In AML, the association of microRNA expression profile and AML genetic categories has been investigated in recent studies [5-11], indeed AML genetic heterogeneity can also be resolved according to their variable miRNA expression signatures. We used quantitative real-time RT-PCR to study the expression of 365 known human miRNAs in a cohort of 9 primary selected AML DNMT3A mutated cases and we compare their miRNA profile to those obtained, in our previous data [10], from AMLs characterized by common cytogenetic and molecular alterations and from three controls of purified CD34+ hematopoietic progenitor cells from healthy donors.

Not surprisingly, AML DNMT3A mutated cases could be distinguished, based on profile of expressed miRNA, from the other genetic groups by an unsupervised clustering method. The miRNA profile in DNMT3A mutated AML was investigated by Marcucci et al. [4], by microarray experiments but the authors did not found any signature associated with DNMT3A mutations and they suggest that these features could be confuse due to the positive association of DNMT3A mutations with NPM1 mutations and the negative association with CEBPA mutations. For the first time, we describe miRNA expression profile derived from a quantitative technology associated with high-risk AML carrying DNMT3A mutations and we found a group of 83 deregulated miRNA and among them two members of the miR-29 family, miR 29a and miR-29c.

The human miR-29 family consists of 3 members: miR-29a, miR-29b and miR-29c; despite the identical seed sequence, recent studies suggest a context-dependent pattern for miR-29s expression and regulation. Consistently, a variety of physiological and disease conditions have been correlated to a differential expression and subcellular organization of miR-29 family members [23].

In this report, to clarify the role of miR-29s deregulation in DNMT3A mutated AML, we consolidate the data that miR-29a and miR-29c, but not miR-29b, are up-regulated in DNMT3A mutated AML. We used quantitative real-time RT-PCR to study miR-29s expression in a large cohort of 100 primary AML characterized by common cytogenetic and molecular alterations, including 50 DNMT3A-mutated AML. According to miRNA profile data, we found miR-29a and miR-29c to be specifically up-regulated in this category. The way leading to aberrant expression of miRNAs is not yet completely understood. The mechanisms through which DNMT3A mutation associates with increased miR-29s expression are almost undefined, although exists indirect evidence for epigenetic regulation of miRNAs from DNA methyltransferase (DNMTs) [24,25]. Moreover miR29a/29b-1 cluster is a candidate for DNA methylation dependent regulation and is epigenetically activated in CLL [26].Growing evidence supports a role for microRNAs (miRNAs) as both targets and effectors in aberrant mechanisms of DNA methylation [27]. Studies in AMLs have provided a wide scenario supporting the role of miR-29s as epi-miRNAs. Disturbances in expression levels of these epi-miRNAs have been linked to tumorigenesis and tumor aggressiveness. Members of the miR 29 family directly target the de novo DNA methyltransferases, DNMT3A and DNMT3B and then miR-29s can repress DNA methylation [28] but they are also able to suppress tumorigenesis by preserving the DNA methylation status. Indeed it was demonstrated that miR-29s repress two important components of the DNA demethylation pathway, namely TET1 and TDG [29]. Therefore, miR-29s repress the activities of both DNA methyltransferases and DNA demethylases, which have opposing functions in the control of DNA methylation.

We, further, corroborated miR-29s findings in our cohort of 100 AML samples demonstrating an inverse behavior between their up-regulation and the down-regulation of the mRNA level of their targets TET1 and TET2.

These data suggest that miR29s act as a crucial regulator of DNA methylation and probably in presence of DNMT3A mutations and TET1 down regulation may cause a perturbation of methylation pattern.

