Platelet-activating factor receptors on B cells of chronic lymphocytic leukaemia patients.

Abstract

Platelet-activating factor (PAF) is an inflammatory phospholipid molecule that acts through membrane PAF receptor (PAF-R). PAF-R is a seven-transmembrane molecule that belongs to the G-protein-coupled family (Ishii & Shimizu, 2000). Reverse transcription polymerase chain reaction (RT-PCR) experiments and pharmacological studies suggested the presence of PAF-R on human tonsillar B lymphocytes (Bastien et al, 1999; Zhuang et al, 2000). However, the low number of available cells did not allow the direct detection of membrane PAF-R. Using multiparameter flow cytometry, we have analysed, for the first time, the putative presence of PAF-R on B cells from chronic lymphocytic leukaemia (B-CLL) patients. Blood was obtained from 28 untreated B-CLL patients (Matutes score > 4) (19 men, nine women, mean age 68 years) according to the Helsinki recommendations. Blood was incubated for 10 min with 5% of AB serum (ETS, Nantes, France) at 37°C before incubation with anti-human PAF-R mouse antibodies (Spi Bio, Massy, France) for 30 min at 4°C. After washing, cells were incubated with fluorescein isothicyanate (FITC)-labelled goat-anti mouse antibodies (Dakopatts, Glostrup, Denmark) (Desplat et al, 1999). Cells were then labelled with phycoerythrin (PE)-labelled anti-CD3, CD5, CD10, CD20, CD22 or CD23 antibodies or PC5-labelled anti-CD19 antibodies (Immunotech, Marseille, France). Cell suspension was then submitted to flow cytometric analysis (XL II, Coulter, Margency, France). As shown in Fig 1, B cells of CLL patients expressed PAF-R. Positive cells were CD3−, CD5+, CD10−, CD19+, CD20+, CD22+ and CD23+ (data not shown). Among the 28 patients, nine (32%) were PAF-Rneg, nine (32%) were PAF-Rdim and 10 (36%) were PAF-Rbright. The vast majority (90 ± 4%, n = 19) of B cells from PAF-R-positive patients expressed PAF-R. Seventy-three per cent (16 out of 22) and 66% (4 out of 6) of patients with a κ- and a λ-chain expressed PAF-R, respectively, indicating that the nature of the light chain did not interfere with the presence or absence of membrane PAF-R. PAF-R on CLL patients. Left panels: lymphocytes gated on physical parameters were analysed for the presence of PAF-R. Right panels: co-expression of CD19 and PAF-R on CLL cells. Cells expressing PAF-R were divided into PAF-Rdim (log fluorescence intensity between the first and second decade) and PAF-Rbright (log fluorescence intensity between the second and third decade). A, B and C are blood samples from PAF-Rbright, PAF-Rdim and PAF-R-negative patients respectively. This clinical study reports, for the first time, membrane PAF-R on B cells of CLL patients. PAF-R is not found in all patients. Stimulation of tonsillar B lymphocytes with PAF increased their interleukin 4 (IL-4) mRNA levels only in a small part of the entire population (Bastien et al, 1999), suggesting that PAF-R was not present on all B cells. Moreover PAF-R mRNA levels are higher in cells of the germinal centre than in the marginal zone. We suggest that the presence of PAF-R in B cells of CLL patients is linked to their presence on the initial cell generating the leukaemic clone. The physiological reason for PAF-R on leukaemic B cells remains an open question. The relationships between PAF and B cells are of importance. Activation of tonsillar B cells through the B-cell receptor downregulates the number of PAF-R and its functional response (Zhuang et al, 2000). Conversely, exogenous PAF abrogates apoptosis induced by IgM cross-linking in a B cell line (Toledano et al, 1997). Clearly, the presence of PAF-R as a prognostic and/or diagnostic marker in the outcome of B-cell malignancies deserves to be elucidated as well as its expression in more immature states such as during acute lymphoblastic leukaemia. The authors are grateful to the Ligue Nationale contre le Cancer (Comité de la Corrèze) for funding this project.