Multiple myeloma is a malignancy of long lived plasma cells. Like normal plasma cells, myeloma cells are dependent on the bone marrow microenvironment for survival. While the specific interactions and downstream signals mediated by the bone marrow stroma have yet to be fully characterized, drug resistance is linked to these pathways, and further understanding will uncover new therapeutic avenues for myeloma.
CD28 and CD86 are best known for their role in T-cell activation; however they have recently been shown to play important roles in the generation and survival of normal long lived plasma cells. CD28 is the canonical costimulatory receptor known to activate the PI3K-Akt pathway in T-cells upon binding to CD80 or CD86 from an antigen-presenting cell. CD28 and CD86 are also expressed by normal plasma and myeloma cells, and we have previously shown that both CD28 and CD86 are necessary for myeloma cell survival. Silencing of either CD28 or CD86 results in cell death in 3 human myeloma cell lines (RPMI8226, MM.1s, and KMS18). Interestingly, in 2 cell lines, knockdown of CD86 results in higher levels of cell death than CD28. In these lines, silencing CD28 or addition of a soluble inhibitor CTLA4-Ig (Abatacept), results in an increase in CD86 expression. Taken together, these data suggest that CD28 and CD86 regulate the survival of myeloma cells through either cis or trans signals, and feedback signals from CD28 control CD86 expression. The data also suggest that signaling events result from ligation of either CD28 or CD86.
To better define the nature of the survival signals emanating from CD28 and CD86, we performed RNA-Seq on myeloma cells where CD28 or CD86 expression had been silenced. For silencing of CD28 we found 1292, 1195, and 1697 transcripts that were significantly changed compared to vector control in KMS18, MM.1s and RPMI8226, and silencing CD86 results in a similar number of changes (1405, 1200 and 1866 transcripts respectively). Since CD28 and CD86 form a receptor-ligand pair, we focused on genes that were common to silencing of both. Of genes that had significant expression changes compared to vector control, we found 229, 221 and 399 transcripts that were commonly regulated by CD28 and CD86 in KMS18, MM.1s and RPMI8226 respectively. Most transcripts were either upregulated (45.4 to 57.5%) or downregulated (28.8 to 41.5%) by silencing of either CD28 or CD86. A subset of transcripts (13.1 to 14%) showed a pattern of expression similar to CD86 - silencing of CD28 and CD86 had opposite effects on expression. This subset of transcripts may represent genes that are regulated by CD86 signaling, and may explain the difference in sensitivity to CD28 vs. CD86 silencing. Curiously, in KMS18 where this difference was not observed, RNA-Seq indicates that these cells are homozygous for a CD86 SNP that is associated with increased cancer susceptibility and lower transplant rejection, and may represent a hypomorphic allele.
Surprisingly, we did not observe any significant changes in either pro- or anti-apoptotic Bcl-2 genes in any cell line except for upregulation in minor transcripts of Bcl2L11 (Bim) in KMS18 cells. This change did not affect overall expression as confirmed by qRT-PCR. However, expression of several cell surface proteins associated with myeloma cell survival did change. Integrin-ß1 (ITGB1) and -ß7 (ITGB7) are surface molecules that facilitate both cell-matrix and cell-cell interactions, and have been implicated in myeloma growth, survival, and drug-resistance. Knockdown of CD28 or CD86 resulted in downregulation of ITGB7 that was confirmed by qRT-PCR. We also saw a reduction of ITGB7 at the cell surface with CD86 knockdown, but not with CD28. ITGB1 expression was reduced at the mRNA and cell surface levels with knockdown of CD86, but was induced with CD28 knockdown. Based on their patterns of expression, ITGB7 may be regulated by CD28 signaling, while ITGB1 may be downstream of CD86 signaling. These data indicate that CD28-86 signaling regulates the expression of integrins on the surface of myeloma cells. Because drug resistance has been linked to the myeloma cells’ interaction with the bone marrow stroma and its resident cells (CAM-DR), these surface molecules could be important mediators of CD28 and CD86 survival signaling. Taken together, our data indicate that targeting CD28-86 signaling is a promising therapeutic approach to CAM-DR, and may be a useful addition to current regimens against myeloma.
Disclosures:
Lonial: Onyx: Consultancy; Celgene: Consultancy; Millennium: Consultancy; Novartis: Consultancy; BMS: Consultancy; Sanofi: Consultancy; Onyx: Consultancy. Boise:Onyx Pharmaceuticals: Consultancy.
Defining the cell surface proteomic landscape of multiple myeloma reveals immunotherapeutic strategies and biomarkers of drug resistance
