Gene expression profile reveals deregulation of genes with relevant functions in the different subclasses of acute myeloid leukemia

Small-molecule α-lipoic acid targets ELK1 to balance human neutrophil and erythrocyte differentiation

BACKGROUND

Erythroid and myeloid differentiation disorders are commonly occurred in leukemia. Given that the relationship between erythroid and myeloid lineages is still unclear. To find the co-regulators in erythroid and myeloid differentiation might help to find new target for therapy of myeloid leukemia. In hematopoiesis, ALA (alpha lipoic acid) is reported to inhibit neutrophil lineage determination by targeting transcription factor ELK1 in granulocyte-monocyte progenitors via splicing factor SF3B1. However, further exploration is needed to determine whether ELK1 is a common regulatory factor for erythroid and myeloid differentiation.

METHODS

In vitro culture of isolated CD34+, CMPs (common myeloid progenitors) and CD34+ CD371- HSPCs (hematopoietic stem progenitor cells) were performed to assay the differentiation potential of monocytes, neutrophils, and erythrocytes. Overexpression lentivirus of long isoform (L-ELK1) or the short isoform (S-ELK1) of ELK1 transduced CD34+ HSPCs were transplanted into NSG mice to assay the human lymphocyte and myeloid differentiation differences 3 months after transplantation. Knocking down of SRSF11, which was high expressed in CD371+GMPs (granulocyte-monocyte progenitors), upregulated by ALA and binding to ELK1-RNA splicing site, was performed to analyze the function in erythroid differentiation derived from CD34+ CD123mid CD38+ CD371- HPCs (hematopoietic progenitor cells). RNA sequencing of L-ELK1 and S-ELK1 overexpressed CD34+ CD123mid CD38+ CD371- HPCs were performed to assay the signals changed by ELK1.

RESULTS

Here, we presented new evidence that ALA promoted erythroid differentiation by targeting the transcription factor ELK1 in CD34+ CD371- hematopoietic stem progenitor cells (HSPCs). Overexpression of either the long isoform (L-ELK1) or the short isoform (S-ELK1) of ELK1 inhibited erythroid-cell differentiation, but knockdown of ELK1 did not affect erythroid-cell differentiation. RNAseq analysis of CD34+ CD123mid CD38+ CD371- HPCs showed that L-ELK1 upregulated the expression of genes related to neutrophil activity, phosphorylation, and hypoxia signals, while S-ELK1 mainly regulated hypoxia-related signals. However, most of the genes that were upregulated by L-ELK1 were only moderately upregulated by S-ELK1, which might be due to a lack of serum response factor interaction and regulation domains in S-ELK1 compared to L-ELK1. In summary, the differentiation of neutrophils and erythrocytes might need to rely on the dose of L-ELK1 and S-ELK1 to achieve precise regulation via RNA splicing signals at early lineage commitment.

CONCLUSIONS

ALA and ELK1 are found to regulate both human granulopoiesis and erythropoiesis via RNA spliceosome, and ALA-ELK1 signal might be the target of human leukemia therapy.

The ferroptosis landscape in acute myeloid leukemia

Ferroptosis induction through the suppression of glutathione peroxidase 4 (GPX4) and apoptosis-inducing factor mitochondria-associated 2 (AIFM2) has proven to be an effective approach in eliminating chemotherapy-resistant cells of various types. However, a comprehensive understanding of the roles of GPX4 and AIFM2 in acute myeloid leukemia (AML) has not yet been achieved. Using cBioPortal, DepMap, GEPIA, Metascape, and ONCOMINE, we compared the transcriptional expression, survival data, gene mutation, methylation, and network analyses of GPX4- and AIFM2-associated signaling pathways in AML. The results revealed that high expression levels of GPX4 and AIFM2 are associated with an adverse prognosis for AML patients. Overexpression of AIFM2 correlated with elevated mutation frequencies in NPM1 and DNMT3A. GPX4 upregulation modulated the following pathways: GO:0045333, cellular respiration; R-HSA-5389840, mitochondrial translation elongation; GO:0009060, aerobic respiration; R-HSA-9609507, protein localization; and R-HSA-8953854, metabolism of RNA. On the other hand, the overexpression of AIFM2 influenced the following processes: GO:0048704, embryonic skeletal system morphogenesis; GO:0021546, rhombomere development; GO:0009954, proximal/distal pattern formation; and GO:0048732, gland development. This study identifies the high expression of GPX4 and AIFM2 as novel biomarkers predicting a poor prognosis for AML patients. Furthermore, ferroptosis induction may improve the stratified treatment of AML.

Explainable artificial intelligence for precision medicine in acute myeloid leukemia

Artificial intelligence (AI) can unveil novel personalized treatments based on drug screening and whole-exome sequencing experiments (WES). However, the concept of “black box” in AI limits the potential of this approach to be translated into the clinical practice. In contrast, explainable AI (XAI) focuses on making AI results understandable to humans. Here, we present a novel XAI method -called multi-dimensional module optimization (MOM)- that associates drug screening with genetic events, while guaranteeing that predictions are interpretable and robust. We applied MOM to an acute myeloid leukemia (AML) cohort of 319 ex-vivo tumor samples with 122 screened drugs and WES. MOM returned a therapeutic strategy based on the FLT3, CBFβ-MYH11, and NRAS status, which predicted AML patient response to Quizartinib, Trametinib, Selumetinib, and Crizotinib. We successfully validated the results in three different large-scale screening experiments. We believe that XAI will help healthcare providers and drug regulators better understand AI medical decisions.

Increased expression of RUNX3 inhibits normal human myeloid development

RUNX3 is a transcription factor dysregulated in acute myeloid leukemia (AML). However, its role in normal myeloid development and leukemia is poorly understood. Here we investigate RUNX3 expression in both settings and the impact of its dysregulation on myelopoiesis. We found that RUNX3 mRNA expression was stable during hematopoiesis but decreased with granulocytic differentiation. In AML, RUNX3 mRNA was overexpressed in many disease subtypes, but downregulated in AML with core binding factor abnormalities, such as RUNX1::ETO. Overexpression of RUNX3 in human hematopoietic stem and progenitor cells (HSPC) inhibited myeloid differentiation, particularly of the granulocytic lineage. Proliferation and myeloid colony formation were also inhibited. Conversely, RUNX3 knockdown did not impact the myeloid growth and development of human HSPC. Overexpression of RUNX3 in the context of RUNX1::ETO did not rescue the RUNX1::ETO-mediated block in differentiation. RNA-sequencing showed that RUNX3 overexpression downregulates key developmental genes, such as KIT and RUNX1, while upregulating lymphoid genes, such as KLRB1 and TBX21. Overall, these data show that increased RUNX3 expression observed in AML could contribute to the developmental arrest characteristic of this disease, possibly by driving a competing transcriptional program favoring a lymphoid fate.

Studying the potential of upregulated PTGS2 and VEGF-C besides hyper-methylation of PTGS2 promoter as biomarkers of Acute myeloid leukemia

Hereby, we aimed to investigate the expression of prostaglandin-endoperoxide synthase 2 (PTGS2) and Vascular Endothelial Factor-C (VEGF-C) besides the methylation of PTGS2 in AML patients. VEGF-C and PTGS2 expression analysis were evaluated in newly diagnosed AML patients and healthy controls by quantitative Reverse Transcriptase PCR method. Also, PTGS2 methylation status was evaluated by Methylation-Sensitive High-Resolution Melting Curve Analysis (MS-HRM). While 34% of patients were female, the mean age of the patients was 43.41 ± 17.60 years suffering mostly from M4 (48.21%) type of AML. Although methylation level between patients and controls was not significantly different, none of the normal controls showed methylation in the PTGS2 promoter. PTGS2 and VEGF-C levels were elevated in AML cases and correlated with WBC, Platelet, and Hemoglobin levels. The survival of patients with overexpressed VEGF-C and PTGS2 was poorer than others. It can be concluded that PTGS2 and especially VEGF-C expression but not PTGS2 methylation can be considered as diagnostic biomarkers for AML.