Increasing knowledge in cancer epigenetics, as well as the availability of experimental tools to study epigenetic modifications, suggests that epigenetic alterations are also involved in the development of human cancer [30]. In general, human tumors are globally hypomethylated and this contributes to genomic instability, transposon activation and accumulation of mutations, however, locus-specific hypermethylation occurring at CpG islands located in the promoter region of TF genes, have been detected in many hematologic neoplasias like AML [31]. DNMT3A constitutes one of the enzymes mainly responsible for de novo methylation. DNMT3A is essential for normal hematopoietic cell differentiation and defects in DNMT3A and DNMT3B lead to impaired self-renewal capacity in mouse hematopoietic stem cell (HSC) [32]. DNMT3A mutations are among the most common mutations in AML, most of the DNMT3A mutations are in heterozigous and located in the catalytic domain at the R882 aminoacidic residue. R882H mutation has recently been shown to act as a dominant negative that inhibits wild-type DNMT3A [33] and its presence have been linked to decrease methylation levels of some genes such as two homeobox genes HOXA11 and HOXB2 [34,35] and the HOX cofactor MEIS1 [36]. However, no clear epigenetic signature has so far emerged that could explain its role in the pathogenesis of AML DNMT3A mutated. We hypotize that the ipomethylation associated to DNMT3A R882H mutation may cause the epigenetic reactivation of miR-29s; overexpression of miR-29s perturbs the global methylation pattern downregulating their target TET1 and TET2.

TETs, methylcytosine dioxygenase protein, are enzymes that convert 5methylCytosina (5mC) to 5-hydroxymethylcytosine (5hmC). 5hmC is not mantained by DNMT1 causing passive demethylation during cell division [37-39]. In myeloid blast cells carrying DNMT3A mutation may be present, despite the ipomethylated genome status, some specific hypermethylation locus located in the promoter region of transcription factor (TF) genes involved in the myeloid differentiation that physiologically will be reactivated during development and cell division, the deregulation of the TETs activity, in this setting, may impair this mechanism. Because DNMT3A mutations seems to cause a reduction of DNA methyltransferase activity, the results of therapeutic benefit provided by demethylating agents [40] is puzzling and may be explain by the putative reactivation of TF that play a key role on myeloid differentiation.

The performed study suggests that the DNMT3A R882H loss of function mutation may contribute to the significant upregulation of miR29s and consequent downregulation of demethylase TETs. This complex deregulation would contribute to epigenetic perturbations of the genome in patients with DNMT3A mutation. This issue may have important implications for treatment and response to hypomethylating drugs in patients affected by alterations in DNMT3A. In this scenario, epi-miRNAs are emerging as excellent candidate epi-therapeutics for their prominent role in the regulation of the epigenetic machinery.

Table 2: miRNA differentially expressed between AML DNMT3A mutated and CD34 healthy donors.

Table 3: The 20 differentially expressed miRNA between AML DNMT3A mutated and DNMT3A unmutated showing the lowest p value.

In conclusion our study provides data on miR-29 family members overexpression associated with DNMT3A mutated AML and describes an opposite trend between these miRNAs and their targets TET1 and TET2 and in particular show an inverse correlation between miR-29a and TET2. These mechanisms may be involved in the molecular pathogenesis of DNMT3A mutated AML and suggest a molecular pathway that, starting from DNMT3A mutation, through miR-29 family members, alter methylation pattern in this disease, providing a rationale for treatment with hypomethylating drugs.

We thank AML patients for their valuable involvement with this study. This work was supported by a grant of Associazione Italiana contro le leucemie, linfomi e mielomi - Palermo (AIL).

1. Haferlach T, Bacher U, Haferlach C, Kern W, Schnittger S. Insight into the molecular pathogenesis of myeloid malignancies. Curr Opin Hematol. 2007; 14: 90-97.

2. Walzer A, Marcucci G. Molecular prognostic factors in cytogenetically normal acute myeloid leucemia. ExpertRev Hematol. 2012; 5: 547– 558.

3. Ley TG, Ding L, Walter MJ, McLellan MD, Lamprecht T, Larson DE, et al. DNMT3A Mutations in Acute Myeloid Leukemia. N Engl J Med 2010; 363: 2424-2433.

