lonal antibody against CD154 or control monoclonal antibody was added in the culture, as indicated. Expression of CD137 on CD19-positive CLL cells was analyzed by flow cytometry. CLL B cells were cultured alone or co-cultured with HeLa control or HeLa-CD154, as indicated, for 24 h before 
analysis of CD137 expression. CLL B cells were analyzed after being co-cultured for 24 h with or without CD32 L cells that had captured agonistic anti-CD40 or control isotype murine IgG antibody, as indicated. RT-PCR was done using various samples as follows: lane 1, CLL B cells co-cultured with HeLa control cells; lane 2, CLL B cells co-cultured with HeLa-CD154; lane 3, HeLa-CD154 cells alone; lane 4, PBMCs from a healthy donor alone; lane 5, PBMCs from a healthy donor stimulated with PMA and ionomycin; lane 6, negative control without the RT reaction. After culture for 48 h, RNA was extracted by Trizol for RT-PCR analysis. RT-PCR products R-7128 chemical information obtained with primers for CD137 or b-actin, are shown as indicated. CLL B cells were cultured with the indicated stimulator. After 24 h, cells were subjected to FACS analysis. MFIR is plotted, which is calculated by dividing the mean fluorescence intensity of cells stained with a PE-CD137 monoclonal antibody by that of cells stained with a PE conjugated isotype control monoclonal antibody. doi:10.1371/journal.pone.0064425.g001 significant effect. We further confirmed the involvement of the CD40CD154 interaction in CD137 induction using CD154-transfected HeLa cells. Co-culture with HeLa-CD154 cells, but not with parental HeLa cells, could strongly induce CD137 expression on CLL B cells. Furthermore, the agonistic anti-CD40 antibody crosslinked with CD32-expressing 26646986 murine fibroblast cells also induced CD137 on CLL B cells, whereas the antibody alone could not induce CD137 expression. The intrinsic induction of CD137 expression in B cells by the CD40 signal was demonstrated by RT-PCR analysis, which revealed that CD137 was induced at the mRNA level in CLL B cells. The sequencing analysis further revealed that the soluble form of CD137, generated by alternative splicing, was also induced in addition to the membranous type in CLL B cells. Next, we checked the inducibility of CD137 by other stimuli for B cells, including LPS, ODNs, PMA, ionomycin, and anti-IgM antibody. CD40 stimulation was the only factor that induced CD137 expression on CLL B cells. The addition of antiIgM could slightly augment CD137 induction by CD154, although it could not induce CD137 by itself. Next, we examined CD137 induction on various types of malignant B cells derived from patients with B-cell acute lymphoblastic leukemia, CLL, diffuse large B-cell lymphoma, Waldenstrom macroglobuline mia, and FL. Each of the primary cells was cocultured with HeLa-CD154 and analyzed by FACS. The induction of CD137 was clear .1.5) in 27 cases but not in 6 cases. Notably, CD137 induction was observed clearly in all CLL cases, and its average value was significantly higher than that of non-CLL cases or healthy donors. This prominent induction of CD137 on CLL B cells prompted us to examine the in vivo induction of CD137 in CLL patients. We analyzed CD137 expression on 106 CLL B cells from the peripheral blood of 7 patients by flow cytometry. In 2 patients, CD137-positive cells could be detected as 2.3% and 0.76% of the CD19-positive and CD3-negative population, respectively. In the other 5 samples, the 9226999 percentage of CD137-positive cells was,0.5% and they were indis