Peripheral blood B-cell death compensates for excessive proliferation in lymphoid tissues and maintains homeostasis in bovine leukemia virus-infected sheep

Abstract

The size of a lymphocyte population is primarily determined by a dynamic equilibrium between cell proliferation and death. Hence, lymphocyte recirculation between the peripheral blood and lymphoid tissues is a key determinant in the maintenance of cell homeostasis. Insights into these mechanisms can be gathered from large-animal models, where lymphatic cannulation from individual lymph nodes is possible. In this study, we assessed in vivo lymphocyte trafficking in bovine leukemia virus (BLV)-infected sheep. With a carboxyfluorescein diacetate succinimidyl ester labeling technique, we demonstrate that the dynamics of lymphocyte recirculation is unaltered but that accelerated proliferation in the lymphoid tissues is compensated for by increased death in the peripheral blood cell population. Lymphocyte homeostasis is thus maintained by biphasic kinetics in two distinct tissues, emphasizing a very dynamic process during BLV infection.

FIG.1.

B-cell trafficking through the lymph node. (A) The efferent lymphatic vessels from the prescapular lymph nodes of a BLV-infected (4535) and a control (4534) sheep were surgically cannulated, allowing chronic sampling of lymph. Animals were injected intravenously with 25 mg of CFSE. At regular intervals of time (2, 22, and 94 h), lymph was collected and B cells were labeled with the anti-IgM 1H4 monoclonal antibody in association with a phycoerythrin conjugate. Ten thousand cells were then analyzed by two-color flow cytometry (x axis = CFSE; y axis = B lymphocytes). The percentages of CFSE⁺ B cells within the total B-lymphocyte population are indicated in the upper right quadrants. (B) Lymph samples from four BLV-infected (107, 1095, 4535, and 4536) and four control (2149, 2152, 4533, and 4534) sheep were continuously collected at regular intervals of time (in hours) post CFSE injection, and the percentages of CFSE⁺ B cells within the total B-cell population were determined. A vertical line is placed at 24 h. (C) Graphic distribution of the percentages of CFSE⁺ B cells in efferent lymph recovered at 26 h after CFSE injection. Individual values (BLV-infected [▴, 107; ⧫, 1095; ▪, 4535; •, 4536] and uninfected [▵, 2149;⋄, 2152; □, 4533; ○, 4534] sheep) and mean values (black lines) are represented. NS means no statistically significant difference by the Student t test. (D) Schematic representation of the change in label over time in the lymph. The model contains two parameters, i.e., the rate of increase in the label (a) and the equilibrium value of the label (k). Neither the rate of recirculation of labeled B cells to the lymph nor the equilibrium value of labeled cells in the lymph varied between BLV-infected and uninfected sheep (P = 0.39 and P = 0.57 for a and k, respectively, according to the Wilcoxon-Mann-Whitney two-tailed test).

FIG.2.

B-cell CFSE kinetics in peripheral blood. (A) CFSE was injected into the jugular vein of a control sheep (117), and blood was recovered at different times (before and 1 min, 2 h, and 6 days after injection) from the other jugular vein. PBMCs were isolated by Percoll gradient centrifugation, and fluorescence was measured by flow cytometry of 10,000 cells (x axis = CFSE; y axis = number of events). (B) One BLV-infected (2091) and one control (2147) sheep were injected intravenously with a bolus of CFSE, and an aliquot of blood was collected by jugular venipuncture before injection and at 1, 6, or 34 days postinjection. PBMCs were purified, and B cells were labeled with the anti-IgM 1H4 monoclonal antibody and stained with a phycoerythrin conjugate. Finally, 10,000 cells were analyzed by flow cytometry (x axis = CFSE; y axis = B lymphocytes). The percentages of CFSE⁺ B cells within the total B-lymphocyte population are indicated in the upper right quadrants. (C) Kinetics of the CFSE⁺ B-cell population in the peripheral blood of three BLV-infected (2091, 4535, and 4536; solid lines) and three control (2147, 4533, and 4534; dotted lines) sheep. The arrows indicate key times of CFSE kinetics (day 27 and day 83) at which the percentages of labeled B cells reached baseline levels in BLV-infected sheep and in the controls, respectively.

