in such patients.
| Patients and Methods |
|---|
|
|
|---|
The study population consisted of adult liver transplant recipients. After liver transplantation, patients were treated initially with corticosteroids, azathioprine, and cyclosporine. The dose of cyclosporine was adjusted to maintain whole-blood levels between 150 and 250 µg/L. Azathioprine was discontinued after 3 months, and steroids were tapered to 5 to 10 mg/day.
Acute rejection was defined by both biochemical and histological criteria, including at least two of the following: portal inflammatory infiltrate, endothelial inflammation, and biliary damage. Rejection was considered resistant when liver function tests improved by <50% of the highest values after three steroid boluses, in addition to persistent histological features of acute rejection on control liver biopsy. Episodes of acute rejection were initially treated for 3 days with a 1-g bolus of intravenous methylprednisolone daily. Cases of steroid-resistant acute rejection were treated with the monoclonal antibody OKT3 (Ortho Pharmaceutical Corp., Raritan, NJ).
Blood Sampling and Liver Biopsy
Blood and liver biopsy specimens were obtained, with their
informed consent, from 29 patients with suspected rejection (Table 1)
. Histological confirmation of acute
rejection was obtained in 21 of these cases. Nine episodes responded to
steroids whereas 12 resisted. The remaining eight patients had no
histological evidence of acute rejection, either initially or during
follow-up. Transplant recipients without acute rejection were assessed
at approximately day 15, so as to be as close as possible to the median
time of acute rejection in the two other groups. A portion of each
biopsy specimen was immediately fixed and embedded in paraffin.
Sections were stained with hematoxylin-eosin-safran for histological
diagnosis. The remaining portion was snap-frozen in liquid nitrogen and
stored at -80°C for immunohistochemistry. Blood and liver samples
were obtained on days 14.5 ± 2.5, 15.1 ± 3.1, and 14.7
± 2.8, respectively, from patients without rejection and patients with
steroid-responsive and steroid-resistant rejection. In the latter two
subgroups, samples were obtained before anti-rejection therapy. Blood
and liver biopsy specimens were also obtained from the same patients,
with their informed consent, during a routine work-up, a mean of 2
years posttransplantation. At that time, the patients were clinically
stable and receiving minimal doses of steroids. Moreover, the total
doses of steroids received since liver transplantation, as well as the
cyclosporine dose and blood concentrations, were similar in the three
groups of patients (Table 1)
. Blood samples were also obtained, with
their informed consent, from 44 healthy volunteers.
|
Five-micrometer cryostat sections of each biopsy specimen were mounted on polysine-coated slides (CML-CBE, Nemours, France). Serial slides were air-dried, fixed in acetone, and stored at -20°C until staining. Cytokine expression was tested using an indirect immunoenzymatic method.30 Anti-IL-1ß monoclonal antibody (Genzyme, Cambridge, MA) and anti-IL-1Ra polyclonal antibody (Genzyme) were used at a dilution of 1:20. In positive controls, some blood mononuclear cells activated by lipopolysaccharide (LPS) (1 µg/ml) (Sigma Chemical Company, St Louis, MO) expressed IL-1ß and IL-1Ra, whereas no inactivated cells expressed IL-1ß and IL-1Ra (data not shown). Cryostat sections were refixed in acetone then incubated with Tris-buffered saline, 20% human AB serum (Sanofi Diagnostics Pasteur, Marnes La Coquette, France), then with the primary antibody. Slides treated with the polyclonal primary antibody (anti-IL-1Ra) were incubated with 1:20 mouse anti-rabbit immunoglobulin (DAKO, Dakopatts A/S, Glostrup, Denmark). All slides were then incubated with 1:20 rabbit anti-mouse immunoglobulin (DAKO) and subsequently with alkaline phosphatase-anti-alkaline phosphatase complexes (DAKO). Alkaline phosphatase activity was revealed after incubation in fast-red TR and napthol-phosphate solution containing levamisole.31 The slides were counterstained with hematoxylin. In negative controls, the primary antibody was replaced by an irrelevant antibody at the same dilution. Anti-CD3 (DAKO) was used as a positive control.
Quantification of Immunohistochemical Staining
The slides were examined by two of the authors (FC and SB), who were blinded to the pathological diagnoses. Positive cells were counted on each slide in a portal and a lobular field at a magnification of x200. The number of positive cells recorded was a consensus between the two observers.