Identification of Potential Novel Prognosis-Related Genes Through Transcriptome Sequencing, Bioinformatics Analysis, and Clinical Validation in Acute Myeloid Leukemia

Background: Acute Myeloid Leukemia (AML) is a complex and heterogeneous hematologic malignancy. However, the function of prognosis-related signature genes in AML remains unclear. Methods: In the current study, transcriptome sequencing was performed on 15 clinical samples, differentially expressed RNAs were identified using R software. The potential interactions network was constructed by using the common genes between target genes of differentially expressed miRNAs with transcriptome sequencing results. Functional and pathway enrichment analysis was performed to identify candidate gene-mediated aberrant signaling pathways. Hub genes were identified by the cytohubba plugin in Cytoscape software, which then expanded the potential interactions regulatory module for hub genes. TCGA-LAML clinical data were used for the prognostic analysis of the hub genes in the regulatory network, and GVSA analysis was used to identify the immune signature of prognosis-related hub genes. qRT-PCR was used to verify the expression of hub genes in independent clinical samples. Results: We obtained 1,610 differentially expressed lncRNAs, 233 differentially expressed miRNAs, and 2,217 differentially expressed mRNAs from transcriptome sequencing. The potential interactions network is constructed by 12 lncRNAs, 25 miRNAs, and 692 mRNAs. Subsequently, a sub-network including 15 miRNAs as well as 12 lncRNAs was created based on the expanded regulatory modules of 25 key genes. The prognostic analysis results show that CCL5 and lncRNA UCA1 was a significant impact on the prognosis of AML. Besides, we found three potential interactions networks such as lncRNA UCA1/hsa-miR-16-5p/COL4A5, lncRNA UCA1/hsa-miR-16-5p/SPARC, and lncRNA SNORA27/hsa-miR-17-5p/CCL5 may play an important role in AML. Furthermore, the evaluation of the immune infiltration shows that CCL5 is positively correlated with various immune signatures, and lncRNA UCA1 is negatively correlated with the immune signatures. Finally, the result of qRT-PCR showed that CCL5 is down-regulated and lncRNA UCA1 is up-regulated in AML samples separately. Conclusions: In conclusion, we propose that CCL5 and lncRNA UCA1 could be recognized biomarkers for predicting survival prognosis based on constructing competing endogenous RNAs in AML, which will provide us novel insight into developing novel prognostic, diagnostic, and therapeutic for AML.

Prognostic Prediction of Cytogenetically Normal Acute Myeloid Leukemia Based on a Gene Expression Model

Acute myeloid leukemia (AML) refers to a heterogeneous group of hematopoietic malignancies. The well-known European Leukemia Network (ELN) stratifies AML patients into three risk groups, based primarily on the detection of cytogenetic abnormalities. However, the prognosis of cytogenetically normal AML (CN-AML), which is the largest AML subset, can be hard to define. Moreover, the clinical outcomes associated with this subgroup are diverse. In this study, using transcriptome profiles collected from CN-AML patients in the BeatAML cohort, we constructed a robust prognostic Cox model named NEST (Nine-gEne SignaTure). The validity of NEST was confirmed in four external independent cohorts. Moreover, the risk score predicted by the NEST model remained an independent prognostic factor in multivariate analyses. Further analysis revealed that the NEST model was suitable for bone marrow mononuclear cell (BMMC) samples but not peripheral blood mononuclear cell (PBMC) samples, which indirectly indicated subtle differences between BMMCs and PBMCs. Our data demonstrated the robustness and accuracy of the NEST model and implied the importance of the immune dysfunction in the leukemogenesis that occurs in CN-AML, which shed new light on the further exploration of molecular mechanisms and treatment guidance for CN-AML.

Bimodal expression of RHOH during myelomonocytic differentiation: Implications for the expansion of AML differentiation therapy

RhoH is an unusual member of the Rho family of small GTP‐binding proteins in that it lacks GTPase activity. Since the RhoH protein is constantly bound by GTP, it is constitutively active and controlled predominantly by changes in quantitative expression. Abnormal levels of RHOH gene transcripts have been linked to a range of malignancies including acute myeloid leukemia (AML). One of the hallmarks of AML is a block in the normal program of myeloid differentiation. Here we investigate how myeloid differentiation is controlled by the quantitative expression of RHOH. Our analysis demonstrates that increasingly mature myeloid cells express progressively lower levels of RHOH. However, as monocytic myeloid cells terminally differentiate into macrophages, RHOH expression is up‐regulated. This up‐regulation is not apparent in AML where myeloid differentiation is blocked at stages of low RHOH expression. Nevertheless, when the up‐regulation of RHOH is forced, then terminal macrophage differentiation is induced and the Cdc42 and Wnt intracellular signalling pathways are repressed. These results indicate that RHOH induction is a driver of terminal differentiation and might represent a means of effecting AML differentiation therapy. The potential of this therapeutic strategy is supported by forced up‐regulation of RHOH reducing the ability of AML cells to produce tumours in vivo.

Function of Deptor and its roles in hematological malignancies

Deptor is a protein that interacts with mTOR and that belongs to the mTORC1 and mTORC2 complexes. Deptor is capable of inhibiting the kinase activity of mTOR. It is well known that the mTOR pathway is involved in various signaling pathways that are involved with various biological processes such as cell growth, apoptosis, autophagy, and the ER stress response. Therefore, Deptor, being a natural inhibitor of mTOR, has become very important in its study. Because of this, it is important to research its role regarding the development and progression of human malignancies, especially in hematologic malignancies. Due to its variation in expression in cancer, it has been suggested that Deptor can act as an oncogene or tumor suppressor depending on the cellular or tissue context. This review discusses recent advances in its transcriptional and post-transcriptional regulation of Deptor. As well as the advances regarding the activities of Deptor in hematological malignancies, its possible role as a biomarker, and its possible clinical relevance in these malignancies.

FGF13 promotes metastasis of triple-negative breast cancer

Triple-negative breast cancer (TNBC) represents 10-20% of all human ductal adenocarcinomas and has a poor prognosis relative to other subtypes, due to the high propensity to develop distant metastases. Hence, new molecular targets for therapeutic intervention are needed for TNBC. We recently conducted a rigorous phenotypic and genomic characterization of four isogenic populations of MDA-MB-231 human triple-negative breast cancer cells that possess a range of intrinsic spontaneous metastatic capacities in vivo, ranging from nonmetastatic (MDA-MB-231_ATCC) to highly metastatic to lung, liver, spleen and spine (MDA-MB-231_HM). Gene expression profiling of primary tumours by RNA-Seq identified the fibroblast growth factor homologous factor, FGF13, as highly upregulated in aggressively metastatic MDA-MB-231_HM tumours. Clinically, higher FGF13 mRNA expression was associated with significantly worse relapse free survival in both luminal A and basal-like human breast cancers but was not associated with other clinical variables and was not upregulated in primary tumours relative to normal mammary gland. Stable FGF13 depletion restricted in vitro colony forming ability in MDA-MB-231_HM TNBC cells but not in oestrogen receptor (ER)-positive MCF-7 or MDA-MB-361 cells. However, despite augmenting MDA-MB-231_HM cell migration and invasion in vitro, FGF13 suppression almost completely blocked the spontaneous metastasis of MDA-MB-231_HM orthotopic xenografts to both lung and liver while having negligible impact on primary tumour growth. Together, these data indicate that FGF13 may represent a therapeutic target for blocking metastatic outgrowth of certain TNBCs. Further evaluation of the roles of individual FGF13 protein isoforms in progression of the different subtypes of breast cancer is warranted.

