Introduction

Immunoglobulin A vasculitis (IgAV) is a small-vessel, immune complex leukocytoclastic vasculitis, characterized by dominant IgA deposits in vascular walls [1]. It represents the most common primary vasculitis in the pediatric population [2, 3]; however, it is also frequent in adults, as recently shown [4].

Clinical signs and symptoms of the disease include cutaneous purpura, joint, gastrointestinal tract (GIT), and renal involvement; however, they differ greatly depending on the age of the patients [5, 6]. Adults develop systemic involvement more frequently compared to children, with greater risk of severe bowel ischemia and chronic renal failure [7,8,9,10,11].

There are currently no diagnostic criteria for IgAV. The updated classification criteria for pediatric IgAV (EULAR/PRINTO/PRES classification criteria) [12], performed well also in adult IgAV [13]. Yet, the final diagnosis leans on histological characterization of IgAV [1, 14]. However, not all the expected significant changes are present in biopsy tissue, neither is IgA deposition pathognomonic for IgAV [14]. In approximately 10–20% of patients, GIT or joint symptoms may occur before the onset of skin purpura and setting an appropriate diagnosis of IgAV is difficult, when there is an atypical presentation [14, 15].

It would be advantageous to find a serological biomarker or group of analytes, which could support clinicians in their diagnosis in such cases. Moreover, there is a need of prognostic markers that would predict patients who are at risk of developing severe GIT and renal involvement, as well as complications during disease progression in order to facilitate appropriate therapeutic management of these patients [14].

The aims of this study were to evaluate serum levels of cytokines and acute phase proteins as well as determine the distribution of neutrophils, lymphocytes, and monocytes, together with neutrophil surface expression of L-selectin (CD62L) and integrin αM (CD11b), among adult IgAV patients.

Materials and methods

Human subjects

We included 62 adult, treatment-naïve IgAV patients, with histologically proven disease [12, 13]. For comparison, sera samples of 53 and peripheral blood of 15 healthy blood donors (HBDs) were obtained. All samples were fully anonymized, before we assessed them, and all subjects signed informed consent. The study was approved by the Slovene National Medical Ethics Committee (#99/04/15 and #65/01/17).

Data collection

The medical records from IgAV patients were analyzed and the following was recorded: age, gender, body mass index, smoking status, duration of clinical symptoms, and signs before sampling, concurrent infections, past malignancy, clinical signs and symptoms, Birmingham vasculitis activity score (BVAS), and results of routine laboratory testing.

Laboratory parameters

Erythrocyte sedimentation rate and concentration of C-reactive protein, complete blood count with differential, basic biochemistry panels including electrolytes, creatinine, urea, serum protein electrophoresis, serum immunoglobulin (Ig) A, M, G levels, and urine analysis were determined.

Cytokine and serum amyloid A measurement

Concentrations of IL-1β, IL-2, IL-8, IL-9, IL-10, IL-17A, IL-23, and TNF-α were measured in sera of IgAV patients and HBDs by AtheNa Multi-Lyte (Luminex xMAP Technology) using human pre-mixed multi-analyte kits (R&D Systems). IL-6 sera proteins levels were measured by ELISA (Invitrogen). The concentrations of serum amyloid A (SAA) were determined using immunonephelometry (BN Prospec System, Siemens). All sera samples were collected following blood clotting and centrifugation at 3000×g for 5 min, aliquoted, and stored at −20 °C until ready for further analysis.

Flow cytometry

Venous blood was drawn from IgAV patients and HBDs and collected into heparin-containing tubes. Whole blood immunophenotyping was performed using 7-Color Immunphenotyping kit (Miltenyi Biotec). Briefly, 100 μl of whole blood was incubated with 10 μl immunophenotyping reagent for 10 min in the dark, at 4 °C. After incubation, erythrocytes were lysed using Red Blood Lysing Solution (Miltenyi Biotec). Neutrophil phenotyping was performed in 50 μl of whole blood, incubated for 30 min at 4 °C in the dark with mouse anti-human antibodies to CD16 (conjugated to PE, clone eBioCB16, eBioscience), CD62L (conjugated to PE-Cyanine 5, clone DREG56, eBioscience), and CD11b (conjugated to APC, clone ICRF44, eBioscience). After incubation, samples were lysed, using Whole Blood Lysing Reagent Kit (Beckman Coulter). All samples were analyzed using flow cytometer MACSQuant Analyzer 10 (Miltenyi Biotec). Analysis of flow cytometry data was performed using MACSQquantify (Analysis Software version 2.8) and FlowLogic (Flow Cytometry Analysis Package, version 700.0a).

Statistical analysis

The normality of distribution was determined by Shapiro–Wilk test. Due to non-normal distribution of the data, summary statistics were expressed as medians and 25–75th percentiles (Q25–Q75), and nonparametric tests were performed. Mann–Whitney U test was used to compare continuous variables between IgAV patients and HBDs, or between IgAV patients with/without different clinical symptoms and signs. Spearman’s rank correlation was calculated to measure the correlation between parameters. All statistical analysis was performed using SPSS statistical software package version 22.0 and Graph Pad Prism software 5.03. All tests were two-tailed and p values of < 0.05 were regarded as statistically significant.

