Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2015 Sep 15;192(6):695-705.
doi: 10.1164/rccm.201501-0107OC.

B Cell-Activating Factor. An Orchestrator of Lymphoid Follicles in Severe Chronic Obstructive Pulmonary Disease

Affiliations
Free PMC article

B Cell-Activating Factor. An Orchestrator of Lymphoid Follicles in Severe Chronic Obstructive Pulmonary Disease

Francesca Polverino et al. Am J Respir Crit Care Med. .
Free PMC article

Abstract

Rationale: Patients with chronic obstructive pulmonary disease (COPD) have increased pulmonary lymphoid follicle (LF) counts. B cell-activating factor of tumor necrosis factor family (BAFF) regulates B cells in health, but its role in COPD pathogenesis is unclear.

Objectives: To determine whether BAFF expression in pulmonary LFs correlates with COPD severity, LF size or number, and/or readouts of B-cell function in LFs.

Methods: We correlated BAFF immunostaining in LFs in lung explants or biopsies from nonsmoking control subjects (NSC), smokers without COPD (SC), and patients with COPD with the number and size of LFs, and LF B-cell apoptosis, activation, and proliferation. We analyzed serum BAFF levels and BAFF expression in B cells in blood and bronchoalveolar lavage samples from the same subject groups. We assessed whether: (1) cigarette smoke extract (CSE) increases B-cell BAFF expression and (2) recombinant BAFF (rBAFF) rescues B cells from CSE-induced apoptosis by inhibiting activation of nuclear factor-κB (NF-κB).

Measurements and main results: Patients with Global Initiative for Chronic Obstructive Lung Disease (GOLD) stage IV COPD had increased numbers and larger pulmonary LFs than patients with GOLD stages I-II COPD and SC. We identified two main types of pulmonary LFs: (1) type A, the predominant type in GOLD stages I-II COPD and SC, characterized by abundant apoptotic but few BAFF-positive cells (mostly B cells); and (2) type B, the main type in GOLD stage IV COPD, characterized by abundant BAFF-positive cells but few apoptotic cells (mostly B cells). BAFF levels were also higher in blood and bronchoalveolar lavage B cells in patients with COPD versus NSC and SC. Surprisingly, rBAFF blocked CSE-induced B-cell apoptosis by inhibiting CSE-induced NF-κB activation.

Conclusions: Our data support the hypothesis that B-cell BAFF expression creates a self-perpetuating loop contributing to COPD progression by promoting pulmonary B-cell survival and LF expansion.

Keywords: B cell–activating factor of tumor necrosis factor family; autoimmunity; chronic obstructive pulmonary disease; cigarette smoke; lymphoid follicles.

