Multiple Levels of Regulation of the Interleukin-6 System in Stroke
Background and Purpose— Serum levels of the cytokine interleukin-6 (IL-6) rise markedly in stroke. IL-6 is a key regulator of inflammatory mechanisms that play an important part in stroke pathophysiology. The action of IL-6 is modified by its soluble receptor subunits sgp130 and sIL-6R. The purpose of this study was to investigate whether serum levels of the receptor subunits are changed after ischemic stroke and to define the role of genetic influences on IL-6 expression in acute stroke.
Methods— In 48 patients with acute stroke and 48 age- and sex-matched control subjects, serum concentrations of IL-6, sgp130, and sIL-6R were measured by enzyme-linked immunosorbent assay. Furthermore, IL-6 promoter haplotypes comprising 4 different polymorphisms (−597G→A, −572G→C, −373A(n)T(n), −174G→C) were determined by DNA sequencing and allele-specific oligonucleotide polymerase chain reaction. The effect of the common haplotypes on IL-6 gene transcription was tested by transfecting reporter fusion genes in the astrocytelike cell line U373.
Results— Whereas serum concentrations of IL-6 significantly rose (P<0.001), sgp130 levels were transiently reduced after stroke (P<0.05), and sIL-6R levels remained unchanged. IL-6 levels depended on the infarct size and the haplotype of the promoter region. The common haplotype A-G-8/12-C was associated with low IL-6 levels after stroke and a reduced induction of IL-6 transcription on stimulation with an adenosine analog in vitro.
Conclusions— The data demonstrate genetic variation in the expression of IL-6 in stroke. Induction of the inflammatory response by IL-6 might be enhanced by a transient downregulation of the potential IL-6 antagonist sgp130.
Stroke triggers an acute-phase response. Elevated serum levels of acute-phase proteins such as C-reactive protein (CRP) or fibrinogen are found in about three quarters of patients with ischemic stroke.1,2 Up to 60% of patients have a core temperature >37.5°C in the first week after symptom onset.3,4 Fever and elevated serum concentrations of CRP or fibrinogen have been linked to poor prognosis.1–3 In infection, the cytokine interleukin-6 (IL-6) is a key regulator of the acute-phase response. IL-6 elevates the temperature set point in the hypothalamus and stimulates the expression of CRP, fibrinogen, and other acute-phase proteins in the liver. Infections secondary to stroke may contribute to the observed activation of the inflammatory response in stroke patients; however, experimental studies support the notion that brain ischemia also triggers an inflammatory response. Cerebral ischemia induces the expression of IL-6 in neurons and astrocytes.5–7 In patients, IL-6 serum concentrations rise within several days. The ischemic brain appears to be a major
See Editorial Comment, page 1869
source of IL-6 in stroke because the serum concentrations correlate with infarct size and because cerebrospinal fluid concentrations exceed the serum concentrations of IL-6.8 The correlations between IL-6 and CRP,9 IL-6 and fibrinogen, or IL-6 and body temperature10 argue for a role of IL-6 in the regulation of the inflammatory response in stroke. However, the fact that no acute-phase response is found in about one quarter of patients shows that the regulation is more complex.1,2
The expression of IL-6 is regulated mainly at the transcriptional level. The promoter of the human IL-6 gene contains several polymorphisms that influence IL-6 gene transcription in selected cell types.11 The −174 G/C polymorphism has been linked to several diseases, including Alzheimer and lacunar infarcts.12,13 However, the effect of these polymorphisms on IL-6 induction in neural cells is unknown.
Extracellular IL-6 exerts its effects through a membrane receptor that consists of the 2 subunits gp130 and IL-6R. Soluble forms of both subunits circulate in blood. IL-6R is missing in many cell types, but sIL-6R can substitute for the resident membrane receptor and thereby acts as an agonist.14 In contrast, sgp130 is a decoy receptor that exerts an antagonistic effect in vitro.15 The serum concentration of sIL-6R has been reported to be elevated in several inflammatory and atherosclerotic diseases,16 but to the best of our knowledge, neither sIL-6R nor sgp130 has been studied in acute stroke. Therefore, we have investigated the serum concentrations of the IL-6 receptors and the effect of polymorphisms in the IL-6 promoter on IL-6 expression in acute stroke.
