CD28 and CD86 Are Necessary for Myeloma Cell Survival.

Abstract 2946

Multiple myeloma (MM) is a hematologic malignancy characterized by the aberrant proliferation of plasma cells. Myeloma cells retain most of the physiological characteristics of their normal counterpart – the long-lived plasma cell. Myeloma cells secrete immunoglobulin and reside in the bone marrow, where they rely heavily on interactions with the stroma for survival signals. While recent advances in therapeutics have led to an increase in median survival post-diagnosis, the disease remains incurable. Understanding the pathways which mediate growth and survival of these cells will help in identifying new targets that can potentially further improve patient outcomes.

CD28 is a receptor better known for its role in T-cell signaling through interaction with its ligands, CD80 or CD86. Interaction between CD28 on T-cells and CD80/86 on antigen-presenting cells leads to survival and proliferation of T-cells. Recent work has shown that the CD80/86-CD28 pathway also plays an important role in normal plasma cell generation and survival. Interestingly, high expression of CD28 and CD86 are poor prognostic markers for myeloma patients. Previous work has shown that CD28 activation provides survival signals for myeloma cells in growth-factor deficient conditions. It has also been shown that CD28 on the myeloma cell interacts with CD80/86 on the dendritic cell, which induces secretion of IL-6 (by the DC), an important myeloma growth factor. However, it is not known if CD28 or CD86 play a role in steady state growth and survival of myeloma cells.

In order to determine the role of each of these 2 molecules in myeloma physiology, we knocked-down either CD28 or CD86 on the myeloma cell via lentivirus-mediated shRNAs. We found that knockdown of CD86 leads to apoptosis in 3 myeloma cell lines (RPMI8226, MM1.s, and KMS18). Four days after infection with the lentivirus containing shCD86, 45.7±4.9 and 60.3±4.6 percent control apoptosis was observed in RPMI8226 and MM1.s respectively, while less death was observed in KMS18 (17.6±1.6). CD28-knockdown resulted in apoptosis as well (24.9±4.3 for RPMI8226, 26.8±4.1 for MM1s, 21.8±3.8 for KMS18, percent control apoptosis). Consistent with these findings, we were unable to establish a myeloma cell line with stable knockdown of either CD28 or CD86. Additionally, RPMI8226 cells stably transfected to over-express either Bcl-2, Bcl-x_L, or Mcl-1 are protected from cell death induced by CD86 or CD28 silencing. These data suggest that CD28 and CD86 are essential to prevent apoptosis of myeloma cells in vitro. To confirm these findings we determined the effects of CTLA4-Ig on myeloma survival. CTLA4-Ig inhibits CD86-CD28 signaling by binding to CD86, blocking its interaction with CD28. We found that treatment of RPMI8226 and MM1.s cells with CTLA4-Ig caused apoptosis in the myeloma cells after 2 days (23.9±3.9 for RPMI8226 and 20.4±6.2 for MM1.s, percent control apoptosis). Thus like normal plasma cells, CD28 and CD86 are required for the survival of myeloma cells.

To determine why silencing of CD86 has a more potent effect than CD28 silencing on myeloma cell survival in 2 out of 3 cell lines, we investigated the effects of silencing on cell surface expression of each of these proteins. CD28 and CD86 mRNA and protein levels were silenced to similar levels by their cognate hairpins. However, in MM.1s and RPMI8226 we found that silencing of CD28 resulted in an increase in CD86 surface expression. This increase was also observed at the mRNA level and in the cells over-expressing Bcl-2 family members, indicating that this is not simply due to the selection of the highest expressing cells. These data suggest a feedback loop exists to regulate CD28-CD86 signaling in myeloma cells. Surprisingly, in the KMS18 cell line, we observe the converse effect, where silencing of CD86 resulted in upregulation of CD28. This provides a likely explanation for why these cells are less susceptible to CD86 silencing than the other two lines. Interestingly, blocking CD86 with CTLA4-Ig treatment also resulted in a modest upregulation in CD28 surface expression of MM.1s and RPMI8226, which suggests that silencing CD86 and binding of CD86 with a soluble receptor are not equivalent, and that multiple signaling feedback pathways exist to regulate the expression of this receptor-ligand pair that is necessary for myeloma cell survival.