Not Only Mutations Matter: Molecular Picture of Acute Myeloid Leukemia Emerging from Transcriptome Studies

The last two decades of genome-scale research revealed a complex molecular picture of acute myeloid leukemia (AML). On the one hand, a number of mutations were discovered and associated with AML diagnosis and prognosis; some of them were introduced into diagnostic tests. On the other hand, transcriptome studies, which preceded AML exome and genome sequencing, remained poorly translated into clinics. Nevertheless, gene expression studies significantly contributed to the elucidation of AML pathogenesis and indicated potential therapeutic directions. The power of transcriptomic approach lies in its comprehensiveness; we can observe how genome manifests its function in a particular type of cells and follow many genes in one test. Moreover, gene expression measurement can be combined with mutation detection, as high-impact mutations are often present in transcripts. This review sums up 20 years of transcriptome research devoted to AML. Gene expression profiling (GEP) revealed signatures distinctive for selected AML subtypes and uncovered the additional within-subtype heterogeneity. The results were particularly valuable in the case of AML with normal karyotype which concerns up to 50% of AML cases. With the use of GEP, new classes of the disease were identified and prognostic predictors were proposed. A plenty of genes were detected as overexpressed in AML when compared to healthy control, including KIT, BAALC, ERG, MN1, CDX2, WT1, PRAME, and HOX genes. High expression of these genes constitutes usually an unfavorable prognostic factor. Upregulation of FLT3 and NPM1 genes, independent on their mutation status, was also reported in AML and correlated with poor outcome. However, transcriptome is not limited to the protein-coding genes; other types of RNA molecules exist in a cell and regulate genome function. It was shown that microRNA (miRNA) profiles differentiated AML groups and predicted outcome not worse than protein-coding gene profiles. For example, upregulation of miR-10a, miR-10b, and miR-196b and downregulation of miR-192 were found as typical of AML with NPM1 mutation whereas overexpression of miR-155 was associated with FLT3-internal tandem duplication (FLT3-ITD). Development of high-throughput technologies and microarray replacement by next generation sequencing (RNA-seq) enabled uncovering a real variety of leukemic cell transcriptomes, reflected by gene fusions, chimeric RNAs, alternatively spliced transcripts, miRNAs, piRNAs, long noncoding RNAs (lncRNAs), and their special type, circular RNAs. Many of them can be considered as AML biomarkers and potential therapeutic targets. The relations between particular RNA puzzles and other components of leukemic cells and their microenvironment, such as exosomes, are now under investigation. Hopefully, the results of this research will shed the light on these aspects of AML pathogenesis which are still not completely understood.

High EVI1 Expression Predicts Poor Outcomes in Adult Acute Myeloid Leukemia Patients with Intermediate Cytogenetic Risk Receiving Chemotherapy

BACKGROUND Acute myeloid leukemia with intermediate cytogenetic risk (ICR-AML) needs to be stratified. The abnormal gene expression might be prognostic, and its cutoff value for patient grouping is pivotal. MATERIAL AND METHODS Ecotropic viral integration site 1 (EVI1) transcripts were assessed in 191 adult ICR-AML patients at diagnosis who received chemotherapy only. MLL-PTD, WT1 transcript levels, FLT3-ITD, and NPM1 mutations were simultaneously evaluated, and 27 normal bone marrow samples were tested to define normal threshold. RESULTS The normal upper limit of EVI1 transcript levels was 8.0%. Receiver operating characteristic curve analysis showed that 1.0% (a 0.9-log reduction from the normal limit) was the EVI1 optimal cutoff value for significantly differentiating relapse (P=0.049). A total of 23 patients (12%) had EVI1 levels ≥1.0%. EVI1 ≥1.0% had no effect on CR achievement, whereas it was significantly associated with lower 2-year relapse-free survival (RFS), disease-free survival (DFS), and overall survival (OS) rates in the entire cohort (P=0.0003, 0.0017, and 0.0009, respectively), patients with normal karyotypes (P=0.0032, 0.0047, and 0.0007, respectively), and FLT3-ITD (-) patients (all P<0.0001). Multivariate analysis showed that EVI1 ≥1.0% was an independent adverse prognostic factor for RFS, DFS, and OS in the entire cohort. In addition, patients with EVI1 transcript levels between 1.0% and 8.0% had 2-year RFS rates similar to those with EVI1 ≥8.0%, and they both had significantly lower RFS rates than those with EVI1 <1.0% (P=0.0005 and 0.027). CONCLUSIONS High EVI1 expression predicts poor outcome in ICR-AML patients receiving chemotherapy. The optimal cutoff value for patient stratification is different from the normal limit.

Gene expression profiling of acute myeloid leukemia samples from adult patients with AML-M1 and -M2 through boutique microarrays, real-time PCR and droplet digital PCR

Acute myeloid leukemia (AML) is the most common and severe form of acute leukemia diagnosed in adults. Owing to its heterogeneity, AML is divided into classes associated with different treatment outcomes and specific gene expression profiles. Based on previous studies on AML, in this study, we designed and generated an AML-array containing 900 oligonucleotide probes complementary to human genes implicated in hematopoietic cell differentiation and maturation, proliferation, apoptosis and leukemic transformation. The AML-array was used to hybridize 118 samples from 33 patients with AML of the M1 and M2 subtypes of the French-American-British (FAB) classification and 15 healthy volunteers (HV). Rigorous analysis of the microarray data revealed that 83 genes were differentially expressed between the patients with AML and the HV, including genes not yet discussed in the context of AML pathogenesis. The most overexpressed genes in AML were STMN1, KITLG, CDK6, MCM5, KRAS, CEBPA, MYC, ANGPT1, SRGN, RPLP0, ENO1 and SET, whereas the most underexpressed genes were IFITM1, LTB, FCN1, BIRC3, LYZ, ADD3, S100A9, FCER1G, PTRPE, CD74 and TMSB4X. The overexpression of the CPA3 gene was specific for AML with mutated NPM1 and FLT3. Although the microarray-based method was insufficient to differentiate between any other AML subgroups, quantitative PCR approaches enabled us to identify 3 genes (ANXA3, S100A9 and WT1) whose expression can be used to discriminate between the 2 studied AML FAB subtypes. The expression levels of the ANXA3 and S100A9 genes were increased, whereas those of WT1 were decreased in the AML-M2 compared to the AML-M1 group. We also examined the association between the STMN1, CAT and ABL1 genes, and the FLT3 and NPM1 mutation status. FLT3⁺/NPM1⁻ AML was associated with the highest expression of STMN1, and ABL1 was upregulated in FLT3⁺ AML and CAT in FLT3⁻ AML, irrespectively of the NPM1 mutation status. Moreover, our results indicated that CAT and WT1 gene expression levels correlated with the response to therapy. CAT expression was highest in patients who remained longer under complete remission, whereas WT1 expression increased with treatment resistance. On the whole, this study demonstrates that the AML-array can potentially serve as a first-line screening tool, and may be helpful for the diagnosis of AML, whereas the differentiation between AML subgroups can be more successfully performed with PCR-based analysis of a few marker genes.

Lineage-specific STAT5 target gene activation in hematopoietic progenitor cells predicts the FLT3(+)-mediated leukemic phenotype

Mutations that activate FMS-like tyrosine kinase 3 (FLT3) are frequent occurrences in acute myeloid leukemia. Two distinct types of mutations have been described: internal duplication of the juxtamembranous domain (ITD) and point mutations of the tyrosine kinase domain (TKD). Although both mutations lead to constitutive FLT3 signaling, only FLT3-ITD strongly activates signal transducer and activator of transcription 5 (STAT5). In a murine transplantation model, FLT3-ITD induces a myeloproliferative neoplasm, whereas FLT3-TKD leads to a lymphoid malignancy with significantly longer latency. Here we report that the presence of STAT5 is critical for the development of a myeloproliferative disease by FLT3-ITD in mice. Deletion of Stat5 in FLT3-ITD-induced leukemogenesis leads not only to a significantly longer survival (82 vs 27 days) of the diseased mice, but also to an immunophenotype switch with expansion of the lymphoid cell compartment. Interestingly, we were able to show differential STAT5 activation in FLT3-ITD(+) myeloid and lymphoid murine progenitors. STAT5 target genes such as Oncostatin M were highly expressed in FLT3-ITD(+) myeloid but not in FLT3-ITD(+) lymphoid progenitor cells. Strikingly, FLT3-TKD expression in combination with Oncostatin M is sufficient to reverse the phenotype to a myeloproliferative disease in FLT3-TKD mice. Thus, lineage-specific STAT5 activation in hematopoietic progenitor cells predicts the FLT3(+)-mediated leukemic phenotype in mice.