Assay of IL-1ß and IL-1Ra by Enzyme-Linked Immunosorbent Assay (ELISA) in Cultures of Blood Cells
Five milliliters of heparinized blood were obtained from each patient, diluted 1:5 in RPMI 1640 medium containing L-glutamine and penicillin/streptomycin (Gibco-BRL, Cergy-Pontoise, France), then cultured for 24 hours with or without LPS (1 µg/ml),21 with various concentrations of dexamethasone (0 to 10-5 mol/L). IL-1ß and IL-1Ra were measured in duplicate in 24-hour supernatants of blood cells, using an ELISA kit (R&D System Europe Ltd., Abingdon, UK). The detection limits were 4 pg/ml IL-1ß and 31 pg/ml IL-1Ra, and the within-run variation was <5% in each case.
Analysis of IL-1Ra and IL-1ß Gene Polymorphisms
Twenty-four patients (nine with steroid-responsive, nine with steroid-resistant acute rejection, and six with no rejection) were investigated for IL-1Ra and IL-1ß gene polymorphisms. Genomic DNA was isolated from blood by the salting-out method.32 IL-1Ra exon 2 polymorphism was analyzed as previously described.33 Oligonucleotides 5'-CTCAGCAACACTCCTAT-3' and 5'-TCCTGGTCTGCAGGTAA-3' were used as primers.34 The polymerase chain reaction products were separated by electrophoresis on 2% agarose gel and stained with ethidium bromide. Allele 1 (four repeats) was 410 bp long and allele 2 (two repeats) was 240 bp long. The region containing the AvaI polymorphic site at position -511 of the IL-1ß gene was amplified by polymerase chain reaction. The oligonucleotides 5'-TGGCAATTGATCTGGTTCATC-3' and 5'-GTTTAGGAATCTTCCCACTT-3' flanking this region were used as primers.35 Polymerase chain reaction products were digested with 6 U of AvaI at 37°C for 3 hours. Fragments were analyzed on 2% agarose gel. This revealed products of 190 bp + 114 bp (allele 1) and 304 bp (allele 2). Position +3953 within exon 5 of the IL-1ß gene exhibited single bp polymorphism.36 The polymorphic region containing the TaqI restriction site was amplified using the following primers: 5'-GTTGTCATCAGACTTTGACC-3' and 5'-TTCAGTTCATATGGACCAGA-3'. TaqI digestion of the 249-bp fragments resulted in products that either remained intact (allele 2) or were cut into two fragments of 135 bp and 114 bp (allele 1).
Statistical Analysis
The immunohistochemical and ELISA results are expressed as means ± SE. Statview IV software (Abacus Concepts, Berkeley, CA) running on a Macintosh computer was used for statistical analyses (analysis of variance followed by Fishers exact test).
| Results |
|---|
|
|
|---|
Positive cells were rare in liver grafts without rejection,
whereas they were abundant during acute rejection (P <
0.03 in the portal tract and P < 0.01 in the lobule),
but there was no difference between patients with steroid-responsive
and steroid-resistant acute rejection (Figure 1a)
. Stained cells were mostly
endothelial, Kupffer, and inflammatory cells, infiltrating the
sinusoids and portal tracts.
|
Positive cells were rare in liver transplants without acute
rejection (Figure 2A)
and abundant in
both steroid-responsive (Figure 2B)
and steroid-resistant acute
rejection (Figure 2C)
. The number of cells expressing IL-1Ra was higher
in patients with acute rejection than without acute rejection
(P < 0.05 in the portal tract and P <
0.02 in the lobule), but there was no difference in this respect
between steroid-responsive and -resistant acute rejection (Figure 1b)
.
Stained cells were Kupffer and inflammatory cells infiltrating the
sinusoids and portal tracts.
|
Few hepatic cells expressed IL-1ß and IL-1Ra, with no significant difference between the three groups of patients (data not shown).
