Increased Circulating Immune Complexes in Acute Stroke
The Triggering Role of Chlamydia pneumoniae and Cytomegalovirus
Background and Purpose— The mechanisms of immune reaction involved in the pathogenesis and clinical course of acute vascular incidents are still not completely understood. The aim of this study was to examine the presence of immune complexes (IC) in the acute stroke setting and the first month thereafter and to characterize IC by analyzing the contents of chlamydial lipopolysaccharide and anti-cytomegalovirus (CMV) antibodies in IC.
Methods— Serum concentration of IC was investigated in 179 stroke patients, 122 “old” controls and 112 “young” controls, by the precipitation method. The presence of chlamydial lipopolisaccharyde and anti-CMV antibodies was investigated in some IC preparations by the ELISA method after earlier dissociation of IC into components by high pH treatment.
Results— Significantly increased serum IC concentration in stroke patients was noticed. Increased serum IC concentration was revealed as an independent strong stroke risk factor and was connected with significantly worse neurological status and increased 30-day mortality rate. A significantly larger proportion of stroke patients than controls had Chlamydia pneumoniae antigen and anti-CMV antibodies in IC.
Conclusions— This study provides the first evidence of an association between increased serum level of IC and the clinical course of cerebral ischemia and identifies a potentially important association of C pneumoniae and CMV-specific IC with stroke incidence.
The mechanisms of immune reaction involved in atherogenesis and acute vascular incidents have been studied extensively during recent years but are still not completely understood. The role of immune complexes (IC) in vessel pathology was evidenced in animal studies. Deposition of IC in the vessel wall of heteroimmunized animals was accompanied by (1) complement system activation and enhanced degranulation of basophils, (2) increased proliferation of smooth muscle cells and elevated synthesis of glucoseaminoglycans and collagen fibers, and (3) enhanced infiltration of monocytes and lymphocytes from peripheral blood to the arterial wall.1,2⇓ The role of IC in atherosclerotic responsiveness increases with age.3 The role of IC has also been supported in human atherogenesis.4–6⇓⇓ The most likely reason for enhanced IC formation accompanying atherosclerosis in humans seems to be cholesterol, 6,7⇓ especially low-density lipoproteins (LDL).8,9⇓ Antigens of infectious agents may be also considered as a possible trigger of increased IC. So far, cytomegalovirus (CMV), Chlamydia pneumoniae, and Helicobacter pylori have been documented as being strongly implicated in human atherosclerosis. In the study of Saikku et al.10 (the Helsinki Heart Study), the presence of IC containing chlamydial LPS (lipopolysaccharide) was an independent risk factor for development of cardiovascular events and allowed for predicting the coronary event 3 to 6 months before the onset. The presence of IC containing chlamydial LPS was documented in patients with myocardial infarction and coronary heart disease11,12⇓
The potential importance of increased IC formation after acute stroke has not been widely described. Although the role of immune response after stroke has been evidenced by many investigators, most studies have been focused on the activation of cellular immunity resulting from brain ischemia and associated with worse poststroke prognosis.13–18⇓⇓⇓⇓⇓
The aim of this study was to (1) examine serum levels of IC in the acute stroke setting and the first month thereafter, (2) characterize IC by analyzing the contents of chlamydial LPS and anti-CMV antibodies in IC, and (3) study the association of serum IC with stroke risk and prognosis.
Between January 1998 and March 1999 we studied patients with ischemic stroke (IS) consecutively admitted to our department of neurology within 24 hours after onset. CT imaging, Doppler sonography, echocardiography, and basal laboratory tests confirmed the diagnosis, established on the basis of history and examination.
Patients with vasculitis, cancer, acute infectious symptoms preceding stroke onset (fever, cough, hoarseness, sore throat, bronchitis, pharyngitis, arthritis, chills), gynecologic or urologic diseases, autoimmunologic disorders, and severe renal and hepatic diseases and those who developed hospital infection were excluded from investigations. The information concerning stroke risk factors and clinical data were collected prospectively according to the Stroke Data Bank (SDB), National Institutes of Health protocol,19 with some minor modifications. According to the Stroke Data Bank, etiologic subtypes of cerebrovascular ischemia were classified as followed: (1) infarct due to atheromatosis in extra- and intracerebral vessels, (2) embolism from the cardiac source, (3) lacunar infarct, and (4) infarct of unknown etiology. The Scandinavian Stroke Scale was used to assess patients’ neurological status (at the 1st, 7th, and 30th day after stroke occurrence). The 30-day stroke fatality rate was also calculated.
