Human peripheral blood age-associated (CD11c+Tbet+) B cells: No association with age.

Based on the well-established observation that aged individuals develop impaired antibody responses to neo-antigens in general, and especially to vaccines, making this age group more susceptible to infectious challenges [1], many groups have focused on the study of the B cell subpopulations that could underly this cohort's humoral deficiency. These studies brought to light age-associated B cell (ABC) phenotypes bearing a putative relation with the declining B cell function. However, different markers and labels describing these atypical subpopulations have been used. Actually, these cells were named ABCs because they were first described in elderly mice, and characterized phenotypically by low CD21 expression, expression of CD11c, and detectable levels of CD95 [2, 3]. A recently published proposal for comprehensive phenotyping of human B cells reports on a small subset of unusual memory B cells lacking CD27, CD21, and CD38 expression, already present in healthy individuals but which would increase with age [4]. These authors argue that others have also described these unusual B cells as age-associated B cells (ABC), atypical memory B cells, tissue-based memory cells, double negative (DN)/late memory B cells, CD11c+ B cells, or memory Tbet+ B cells. However, it is unclear whether they all refer to the same subset, mainly because alternative phenotype markers have been used (e.g., IgD, CD10, CD11c, Blimp, among others). B cells are not a uniform population, they diverge according to their differential pathway, which relates to a variety of phenomena such as the strength and length of engagement with their antigen receptors, the presence of other stimuli, and others [5]. Age associated B cell were first described in mice, in which our knowledge progressed significantly while in humans it has been hampered by issues such as different phenotypic definitions for the ABC population, diverse combinations of surface and intracellular markers and names. Although there are several classifications of B cells, the most common classification starts with the division of B cells into four subpopulations using IgD and CD27 expression: Naive (IgD+CD27−), Switched (IgD−CD27+), Unswitched (IgD+CD27+), and Double negative (IgD−CD27−). In general DN cells are pointed as age associated B cells, but according to recent reviews [4, 6, 7], these cells comprise more phenotypes than the ABCs, since not all DN are necessarily CD21−, CD11c+ and TbeT+. ABC has recently been the subject of more in-depth reviews, especially those cells bearing the CD11c+Tbet+ phenotype [8-10]. These cells can be found in the blood, spleen, and bone marrow, but rarely in lymph nodes [5, 11]. Because of low levels of CXCR5 and CCR7 expression they are not located in lymph nodes [5]. Despite the heterogeneity of phenotype markers, these B cells share certain features, suggesting that they are expanded in various settings of chronic immune activation, a feature also associated with aging. They would represent a population of functionally exhausted B cells based on their significant replication history, accumulation of somatic hypermutations, and poor ex vivo responses to activation [12, 13]. However, the issue of whether these cells increase with and their role in immunosenescence remains inconclusive [8-10]. We tried to resolve in part this issue by assessing the distribution of peripheral blood CD11c+Tbet+ ABC in subjects with healthy and pathological aging, compared with that in young adults. We also assessed the distribution of the recently described atypical and activated phenotypes of ABC [14]. Three groups of aged (>60 years old) individuals and a young adult group were recruited: (a) non-smoking elderly (n = 20) without any severe disease or diseases that could affect the immune system; (b) smokers with ≥60 years old, without pulmonary function alterations or diseases that could affect the immune system (n = 17); (c) COPD (Chronic obstructive pulmonary disease) patients with controlled disease, without exacerbation for at least 30 days (n = 20); these patients were all ex-smokers who quitted smoking for ≥24 months prior admission to the study; (d) adults 18–40 years-old, without severe diseases or diseases that could affect the immune system (n = 20). The detailed criteria for recruitment and admission in the study of both the aged and young adult donors have been described previously [15]. The protocol was approved by the Institutional Ethics Committee (#5.264.948) and all individuals signed a written consent form. Peripheral blood mononuclear cells (PBMCs) were obtained from blood samples through Ficoll–Hypaque gradient, and cryopreserved at 1 × 107 cell suspensions in fetal calf serum/10% DMSO in liquid nitrogen. Cells were defrosted and incubated overnight in RPMI/10% normal human AB serum to recover homeostasis. Trypan blue staining was used to assess cell viability and only cell suspensions with over 90% viability were used. Evaluation of B cell markers was performed ex vivo by Flow cytometry staining, after cell thawing and overnight resting. Cells were resuspended in 50 μL PBS with the following monoclonal antibodies for extracellular staining: CD3 APC (BD Biosciences), CD19 PE (BD Biosciences), CD27 APC-H7 (BD Biosciences), CD21 PeCy-7 (BD Biosciences), CD11c Alexa Fluor 700 (Biolegend) and Live/DeadTM fixable violet dead cell stain kit (Invitrogen). Cells were incubated in the dark at 4°C for 30 min, washed twice with PBS. Intracellular staining was performed sequentially using BD Cytofix/Cytoperm fixation permeabilization kit. Extracellular stained cells were fixed and permeabilized for 20′, then washed twice with Permwash