Objectives: Lung transplant is a last treatment option for patients with end-stage pulmonary disease. Chronic lung allograft dysfunction, which generally manifests as bronchiolitis obliterans syndrome, is a major long-term survival limitation. Bronchiolitis obliterans syndrome is diagnosed when forced expiratory volume in 1 second declines > 20% in the absence of known causes. B cells can either contribute or restrain the development of bronchiolitis obliterans syndrome (eg, via induction of alloimmune antibodies, regulation of cellular immunity, and induction of tolerance). Here, we explored how peripheral B-cell subsets were altered in lung transplant recipients with bronchiolitis obliterans syndrome.
Materials and Methods: Fresh whole blood samples were analyzed from 42 lung transplant recipients, including 17 with bronchiolitis obliterans syndrome; samples from these groups were compared with 10 age-matched healthy control samples. B-cell subsets were analyzed using flow cytometry, and relative distributions of subsets were compared. Changes in forced expiratory volume in 1 second were also determined.
Results: Absolute B-cell count was significantly increased in transplant recipients with bronchiolitis obliterans syndrome. Transitional (CD24+CD38+) and naïve (CD27-IgD+) B cells were decreased in lung transplant patients, with transitional B cells almost absent in those with bronchiolitis obliterans syndrome. Double-negative (CD27-IgD-) memory B cells were significantly increased (P < .001). No differences were found for plasmablasts (CD38+CD24-) and switched (CD27+IgD-) and non-switched (CD27+IgD+) memory B cells. Correlation analyses showed positive correlations between lung function and naïve B cells in transplant recipients (P = .0245; r = -0.458).
Conclusions: Peripheral B-cell count and subset distribution were altered in lung transplant recipients with and without bronchiolitis obliterans syndrome compared with healthy controls. Transitional and naïve B-cell decreases may be caused by differentiation toward double-negative B-cells, which were increased. The correlation between forced expiratory volume and naïve B cells during follow-up care may be clinically interesting to investigate.
Key words : Bronchiolitis obliterans syndrome, End-stage pulmonary disease, Forced expiratory volume
Lung transplant is a last treatment option for patients with end-stage pulmonary disease.1,2 However, after lung transplant, patients have a considerably lower survival rate than patients with other types of organ transplants, with only 55% of patients surviving the first 5 years after the procedure.1,3 The lung allograft is in constant contact with ambient air and therefore also with pathogens and irritants. The allograft is considerably more prone to infections, caused by a decrease in mucous production, medication, and suppressed cough reflex due to the de-enervation of the transplanted lung.4,5 This leads to a triggering of immunologic reactions in the lungs that could, if not controlled, ultimately lead to chronic lung allograft dysfunction (CLAD).3 This dysfunction can manifest as a restrictive CLAD (restrictive allograft dysfunction) or as an obstructive CLAD (bronchiolitis obliterans syndrome [BOS]).3 Bronchiolitis obliterans syndrome is the most common cause of mortality in lung transplant patients.6 This syndrome involves inflammation of the bronchi, which leads to total destruction of the small airways and ultimately leads to airway-directed rejection.1,2,4,5,7 The severity of BOS (grades 0-3) is determined by the extent of decreased lung function in forced expiratory volume in the first second of expiration (FEV1) from a patient’s posttransplant baseline measurement.8
To prevent development of allograft dysfunction, patients receive a regimen of immunosuppressive drugs that reduce differentiation and proliferation of lymphocytes, including B cells.9 B cells are capable of regulating cellular immunity and inducing accommodation and tolerance.10,11 Tolerance is characterized by the nonresponsiveness of the host immune system toward the transplanted organ.12 B cells go through a series of developmental stages wherein they express different cell markers until reaching full maturation. Flow cytometric immunophenotyping is used to delineate the different stages of peripheral B-cell maturation in humans.13 The most immature peripheral B cell, termed transitional, can be identified by high expression of CD24 and CD38.13-16 With the use of immunoglobulin D (IgD) and CD27 as markers of human memory, B cells can be divided into 4 different populations: naïve (IgD+CD27-), non-switched memory (IgD+CD27+), switched memory (IgD-CD27+), and CD27-negative (IgD-CD27-) B cells, also known as double-negative B cells (DN B cells).13,17,18 Peripheral plasmablasts can be identified by increased expression of CD38.13,19-21 Whereas memory B cells and plasmablasts are associated with early acute antibody-mediated rejection,22 naïve and transitional B cells are associated with tolerance in kidney and liver transplant.23-25
Although B cells are becoming more intensively researched in solid-organ transplant, the effects of immunosuppressive medication and chronic rejection after lung transplant are still poorly understood. Therefore, in this study, our aim was to investigate peripheral B-cell subsets in lung transplant recipients. We compared B-cell subsets among healthy controls, lung transplant recipients diagnosed with BOS, and lung transplant recipients without BOS. Because diagnosis of BOS is based on the decrease of lung function (decline of FEV1 ≥ 20%), correlation analyses were performed for changes in FEV1 and peripheral B-cell subsets.
