Cooperative STAT/NF-κB signaling regulates lymphoma metabolic reprogramming and aberrant GOT2 expression

Abstract

Knowledge of stromal factors that have a role in the transcriptional regulation of metabolic pathways aside from c-Myc is fundamental to improvements in lymphoma therapy. Using a MYC-inducible human B-cell line, we observed the cooperative activation of STAT3 and NF-κB by IL10 and CpG stimulation. We show that IL10 + CpG-mediated cell proliferation of MYC^low cells depends on glutaminolysis. By ¹³C- and ¹⁵N-tracing of glutamine metabolism and metabolite rescue experiments, we demonstrate that GOT2 provides aspartate and nucleotides to cells with activated or aberrant Jak/STAT and NF-κB signaling. A model of GOT2 transcriptional regulation is proposed, in which the cooperative phosphorylation of STAT3 and direct joint binding of STAT3 and p65/NF-κB to the proximal GOT2 promoter are important. Furthermore, high aberrant GOT2 expression is prognostic in diffuse large B-cell lymphoma underscoring the current findings and importance of stromal factors in lymphoma biology.

Fig. 1

Activation of cell metabolism and global gene expression by combined stimulation of P493-6 cells with IL10 and CpG. a Heat map of changes in intracellular metabolite concentrations depicted as Log2FC after IL10 and/or CpG stimulation of P493-6 MYC^low cells in relation to unstimulated cells. Heat map of gene expression changes associated with b glycolysis (KEGG-term) and c glutaminolysis (KEGG-term), respectively, after IL10 and/or CpG stimulation of P493-6 MYC^low cells presented as Log2FC. Mean of three independent experiments is shown in all heat maps. Effects on gene expression were analyzed by linear regression and p-values for the IL10:CpG interaction terms were calculated (adjusted by Benjamini–Hochberg). Positive synergistic interactions are marked with a star (*p < 0.05). Relative cell counts of d IL10 + CpG-stimulated P493-6 MYC^low cells and e unstimulated P493-6 MYC^high cells grown in media with and without either glucose (–Glc) or glutamine (–Gln). Mean ± SD of three independent experiments are given

Fig. 2

Proliferation of IL10 + CpG-stimulated and MYC-overexpressing cells is dependent on different glutamine-derived metabolites. a Mean isotopic enrichment of metabolites in unstimulated and IL10 + CpG-stimulated P493-6 MYC^low, as well as unstimulated P493-6 MYC^high cells treated with ¹³C₅-glutamine within the text only 13C5-Gln is used?. b Scheme of oxidative decarboxylation (blue) and reductive carboxylation (red). Both can be distinguished by isotopic labelling in mass spectrometry. c Isotopic fractions of citrate in IL10 + CpG-stimulated P493-6 MYC^low and P493-6 MYC^high cells. Mass spectra were analysed according to the different masses of citrate. Data on 1-¹³C-glutamine within the main text 1-13C-Gln is used ? labeling are summarized in Supplementary Fig. 5. Fractions of d m + 1 ¹⁵N isotopologue in selected amino acids and e ¹⁵N-enriched nucleobases in unstimulated and IL10 + CpG-stimulated P493-6 MYC^low, as well as P493-6 MYC^high cells. f,g Relative cell counts of IL10 + CpG-stimulated P493-6 MYC^low and P493-6 MYC^high cells grown in Gln-deprived and metabolite-treated media. Cells were treated with either f TCA metabolites (Pyr = pyruvate, α-KG = dimethyl 2-oxoglutarate, Mal = diethyl malate, OAA = oxaloacetate) or g aspartate (Asp) and nucleobases (A = adenine, T = thymine) for 24 h. For all plots, error bars represent mean ± SD of three independent experiments and results from Bonferroni post hoc tests on a one-way ANOVA results are given (*p < 0.05, **p < 0.01, ***p < 0.001)

Fig. 3

Proliferation of P493-6 and B-cell lymphoma cells requires transaminase activity. a Inhibition of cell doubling of IL10 + CpG-stimulated P493-6 MYC^low and P493-6 MYC^high cells after 24 h of treatment with CB-839 or AOA. b Inhibition of cell doubling of the B-cell lymphoma cell lines CA-46, OCI-Ly3, and L-428 using CB-839 or AOA treatment for 24 h. c Intracellular amino acid abundance in IL10 + CpG-stimulated MYC^low and MYC^high cells 24 h after AOA treatment and stimulation. For each amino acid data were normalized relative to the corresponding values of DMSO-treated cells. For all plots, error bars represent mean ± SD of three independent experiments and results from Bonferroni post hoc tests on a one-way ANOVA are given (*p < 0.05, **p < 0.01, ***p < 0.001)