reagent. Cells were resuspended in 50 μL of Permwash containing monoclonal antibody Tbet FITC (BD Bioscience), incubated in the dark at 4°C for 30 min, washed twice with PBS. Cells were then resuspended in FACSFlow (BD Bioscience) for acquisition on an LSR Fortessa™ cytometer (BD Biosciences). Data were analyzed using FlowJo Software version 10.0. At least 200.000 events within lymphocyte gate were acquired for each analysis. The subsets defined within CD19+ cells after flow cytometry analysis were defined as: Total ABC cells CD21−CD11+Tbet+; ABC activated CD21−CD11+Tbet+CD27+; ABC atypical CD21−CD11+Tbet+CD27−. Gating strategy is shown in Figure 1. In Table 1, we show the clinical and demographic data of the participants. All three age groups were comparable regarding age range (medians around 65 years old), while the median age of the young adult group was 32 years old. BMI was similar among all groups, as was the predominance of women donors. While smoking history was not significantly different between the Smoker and COPD groups, the latter had, as expected, a significantly reduced pulmonary function. B cell numbers and percentages were not significantly different among the groups. ABC were a small fraction of the total B cells in all three age groups, in a similar fashion to the observed in young adults. Analysis of the absolute ABC numbers reinforced the overlapping results among the groups, with a median of ~70 cells/mm3 in all groups. In addition, there was no correlation between donors' ages and either the percentage or the absolute numbers of ABC cells (Spearman test, p = 0.81). ABC has been reported to comprise two subsets, activated and atypical ABC, according to the expression or lack of expression of the B cell memory marker CD27, respectively [14]. We examined the proportions of both subtypes within the ABC. Figure 2 shows that activated ABC overcomes atypical ABC irrespective of age or pathological condition. Unexpected populations of human memory B cell subsets have been first described over 15 years ago. These B cells have subsequently been identified because, despite bearing typical features of memory cells, they have abnormal expression of CD21, CD11c and CD27. More recently, different groups have also added Tbet molecule expression to define the ABC subset [9, 10]. These cells have frequently been termed age-associated B cells because of some evidence that they accumulate with aging [10], although there were also reports that they did not [16]. Many aspects of ABC remain unsettled because of the heterogeneity of markers used to identify these cells in humans. Functionally, it has been shown that, differently from conventional memory B cells, ABC respond poorly to ex-vivo stimulation through the B cell receptor, CD40 and toll like receptors [9, 10], suggesting they could correspond to exhausted B cells due to chronic antigen stimulation. In healthy individuals, ABC plays a role in protective antiviral responses by secreting virus-specific Ig2a [17, 18]. In fact, atypical memory B cells have been shown to accumulate in several infectious conditions that run a protracted course associated with chronic antigen stimulation, such as HIV, malaria, hepatitis B, etc. It has recently suggested that atypical, but not activated, ABC were linked to a deficient vaccine response of the elderly [14]. On the other hand, ABC has been reported to expand during disease activity in several autoimmune conditions and to play a role in the autoinflammatory process. In these conditions, in addition to be a marker of disease activity, they may display a disease promoting function as they can differentiate into autoreactive antibody secreting cells [16]. In a SLE mice model, Tbet targeting improved the disease status, reduced the autoantibodies and serum IgG2a titers, resulting in better survival [19]. Of note, however, age was not associated with further expansion of ABC in patients with autoimmune diseases [12]. We thus assessed the ABC in healthy aging subjects and in subjects with two conditions, tobacco exposure and COPD. Patients with COPD are known to have a persistent inflammation, a consequence of an abnormal immune response caused by lung remodeling and an exacerbated inflammatory response. Smokers who did not develop COPD also show evidence of an inflammatory status, which is nevertheless milder than COPD patients [20]. In both conditions the pro-inflammatory status does not relate to an autoimmune process or to chronic infections. We found no expansion of CD11c+Tbet+ ABC in the peripheral blood of all aged groups as compared with the healthy young adult group. We also found that, irrespective of age, the active (CD27+) but not the atypical (CD27−), CD11c+Tbet+ B cells corresponded to the major phenotype. Our findings demonstrate that ABC do not necessarily increase with healthy and pathological aging, and that this designation may be misguiding, as already suggested by others [12]. While several observations show that these memory B cells play a significant role in chronic infections and autoimmune conditions, their contribution to immunosenescence remains unclear. We reinforce previous suggestions that ABC should be preferentially termed CD11c+Tbet+ cells. We thank Dr. Jefferson Russo Victor for writing support during the commentary preparation. We also thank the LIM-56 structure, where all experimental procedures were performed. Further, we express our gratitude to the volunteers for their blood donation. As well, we thank the ‘Fundacão de Amparo à Pesquisa do Estado de São Paulo (FAPESP)’ #2020/15847-2 grant that supported this work. JRF receive postdoctoral scholarship from FAPESP. TNCP is doctoral scholarship holder from CAPES. GB and AJSD are senior researchers from CNPq.