Materials and Methods
Blood samples (in EDTA) were obtained from 42 lung transplant recipients, including 17 recipients diagnosed with BOS. Samples were also obtained from 10 healthy control individuals. Lung transplant recipients included in this study were transplanted at the Heart Lung Centre (Utrecht, the Netherlands), received follow-up care at the St. Antonius Hospital, and provided Biobank informed consent. The diagnosis of BOS was defined as a decline in FEV1 of more than 20% from the baseline determined by an average of 2 measurements made at least 3 weeks apart in the absence of known causes.26 Patients were chosen at random when coming in for follow-up care and were only excluded from the study if they presented with severe complications and/or needed to be hospitalized. All patients with the diagnosis BOS were a BOS grade I or higher and were excluded for restrictive allograft dysfunction.8 Change in FEV1 was determined for each transplant recipient by calculating the difference between the highest FEV1 after lung transplant and FEV1 at time of blood sampling.
Fluorescence-activated cell sorting analysis
A total of 100 μL of fresh whole blood was stained for 15 minutes at room temperature in the dark. Antibodies used to identify the different stages were CD45 PE-Cy7 (HI30; Biolegend, San Diego, CA, USA), CD20 BV421 (2H7; Biolegend), CD19 PE (HIB19; Biolegend), CD27 APC (L128; BD Biosciences, San Diego, CA, USA), CD24 APC efluor 780 (SN3 A5 2H10; eBioscience, Waltham, MA, USA), CD38 PerCP-eFluo710 (HB7; eBioscience), and IgD FITC (IA6-2; eBioscience). An overview of the analysis is given in Figure 1. Samples were lyse-fixed with BD FACS lysing solution (BD Biosciences), which included < 15% formaldehyde and < 50% diethylene glycol. Samples were run on a FACS CantoII (BD Biosciences), and analyses were performed using Flowjo software (version 7.6.5; BD Biosciences).
Data were analyzed using GraphPad Prism software (version 7; GraphPad Software, San Diego, CA, USA). Case control comparisons between 2 groups were performed using Mann-Whitney U test. The Bonferroni correction was used to counteract the problem of multiple comparisons. P ≤ .016 was considered significant. A Spearman correlation test was performed for the change in FEV1, transitional B cells, naïve B cells, and DN B cells.
Our study included 42 lung transplant recipients, which included 17 patients diagnosed with BOS. Patient characteristics are presented in Table 1. Pretransplant diagnosis varied in the study groups, with largest representation being chronic obstructive pulmonary disease, which was 68% and 47.1% for the lung transplant-only and BOS group, respectively. In the lung transplant-only group, 60% were men, with an average age just above 60 years. In the BOS group, only 37% were men, with an average age of almost 52 years. The average time between lung transplant and blood sampling was 5.5 years for recipients with BOS and 6.8 years for the lung transplant-only group. During this time, the BOS cohort showed a change in FEV1 with an average of 1.6 L, whereas the lung transplant-only cohort showed an average change in FEV1 of 0.54 L.
After lung transplant, standard triple maintenance therapy involved tacrolimus, mycophenolic acid, and corticosteroids. All included transplant recipients received corticosteroids (100%), with 88% receiving mycophenolate acid in combination with tacrolimus (95.2%) and/or sirolimus (7.1%) at time of blood sampling. Blood material was taken and analyzed years after the initial BOS diagnosis; therefore, there were no additional treatments and increased immunosuppressive dosing at the time of blood sampling.
HLA typing and antibody screening
For each individual patient, HLA typing and antibody screening were performed before transplant (Table 2). Data were missing for 5 patients (2 in the BOS cohort and 3 in the lung transplant-only cohort). Every patient had reduced matched phenotypes with their donor. In the BOS cohort, only 1 patient had unacceptable HLA mismatches (DQ) and 78% panel reactive antibody (PRA) values. Luminex single antigen results showed only positive results for antikeratin antibody (KL1). The remaining patients in the BOS cohort had 0% PRA and no unacceptable mismatches pretransplant. Within the lung transplant-only cohort, 1 patient showed 100% PRA. This patient showed positive Luminex single antigen results for B82, B58, and B57. None of these were against the donor. Two patients had unacceptable mismatches (A, B, and DR) but did not have any HLA antibodies prior to transplant (0% PRA).