RUNX3 promoter hypermethylation is frequent in leukaemia cell lines and associated with acute myeloid leukaemia inv(16) subtype

Correlative and functional studies support the involvement of the RUNX gene family in haematological malignancies. To elucidate the role of epigenetics in RUNX inactivation, we evaluated promoter DNA methylation of RUNX1, 2, and 3 in 23 leukaemia cell lines and samples from acute myeloid leukaemia (AML), acute lymphocytic leukaemia (ALL) and myelodysplatic syndromes (MDS) patients. RUNX1 and RUNX2 gene promoters were mostly unmethylated in cell lines and clinical samples. Hypermethylation of RUNX3 was frequent among cell lines (74%) and highly variable among patient samples, with clear association to cytogenetic status. High frequency of RUNX3 hypermethylation (85% of the 20 studied cases) was found in AML patients with inv(16)(p13.1q22) compared to other AML subtypes (31% of the other 49 cases). RUNX3 hypermethylation was also frequent in ALL (100% of the six cases) but low in MDS (21%). In support of a functional role, hypermethylation of RUNX3 was correlated with low levels of protein, and treatment of cell lines with the DNA demethylating agent, decitabine, resulted in mRNA re-expression. Furthermore, relapse-free survival of non-inv(16)(p13.1q22) AML patients without RUNX3 methylation was significantly better (P = 0·016) than that of methylated cases. These results suggest that RUNX3 silencing is an important event in inv(16)(p13.1q22) leukaemias.

HIF factors cooperate with PML-RARα to promote acute promyelocytic leukemia progression and relapse

Acute promyelocytic leukemia (APL) is epitomized by the chromosomal translocation t(15;17) and the resulting oncogenic fusion protein PML-RARα. Although acting primarily as a transcriptional repressor, PML-RARα can also exert functions of transcriptional co-activation. Here, we find that PML-RARα stimulates transcription driven by HIF factors, which are critical regulators of adaptive responses to hypoxia and stem cell maintenance. Consistently, HIF-related gene signatures are upregulated in leukemic promyelocytes from APL patients compared to normal promyelocytes. Through in vitro and in vivo studies, we find that PML-RARα exploits a number of HIF-1α-regulated pro-leukemogenic functions that include cell migration, bone marrow (BM) neo-angiogenesis and self-renewal of APL blasts. Furthermore, HIF-1α levels increase upon treatment of APL cells with all-trans retinoic acid (ATRA). As a consequence, inhibiting HIF-1α in APL mouse models delays leukemia progression and exquisitely synergizes with ATRA to eliminate leukemia-initiating cells (LICs).

Regulation of CDX4 gene transcription by HoxA9, HoxA10, the Mll-Ell oncogene and Shp2 during leukemogenesis

Cdx and Hox proteins are homeodomain transcription factors that regulate hematopoiesis. Transcription of the HOX and CDX genes decreases during normal myelopoiesis, but is aberrantly sustained in leukemias with translocation or partial tandem duplication of the MLL1 gene. Cdx4 activates transcription of the HOXA9 and HOXA10 genes, and HoxA10 activates CDX4 transcription. The events that break this feedback loop, permitting a decreased Cdx4 expression during normal myelopoiesis, were previously undefined. In the current study, we find that HoxA9 represses CDX4 transcription in differentiating myeloid cells, antagonizing activation by HoxA10. We determine that tyrosine phosphorylation of HoxA10 impairs transcriptional activation of CDX4, but tyrosine phosphorylation of HoxA9 facilitates repression of this gene. As HoxA9 and HoxA10 are phosphorylated during myelopoiesis, this provides a mechanism for differentiation stage-specific Cdx4 expression. HoxA9 and HoxA10 are increased in cells expressing Mll-Ell, a leukemia-associated MLL1 fusion protein. We find that Mll-Ell induces a HoxA10-dependent increase in Cdx4 expression in myeloid progenitor cells. However, Cdx4 decreases in a HoxA9-dependent manner on exposure of Mll-Ell-expressing cells to differentiating cytokines. Leukemia-associated, constitutively active mutants of Shp2 block cytokine-induced tyrosine phosphorylation of HoxA9 and HoxA10. In comparison with myeloid progenitor cells that are expressing Mll-Ell alone, we find increased CDX4 transcription and Cdx4 expression in cells co-expressing Mll-Ell plus constitutively active Shp2. Increased Cdx4 expression is sustained on exposure of these cells to differentiating cytokines. Our results identify a mechanism for increased and sustained CDX4 transcription in leukemias co-overexpressing HoxA9 and HoxA10 in combination with constitutive activation of Shp2. This is clinically relevant, because MLL1 translocations and constitutive Shp2 activation co-exist in human myeloid leukemias.

Phenotypic, genomic and functional characterization reveals no differences between CD138++ and CD138low subpopulations in multiple myeloma cell lines

Despite recent advances in the treatment of multiple myeloma (MM), it remains an incurable disease potentially due to the presence of resistant myeloma cancer stem cells (MM-CSC). Although the presence of clonogenic cells in MM was described three decades ago, the phenotype of MM-CSC is still controversial, especially with respect to the expression of syndecan-1 (CD138). Here, we demonstrate the presence of two subpopulations--CD138++ (95-99%) and CD138low (1-5%)--in eight MM cell lines. To find out possible stem-cell-like features, we have phenotypically, genomic and functionally characterized the two subpopulations. Our results show that the minor CD138low subpopulation is morphologically identical to the CD138++ fraction and does not represent a more immature B-cell compartment (with lack of CD19, CD20 and CD27 expression). Moreover, both subpopulations have similar gene expression and genomic profiles. Importantly, both CD138++ and CD138low subpopulations have similar sensitivity to bortezomib, melphalan and doxorubicin. Finally, serial engraftment in CB17-SCID mice shows that CD138++ as well as CD138low cells have self-renewal potential and they are phenotypically interconvertible. Overall, our results differ from previously published data in MM cell lines which attribute a B-cell phenotype to MM-CSC. Future characterization of clonal plasma cell subpopulations in MM patients' samples will guarantee the discovery of more reliable markers able to discriminate true clonogenic myeloma cells.

geneCommittee: a web-based tool for extensively testing the discriminatory power of biologically relevant gene sets in microarray data classification

BACKGROUND

The diagnosis and prognosis of several diseases can be shortened through the use of different large-scale genome experiments. In this context, microarrays can generate expression data for a huge set of genes. However, to obtain solid statistical evidence from the resulting data, it is necessary to train and to validate many classification techniques in order to find the best discriminative method. This is a time-consuming process that normally depends on intricate statistical tools.

RESULTS

geneCommittee is a web-based interactive tool for routinely evaluating the discriminative classification power of custom hypothesis in the form of biologically relevant gene sets. While the user can work with different gene set collections and several microarray data files to configure specific classification experiments, the tool is able to run several tests in parallel. Provided with a straightforward and intuitive interface, geneCommittee is able to render valuable information for diagnostic analyses and clinical management decisions based on systematically evaluating custom hypothesis over different data sets using complementary classifiers, a key aspect in clinical research.

CONCLUSIONS

geneCommittee allows the enrichment of microarrays raw data with gene functional annotations, producing integrated datasets that simplify the construction of better discriminative hypothesis, and allows the creation of a set of complementary classifiers. The trained committees can then be used for clinical research and diagnosis. Full documentation including common use cases and guided analysis workflows is freely available at http://sing.ei.uvigo.es/GC/.

Runx3 deficiency results in myeloproliferative disorder in aged mice

Runx3 conditional knockout mice develop a myeloproliferative disorder when aged. Runx3-deficient cells show hypersensitivity to G-CSF.

Comparison of feature selection methods for cross-laboratory microarray analysis

The amount of gene expression data of microarray has grown exponentially. To apply them for extensive studies, integrated analysis of cross-laboratory (cross-lab) data becomes a trend, and thus, choosing an appropriate feature selection method is an essential issue. This paper focuses on feature selection for Affymetrix (Affy) microarray studies across different labs. We investigate four feature selection methods: $(t)$-test, significance analysis of microarrays (SAM), rank products (RP), and random forest (RF). The four methods are applied to acute lymphoblastic leukemia, acute myeloid leukemia, breast cancer, and lung cancer Affy data which consist of three cross-lab data sets each. We utilize a rank-based normalization method to reduce the bias from cross-lab data sets. Training on one data set or two combined data sets to test the remaining data set(s) are both considered. Balanced accuracy is used for prediction evaluation. This study provides comprehensive comparisons of the four feature selection methods in cross-lab microarray analysis. Results show that SAM has the best classification performance. RF also gets high classification accuracy, but it is not as stable as SAM. The most naive method is $(t)$-test, but its performance is the worst among the four methods. In this study, we further discuss the influence from the number of training samples, the number of selected genes, and the issue of unbalanced data sets.

