Identification of the Integrated Prognostic Signature Associated with Immuno-relevant Genes and Long Non-coding RNAs in Acute Myeloid Leukemia.

BACKGROUND: Like other cancers, considerable effort has been made in acute myeloid leukemia (AML) to identify prognostic genes and long noncoding RNAs (lncRNAs) with their potential clinical applications. However, to date, no integrated prognostic model has been developed that combines both gene expression and lncRNAs as a singular approach in AML. METHOD: Comprehensive bioinformatic approaches (Weighted gene co-expression network analysis, Univariate Cox regression analyses, Pearson correlation, LASSO-Cox regression, Wilcoxon test) were used to construct the signature and to define high- and low-risk groups in AML datasets. ESTIMATE and CIBERSORT algorithms were applied to investigate the potential impact of infiltrating immune cells based on the obtained signature in tumor microenvironment. In addition, gene ontology (GO) and KEGG enrichment were applied to explore the potential function of the signature. RESULTS: Herein, we focused on immune-related genes (IRGs) and immune-related long noncoding RNAs (IRlncRNAs) and constructed an integrated prognostic immunorelevant signature in AML. The obtained signature exhibit five IRGs (DAXX, PSMB8, CSRP1, RAC2 and PTPN6) and one IRlncRNA (AC080037.2) and is strictly associated with age and FAB (French-American-British classification). Importantly, the high-risk AML group (defined by the signature) correlated positively with three types of scores (immune score, stroma score, and ESTIMATE score). We also identified a few immune cells (resting mast cells and monocytes) potentially involved in the correlation between signature and survival of AML patients. The prognostic ability of the obtained signature was tested in the training cohort and then validated in both test and total cohorts. The pathway enrichment analysis confirmed the possible immune- related role of the signature. CONCLUSION: We constructed an integrated prognostic signature comprising five immune-related protein-coding genes (IRPCG) (DAXX, PSMB8, CSRP1, RAC2, and PTPN6) and one immune-related lncRNA (AC080037.2) that may serve as potential biomarkers for predicting survival and further stratifying AML patients.

mTORC1-Activated Monocytes Increase Tregs and Inhibit the Immune Response to Bacterial Infections.

The TSC1/2 heterodimer, a key upstream regulator of the mTOR, can inhibit the activation of mTOR, which plays a critical role in immune responses after bacterial infections. Monocytes are an innate immune cell type that have been shown to be involved in bacteremia. However, how the mTOR pathway is involved in the regulation of monocytes is largely unknown. In our study, TSC1 KO mice and WT mice were infected with E. coli. When compared to WT mice, we found higher mortality, greater numbers of bacteria, decreased expression of coactivators in monocytes, increased numbers of Tregs, and decreased numbers of effector T cells in TSC1 KO mice. Monocytes obtained from TSC1 KO mice produced more ROS, IL-6, IL-10, and TGF-ß and less IL-1, IFN-γ, and TNF-α. Taken together, our results suggest that the inhibited immune functioning in TSC1 KO mice is influenced by mTORC1 activation in monocytes. The reduced expression of coactivators resulted in inhibited effector T cell proliferation. mTORC1-activated monocytes are harmful during bacterial infections. Therefore, inhibiting mTORC1 signaling through rapamycin administration could rescue the harmful aspects of an overactive immune response, and this knowledge provides a new direction for clinical therapy.