Increased absolute count in patients with bronchiolitis obliterans syndrome
Total B cells were determined by CD45+ and CD19+ expression. Figure 2 plots the absolute counts for BOS-free lung transplant recipients, transplant recipients with BOS, and healthy controls. B cells were decreased in the lung transplant-only group compared with healthy controls (P < .001). A significant increase was also shown for recipients with BOS compared with the lung transplant-only group (P < .001).
Significantly decreased transitional and naïve B cells in lung transplant
recipients with and without bronchiolitis obliterans syndrome
Transitional B cells (recent bone marrow emigrant B cells) are the most immature peripheral B cells in the human blood and are characterized by high CD24 and CD38 expression.13,27 Figure 3A shows the percentage of transitional B cells of total B cells for the 3 study groups. The median percentage significantly decreased in transplant recipients. In patients with BOS, the median percentage of transitional B cells was even lower (P = .0024). After transitional B cells, mature naïve B cells were distinguished from memory B cells by expression of CD27 and IgD. Figure 3B shows the percentage of mature naïve B cells (CD27-IgD+) for the 3 study groups. Although percentages of naïve B cells varied widely between transplant patients, median percentages were significantly decreased compared with that shown in the healthy controls. No differences were observed between transplant patients without and with BOS.
Significantly increased frequency of double-negative memory B cells
Memory B cells can be distinguished based on CD27 and IgD expression. The loss of surface IgD expression on B cells is a feature of classical switched memory cells.13 We differentiated between non-switched memory B cells (CD27+IgD+), also known as marginal zone-like B cells, and switched memory B cells (CD27+IgD-).13 Figure 4 shows that no differences were found in percentage of non-switched and switched memory B cells of total B-cell population. As shown in Figure 5, DN memory B cells (CD27-IgD-) increased in transplant recipients (P < .001) compared with healthy controls, despite the considerable within-group variation. However, no differences were found between transplant patients with and without BOS.
Stable frequency of plasmablasts in lung transplant patients
Plasmablasts or immature plasma cells can also be determined in the peripheral blood. Plasmablasts can eventually become plasma cells and are able to produce large volumes of antibodies.13 Figure 6 shows the percentage of plasmablasts of total B cells; as shown, no differences were detected.
Changes in forced expiratory volume in the first second of expiration were
correlated with naïve B cell levels
To test whether the observed shift in peripheral B-cell subsets correlated with clinical features of the patients, transitional and naïve B cells were compared with changes in FEV1 lung function. Figure 7 shows the linear regression for the decrease in naïve B cells and the increase of DN B cells with increasing changes in FEV1. Spearman analyses showed that an increase of FEV1 delta correlated with a decrease in naïve B cells (P = .0245, r = -0.458), whereas the opposite trend was observed for DN B cells. No correlation was found for lung transplant recipients with BOS, due to the small cohort size.
Obstructive chronic rejection or BOS is one of the major causes of lung allograft failure.28,29 Although diagnosis is made based on the decline of FEV1, BOS can be considered as a continuing process that starts with the dysregulation of the immune system, eventually leading to damage of the endothelium and development of fibrosis.30,31 In this study, we investigated peripheral B-cell subsets in lung transplant recipients with and without BOS and in healthy controls. The absolute count in recipients with BOS was significantly increased. Young peripheral B cells, including transitional and naïve B cells, were significantly decreased in transplant recipients compared with healthy control. Transitional B cells were hardly detectable in patients diagnosed with BOS. Although frequencies of mature non-switched memory B cells, switched memory B cells, and plasmablasts remained stable, DN memory B cells were significantly increased. When these results were combined with changes in FEV1 at blood sampling times, a correlation was found between FEV1 and naïve B cells.
The increase in absolute count of B cells was counterintuitive because, with the use of triple maintenance immunosuppressive therapy, a decrease was expected. This increase may indicate that the current immunosuppression regimen was not enough. In the peripheral blood, naïve B cells form the most prominent subset of B cells, with percentages ranging from 60% to 90%. Our data showed a decrease of young transitional and naïve B cells in transplant recipients. This also was counterintuitive. The immunosuppressive regimen in our cohort consisted of corticosteroids along with mycophenolic acid in combination with tacrolimus and/or sirolimus. Most of the medications are aimed to prevent proliferation of B cells (on which we will elaborate more). Therefore, the expectation was to find a relatively large group of young B cells in transplant recipients. The use of glucocorticoids has not been shown to cause significant acute changes in B-cell gene expression or in number of circulating B cells,32 and the prolonged use of corticosteroids may have a reducing effect on the total numbers of circulating B cells.33 In vitro, prednisolone inhibits the differentiation into plasma cells at high concentrations.34 Mycophenolic acid is reported to target lymphocytes and block proliferation.35-38 The effect of mycophenolic acid on B cells in vitro shows efficient suppression of cell proliferation and differentiation toward the memory B-cell phase. The effect of tacrolimus and sirolimus on B cells was tested in vitro by Traitanon and associates, with tacrolimus almost having no effect on the activation and proliferation of B cells.9 However, tacrolimus was found to inhibit T-cell-dependent B-cell activation.9,32
Our analyses of mature memory B cells revealed stable frequencies between transplant recipients and healthy controls. Overall, the shift in B-cell subsets observed in transplant recipients may in part be due to the use of a prolonged combined use of immunosuppressive medications; however, the results did not match the expected effect. The use of prolonged combined immunosuppressive medications on peripheral B-cell subsets in vivo requires further investigations.