Results

Characteristics of IgAV patients

Sixty-two IgAV patients (52% males) with median (Q25–Q75) age of 56.0 (38.9–74.8) years were recruited in the study. Patients were included after a median (Q25–Q75) duration of symptoms of 10 (5–19) days. All patients presented with skin purpura (necrotic 45%, generalized 56%). Joint, GIT, and renal involvement were present in 40%, 39%, and 35%, respectively (Table 1).

Table 1 Demographics, clinical, and laboratory data of IgAV patients
Full size table

Routine laboratory findings

Results of routine laboratory tests are present in Table 1. Sera levels of IgA were elevated in 29/62 (47%) of patients above 4 g/l.

Inflammatory cytokine profile

Levels of SAA (12-fold), an acute phase protein with cytokine-like properties, IL-6 (3-fold), IL-8 (2-fold), and TNF-α (2-fold) were statistically significantly elevated in sera of IgAV patients compared to HBDs (Table 2). IL-1β, IL-9, IL-10, and IL-23 were found to be unchanged in IgAV patients compared to HBDs (Table 2) and concentrations of IL-2 and IL-17A could not be determined due to their levels below the limits of detection of this assay.

Table 2 Cytokine profile in sera of adult IgAV patients compared to healthy blood donors
Full size table

Inflammatory cell profile and neutrophil phenotype

In peripheral blood of IgAV patients, there was a higher percentage of CD16+ neutrophils (p < 0.001) and lower percentages of CD3+ (p < 0.001), including CD4+ (p = 0.003) and CD8+ (p = 0.002) T-lymphocytes and CD20+ B-lymphocytes (p = 0.006), as compared to HBDs (Figs. 1 and 2; Supplemental Table 1). IgAV neutrophils showed significantly higher L-selectin (CD62L) surface expression with a median (Q25–Q75) of 86.5 (45.4–107.5) vs. 55.5 (44.3–64.9) p = 0.008 in HBDs (Fig. 2; Supplemental Table 1).

Fig. 1
figure 1

Cellular immunophenotyping in HBD (a) and IgAV patients (b) Median percentages are presented of neutrophils, T-lymphocytes, B-lymphocytes, NK cells, and monocytes in HBD (n = 15) and IgAV patients (n = 15); HBD, healthy blood donors; IgAV, immunoglobulin A vasculitis; NK, natural killer

Full size image
Fig. 2
figure 2

Cell profiles and neutrophil surface protein expression in adult IgAV patients as compared to HBD. Medians with Q25–Q75 are shown for percentages of T-lymphocytes, CD4+ T-lymphocytes, CD8+ T-lymphocytes, B-lymphocytes, NK cells, monocytes, neutrophils, and median fluorescence intensity of CD62L and CD11b expression on CD16+ neutrophils. Numbers of IgAV patients and HBD are indicated in brackets; **p < 0.01; ***p < 0.001. IgAV, patients with immunoglobulin A vasculitis; HBD, healthy blood donors; NK, natural killer cells; MFI, median fluorescence intensity

Full size image

IgAV patients with GIT involvement showed a significantly higher number of leukocytes (p < 0.0001), neutrophils (p = 0.001), and calculated neutrophil to lymphocyte ratio (NLR) (p = 0.02), while showing lower neutrophil CD11b expression (p = 0.005), as compared to IgAV patients without GIT involvement (Fig. 3; Supplemental Table 2).

Fig. 3
figure 3

Number of leukocytes, neutrophils, NLR, and CD11b neutrophil expression in IgAV patients with GIT involvement as compared to patients without GIT involvement. Medians with Q25–Q75 are shown for number of leukocytes a, neutrophils b, ratio between # of neutrophils and lymphocytes c, and median fluorescence intensity for CD11b expression on CD16+ neutrophils d. Numbers of included IgAV patients are indicated in brackets. *p < 0.05; **p < 0.01; ***p < 0.001. GIT, gastrointestinal tract; IgAV, patients with immunoglobulin A vasculitis; NLR, neutrophil to lymphocyte ratio; MFI, median fluorescence intensity

Full size image

NLR positively correlated with SAA levels (r = 0.603, 95% CI = 0.410–0.744, p < 0.0001) and neutrophil CD11b expression showed good positive correlation with sera levels of IgA (r = 0.514, 95% CI = 0.177–0.743, p < 0.01) (Fig. 4).