4. Marcucci G, Metzeler KH, Schwind S, Becker H, Maharry K, Mròzek K et al. Age-Related Prognostic Impact of Different Types of DNMT3A Mutations in Adults With Primary Cytogenetically Normal Acute Myeloid Leukemia. J Clin Oncol. 2012; 30: 742-750.

5. Isken F, Steffen B, Merk S, Dugas M, Markus B, Tidow N, et al. Identification of acute myeloid leukaemia associated microRNA expression patterns. Br J Haematol. 2008; 140: 153-161.

6. Garzon R, Volinia S, Liu CG, Fernandez-Cymering C, Palumbo T, Picchiorri F, et al. MicroRNA signatures associated with cytogenetics and prognosis in acute myeloid leukemia. Blood. 2008; 111: 3183- 3189.

7. Marcucci G, Radmacher MD, Maharry K, Mrózek K, Ruppert AS, Paschka P, et al. MicroRNA expression in cytogenetically normal acute myeloid leukemia. N Engl J Med 2008; 358:1919-1928.

8. Jongen-Lavrencic M, Sun SM, DijkstraMK ,Valk PJM, Lowemberg B. MicroRNA expression profiling in relation to the genetic heterogeneity of acute myeloid leukaemia. Blood. 2008; 111: 5078-5085.

9. Li Z, Lu J, Sun M, Mi S, Zhang H, Luo RT, et al. Distinct microRNA expression profiles in acute myeloid leukemia with common translocations. Proc Natl Acad Sci USA. 2008; 105: 15535-15540.

10. Cammarata G, Augugliaro L, Salemi D, Agueli C, La Rosa M, Dagnino L, et al. Differential expression of specific microRNA and their targets in acute myeloid leukemia. Am J Hematol. 2010; 85: 331-339.

11. Bartel D. MicroRNAs: genomics, biogenesis, mechanism, and function. Cell. 2004; 116: 281-297.

12. Lu J, Getz G, Miska EA, Alvarez-Saavedra E, Lamb J, Peck D, et al. MicroRNA expression profiles classify human cancers. Nature. 2005; 435: 834-838.

13. Garzon R, Croce CM. MicroRNAs in normal and malignant hematopoiesis. Curr Opin Hematol. 2008; 15: 352-358.

14. Fazi F, Rosa A, Fatica A, Gelmetti V, De Marchis ML, Nervi C, et al. A minicircuitry comprised of microRNA-223 and transcription factors NFI-A and C/EBPalpha regulates human granulopoiesis. Cell. 2005; 123: 819-831.

15. Debernardi S, Skoulasis S, Molloy G, Chaplin T, Dixon-Mclver A, Young BD. MicroRNA miR-181a correlates with morphological sub-class of acute myeloid leukaemia and the expression of its target genes in global genome-wide analysis. Leukemia. 2007; 21: 912-916.

16. Felli N, Fontana L, Pelosi E, Botta R, Bonci D, Facchiano F, et al. MicroRNAs 221 and 222 inhibit normal erythropoiesis and erythroleukemic cell growth via kit receptor down-modulation. Proc Natl Acad Sci USA. 2005; 102: 18081-18086.

17. Vardiman JW, Thiele J, Arber DA, Brunning RD, Borowitz MJ, Porwit A, et al. The 2008 revision of the World Health Organization (WHO) classification of myeloid neoplasms and acute leukemia: rationale and important changes. Blood. 2009; 114: 937-951.

18. Mitelman F. An international system for human cytogenetic nomenclature. 2nd ed. ISCN Basel: S. Karger 1995.

19. Gabert J, Beillard E, van der Velden VHJ, Bi W, Grimwade D, Pallisgaard N, et al. Standardization and quality control studies of ‘real-time’ quantitative reverse transcriptase polymerase chain reaction of fusion gene transcripts for residual disease detection in leukemia - a Europe Against Cancer program. Leukemia. 2003; 17: 2318–2357.

20. Meshinchi S, Woods WG, Stirewalt DL, Sweetser DA, Buckley JD, Tjoa TK, et al. Prevalence and prognostic significance of Flt3 internal tandem duplication in pediatric acute myeloid leukemia. Blood. 2001; 97: 89-94.

