Abstract
The BCR-ABL oncogene activates several signaling pathways, most notably by constitutive phosphorylation of the signal transducer and activator of transcription protein 5 (STAT5). After phosphorylation and nuclear translocation, STAT5 transcriptionally activates numerous genes responsible for proliferation, survival and differentiation of hematopoietic stem and progenitor cells. Among the STAT5 target genes are suppressor of cytokine signaling (SOCS) proteins. SOCS proteins inhibit JAK kinases by multiple mechanisms, thereby terminating cytokine signaling in a classical negative feedback loop. The SOCS family of proteins comprises eight members: cytokine-inducible SRC homology 2 (SH2) domain protein (CIS) and SOCS1–SOCS7. SOCS1 is frequently silenced by hypermethylation in multiple myeloma and inactivating mutations have been found in Hodgkin lymphoma with consecutive increase in JAK2 kinase activity. More recently, we identified SOCS1 as a “conditional oncogene” in the context of the FLT3-ITD oncogene (Reddy et al, Blood 2012): SOCS1 significantly enhanced FLT3-ITD-mediated myeloid transformation, both in vitro and in vivo. We hypothesized that this may be a more general mechanism of transformation and therefore analyzed the role of SOCS proteins in BCR-ABL mediated transformation and leukemogenesis.
First, we investigated gene expression levels of SOCS proteins in BCR-ABL positive (versus BCR-ABL negative) cell lines and primary ALL long term-cultured cells. Upon treatment with the BCR-ABL inhibitor imatinib, mRNA expression levels of CIS and SOCS1-4 were reduced. SOCS5-7 did not exhibit any changes and were non-responsive to ABL-kinase inhibition. In lineage-depleted primary murine bone marrow retrovirally transduced with BCR-ABL, high induction of CIS and SOCS1-3 mRNA was detected, while SOCS4-7 showed only minor changes. When overexpressed in IL-3 dependent cell lines, SOCS1 led to a very rapid cell death within few days. Similar effects were demonstrated for CIS and SOCS2 overexpression, however, with a slower kinetics. In contrast, BCR-ABL transduced cells were insensitive to SOCS1 overexpression. In colony formation assays performed with primary hematopoietic cells, expression of SOCS1 led to a significant decrease of colony numbers. Interestingly, co-expression of SOCS1 and BCR-ABL (hereafter abbreviated as SOCS1/BCR-ABL) also lowered colony numbers compared to cells expressing BCR-ABL alone. However, when cells were subjected to interferon alpha or interferon gamma treatment, SOCS1/BCR-ABL positive cells displayed higher colony numbers and gained a growth advantage over BCR-ABL expressing cells, since anti-proliferative effects of the cytokines were inhibited by the presence of SOCS1. A careful analysis of the downstream signaling cascade of BCR-ABL and SOCS1/BCR-ABL expressing cells did not demonstrate any differences in the phosphorylation of AKT, ERK1/2 and STAT5. However, when BCR-ABL was inhibited by imatinib, STAT5 phosphorylation was significantly decreased in SOCS1/BCR-ABL transduced cells. Finally, the influence of SOCS1 in BCR-ABL mediated leukemia was investigated in a murine bone marrow transplantation model. BCR-ABL or SOCS1/BCR-ABL expressing cells led to disease formation with a chronic myeloid leukemia-like phenotype. Interestingly, the co-expression of SOCS1 and BCR-ABL prolonged disease latency, as opposed to the phenotype observed with FLT3-ITD (where SOCS1 co-expression shortened latency). In this setting SOCS1 acts as a tumor suppressor, protecting BCR-ABL transformed cells from rapid disease development, and a molecular analysis will be presented.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.