Subjects and Methods
In this study, 48 patients treated for acute ischemic stroke in our clinic were included prospectively over the course of 7 months. All patients were admitted within 24 hours after onset of symptoms. Patients were not included in the study if symptoms lasted <24 hours, if they had a history of a malignant tumor or stroke within the last year, if they were currently receiving anti-inflammatory or immunosuppressive medication, or if they presented with signs or symptoms of acute or chronic infection.17 Patients were excluded from the study if they developed signs of infection (4 patients), if intensive care therapy became necessary (4 patients), or if they were discharged before day 3 or 7 (17 patients). The control group was recruited simultaneously. Control subjects were sex and age matched and were electively admitted to the Department of Ophthalmology for cataract surgery or glaucoma therapy. During hospitalization, stroke patients were treated with anticoagulants (heparin or phenprocoumon, 17 patients), antiplatelet drugs (20 patients), statins (18 patients), or antihypertensive agents (31 patients). All subjects gave informed consent. The study was approved by the local ethics committee.
Imaging and Clinical Examinations
The neurological status of stroke patients was assessed by the National Institutes of Health Stroke Scale (NIHSS) on days 1 and 7 and later than day 90 after symptom onset. Neurological improvement or deterioration was defined as a change of NIHSS of at least 4 between days 7 and 90 or later compared with day 1. The origin of stroke was classified according to the Trial of Org 10172 in Acute Stroke Treatment (TOAST) criteria.18 The aural temperature of patients was measured on days 1, 3, and 7 between 7 and 8 am. In control subjects, temperature was measured on admission. We measured the infarct volume on CT or MRI scans performed between days 1 and 7. Stroke volume was assessed as described elsewhere.19 Ultrasonographic measurement of the intima-media thickness (IMT) of the common carotid artery is also described elsewhere.20
In stroke patients, blood samples were taken between 7 and 8 am on day 1 (<24 hour after onset of symptoms) or day 2 (between 24 and 48 hours after onset), day 3, day 7, and after 90 days. In the control group, blood was taken on admission before surgery. CRP was determined by an ultrasensitive turbidimetric test that uses latex beads (Quantex CRP Ultrasensitiv, Instrumentation Laboratory). The threshold of detectability was 1.0 mg/L. Fibrinogen was measured by functional coagulation testing (derived fibrinogen) with Recombiplastin (Instrumentation Laboratory) as reagent. For cytokine determination, samples were immediately put on ice and centrifuged within 30 minutes. Serum was then stored at −80°C until the enzyme-linked immunosorbent assays (ELISAs) were performed. For determination of IL-6, sIL-6R, and sgp130, commercial ELISA kits (R&D) were used.
The primers IL-6-00501 (5′-AGT GGG CTG AAG CAG GTG AAG AAA-3′) and IL-6-11028 (5′-CTG ATT GGA AAC CTT ATT AAG ATT GT-3′) were used for amplification of the IL-6 promoter region, which was sequenced with standard techniques. Eight different IL-6 promoter haplotypes have been described by Terry et al11 in whites (haplotypes A through G and X, with X corresponding to a deletion of a G residue immediately 5′ to the AT run in haplotype D). On the basis of this work, we interpreted the sequence data from the promoter region of 34 patients and 21 control subjects. If homozygosity was observed for all 4 polymorphic positions (−597G→A, −572G→C, −373A(n)T(n), −174G→C) or for 3 of them, the 2 haplotypes were unequivocally established. When ≥2 of the 4 promoter polymorphisms were found to be heterozygous, the combination of individual genotypes made it possible to deduce 2 haplotypes on the basis of the 8 haplotypes described by Terry et al.11 Some of these deduced haplotypes were confirmed by allele-specific oligonucleotide polymerase chain reaction (PCR) with the following primers: AAGTAACTGCACGAAATTTGAGGG, AAGTAACTGCACGAAATTTGAGGA, GCAATGTGACGTCCTTTAGCATC, and GCAATGTGACGTCCTTTAGCATG. Sequence analysis of the allele-specific PCR products consistently confirmed the deduced haplotypes on the basis of the 8 known, naturally occurring haplotypes.