Figures

Figure 1.
Figure 1.
The number and the size of pulmonary lymphoid follicles (LFs) were increased in patients with chronic obstructive pulmonary disease (COPD), especially in the very severe stages of the disease. Quantitation of the number (A) and area (B) of LFs in the lungs of nonsmoking control subjects (NSC), smokers without COPD (SC), patients with Global Initiative for Chronic Obstructive Lung Disease (GOLD) stages I–II COPD, and patients with GOLD stage IV COPD (n = 5/group). The Student's t test was used to analyze the data. In A and B, data are mean + SEM; *P < 0.05. (CH) Representative images of LFs in the lungs of NSC (C), SC (D), patients with GOLD stages I–II COPD (E), and patients with GOLD stage IV COPD (FH), in which B cells are identified by staining with a red fluorophore for CD20. In CG, the magnification is ×100. (H) Inset of an image of typical pulmonary LFs in the lung of the patient with GOLD stage IV COPD shown in G.
Figure 2.
Figure 2.
Lymphoid follicles (LFs) from patients with chronic obstructive pulmonary disease (COPD) with Global Initiative for Chronic Obstructive Lung Disease (GOLD) stage IV disease had more B cell–activating factor of tumor necrosis factor family (BAFF)-positive B cells and fewer apoptotic cells than patients with GOLD stage I–II COPD and smokers without COPD (SC). (AC) Confocal images of triple-color immunofluorescence staining of representative pulmonary LFs from a patient with GOLD stage IV COPD (A), a patient with GOLD stage II COPD (B), and an SC (C). B cells were stained with a red fluorophore for CD20, BAFF-positive cells are identified with a green fluorophore, and active caspase-3–positive cells with a gray color. 4′,6-Diamidino-2-phenylindole (blue) was used to counterstain the nuclei. The images shown are representative of LFs in 5 to 10 subjects/group. Two out of 10 patients with GOLD stage IV COPD, 3 out of 9 patients with GOLD stages I–II COPD, 1 out of 5 SC, and 3 out of 5 nonsmoking control subjects had no LFs in the lung sections studied.
Figure 3.
Figure 3.
The percentage of B cell–activating factor of tumor necrosis factor family (BAFF)-positive B cells in pulmonary lymphoid follicles (LFs) in patients with chronic obstructive pulmonary disease (COPD) was correlated indirectly with the percentage of apoptotic B cells in LFs and directly with LF area. The percentages of BAFF-positive B cells (A) and active caspase-3–positive B cells (B) in LFs in the lungs of nonsmoking control subjects (NSC), smokers without COPD (SC), Global Initiative for Chronic Obstructive Lung Disease (GOLD) stages I–II COPD, and GOLD stage IV COPD are shown. A Student's t test (A) and a Mann-Whitney test (B) were used to perform the statistical analysis of the groups. In A, data are mean + SEM. In B, the median and 5% and 95% confidence intervals are shown in the box plots, and the error bars in the box plots are SDs in B. In A and B, *P < 0.05 versus NSC or versus the group indicated. In C, the correlation between percentage of BAFF-positive B cells and percentage of active caspase-3–positive B cells in LFs in the COPD group is shown. (D) Correlation between the percentage of BAFF-positive B cells and the area of pulmonary LFs in the COPD group. These results are nonlinear, and the best fit for the data was a three-parameter single exponential rise to maximum curve. (E) Correlation between percentage of active caspase-3–positive B cells and the FEV1% predicted in patients with COPD is shown. The Pearson (C and E) or Durbin-Watson (D) statistical correlation tests were used to analyze the data. In AE, a total of 29 subjects were analyzed; 4 LFs were analyzed in NSC, 9 LFs in SC, 10 LFs in GOLD stages I–II COPD, and 17 LFs in patients with GOLD stage IV COPD. In CE, patients with COPD with GOLD I–II disease are represented by open triangles, and those with GOLD IV disease by solid triangles.
Figure 4.
Figure 4.
B cell–activating factor of tumor necrosis factor family (BAFF) was produced by B cells incubated with low concentrations of cigarette smoke extract (CSE), and exogenous recombinant BAFF rescues B cells from apoptosis induced by high concentrations of CSE in a nuclear factor-κB–dependent manner. (A) B cells were isolated from the blood of a healthy nonsmoker and incubated at 37°C with or without 2.5% or 5% CSE for 6, 18, 24, and 48 hours. Cells were then immunostained with Alexa-546 for CD20 and Alexa-488 for BAFF. BAFF staining in CD20+ B cells was quantified as described in Methods. Incubating human blood B cells with CSE concentrations greater than 5% did not increase B-cell BAFF expression (data not shown). (B, D, and F) Purified blood B cells from a healthy human nonsmoker were incubated at 37°C for up to 18 hours with or without 10% CSE and with or without 100 nM recombinant human BAFF (rhBAFF). In B, cells were immunostained with Alexa-546 for CD20 and Alexa-488 for active caspase-3. Active caspase-3 staining was quantified as described in Methods. Human B cells incubated with or without CSE, BAFF, and CSE with BAFF as outlined above, and DNA fragmentation were assayed in the B cells using a terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay (D), and the B cells were stained with propidium iodide (PI) (F). (C, E, and G) B cells were isolated from the spleens of unchallenged C57BL/6 wild-type mice, and equal numbers of B cells per experimental condition were incubated at 37°C with or without 2% CSE and with or without 100 nM recombinant murine BAFF (rmBAFF) for 12 hours. Increases in intracellular levels of active caspase-3 relative to cells incubated without BAFF or CSE were quantified in cell extracts using a quenched fluorogenic substrate specific for caspase-3 and assay standards of known concentrations of active caspase-3 (C), as described in the Methods section of the online supplement. Apoptosis of murine B cells was also measured by staining cells with either a TUNEL kit (E) or PI (G). Either Student's t tests (A and B) or Mann-Whitney tests (CG) were used to analyze the data. Data are expressed as mean ± SEM (A and B) or presented as box plots showing medians and 5% and 95% confidence intervals, with solid circles indicating outliers (CG). In AG, *P < 0.05 versus baseline or versus the group indicated. The results shown in A and B are representative of six and three independent experiments, respectively.
