JAM-A overexpression is related to disease progression in diffuse large B-cell lymphoma and downregulated by lenalidomide

Abstract

Cancer stem cells play an important role on tumor progression. Biomarkers of stem cell property and their relationship to extranodal involvement of malignant lymphocytes are undefined in diffuse large B-cell lymphoma (DLBCL). Here we showed that junctional adhesion molecule-A (JAM-A) was highly expressed in DLBCL patients with multiple extranodal lesions. JAM-A maintained B-lymphoma cell stemness and was associated with cell invasion and epithelial-to-mesenchymal transition both in vitro and in vivo. As mechanism of action, JAM-A overexpression selectively activated transforming growth factor-β (TGF-β)/NODAL signaling, thereby enhanced B-lymphoma cell aggressiveness and induced extranodal involvement to mesoendoderm-derived organs in DLBCL. Lenalidomide downregulated JAM-A and downstream NODAL expression, resulting in inhibition of B-lymphoma cell invasion and epithelial-to-mesenchymal transition. In a murine xenograft model established with subcutaneous injection of JAM-A-overexpressing B-lymphoma cells, lenalidomide retarded tumor growth and prevented cell invasion to mesoendoderm-derived organs, consistent with the downregulation of JAM-A and NODAL expression. Collectively, these findings indicated that JAM-A was related to extranodal involvement in DLBCL through modulating TGF-β/NODAL signaling. Identified as a biomarker of stem cell property, JAM-A indicated the sensitivity of B-lymphoma cells to lenalidomide. Therapeutic targeting of JAM-A/NODAL axis could thus be a promising clinical strategy to impede tumor progression in DLBCL.

Figure 1

JAM-A overexpression was related to extranodal involvement and poor disease outcome in DLBCL. (A,B) JAM-A gene (A) and JAM-A protein (B) were overexpressed in DLBCL. (C) JAM-A gene expression correlated well with JAM-A protein expression. (D) DLBCL patients with extranodal involvement had higher JAM-A gene and JAM-A protein expression than those only with nodal lesions. (E) Patients in high JAM-A expression group had poor progression-free survival (PFS).

Figure 2

JAM-A indicated B-lymphoma cell stemness. (A) As assessed by Whole-mount in Situ Hybridization and real-time quantitative RT-PCR, c-myb and runx1 were overexpressed after 26 h microinjecting jam-1 mRNA (jam-1) in Zebrafish embryos. WT, wild type. (B,C) B-lymphoma cell line DB and SU-DHL-4 were transfected with control vector (Vector) and JAM-A vector (JAM-A). Ectopic expression of JAM-A was associated with increased colony formation (B) and stem-cell marker CD133 and CD34 expression (C). (D) DLBCL patients with high JAM-A expression displayed increased CD133 positivity. (E) Patients in high JAM-A expression group were enriched for a stem cell gene signature, as revealed by RNA sequencing. Data in (A), (B) and (C) are representative of three independent experiments.

Figure 3

JAM-A induced B-lymphoma cell invasion. (A,B) Epithelial-mesenchymal transition (EMT) was observed in JAM-A-overexpressing DB cells (A) and in DLBCL patients with high JAM-A expression (B). (C,D) JAM-A-transfected DB cells acquired increased cell invasion (C), which was inhibited in JAM-A-ShRNA-transfected cells (D). Data in (A,C and D) are representative of three independent experiments.

Figure 4

JAM-A activated TGF-β/NODAL signaling. (A,B) In JAM-A-transfected DB cells (A) and DLBCL patients (B), JAM-A overexpression was related to activation of TGF-β/NODAL signaling. (C) As assessed by Whole-mount in Situ Hybridization and real-time quantitative RT-PCR, ndr-1 and ndr-2 were increased in jam-1-overexpressing zebrafish. WT, wild type. (D) JAM-A upregulated NODAL expression, which was restored by specific TGF-β/NODAL/Smad inhibitor SB431542 or JAM-A ShRNA. SB431542 abrogated JAM-A-induced lymphoma cell invasion (E) JAM-A was the highest in patients with involvement of endoderm-derived organs, followed by those with involvement of mesoderm- and ectoderm-derived organs. (F) JAM-A correlated with increased NODAL expression in primary lymph nodes, as well as in secondary lesions. Data in (C) are representative of three independent experiments.

Figure 5

Lenalidomide inhibited lymphoma progression. (A) Lenalidomide (1 μM) downregulated JAM-A and NODAL expression more significantly in JAM-A-transfected DB cells as compared to vector-transfected cells. (B,C) Lenalidomide (1 μM) inhibited JAM-A-transfected cell invasion (B) and EMT (C). (D) In murine xenograft model established with subcutaneous injection of DB cells, tumor size was similar between vector-transfected (Vector) and JAM-A-transfected (JAM-A) group. Lenalidomide (25 mg/kg/day, JAM-A + lenalidomide) retarded tumor growth. *P < 0.05 and ***P < 0.001 comparing with the JAM-A group. (E) By micro-PET-CT, JAM-A overexpression contributed to tumor metastasis, all involving endoderm- and mesoderm-associated organs, which was inhibited by lenalidomide. (F) Lenalidomide downregulated JAM-A and NODAL expression. Data in (B) and (C) are representative of three independent experiments.