21. Falini B, Mecucci C, Tiacci E, Alcalay M, Rosati R, Pasqualucci L, et al. Cytoplasmicnucleophosmin in acute myelogenousleukemia with a normal karyotype. N Engl J Med. 2005; 352: 254-266.

22. Pfaffl MW, Horgan GW, Dempfle L. Relative expression software tool (REST) for group-wise comparison and statistical analysis of relative expression results in real-time PCR. Nucleic Acids Res. 2002; 30: e36.

23. Amodio N, Rossi M, Raimondi L, Pitari MR, Botta C, Tagliaferri P, et al. miR-29s: a family of epi-miRNAs with therapeutic implications in hematologic malignancies. Oncotarget. 2015; 6: 12837-12861.

24. Lujambo A, Ropero S, Ballestar E, Fraga MF, Cerrato C, Setien F, et al. Genetic unmaskin o fan epigenetically silenced microrNA in human cancer cell. Cancer Res. 2007; 67: 1424-1429.

25. Suzuki H, Takatsuka S, Akashi H, Yamamoto E, Nojima M, Maruyama R, et al. Genome-wide profiling of chromatin signature reveals epigenetic regulation of microRNA genes in colorectal cancer. Cancer Res. 2011; 71: 5646-5658.

26. Baer C, Claus R, Frenzel LP, Zucknick M, Park YJ, Gu L, et al. Extensive Promoter DNA Hypermethylation and Hypiomethylation is Associated with Aberrant MicroRNA Expression in Chronic Lymphocytic Leukemia. Cancer Res. 2012; 72: 3775-3785.

27. Fabbri M, Garzon R, Cimmino A, Liu Z, Zanesi N, Callegari E, et al. MicroRNA-29 family reverts aberrant methylation in lung cancer by targeting DNA methyltransferases 3A and 3B. Proc Natl Acad Sci USA. 2007; 104: 15805-15810.

28. Garzon R, Liu S, Fabbri M, Liu Z, Heaphy CE, Callegari E, et al. MicroRNA 29b induces global DNA hypomethylation and tumor suppressor gene reexpression in acute myeloid leukemia by targeting directly DNMT3A and 3B and indirectly DNMT1. Blood. 2009; 113: 6411-6418.

29. Sumiyo M, Takuro H, Mika K, Takahiro O, Shoji T, Izuho H. miR 29 Represses the Activities of DNA Methyltransferases and DNA Demethylases 1. Int J Mol Sci . 2013; 14: 14647-14658.

30. Virani S, Colacino JA, Kim JH, Rozek LS. Cancer epigenetics: a brief review. ILAR J. 2012; 53: 359-369.

31. Toyota M, Kopecky KJ, Toyota MO, Jair KW, Willman CL, Issa JP. Methylation profiling in acute myeloid leukemia. Blood. 2001. 97: 2823-2839.

32. Yang L, Rau R, Goodell MA. DNMT3A in haematological malignancies. Nat Rev Cancer. 2015; 15: 152-165.

33. Russler-Germain DA, Spencer DH, Young MA, Lamprecht TL, Miller CA, Fulton R, et al. The R882H DNMT3A mutation associated with AML dominantly inhibits wild-type DNMT3A by blocking its ability to form active tetramers. Cancer Cell. 2014; 25: 442-454.

34. Hájková H, Marková J, Haškovec C, Sárová I, Fuchs O, Koste?ka A, et al. Decreased DNA methylation in acute myeloid leukemia patients with DNMT3A mutations and prognostic implications of DNA methylation. Leuk Res. 2012; 36: 1128-1133.

35. Qu Y, Lennartsson A, Gaidzik VI, Deneberg S, Karimi M, Bengtzén S, et al. Differential methylation in CN-AML preferentially target non-cgiregions and is dictated by DNMT3A mutational status and associated with predominant hypometylation of HOX genes. Epigenetics. 2014; 9: 1108-1119.