We hypothesize that the decrease in transitional and naïve B cells and the increase of DN B cells might be continued pressure of allograft antigens, stimulating B cells to proliferate, especially during chronic inflammation and rejection. With the use of immunosuppressive medication, this indeed might lead to more DN memory-like B cells. The function of DN B cells and their relationship to CD27+ cells is not clear. The DN B-cell population has been shown to be present with greater frequency in immuno-senescence, systemic lupus erythematosus, rotavirus infection, and human immunodeficiency virus infection.33,39-41 Wu and associates reported that DN B cells are memory cells closely related to switched memory B cells.42 Wei and associates found similar results in patients with systemic lupus erythematosus.33 The distribution of mutations indicates that DN memory B cells are antigen experienced and show similar positive response in the absence of B-cell receptor costimulation.33 Palma and colleagues hypothesized that DN B cells may have been naïve B cells triggered by polyclonal stimuli due to increased circulation of antigen.43 Solid-organ transplant patients experience chronic immune activation of foreign antigen through the allograft; therefore, it is plausible that this is the main reason for the increase of DN B cells observed in our lung transplant patients.
When the shift in B-cell subsets was compared with clinical features, a negative correlation was found for changes of FEV1 at blood sampling times and naïve B-cell population (P = .0245, r = -0.458) in the lung transplant-only cohort. This suggests that the decrease in young B cells perhaps already starts when lung function decline begins during the BOS-free period. The shift that we found in B-cell subsets in transplant patients corresponds with the findings from Pereira and colleagues in patients with primary immunodeficiency.44 This study found that the proportions of transitional and naïve B cells were reduced, whereas atypical memory B cells were present at a higher proportion, also hypothesizing being caused by chronic stimulation of the immune system.44
Our study had several limitations, including being a single-center study with a small cohort. In addition, although HLA mismatching has been shown to be correlated with development of chronic rejection,45-47 because of limited donor organs available at our center, HLA mismatching and HLA antibody analyses are not part of routine screening. For our study cohort, HLA mismatching and antibody screening were performed before transplant. However, we found no differences between the BOS and liver transplant-only cohort. All patients had reduced matching HLA phenotypes, but only 3 patients had HLA mismatches, with 2 of these being free of chronic rejection. The change measured in peripheral B cells neither correlated with HLA mismatching nor correlated with pretransplant antibody screening.
Our results showed that young B cells are reduced and DN B cells are increased in lung transplant patients. We hypothesize that, when the lung transplant recipient’s immune system is exposed to abundant antigen, which is even higher during BOS due to chronic inflammation and repair processes, it is likely that that naïve and transitional B cells differentiate and their peripheral frequency decreases, while DN B-cell numbers increase. In our study, a decrease in naïve B cells correlated with a decrease in lung function in recipients who were not yet diagnosed with allograft dysfunction. This is an interesting clinical concept during follow-up care. In the future, we hope to confirm this hypothesis in retrospective and prospective research investigations.
Volume : 18
Issue : 2
Pages : 234 - 241
DOI : 10.6002/ect.2019.0240
From the 1Interstitial Lung Diseases, Centre of Excellence, Department of
Pulmonology, St. Antonius Hospital, and the 2Department of Medical Microbiology
& Immunology, St. Antonius Hospital, Nieuwegein, The Netherlands; and the
3Department of Respiratory Medicine University Centre Utrecht, Utrecht, The
Acknowledgements: The authors have no sources of funding for this study and have no conflicts of interest to declare. We give special acknowledgements to the medical doctors who saw the patients.
Corresponding author: Inge Schreurs, Koekoekslaan 1, 3435CM, Nieuwegein, The Netherlands
Figure 1. Flow Cytometry Gating Strategy for B Cells
Figure 2. Absolute Count of Total B Cells
Figure 3. Transitional (A) and Naïve (B) B Cells
Figure 4. Non-switched and Switched Memory B Cells
Figure 5. Double-Negative B Cells
Figure 6. Plasmablasts
Figure 7. Linear Regression B-Cell Subsets and Forced Expiratory Volume in 1 Second
Table 1. Patient Characteristics
Table 2. Unacceptable HLA Mismatch and Antibody Screening Pretransplant