Abstract
A large community outbreak of Q fever occurred in the Netherlands in the period 2007 to 2010. Some of the infected patients developed chronic Q fever, which typically includes pathogen dissemination to predisposed cardiovascular sites, with potentially fatal consequences. To identify the immune mechanisms responsible for ineffective clearance of Coxiella burnetii in patients who developed chronic Q fever, we compared serum concentrations of 47 inflammation-associated markers among patients with acute Q fever, vascular chronic Q fever, and past resolved Q fever. Serum levels of gamma interferon were strongly increased in acute but not in vascular chronic Q fever patients, compared to past resolved Q fever patients. Interleukin-18 levels showed a comparable increase in acute as well as vascular chronic Q fever patients. Additionally, vascular chronic Q fever patients had lower serum levels of gamma interferon-inducible protein 10 (IP-10) and transforming growth factor β (TGF-β) than did acute Q fever patients. Serum responses for these and other markers indicate that type I immune responses to C. burnetii are affected in chronic Q fever patients. This may be attributed to an affected immune system in cardiovascular patients, which enables local C. burnetii replication at affected cardiovascular sites.
INTRODUCTION
A large community outbreak of Query fever (Q fever), which lasted from 2007 to 2010, took place in the Netherlands (1). In this period, 4,026 cases were reported, of which the majority (2,354) occurred in 2009. Q fever is a zoonotic infection caused by the omnipresent Gram-negative obligate intracellular coccobacillus Coxiella burnetii, which infects macrophages and monocytes. After passive entry into the host cell, C. burnetii is internalized into phagosomes, which fuse rapidly with lysosomes to form phagolysosomes (2, 3). After exposure, approximately 60% of patients remain asymptomatic. In symptomatic patients with acute Q fever, the most common clinical presentations are isolated fever, pneumonia, and hepatitis. In general, it is a mild disease that resolves spontaneously within 2 weeks. A chronic fatigue syndrome, which may persist for years, has been reported as a clinical complication following acute Q fever. This disease entity should be distinguished from chronic Q fever, which is typically a dissemination of the infection to predisposed sites like defective cardiac valves or aneurysms and vascular grafts, causing potentially fatal endocarditis or vascular infection, respectively (3, 4). The all-cause mortality of chronic Q fever in the Netherlands is 13%, and even 18% in patients with vascular chronic Q fever (5). Because of this high mortality, patients with cardiovascular disease were actively screened for the presence of chronic Q fever during and following the outbreak (6, 7). This screening contributed to the identification of the largest group of chronic Q fever patients diagnosed in a single outbreak.
In chronic Q fever patients, the immune response to C. burnetii infection is considered ineffective, possibly due to a defective cell-mediated immune response and absence of formation of granulomas (3, 4, 8, 9). However, the exact mechanism that underlies this defective response has not been elucidated yet. Type I immunity is believed to play a crucial role in the host response against C. burnetii. Activation of type I immunity leads to macrophage activation and killing of intracellular bacteria. Gamma interferon (IFN-γ), the key type I cytokine, is produced by T cells and natural killer (NK) cells, which play an important role in host defense against intracellular pathogens (10). In mice, it has been shown that IFN-γ is essential for the early control of C. burnetii infection (11). In literature case reports, a sustained clinical response of immunomodulatory therapy with IFN-γ was reported in a child with dual-antibiotic-therapy-resistant chronic multifocal Q fever osteomyelitis (12) as well as a child with multiple-antibiotic-resistant chronic Q fever (13). In addition, it has been shown in vitro that C. burnetii multiplied in untreated human monocytic cells, while reduced bacterial viability was observed after 24 h of IFN-γ treatment (14). These observations point to an essential role of IFN-γ in immunity against C. burnetii. However, whether a defect in type I immunity and IFN-γ production is involved in the development of chronic Q fever is not known.
Only a small proportion of C. burnetii-infected patients with cardiovascular risk factors go on to develop chronic Q fever (6, 7). Apparently, only a subgroup of these patients is unable to cope with this infection. The present study was designed to identify the immune mechanisms involved in efficient clearance of C. burnetii and the nature of the immune defect that hampers clearance in patients that go on to develop chronic Q fever. To this end, we analyzed the immune response of Q fever patients by measuring circulating serum concentrations of a set of cytokines and comparing the responses of patients with acute Q fever, vascular chronic Q fever, and past resolved Q fever.
MATERIALS AND METHODS
Serum samples.
Included in this retrospective study were sera, obtained from 22 May to 31 May 2009 (weeks 21 and 22), of adult patients (≥18 years) who were diagnosed with PCR-proven seronegative acute Q fever by the Department of Medical Microbiology and Infection Control of the Jeroen Bosch Hospital (JBH; 's-Hertogenbosch, the Netherlands). All sera were referred because of clinical suspicion of acute Q fever. This 10-day period lies in the peak incidence weeks of the 2007-to-2010 acute Q fever outbreak (data from the National Institute for Public Health and the Environment). Serum samples of these patients were screened for IgM phase II antibodies with an enzyme-linked immunosorbent assay (ELISA) (Institut Virion/Serion GmbH, Würzburg, Germany). Following a negative outcome of the IgM phase II ELISA, an in-house PCR for C. burnetii DNA had been performed on serum with a positive outcome establishing the diagnosis of acute Q fever (15).