36. Ferreira HJ, Heyn H, Vizoso M, Moutinho C, Vidal E, Gomez A, et al. DNMT3A mutations mediate the epigenetic reactivation of the leukemogenic factor MEISI in acute myeloid leukemia. Oncogene. 2016; 35: 3079-3082.

37. Tahiliani M, Koh KP, Shen Y, Pastor WA, Bandukwala H, Brudno Y, et al. Conversion of 5-methylcytosine to 5-hydroxymethylcitosine in mammalian DNA by MLL partner TET1. Science. 2009; 324: 930-935.

38. William A. Pastor WA, Pape UJ, Huang Y, Henderson HR, Lister R, Ko M, et al. Genome-wide mapping of 5-hydroxymethylcytosine in embryonic stem cells. Nature. 2011; 473: 394-397.

39. Inoue A, Zhang Y. Replication dependent loss of 5-hydroxymethylcytosine in mouse preimplantation embryos. Science. 2011; 334:194.

40. LaRochelle O, Bertoli S, Vergez F, Sarry JE, Mansat-De Mas V, Dobbelstein S, et al. Do AML patients with DNMT3A exon 23 mutation benefit from idarubicin as compared to daunorubicin ? A single center experience. Oncotarget. 2011; 2: 850-861.

Background: Acute myeloid leukemia (AML) represents a heterogeneous disorder with recurrent chromosomal alterations and molecular abnormalities. A number of molecular abnormalities have been described involving several genes such as FLT3, NPM1, CEBPA, IDH1, IDH2 and DNMT3A among AML with normal kariotype (NK-AML). DNMT3A, a member of DNA methyltransferases, is mutated in approximately 22% of de novo NK-AML; the mutation at codon R882 (R882-DNMT3A mutations) confers unfavorable outcome in this setting. Our previous data demonstrated distinctive miRNA expression patterns in some genetic groups, the aim of the study is to investigate about the miRNA profile and its role in the pathogenesis of DNMT3A mutated AML.

Methods and Results: To indagate about miRNA signature in NK-AML R882-DNMT3A mutated we performed quantitative real-time PCR (TaqMan Human MicroRNA Array, A); we studied the expression of 365 known human miRNA in 9 selected de-novo AML cases showing DNMT3A mutations. We compare miRNA expression data with our previous results obtained in 31 AML DNMT3A unmutated and we focused on an up-regulation of mir-29 family: miR-29a, miR-29b and miR-29c. So we decide to investigate expression levels of these miRNAs in additional 50 new DNMT3A mutated AML patients and we confirm the up-regulation of 2 of them: miR-29a and miR-29c. The most interesting issue is that miR-29s have been demonstrated to directly target 3’-UTR of DNMT3A and DNMT3B resulting in a global hypomethylation but also miR-29s is able to directly suppress the major DNA demethylases, TET1 and TDG. In order to better understand the pathogenesis of the subgroup of AML DNMT3A mutated and the existing correlation between miR-29s and its targets, we evaluated the expression levels of miR-29s targets DNMT3A, DNMT3B, TET1, TET2 and TDG in 50 AML DNMT3A mutated patients and in 50 patients AML DNMT3A wild type (control group) using qRT-PCR with specific TaqMan assay (Applied Biosystems). Results obtained revealed a no significant differences in expression of DNMT3A, DNMT3B and TDG; however we found significant down-regulation of the demethylases TET1 and TET2.

Conclusion: These data suggest that miR-29s act as crucial regulators of DNA methylation and probably in presence of DNMT3A mutations and TET1 down-regulation may cause a perturbation of methylation pattern. This issue may have important implications for treatment and response to hypomethylating drugs in patients affected by alterations in DNMT3A.

Randazzo V, Salemi D, Agueli CCannella S, Marfia A, et al. (2016) Upregulation of Mir-29 in Normal Karyotype Aml Showing Dnmt3a Mutation. J Hematol Transfus 4(2): 1048.