Two control groups, “old” and “young,” free of clinical signs of infection, other systemic diseases, and IS in history were included in the investigation. No special studies have been performed to assess the progress of atherosclerotic process in patients’ groups. The local Ethical Comittee approved the study.
Blood Sampling and Laboratory Procedures
Venous blood samples were obtained from stroke patients within 24 hours and at 7 and 30 days after the onset; samples were obtained once from control subjects. Serum was stored at −70°C until analysis.
Glucose concentration and total cholesterol, LDL, high-density cholesterol, and triglyceride levels were measured by Abbott Spectrum. Circulating IC were precipitated with the use of 5% polyethylene glycol (PEG) (FERAK Laboratories GMBH Berlin) according to the method of Creighton et al,20 with some minor modifications. Briefly, 0.3 mL of the sample was added to the 6 mL of 5% PEG in sodium borate buffer, pH 8.4, and incubated overnight at 4°C; next, centrifugation at 2500 rpm for 20 minutes at 4°C was performed. Pellets were washed twice with 3.5% PEG and dissolved with 0.3 mL of distilled H2O with the addition of 2.7 mL of 0.1 mol/L NaOH. The blind sample consisted of 0.3 mL of H2O distillated and 2.7 mL of 0.1 mol/L NaOH. Extinction was measured on a spectrophotometer at 280 nm. Results were expressed as OD280 values and considered positive when they exceeded the geometric mean OD280 value (calculated from the log-transformed distribution) in the young control group. Serum concentrations of IC were determined in 179 patients with IS, 122 old controls, and 112 young controls.
To investigate the species specificity of isolated IC, pellets obtained after PEG precipitation were resuspended in 100 μL of 0.1 mol/L sodium borate, pH 10.2. This dissociated IC into its components, which could be analyzed by ELISA for Chlamydia antigen (LPS) and anti-CMV antibodies.21
The analysis of chlamydial LPS contents in IC (isolated from the sera obtained within the first 24 hours after stoke onset) was performed in 44 patients, 75 old controls, and 56 young controls (subjects recruited from the group of patients with increased values of serum IC concentrations have been included into the study). The presence of chlamydial LPS in IC was investigated with the use of an ELISA Kit (DAKO) (two wells were used to get a mean value for a serum sample). Absorbance at 492 nm was measured on a Stat-Fax 2100 microplate reader (Awareness Technology Inc). The results were expressed as OD492 values and scored positive when they exceeded the geometric mean value (calculated from the log-transformed distribution) noticed in young controls (0.393).
The content of IC for anti-CMV antibodies was investigated in 56 stroke patients (within the first 24 hours after onset), 53 old controls, and 57 young controls (the subjects’ qualifications for this study were performed in a similar manner as in the case of the “ Chlamydial” group). The Vironostica anti-CMV II Kit (Organon Teknika) was used in that investigation (two wells were used to get a mean for a serum sample). Absorbance at 450 nm was measured on a Stat-Fax 2100 microplate reader (Awareness Technology Inc). The results were expressed as mean OD450 values and considered reactive if the sample absorbance was greater than the mean value noticed in young controls (0.395).
Statistical analysis was carried out with the STATISTICA PL. OD values representing serum concentrations of IC and the presence of chlamydial LPS in IC had distributions with a marked positive skew and were log transformed throughout. Serum IC level was analyzed both as a categorical variable (quintiles) and as a continuous variable. Student’s t test, Mann-Whitney U test, and Wilcoxon’s test were applied to assess significant differences of continuous variables among groups. For frequency distribution proportional tests-likelihood ratio, χ2 was performed. Spearman’s correlation test was used to study correlations between continuous variables. Age, hypertension, smoking, atrial fibrillation, congestive heart failure, myocardial infarction, and serum IC were investigated as predictors of stroke incidence using multiple regression techniques. The association between incident of stroke and serum levels of IC was investigated using logistic regression in STATISTICA. Similar logistic regression model was applied in analysis of the association of C pneumoniae-specific and CMV-specific IC with the risk of IS incidence.
We examined 179 patients with IS: 88 men and 91 women, with mean age of 74 (range, 42 to 97) years. The old control group consisted of 56 men and 66 women, with mean age 66 (range, 54 to 91) years. A characteristic of risk factors profile in stroke patients and old controls is presented in Table 1. The young control group comprised 112 individuals: 45 men and 67 women, with mean age of 35 years (range, 19 to 41) years.
Serum IC Concentrations in Stroke
As compared with controls, significantly elevated serum IC concentrations were noticed in stroke patients within 24 hours and at days 7 and 30 after onset. The highest level of serum IC was observed within the first 24 hours after stroke (Figure).