Sera of adult vascular chronic Q fever patients were obtained as part of a call/recall screening program for high-risk patients with an aortic aneurysm or central vascular reconstruction, initiated during and following the outbreak in the areas of the JBH and the neighboring Bernhoven Hospital (locations in Oss and Veghel, the Netherlands) (7). These regions correspond to the center of the epidemic. Screening for IgG phase I and IgG phase II antibodies was performed on serum samples with an immunofluorescence assay (IFA) (Focus Diagnostics, Inc., Cypress, CA, USA). In the case of an IgG phase I titer of ≥1:512, subsequent C. burnetii PCR on serum was performed. According to diagnostic criteria established by the Dutch Q Fever Consensus Group, a positive serum PCR in combination with a corresponding serologic profile in a patient with a vascular risk factor is considered diagnostic for proven chronic Q fever (16). Only patients diagnosed with PCR-proven vascular chronic Q fever in a clinically stable situation were analyzed in this study (excluding patients presenting with vascular complications, after recent surgery, or with coinfections).
Sera of adult patients with past resolved Q fever obtained as part of the above-mentioned screening program among patients with an aortic aneurysm or central vascular reconstruction were included in the control group. In these sera, screening for IgG phase I and IgG phase II antibodies revealed low antibody titers indicative of past resolved Q fever infection without indication for chronic Q fever.
After collection, all sera were stored at −20°C.
Ethics statement.
In 2009, the Internal Review Board of the JBH approved the anonymous use of discarded blood for validation or research purposes. All patients who donated blood were informed of this possibility and had the right of refusal.
A regional medical ethics committee (Medisch-Ethische Toetsing Patiënten en Proefpersonen [METOPP]) waived the need for informed consent as far as testing of high-risk groups for chronic Q fever was concerned.
Multimarker analysis.
From each of the three clinical groups (acute Q fever, vascular chronic Q fever, and past resolved Q fever), 10 serum samples were randomly selected for multianalyte serum analysis. Next, vials containing 125 μl of serum from these samples were blinded and randomized before shipment to the biomarker testing laboratory, Rules-Based Medicine (RBM; Austin, TX, USA). There, a set of 47 inflammation-associated markers were measured by bead-based immunoassay using the Human InflammationMAP v.1.0 marker panel on a Luminex 100 instrument (Luminex, Austin, TX, USA).
For data analysis, eight markers which did not have at least five values in the detectable range in at least one of the experimental groups were discarded from further statistical analysis. Values flagged as below the measurement range were replaced for further calculations by an estimated concentration of 80% of the lowest measurable concentration for that analyte. Further statistical analysis was performed using the statistical software package R (www.r-project.org). Per group, the geometric average concentration per marker was calculated. Differences in marker concentrations between the three groups were determined using a one-way analysis of variance (ANOVA) on log-transformed data. Resulting P values were corrected for multiple testing using the false discovery rate (FDR) according to Benjamini and Hochberg (17). An FDR threshold of 10% was used to identify markers with significantly different serum concentrations. For these markers, post hoc t tests were performed on the pairwise contrasts among the three experimental groups, again with an FDR threshold of 10% as a significance threshold. Data were visualized as a heat map and by principal component analysis (PCA).
Targeted serum analysis.
For both validation of previous results and testing of new hypotheses, serum levels of a smaller number of markers were determined in-house, using sera from 43 acute Q fever patients, 31 vascular chronic Q fever patients, and 10 patients with past resolved Q fever. The vascular chronic Q fever group and the past resolved Q fever groups both included 10 sera also used in the multimarker analysis. Bead-based immunoassays for IFN-γ (Bio-Rad, Veenendaal, the Netherlands), IFN-γ-inducible protein 10 (IP-10) (R&D Systems, Minneapolis, MN, USA), transforming growth factor β1 (TGF-β1) (R&D Systems), and TGF-β2 (R&D Systems) were performed on a Luminex 100 instrument. Serum volumes used were 25 μl for IFN-γ and IP-10 and 20 μl for TGF-β1 and TGF-β2. Interleukin-18 (IL-18) was determined by ELISA (MBL, Woburn, MA, USA) using 20 μl serum. All assays were performed according to the manufacturer's instructions.
Statistical analysis was performed by pairwise comparisons between all groups, using a t test on log-transformed values. For this second (targeted) step, a more stringent FDR-corrected P value of 5% was used as a significance threshold.
RESULTS
Patients.
In the period from 22 May to 31 May 2009, we diagnosed 43 adult patients (24 males [55.8%], mean age of 50 [range, 19 to 82] years) during the early acute stage of acute Q fever as illustrated by the presence of circulating C. burnetii DNA (positive PCR) but negative serology (negative IgM phase II ELISA). Follow-up serum samples were received from 30 of these 43 patients. IFA revealed the presence of IgG phase I and/or IgG phase II in all of these follow-up samples, indicating subsequent seroconversion after the initial seronegative phase of the infection.
Screening for chronic Q fever in patients with vascular risk factors identified 31 adult patients (26 males [83.9%], mean age of 71 [range, 53 to 90] years) with PCR-proven vascular chronic Q fever in a clinically stable situation. Vascular chronic Q fever patients were older and significantly more often male than acute Q fever patients.
In addition, this screening program identified 88 patients with vascular risk factors and past resolved Q fever (74 males [84.1%], mean age of 71 [range, 34 to 91] years). Patient characteristics are shown in Table 1.
TABLE 1.
Patient characteristics in acute and vascular chronic Q fever patients
| Characteristic | Value for Q fever group (n): |
P value, acute vs vascular chronic Q fever | ||
|---|---|---|---|---|
| Past resolved (88) | Acute (43) | Vascular chronic (31) | ||
| Gender (% male) | 84.1 | 55.8 | 83.9 | 0.013 |
| Age (yr) | 71 ± 9 | 50 ± 16 | 71 ± 9 | <0.001 |
| C. burnetii DNA load (qPCR CT)a | Not determined | 35.9 ± 3.9 | 36.8 ± 4.1 | 0.36 |
qPCR threshold cycle (CT) values are based on the 2-logarithm of the C. burnetii DNA quantity, with a lower CT value indicating a higher load.