ELISA IL-1ß Levels at the Time of Rejection
IL-1ß concentrations were low in the supernatants of
unstimulated blood cells from patients without acute rejection, with
steroid-responsive and with steroid-resistant acute rejection (46
± 29 pg/ml, 46 ± 29 pg/ml, and 64 ± 46 pg/ml,
respectively). After LPS stimulation, the corresponding IL-1ß
concentrations rose to 1,634 ± 724 pg/ml, 1,027 ± 612
pg/ml, and 1,959 ± 264 pg/ml (NS). IL-1ß production was
inhibited by dexamethasone in a concentration-dependent manner and to a
similar extent in each group (Figure 3a)
.
|
Supernatants of unstimulated cells contained 1,824 ± 394
pg/ml, 1,002 ± 583 pg/ml, and 1,422 ± 417 pg/ml IL-1ß,
respectively, in patients without rejection and those with
steroid-responsive and -resistant acute rejection (NS). After LPS
stimulation, the corresponding IL-1ß concentrations were 6,728
± 1,497 pg/ml, 4,155 ± 1,231 pg/ml, and 5,149 ± 1,098
pg/ml (NS). IL-1ß production was inhibited by dexamethasone to a
similar extent in each group (Figure 3b)
.
ELISA IL-1Ra Levels at the Time of Rejection
IL-1Ra concentrations were 565 ± 275 pg/ml, 2,012 ±
1,486 pg/ml, and 1,034 ± 496 pg/ml, respectively, in the
supernatants of unstimulated cells from patients without rejection and
those with steroid-responsive or steroid-resistant rejection. After
stimulation, the corresponding IL-1Ra concentrations rose to 6,461
± 1,386 pg/ml, 9,010 ± 2,119 pg/ml, and 3,129 ± 349 pg/ml
(P < 0.01 between the steroid-responsive and
steroid-resistant acute rejection groups). Dexamethasone inhibited
LPS-stimulated IL-1Ra production to a similar extent in the three
groups (Figure 3c)
.
ELISA IL-1Ra Levels after 2 Years
IL-1Ra concentrations were 3,573 ± 573 pg/ml, 3,156 ±
798 pg/ml, and 1,934 ± 250 pg/ml in the supernatants of
unstimulated cells from patients without rejection and those with
steroid-responsive and -resistant acute rejection, respectively. After
stimulation, the corresponding IL-1Ra concentrations rose to 6,331
± 1,210 pg/ml, 7,269 ± 1,301 pg/ml, and 3,661 ± 366 pg/ml
(Figure 3d)
. IL-1Ra secretion was inhibited by corticosteroids to a
similar extent in the three groups of patients. IL-1Ra secretion was
lower in patients with a history of steroid-resistant rejection than in
the other two groups, both in the absence and presence of dexamethasone
(P < 0.01). In contrast, IL-1Ra secretion did
not significantly differ between patients with a past history of acute
rejection (5,208 ± 701 pg/ml) and those with no such history
(6,331 ± 121 pg/ml). Taking a cut-off value of 5,750 pg/ml, eight
of the nine patients with steroid-responsive acute rejection exhibited
IL-1Ra levels above this value, whereas all 12 patients with resistant
acute rejection had values below the cut-off (Figure 4)
.
|
The distribution of IL-1ß and IL-1Ra genotypes and corresponding
allelic frequencies were similar to those previously reported (Table 2)
. None of the alleles examined was
associated with the resistance of acute rejection to steroid therapy in
the patients tested. In addition, the association of IL-1Ra allele 2
with IL-1ß -511 allele 2 or IL-1ß +3,953 allele 2 had no effect on
steroid resistance (Table 3)
.
|
|
To determine whether the low IL-1Ra levels observed in transplant
patients with steroid resistance were also found in the general
population, IL-1Ra production was measured in blood samples from 44
healthy volunteers. Large interindividual differences were found in
IL-1Ra secretion (Figure 4)
. Low IL-1Ra production (<5,750 pg/ml) by
LPS-stimulated mononuclear cells, similar to that found in the
steroid-resistant acute rejection group, was observed in 34% of
healthy volunteers.
| Discussion |
|---|
|
|
|---|
In all of the patients with acute rejection, whatever their response to steroid therapy, a high level of IL-1ß production was found by immunohistochemistry on liver biopsies and by ELISA in the supernatants of stimulated blood cells. These data are in agreement with those previously obtained by reverse transcriptase-polymerase chain reaction in the same clinical setting.11,12 Because steroid-resistant and steroid-responsive acute rejection did not differ in terms of IL-1ß production by blood mononuclear cells, a defect in IL-1ß sensitivity to steroid therapy is probably not involved in steroid resistance, although a lower sensitivity of hepatocytes and sinusoidal endothelial cells to steroids could not be excluded.
