Microglia are highly dynamic cells in the brain. Their functional diversity and phenotypic versatility brought microglial energy metabolism into the focus of research. Although it is known that microenvironmental cues shape microglial phenotype, their bioenergetic response to local nutrient availability remains unclear. In the present study effects of energy substrates on the oxidative and glycolytic metabolism of primary - and BV-2 microglial cells were investigated. Cellular oxygen consumption, glycolytic activity, the levels of intracellular ATP/ADP, autophagy, mTOR phosphorylation, apoptosis and cell viability were measured in the absence of nutrients or in the presence of physiological energy substrates: glutamine, glucose, lactate, pyruvate or ketone bodies. All of the oxidative energy metabolites increased the rate of basal and maximal respiration. However, the addition of glucose decreased microglial oxidative metabolism and glycolytic activity was enhanced. Increased ATP/ADP ratio and cell viability, activation of the mTOR and reduction of autophagic activity were observed in glutamine-supplemented media. Moreover, moderate and transient oxidation of ketone bodies was highly enhanced by glutamine, suggesting that anaplerosis of the TCA-cycle could stimulate ketone body oxidation. It is concluded that microglia show high metabolic plasticity and utilize a wide range of substrates. Among them glutamine is the most efficient metabolite. To our knowledge these data provide the first account of microglial direct metabolic response to nutrients under short-term starvation and demonstrate that microglia exhibit versatile metabolic machinery. Our finding that microglia have a distinct bioenergetic profile provides a critical foundation for specifying microglial contributions to brain energy metabolism.

CD49d and CXCR4 are key determinants of interactions between chronic lymphocytic leukemia (CLL) tumor cells and their microenvironment. In this study, we investigated the effect of CD49d and CXCR4 expressions on survival of CLL cells. Primary CLL cells were cultured with CD49d ligand, VCAM-1, or bone marrow stromal cells (BMSCs); then, apoptosis and immunophenotype analyses were performed. VCAM-1 treatment could not induce direct apoptosis protection or immunophenotype change on the CD49d-expressing CLL cells, but resulted in actin reorganization. The BMSC-induced apoptosis protection was independent from the presence of CD49d expression of CLL cells, but showed an inverse correlation with their CXCR4 expression level. We suppose that CD49d contributes to enhanced survival of leukemic cells by mediating migration to the protective microenvironment, not by direct prevention of apoptosis. Moreover, CLL cells with low CXCR4 expression represent a subpopulation that is more dependent on the microenvironmental stimuli for survival, and show increased "death by neglect" when separated from the supportive niche.

Chronic lymphocytic leukemia (CLL) is characterized by a neoplastic B-cell population coexpressing CD5 and CD23; however, the expression of CD23 is variable. In human, two isotypes of CD23 have been identified and related to different functions. The aim of our study was to investigate the relative expression of the two CD23 isotypes in CLL and find possible correlation with other prognostic factors. The expression of CD23 isotypes was analyzed in 54 cases of CLL by polymerase chain reaction (PCR) and quantitative real-time PCR. The immunophenotype of CLL cells was characterized by flow cytometry. We demonstrated a higher CD23a than CD23b expression of CLL cells. Our results also revealed two subsets of CLL cases with a distinct CD23 isotype expression pattern. Thirty-two percent of the cases (group CLL1) showed both low mRNA level of CD23 isotypes and high protein levels of CD20 and CD38 in contrast to group CLL2 with high CD23 mRNA levels. By correlating these results to the presence of prognostic factors determined by fluorescence in situ hybridization, we found that the majority of the cases of group CLL1 (14/17) carried trisomy 12. In summary, our results confirm a high CD23a/CD23b ratio of the CLL cells and demonstrate that in a subset of CLL cases, low CD23 expression together with high CD20 and CD38 expressions may serve as a surrogate for trisomy 12.

BackgroundMultiple studies concluded that oncometabolites (e.g. D-2-hydroxyglutarate (2-HG) related to mutant isocitrate dehydrogenase 1/2 (IDH1/2) and lactate) have tumour promoting potential. Regulatory mechanisms implicated in the maintenance of oncometabolite production have great interest. mTOR (mammalian target of rapamycin) orchestrates different pathways, influences cellular growth and metabolism. Considering hyperactivation of mTOR in several malignancies, the question has been addressed whether mTOR operates through controlling of oncometabolite accumulation in metabolic reprogramming.MethodsHT-1080 cells – carrying originally endogenous IDH1 mutation – were used in vitro and in vivo. Anti-tumour effects of rapamycin were studied using different assays. The main sources and productions of the oncometabolites (2-HG and lactate) were analysed by 13C-labeled substrates. Alterations at protein and metabolite levels were followed by Western blot, flow cytometry, immunohistochemistry and liquid chromatography mass spectrometry using rapamycin, PP242 and different glutaminase inhibitors, as well.ResultsRapamycin (mTORC1 inhibitor) inhibited proliferation, migration and altered the metabolic activity of IDH1 mutant HT-1080 cells. Rapamycin reduced the level of 2-HG sourced mainly from glutamine and glucose derived lactate which correlated to the decreased incorporation of 13C atoms from 13C-substrates. Additionally, decreased expressions of lactate dehydrogenase A and glutaminase were also observed both in vitro and in vivo.ConclusionsConsidering the role of lactate and 2-HG in regulatory network and in metabolic symbiosis it could be assumed that mTOR inhibitors have additional effects besides their anti-proliferative effects in tumours with glycolytic phenotype, especially in case of IDH1 mutation (e.g. acute myeloid leukemias, gliomas, chondrosarcomas). Based on our new results, we suggest targeting mTOR activity depending on the metabolic and besides molecular genetic phenotype of tumours to increase the success of therapies.Electronic supplementary materialThe online version of this article (doi:10.1186/s13046-017-0544-y) contains supplementary material, which is available to authorized users.

The primary aim of this study was to determine mTOR-pathway activity in primary central nervous system lymphoma (PCNSL), which could be a potential target for therapy. After demonstrating that p-S6 positivity largely exceeded mTOR activity, we aimed to identify other pathways that may lead to S6 phosphorylation. We measured mTOR activity with immunohistochemistry for p-mTOR and its downstream effectors p(T389)-p70S6K1, p-S6, and p-4E-BP1 in 31 cases of PCNSL and 51 cases of systemic diffuse large B-cell lymphoma (DLBCL) and evaluated alternative S6 phosphorylation pathways with p-RSK, p(T229)-p70S6K1, and PASK antibodies. Finally, we examined the impact of PASK inhibition on S6 phosphorylation on BHD1 cell line. mTOR-pathway activity was significantly less frequent in PCNSL compared with DLBCL. p-S6 positivity was related to mTOR-pathway in DLBCL, but not in PCNSL. Among the other kinases potentially responsible for S6 phosphorylation, PASK proved to be positive in all cases of PCNSL and DLBCL. Inhibition of PASK resulted in reduced expression of p-S6 in BHD1-cells. This is the first study demonstrating an mTOR independent p-S6 activity in PCNSL and that PASK may contribute to the phosphorylation of S6. Our findings also suggest a potential role of PASK in the pathomechanism of PCNSL and in DLBCL.