The type III receptor tyrosine kinase FLT3 is frequently mutated in acute myeloid leukemia. Oncogenic FLT3 mutants display constitutive activity leading to aberrant cell proliferation and survival. Phosphorylation on several critical tyrosine residues is known to be essential for FLT3 signaling. Among these tyrosine residues, Y842 is located in the so-called activation loop. The position of this tyrosine residue is well conserved in all receptor tyrosine kinases. It has been reported that phosphorylation of the activation loop tyrosine is critical for catalytic activity for some but not all receptor tyrosine kinases. The role of Y842 residue in FLT3 signaling has not yet been studied. In this report, we show that Y842 is not important for FLT3 activation or ubiquitination but plays a critical role in regulating signaling downstream of the receptor as well as controlling receptor stability. We found that mutation of Y842 in the FLT3-ITD oncogenic mutant background reduced cell viability and increased apoptosis. Furthermore, the introduction of the Y842 mutation in the FLT3-ITD background led to a dramatic reduction in in vitro colony forming capacity. Additionally, mice injected with cells expressing FLT3-ITD/Y842F displayed a significant delay in tumor formation, compared to FLT3-ITD expressing cells. Microarray analysis comparing gene expression regulated by FLT3-ITD versus FLT3-ITD/Y842F demonstrated that mutation of Y842 causes suppression of anti-apoptotic genes. Furthermore, we showed that cells expressing FLT3-ITD/Y842F display impaired activity of the RAS/ERK pathway due to reduced interaction between FLT3 and SHP2 leading to reduced SHP2 activation. Thus, we suggest that Y842 is critical for FLT3-mediated RAS/ERK signaling and cellular transformation.Electronic supplementary materialThe online version of this article (doi:10.1007/s00018-017-2494-0) contains supplementary material, which is available to authorized users.

The type III receptor tyrosine kinase FLT3 is one of the most commonly mutated oncogenes in acute myeloid leukemia (AML). Inhibition of mutated FLT3 in combination with chemotherapy has displayed promising results in clinical trials. However, one of the major obstacles in targeting FLT3 is the development of resistant disease due to secondary mutations in FLT3 that lead to relapse. FLT3 and its oncogenic mutants signal through associating proteins that activate downstream signaling. Thus, targeting proteins that interact with FLT3 and their downstream signaling cascades can be an alternative approach to treat FLT3-dependent AML. We used an SH2 domain array screen to identify novel FLT3 interacting proteins and identified ABL2 as a potent interacting partner of FLT3. To understand the role of ABL2 in FLT3-mediated biological and cellular events, we used the murine pro-B cell line Ba/F3 as a model system. Overexpression of ABL2 in Ba/F3 cells expressing an oncogenic mutant of FLT3 (FLT3-ITD) resulted in partial inhibition of FLT3-ITD-dependent cell proliferation and colony formation. ABL2 expression did not alter the kinase activity of FLT3, its ubiquitination or its stability. However, it partially blocked FLT3-induced AKT phosphorylation without affecting ERK1/2 and p38 activation. Taken together our data suggest that ABL2 acts as negative regulator of signaling downstream of FLT3.

Insulin action initiates a series of phosphorylation events regulating cellular differentiation, growth and metabolism. We have previously discovered, in a mass spectrometry-based phosphoproteomic study, that insulin/IGF-1 signalling induces phosphorylation of retinoid x receptor alpha (RXRα) at S22 in mouse brown pre-adipocytes. Here, we show that insulin induces the phosphorylation of RXRα at S22 in both brown precursor and mature adipocytes through a pathway involving ERK, downstream of IRS-1 and −2. We also found that RXRα S22 phosphorylation is promoted by insulin and upon re-feeding in brown adipose tissue in vivo, and that insulin-stimulated S22 phosphorylation of RXRα is dampened by dietinduced obesity. We used Rxra knockout cells re-expressing wild type (WT) or S22A nonphosphorylatable forms of RXRα to further characterize the role of S22 in brown adipocytes. Knockout of Rxra in brown pre-adipocytes resulted in decreased lipid accumulation and adipogenic gene expression during differentiation, and re-expression of RxraWT alleviated these effects. However, we observed no significant difference in cells re-expressing the RxraS22A mutant as compared with the cells reexpressing RxraWT. Furthermore, comparison of gene expression during adipogenesis in the WT and S22A re-expressing cells by RNA sequencing revealed similar transcriptomic profiles. Thus, our data propose a dispensable role for RXRα S22 phosphorylation in adipogenesis and transcription in differentiating brown pre-adipocytes.

Activation of thermogenic adipose tissue is linked to improved metabolic outcomes in mice and humans. Dissipation of energy as heat during thermogenesis relies on sufficient innervation of fat by sympathetic nerve fibers, a process recently proposed to be regulated by the adipose-specific calsyntenin3b (Clstn3b)-S100b axis. Here we aimed 1) to assess enrichment patterns of CLSTN3b, S100b as well as the previously annotated neuronal CLSTN3a in perirenal brown and subcutaneous white human fat specimens, and 2) to investigate if the novel Clstn3b is dynamically regulated by changes in environmental temperatures and nutritional stress in thermogenic adipose tissues in mice. We provide evidence for CLSTN3b enrichment in multilocular perirenal fat located anatomically in the proximity to both the adrenal gland and sympathetic nerve bundles innervating the kidney in humans. Moreover, transcript levels of CLSTN3b, but not S100b or CLSTN3a, positively correlate with uncoupling protein 1 (UCP1) expression in human adipose tissue. Our results further show that Clsnt3b is preferentially expressed in brown adipocytes and is highly responsive to changes in environmental temperature and obesity state in mice. Collectively, this brief communication highlights CLSTN3b as a hallmark of thermogenic adipose depots in mice and humans.