Tribbles in acute leukemia

There is growing research interest in the mammalian Tribbles (Trib) family of serine/threonine pseudokinases and their oncogenic association with acute leukemias. This review is to understand the role of Trib genes in hematopoietic malignancies and their potential as targets for novel therapeutic strategies in acute myeloid leukemia and acute lymphoblastic leukemia. We discuss the role of Tribs as central signaling mediators in different subtypes of acute leukemia and propose that inhibition of dysregulated Trib signaling may be therapeutically beneficial.

Deregulation of protein phosphatase expression in acute myeloid leukemia

Acute myeloid leukemia (AML) is a highly malignant disease of myeloid cell line. AML is the most frequent adult leukemia with inadequate treatment possibility. The protein phosphatases are critical regulators of cell signaling, and deregulation of protein phosphatases always contribute to cell transformation. Although many studies established a relationship between protein phosphatases and leukemia, little is known about the role of this group of proteins in AML. To address this issue, we initially identified the complete catalog of human protein phosphatase genes and used this catalog to study deregulation of protein phosphatases in AML. Using mRNA expression data of AML patients, we show that 11 protein phosphatases are deregulated in AML within 174 protein phosphatases. The GO enrichment study suggests that these genes are involved in multiple biological processes other than protein de-phosphorylation. Expression of DUSP10, PTPRC, and PTPRE was significantly higher than average expression in AML, and a linear combination of DUSP10, MTMR11, PTPN4, and PTPRE expressions provides important information about disease subtypes. Our results provide an overview of protein phosphatase deregulation in AML.

Genome-scale technology driven advances to research into normal and malignant haematopoiesis

Haematopoiesis or blood development has long served as a model system for adult stem cell biology. Moreover, when combined, the various cancers of the blood represent one of the commonest human malignancies. Large numbers of researchers have therefore dedicated their scientific careers to studying haematopoiesis for more than a century. Throughout this period, many new technologies have first been applied towards the study of blood cells, and the research fields of normal and malignant haematopoiesis have also been some of the earliest adopters of genome-scale technologies. This has resulted in significant new insights with implications ranging from basic biological mechanisms to patient diagnosis and prognosis and also produced lessons likely to be relevant for many other areas of biomedical research. This paper discusses the current state of play for a range of genome-scale applications within haemopoiesis research, including gene expression profiling, ChIP-sequencing, genomewide association analysis, and cancer genome sequencing. A concluding outlook section explores likely future areas of progress as well as potential technological and educational bottlenecks.

Imatinib therapy of chronic myeloid leukemia restores the expression levels of key genes for DNA damage and cell-cycle progression

BACKGROUND

Chronic myeloid leukemia (CML) is a malignant clonal disorder of the hematopoietic system caused by the expression of the BCR/ABL fusion oncogene. It is well known that CML cells are genetically unstable. However, the mechanisms by which these cells acquire genetic alterations are poorly understood. Imatinib mesylate is the standard therapy for newly diagnosed CML patients. Imatinib mesylate targets the oncogenic kinase activity of BCR-ABL.

OBJECTIVE

To study the gene expression profile of bone marrow hematopoietic cells in the same patients with CML before and 1 month after imatinib therapy.

METHODS

Samples from patients with CML were analyzed using Affymetrix GeneChip Expression Arrays.

RESULTS

A total of 594 differentially expressed genes, most of which (393 genes) were downregulated, as a result of imatinib therapy were observed.

CONCLUSION

The blockade of oncoprotein Bcr-Abl by imatinib could cause a decrease in the expression of key DNA repair genes and substantially modify the expression profile of the bone marrow cells in the first days of therapy.

Identification of a novel recurrent gain on 20q13 in chronic lymphocytic leukemia by array CGH and gene expression profiling

BACKGROUND

The presence of genetic changes is a hallmark of chronic lymphocytic leukemia (CLL). The most common cytogenetic abnormalities with independent prognostic significance in CLL are 13q14, ATM and TP53 deletions and trisomy 12. However, CLL displays a great genetic and biological heterogeneity. The aim of this study was to analyze the genomic imbalances in CLL cytogenetic subsets from both genomic and gene expression perspectives to identify new recurrent alterations.

PATIENTS AND METHODS

The genomic imbalances and expression levels of 67 patients were analyzed. The novel recurrent abnormalities detected with bacterial artificial chromosome array were confirmed by FISH and oligonucleotide microarrays. In all cases, gene expression profiling was assessed.

RESULTS

Copy number alterations were identified in 75% of cases. Overall, the results confirmed FISH studies for the regions frequently involved in CLL and also defined a new recurrent gain on chromosome 20q13.12, in 19% (13/67) of the CLL patients. Oligonucleotide expression correlated with the regions of loss or gain of genomic material, suggesting that the changes in gene expression are related to alterations in copy number.

CONCLUSION

Our study demonstrates the presence of a recurrent gain in 20q13.12 associated with overexpression of the genes located in this region, in CLL cytogenetic subgroups.

The significance of low PU.1 expression in patients with acute promyelocytic leukemia

BACKGROUND

Although the importance of the hematopoietic transcription factor PU.1 in acute myeloid leukemia (AML) has been demonstrated, the expression of PU.1 in acute promyelocytic leukemia (APL) patient samples awaits further investigation. The current study used APL patient samples to assess the expression pattern of PU.1 in the initiation and progression of APL.

FINDINGS

We used real-time RT-PCR to compare PU.1 expression between de novo APL patient samples and normal blood specimens, and the results indicated that PU.1 expression was significantly lower in newly diagnosed APL patient samples as compared to normal hematopoietic cells. Further evidence showed a significant inverse correlation between the expression level of PML-RARα and that of PU.1. In addition, we analyzed the correlation between PML-RARα and PU.1 expression in a large population of AML patients retrieved from the expression profiles. The results showed that PU.1 expression was lower in patients with APL than other AML subtypes and there was also a trend towards increasing PU.1 expression from AML-M0 to AML-M5, with the exception of AML-M3 (APL). These observations suggested that PU.1 expression was reduced by PML-RARα in APL patients. Furthermore, we measured PU.1 expression in APL-initiating cells isolated from de novo APL patients by side population cell analysis and found that suppression of PU.1 expression occurred concurrently with PML-RARα expression, indicating the pivotal role of PU.1 in APL initiation.

CONCLUSION

Our findings provide evidence that low PU.1 expression in APL patients is required for disease initiation and progression.

Evi1 forms a bridge between the epigenetic machinery and signaling pathways

Recent studies have demonstrated the significance of the leukemia oncogene Evi1 as the regulator of hematopoietic stem cells and marker of poor clinical outcomes in myeloid malignancies. Evi1-mediated leukemogenic activities include a wide array of functions such as the induction of epigenetic modifications, transcriptional control, and regulation of signaling pathways. We have recently succeeded in comprehensively elucidating the oncogenic function of Evi1 in a model of the polycomb-Evi1-PTEN/AKT/mTOR axis. These results may provide us with novel therapeutic approaches to conquer the poor prognosis associated with Evi1-activated leukemia or other solid tumors with high Evi1 expression. Here, we review the current understanding of the role of Evi1 in controlling the development of leukemia and highlight potential modalities for targeting factors involved in Evi1-regulated signaling.

CD20 positive cells are undetectable in the majority of multiple myeloma cell lines and are not associated with a cancer stem cell phenotype

Although new therapies have doubled the survival of multiple myeloma patients, this remains an incurable disease. It has been postulated that the so-called myeloma cancer stem cells would be responsible for tumor initiation and relapse but their unequivocal identification remains unclear. Here, we investigated in a panel of myeloma cell lines the presence of CD20(+) cells harboring a stem-cell phenotype. Thus, only a small population of CD20(dim+) cells (0.3%) in the RPMI-8226 cell line was found. CD20(dim+) RPMI-8226 cells expressed the plasma cell markers CD38 and CD138 and were CD19(-)CD27(-). Additionally, CD20(dim+) RPMI-8226 cells did not exhibit stem-cell markers as shown by gene expression profiling and the aldehyde dehydrogenase assay. Furthermore, we demonstrated that CD20(dim+) RPMI-8226 cells are not essential for CB17-SCID mice engraftment and show lower self-renewal potential than the CD20(-) RPMI-8226 cells. These results do not support CD20 expression for the identification of myeloma cancer stem cells.