Fig. 4

GOT2 supports the proliferation of IL10 + CpG-stimulated and MYC-overexpressing cells. a Transient knockdown of GOT1 and GOT2 in P493-6 cells was performed by siRNA transfection. Knockdown efficiencies (GOT*/tubulin) relative to scrb control are provided under the images. b Relative cell numbers of IL10 + CpG-stimulated P493-6 MYC^low and P493-6 MYC^high after GOT1 or GOT2 knockdown depicted in a. Relative cell numbers of c IL10 + CpG-stimulated P493-6 MYC^low and d unstimulated P493-6 MYC^high cells after GOT2 knockdown and indicated α-KG, Asp, A or T addition as described in Fig. 2. e Transient knockdown of GPT2 in P493-6 cells performed by siRNA transfection. Knockdown efficiencies (GPT2/tubulin) relative to scrb control are provided under the images. f Relative cell numbers of IL10 + CpG-stimulated P493-6 MYC^low and P493-6 MYC^high cells after GPT2 knockdown as depicted in e. All experiments were performed in triplicates. For b–d and f, error bars represent mean ± SD of three independent experiments and results from Bonferroni post hoc tests on a one-way ANOVA are given (*p < 0.05, **p < 0.01, ***p < 0.001)

Fig. 5

Context dependent role of GOT2 in lymphoma cells. a Immunoblot of GOT2 in CA-46, OCI-Ly3, and L-428 lymphoma cells 24 h after siRNA-mediated knockdown. Knockdown efficiencies (GOT2/tubulin) relative to scrb control are presented under the images. b Relative cell numbers of CA-46, OCI-LY3, and L-428 cells after GOT2 knockdown and metabolite treatment. Metabolites were used as described in Fig. 2. Error bars represent mean ± SD of three independent experiments and results from Bonferroni post hoc tests on a one-way ANOVA are given (*p < 0.05, **p < 0.01, ***p < 0.001)

Fig. 6

GOT2 expression in B-cells is up-regulated by NF-κB/STAT3 activation or MYC. a GOT2 mRNA expression in 24h-stimulated P493-6 MYC^low and unstimulated P493-6 MYC^high cells after inhibition of NF-κB signalling by ACHP or JAK/STAT signalling by Ruxolitinib. b Representative immunoblot of STAT3 and p65 protein levels 24 h after STAT3 (STAT3si) or RELA siRNA (RELAsi) transfection of P493-6 cells. Knockdown efficiencies (protein/GAPDH) relative to scrb control are presented under the images. c GOT2 mRNA expression in 24h-stimulated P493-6 MYC^low cells after siRNA transfection shown in b. d Chromatin immunoprecipitation of STAT3 and p65 1 h after stimulation of P493-6 MYC^low cells. Displayed are average qPCR results and technical errors (SD) of the GOT2 promoter (– 20 bp) from a representative stimulation out of three. Enrichment was calculated relative to an inactive control region (PRAME). e Full tyrosine and serine phosphorylation of STAT3 is only achieved after IL10 + CpG costimulation as revealed by immunoblot analysis. Cells were stimulated with IL10, CpG or both for 30 min. f GOT2 mRNA expression of TMD-8 and OCI-Ly3 cells in comparison with BJAB and CA-46 cells 24 h after ACHP or rRuxolitinib treatment. For additional details about the used cell lines also refer to Supplementary Fig. 7a. With the exception of d, all values are given as mean ± SD of three independent replicates. Results from Bonferroni post hoc test on a two-way ANOVA are presented (***p < 0.001, **p < 0.01)

Fig. 7

GOT2 is overexpressed in lymphoma subtypes and correlates with worse overall survival in DLBCL patients. a Microarray gene expression data of GOT2 obtained at ONCOMINE^,. Expression of normal B-cell controls (1 = B lymphocytes (n = 5), 2 = centroblasts (n = 5), 3 = memory B -cells (n = 5), 4 = naive pre-germinal B- cells (n = 5)), and lymphoma samples (5 = BL (n = 17), 6 = CLL (n = 34), 7 = DLBCL (n = 32), and 8 = MCL (n = 8)) are given. Boxes are showing median (middle line), upper and lower quantile (edges), minimum and maximum (bars), and outliners (dots). b Immunohistological staining of GOT2 in representative samples of tonsils, BLs, DLBCLs, and HLs (see also Table 1). The scale bar represents 50 µm. In Supplementary Fig. 12, a corresponding GOT2 staining for the liver, lung, muscles, and skin is provided showing a homogeneous GOT2 expression in hepatocytes. c Kaplan–Meier curves of DLBCL patients (n = 157) treated with R-CHOP. Patients were grouped according to their GOT2 expression relative to the median of the group (high > median and low ≤ median). Additional details using a Cox proportional hazards regression model to estimate the clinical outcome of patients with DLBCL in relation to the expression of GOT2 are presented within Supplementary Table 1