Circulating C. burnetii DNA loads as determined by quantitative PCR (qPCR) during the routine clinical diagnostics were not significantly different between acute and vascular chronic Q fever patients (Table 1).
Immune response.
By measuring a large set of cytokines in a small group of patients, we first identified cytokines and other inflammatory markers that were differentially expressed in patients with acute and vascular chronic Q fever. This analysis showed differences in 16 out of 39 analyzed markers (Table 2). Upon comparison to patients with past resolved Q fever, changes in serum concentrations were for most of these markers more pronounced in acute than in vascular chronic Q fever patients. In a PCA visualization, which visualizes samples based on their degree of dissimilarity, the acute Q fever group was clearly distinct from the past resolved Q fever group, whereas the group with vascular chronic Q fever showed a similar overall trend but was not readily distinguishable (Fig. 1).
TABLE 2.
Serum marker analysis resultsb
| Marker | Geometric avg/group: |
FDRa | Post hoc testa | ||
|---|---|---|---|---|---|
| Past resolved | Acute | Vascular chronic | |||
| Alpha-2-macroglobulin (mg/ml) | 1.17 | 1.99 | 1.33 | 0.158 | |
| Alpha-1-antitrypsin (mg/ml) | 2.25 | 3.72 | 3.55 | 0.037 A | BC |
| Beta-2-microglobulin (μg/ml) | 2.03 | 2.13 | 2.70 | 0.337 | |
| Brain-derived neurotrophic factor (ng/ml) | 29.3 | 21.6 | 21.9 | 0.133 | |
| Complement C3 (mg/ml) | 1.44 | 2.45 | 1.67 | 0.002 A | BD |
| C-reactive protein (μg/ml) | 3.2 | 157.0 | 11.4 | <0.001 A | BD |
| Eotaxin-1 (pg/ml) | 369 | 160 | 334 | 0.002 A | BD |
| Factor VII (ng/ml) | 661 | 373 | 432 | 0.037 A | B |
| Fibrinogen (μg/ml) | 13.9 | 25.6 | 14.5 | 0.067 A | |
| Ferritin (ng/ml) | 161 | 638 | 99 | 0.001 A | BD |
| Haptoglobin (mg/ml) | 1.87 | 4.17 | 2.43 | 0.163 | |
| Intercellular adhesion molecule 1 (ng/ml) | 113 | 201 | 199 | 0.002 A | BC |
| Gamma interferon (pg/ml) | 0.69 | 8.48 | 0.73 | <0.001 A | BD |
| Interleukin-1 alpha (pg/ml) | 1.27 | 1.97 | 1.48 | 0.163 | |
| Interleukin-1 beta (pg/ml) | 1.69 | 2.16 | 1.56 | 0.343 | |
| Interleukin-1 receptor antagonist (pg/ml) | 112 | 403 | 133 | <0.001 A | BD |
| Interleukin-6 (pg/ml) | 4.87 | 35.26 | 7.69 | 0.004 A | BD |
| Interleukin-8 (pg/ml) | 35.2 | 24.6 | 94.3 | 0.129 | |
| Interleukin-10 (pg/ml) | 2.92 | 6.39 | 3.48 | 0.159 | |
| Interleukin-12 subunit p70 (pg/ml) | 9.9 | 19.0 | 10.6 | 0.082 A | B |
| Interleukin-15 (ng/ml) | 0.37 | 0.30 | 0.33 | 0.723 | |
| Interleukin-17 (pg/ml) | 2.42 | 2.31 | 1.67 | 0.413 | |
| Interleukin-18 (pg/ml) | 177 | 291 | 290 | 0.082 A | B |
| Interleukin-23 (ng/ml) | 1.84 | 1.85 | 1.79 | 0.989 | |
| Monocyte chemotactic protein 1 (pg/ml) | 263 | 174 | 221 | 0.300 | |
| Macrophage inflammatory protein-1 alpha (pg/ml) | 56.2 | 24.9 | 103.9 | 0.004 A | BD |
| Macrophage inflammatory protein-1 beta (pg/ml) | 318 | 489 | 676 | 0.103 | |
| Matrix metalloproteinase-2 (ng/ml) | 2.01 | 1.53 | 1.46 | 0.163 | |
| Matrix metalloproteinase-3 (ng/ml) | 8.16 | 6.05 | 8.82 | 0.447 | |
| Matrix metalloproteinase-9 (ng/ml) | 15.6 | 26.0 | 23.4 | 0.098 A | |
| T-cell-specific protein RANTES (ng/ml) | 16.2 | 11.8 | 24.8 | 0.037 A | D |
| Stem cell factor (pg/ml) | 268 | 199 | 242 | 0.244 | |
| Tissue inhibitor of metalloproteinases 1 (ng/ml) | 180 | 210 | 197 | 0.273 | |
| Tumor necrosis factor alpha (pg/ml) | 2.38 | 2.84 | 5.67 | 0.106 | |
| Tumor necrosis factor receptor 2 (ng/ml) | 9.5 | 13.6 | 14.1 | 0.129 | |
| Vascular cell adhesion molecule 1 (ng/ml) | 698 | 847 | 805 | 0.413 | |
| Vitamin D-binding protein (μg/ml) | 224 | 239 | 309 | 0.343 | |
| Vascular endothelial growth factor (pg/ml) | 807 | 912 | 957 | 0.723 | |
| von Willebrand factor (μg/ml) | 44.8 | 52.3 | 54.9 | 0.337 | |
Significance is shown by uppercase letters as follows: A, significant in one-way ANOVA at an FDR of <10%; B, significant for post hoc comparison on past resolved versus acute Q fever group (FDR, <10%); C, significant for post hoc comparison on past resolved versus vascular chronic Q fever group (FDR, <10%); D, significant for post hoc comparison on acute versus vascular chronic Q fever group (FDR, <10%).