Cell Culture and Transfection
The human astrocytoma cell line U373 was cultured in Earle’s modified Eagle medium (PAA) containing 10% fetal calf serum, penicillin (50 IU/mL), streptomycin (50 μg/mL), and 1% nonessential amino acids (GIBCO BRL). One day before transfection, cells were plated on a 24-well plate and transfected with DEAE dextran. Wells were washed with phosphate-buffered saline, and then 1 μg of the first reporter plus 0.1 μg pRL-SV40 as a second reporter (Promega) were added in Earle’s modified Eagle medium containing 75 μg/mL DEAE-dextran but no serum for 4 hours. Then cells were shocked by the addition of 10% dimethyl sulfoxide in phosphate-buffered saline for 1 minute. After incubation for 42 hours in full medium, cells were stimulated for 6 hours with 10 μmol/L 2-chloroadenosine (RBI, Biotrend). Cells were harvested, and the activities of firefly and renilla luciferases were measured with the Dual Luciferase Reporter Assay (Promega). The reporter fusion genes containing haplotypes of the human IL-6 promoter and the coding sequence of firefly luciferase have been described before.11
Correlation was quantified by use of Spearman’s correlation. Data are presented as median and percentiles (10th, 25th, 75th, and 90th in box plots) because not all parameters were normally distributed. We used the Mann-Whitney U test and the Wilcoxon test to analyze differences between groups and time points, respectively. Analysis of variance (ANOVA) was performed to test for differences between groups. Allele frequencies were compared by the exact χ2 test for small sample size. A linear regression model was used to adjust the log10-transformed IL-6 concentrations for log10-transformed infarct volume and to compare IL-6 and IL-6 receptor concentrations in patients and control subjects, with age, sex, and IMT of the common carotid artery as covariants.
Clinical and demographic data of our study population are summarized in Table 1. Although we did not include patients with clinical signs of infection, the acute-phase proteins fibrinogen (Figure 1A) and CRP (Figure 1B) were increased in stroke patients, as has been reported before.1 In our group of patients, 5 had a body temperature >37.4°C on day 3 compared with no control subjects (P<0.025); the average body temperature did not differ from that of the control group (Figure 1C). IL-6 serum concentrations were markedly elevated after stroke (P<0.001) (Figure 1D). In contrast, the serum concentration of sgp130 was significantly reduced the first week after stroke (P<0.05) but not at follow-up >3 months later (Figure 1E). IL-6 and sgp130 remained different after adjustment for age, sex, and IMT, a marker of atherosclerosis (IL-6, P<0.001; sgp130, P=0.003). Serum concentration of sIL-6R did not differ between control subjects and patients with acute stroke (Figure 1F). There were no significant correlations between IL-6 serum levels and its receptors in stroke patients.
The kinetics of the increase in IL-6 and CRP differed: IL-6 serum concentration had already increased in the first 24 hours after stroke and before the peak of the acute-phase protein CRP on day 3 (Figure 1B and 1D). On days 3 and 7, serum concentration of IL-6 closely correlated with CRP and fibrinogen concentrations (IL-6 day 1 and CRP day 3, r=0.35, P=0.03; IL-6 day 3 and CRP day 3, r=0.36, P=0.04; IL-6 day 7 and CRP day 7, r=0.6, P=0.005; IL-6 day 1 and fibrinogen day 3, r=0.35, P=0.04; IL-6 day 7 and fibrinogen day 7, r=0.5, P=0.02) but not with leukocyte count or body temperature (data not shown). CRP concentrations were significantly higher in patients on days 3 and 7 compared with control subjects but not on day 1, possibly because of the limited sensitivity of the CRP assay (Figure 1B).