Routine use of microarray-based gene expression profiling to identify patients with low cytogenetic risk acute myeloid leukemia: accurate results can be obtained even with suboptimal samples

Background

Gene expression profiling has shown its ability to identify with high accuracy low cytogenetic risk acute myeloid leukemia such as acute promyelocytic leukemia and leukemias with t(8;21) or inv(16). The aim of this gene expression profiling study was to evaluate to what extent suboptimal samples with low leukemic blast load (range, 2-59%) and/or poor quality control criteria could also be correctly identified.

Methods

Specific signatures were first defined so that all 71 acute promyelocytic leukemia, leukemia with t(8;21) or inv(16)-AML as well as cytogenetically normal acute myeloid leukemia samples with at least 60% blasts and good quality control criteria were correctly classified (training set). The classifiers were then evaluated for their ability to assign to the expected class 111 samples considered as suboptimal because of a low leukemic blast load (n = 101) and/or poor quality control criteria (n = 10) (test set).

Results

With 10-marker classifiers, all training set samples as well as 97 of the 101 test samples with a low blast load, and all 10 samples with poor quality control criteria were correctly classified. Regarding test set samples, the overall error rate of the class prediction was below 4 percent, even though the leukemic blast load was as low as 2%. Sensitivity, specificity, negative and positive predictive values of the class assignments ranged from 91% to 100%. Of note, for acute promyelocytic leukemia and leukemias with t(8;21) or inv(16), the confidence level of the class assignment was influenced by the leukemic blast load.

Conclusion

Gene expression profiling and a supervised method requiring 10-marker classifiers enable the identification of favorable cytogenetic risk acute myeloid leukemia even when samples contain low leukemic blast loads or display poor quality control criterion.

Elevated calpain activity in acute myelogenous leukemia correlates with decreased calpastatin expression

Calpains are intracellular cysteine proteases that have crucial roles in many physiological and pathological processes. Elevated calpain activity has been associated with many pathological states. Calpain inhibition can be protective or lethal depending on the context. Previous work has shown that c-myc transformation regulates calpain activity by suppressing calpastatin, the endogenous negative regulator of calpain. Here, we have investigated calpain activity in primary acute myelogenous leukemia (AML) blast cells. Calpain activity was heterogeneous and greatly elevated over a wide range in AML blast cells, with no correlation to FAB classification. Activity was particularly elevated in the CD34+CD38− enriched fraction compared with the CD34+CD38+ fraction. Treatment of the cells with the specific calpain inhibitor, PD150606, induced significant apoptosis in AML blast cells but not in normal equivalent cells. Sensitivity to calpain inhibition correlated with calpain activity and preferentially targeted CD34+CD38− cells. There was no correlation between calpain activity and p-ERK levels, suggesting the ras pathway may not be a major contributor to calpain activity in AML. A significant negative correlation existed between calpain activity and calpastatin, suggesting calpastatin is the major regulator of activity in these cells. Analysis of previously published microarray data from a variety of AML patients demonstrated a significant negative correlation between calpastatin and c-myc expression. Patients who achieved a complete remission had significantly lower calpain activity than those who had no response to treatment. Taken together, these results demonstrate elevated calpain activity in AML, anti-leukemic activity of calpain inhibition and prognostic potential of calpain activity measurement.

Halofuginone has anti-proliferative effects in acute promyelocytic leukemia by modulating the transforming growth factor beta signaling pathway

Promyelocytic leukemia-retinoic acid receptor alpha (PML-RARα) expression in acute promyelocytic leukemia (APL) impairs transforming growth factor beta (TGFβ) signaling, leading to cell growth advantage. Halofuginone (HF), a low-molecular-weight alkaloid that modulates TGFβ signaling, was used to treat APL cell lines and non-obese diabetic/severe combined immunodeficiency (NOD/SCID) mice subjected to transplantation with leukemic cells from human chorionic gonadotrophin-PML-RARα transgenic mice (TG). Cell cycle analysis using incorporated bromodeoxyuridine and 7-amino-actinomycin D showed that, in NB4 and NB4-R2 APL cell lines, HF inhibited cellular proliferation (P<0.001) and induced apoptosis (P = 0.002) after a 24-hour incubation. Addition of TGFβ revealed that NB4 cells were resistant to its growth-suppressive effects and that HF induced these effects in the presence or absence of the cytokine. Cell growth inhibition was associated with up-regulation of TGFβ target genes involved in cell cycle regulation (TGFB, TGFBRI, SMAD3, p15, and p21) and down-regulation of MYC. Additionally, TGFβ protein levels were decreased in leukemic TG animals and HF in vivo could restore TGFβ values to normal. To test the in vivo anti-leukemic activity of HF, we transplanted NOD/SCID mice with TG leukemic cells and treated them with HF for 21 days. HF induced partial hematological remission in the peripheral blood, bone marrow, and spleen. Together, these results suggest that HF has anti-proliferative and anti-leukemic effects by reversing the TGFβ blockade in APL. Since loss of the TGFβ response in leukemic cells may be an important second oncogenic hit, modulation of TGFβ signaling may be of therapeutic interest.

[From conventional cytogenetics to microarrays. Fifty years of Philadelphia chromosome]

In 1960 Ph-chromosome was found associated with the presence of chronic myelogenous leukemia. In these 50 years an increasing number of cytogenetic abnormalities have been found associated with hematological malignancies. The presence of these abnormalities is not only important for the diagnosis of the patient, but it also contributes to the prognosis of patients with leukemia or lymphoma. For this reason the WHO classification of hematological disease has included these studies for the correct characterization of leukemias and lymphomas. In addition, the use of FISH and micromatrix methodologies have refined the genetic lesions present in these malignancies. The cytogenetic changes observed also provide further information in relation to the therapy.

Gene expression profiling in MDS and AML: potential and future avenues

Today, the classification systems for myelodysplastic syndromes (MDS) and acute myeloid leukemia (AML) already incorporate cytogenetic and molecular genetic aberrations in an attempt to better reflect disease biology. However, in many MDS/AML patients no genetic aberrations have been identified yet, and even within some cytogenetically well-defined subclasses there is considerable clinical heterogeneity. Recent advances in genomics technologies such as gene expression profiling (GEP) provide powerful tools to further characterize myeloid malignancies at the molecular level, with the goal to refine the MDS/AML classification system, incorporating as yet unknown molecular genetic and epigenetic pathomechanisms, which are likely reflected by aberrant gene expression patterns. In this study, we provide a comprehensive review on how GEP has contributed to a refined molecular taxonomy of MDS and AML with regard to diagnosis, prediction of clinical outcome, discovery of novel subclasses and identification of novel therapeutic targets and novel drugs. As many challenges remain ahead, we discuss the pitfalls of this technology and its potential including future integrative studies with other genomics technologies, which will continue to improve our understanding of malignant transformation in myeloid malignancies and thereby contribute to individualized risk-adapted treatment strategies for MDS and AML patients.

Evi1 represses PTEN expression and activates PI3K/AKT/mTOR via interactions with polycomb proteins. Blood. 2011;117:3617-3628. [PMID: 21289308 DOI: 10.1182/blood-2009-12-261602]

Evi1 (ecotropic viral integration site 1) is essential for proliferation of hematopoietic stem cells and implicated in the development of myeloid disorders. Particularly, high Evi1 expression defines one of the largest clusters in acute myeloid leukemia and is significantly associated with extremely poor prognosis. However, mechanistic basis of Evi1-mediated leukemogenesis has not been fully elucidated. Here, we show that Evi1 directly represses phosphatase and tensin homologue deleted on chromosome 10 (PTEN) transcription in the murine bone marrow, which leads to activation of AKT/mammalian target of rapamycin (mTOR) signaling. In a murine bone marrow transplantation model, Evi1 leukemia showed modestly increased sensitivity to an mTOR inhibitor rapamycin. Furthermore, we found that Evi1 binds to several polycomb group proteins and recruits polycomb repressive complexes for PTEN down-regulation, which shows a novel epigenetic mechanism of AKT/mTOR activation in leukemia. Expression analyses and ChIPassays with human samples indicate that our findings in mice models are recapitulated in human leukemic cells. Dependence of Evi1-expressing leukemic cells on AKT/mTOR signaling provides the first example of targeted therapeutic modalities that suppress the leukemogenic activity of Evi1. The PTEN/AKT/mTOR signaling pathway and the Evi1-polycomb interaction can be promising therapeutic targets for leukemia with activated Evi1.