FIG.2.

B-cell CFSE kinetics in peripheral blood. (A) CFSE was injected into the jugular vein of a control sheep (117), and blood was recovered at different times (before and 1 min, 2 h, and 6 days after injection) from the other jugular vein. PBMCs were isolated by Percoll gradient centrifugation, and fluorescence was measured by flow cytometry of 10,000 cells (x axis = CFSE; y axis = number of events). (B) One BLV-infected (2091) and one control (2147) sheep were injected intravenously with a bolus of CFSE, and an aliquot of blood was collected by jugular venipuncture before injection and at 1, 6, or 34 days postinjection. PBMCs were purified, and B cells were labeled with the anti-IgM 1H4 monoclonal antibody and stained with a phycoerythrin conjugate. Finally, 10,000 cells were analyzed by flow cytometry (x axis = CFSE; y axis = B lymphocytes). The percentages of CFSE⁺ B cells within the total B-lymphocyte population are indicated in the upper right quadrants. (C) Kinetics of the CFSE⁺ B-cell population in the peripheral blood of three BLV-infected (2091, 4535, and 4536; solid lines) and three control (2147, 4533, and 4534; dotted lines) sheep. The arrows indicate key times of CFSE kinetics (day 27 and day 83) at which the percentages of labeled B cells reached baseline levels in BLV-infected sheep and in the controls, respectively.

FIG. 3.

Kinetic analyses of B-cell subsets. At regular time intervals of the CFSE kinetic analysis of three BLV-infected sheep (2091, 4535, and 4536) and three controls (2147, 4333, and 4534) (Fig. 2), PBMCs were labeled with a B-lymphocyte-specific antibody and with a peridinin-chlorophyll-protein conjugate. The PBMCs were then incubated with monoclonal antibodies directed against L-selectin, CD21, CD5, or CD11b and with a phycoerythrin conjugate. Finally, 10,000 events of thrice-labeled cells were analyzed by flow cytometry. The percentages of CFSE⁺ B cells expressing L-selectin, CD21, CD5, or CD11b in the corresponding B-lymphocyte subsets were determined at different time intervals (in days). At day 26, the percentages of CFSE⁺ CD11b⁻ B cells reached baseline levels in both infected sheep and controls (arrow).

FIG. 4.

Theoretical fit of the model to the CFSE data. The percentages of CFSE⁺ B cells (P [black squares]) and the ratio of the MFI of B⁺ CFSE⁺ cells to the MFI of B⁺ CFSE⁻ cells (I [open triangles]) were determined by flow cytometry analyses. The data corresponding to the P and I values were fitted simultaneously to a mathematical model, yielding theoretical curves.

FIG. 5.

Model summarizing the dynamic parameters studied. (A) PBMCs from two BLV-infected sheep were isolated from total blood, and cells were collected in parallel from lymph. Cells were cultivated for 24 h in RPMI medium supplemented with 10% fetal calf serum, 2 mM glutamine, and penicillin-streptomycin. B cells and viral capsid protein p24 were then labeled with specific monoclonal antibodies and analyzed by flow cytometry. The percentages of CFSE⁺ B cells expressing p24 in blood are represented by black bars, and data from lymph are represented by gray bars. The data are mean values and standard deviations resulting from more than five samplings taken over a period of 46 h post CFSE injection. (B) Quantification of proliferation and death rates in BLV-infected sheep. Excessive proliferation in the lymphoid tissues measured by BrdU incorporation can be compensated for by an increase in cell death in the peripheral blood, as deduced from the CFSE kinetics. The percentages shown represent the proliferation or death rates determined in BLV-infected or control sheep. **, statistically significant difference (P < 0.01) by the two-tailed unpaired Student t test.