Genetic Screening for Myeloma Cell Susceptibility to NK Cell Mediated Lysis Using Large Lentiviral shRNA Libraries

Multiple myeloma (MM) is a B cell neoplasm characterized by clonal expansion of malignant plasma cells in the bone marrow. Despite the use of new drugs such as thalidomide and bortezomib, MM remains an incurable disease. Successful treatment of MM with allogeneic stem cell transplantation suggests that MM is susceptible to immunologic approaches. In this setting, donor immune cells including T, B and NK cells are able to mediate a potent GVM effect resulting in tumor regression. However in contrast to other hematologic malignancies in which responses are more durable, only 10–20% of patients with MM are disease free at 5 years after transplant and the majority of patients develop resistant disease and relapse in the first year after treatment. The mechanisms why MM cells become resistant and escape immunologic control are unknown. RNA interference (RNAi) is a new powerful tool for analysis of specific gene functions in mammalian cells and the development of shRNAi libraries targeting thousands of genes permits large-scale loss-of-function studies in human cells. To study whether this approach can be used to identify mechanisms of immunologic resistance and develop new methods to enhance susceptibility of MM to immunologic rejection, we undertook a cell-cell interaction based screen using a large lentiviral shRNA library containing 476 protein kinases and 180 phosphatases that represent 88% and 80%, respectively, of known NCBI sequences with these functions. The library also included 372 genes representing tumor suppressors, DNA binding proteins as well as irrelevant shRNAs as controls. Each gene was targeted by 5 or more shRNAs tested individually in a 384 well format using robotic manipulations. A total of 6,144 individual shRNAs were included in this screen. 5,000 IM-9 MM cells/well were plated in 384 well plates in 5 replicate sets and infected with each shRNA. After 24 hours of infection, puromycin was added to 1 set. After 48 hours, cell viability was determined in the puromycin and 1 control set using luminescence assay to assess infection efficiency and potential toxicity for each shRNA. NKL effector cells (CD3 negative, IL-2-dependent natural killer cell line) were added to each well in the remaining 3 sets. After 12 hours of incubation, supernatants were harvested and the level of interferon gamma (INF-g) released by the NKL cells was measured using anti-INF-g beads and a BD FACSCanto II flow cytometer equipped with a high-throughput platform. Using the viability measurements in the 2 replicates treated with +/− puromycin, we excluded from further analysis all wells in which the shRNA infection rate was less than 70%. Levels of IFN-g secretion by NKL cells in response to IM-9 MM target cells in the remaining wells were normalized and the top 10 percentile of shRNAs inducing increased secretion of IFN-g was identified. To reduce the likelihood that effects identified in this screen were due to off-target effects of the shRNAs, only genes that were positive for 2 or more independent shRNAs were selected for further analysis. With this method of ranking, a total of 121 genes were selected for further analysis. Several common signaling pathways including ERK, PI3K and TGFB receptor family were each identified by multiple different genes. The screening also identified several membrane proteins including 2 members of the Ephrin family suggesting that these genes or pathways can play a role in increased resistance of MM cells to NK mediated lysis. To validate the screening results, the same shRNAs were used to produce stable IM-9 cells with constitutive knockdown of the genes of interest and re-tested by INF-g ELISA or 51Cr release assay. The selective down-regulation of the specific proteins was also evaluated using western blot analysis. For genes in the ERK pathway, the screening results were further validated using pharmacological inhibitors of this pathway. Cells were treated for 2 hours with different drug concentrations and incubated with NKL cells in a 4 hour 51Cr release assay. We have begun to investigate the role of some of these genes in other MM cell lines as well as in other tumors using different NK effector cells. In summary, these studies demonstrate that large-scale lentiviral shRNA libraries can be used in high-throughput immune functional screens to identify new pathways that may be involved in modulating the sensitivity of tumor cells to immune mediated destruction and may lead to the identification of new approaches for enhancing susceptibility of tumor cells to NK cell killing in vivo.