Fig. 4
figure 4

Correlation between NLR and SAA concentration (a) and neutrophil CD11b expression and IgA levels (b) in IgAV patients. Shown are Spearman correlation coefficient (r), 95% confidence interval, p value, and numbers of included patients. CI, confidence interval; IgA, patients with immunoglobulin A vasculitis; NLR, neutrophil to lymphocyte ratio; MFI, median fluorescence intensity; SAA; serum amyloid A

Full size image

Discussion

IgAV was thought to be mainly the disease of children; however, it can frequently occur in adults [4] whose disease course is usually more severe as compared to children [16]. Age, gender, and extent of purpura seem to forecast outcomes of IgA vasculitis in adults [17], but prompt diagnosis and identification of patients at risk, are a considerable challenge for clinicians, with no clinically useful laboratory biomarkers available to date [14].

Serological reports in IgAV have indicated dysregulated inflammatory cytokine production and elevated acute phase proteins. Namely, increased levels of systemic SAA (36-fold) were previously found in IgAV pediatric patients [18, 19] and associated with GIT manifestations [18]. The current report exposes significantly elevated levels of SAA (12-fold) in the sera of adult IgAV patients, indicating a low–medium inflammatory response.

While Th1-, as well as Th2-related cytokines, have been extensively studied in pediatric IgAV patients [19,20,21,22,23], data available on the adult population are scarce. Berthelot et al. recently found higher concentrations of IL-1β, IL-6, and IL-8 in the sera of adult IgAV patients, whereas levels of IL-10, IL-12p70, and TNF-α were similar between patients and control group [24]. Similarly, another study also reported on higher levels of IL-6 and IL-8 in IgAV patients, which were associated with GIT and renal involvement, respectively [25]. In the current study, we observed increased levels of pro-inflammatory cytokines IL-6, IL-8, and TNF-α in the sera of adult IgAV compared to HBDs; however, there was no association with clinical signs and symptoms.

Together with elevated sera concentrations of IgA and higher percentages of neutrophils, our results support the theory of the central role of IgA, endothelial cells, and neutrophils in IgAV pathogenesis, as recently reviewed in Heineke et al. [26]. Although, the origin and specific antigen targets of IgA remain unknown, these antibodies were found to bind antigens on the surface of endothelial cells [27], such as β2-glycoprotein I [21] and phosphatidylserine [25]. IgA binding to endothelial cells, together with the systemic inflammatory environment (with elevated IL-6 and SAA) promote endothelial activation and production of IL-8 [28, 29], a major chemoattractant for neutrophils leading to their recruitment and activation [26]. Indeed, SAA has been shown to induce paracrine IL-8 via Toll-like receptor 2, synergizing to recruit neutrophils [30]. In addition, increased levels of TNF-α in the sera of IgAV patients could further accelerate inflammation, as this cytokine can enhance binding of antibodies to endothelial cells and further induce IL-8 release [21, 29]. Moreover, IgA bound to Fc receptors (FcαRI) on neutrophils, leads to antibody-dependent cell-mediated cytotoxicity (ADCC) and production of reactive oxygen species. Thus, the cross-talk between IgA, endothelial cells, and neutrophils, even in a low inflammatory environment, can lead to substantial vascular damage [26].

We also found elevated expression of L-selectin (CD62L), an adhesion molecule expressed on neutrophils, that could be involved in early pathological events of IgAV, since it is important for the initial adhesion of neutrophils to vascular surfaces [31]. Indeed, a prominent role for neutrophils, with an escaped proinflammatory phenotype was previously proposed for another systemic vasculitis [32], with both neutrophil markers CD62L and CD11b indicating progression of the disease.

Interestingly, in our study, CD11b expression was elevated in IgAV and showed good positive correlation with IgA levels. CD11b (in complex with CD18 is named Mac-1) represents a predominant β2 integrin expressed on neutrophils. Mac-1 is not only important for firm adhesion of neutrophils to endothelial cells [33], but also plays a role in ADCC by interacting with Fc receptors on neutrophils [34]. This could indicate an independent pathogenic mechanism in IgAV, since increased levels of IgA and CD11b expression could lead to enhanced ADCC and vascular damage. On the other hand, the current report also shows a negative association of neutrophil CD11b expression with GIT involvement. Migration of neutrophils across cellular barriers, such as vascular endothelium and epithelium occurs due to the effects of chemoattractants and cell-adhesion molecule interactions; however it seems, through different mechanisms. Specifically, neutrophils can also migrate across the intestinal epithelium using a CD11b/CD18-independent mechanism [35], as suggested through the CD47 interaction with the signal regulatory protein α [36, 37]. It is therefore possible that neutrophils use different adhesion molecules to enter the site of inflammation in GIT of patients with IgAV, requiring a neutrophil phenotype distinct from the one observed in patients with skin limited involvement.

Conclusion

Taken together, and to the best of our knowledge, this is the first study to date to determine both, serological and cellular markers in histologically-proven adult treatment-naïve IgAV. We confirmed the association of the NLR with GIT involvement [38,39,40,41] that could be a viable marker in the future. Neutrophils and their surface molecules (CD62L and CD11b), in the presence of low–medium grade inflammation, may all contribute to the vicious cycle of IgAV pathogenesis.