LNK/SH2B3 Loss Exacerbates the Development of Myeloproliferative Neoplasms in CBL-deficient Mice

Abstract

Genetic variations of signaling modulator protein LNK (also called SH2B3) are associated with relatively mild myeloproliferative phenotypes in patients with myeloproliferative neoplasms (MPN). However, these variations can induce more severe MPN disease and even leukemic transformation when co-existing with other driver mutations. In addition to the most prevalent driver mutation JAK2V617F, LNK mutations have been clinically identified in patients harboring CBL inactivation mutations, but its significance remains unclear. Here, using a transgenic mouse model, we demonstrated that mice with the loss of both Lnk and Cbl exhibited severe splenomegaly, extramedullary hematopoiesis and exacerbated myeloproliferative characteristics. Moreover, a population of Mac1⁺ myeloid cells expressed c-Kit in aged mice. Mechanistically, we discovered that LNK could pull down multiple regulatory subunits of the proteosome. Further analysis confirmed a positive role of LNK in regulating proteasome activity, independent of its well-established function in signaling transduction. Thus, our work reveals a novel function of LNK in coordinating with the E3 ligase CBL to regulate myeloid malignancies.

Keywords: CBL; Leukemia; MPN; Proteasome; SH2B3.

Fig. 1

Cbl;Lnk DKO mice develop a severe myeloproliferative disease. A Complete blood count analysis of the periphery blood (PB) of the indicated mice. B Percentage of neutrophils (NE) and lymphocytes in PB of the indicated mice. C Red blood cell (RBC) and platelet count in PB of the indicated mice. Data from A-C were obtained from 5-month-old mice (WT, n = 4; Cbl^−/−, n = 3; Lnk^−/−, n = 4; DKO, n = 4). Each symbol represents a mouse. Error bars represent SE. p values were calculated by unpaired student’s t-test. *p < 0.05; **p < 0.01; ***p < 0.001; ns, not significant

Fig. 2

Leukemia transformation in old Cbl;Lnk DKO mice. A Total BM cellularity in the indicated mice. B Left panel shows the representative image of spleens from the indicated mice; Right panel is the statistical data of spleen weight. C Statistics of total cell numbers in the spleen from indicated mice. D-F Percentage of lineage cells in the BM (D), spleen (E), and PB (F) as analyzed by flow cytometry. G, H Percentage of lineage cells in the lung (G) and liver (H) as analyzed by flow cytometry. All data from A-H were obtained from 5-month-old mice (n = 3 for each genotype). Each symbol represents a mouse. Error bars represent SE. p values were calculated by unpaired student’s t-test. *p < 0.05; **p < 0.01; ***p < 0.001; ns, not significant

Fig. 3

Cbl;Lnk DKO HSPCs show superior transplantability. A-B Frequency (A) and number (B) of HSPC compartments in the BM of Cbl^−/− and DKO mice. C-D Frequency (C) and number (D) of HSPC compartments in the spleen of Cbl^−/− and DKO mice. E Competitive transplantation and analysis strategy of Cbl^−/− and DKO mice. F Donor-derived cell percentage in the PB of recipient mice at 4, 8, 12, 16 weeks post-transplantation. G Lineage cell percentage in donor-derived cells at 16 weeks post-transplantation. H Donor-derived LT-HSC cell number in the spleen of recipient mice after 16 weeks posttransplantation. Data from A-D were obtained from 5-month-old mice (n = 3 for each genotype). Data from F–H were obtained from PB of recipient mice (n = 6 for each group). Each symbol represents a mouse. Error bars represent SE. p values were calculated by unpaired student’s t-test. *p < 0.05; **p < 0.01; ***p < 0.001; ns, not significant

Fig. 4

LNK positively regulates proteasome activity. A Proteasome subunits identified in LNK immunoprecipitation/mass spectrometry (IP/MS). B Immunoblot of PSMC3, PSMC5 and PSMD13 in shLuc control and two separate LNK-deficient TF-1 cell lines. C The chymotrypsin-like and caspase-like proteolytic activities of the proteasome from shLuc control and LNK-deficient TF-1 cell lines. Data from C were independently repeated three times. Error bars represent SE. p values were calculated by unpaired student’s t-test. **p < 0.01; ***p < 0.001; ns, not significant

Fig. 5

LNK loss leads to an altered proteome. A Immunoblot of LNK in shLuc control and shLNK (KD) TF-1 cells. JAK2 is a positive control that is known to be upregulated in LNK KD cells. B Schematic representation of SILAC technique and mass spectrometry. to determine the proteomes of shLuc control and LNK KD cells. C Gene pathway enrichment of differentially expressed genes (DEG) identified by SILAC/MS in LNK KD cells compared with control cells. Proteins with fold changes larger or smaller than 50% were defined as DEG. Red and blue bars indicate up- and down-regulated pathways, respectively. D Proteins with most significantly up- (red dots) and down-regulated (blue dots) changes in LNK KD cells were indicated

Fig. 6

LNK loss does not affect the protein synthesis rate of HSPCs. A Representative flow plots of HSPC subpopulations in bone marrows of WT and Lnk KO mice. B Percentages of HSCs and MPPs in bone marrows of WT (n = 6) and Lnk KO (n = 6) mice as shown in A. C Statistics of relative protein synthesis rates in the indicated populations of bone marrow cells from WT (n = 6) and Lnk KO (n = 6) mice. Each symbol in B and C represents a mouse. Error bars represent SE. p values were calculated by unpaired student’s t-test. *p < 0.01; ***p < 0.001; ns, not significant