Interleukin 23 Levels Are Increased in Carotid Atherosclerosis
Possible Role for the Interleukin 23/Interleukin 17 Axis
Background and Purpose—Interleukin (IL)-23 is a cytokine in the IL-12 family, mainly produced by antigen-presenting cells with a central role in inflammation. We hypothesize that IL-23 is also important in atherogenesis and investigate this in a population with carotid atherosclerosis.
Methods—Plasma levels of IL-23 were measured in patients with carotid artery stenosis and in healthy controls. The mRNA levels of IL-23 and its receptor, IL-23R, were measured in atherosclerotic plaques, nonatherosclerotic vessels, peripheral blood mononuclear cells, and plasmacytoid dendritic cells.
Results—Our findings were as follows: (1) patients with carotid atherosclerosis (n=177) had significantly raised plasma levels of IL-23 when compared with healthy controls (n=24) with particularly high levels in those with the most recent symptoms. (2) mRNA levels of IL-23 and IL-23R were markedly increased in carotid plaques (n=68) when compared with nonatherosclerotic vessels (n=8–10). Immunostaining showed colocalization to plaque macrophages. (3) Patients with carotid atherosclerosis had increased mRNA levels of both IL-23 and IL-23R in plasmacytoid dendritic cells, but not in peripheral blood mononuclear cells. (4) IL-23 increased IL-17 release in monocytes and particularly in peripheral blood mononuclear cells from patients with carotid atherosclerosis, but not in cells from healthy controls. (5) IL-23 gave a prominent tumor necrosis factor release in monocytes from patients with carotid atherosclerosis but not in cells from healthy controls. (6) High plasma levels of IL-23 were associated with increased mortality during follow-up.
Conclusions—We have shown an association between IL-23 and disease progression in patients with carotid atherosclerosis, potentially involving IL-17-related mechanisms.
Stroke is one of the major causes of death and disability worldwide. Approximately 85% of all strokes are ischemic, and 20% to 30% of these are caused by carotid atherosclerosis. Moderate- to high-grade carotid artery stenosis can be detected in 1% to 3% of adults, and the incidence increases with age.1–3 Along with the degree of stenosis, inflammation and plaque composition are important in predicting the risk of clinical symptoms.4 The atherosclerotic plaque is composed of infiltrating inflammatory cells (eg, monocytes/macrophages, T cells, and dendritic cells [DCs]), smooth muscle cells, and lipids. Plaques prone to rupture have thin fibrous caps and high-grade inflammation. The balance between pro- and anti-inflammatory mediators is therefore of major importance for the fate of the plaque and for the occurrence of adverse events.5,6
Interleukin (IL)-23 belongs to the IL-12 family of cytokines and consists of the 2 subunits p19 and p40.7 IL-23 is mainly produced by macrophages and DCs, and through its interaction with the IL-23 receptor (IL-23R), composed of the IL-12Rβ1 unit and the specific IL-23R unit, plays a central role in inflammation including the induction of Th17 cells.8–10 IL-23 has been connected with various inflammatory disorders, such as inflammatory bowel diseases,11 rheumatoid arthritis,12 and psoriasis.13 IL-23 has also been implicated in atherogenesis.14–16 Thus, patients with peripheral arterial disease have raised serum levels of IL-23, and increased IL-23 expression is seen in human carotid lesions.14,16 Zhang et al17 showed that carriers of a polymorphism in the IL-23R gene, which also gave increased expression of IL-23R in peripheral blood mononuclear cells (PBMCs), had increased risk of coronary artery disease. However, decreased IL-23 in PBMCs from patients with coronary artery disease has also been reported.18 In a mouse model of ischemic stroke, IL-23 levels were upregulated in both the circulation and in the brain.15 However, the role for IL-23 in atherosclerosis is still unclear.
In this study, we examined the expression of IL-23 in patients with carotid atherosclerosis both systemically and within the atherosclerotic plaque as well as its relation to adverse events during follow-up. We also examined the effect of IL-23 in PBMCs and monocytes from patients with carotid atherosclerosis and healthy controls.
Materials and Methods
A description of tissue sampling from carotid plaque and nonatherosclerotic vessels, immunohistochemistry, immunofluorescence, isolation of PBMCs, monocytes and plasmacytoid DCs (pDCs), stimulation of cells, and real-time quantitative polymerase chain reaction is given in the online-only Data Supplement.
Patients and Control Subjects
One hundred and seventy-seven patients with moderate (50%–69%) or severe (≥70%) internal carotid artery stenosis, treated with carotid endarterectomy (n=147), carotid artery stenting (n=10), or conservatively (n=20), were recruited at Department of Neurology, Oslo University Hospital, Rikshospitalet (Table I in the online-only Data Supplement). The patients were classified into 2 groups according to their symptoms: (1) 69 patients (39%) had a stroke, transient ischemic attack, or amaurosis fugax ipsilateral to the stenotic internal carotid artery in the previous 2 months, and (2) 108 patients (61%) had symptoms >2 months ago or no relevant symptoms detected during clinical examinations of patients with coronary artery disease or peripheral artery disease. The groups are further denoted as symptomatic and asymptomatic, respectively. Carotid stenoses were diagnosed and classified by precerebral color duplex and computed tomographic angiography. Ultrasound plaque appearance in terms of echogenicity was classified according to consensus criteria.19,20 Exclusion criteria were concomitant disease that could influence inflammation, such as infections, connective tissue disease, or malignancies, heart failure, and liver or kidney disease. For comparison, blood samples were collected from 24 healthy control subjects recruited from the same area of Norway as the patients. All controls were apparently healthy individuals as assessed by disease history, clinical examination, and normal levels of C-reactive protein. The protocols were approved by the Regional Committee for Medical and Research Ethics, South-East, Norway, ref S-0923a 2009/6065. Signed informed consent for participation in the study was obtained from all individuals.
Blood Sampling Protocol
Venipuncture of a forearm vein was performed within 2 days before carotid endarterectomy/carotid artery stenting with minimal stasis. Blood was drawn into pyrogen-free tubes with ethylenediaminetetraacetic acid as anticoagulant. The tubes were immediately immersed in melting ice and centrifuged within 30 minutes at 2500g for 20 minutes to obtain platelet-poor plasma. All samples were stored at −80ºC.
Concentrations of IL-23 in plasma and IL-17A/F (denoted IL-17) and tumor necrosis factor (TNF) in PBMC (IL-17) and monocyte (TNF) supernatants were analyzed by enzyme immunoassay (R&D Systems). IL-17 levels in monocyte supernatants and plasma were measured by Bioplex technology (BioRad Laboratories, Hercules, CA).
For comparison of 2 groups of individuals, the Mann–Whitney U test was used. The χ2 test was used for analyzing contingency data. In the in vitro studies, Wilcoxon signed-rank test and Mann–Whitney U test were used for paired and unpaired data, respectively. Coefficient of correlation was calculated by the Pearson or Spearman rank test depending on the distribution of data. Kaplan–Meier analysis with log-rank test was performed to compare the number of events in relation to dichotomized IL-23 levels. IL-23 was dichotomized according to the median value in the patient group as a whole. The importance of IL-23 as a risk factor for all-cause mortality was further investigated by multivariable Cox regression including variables that were imbalanced (P<0.05) between survivors and nonsurvivors (ie, age and total leukocyte counts). Probability values (2-sided) were considered significant at P<0.05. All calculations were performed with SPSS for Windows statistical software (Version 18.0; SPSS Inc, Chicago, IL) or Graphpad Prism (Version 6.0; GraphPad Software, Inc., San Diego, CA).
Increased Plasma Levels of IL-23 in Patients With Carotid Atherosclerosis
Patients with carotid atherosclerosis (n=177) had markedly raised plasma levels of IL-23 compared with healthy controls (n=24) (median: 295 versus 94 pg/mL; P<0.01), with particularly high levels in patients with the most recent symptoms (ie, symptoms within the last 2 months, n=69; median and [range]: 430 [1–8779] pg/mL; P<0.001 versus controls 94 [3–391] pg/mL; Figure 1). Also when dividing the patients in other subgroups (symptoms <1 months ago [n=55]; symptoms 2–6 months ago [n=36]; symptoms >6 months ago or no symptoms [n=72]), the patients still had higher levels of IL-23 than healthy controls (IL-23, median and [ranges]: 438 [1–8779] pg/mL; P<0.001 versus controls; 345 [1–8638] pg/mL; P=0.007 versus controls and 173 [1–6268] pg/mL; P=0.05 versus controls, respectively). Thus, regardless of division, all subgroups of patients had higher levels of IL-23 than controls.
Of all baseline variables outlined in Table I in the online-only Data Supplement, including C-reactive protein, only IL-23 and use of clopidogrel were significantly associated with recent ischemic symptoms (ie, within 2 months, n=69) when compared with asymptomatic patients (ie, symptoms >2 months ago or no relevant symptoms, n=108). In the patient group as a whole, IL-23 levels were moderately correlated with age (r=0.2; P<0.05) and total leukocyte count (r=0.17; P<0.05). Although the patients were older than the controls (P<0.001), plasma levels of IL-23 were significantly higher in patients with carotid atherosclerosis also after adjusting for age (P<0.005). We found no association between IL-23 and the use of statins or other medications outlined in Table I in the online-only Data Supplement. For statins, this may reflect that 90% of the patients were on these medications.
Increased Expression of IL-23 and IL-23R in Carotid Atherosclerotic Plaques
Gene expression of IL-23 and IL-23R was quantified by quantitative polymerase chain reaction in plaques from patients with carotid atherosclerosis (n=68) and in nonatherosclerotic control vessels (common iliac artery of organ donors, n=8–10). There were markedly increased levels of both transcripts in plaques compared with control samples with particularly increased levels of IL-23R (18-fold increase; Figure 2). However, there were no differences in expression of either IL-23 or IL-23R between asymptomatic (n=13) and symptomatic (n=55) patients (data not shown). Within the plaques, expression of IL-23 and IL-23R was confirmed by immunohistochemistry (Figure I in the online-only Data Supplement). In carotid plaques, IL-23/IL-23R were colocalized to CD68+ macrophages (Figure 3), with only weak, sporadically, staining in relation to T cells and DCs (data not shown). For DCs, this could reflect the low abundance of these cells within the lesion.
Increased Expression of IL-23 and IL-23R in pDCs From Patients With Carotid Atherosclerosis
DCs are an important cellular source of IL-23, and this cytokine may in turn modulate the function of these cells.10 As shown in Figure 4, pDCs from patients with carotid atherosclerosis (n=10) had markedly increased mRNA levels of IL-23 (≈25-fold increase) and IL-23R (≈8-fold increase) compared with pDCs from healthy controls (n=14). In contrast, expression levels of IL-23 and IL-23R in PBMCs, primarily containing lymphocytes and monocytes, showed no significant difference between patients (n=23–24) and controls (n=35–36).
IL-23 Enhances IL-17 Release From PBMCs and Monocytes
IL-23 is a known inducer of the Th17 cell subset, and hence the production of IL-17.9 To explore this interplay in relation to atherosclerosis, we stimulated PBMCs from patients with carotid atherosclerosis (n=9) and healthy controls (n=8) with IL-23 (100 ng/mL) alone or in combination with phytohaemagglutinin (50 μg/mL) for 48 hours. Although IL-23 had no effect in PBMCs from healthy controls, it significantly increased IL-17 release in cells from patients with carotid atherosclerosis both when given alone and in phytohaemagglutinin-stimulated cells (Figure 5A).
To further elucidate the IL-23/IL-17 axis, we isolated monocytes from patients with carotid atherosclerosis (n=6) and healthy controls (n=5–6). The cells were preactivated by TNF to mimic the situation within an inflammatory microenvironment,21 and thereafter stimulated with IL-23 (100 ng/mL) alone or in combination with lipopolysaccharide (10 ng/mL) for 24 hours. The release of IL-17 from monocytes was in general low, and in some subjects absent. However, IL-23 induced a modest but significant increase in IL-17 release from patient cells, but not from control cells (Figure 5B). Our findings could suggest an interaction between IL-23 and IL-17, but we found no correlation between IL-17 and IL-23 either within the lesion (r=−0.2; P=0.09) or in plasma (r=−0.02; P=0.8). In plasma, the lack of correlation could at least partly reflect that several of the samples were low or below the detection limit of the IL-17 assay.
IL-23 Promotes TNF Release in Monocytes From Patients With Carotid Atherosclerosis
To further examine the inflammatory potential of IL-23 in monocytes, we examined the release of TNF in these cells after stimulation with IL-23 with or without costimulation with lipopolysaccharide. These experiments show that IL-23 enhanced the lipopolysaccharide-stimulated release of TNF in patients but not in controls (Figure 5C). Our data show that IL-23 has potent inflammatory effects in monocytes from patients with carotid atherosclerosis, beyond that of IL-17 release.
High Plasma Levels of IL-23 Are Associated With Mortality
To further elucidate the relationship of IL-23 to carotid atherosclerosis, we finally examined the association of plasma IL-23 levels to adverse outcome in the present study population (n=177; Table I in the online-only Data Supplement). During follow-up (mean 3.5 years), 28 of the patients died. About half of the mortalities (n=13) were caused by cerebrovascular or cardiovascular events. High IL-23 levels in plasma (ie, >median 295 pg/mL) were significantly associated with increased mortality also when adjusting for variables that were imbalanced (P<0.05) between survivors and nonsurvivors (age and leukocyte count; Figure 6).
Previous studies have shown increased expression of IL-23 in carotid lesions with particularly high levels in symptomatic patients.14 In this study, we extend these findings by showing that the increased expression of IL-23 in carotid plaques is accompanied by an enhanced expression of its receptor IL-23R, both primarily located to macrophages. These patients also had increased levels of IL-23 in plasma and markedly increased expression of IL-23/IL-23R in pDCs. High plasma levels of IL-23 were associated with recent symptoms and with increased mortality during follow-up, underscoring the association between IL-23 and carotid atherosclerosis.
DCs are key modulators of immunity, pivotal in promoting innate and adaptive immune responses. pDCs have been suggested to promote atherogenesis through interferon-α-related mechanisms.22 In this study, we have shown that patients with carotid atherosclerosis have markedly enhanced expression of IL-23/IL-23R in pDCs, pointing to a role of IL-23 in relation to pDC activity in these patients.
To elucidate the effect of the increased IL-23/IL-23R levels in patients with carotid atherosclerosis, we stimulated PBMCs from patients and controls with IL-23. In vitro stimulation of PBMCs gives the opportunity to study the interplay between different immune cells and could thereby mimic the in vivo situation of plaque inflammation. We found that IL-23 significantly increased IL-17 release from patients, but not from control PBMCs. Also, IL-23 further potentiated the phytohaemagglutinin-inducing effect on IL-17 release, pointing toward an acceleration of a preactivated inflammatory status in these patients. An IL-23-inducing effect on IL-17 release was also found in monocytes from patients with carotid atherosclerosis, but not in monocytes from healthy controls. However, the IL-17 response from monocytes was in general low suggesting that monocytes are not an important cellular source of IL-17 within the PBMCs. Whereas the IL-17-inducing effect of IL-23 in monocytes was rather low, IL-23 markedly increased TNF release in these cells when costimulated with lipopolysaccharide, and again, this was only seen in patient cells. Our data suggest an inflammatory potential of IL-23 in PBMCs and monocytes, and these data also underscore the importance of examining cells from both patients and healthy controls when studying functional properties of cells in relation to atherosclerosis.
The role of IL-17 in atherosclerosis has been debated,23 but our in vitro results may strengthen its position as a potential proatherogenic mediator. Thus, we suggest that the IL-23/IL-17 axis, well characterized in many other inflammatory conditions, also may play a role in atherosclerotic disease.24,25 Recently, however, IL-23 has also been shown to have important proatherogenic effects on macrophages, independently of IL-17.26 In our study, IL-23 significantly increased TNF release in patient monocytes, and we also found colocalization of IL-23/IL-23R and macrophages within the atherosclerotic lesion, supporting an interplay between monocytes/macrophages and IL-23 in atherosclerotic disease. In contrast to the increased expression of IL-23/IL-23R in carotid plaques, this was not seen in PBMCs from these patients, suggesting that the upregulation of IL-23/IL-23R in PBMC-related cells, such as macrophages, within the atherosclerotic lesion reflects processes within the plaque. However, the relationship between IL-23 and different immune cells in atherosclerosis needs further investigation.
This study has some limitations, such as a relatively low number of patients and controls, and the controls were not perfectly matched with the patient group. The lack of functional data from pDCs and T cells limits the effect of our data, and the mortality data should be interpreted with caution because of a low number of events. Moreover, the association between IL-23 and carotid atherosclerosis does not necessarily imply any causal relationship. Even so, our findings may suggest a potential proatherogenic role of IL-23, and this possibility should be further investigated in larger clinical studies as well as in more mechanistic studies, including studies in experimental atherosclerosis.
We thank Ellen Lund Sagen at the Institute for Internal Medicine, Oslo University Hospital, and the transplantation surgeons at Oslo University Hospital for their contribution to this work.
Sources of Funding
This work was supported by grants from the Norwegian Council of Cardiovascular Research, Research Council of Norway, University of Oslo, and South-Eastern Norway Regional Health Authority.
↵† Drs Halvorsen and Skjelland are joint senior authors.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.114.006516/-/DC1.
- Received June 24, 2014.
- Revision received December 5, 2014.
- Accepted December 24, 2014.
- © 2015 American Heart Association, Inc.
- Roger VL,
- Go AS,
- Lloyd-Jones DM,
- Adams RJ,
- Berry JD,
- Brown TM,
- et al
- Mathiesen EB,
- Bønaa KH,
- Joakimsen O
- Parham C,
- Chirica M,
- Timans J,
- Vaisberg E,
- Travis M,
- Cheung J,
- et al
- Langrish CL,
- Chen Y,
- Blumenschein WM,
- Mattson J,
- Basham B,
- Sedgwick JD,
- et al
- Breland UM,
- Michelsen AE,
- Skjelland M,
- Folkersen L,
- Krohg-Sørensen K,
- Russell D,
- et al
- Qu N,
- Xu M,
- Mizoguchi I,
- Furusawa J-i,
- Kaneko K,
- Watanabe K,
- et al
- Subramanian M,
- Thorp EB,
- Tabas I