A set of 39 serum markers were measured in three clinical groups (each for n = 10 samples): past resolved Q fever group, acute Q fever group, and vascular chronic Q fever group.
FIG 1.
Principal component analysis of markers with different serum levels in patients with acute Q fever, vascular chronic Q fever, and past resolved Q fever. Distances between samples indicate the degree of dissimilarity. Shades of gray indicate patient groups: light gray, past resolved; dark gray, acute; black, vascular chronic Q fever.
More detailed data visualization in a heat map indicated that differentially expressed markers could be arranged into groups with distinct expression patterns (Fig. 2). C-reactive protein (CRP) and IL-6 serum levels were very strongly increased in the acute Q fever group compared to a smaller increase in the vascular chronic Q fever group. Four markers (IL-18, matrix metalloproteinase-9 [MMP9], alpha-1-antitrypsin [AAT], and intercellular adhesion molecule 1 [ICAM1]) had increased expression at comparable levels in both acute and vascular chronic Q fever patients. Several markers showed an increase in the acute Q fever group but no substantial change in the vascular chronic Q fever group. These markers included IL-1 receptor antagonist (IL-1RA), complement C3, IL-12 subunit p70 (IL-12p70), ferritin, IFN-γ, and fibrinogen. The difference was most pronounced for IFN-γ. Two markers, macrophage inflammatory protein-1 alpha (MIP1α) and Regulated on Activation, Normal T cell Expressed and Secreted (RANTES), had lower serum levels in acute Q fever patients but increased levels in the vascular chronic Q fever group. Finally, serum levels of factor VII and eotaxin-1 were decreased in both acute and vascular chronic Q fever groups compared to patients with past resolved Q fever (Table 2; Fig. 2).
FIG 2.
Heat map of serum markers with significantly different (FDR, 10%) serum concentrations in patients with acute Q fever (ACU01 to -10), vascular chronic Q fever (CHR01 to -10), and past resolved Q fever (PAS01 to -10). Serum levels are given in comparison to the average of the past resolved Q fever patient group. The color scale runs from red (increased levels) via white (unchanged) to blue (decreased levels). Colors in the top bar indicate patient groups and correspond to those in Fig. 1. AAT, alpha-1-antitrypsin; C3, complement C3; CRP, C-reactive protein; ICAM1, intercellular adhesion molecule 1; IFN-g, gamma interferon; IL-1RA, interleukin-1 receptor antagonist; IL-6, interleukin-6; IL-12p70, interleukin-12 subunit p70; IL-18, interleukin-18; MIP1a, macrophage inflammatory protein-1 alpha; MMP9, matrix metalloproteinase-9; RANTES, Regulated on Activation, Normal T cell Expressed and Secreted.
Because C. burnetii is an intracellular pathogen and its clearance is probably induced via the Th1 route, we further explored our finding of low IFN-γ levels in the vascular chronic Q fever group in larger groups of patient sera. In addition, several other cytokines involved in IFN-γ signaling were measured in these larger patient groups. These included IL-18, a cytokine known to induce the production of IFN-γ by NK cells and T cells; IP-10, a chemoattractant for activated T cells expressed in many Th1-type inflammatory diseases (18) which is also regulated by IL-18; and TGF-β1, because it is known to reduce IFN-γ production (19).
These analyses again showed significantly higher levels of IFN-γ in the acute Q fever group (12.30 pg/ml) than in the vascular chronic Q fever (2.72 pg/ml) or the past resolved Q fever (1.75 pg/ml) group, both differences being significant at an FDR-corrected P value of <0.001. Differences between vascular chronic Q fever and past resolved Q fever patients were not significant (FDR-corrected P value of 0.32) (Table 3).
TABLE 3.
Targeted serum analysisa
| Marker | Geometric avg/group (n): |
FDR for comparison: |
||||
|---|---|---|---|---|---|---|
| Past resolved (10) | Acute (43) | Vascular chronic (31) | Past resolved with acute | Past resolved with vascular chronic | Acute with vascular chronic | |
| IFN-γ (pg/ml) | 1.75 | 12.30 | 2.72 | <0.001 | 0.320 | <0.001 |
| IL-18 (pg/ml) | 325 | 593 | 761 | 0.005 | <0.001 | 0.062 |
| IP-10 (pg/ml) | 43.0 | 161.3 | 93.3 | <0.001 | 0.023 | 0.004 |
| TGF-β1 (ng/ml) | 112.5 | 141.4 | 11.9 | 0.055 | 0.029 | <0.001 |
| TGF-β2 (pg/ml) | 577 | 769 | 209 | <0.001 | 0.050 | <0.001 |
Serum markers were measured in three clinical groups: past resolved Q fever group, acute Q fever group, and vascular chronic Q fever group.
In the analysis of the larger groups, IL-18 again showed significantly increased levels in acute and vascular chronic Q fever patients, compared to past resolved Q fever patients. Levels in acute and vascular chronic Q fever patients were 593 and 761 pg/ml, respectively, which were not significantly different (FDR = 0.062).
Serum levels of IP-10 were higher in patients with acute Q fever (161 pg/ml) than in vascular chronic Q fever patients (93 pg/ml) and past resolved Q fever patients (43 pg/ml) (both significant at an FDR of <0.01). The difference between the vascular chronic Q fever and past resolved Q fever groups was also significant (FDR = 0.023).
TGF-β1 levels were decreased in vascular chronic Q fever patients (11.9 ng/ml) compared to acute Q fever patients (141.4 ng/ml) or past resolved Q fever patients (112.5 ng/ml). For this marker, the difference between the vascular chronic Q fever group and both other groups was significant at an FDR of <5% (Table 3). Also, TGF-β2 levels were decreased in vascular chronic Q fever patients (209 pg/ml) compared to acute Q fever patients (769 pg/ml) or past resolved Q fever patients (577 pg/ml), with the levels in the acute Q fever group significantly different from both other groups (Table 3).
DISCUSSION
In this study, we set out to identify the immune mechanisms involved in clearance of C. burnetii, and the nature of the immune defect that hampers such clearance in patients who go on to develop chronic Q fever, within the context of this study more specifically vascular chronic Q fever. A particular strength of our study is that we determined serum levels of over 40 different cytokines and other inflammatory markers, allowing us to relate the various responses to each other. To our knowledge, this study is the first one describing such a comprehensive serum profiling in Q fever patients.
A considerable number of markers showed different levels in acute Q fever patients compared to vascular chronic Q fever patients. Of these markers, CRP and IL-6 were strongly increased in the acute Q fever group and to a lesser extent in the vascular chronic Q fever group compared to the past resolved Q fever group. Recent findings obtained in acute Q fever patients from the same outbreak also showed correlated responses between these two acute-phase response markers (20). We also show here that a set of CC-type chemokines (MIP1α, RANTES, and eotaxin-1) shows downregulation in acute Q fever patients and a heterogeneous response in the group of vascular chronic Q fever patients. These chemokines share their capacity to attract eosinophils, and their common downregulation in the acute Q fever group points toward a skewing of the systemic chemokine profile away from a Th2 toward a Th1 response, which is in agreement with the type I response found in acute Q fever patients. Most other inflammatory markers had increased levels in the acute and vascular chronic Q fever groups, with levels in the vascular chronic patients either similar to the acute Q fever group or intermediate between the acute and past resolved Q fever groups. These, together, indicate a weaker overall inflammatory response in vascular chronic Q fever patients.
The main finding, however, of this study was that patients with vascular chronic Q fever display reduced IFN-γ production. This could indicate a potential cause of the inadequate immune response postulated in patients with chronic Q fever (3, 4). As is shown in mice, IFN-γ seems to play a key role in the early control of Q fever. In mice, low levels of IFN-γ resulted in aberrant macrophage activation and the lack of killing of C. burnetii (11).
One of the cytokines consistently induced in both acute and vascular chronic Q fever is IL-18. This cytokine, formerly known as IFN-γ-inducing factor, is mainly produced by cells of the macrophage lineage with a wide range of biologic activities that include the induction of IFN-γ production in T cells and NK cells (21, 22). Increased circulating IL-18 levels have been reported in viral infections, in which context recently markedly elevated levels during acute Epstein-Barr virus (EBV) infections and EBV-associated diseases were shown (23). Phylogenetically, the nearest relatives to C. burnetii are the Legionellaceae. These microorganisms share similar intracellular lifestyles and utilize common genes to infect alveolar macrophages (24). It has been shown previously that IL-18 plays a key role in modulating induction of IFN-γ in the murine lung in response to Legionella pneumophila infection (25). The role of IL-18 in immunity against C. burnetii has, however, not been investigated.
In our study, IL-18 serum levels were increased at comparable levels in both acute and vascular chronic Q fever patients compared to past resolved Q fever patients, a finding which was validated in the second larger serum analysis. Thus, it appears that although one of the cytokines involved in activation of IFN-γ production was present, this did not result in IFN-γ production in vascular chronic Q fever patients, whereas acute patients readily produced IFN-γ.
In addition to IL-18, also IL-12p70, which is known to induce the production of IFN-γ by NK cells and T cells (26), and IL-23, another cytokine involved in the Th1 immune response by activating inflammatory cells that is required for induction of chronic inflammation and granuloma formation (27), were measured in the initial biomarker panel. The latter did not show any signs of differential expression among the three patient groups. (FDR-corrected P value, 0.98). IL-12p70 showed an increased level in the acute Q fever group compared to the other two groups. However, it should be noted that this difference was only borderline significant (FDR = 8%) and that this cytokine was not consistently present within the detectable range (8 times in acute Q fever patients, 4 times in vascular chronic Q fever patients, and 4 times in past resolved Q fever patients), so some caution needs to be taken in interpreting these findings.
Taken together, IL-18, IL-12p70, and IL-23 did not show clearly significant differences in serum levels between acute and vascular chronic Q fever patients, and therefore, these cytokines cannot be held responsible for the observed difference in IFN-γ levels in patients with acute and vascular chronic Q fever.
Furthermore, we have found significantly lower levels of IP-10 in patients with vascular chronic Q fever than in patients with acute Q fever. It is known that inflammation is associated with secretion of IP-10 from leukocytes like neutrophils, eosinophils, and monocytes as well as epithelial, endothelial, and stromal cells and keratinocytes in response to IFN-γ (28). For instance, it has been shown that impaired IP-10 production led to susceptibility to L. pneumophila infection (29). Apparently, induction of IFN-γ and IP-10 is reduced in vascular chronic Q fever, even though the signal from the IFN-γ- and IP-10-inducing cytokine IL-18 is present, indicating a dysregulation of the type I cytokine response. A hypothesis that would fit with these observations is that IL-18 responsiveness is reduced in (vascular) chronic Q fever, at least as far as the response to C. burnetii is involved.
One of the cytokines that is implicated in modulating IL-18 responsiveness is TGF-β, which is a pleiotropic cytokine produced by T cells and macrophages (30). Under inflammatory conditions, in the presence of other cytokines such as IL-6 and IL-4, TGF-β can promote further inflammation and augment autoimmune conditions (31). In mice, TGF-β interferes with IL-18-induced IFN-γ production by reducing the number of IL-18 receptors at the cell surface and, thereby, preventing induction of T-bet expression (32). In addition, TGF-β reduces IFN-γ production in response to the engagement of the Fc receptor CD16 on NK cells (19). However, by enhancing survival of memory CD8+ T cells, once an antigen is detected, it increases the production of IL-17 and IFN-γ (33). Furthermore, it has been shown that TGF-β inhibits the killing of the intracellular microorganisms Trypanosoma cruzi, Mycobacterium avium, and Mycobacterium tuberculosis. In supernatants of unstimulated peripheral blood mononuclear cells from patients with Q fever endocarditis, TGF-β levels were significantly increased compared with control supernatants (34).
We showed that the levels of TGF-β in patients with vascular chronic Q fever were significantly lower than those in patients with acute Q fever. Our results seem to be contrary to what could be expected based on previous research. However, our findings are shown to be highly significant (FDR-corrected P value, <10−5). A possible explanation may be that in vivo TGF-β is acting locally, which might explain differences between literature findings based on cytokine production by specific types of immune cells under controlled conditions as opposed to systemic levels of serum cytokines measured in our study.
A difference between local and systemic response in chronic Q fever can most likely be attributed to perturbed responses by C. burnetii-infected macrophages. Such a defect in local macrophage cellular immunity points toward a change in macrophage M1/M2 polarization. A relation between macrophage polarization and disease course has, for example, been observed in leprosy (35) as well as Whipple's disease (36), both caused by intracellular pathogens.
Indeed, literature data indicate that C. burnetii activates an atypical M2 program in macrophages, which may account for the capacity of C. burnetii to survive in macrophages (37). C. burnetii-stimulated macrophages produced high levels of TGF-β1 (37), and we hypothesize that this may account for the increase in serum TGF-β found in acute Q fever patients. The lack of systemic increase in serum TGF-β in vascular chronic Q fever patients might be explained by such responses being confined to sites associated with chronic Q fever like aneurysms, or cardiac valves in the case of Q fever endocarditis. Interestingly, it has been reported that patients with cardiac valve lesions have elevated levels of apoptotic leukocytes in the circulation and that apoptotic cell binding induces a nonprotective M2 program in C. burnetii-stimulated monocytes and macrophages (38). Patients with cardiovascular disease may, therefore, have a skewed immune system as a consequence of their existing illness, which creates a suitable environment for C. burnetii replication and the pathogenesis of chronic Q fever (38). This provides an explanation for the incidence of chronic Q fever among cardiovascular patients. Interestingly, in vitro data show that IFN-γ redirects monocytes and macrophages toward an M1 program and leads to reduced counts of C. burnetii (38). Moreover, two isolated case studies have been reported in which IFN-γ was used to treat forms of chronic Q fever (12, 13). This suggests that research into IFN-γ therapy in chronic Q fever patients might be valuable.
The multimarker serum screening approach used in this study allowed us to identify several markers in type I immunity that were affected in vascular chronic Q fever. Despite the significant differences between specific cytokines in acute and vascular chronic Q fever, a limitation of our study is the lack of consecutive determination of these cytokines in subsequent sera. These results could have affected the precise understanding of the immune response in acute and vascular chronic Q fever.
In conclusion, we found significantly lower serum levels of IFN-γ in patients suffering from vascular chronic Q fever, despite the presence of similar IL-18 levels and significantly reduced levels of TGF-β, in comparison to acute Q fever patients. This cytokine profile indicates a dysregulation of the systemic and probably also local type I response to C. burnetii, which could explain the inadequate immune response in these patients.
REFERENCES
- 1.van der Hoek W, Schneeberger PM, Oomen T, Wegdam-Blans MC, Dijkstra F, Notermans DW, Bijlmer HA, Groeneveld K, Wijkmans CJ, Rietveld A, Kampschreur LM, van Duynhoven Y. 2012. Shifting priorities in the aftermath of a Q fever epidemic in 2007 to 2009 in The Netherlands: from acute to chronic infection. Euro Surveill 17:20059 http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId=20059. [PubMed] [Google Scholar]
- 2.Derrick E. 1964. The query fever—the Elkington oration. Qld Health 1:1–20. [Google Scholar]
- 3.Maurin M, Raoult D. 1999. Q fever. Clin Microbiol Rev 12:518–553. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Raoult D, Marrie T, Mege J. 2005. Natural history and pathophysiology of Q fever. Lancet Infect Dis 5:219–226. doi: 10.1016/S1473-3099(05)70052-9. [DOI] [PubMed] [Google Scholar]
- 5.Kampschreur LM, Delsing CE, Groenwold RH, Wegdam-Blans MC, Bleeker-Rovers CP, de Jager-Leclercq MG, Hoepelman AI, van Kasteren ME, Buijs J, Renders NH, Nabuurs-Franssen MH, Oosterheert JJ, Wever PC. 2014. Chronic Q fever in the Netherlands 5 years after the start of the Q fever epidemic: results from the Dutch chronic Q fever database. J Clin Microbiol 52:1637–1643. doi: 10.1128/JCM.03221-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Kampschreur LM, Oosterheert JJ, Hoepelman AI, Lestrade PJ, Renders NH, Elsman P, Wever PC. 2012. Prevalence of chronic Q fever in patients with a history of cardiac valve surgery in an area where Coxiella burnetii is epidemic. Clin Vaccine Immunol 19:1165–1169. doi: 10.1128/CVI.00185-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Hagenaars JC, Wever PC, van Petersen AS, Lestrade PJ, de Jager-Leclercq MG, Hermans MH, Moll FL, Koning OH, Renders NH. 2014. Estimated prevalence of chronic Q fever among Coxiella burnetii seropositive patients with an abdominal aortic/iliac aneurysm or aorto-iliac reconstruction after a large Dutch Q fever outbreak. J Infect 69:154–160. doi: 10.1016/j.jinf.2014.03.009. [DOI] [PubMed] [Google Scholar]
- 8.Meghari S, Capo C, Raoult D, Mege JL. 2006. Deficient transendothelial migration of leukocytes in Q fever: the role played by interleukin-10. J Infect Dis 194:365–369. doi: 10.1086/505227. [DOI] [PubMed] [Google Scholar]
- 9.Hagenaars JC, Koning OH, van den Haak RF, Verhoeven BA, Renders NH, Hermans MH, Wever PC, van Suylen RJ. 2014. Histological characteristics of the abdominal aortic wall in patients with vascular chronic Q fever. Int J Exp Pathol 95:282–289. doi: 10.1111/iep.12086. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Murray HW. 1988. Interferon-gamma, the activated macrophage, and host defense against microbial challenge. Ann Intern Med 108:595–608. doi: 10.7326/0003-4819-108-4-595. [DOI] [PubMed] [Google Scholar]
- 11.Andoh M, Zhang G, Russell-Lodrigue KE, Shive HR, Weeks BR, Samuel JE. 2007. T cells are essential for bacterial clearance, and gamma interferon, tumor necrosis factor alpha, and B cells are crucial for disease development in Coxiella burnetii infection in mice. Infect Immun 75:3245–3255. doi: 10.1128/IAI.01767-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Neth OW, Falcon D, Peromingo E, Soledad Camacho M, Rodriguez-Gallego C, Obando I. 2011. Successful management of chronic multifocal Q fever osteomyelitis with adjuvant interferon-gamma therapy. Pediatr Infect Dis J 30:810–812. doi: 10.1097/INF.0b013e31821487f5. [DOI] [PubMed] [Google Scholar]
- 13.Morisawa Y, Wakiguchi H, Takechi T, Kurashige T, Nagaoka H. 2001. Intractable Q fever treated with recombinant gamma interferon. Pediatr Infect Dis J 20:546–547. doi: 10.1097/00006454-200105000-00018. [DOI] [PubMed] [Google Scholar]
- 14.Dellacasagrande J, Capo C, Raoult D, Mege JL. 1999. IFN-gamma-mediated control of Coxiella burnetii survival in monocytes: the role of cell apoptosis and TNF. J Immunol 162:2259–2265. [PubMed] [Google Scholar]
- 15.Schneeberger PM, Hermans MH, van Hannen EJ, Schellekens JJ, Leenders AC, Wever PC. 2010. Real-time PCR with serum samples is indispensable for early diagnosis of acute Q fever. Clin Vaccine Immunol 17:286–290. doi: 10.1128/CVI.00454-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Wegdam-Blans MC, Kampschreur LM, Delsing CE, Bleeker-Rovers CP, Sprong T, van Kasteren ME, Notermans DW, Renders NH, Bijlmer HA, Lestrade PJ, Koopmans MP, Nabuurs-Franssen MH, Oosterheert JJ, Dutch Q Fever Consensus Group . 2012. Chronic Q fever: review of the literature and a proposal of new diagnostic criteria. J Infect 64:247–259. doi: 10.1016/j.jinf.2011.12.014. [DOI] [PubMed] [Google Scholar]
- 17.Benjamini Y, Hochberg Y. 1995. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Series B Methodol 57:289–300. [Google Scholar]
- 18.Dufour JH, Dziejman M, Liu MT, Leung JH, Lane TE, Luster AD. 2002. IFN-gamma-inducible protein 10 (IP-10; CXCL10)-deficient mice reveal a role for IP-10 in effector T cell generation and trafficking. J Immunol 168:3195–3204. doi: 10.4049/jimmunol.168.7.3195. [DOI] [PubMed] [Google Scholar]
