SH2B3 (LNK) mutations from myeloproliferative neoplasms patients have mild loss of function against wild type JAK2 and JAK2 V617F

Abstract

Somatic point mutations in the PH domain of SH2B3 (LNK), an adaptor protein that is highly expressed in haematopoietic cells, were recently described in patients with myeloproliferative neoplasms. We studied the effect of these mutations on the JAK2 signalling pathway in cells expressing either wild type JAK2 or the JAK2 V617F mutation. Compared to wild type SH2B3, PH domain mutants have mild loss of function, with no evidence for a dominant-negative effect. Mutants retain binding capacity for JAK2, an established SH2B3 target, as well as for the adaptor proteins 14-3-3 and CBL. Our data suggest that the loss of SH2B3 inhibitory function conferred by the PH domain mutations is mild and may collaborate with JAK2 V617F and CBL mutations in order to promote either the development or the progression of myeloproliferative neoplasms.

Figure 1. Sh2b3 PH domain mutants compared to WT Sh2b3 have a decreased inhibitory effect on BaF3-E and BM colonies

(A,B) BaF3-E cells were transfected with either Sh2b3 wild type (WT) or mutant Sh2b3. GFP-positive cells were sorted 2 days after transfection and plated in equal cell densities with either 10 iu Epo/ ml (A) or 10 iu IL3/ ml (B) in soft agar. Colonies were counted after 10 days. Results represent the mean ± SD of triplicate samples. (C,D) BM enriched progenitor cells from Sh2b3 knock-out (KO) (C) or WT (D) mice were infected on 2 consecutive days with either Sh2b3 WT or mutant Sh2b3. GFP positive cells were sorted 2 days after the second infection and equal cell number plated in cytokine rich methylcellulose media (M3434, STEMCELL Technologies Inc., Vancouver, BC). Total colonies number was scored after 8–9 days. Results represent the mean ± SD of triplicate samples. (E) Western blot demonstrating SH2B3 expression in sorted BaF3-E cells transfected with Sh2b3 variants. * Significantly different from WT SH2B3, p<0.05.

Figure 2. Epo-dependent phosphorylation of Stat5 and Erk is inhibited by the Sh2b3 PH domain mutants less than WT Sh2b3

BaF3-E cells transfected with either WT or mutant Sh2b3 were serum-starved for 6 h and stimulated with 10 iu/ml Epo for 15 and 60 min. Phosphorylation of Stat5 and Erk was assessed in GFP-positive cells by flow analysis, using PE-conjugated STAT5 (pY694) and Alexa Fluor 647 conjugated ERK 1/2 (pY202/204) antibodies. A representative of 2 independent experiments is shown. (A) Gating of BaF3-E GFP-positive cells. (B) STAT5 and ERK phosphorylation. Mean fluorescence intensities (MFIs) for each time point were normalized to un-stimulated cells and are represented as fold-change. (C) Western blot of proteins from BaF3-E transduced with Sh2b3.

Figure 3. IL3-dependent STAT5 phosphorylation is inhibited less by the Sh2b3 PH domain mutants compared to WT Sh2b3

Murine 5FU-enriched BM progenitor cells transduced with either WT or mutant Sh2b3 were serum-starved for 2 h and stimulated with 10 iu/ml IL3 for 15 and 60 min. Phosphorylation of STAT5 and ERK was assessed in GFP-positive cells by flow analysis, using PE-conjugated STAT5 (pY694) and Alexa Fluor 647 conjugated ERK 1/2 (pY202/204) antibodies. A representative of 2 independent experiments is shown. (A) Gating of GFP-positive cells. (B) STAT5 phosphorylation. MFIs for each time point were normalized to unstimulated cells and are represented as fold-change.

Figure 4. Growth and signal transduction of JAK2 V617F HEL cells is inhibited less by the Sh2b3 PH domain mutants compared to WT Sh2b3

HEL cells (JAK2 V617F) were transduced with either WT or mutant Sh2b3. (A) GFP-positive cells were sorted, plated at equal cell densities in liquid culture and counted every 2 days. Results represent the mean ± SD of triplicate samples. Transduced cells were serum-starved for 8 h and studied for STAT5 and ERK phosphorylation. (B) Gating of GFP-positive cells. (C) STAT5 and ERK phosphorylation. MFIs were normalized to the vector control transduced cells and are represented as fold-change. (D) Western blot analysis of proteins from sorted GFP-positive HEL cells transduced with either a vector control, WT Sh2b3, Sh2b3 RE or EQ. Cells were serum-starved for 8 h and analysed for STAT5 and ERK phosphorylation. Band intensity was normalized to ACTB and represent fold change compared to the vector control.

Figure 5. SH2B3 PH domain mutants retain their binding to JAK2, CBL and 14-3-3

293T cells were co-transfected with either WT JAK2 (A) or CBL (C) and either WT or mutant SH2B3. Protein lysates were immunoprecipitated with V5 antibodies and analysed by Western blot as indicated. (B) 293T cells were transfected with either WT or mutant SH2B3. Protein lysates were immunoprecipitated with V5 antibodies and analysed by Western blot as indicated.