IL-1Ra expression in the liver grafts was also increased in patients with acute rejection relative to patients without rejection. To our knowledge, this is the first clear evidence that hepatic IL-1Ra production is enhanced during acute liver graft rejection. This IL-1Ra overexpression may result from an increased stimulation through the local production of cytokines, and may restrict the rejection process by inhibiting the effects of IL-1.37,38 In those patients, IL-1Ra was only expressed by the inflammatory cells infiltrating the sinusoids and portal tracts. In our study, hepatocytes, which may secrete IL-1Ra, especially when stimulated by IL-1 and IL-6,37 were consistently found to be negative for IL-1Ra expression by immunohistochemistry. It is noteworthy that the number of the infiltrating cells expressing IL-1Ra was similar during steroid-responsive and -resistant episodes of acute rejection. In contrast, IL-1Ra production by blood cells was lower in steroid-resistant than in steroid-sensitive patients, both at the time of acute rejection and throughout the long term.
To determine whether this difference reflected the existence of high
and low IL-1Ra producers within the general population, we compared
LPS-stimulated IL-1Ra production by blood cells from the transplant
patients with that of healthy volunteers, and found the same broad
range of responses in both groups. All steroid-resistant patients
exhibited IL-1Ra levels <5,750 pg/ml, as did
30% of the healthy
volunteers, who may therefore be at risk of steroid resistance. It
remains to be determined if this distinction between weak and strong
IL-1Ra producers could be used to predict an increased risk of steroid
resistance in other transplant settings, as well as in other
inflammatory or autoimmune diseases.
We then investigated whether the variability in IL-1Ra production could be explained by the recently described IL-1ß and IL-1Ra gene polymorphisms. Indeed, certain IL-1ß or IL-1Ra alleles are associated with severe forms of inflammatory or autoimmune diseases.35,36,39-43 In addition, allele 2 of the IL-1Ra gene is associated with higher concentrations of the molecule in plasma.34 We found no link between the known IL-1ß and IL-1Ra polymorphisms and steroid resistance. The number of patients tested was however small, and this result needs to be confirmed. Other polymorphisms of IL-1Ra gene should also be investigated.
The mechanisms of steroid resistance are still poorly understood. In steroid-resistant asthma, neither altered steroid metabolism nor an abnormality of the hypothalamic-pituitary-adrenal axis seems to be involved.44-47 Impaired liver graft function might result in reduced conversion of methylprednisolone to its active form, without a corresponding reduction in its rate of elimination.48 However, we found no difference in the conversion rate of prednisolone to its active form among our patients (V. Furlan, manuscript in preparation). A relative insensitivity of T lymphocytes to corticosteroids has also been suspected.46,47,49,50 The present study suggests that the clinical resistance of acute liver rejection to steroids is at least partly due to an IL-1Ra defect. Clinical steroid resistance was probably not related to a stronger inhibition of IL-1Ra production by steroids in the clinically resistant group, as the inhibition curves ran parallel in all groups. The use of the term "steroid resistance" in the transplant setting may thus be misplaced, as it does not correspond to true steroid resistance at the cellular level. IL-1 production is incompletely inhibited by corticosteroids during acute rejection, as substantial amounts of IL-1 can be found in the graft. IL-1Ra may prevent residual IL-1 from binding to its receptor. The inability of some individuals to mount an adequate IL-1Ra response may result in the so-called steroid resistance. Clinical resistance to steroids is probably potentiated by the inhibitory effect of steroids on IL-1Ra production.
This study defines a new concept in the pathophysiology of steroid resistance and suggests IL-1Ra as a potential treatment. Indeed, IL-1Ra could be used to treat steroid-resistant acute rejection in humans, as in experimental models.51-53 The use of IL-1Ra to treat rheumatoid arthritis and graft-versus-host disease has proved to be safe and effective.54-56 Moreover, constitutively low IL-1Ra production can be detected before transplantation, so that IL-1Ra could be included in the immunosuppressive regimen of such patients.
| Acknowledgements |
|---|
| Footnotes |
|---|
Supported by a grant from Novartis Pharma France. F. C. is the recipient of a fellowship from the Association Française pour lEtude du Foie.
Accepted for publication July 13, 2000.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. Xu, R. Lucas, M. Schuchmann, S. Kuhnle, T. Meergans, A. P. Barreiros, A. W. Lohse, G. Otto, and A. Wendel GM-CSF Restores Innate, But Not Adaptive, Immune Responses in Glucocorticoid-Immunosuppressed Human Blood In Vitro J. Immunol., July 15, 2003; 171(2): 938 - 947. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |