Investigational Analysis Reveals a Potential Role for Neutrophils in Giant-Cell Arteritis Disease ProgressionNovelty and Significance
Rationale: Giant-cell arteritis (GCA) is a large-vessel vasculitis characterized by immune cell infiltration, yet the potential involvement of neutrophils has rarely been studied.
Objective: We investigated whether alterations in neutrophil reactivity occurred in the pathogenesis of GCA or during its clinical management with a canonical glucocorticoid dose regimen during a 6-month period.
Methods and Results: Blood samples were taken within 48 hours of therapy commencement and at weeks 1, 4, and 24 after glucocorticoid dose. Flow cytometric analysis revealed 3 distinct neutrophil populations and phenotypes. Within 48 hours of steroid treatment, neutrophils displayed an AnxA1hiCD62LloCD11bhi phenotype, whereas week 1 neutrophils were AnxA1hiCD62LloCD11blo and displayed minimal adhesion to endothelial monolayers under flow, and week 24 (ie, lowest glucocorticoid dose) neutrophils were AnxA1hiCD62LhiCD11bhi with increased endothelial adhesion under flow. Week 24 plasma analyses showed high levels of C-X-C motif chemokine ligand 5, interleukin (IL) 8, IL-17, and IL-6. Importantly, comparison of week 1 and week 24 samples revealed a suppressive neutrophil effect on T-cell proliferation at the former time point only. Finally, in vitro incubation of naive neutrophils with concentrations of IL-6 and IL-17 quantified in GCA plasma at weeks 1 and 24 replicated this differential modulation of lymphocyte proliferation.
Conclusions: This translational study highlights a novel clinical manifestation of GCA, with evidence for a neutrophil component and an escaped proinflammatory phenotype when glucocorticoid therapy is tapered. These results indicate potential involvement of neutrophils in GCA pathogenesis.
Giant-cell arteritis (GCA) is a systemic inflammatory disease associated with focal granulomatous panarteritis predominantly involving extracranial branches of the aorta. The most feared complication is critical ischemia leading to anterior ischemic optic neuropathy and permanent sight loss (≈20% of cases). Clinical management of GCA is with immediate high-dose glucocorticoids started on suspicion, with tapering over several months1 depending on the clinical response of patients to treatment.
GCA is considered a T helper (Th)-1 and Th17 T-cell–mediated disease. Interferon-γ (IFN-γ)–secreting Th1 cells are relatively glucocorticoid-resistant and largely responsible for chronic disease activity. In contrast, increased plasma interleukin (IL) 17 levels and Th17 cell arterial wall infiltration are steroid-sensitive.2 It is noteworthy that a pivotal property of IL-17 (referred to herein as IL-17) is neutrophil activation, yet studies of neutrophil phenotype in GCA pathology are scant.3,4
Detailed Methods section is given in the online-only Data Supplement.
This study was conducted in accordance with the Declaration of Helsinki. Patients gave informed consent, and samples were collected from Southend University and Hammersmith Hospitals (protocol approved by the East London and the City Local Research Ethics Committee; see Table for patient demographics).
A whole-blood staining protocol was performed as described.5
Flow Chamber Assay
Human umbilical vein endothelial cells (ethics as stated) were stimulated with tumor necrosis factor-α (10 ng/mL, 4 hours). Blood neutrophils were isolated via density gradient,6 and analysis of total cell capture and rolling and firmly adherent neutrophils was performed off-line.6
Determination of Plasma Cytokine Levels
Plasma prepared from blood of patients was tested for C-X-C motif chemokine ligand 5, IL-8, IL-6, sIL-6R, IL-17, and IFN-γ by enzyme-linked immunosorbent assay.
Either paired Student t test (for 48 hours and week 1 poststeroid samples) or 1-way repeated-measures ANOVA for longitudinal analyses was performed. Statistical differences were accepted if P<0.05.
Longitudinal Changes in GCA Neutrophil Phenotype and Circulating Numbers
Neutrophilia was observed within 48 hours of prednisone commencement (≈5×106 neutrophils/mL) and at 1 and 24 weeks after steroids (≈4×106/mL; Table) compared with both patient control groups (≈1.2×106/mL; P<0.05). Longitudinal expression of the glucocorticoid-regulated protein annexin A1 (AnxA1) revealed high neutrophil surface expression as early as 48 hours after therapy. At week 1, ≈3- to 4-fold increase higher than that for osteroarthritis and high-dose steroid control group patients (Figure 1A) was found; this peak declined steadily by week 4. However, an increase was detected again at week 24, corresponding to glucocorticoid tapering (Figure 1A). This is a nongenomic response because no significant difference in AnxA1 mRNA was observed across the groups (Online Figure II). Expression of the AnxA1 receptor lipoxin A4 receptor/formyl-peptide receptor 2 did not change at any time point (Figure 1B).
We next analyzed the longitudinal expression of CD62L and CD11b. GCA neutrophils expressed low levels of CD62L compared with the 2 control groups, with reduction evident as early as 48 hours after therapy commencement. Values began to increase from week 4, with higher cellular expression by week 24 (Figure 1A and 1B; P<0.05). Although CD11b expression was high at 48 hours, it rapidly decreased by week 1 (Figure 1A), with no significant difference from control group patients. Again, there was a 3-fold selective increase in CD11b on GCA neutrophils at week 24 (Figure 1A).
Therefore, 48 hours post steroid treatment, neutrophils displayed an AnxA1hiCD62LloCD11bhi phenotype, 1 week post steroid CD16+ neutrophils displayed an AnxA1hiCD62LloCD11blo phenotype, and 24 weeks post steroid, CD16+ neutrophils displayed an AnxA1hiCD62LhiCD11bhi phenotype (Online Figure III). These phenotypes correlated with neutrophil–endothelial cell interactions under flow. There was a 3-fold decrease in GCA–neutrophil interactions at week 1 (compared with osteroarthritis) as a result of marked attenuation in rolling and adhesion (Figure 1C). In contrast, week 24 GCA neutrophils exhibited significantly increased capture and adhesion (Figure 1C).
Suppressor Neutrophil Reduction at 24 Weeks After Glucocorticoid
Plasma analyses of the chemokines IL-8 and C-X-C motif chemokine ligand 5 indicated a significant increase within 48 hours of steroid commencement, declining at 1 and 4 weeks after steroid. However, there was a significant augmentation of both neutrophil chemoattractants at 24 weeks after steroid, with levels similar to that observed at 48 hours (Figure 2A). Plasma IFN-γ levels were significantly high 48 hours after therapy compared with control groups and 1 week after steroid, but no further significant changes in IFN-γ levels were observed during the rest of the time course (Figure 2A). In contrast to IFN- γ, IL-6 levels increased in patients with GCA 24 weeks after steroid, which coincided with an increase in IL-17 (Figure 2A).
A recent study described a novel CD16brightCD62Ldim population of neutrophils that suppress T-cell proliferation.7 Reanalysis of our data revealed a significant reduction in CD16hiCD62Llo events (equivalent to CD16bightCD62Ldim suppressors) at week 24 compared with week 1 (Figure 2B), with higher levels of AnxA1 compared with week 1 (Figure 2B). Incubation of neutrophils from healthy donors with concentrations of IL-17 and IL-6 measured in the plasma of patients with GCA at weeks 1 and 24 replicated the difference in the neutrophil suppressor population (Figure 3A) and their ability to suppress T-cell proliferation: neutrophils treated with week 1 levels of the 2 cytokines were able to effectively suppress T-cell proliferation but not when treated with week 24 levels of IL-6 and IL-17 (Figure 3B).
Finally, we quantified chemokine receptor expression on T cells after coculture with neutrophils treated with IL-6 and IL-17 in combination. Whereas C-X-C motif chemokine receptor-4 expression did not significantly change (data not shown), a 2-fold to 3-fold increase in T-cell C-X-C motif chemokine receptor-3 expression was observed on coculture with week 24, but not week 1, concentrations of IL-6 and IL-17 (Figure 3C).
The recent identification of IL-17–producing T cells in patients with GCA2 suggests a potential role for neutrophils because this cytokine promotes bone marrow mobilization as well as activation and trafficking of neutrophils into perivascular tissue,8 yet there is scant evidence for the role of neutrophils in GCA.3,4 We monitored neutrophil function and phenotypes during a canonical 6-month glucocorticoid treatment and provide evidence for the role of neutrophil phenotypic changes in GCA pathology.
Our initial interest in neutrophils and GCA stemmed from the neutrophilia typically seen in patients using steroid therapy (Table). Persistent neutrophilia observed at 24 weeks (a time when most patients have achieved clinical remission) suggested existence of a subclinical vascular inflammatory state that might explain disease reemergence. To test this hypothesis, we analyzed neutrophil phenotypes as early as 48 hours after steroids and at 1, 4, and 24 weeks after therapy. GCA neutrophils display a classically activated CD16hiAnxA1hiCD62LloCD11bhi phenotype at 48 hours. This phenotype comes under rapid control within 1 week of treatment, despite stable neutrophilia, with a CD16hiAnxA1hiCD62LloCD11blo signature. These neutrophils were hyporeactive, as confirmed by minimal interaction with an inflamed endothelial monolayer under flow conditions similar to the CD16brightCD62Ldim neutrophil reactivity previously described.9 This neutrophil phenotype is similar to that reported after estrogen treatment.5 In stark contrast, neutrophils at 24 weeks after glucocorticoids exhibited a CD16hiAnxA1hiCD62LhiCD11bhi phenotype correlating with marked adhesion to endothelial monolayers.
We initially postulated that the neutrophil phenotype observed at week 1 was a direct consequence of high-dose steroid therapy because AnxA1 is glucocorticoid regulated10 and there is evidence for glucocorticoid-induced CD62L shedding.11 However, this protective neutrophil phenotype was specific to steroid-treated GCA because cells from the high-dose steroid control group did not display the same hyporeactive phenotype. Furthermore, despite the high AnxA1 levels in week 24 neutrophils, there was still significantly increased firm adhesion, suggesting either a defective protein5,6 or an inability to counteract the cellular hyperactivity. The molecular mechanisms behind AnxA1 mobilization at weeks 1 and 24 warrant further investigation.
The emerging hypothesis of a neutrophil component in GCA was confirmed by cytokine measurements: the highest circulating levels of C-X-C motif chemokine ligand 5 and IL-8, together with IL-6 and IL-17, were observed at week 24 (Figure 2). It should also be noted that levels of both neutrophil chemoattractants and IFN-γ were significantly augmented within 48 hours of steroid commencement compared with high-dose steroid control group patients who had been using steroid therapy for a similar length of time. Therefore, taken together, this neutrophil component seems to be specific to GCA. Increased IL-17 expression after therapeutic control is congruent with a model whereby T-cell/neutrophil crosstalk is key to GCA progression, possibly exacerbating vascular inflammation.
Near completion of this study, 2 neutrophil phenotypes were reported in the blood of volunteers infused with lipopolysaccharide with a novel suppressor pool, identified as CD16brightCD62LdimCD11bbright, able to dampen T-cell activation.7 This suppressor pool was detected in our week 1 samples and almost halved by week 24 after steroids, making the present GCA study the first to identify this neutrophil subset in disease. Combining our functional data with those of Pillay et al,7 we hypothesize that week 24 GCA neutrophils are unable to suppress T-cell responses, favoring loss of glucocorticoid control and, in time, re-emergence of vascular inflammation.
Intriguingly, we could reproduce in vitro the neutrophil dichotomy using concentrations of IL-6 and IL-17 equivalent to those measured in GCA plasma samples. The reduction in suppressor neutrophils after treatment with these cytokines correlated with attenuated inhibition of lymphocyte proliferation. Furthermore, analysis of T cells cocultured with these neutrophils demonstrated high levels of the chemokine receptor C-X-C motif chemokine receptor-3, an important determinant of Th1 and Th17 cell trafficking to inflamed tissues.12,13 C-X-C motif chemokine receptor-3+ T cells have been identified in the temporal arteries of patients with GCA.14
In conclusion, we report potential involvement of neutrophils in GCA pathogenesis and relapse. Our data support the concept that the disease process is incompletely controlled by glucocorticoid therapy because tapering leads to loss of the neutrophil suppressor subset. This, in turn, may be the prelude to lymphocyte proliferation and disease relapse with an associated increased risk of vascular complications (Figure 4). Thus, monitoring neutrophil phenotype might inform disease status, predict risk of relapse, and facilitate steroid tapering in individual patients.
We thank Dr Neil Dufton for Figure 4.
Sources of Funding
This work was supported by British Heart Foundation (PG/09/060) and partly by the Wellcome Trust (086867/Z/08) and Imperial College Biomedical Resource Centre.
In September 2013, the average time from submission to first decision for all original research papers submitted to Circulation Research was 13.2 days.
The online-only Data Supplement is available with this article at http://circres.ahajournals.org/lookup/suppl/doi:10.1161/CIRCRESAHA.114.301374/-/DC1.
- Nonstandard Abbreviations and Acronyms
- annexin A1
- giant-cell arteritis
- Received March 14, 2013.
- Revision received October 21, 2013.
- Accepted October 21, 2013.
- © 2013 American Heart Association, Inc.
- Dasgupta B,
- Borg FA,
- Hassan N,
- Alexander L,
- Barraclough K,
- Bourke B,
- Fulcher J,
- Hollywood J,
- Hutchings A,
- James P,
- Kyle V,
- Nott J,
- Power M,
- Samanta A
- Deng J,
- Younge BR,
- Olshen RA,
- Goronzy JJ,
- Weyand CM
- Chatelain D,
- Duhaut P,
- Schmidt J,
- Loire R,
- Bosshard S,
- Guernou M,
- Pellet H,
- Piette JC,
- Sevestre H,
- Ducroix JP
- Nadkarni S,
- Cooper D,
- Brancaleone V,
- Bena S,
- Perretti M
- Hayhoe RP,
- Kamal AM,
- Solito E,
- Flower RJ,
- Cooper D,
- Perretti M
- Kamp VM,
- Pillay J,
- Lammers JW,
- Pickkers P,
- Ulfman LH,
- Koenderman L
- Burton JL,
- Kehrli ME Jr.,
- Kapil S,
- Horst RL
- Bonecchi R,
- Bianchi G,
- Bordignon PP,
- D’Ambrosio D,
- Lang R,
- Borsatti A,
- Sozzani S,
- Allavena P,
- Gray PA,
- Mantovani A,
- Sinigaglia F
- Brühl H,
- Vielhauer V,
- Weiss M,
- Mack M,
- Schlöndorff D,
- Segerer S
Novelty and Significance
What Is Known?
Giant-cell arteritis (GCA) is a large-vessel disease characterized by inflammation of large and medium arteries, commonly the temporal artery.
GCA is an important cause of acute vision loss.
GCA is currently treated with high-dose glucocorticoid therapy, which, on improvement in clinical symptoms, tapers over 6 months.
T-cell activation and recruitment into the involved artery are implicated in the pathogenesis of GCA.
What New Information Does This Article Contribute?
Analyses of circulating neutrophils showed activation during the early stage of the disease.
Administration of high-dose glucocorticoids controls neutrophil activation.
Within 48 hours after treatment with high-dose steroids, neutrophils showed minimal adhesion to endothelial monolayers.
However, on tapering of steroids after 6 months of therapy, neutrophils showed increased endothelial adhesion under flow conditions.
Similarly, after 6 months of therapy, the suppressor phenotype of neutrophil is less expressed, and there is re-emergence of signs of neutrophil activation coupled to higher levels of proinflammatory cytokines.
Previous studies have shown that T cells are involved in the pathogenesis of GCA. However, mechanisms regulating their hyper-reactivity are poorly understood. We examined the profile and activity of the circulating leukocytes in patients with GCA longitudinally during a 6-month glucocorticoid therapy period. Administration of high-dose steroids was associated with a low inflammatory profile of the circulating neutrophil, indicated by the presence of a larger proportion of suppressor neutrophils that inhibited T-cell proliferation. However, tapering of steroids was associated with clear signs of neutrophil activation and high levels of 2 proinflammatory cytokines, interleukin 6 and interleukin 17A. These changes were associated with reduced proportion of the neutrophils with the suppressor phenotype. Collectively, these results suggest a potential involvement of neutrophils in the pathogenesis of GCA and might indicate novel interactions between neutrophils and T cells. High doses of glucocorticoids control these interactions favorably. However, tapering of the steroid dose is associated with subliminal signs of vascular inflammation, possibly conducive to subsequent disease relapse and the associated vascular complications. Monitoring neutrophil phenotype might provide novel information on disease status, risk of relapse, and steroid dosage.