A significantly higher percentage of stroke patients were IC positive when compared with old and young controls (Table 2). Multiple logistic analyses revealed increased serum IC concentration as an independent stroke risk factor (Table 3). There was a strong trend for increasing risk of IS incidence with increasing serum levels of IC (Table 4).
There was no significant difference in total cholesterol and LDL levels between patients with increased versus normal serum IC concentration (total cholesterol 5.59±1.25 mmol/L versus 6.45± 1.06 mmol/L, respectively; LDL 4.02± 1.59 mmol/L versus 4.76±0.91 mmol/L, respectively). However, in the group of patients with increased serum IC concentration, a significant correlation has been noticed between serum IC concentration and (1) total cholesterol level (r=0.19; P=0.030) and (2) LDL (r=0.22; P=0.015).
Serum IC Concentration and Stroke Patients’ Neurological Statuses and Prognoses
Worse neurological status (on the Scandinavian Scale) in patients with elevated serum IC concentrations was noticed on the first day (32.0 [95% CI, 30 to 36] versus 41.0 [95% CI, 37 to 46], respectively, P<0.05) and seventh day (40.5 [95% CI, 38 to 43] versus 46.0, [95% CI, 42 to 50], P<0.05) after stroke onset in comparison with patients with normal IC levels.
The 30-day fatality rate was statistically higher in patients with elevated serum IC concentrations noticed on the first day after stroke onset when compared with patients with nonelevated IC levels. In the group of 142 patients with elevated serum IC concentrations, 24 persons died (16.9% [95% CI, 10.7 to 23.1]), whereas in the normal IC group only 1 of 37 individuals died (2.7% [95% CI, 2.5 to 7.9]) (P=0.0163).
Serum IC Concentration and Stroke Subtypes
In our group of patients, 31.3% (95% CI, 24.5 to 38.1) had atherothrombotic stroke, 21.2% (95% CI, 15.2 to 27.2) had lacunar stroke, 20.1% (95% CI, 14.2 to 26.0) had cardioembolic stroke, and 27.4% (95% CI, 20.8 to 34.9) had stroke of unknown etiology. We did not observe any differences in the levels of serum IC in respective groups with different etiologic stroke subtypes (Table 5).
The Species Specificity of Isolated IC
A significantly larger percentage of stroke patients than old and young controls had chlamydial LPS and anti-CMV antibodies in IC (Table 6). In the group of patients with chlamydial LPS in IC, more frequent history of atrial fibrillation (32% versus 0%, respectively; P<0.01) and previous stroke (24% versus 5.3%, respectively; P<0.05) was noticed when compared with patients without chlamydial LPS in IC. Patients with anti-CMV antibodies in IC more often presented earlier transient ischemic attack incidence (9.3% versus 0%, respectively; NS).
Increased levels of serum IC C pneumoniae-specific and CMV-specific IC were connected with increased risk of stroke incidence (odds ratio, 6.00; 95% CI, 1.61 to 22.29; P=0.0079; odds ratio, 7.60; 95% CI, 3.21 to 17.96; P=0.00001, respectively).
We demonstrated significantly increased serum IC concentrations at every time point of the study. The greatest increase in serum level of IC was noticed within the period from 24 hours to 7 days after stroke. Statistically significant differences were noticed in serum IC between patients and both groups of controls.
The difference in serum IC between patients and controls could be regarded as being a result of the older age of patients rather than of stroke incidence. However, firstly, no significant difference in serum IC concentration has been noticed between young and old controls; secondly, serum IC dropped continuously from 24 hours to 30 days, approaching levels noticed in controls. It is difficult to ultimately establish the reason for increased serum IC in stroke. When considering the lack of differences between serum levels of IC among different stroke subtypes and the continuous drop in IC concentration from 24 hours to 30 days, nearly approaching values noticed in old controls, it seems that elevated IC concentrations are a result of an event that occurs at the time of or just prior to the acute stroke.
The observation of the mostly increased serum IC concentration within the first 24 hours after onset and persisting until day 7 could be considered as representing an immune response to an acute infection directly preceding stroke. However, such explanation seems to be of minimal probability. All patients with acute infection symptoms preceding stroke onset (fever, cough, hoarseness, sore throat, bronchitis, pharyngitis, arthritis, chills) and those who developed infection during hospitalization were excluded from investigations.
