CD31/PECAM-1 Is Overexpressed On Mantle-Cell Lymphoma (MCL) Primary Cells and Positively Regulates Engraftment and Growth of MCL Cell Lines in Immunodeficient Mice.

Abstract 2681

Mantle-cell lymphoma (MCL) is a lymphoma subtype with poor prognosis. Recently, new drugs targeting at least partially the angiogenesis cascade have been successfully tested in the therapy of relapsed MCL patients, including temsirolimus, enzastaurin and lenalidomide. To reveal molecular mechanisms that regulate MCL-induced angiogenesis (and that might represent potential new druggable targets), we established and analyzed two mouse models of human MCL. Immunodeficient mice were subcutaneously (s.c.) xenografted with MCL cell lines JEKO-1 and HBL-2, and when tumor diameters reached 3cm in any size, the mice were sacrificed, and the excized tumors subjected to immunohistochemical (IHC) analysis. Alternatively, ex vivo obtained MCL cells were magnetically sorted using CD45 microbeads, and subjected to gene expression and flow cytometry analyses compared to the in vitro growing controls.

IHC analysis proved that the tumors were neovascularized. Gene expression profiling by TaqMan Human Angiogenesis Array revealed that the most upregulated gene in both JEKO-1 and HBL-2 in vivo vs. in vitro growing cells was platelet/endothelial cell adhesion molecule CD31/PECAM-1 (fold change 148.9 ± 19.4, and 127.6 ± 10.5, respectively). Recently, Boyd et al. reported upregulation of CD31 in three of five primary MCL samples compared to normal B cells by Western blotting. By flow cytometry we assessed surface expression of CD31 on primary cells obtained from peripheral blood and bone marrow of 25 MCL patients before therapy. The percentage of CD31 positive cells was significantly higher in the subset of CD19+CD5+ peripheral blood MCL cells compared to CD19+CD5+ healthy donor B-cells (78.6± 4.4 vs. 14.7±2.6, p< 0.001), as well as in the subset of CD19+CD5+ bone marrow MCL cells compared to CD19+CD5+ B-cells obtained from the bone marrow of patients with various lymphoma subtypes with no detectable bone marrow involvement (81.9±3.9 vs. 20.5±4.8, p< 0.0001).

In addition to surface CD31 we asked if patients with MCL have increased levels of soluble form of CD31 (sCD31) compared to healthy volunteers. We measured concentrations of sCD31 in plasma samples obtained from 17 MCL patients before therapy by ELISA. Despite the fact that MCL patients demonstrated higher variability in sCD31 concentrations (range 10.7–135.6 ng/ml) compared to healthy volunteers (range 43.3– 92.0 ng/ml), the medians were not statistically different (56.4 vs. 53.3 ng/ml). We also measured concentrations of sCD31 in plasma of two patients with MCL before treatment and after three cycles of chemotherapy, and found that the post-chemotherapy levels of sCD31 were lower in both patients (91.8 vs. 60.7 ng/ml; 56.4 vs. 30.4 ng/ml).

To investigate the role of CD31 in the biology of MCL we derived JEKO-1 and HBL-2 clones with stable downregulation of CD31 by siRNA approach. In addition, two JEKO-1 subclones with upregulated CD31 were established by limiting dilution from the original cell line (=controls). The limiting dilution approach was not feasible in HBL-2 cells, which completely lack CD31 expression. While HBL-2 clones with downregulated CD31 engrafted constantly, JEKO-1 clones with downregulated CD31 engrafted only in 2 out of 8 mice. Growth of the tumors derived from both HBL-2 and JEKO-1 clones with downregulated CD31 was significantly slower compared to that of control tumors (HBL-2: 2.1±0.4 g versus 4.1±0.2 g, p=0.002; JEKO-1: 0.5±0.1 g versus 2.2±0.2 g, p=0.0019). Xenotransplantation of JEKO-1 clones with upregulated CD31 resulted in accelerated tumor growth compared to controls (3.4±0.2 g versus 2.2±0.2 g, p=0.0027). Importantly, the in vitro proliferation rate between the clones with changed CD31 expression and the original cell lines were not statistically different suggesting that the different growth pattern of tumors was a consequence of altered interaction between the tumor cells and the murine microenvironment.

In summary, CD31/PECAM-1 antigen is overexpressed on primary CD19+CD5+ MCL cells obtained from the peripheral blood and bone marrow of MCL patients before therapy compared to control CD19+CD5+ B-cells. The upregulated CD31/PECAM-1 appears to play important role in MCL biology, and might represent potential druggable target.

Financial Support: IGA-MZ NT13201-4/2012, GAUK 259211/110709, GAUK 446211, UNCE 204021, PRVOUK P24/LF1/3, PRVOUK 1–5101–280002 PVK, SVV-2012–254260507