• Acute myeloid leukemia; DNMT3A; miRNA; Gene 
expression

AML: Acute Myeloid Leukemia; NK-AML: Normal Karyotype

Acute myeloid leukaemia is a cytogenetically heterogeneous disorder with acquired recurrent chromosomal alterations detected in about 55% of adult patients. In the remaining cases showing normal karyotype (NK-AML) a number of molecular abnormalities with a prognostic value have been described such as mutations in genes FLT3, MLL, CEBPA, NPM1, WT1, IDH1, IDH2 and DNMT3A [1,2]. DNMT3A, a member of DNA methyltransferases, is mutated in approximately 22% of de novo AML patients with normal karyotype leading to adverse overall survival (OS), independently from age and presence of FLT3 mutations. There are two major types of DNMT3A mutations: the first type is the highly recurrent set of mutations at codon R882 (R882-DNMT3A mutations); the second type is represented by all the other mutations in this gene (no-R882- DNMT3A mutations) [3,4]. The heterogeneity of AML can also be resolved according to their microRNA signatures [5-10]. MicroRNA (miRNA) are 19–24 nucleotides noncoding RNA which regulate the expression of target mRNAs at transcriptional and translational level; miRNA deregulation has been found in different human diseases, including cancer, leukaemia, diabetes, immuno- or neurodegenerative disorders [11-13]. Their role in normal hematopoiesis elucidated by recent studies, reveal specific variations of the miRNome during the commitment and development of the hematological stem cells in the different lineage [14-16]. Our previous data demonstrated distinctive miRNA expression patterns in some genetic groups [10]. The aim of this study is to investigate about the miRNA profile and its role in the pathogenesis of DNMT3A mutated AML.

Patients and specimens

100 cases of AML, genetically characterized at “Cervello Hospital” of Palermo, were enrolled. Diagnosis of AML was established according to WHO classification criteria by standard morphological, cytogenetic, immunophenotypic and molecular studies [17]. AML patients were selected to obtain comparable genetic groups: 50 DNMT3 mutated AML and 50 AML with other genetic alterations; patient’s characteristics are shown in (Table 1). Genes and miR-29s expression levels were evaluated in samples from bone marrow obtained at diagnosis or from samples of peripheral blood showing more than 70% leukemic cells. All AML samples were taken for diagnostic purposes from patients registered in the GIMEMA protocols AML12 (ISRCTN number 17833622) and AML 1310 (Eudra CT number 2010- 023809-36) who gave informed written consent for usage of biological material not necessary only for diagnostic purposes but also for scientific purposes. GIMEMA protocols were approved by Medical Ethics Committee of AOOR Villa Sofia Cervello “Comitato Etico Palermo 2”.

Cytogenetic and molecular analysis

Cytogenetic studies from bone marrow specimens were performed according standard procedures. For each culture 15-20 metaphases were karyotyped. Chromosomal abnormalities were classified according to the International System for Cytogenetic Nomenclature (ISCN) [18]. Molecular studies of leukemic cells were performed to detect gene fusion-transcripts associated with AML as described by EAC protocols [19]. FLT3 and NPM1 gene mutations analysis were performed according to Meshinchi S, et al. (2001) [20], and to Falini B, et al. (2005) [21], respectively. DNMT3A mutation were studied by standard sequencing of exon 23 and DNMT3A-R882-mutated samples were examined using Sequencing Analysis and SeqScape software analysis on a AB3130 Genetic Analyzer (Applied Biosystems, Foster City, CA).