Genome-wide analysis of transcriptional reprogramming in mouse models of acute myeloid leukaemia

Acute leukaemias are commonly caused by mutations that corrupt the transcriptional circuitry of haematopoietic stem/progenitor cells. However, the mechanisms underlying large-scale transcriptional reprogramming remain largely unknown. Here we investigated transcriptional reprogramming at genome-scale in mouse retroviral transplant models of acute myeloid leukaemia (AML) using both gene-expression profiling and ChIP-sequencing. We identified several thousand candidate regulatory regions with altered levels of histone acetylation that were characterised by differential distribution of consensus motifs for key haematopoietic transcription factors including Gata2, Gfi1 and Sfpi1/Pu.1. In particular, downregulation of Gata2 expression was mirrored by abundant GATA motifs in regions of reduced histone acetylation suggesting an important role in leukaemogenic transcriptional reprogramming. Forced re-expression of Gata2 was not compatible with sustained growth of leukaemic cells thus suggesting a previously unrecognised role for Gata2 in downregulation during the development of AML. Additionally, large scale human AML datasets revealed significantly higher expression of GATA2 in CD34+ cells from healthy controls compared with AML blast cells. The integrated genome-scale analysis applied in this study represents a valuable and widely applicable approach to study the transcriptional control of both normal and aberrant haematopoiesis and to identify critical factors responsible for transcriptional reprogramming in human cancer.

A conceptual framework for the identification of candidate drugs and drug targets in acute promyelocytic leukemia

Chromosomal translocations of transcription factors generating fusion proteins with aberrant transcriptional activity are common in acute leukemia. In acute promyelocytic leukemia (APL), the promyelocytic leukemia-retinoic-acid receptor alpha (PML-RARA) fusion protein, which emerges as a consequence of the t(15;17) translocation, acts as a transcriptional repressor that blocks neutrophil differentiation at the promyelocyte (PM) stage. In this study, we used publicly available microarray data sets and identified signatures of genes dysregulated in APL by comparison of gene expression profiles of APL cells and normal PMs representing the same stage of differentiation. We next subjected our identified APL signatures of dysregulated genes to a series of computational analyses leading to (i) the finding that APL cells show stem cell properties with respect to gene expression and transcriptional regulation, and (ii) the identification of candidate drugs and drug targets for therapeutic interventions. Significantly, our study provides a conceptual framework that can be applied to any subtype of AML and cancer in general to uncover novel information from published microarray data sets at low cost. In a broader perspective, our study provides strong evidence that genomic strategies might be used in a clinical setting to prospectively identify candidate drugs that subsequently are validated in vitro to define the most effective drug combination for individual cancer patients on a rational basis.

A differentiation-based phylogeny of cancer subtypes

Histopathological classification of human tumors relies in part on the degree of differentiation of the tumor sample. To date, there is no objective systematic method to categorize tumor subtypes by maturation. In this paper, we introduce a novel computational algorithm to rank tumor subtypes according to the dissimilarity of their gene expression from that of stem cells and fully differentiated tissue, and thereby construct a phylogenetic tree of cancer. We validate our methodology with expression data of leukemia, breast cancer and liposarcoma subtypes and then apply it to a broader group of sarcomas. This ranking of tumor subtypes resulting from the application of our methodology allows the identification of genes correlated with differentiation and may help to identify novel therapeutic targets. Our algorithm represents the first phylogeny-based tool to analyze the differentiation status of human tumors.

Gene expression profiling of malignancies is often held to demonstrate genes that are “up-regulated” or “down-regulated”, but the appropriate frame of reference against which observations should be compared has not been determined. Fully differentiated somatic cells arise from stem cells, with changes in gene expression that can be experimentally determined. If cancers arise as the result of an abruption of the differentiation process, then poorly differentiated cancers would have a gene expression more similar to stem cells than to normal differentiated tissue, and well differentiated cancers would have a gene expression more similar to fully differentiated cells than to stem cells. In this paper, we describe a novel computational algorithm that allows orientation of cancer gene expression between the poles of the gene expression of stem cells and of fully differentiated tissue. Our methodology allows the construction of a multi-branched phylogeny of human malignancies and can be used to identify genes related to differentiation as well as novel therapeutic targets.

Two distinct clonal populations in acute promyelocytic leukemia, one involving chromosome 17 and the other involving an isochromosome 17

Acute promyelocytic leukemia (APL) is characterized by a t(15;17)(q22;q21) rearrangement. Additional chromosomal rearrangements have been reported in 25-40% of APL patients. The most common abnormality involving chromosome 17 is ider(17). Here we report the case of a patient with APL with isochromosome 17q combined with ider(17), confirmed by fluorescence in situ hybridization. Cytogenetic data strongly suggest that the involvement of chromosomes 15 and 17 in translocation occurs after formation of the isochromosome 17. The case reported here presents the novel finding of two separate clonal events apparently occurring at the same time in an APL patient.

Integrative meta-analysis of differential gene expression in acute myeloid leukemia

Background

Acute myeloid leukemia (AML) is a heterogeneous disease with an overall poor prognosis. Gene expression profiling studies of patients with AML has provided key insights into disease pathogenesis while exposing potential diagnostic and prognostic markers and therapeutic targets. A systematic comparison of the large body of gene expression profiling studies in AML has the potential to test the extensibility of conclusions based on single studies and provide further insights into AML.

Methodology/Principal Findings

In this study, we systematically compared 25 published reports of gene expression profiling in AML. There were a total of 4,918 reported genes of which one third were reported in more than one study. We found that only a minority of reported prognostically-associated genes (9.6%) were replicated in at least one other study. In a combined analysis, we comprehensively identified both gene sets and functional gene categories and pathways that exhibited significant differential regulation in distinct prognostic categories, including many previously unreported associations.

Conclusions/Significance

We developed a novel approach for granular, cross-study analysis of gene-by-gene data and their relationships with established prognostic features and patient outcome. We identified many robust novel prognostic molecular features in AML that were undetected in prior studies, and which provide insights into AML pathogenesis with potential diagnostic, prognostic, and therapeutic implications. Our database and integrative analysis are available online (http://gat.stamlab.org).

Deregulation of microRNA expression in the different genetic subtypes of multiple myeloma and correlation with gene expression profiling

Specific microRNA (miRNA) signatures have been associated with different cytogenetic subtypes in acute leukemias. This finding prompted us to investigate potential associations between genetic abnormalities in multiple myeloma (MM) and singular miRNA expression profiles. Moreover, global gene expression profiling was also analyzed to find correlated miRNA gene expression and select miRNA target genes that show such correlation. For this purpose, we analyzed the expression level of 365 miRNAs and the gene expression profiling in 60 newly diagnosed MM patients, selected to represent the most relevant recurrent genetic abnormalities. Supervised analysis showed significantly deregulated miRNAs in the different cytogenetic subtypes as compared with normal PC. It is interesting to note that miR-1 and miR-133a clustered on the same chromosomal loci, were specifically overexpressed in the cases with t(14;16). The analysis of the relationship between miRNA expression and their respective target genes showed a conserved inverse correlation between several miRNAs deregulated in MM cells and CCND2 expression level. These results illustrate, for the first time, that miRNA expression pattern in MM is associated with genetic abnormalities, and that the correlation of the expression profile of miRNA and their putative mRNA targets is useful to find statistically significant protein-coding genes in MM pathogenesis associated with changes in specific miRNAs.

