Discussion
In the present study we report differences in circulatory B cell subsets, monocyte subsets and IL-6 levels in IgAN patients, compared to healthy controls (HC) and/or disease controls i.e. patients with ADPKD. We also found associations between sCD40L levels and the proportion of pre-switched B cells, as well as sCD40L and MCP-1 levels and albuminuria in IgAN patients.
B cell development is an important subject to understand the pathophysiology of IgAN. IgD is expressed on naïve B cells before they are exposed to antigen and perform class-switching. IgD together with IgA are the only human immunoglobulins undergoing O-glycosylation (9, 10). Changes in IgA O-glycosylation occur late during B cell maturation and imply production of Gd-IgA1 (Hit 1) and autoantibodies against Gd-IgA1 in IgAN (Hit 2) (4, 10), both events crucial in the pathological process (4, 11, 12). Several immunological effector pathways are involved and orchestrate these events, for example T-lymphocytes, APCs and the cytokine/chemokine network. Given this immunological network, we aimed to simultaneously identify different B cell subsets together with T-cell, monocyte and cytokine/chemokine signatures with a method that minimize ex vivo manipulation.
We observed a significant lower proportion of pre-switched B cells and a higher trend in the naïve B cell population in IgAN compared to both HC and ADPKD. To validate these observations, we analyzed individual ratios between naïve and pre-switched B cells and disclosed an increased ratio in IgAN compared to HC and ADPKD. This may suggest an altered balance in B cell maturation in IgAN which could reflect a migration of pre-switched B cells from the circulation to for example follicles in the gut mucosa and Payer´s patches for further maturation and class-switching (13).
The proportions of plasmablasts in the peripheral circulation of IgAN patients were lower compared to HC. Plasma cells constitute an important B cell subset responsible for antibody production and their life-cycle in IgAN has been debated. Our data indicate a translocation of circulatory plasmablasts to lymphoid tissues for further maturation into antibody-producing cells. This assumption is supported by data that demonstrate migration of plasma cells to central lymphoid tissue, such as the bone marrow where they continue to produce underglycosylated antibodies (14, 15). A lower number of polymeric IgA-secreting plasma cells in the duodenal mucosa but higher in systemic sites, for example in the bone marrow have been reported, which support a dynamic view of the cellular compartmentalize process in IgAN (4, 16). A lower proportion of circulating plasmablasts in IgAN patients has also recently been reported by Cols et al (17). And their cohort was comparable to ours in terms of age and eGFR. However in the study by Si and co-workers the percentage of plasmablasts was elevated in IgAN (18). In contrast their cohort included younger patients with more advanced disease compared to our patient group. This discrepancy indicate that age and clinical status are crucial parameters to consider when B cell subtype data are interpreted.
Th2 cells play a major role in the orchestration of class-switching and maturation of B cells. We calculated the relationship between the plasmablasts and Th2 cells and noticed a lower ratio in IgAN which indicates a relative higher available circulating Th2 count per plasmablast. However, whether this relationship is valid also at the lymphoid tissue or whether it has any impact on the actual antibody production can only be speculated on. Overall, we found no significant differences in various subsets of T cells in IgAN compared to HC.
We report a higher proportion of long-lived plasma cells (LLPC) in IgAN patients compared to HC. LLPC contribute to the humoral response by providing antibody production independent of remaining antibody producing plasma cells (19). LLPCs consist partly of CD19-negative cells and reside mostly in bone marrow (19, 20) but can also be found in inflamed tissue such as in kidneys (20) or the intestinal mucosa (21). To the best of our knowledge the CD19-negative LLPCs have not previously been studied in IgAN and it is currently not known in what way this subpopulation of B cells contributes to the pathological processes. An interesting finding is that polycystic kidney disease patients also had an expanded population of CD19-negative LLPC. A plausible explanation for this finding is not imminent but there is growing interest in how vascularization and lymphangiogenesis might affect cyst formation and growth, a process in which B cells are involved (22).
Our data showed an altered monocyte phenotype in patients with IgAN with an increased proportion of non-classical monocytes. Peripheral monocytes can be divided into three subgroups, classical (CD14++ CD16), intermediate (CD14++, CD16+) and non-classical (CD14+, CD16++) monocytes (23, 24) and our data go in line with a study by Cox and co-workers (25). A noteworthy difference between our studies is that the patients had a more preserved renal function in the study by Cox and co-workers. An interpretation of this observation could be that changes in monocyte profile appear at an early stage of IgAN disease. In our study increased levels of non-classical monocytes were accompanied by elevated IL-6. Cox et al did not analyze IL-6, and therefore we cannot further speculate whether an expanded population of non-classical monocytes precede an increase in IL-6.
CD16+ monocytes have previously gained attention and our group has reported increased CD16+ peripheral monocytes as well as their accumulation at the site of induced inflammation in patients with CKD (26). Zhu and co-workers reported a pro-inflammatory role of these cells in systemic lupus erythematosus (27) and Cols et al reported a potential role of CD89Low non-classical monocytes in the IgAN pathogenesis, which together set attention towards this monocyte subset (17). Of interest is that we found a significant relationship between MCP-1 levels, a known chemotactic factor for monocytes, and albuminuria in the current study, which further indicate a role for these cells in the pathophysiology of IgAN.
We found a significant correlation between sCD40L levels and albuminuria and higher levels of sCD40L were related to a lower proportion of pre-switched B cells. This might indicate a role for the CD40-CD40L pathway in the IgAN pathophysiology, even though the exact in vivo mechanism of sCD40 is not fully understood. However, sCD40L has been suggested to act as a suppressor of the immune response by blocking the CD40-CD40L interaction between B- and T-cells, thereby affecting the B cell maturation and class-switch and inhibiting immunoglobulin production (28-30).
In this study we used a method that admits simultaneous analysis of several immune cells and soluble factors with a minimum of ex vivo manipulation. Even though the number of subjects is low, the study includes a well-defined group of IgAN patients, HC and a disease control group (ADPKD) to validate our results.
In conclusion, we applied an easy-access method to analyze subsets of immune cells as well as relevant inflammatory mediators in IgAN patients, healthy controls (HC) and disease controls with ADPKD. Our data demonstrate an altered B cell profile that indicates a pathophysiological role of the B cell linage in IgAN and an increased proportion of non-classical monocytes suggesting their role in the disease process.