miRNA profile

Total RNA was extracted from blast cells from bone marrow or peripheral blood at diagnosis by the QIAsymphony System (QIAGEN). To study miRNA expression levels, we used a lowdensity array system, the TaqMan Custom Array-TLDA human miRNA panel (4342265, Applied Biosystems). This technology is a method for quantitative RT-PCR (Q-RT-PCR) that can simultaneously assay the expression levels of up to 365 different miRNA genes on a single card (about 50% of the miRNA currently known, the composition of the array is viable at www. appliedbiosystems.com). The low-density arrays were performed in accordance to manufacturer’s protocols as previously described [10]. We utilized our previous data for the control categories (CD34 healthy donors and AML DNMT3A Wt) and we performed 3 new arrays with a pool of three cases each for a total of 9 new diagnosed DNMT3A mutated AML. Data were quantified using the SDS 2.3 software and normalized using the RNU48 as endogenous control. The cycle threshold (Ct) value, which was calculated relatively to the endogenous control, was used for our analysis (ΔCt). The 2-ΔΔCT (delta-delta-Ct algorithm) method was used to calculate the relative changes in gene expression among different groups. Heat map was constructed by RStudio software using ΔCt values; in particular, to increase the stability of results we applied filtering criteria that included miRNA which were reliably quantifiable in at least 70% of single categories.

miRNA expression

Total RNA was extracted from blast cells derived from bone marrow at diagnosis by the QIA symphony System (QIAGEN). To study miR-29s expression levels, we used TaqMan assay for miR29a, miR-29b and miR-29c (002447, 000413, 000587 Applied Biosystems). Briefly, reverse transcription reactions were performed for each sample using the High Capacity cDNA Archive Kit (Applied Biosystems) with the specific primer 5X for single miR-29s. One μL of cDNA from each sample was added to a 6.5 μL of TaqMan® Universal PCR Master Mix and 0.65 μL of 20X miR29s probe in a final volume of 13 μL. Q-PCR amplifications were performed on an ABI 7900HT Q-RT-PCR; data were quantified using the SDS 2.3 software and normalized using the RNU48 as endogenous control. The cycle threshold (Ct) values, which were calculated relatively to the endogenous control, were used for our analysis (ΔCt). The 2-ΔΔCT method was used to calculate relative changes in gene expression among different samples.

miR-29 targets prediction analysis

To predict potential target genes of miR-29s we utilized miRBase (http:// http://www.mirbase.org/). miRBase is a searchable database of published miRNA sequences and annotations. This resources is linked to TargetScanHuman application, allowing us to identify target gene of 29 miRNAs family simultaneously, applying a “consensus” aggregate PCT score> 0.8. This score reflected the Bayesian estimate of the probability that a site is conserved due to selective maintenance of miRNA targeting rather than by chance or any other reason not pertinent to miRNA targeting.

miR-29s targets gene expression

To determinate mRNA expression levels of miR-29s targets we used Q-RT-PCR with the 7900 TAQMAN system and predesigned available assay (Assay on Demand, Applied Biosystems). Q-RT-PCR data were quantified using the SDS 2.3 software and normalized using the ABL as endogenous control. To calculate relative changes in gene expression among different samples we used the 2-ΔΔCT method.

Statistical analysis

Results from each TaqMan run were quantified separately. To increase the accuracy of the results, we applied filtering criteria that included gene control expression which were reliably quantifiable (cutoff < 35 Ct). Undetermined values of Ct were estimated at 50 Ct (the last cycle of the reaction). Resulting ΔCts were used for our analysis. Differential expressions of miRNA and mRNA targets were defined using the Relative Expression Software Tool (REST) proposed by Pfaffl (2002) [22]. Fold change was estimated by REST. Namely, a gene was defined differentially expressed when estimated p-value was < 0.05.

To asses correlation between targets and miR-29s we utilized Spearman’s rank correlation test also.