Serum concentration of IL-6 correlated with infarct volume (r=0.4, P=0.003) and clinical status of stroke patients as expressed by NIHSS on day 1, on day 7, and after day 90 (r=0.4, P=0.003; r=0.4, P=0.02; r=0.5, P=0.012, respectively). The concentrations of the 2 receptor subunits did not correlate with infarct size or NIHSS (data not shown), nor was there any obvious difference in infarct volume (1-way ANOVA, P=0.2) or CRP, IL-6, and IL-6 receptor concentrations in the etiological subgroups. We found no correlation of temperature, CRP, fibrinogen, IL-6, and IL-6 soluble receptors on day 1 and neurological deterioration or improvement.
To search for genetic influences on IL-6 expression, we determined 4 polymorphisms in the IL-6 promoter region in 34 stroke patients and 21 control subjects. The frequencies of the 8 haplotypes of IL-6 were comparable, as has been described previously.11 The frequency distribution between patients and control subjects did not differ in general (P=0.19), although an exploratory analysis for specific haplotypes indicated a different frequency of haplotype C between control subjects and patients (Table 2).
The haplotype found most frequently was F. IL-6 serum concentrations were significantly lower in patients with haplotype F than in patients with other haplotypes in a multivariate ANOVA adjusted for infarct size (P=0.019; Figure 2A). Control subjects and patients with haplotype F had lower body temperature and CRP values, which was significant in control subjects (Figure 2B and 2C).
In cerebral ischemia, astrocytes express IL-6,5 and adenosine is considered a major stimulus of IL-6 transcription.21 Therefore, we investigated the effect of the various haplotypes of the IL-6 promoter on the induction of IL-6 transcription in the human astrocytelike cell line U373. Basal transcription did not differ between the 8 common haplotypic variations of the IL-6 promoter (data not shown). Induction by the stable adenosine analog 2-chloroadenosine significantly depended on the haplotype of the IL-6 promoter, with haplotype F mediating the lowest induction of IL-6 transcription (Figure 3).
Activation of the inflammatory response plays a pivotal role in acute stroke. An increase in serum levels of CRP and fibrinogen and in body temperature predicts poor prognosis. Moreover, activation of the acute-phase response has been involved in the pathogenesis of ischemic stroke, suggesting that intervention might be beneficial. Therefore, it is important to understand the regulation of the acute-phase response in stroke.
Because stroke is often associated with infections,17 activation of the acute-phase response might be secondary to the accompanying infection rather than the cerebral ischemia itself. To minimize this confounding factor, we excluded patients with signs of infection. Still, there was a significant increase in CRP and fibrinogen levels and in body temperature in part of the study sample. Although clinical data allow no definite conclusions on the cause of this increase, experimental work has elucidated a mechanism through which cerebral ischemia is able to activate the acute-phase reaction. IL-6, a major regulator of the acute-phase response, is expressed in the ischemic brain and modulates body temperature.5–7 In accordance with this concept, IL-6 serum concentrations rapidly rise after stroke and correlate with infarct size8,9,22 (Figures 1D and 2⇑A). Further support for the alleged role of IL-6 as a key regulator of the acute-phase response in acute stroke comes from the fact that the elevation in IL-6 levels precedes the CRP increase and correlates with the CRP and fibrinogen levels later (Figure 1).
The effects of IL-6 are modulated by the 2 soluble receptor subunits sIL-6R and sgp130. In vitro, sIL-6R promotes the effect of IL-6 as a coactivator, but sgp130 captures the soluble complex of IL-6/sIL-6R and prevents it from binding to the membrane-bound gp130 that mediates cellular effects.16 We report for the first time that sIL-6R levels are unchanged and that sgp130 serum concentration is reduced after stroke. However, the reduction after acute stroke was transient, suggesting that it is due to the cerebral ischemia and not related to the underlying vascular disease. The biochemical action of sgp130 predicts that reduced serum concentrations of sgp130 would promote the proinflammatory action of IL-6. However, in our study, there was no correlation between serum levels of sgp130 and activation of the acute-phase response. To detect a functional role of sgp130, a larger sample size would probably be required.
