CD27-CD38lowCD21low B-Cells Are Increased in Axial Spondyloarthritis

Abstract

B-cells have received little attention in axial spondyloarthritis (axSpA) and for this reason their role in pathogenesis remains unclear. However, there are indications that B-cells may be involved in the disease process. Our objective was to obtain insights into the composition of the peripheral B-cell compartment of axSpA patients compared to healthy donors (HD) and patients with primary Sjögren's syndrome (pSS), a typical B-cell-associated autoimmune disease. Special emphasis was given to CD27-negative B-cells expressing low levels of CD21 (CD21^low B-cells), since this subset is implicated in autoimmune diseases with strong involvement of B-cells. Transitional B-cells (CD38^hi) were excluded from the analysis of the CD27^-CD21^low B-cell compartment. This study included 45 axSpA patients, 20 pSS patients and 30 HDs. Intriguingly, compared to HDs the frequency of CD27^-CD38^lowCD21^low B-cells was significantly elevated in both axSpA and pSS patients (P<0.0001 for both comparisons). The frequency of CD27^-CD38^lowCD21^low B-cells expressing the activation-induced immune markers T-bet and CD11c was decreased in axSpA patients compared to HDs. A higher proportion of CD27^-CD38^lowCD21^low B-cells expressed the chemokine receptor CXCR3 in axSpA compared to HDs, suggestive for active involvement of these cells in an inflammatory process. The frequency of CD27^-CD38^lowCD21^low B-cells in axSpA patients correlated positively with age and erythrocyte sedimentation rate. Furthermore, axSpA patients with extra-skeletal manifestations (ESM) showed increased frequencies of CD27^-CD38^lowCD21^low B-cells compared to patients without ESM. In conclusion, our findings are suggestive of active B-cell involvement in the pathogenesis of axSpA, against prevailing dogma.

Keywords: B-cells; CD21low B-cells; Sjögren’s syndrome; ankylosing spondylitis; autoimmunity; axial spondyloarthritis.

Figure 1

Frequencies of total B-cells and B-cell subsets in patients with axSpA, pSS and healthy donors. Global gating strategy is shown with arrows pointing towards the next level of gating. Frequencies of (A) total B-cells (CD19⁺), relative to total lymphocytes, and the following B-cell subsets, relative to total B-cells, are shown: (B) memory B-cells (CD27⁺IgD^-), (C) double negative (CD27^-IgD^-) B-cells, (D) naïve (CD27^-IgD⁺) B-cells, (E) plasmablasts (CD27⁺CD38^hi), (F) transitional (CD24^hiCD38^high) B-cells, (G) CD27⁺CD38^lowCD21^low B-cells and (H) CD27^-CD38^lowCD21^low B-cells. Patients with axial spondyloarthritis (axSpA; n = 45) are indicated with red circles, healthy donors (HD; n = 30) are indicated with green squares and patients with primary Sjögren’s Syndrome (pSS; n = 20) are indicated with blue triangles. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns, not significant and horizontal lines indicate the medians.

Figure 2

Phenotypical characteristics of CD27^-CD38^lowCD21^low B-cells in patients with axSpA, pSS and healthy donors. The proportions of (A) T-Bet-positive, (B) CD11c-positive, (C) CXCR3-positive, (D) CXCR5-positive, (E) CD86-positive cells were analyzed within the CD27^-CD38^lowCD21⁺ and CD27^-CD38^lowCD21^low B-cell subsets. In addition, (F) an illustration of T-bet/CD11c co-expression by CD27^-CD38^lowCD21^low B-cells is displayed. Patients with axial spondyloarthritis (axSpA; n = 45) are indicated with red circles, healthy donors (HD; n = 30) are indicated with green squares and patients with primary Sjögren’s Syndrome (pSS; n = 20) are indicated with blue triangles. **P < 0.01, ***P < 0.001, ****P < 0.0001, ns, not significant and horizontal lines indicate the medians.

Figure 3

Distribution of immunoglobulins IgD and IgM by CD27CD38^lowCD21 B-cell compartment in patients with axSpA, pSS and HDs. Analysis of the distribution of immunoglobulins IgD and IgM (A) is shown for CD27⁺CD38^lowCD21^low, CD27⁺CD38^lowCD21⁺, CD27^-CD38^lowCD21^low, CD27^-CD38^lowCD21⁺ B-cells respectively. A more detailed analysis of the immune marker expression (B) by the naïve (IgD⁺IgM⁺) in orange and switched (IgD^-IgM^-) B-cell populations in blue shown as histogram plots comparing CD27^-CD38^lowCD21^low B-cell populations of one axial spondyloarthritis patient, one primary Sjögren’s syndrome patient and a healthy donor. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns, not significant and horizontal lines indicate the medians.

Figure 4

Relationship between frequencies of CD27^-CD38^lowCD21^low B-cells and clinical parameters in patients with axSpA. (A) Patients with axial spondyloarthritis (axSpA) were divided into separate groups based on subcategories of axSpA and the following clinical parameters: Biological DMARD naïve, nr-axSpA or axSpA, sex, symptom duration, C-reactive protein (CRP) and Ankylosing Spondylitis Disease Activity Score (ASDAS). (B) To explore associations between the CD27^-CD38^lowCD21^low B-cell compartment and several continuous clinical parameters in patients with axSpA, the frequency of CD27^-CD38^lowCD21^low B-cells was plotted against the following clinical parameters: Age, Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), ASDAS, CRP, erythrocyte sedimentation rate (ESR), symptom duration. Patients with axSpA(n=45) are indicated with red circles, healthy donors (HD; n = 30) are indicated with green squares. *P < 0.05, **P < 0.01, ****P < 0.0001, significant values in bold, ns, not significant and horizontal lines indicate the medians.