Significantly differentially expressed miRNAs in DNMT3A-mutated AML

To identify the most significantly differentially expressed miRNA between purified CD34+ hematopoietic progenitor cells from healthy donors and AML DNMT3A mutated samples, we performed a supervised analysis by REST. We identified a set of 38 differential expressed miRNA, 2 of them being decremented and 36 upregulated in DNMT3A mutate samples (Table 2). Another step was finalized to study the differentially expressed miRNA in AML DNMT3A mutated compared to different genetic categories of AML, for this reason, CD34 samples were excluded from further analysis. Unsupervised hierarchical cluster analysis, generated on all 365 miRNAs, failed to clearly group the DNMT3A mutated in the same cluster (data not show). To find miRNAs with statistically significant differences in expression level among the major AML genetic groups and DNMT3A mutated group, we performed a REST analysis and we observed 83 differentially expressed miRNAs. Unsupervised hierarchical cluster analysis, generated on the differentially expressed miRNAs, clearly grouped the DNMT3A mutated in the same cluster, in particular those samples DNMT3A+ve/NPM1neg are grouped in the same subcluster (Figure 1). Furthermore we observed 20 miRNAs up regulated with a stronger significant p-value (Table 3). Due to the relevant role on epigenetic regulation of miR29s, that we found over expressed in DNMT3A compared to both control groups (CD34 and AML DNMT3a Wt), we decide to investigate expression levels of these miR29s in additional 50 newly diagnosed DNMT3A mutated AML patients.

Up-regulation of miR-29s in DNMT3A-R882-mutated AML

The analysis was finalized to study the expression level of miR-29s in a cohort of 100 AML including 50 DNMT3A mutated and 50 DNMT3A Wt samples; patient’s characteristics are shown in (Table 1). Statistical analysis performed by t-test, showed up-regulation of miR-29s (p<0.02) in DNMT3A mutated AMLs compared with AML without DNMT3A mutations. We evaluated the expression of single family member’s miR-29a, miR-29b and miR-29c in our cohort. ΔCt values of single samples (miR-29s Ct - RNU48 Ct) of both groups are reported respectively in (figure 2A, 3A and 4A). ΔCt values are plotted in ascending order that inversely correlate to expression levels. (Figure 2A and 4A) clearly showed a lower ΔCt in DNMT3A mutated group. Statistical analysis performed by REST showed a 2,4 fold up-regulation of miR-29a (p<0.01) in DNMT3A mutated AMLs compared with AML without DNMT3A mutations (Figure 2B), no difference in expression of miR-29b (Figure 3B) and a 7,5 fold up-regulation of miR-29c (p<0.001) in DNMT3A mutated AMLs compared with AML without DNMT3A mutations (Figure 3B).

miR-29 targets prediction analysis

We used TargetScanHuman to predict target gene of miR 29 family and we found 679 miRNAs with a PCT> 0.8 score. Next in this selection we looked for targets with a central role in the determination of methylation cell status and we identified the demethylases: TET2 (PCT> 0.99), TET1 (PCT> 0.99) and TDG (PCT: 0.82) and the methylases DNMT3B (PCT: 0.94) andDNMT3A (PCT: 0.80).

Expression levels of miR-29s targets in AML blast cells

We tested the expression of predicted and known to be mirR29s targets genes: the methylases DNMT3A and DNMT3B, the demethylases TET1, TET2 and TDG in our cohort of 100 AML patients. mRNA levels of target genes were obtained by Q-RTPCR, ΔCt values of single samples (target gene Ct – ABL Ct) of both groups are reported respectively in (Figure 5A,C and 6A,C,E). ΔCt values are plotted in ascending order that inversely correlate to expression levels. TET1 and TET2 expression are reported in (Figure 6A & 6C) that clearly showed a higher ΔCt in DNMT3A mutated group. We found that DNMT3A and DNMT3B genes do not exhibit a significant inverse correlation with miR-29s up regulation (Figure 5 B,D), on the contrary demethylases TET1 and TET2, but not TGD, exhibit an inverse trend compared with miR 29s up-regulation, being down-regulated in DNMT3A-mutated AML versus others, with a fold of 0.6 fold (p<0.009) and of 0.36 (p<0.001) respectively (Figure 6 B,D,F). Moreover we found an inverse correlation between TET2 and miR-29a with a Spearman’s rank correlation coefficient of -0,39. We also found low negative value (<-0,3) of Spearman’s rank correlation coefficient among TET1 vs miR-29a and TET1 and TET2 vs. miR29c.

Table 1: Patient’s characteristics.