Of note, the expression of IL-6 in acute stroke not only is determined by infarct size but also is under genetic control. The IL-6 promoter contains 4 polymorphic sites that are associated in 8 common haplotypes because of linkage disequilibrium. The distribution of haplotypes in our group of patients and control subjects was similar to that reported in another white sample.11 Patients and control subjects differed only in the allelic frequency of the rare haplotype C (G-G-10/10-G), but an association of haplotype C with cerebrovascular disease would have to be verified in a larger sample. The most common haplotype of the IL-6 promoter (A-G-8/12-C), F, was associated with low IL-6 levels in stroke patients. The IL-6 promoter is known to regulate IL-6 expression in neural cells in response to adenosine and other stimuli involved in cerebral ischemia.21,23 This suggests that the combination of polymorphisms found in haplotype F might interfere with IL-6 induction in stroke or conversely that non–F haplotypes favor IL-6 induction. Indeed, the stimulation of IL-6 gene transcription by an analog of adenosine, which is involved in the ischemic pathophysiology,24 clearly depended on the haplotype of the IL-6 promoter in astrocytelike cells. Haplotypes B and E led to a high induction of IL-6 gene transcription, whereas haplotype F mediated only a low induction of IL-6 gene transcription in vitro. The relative rarity of haplotypes B and E in our clinical sample might have prevented a significant association of these haplotypes with high IL-6 serum concentrations. In contrast to IL-6 gene transcription stimulated by 2-chloroadenosine, the unstimulated IL-6 transcription in astrocytelike cells was not influenced by the haplotypic combination of polymorphisms. Likewise, the IL-6 serum concentrations in control subjects did not depend on the presence of haplotype F, although differences in body temperature and CRP levels suggest that polymorphisms in the IL-6 promoter might also modulate IL-6 expression before stroke. Because of the modular organization of eukaryotic promoters, polymorphisms may interfere with a distinct stimulus or condition without affecting another.
Previous association studies have focused on a single polymorphism at −174 in the IL-6 promoter but have not evaluated other polymorphisms or their haplotypic combination. The 4 polymorphisms, however, exert a cooperative effect on IL-6 gene transcription.11 The −174G polymorphism is present in haplotypes B, D, E, C, X, and G. Its effect on IL-6 transcription depends heavily on the haplotypic promoter context (Figure 3). Therefore, our results are not directly comparable with previous work. However, given the rarity of haplotype A (Table 2), haplotype F should be present in most individuals with the −174C polymorphism. Previous population genetic studies reported both equal and lower IL-6 levels in healthy control subjects with −174C.25,26 In other medical conditions, the −174C polymorphism was associated with equal and higher serum concentrations of IL-6.16,27
In summary, our data suggest 2 further levels of regulation of the important acute-phase response in stroke. The genetic control of IL-6 expression and the acute reduction in the potential IL-6 antagonist sgp130 may modulate the acute-phase response and outcome of stroke.
This study was supported by a grant from the DAAD (D.A.) and DFG (M.S.). We thank Inge Werner for excellent technical assistance; Professor Fiehn, Central Laboratory, University of Heidelberg, for measurement of CRP and fibrinogen; and Michael Preusch for help with the PCR.
- Received February 6, 2003.
- Revision received March 6, 2003.
- Accepted March 11, 2003.
Di Napoli M, Papa F, Bocola V. Prognostic influence of increased c-reactive protein and fibrinogen levels in ischemic stroke. Stroke. 2001; 32: 133–138.
Di Napoli M, Papa F, Bocola V. C-reactive protein in ischemic stroke: an independent prognostic factor. Stroke. 2001; 32: 917–924.