Three other possibilities could be considered to explain enhanced IC formation observed shortly after stroke. One is neuronal injury accompanying stroke and stimulating increased IC formation. The second is liberation of antigens from damaged areas of brain vasculature into circulation. Two recent publications confirmed the presence of C pneumoniae in cerebral vessels. Virok et al22 detected the presence of C pneumoniae DNA in 33% of the atherosclerotic middle cerebral arteries. Nadareishvili et al23 noticed the presence of C pneumoniae in carotid aterosclerotic plaque. So it seems possible that elevated serum levels of IC, observed in our study, could be a result of liberation of chlamydial antigens into circulation from the damaged areas of vasculature. It cannot be excluded that IC of any other specificity could also arise as a consequence of a similar mechanism. The third possibility exists as a potential explanation of increased IC levels noticed shortly after stroke: silent exacerbation of a chronic infection directly preceding onset of symptoms. Because we could not exclude that the increased IC formation observed in our study directly precedes IS incidence, we included values of serum IC measured within 24 hours after IS into regression analysis, which revealed increased serum IC as a strong independent risk factor for IS. Increased serum levels of C pneumoniae-specific and CMV-specific IC were also connected with increased risk of stroke incidence. Nadareishvili et al23 suggest that the presence of C pneumoniae could modify the immune system in atherosclerotic plaque leading to plaque destabilization. In their study, a strong association between the presence of C pneumoniae and the accumulation of T lymphocytes (CD4+ and CD8+) was noticed in symptomatic versus asymptomatic patients.23 It could not be excluded that infection by C pneumoniae could also maintain destabilization of atherosclerotic plaque by induction of increased deposition of IC in atherosclerotic vessels. However, we did not study serum levels of IC before stroke incidence; therefore, the role of chronic process influence on increased IC formation is only hypothetic.
A significant part of isolated IC contained antigens of infectious agents: C pneumoniae and CMV. However, the possibility could not be excluded that increased IC formation could be triggered by other antigens, for example, antigens of endothelial cells, cardiolipins, heat shock proteins (HSP), or oxidatively modified LDL. In previous studies, we observed significantly elevated serum levels of anti-heat shock proteins and anti-cardiolipin antibodies in stroke patients during the first 48 hours after onset.24,25⇓ Our observation of a correlation between serum IC and total cholesterol and LDL levels in the group of patients with increased serum IC concentration could suggest that cholesterol or LDL could also trigger enhanced IC formation in stroke. The presence of IC containing cholesterol has been observed in patients with hypercholesterolemia (it was accompanied by significant increase of sC5b-9 and increase of soluble intercellular adhesion molecule-1 as a sign of endothelial dysfunction), atherosclerosis, and coronary stenosis.6,7⇓
One of the important findings of our study, which deserves special comment, seems to be the first evidence of worse mean neurological status and higher 30-day mortality rate in patients with elevated IC. Different hypotheses could be created to explain the influence of increased IC formation on the clinical course of IS. It is possible that increased IC formation after stroke could contribute to the enhanced postischemic brain tissue damage by complement system activation, with subsequent recruitment and stimulation of inflammatory cells such as neutrophils and macrophages within the ischemic area.6 The second mechanism of IC action enhancing postischemic brain tissue damage could be an upregulation of interleukin (IL)-8. IL-8 is a potent chemoattractant produced by mononuclear cells that is involved in polymorphonuclear neutraphil recruitment and activation. Upregulation of IL-8 mRNA expression was noticed already within the first few days after onset of stroke symptoms and remained elevated for up to 1 month.26
A limitation of this study is the method that has been used to measure serum IC concentration. The increase in optical density measured in our PEG preparates could be regarded as an evidence of methodological artifact due to the use of a common reagent (PEG-6000). It is possible that IC established only a part of the precipitates obtained after serum PEG treatment, but it seems impossible that more/less artifacts were pelleted in the sera obtained on the first, 7th, or 30th day after stroke onset. The evidence of IC contents in precipitates obtained after serum PEG treatment could be detection of chlamydial LPS and anti-CMV antibodies in dissolved pellets.
Our study demonstrates an association of increased serum IC with stroke incidence. We provided the first evidence of increased serum IC concentration influence on the clinical outcome of cerebral ischemia. We partially characterized isolated IC and documented the C pneumoniae and CMV specificity of IC. That observation further implicates C pneumoniae and CMV with the occurrence of stroke. Further investigations are needed to explain the mechanisms of IC action in stroke pathogenesis and course, taking into consideration the induction of proinflammatory cytokines and the participation of IC in complement system activation observed in acute vascular incidents.
- Received May 28, 2001.
- Revision received December 18, 2001.
- Accepted December 19, 2001.
- ↵Minick CR. The role of immunologically induced arterial injury in atherogenesis.In: Constantinides P, Pratesi F, Cavallero C, et al, eds. Immunity and Atherosclerosis. London: Academic Press; 1980: 111–120.