Genes predictive of outcome and novel molecular classification schemes in adult acute myeloid leukemia

The pretreatment karyotype of leukemic blasts is currently the key determinant in therapy decision making in acute myeloid leukemia (AML). The World Health Organization (WHO) has recognized this important information by including, besides clinical, cytological, cytochemical, and immunophenotypical features, recurrent cytogenetic abnormalities in its classification (Table 1). However, although the WHO defines important biologically and clinically relevant entities, the prognostic value of some of the well-defined cytogenetic subgroups is partially masked in the WHO classification. Moreover, in the recent past a number of novel molecular aberrations with marked prognostic value, which are not yet incorporated in the WHO classifications have been identified. These molecular abnormalities include mutations (e.g., in FLT3, c-KIT, and NPM1), partial duplications (e.g., of MLL and FLT3), and abnormal expression of pathogenetic genes (e.g., EVI1, WT1, BCL2, MDR1, BAALC, and ERG). In addition, novel molecular approaches in genomics, like monitoring the expression levels of thousands of genes in parallel using DNA microarray technology, open possibilities for further refinement of prognostication of AML. Gene expression profiling in AML is already well established and has proven to be valuable to recognize various cytogenetic subtypes, discover novel AML subclasses, and predict clinical outcome. The current advances made in molecular understanding of AML will ultimately lead to a further refinement of prognostics of AML.

Cbfb/Runx1 repression-independent blockage of differentiation and accumulation of Csf2rb-expressing cells by Cbfb-MYH11

It is known that CBFB-MYH11, the fusion gene generated by inversion of chromosome 16 in human acute myeloid leukemia, is causative for oncogenic transformation. However, the mechanism by which CBFB-MYH11 initiates leukemogenesis is not clear. Previously published reports showed that CBFB-MYH11 dominantly inhibits RUNX1 and CBFB, and such inhibition has been suggested as the mechanism for leukemogenesis. Here we show that Cbfb-MYH11 caused Cbfb/Runx1 repression-independent defects in both primitive and definitive hematopoiesis. During primitive hematopoiesis, Cbfb-MYH11 delayed differentiation characterized by sustained expression of Gata2, Il1rl1, and Csf2rb, a phenotype not found in Cbfb and Runx1 knockout mice. Expression of Cbfb-MYH11 in the bone marrow induced the accumulation of abnormal progenitor-like cells expressing Csf2rb in preleukemic mice. The expression of all 3 genes was detected in most human and murine CBFB-MYH11(+) leukemia samples. Interestingly, Cbfb-MYH11(+) preleukemic progenitors and leukemia-initiating cells did not express Csf2rb, although the majority of leukemia cells in our Cbfb-MYH11 knockin mice were Csf2rb(+). Therefore Csf2rb can be used as a negative selection marker to enrich preleukemic progenitor cells and leukemia-initiating cells from Cbfb-MYH11 mice. These results suggest that Cbfb/Runx1 repression-independent activities contribute to leukemogenesis by Cbfb-MYH11.

A high number of losses in 13q14 chromosome band is associated with a worse outcome and biological differences in patients with B-cell chronic lymphoid leukemia

BACKGROUND

Among patients with B-cell chronic lymphoid leukemia, those with 13q14 deletion have a favorable outcome. However, whether the percentage of cells with 13q- influences the prognosis or the biological characteristics of this disease is unknown. We analyzed the clinico-biological characteristics and outcome of patients with B-cell chronic lymphoid leukemia with loss of 13q as the sole cytogenetic aberration.

DESIGN AND METHODS

Three hundred and fifty patients with B-cell chronic lymphoid leukemia were studied. Clinical data were collected and fluorescence in situ hybridization and molecular studies were carried out. In addition, a gene expression profile was obtained by microarray-based analysis.

RESULTS

In 109 out of the 350 cases (31.1%) loss of 13q was the sole cytogenetic aberration at diagnosis. In the subgroup of patients with 80% or more of cells with loss of 13q (18 cases), the overall survival was 56 months compared with not reached in the 91 cases in whom less than 80% of cells had loss of 13q (p< 0.0001). The variables included in the multivariate analysis for overall survival were the percentage of losses of 13q14 (p=0.001) and B symptoms (p=0.007). The time to first therapy in the group with 80% or more vs. less than 80% of losses was 38 months vs. 87 months, respectively (p=0.05). In the multivariate analysis the variables selected were unmutated status of IgV(H) (p=0.001) and a high level of beta(2)microglobulin (p=0.003). Interestingly, these differences regarding overall survival and time to first therapy were also present when other cut-offs were considered. The gene expression profile of patients with a high number of losses in 13q14 showed a high proliferation rate, downregulation of apoptosis-related genes, and dysregulation of genes related to mitochondrial functions.

CONCLUSIONS

Patients with B-cell chronic lymphoid leukemia with a high number of losses in 13q14 as the sole cytogenetic aberration at diagnosis display different clinical and biological features: short overall survival and time to first therapy as well as more proliferation and less apoptosis. A quantification of the number of cells showing a genetic abnormality should, therefore, be included in the study of the prognostic factors of B-cell chronic lymphoid leukemia.

High throughput digital quantification of mRNA abundance in primary human acute myeloid leukemia samples

Acute promyelocytic leukemia (APL) is characterized by the t(15;17) chromosomal translocation, which results in fusion of the retinoic acid receptor alpha (RARA) gene to another gene, most commonly promyelocytic leukemia (PML). The resulting fusion protein, PML-RARA, initiates APL, which is a subtype (M3) of acute myeloid leukemia (AML). In this report, we identify a gene expression signature that is specific to M3 samples; it was not found in other AML subtypes and did not simply represent the normal gene expression pattern of primary promyelocytes. To validate this signature for a large number of genes, we tested a recently developed high throughput digital technology (NanoString nCounter). Nearly all of the genes tested demonstrated highly significant concordance with our microarray data (P < 0.05). The validated gene signature reliably identified M3 samples in 2 other AML datasets, and the validated genes were substantially enriched in our mouse model of APL, but not in a cell line that inducibly expressed PML-RARA. These results demonstrate that nCounter is a highly reproducible, customizable system for mRNA quantification using limited amounts of clinical material, which provides a valuable tool for biomarker measurement in low-abundance patient samples.

Identification of protein-coding and intronic noncoding RNAs down-regulated in clear cell renal carcinoma

The clear cell subtype of renal cell carcinoma (RCC) is the most lethal and prevalent cancer of the urinary system. To investigate the molecular changes associated with malignant transformation in clear cell RCC, the gene expression profiles of matched samples of tumor and adjacent non-neoplastic tissue were obtained from six patients. A custom-built cDNA microarray platform was used, comprising 2292 probes that map to exons of genes and 822 probes for noncoding RNAs mapping to intronic regions. Intronic transcription was detected in all normal and neoplastic renal tissues. A subset of 55 transcripts was significantly down-regulated in clear cell RCC relative to the matched nontumor tissue as determined by a combination of two statistical tests and leave-one-out patient cross-validation. Among the down-regulated transcripts, 49 mapped to untranslated or coding exons and 6 were intronic relative to known exons of protein-coding genes. Lower levels of expression of SIN3B, TRIP3, SYNJ2BP and NDE1 (P < 0.02), and of intronic transcripts derived from SND1 and ACTN4 loci (P < 0.05), were confirmed in clear cell RCC by Real-time RT-PCR. A subset of 25 transcripts was deregulated in additional six nonclear cell RCC samples, pointing to common transcriptional alterations in RCC irrespective of the histological subtype or differentiation state of the tumor. Our results indicate a novel set of tumor suppressor gene candidates, including noncoding intronic RNAs, which may play a significant role in malignant transformations of normal renal cells.

Effects of imatinib mesylate on normal bone marrow cells from chronic myeloid leukemia patients in complete cytogenetic response

Information on the effects of imatinib mesylate (IM) on the non-clonal bone marrow (BM) cell compartment is scanty. We have analyzed the gene expression profile of BM hematopoietic cells after IM therapy in 20 patients with chronic myeloid leukaemia (CML) in complete cytogenetic response (CCyR) and compared it with that of normal volunteer donors by oligonucleotide microarrays. In CCyR CML samples, IM induces a decrease in proliferation as well as increase in apoptosis and ubiquitination in residual non-clonal BM cells. In addition, IM diminishes cell-to-cell adhesion and downregulates the expression of the erythropoietin (EPO) receptor gene. The latter was confirmed by RT-PCR.