Figure 5.
Figure 5.
Exogenous recombinant B cell–activating factor of tumor necrosis factor family (BAFF) (rBAFF) rescues B cells from apoptosis induced by high concentrations of cigarette smoke extract (CSE) in a nuclear factor (NF)-κB–dependent manner. (A and B) Aliquots of B cells isolated from unchallenged wild-type mice were incubated with or without 2% CSE and with or without 100 nM rBAFF. Nuclear proteins were isolated, and equal amounts of protein (4 μg/sample) in the nuclear extracts were subjected to electrophoretic mobility shifts assays (EMSAs) using a labeled oligonucleotide probe containing the NF-κB consensus sequence. Assays were performed in the presence or absence of excess unlabeled probe to identify NF-κB protein bound specifically to the probe. A shows an image of an EMSA analysis of nuclear protein extracts from all experimental groups and is representative of the results from eight experiments (with two authors [M.L.-C. and P.T.] each performing four experiments independently). Note the marked reduction in signal of the band corresponding to the NF-κB–oligonucleotide complex when excess unlabeled probe is added, indicating specific binding of NF-κB present in nuclear extracts to the labeled oligonucleotide probe. In B, the intensities of the bands corresponding to NF-κB–oligonucleotide complexes were quantified using densitometry, and band intensities for all groups were normalized to signals in the B cells incubated without CSE or recombinant murine BAFF (rmBAFF). Box plots show the median values and the 5% and 95% confidence intervals. The error bars in the box plots (which are too small to be seen) are SDs; n = 8 independent experiments. In B, *P < 0.05 compared with unstimulated cells or with the group indicated. In C and D, aliquots of B cells isolated from unchallenged wild-type mice were incubated with or without 2% CSE and with or without 100 nM rmBAFF. After 12 h, cells were immunostained with Alexa-546 for CD20 and Alexa-488 for the p65 subunit of NF-κB. P65 staining in the nuclear versus cytoplasmic areas of the cells was quantified as described in Methods. C shows quantitation of staining for p65 in the nuclei of the cells. D shows quantitation of the ratio of nuclear:cytoplasmic staining for p65 in B cells. Either a Kruskall-Wallis one-way analysis of variance (ANOVA) by ranks test (B) or a one-way ANOVA test (C and D) was used to analyze the data. In C and D, *P < 0.05 compared with the group indicated. Data are presented as box plots showing medians and 5% and 95% confidence intervals (B) or as mean + SEM (C and D). AU = arbitrary units.
Figure 6.
Figure 6.
Nuclear localization of the p65 subunit of nuclear factor (NF)-κB is reduced in B cells in Global Initiative for Chronic Obstructive Lung Disease (GOLD) stage IV chronic obstructive pulmonary disease (COPD) pulmonary lymphoid follicles (LFs) versus B cells in LFs of smokers without COPD (SC) and GOLD stage I–II COPD lungs. (AC) Confocal images of triple-color immunofluorescence staining of pulmonary LFs from an SC (A), a patient with GOLD stage I COPD (B), and a patient with GOLD stage IV COPD (C). B cells were stained with a red fluorophore for CD20, a green fluorophore for B cell–activating factor of tumor necrosis factor family (BAFF), and a gray color for the p65 subunit of NF-κB. 4′,6-Diamidino-2-phenylindole (blue) was used to counterstain the nuclei. The images are representative of LFs in five to seven subjects/group. In D, a lung section from a patient with GOLD IV stage COPD was stained with isotype-matched nonimmune control antibodies. White arrows indicate nuclear localization of p65 staining in LF B cells, especially in the SC and patients with GOLD stage I COPD. Ms = murine; Rb = rabbit.
Figure 7.
Figure 7.
Cartoon illustrating potential roles for B cell–activating factor of tumor necrosis factor family (BAFF) in the regulation of B-cell homeostasis and lymphoid follicle (LF) expansion in chronic obstructive pulmonary disease (COPD) lungs. (A) Under mild chronic inflammatory conditions, such as those occurring in the lungs and other compartments of healthy smokers or patients with Global Initiative for Chronic Obstructive Lung Disease (GOLD) stages I and II COPD, limited quantities of BAFF are produced mainly by T cells, macrophages, and dendritic cells. Antigen recognition by the B-cell receptor (not shown) and engagement of CD19 on B cells with CD21 on dendritic cells (not shown) coupled with BAFF binding to its receptors on B cells promotes activation, proliferation, and differentiation of B cells. Limited BAFF signaling through its receptors on B cells is associated with increased nuclear factor (NF)-κB–induced apoptosis, leading to the formation of LFs that are limited in size and number. (B) Under severe chronic inflammatory conditions, such as those occurring in the severe stages of COPD, B cells themselves produce BAFF, and BAFF signaling via BAFF receptor (BAFF-R) in B cells inhibits nuclear translocation and activation of NF-κB, thereby preventing B cells in the LFs from undergoing apoptosis. Our data support the hypothesis that this process establishes a self-perpetuating loop of B-cell activation and increased B-cell survival, causing excessive expansion of the B-cell pool in the lung and especially within pulmonary LFs. These events may promote the growth of existing B-cell–rich LFs and the formation of additional LFs. BCMA = B-cell maturation antigen; TACI = transmembrane activator and calcium modulator and cytophilin ligand interactor.

Comment in

Similar articles

See all similar articles

Cited by 30 articles

See all "Cited by" articles

Publication types

MeSH terms

Feedback