- ↵Tereshina OP, Kopylova GV, Butenko GM. Role of immune complexes in the pathogenesis of atherosclerosis: age-related aspect. Patol Fiziol Eksp Ter. 1994; 2: 8–10.
- ↵Mustafa A, Nityanand S, Berglund L, Lithell H, Lefvert AK. Circulating immune complexes in 50-year-old men as a strong and independent risk factor for myocardial infarction. Circulation. 2000; 102: 2576–2581.
- ↵Glader CA, Boman J, Saikku P, Stenlund H, Weinehall L, Hallmanns G, Dahlen GH. The proatherogenic properties of lipoprotein(a) may be enhanced through the formation of circulating immune complexes containing Chlamydia pneumoniae-specific IgG antibodies. Eur Heart J. 2000; 21: 639–646.
- ↵Lecomte E, Herbeth B, Clerc G, Khalife K, Siest G, Artur Y. Cholesterol content of circulating immune complexes in patients with coronary stenosis and subjects without evidence of atherosclerosis. Clin Chem. 1995; 41: 1526–1531.
- ↵Kalenich OS, Tertov VV, Liakishev AA, Ruda MI, Orekhov AN. The cholesterol content in immune complexes as a marker of coronary and peripheral atherosclerosis. Ter Arkh. 1991; 63: 59–61.
- ↵Kiener PA, Rankin BM, Davis PM, Yocum SA, Warr GA, Grove RI. Immune complexes of LDL induce atherogenic responses in human monocytic cells. Arterioscler Thromb Vasc Biol. 1995; 15: 990–999.
- ↵Saikku P, Leinonen M, Tenkanen L, Linnanmaki E, Ekman MR, Manninen V, Manttari M, Frick MH, Huttunen JK. Chronic Chlamydia pneumoniae infection as a risk factor for coronary heart disease in the Helsinki Heart Study. Ann Intern Med. 1992; 116: 273–278.
- ↵Linnanmaki E, Leinonen M, Mattila K, Nieminen MS, Valtonen V, Saikku P. Chlamydia pneumoniae-specific circulating immune complexes in patients with chronic coronary heart disease. Circulation. 1993; 87: 1130–1134.
- ↵Akopov SE, Simonian NA, Grigorian GS. Dynamics of polymorphnuclear leukocyte accumulation in acute cerebral infarction and their correlation with brain tissue damage. Stroke. 1996; 27: 1739–1743.
- ↵Del Zoppo GJ, Schmid-Schonbein GW, Mori E, Copeland BR, Chang CM. Polymorphnuclear leukocytes occlude capillaries following middle cerebral artery occlusion and reperfusion in baboons. Stroke. 1991; 22: 1276–1283.
- ↵Elneihoum AM, Falke P, Axelsson L, Lundberg E, Lindgarde F, Ohlsson K. Leukocyte activation detected by increased plasma levels of inflammatory mediators in patients with ischemic cerebrovascular diseases. Stroke. 1996; 27: 1734.
- ↵Kochanek PM, Hallenbeck JM. Polymorphnuclear leukocytes and monocytes/macrophages in the pathogenesis of cerebral ischemia and stroke. Stroke. 1992; 23: 1367–1379.
- ↵Tarkowski E, Naver H. Wallin BG, Blomstrand C, Tarkowski A. Lateralization of T-lymphocyte response in patients with stroke: effect of sympathetic dysfunction? Stroke. 1995; 28: 57.
- ↵Foulkes M, Wolf P, Price T, Stroke Data Bank. Design, methods, and baseline characteristic. Stroke. 1988; 19: 547–554.
- ↵Creighton WD, Lambert PH, Miescher PA. Detection of antibodies and soluble antigen-antibody complexes by precipitation with polyethylene glycol. J Immunol. 1973; 111: 1219–1227.
- ↵Virok D, Kis Z, Karai L, Intzedy L, Burian K, Szabo A, Ivanyi B, Gonczol E. Chlamydia pneumoniae in atherosclerotic middle cerebral artery. Stroke. 2001; 32: 1973–1976.
- ↵Nadareishvili ZG, Koziol DE, Szekely B, Ruetzler C, LaBiche R, McCarron R, DeGraba TJ. Increased CD8(+) T cells associated with Chlamydia pneumoniae in symptomatic carotid plaque. Stroke. 2001; 32: 1966–1972.
- ↵Kostulas N, Pelidou SH, Kivisakk P, Kostulas V, Link H. Increased IL-1B, IL-8 and IL-17 mRNA expression in blood mononuclear cells observed in a prospective ischemic stroke study. Stroke. 1999; 30: 2174–2179.