- 19.Trotta R, Dal Col J, Yu J, Ciarlariello D, Thomas B, Zhang X, Allard J II, Wei M, Mao H, Byrd JC, Perrotti D, Caligiuri MA. 2008. TGF-beta utilizes SMAD3 to inhibit CD16-mediated IFN-gamma production and antibody-dependent cellular cytotoxicity in human NK cells. J Immunol 181:3784–3792. doi: 10.4049/jimmunol.181.6.3784. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Kremers MN, Janssen R, Wielders CC, Kampschreur LM, Schneeberger PM, Netten PM, de Klerk A, Hodemaekers HM, Hermans MH, Notermans DW, Wever PC. 2014. Correlations between peripheral blood Coxiella burnetii DNA load, interleukin-6 levels, and C-reactive protein levels in patients with acute Q fever. Clin Vaccine Immunol 21:484–487. doi: 10.1128/CVI.00715-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Dinarello CA. 2000. Interleukin-18, a proinflammatory cytokine. Eur Cytokine Netw 11:483–486. [PubMed] [Google Scholar]
- 22.Okamura H, Tsutsi H, Komatsu T, Yutsudo M, Hakura A, Tanimoto T, Torigoe K, Okura T, Nukada Y, Hattori K, Akita K, Namba M, Tanabe F, Konishi K, Fukuda S, Kurimoto M. 1995. Cloning of a new cytokine that induces IFN-gamma production by T cells. Nature 378:88–91. doi: 10.1038/378088a0. [DOI] [PubMed] [Google Scholar]
- 23.van de Veerdonk FL, Wever PC, Hermans MH, Fijnheer R, Joosten LA, van der Meer JW, Netea MG, Schneeberger PM. 2012. IL-18 serum concentration is markedly elevated in acute EBV infection and can serve as a marker for disease severity. J Infect Dis 206:197–201. doi: 10.1093/infdis/jis335. [DOI] [PubMed] [Google Scholar]
- 24.Fields BS, Benson RF, Besser RE. 2002. Legionella and Legionnaires' disease: 25 years of investigation. Clin Microbiol Rev 15:506–526. doi: 10.1128/CMR.15.3.506-526.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Brieland JK, Jackson C, Hurst S, Loebenberg D, Muchamuel T, Debets R, Kastelein R, Churakova T, Abrams J, Hare R, O'Garra A. 2000. Immunomodulatory role of endogenous interleukin-18 in gamma interferon-mediated resolution of replicative Legionella pneumophila lung infection. Infect Immun 68:6567–6573. doi: 10.1128/IAI.68.12.6567-6573.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Bastos KR, Barboza R, Sardinha L, Russo M, Alvarez JM, Lima MR. 2007. Role of endogenous IFN-gamma in macrophage programming induced by IL-12 and IL-18. J Interferon Cytokine Res 27:399–410. doi: 10.1089/jir.2007.0128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Langrish CL, McKenzie BS, Wilson NJ, de Waal Malefyt R, Kastelein RA, Cua DJ. 2004. IL-12 and IL-23: master regulators of innate and adaptive immunity. Immunol Rev 202:96–105. doi: 10.1111/j.0105-2896.2004.00214.x. [DOI] [PubMed] [Google Scholar]
- 28.Liu M, Guo S, Hibbert JM, Jain V, Singh N, Wilson NO, Stiles JK. 2011. CXCL10/IP-10 in infectious diseases pathogenesis and potential therapeutic implications. Cytokine Growth Factor Rev 22:121–130. doi: 10.1016/j.cytogfr.2011.06.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Lettinga KD, Weijer S, Speelman P, Prins JM, Van Der Poll T, Verbon A. 2003. Reduced interferon-gamma release in patients recovered from Legionnaires' disease. Thorax 58:63–67. doi: 10.1136/thorax.58.1.63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Mims CA, Dockrell H, Goering R, Roitt I, Wakelin D, Zuckerman M. 2004. Medical microbiology, 3rd ed, p 106–110. Mosby, Edinburgh, United Kingdom. [Google Scholar]
- 31.Sanjabi S, Zenewicz LA, Kamanaka M, Flavell RA. 2009. Anti-inflammatory and pro-inflammatory roles of TGF-beta, IL-10, and IL-22 in immunity and autoimmunity. Curr Opin Pharmacol 9:447–453. doi: 10.1016/j.coph.2009.04.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Koutoulaki A, Langley M, Sloan AJ, Aeschlimann D, Wei XQ. 2010. TNFalpha and TGF-beta1 influence IL-18-induced IFNgamma production through regulation of IL-18 receptor and T-bet expression. Cytokine 49:177–184. doi: 10.1016/j.cyto.2009.09.015. [DOI] [PubMed] [Google Scholar]
- 33.Filippi CM, Juedes AE, Oldham JE, Ling E, Togher L, Peng Y, Flavell RA, von Herrath MG. 2008. Transforming growth factor-beta suppresses the activation of CD8+ T-cells when naive but promotes their survival and function once antigen experienced: a two-faced impact on autoimmunity. Diabetes 57:2684–2692. doi: 10.2337/db08-0609. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Capo C, Zaffran Y, Zugun F, Houpikian P, Raoult D, Mege JL. 1996. Production of interleukin-10 and transforming growth factor beta by peripheral blood mononuclear cells in Q fever endocarditis. Infect Immun 64:4143–4147. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Bleharski JR, Li H, Meinken C, Graeber TG, Ochoa MT, Yamamura M, Burdick A, Sarno EN, Wagner M, Rollinghoff M, Rea TH, Colonna M, Stenger S, Bloom BR, Eisenberg D, Modlin RL. 2003. Use of genetic profiling in leprosy to discriminate clinical forms of the disease. Science 301:1527–1530. doi: 10.1126/science.1087785. [DOI] [PubMed] [Google Scholar]
- 36.Desnues B, Raoult D, Mege JL. 2005. IL-16 is critical for Tropheryma whipplei replication in Whipple's disease. J Immunol 175:4575–4582. doi: 10.4049/jimmunol.175.7.4575. [DOI] [PubMed] [Google Scholar]
- 37.Benoit M, Barbarat B, Bernard A, Olive D, Mege JL. 2008. Coxiella burnetii, the agent of Q fever, stimulates an atypical M2 activation program in human macrophages. Eur J Immunol 38:1065–1070. doi: 10.1002/eji.200738067. [DOI] [PubMed] [Google Scholar]
- 38.Benoit M, Ghigo E, Capo C, Raoult D, Mege JL. 2008. The uptake of apoptotic cells drives Coxiella burnetii replication and macrophage polarization: a model for Q fever endocarditis. PLoS Pathog 4:e1000066. doi: 10.1371/journal.ppat.1000066. [DOI] [PMC free article] [PubMed] [Google Scholar] [Retracted]