Castillo J, Dávalos A, Marrugat J, Noya M. Timing for fever-related brain damage in acute ischemic stroke. Stroke. 1998; 29: 2455–2460.
Maeda Y, Matsumoto M, Hori O, Kuwabara K, Ogawa S, Yan SD, Ohtsuki T, Kinoshita T, Kamada T, Stern DM. Hypoxia/reoxygenation-mediated induction of astrocyte interleukin 6: a paracrine mechanism potentially enhancing neuron survival. J Exp Med. 1994; 180: 2297–2308.
Tarkowski E, Rosengren L, Blomstrand C, Wikkelsö C, Jensen C, Ekholm S, Tarkowski A. Early intrathecal production of interleukin-6 predicts the size of brain lesion in stroke. Stroke. 1995; 26: 1393–1398.
Vila N, Castillo J, Dávalos A, Chamorro A. Proinflammatory cytokines and early neurological worsening in ischemic stroke. Stroke. 2000; 31: 2325–2329.
Terry CF, Loukaci V, Green FR. Cooperative influence of genetic polymorphisms on interleukin 6 transcriptional regulation. J Biol Chem. 2000; 275: 18138–18144.
Schöbitz B, Pezeshki G, Pohl T, Hemmann U, Heinrich PC, Holsboer F, Reul JM. Soluble interleukin-6 (IL-6) receptor augments central effects of Il-6 in vivo. FASEB J. 1995; 9: 659–664.
Jones SA, Horiuchi S, Topley N, Yamamoto N, Fuller GM. The soluble interleukin 6 receptor: mechanisms of production and implications in disease. FASEB J. 2001; 15: 43–58.
Grau AJ, Buggle F, Heindl S, Steichen-Wiehn C, Banerjee T, Maiwald M, Rohlfs M, Suhr H, Fiehn W, Becher H, et al. Recent infection as a risk factor for cerebrovascular ischemia. Stroke. 1995; 26: 373–379.
Adams HP Jr, Bendixen BH, Kappelle LJ, Biller J, Love BB, Gordon DL, Marsh EE3rd. Classification of subtype of acute ischemic stroke: definitions for use in a multicenter clinical trial: Toast: Trial of Org 10172 in Acute Stroke Treatment. Stroke. 1993; 24: 35–41.
Fassbender K, Rossol S, Kammer T, Daffertshofer M, Wirth S, Dollman M, Hennerici M. Proinflammatory cytokines in serum of patients with acute cerebral ischemia: kinetics of secretion and relation to the extent of brain damage and outcome of disease. J Neurol Sci. 1994; 122: 135–139.
Sallmann S, Juttler E, Prinz S, Petersen N, Knopf U, Weiser T, Schwaninger M. Induction of interleukin-6 by depolarization of neurons. J Neurosci. 2000; 20: 8637–8642.
Fishman D, Faulds G, Jeffery R, Mohamed-Ali V, Yudkin JS, Humphries S, Woo P. The effect of novel polymorphisms in the interleukin-6 (Il-6) gene on Il-6 transcription and plasma Il-6 levels, and an association with systemic-onset juvenile chronic arthritis. J Clin Invest. 1998; 102: 1369–1376.
Rauramaa R, Vaisanen SB, Luong LA, Schmidt-Trücksäss A, Penttilä IM, Bouchard C, Töyry J, Humphries SE. Stromelysin-1 and interleukin-6 gene promoter polymorphisms are determinants of asymptomatic carotid artery atherosclerosis. Arterioscler Thromb Vasc Biol. 2000; 20: 2657–2662.
Brull DJ, Montgomery HE, Sanders J, Dharmrait S, Luong L, Rumley A, Lowe GDO, Humphries SE. Interleukin-6 gene −174g→c and −572g→c promoter polymorphisms are strong predictors of plasma interleukin-6 levels after coronary artery bypass surgery. Arterioscler Thromb Vasc Biol. 2001; 21: 1458–1463.