Aberrant AML1 gene expression in the diagnosis of childhood leukemias not characterized by AML1 -involved cytogenetic abnormalities

Abstract

The AML1 (acute myeloid leukemia 1) gene, a necessary prerequisite of embryonic hematopoiesis and a critical regulator of normal hematopoietic development, is one of the most frequently mutated genes in human leukemia, involving over 50 chromosome translocations and over 20 partner genes. In the few existing studies investigating AML1 gene expression in childhood leukemias, aberrant upregulation seems to specifically associate with AML1 translocations and amplifications. The aim of this study was to determine whether overexpression also extends to other leukemic subtypes than the ones karyotypically involving AML1. We use quantitative real-time polymerase chain reaction methodology to investigate gene expression in 100 children with acute leukemias and compare them to those of healthy controls. We show that in childhood acute lymphoblastic leukemia, AML1 gene overexpression is associated with a variety of leukemic subtypes, both immunophenotypically and cytogenetically. Statistically significantly higher transcripts of the gene were detected in the acute lymphoblastic leukemia group as compared to the acute myeloid leukemia group, where AML1 overexpression appeared to associate with cytogenetic abnormalities additional to those that engage the AML1 gene, or that are reported as showing a “normal” karyotype. Collectively, our study shows that AML1 gene overexpression characterizes a broader range of leukemic subtypes than previously thought, including various maturation stages of B-cell acute lymphoblastic leukemia and cytogenetic types additional to those involving the AML1 gene.

Keywords

childhood leukemia gene overexpression gene amplification overall survival

Introduction

The acute myeloid leukemia 1 gene (AML1; also known as AML1B, RUNX1, CBFA2, or PEBP2aB1) was initially cloned as a target of the translocation breakpoint of chromosome 21 in the t(8;21)(q22;q22) AML1-ETO translocation, most frequently found in acute myeloid leukemia (AML).¹ Subsequently, several lines of evidence have shown AML1 to be a necessary prerequisite of embryonic hematopoiesis^2–6 and a critical regulator of normal hematopoietic development, as it is constitutively expressed in all lineages of the hematopoietic system, except for mature erythroid cells.⁷ Consequently, AML1 is normally involved in regulating the expression of a number of genes that are important for the development, function, and differentiation of hematopoietic cells; it therefore activates the transcription of several hematopoietic genes, including genes encoding cytokines and cytokine receptors, such as IL-3, GM-CSF, CSF-1R, the defensin protein, NP3, receptors of T- and B-cells, and genes specific to myeloid cells (e.g. neutrophil elastase and myeloperoxidase).⁸ However, because the AML1 binding sites are necessary but not sufficient for transcriptional activation, the gene has been hypothesized to act as a coordination agent that facilitates the assembly and cooperation of other transcriptional activation complexes.⁸ When AML1 cooperates with tissue-specific factors that regulate cell proliferation and differentiation (e.g. C/EBPa, AP-1, 1-ETS, PU.1, and c-Myb), it may act as an activator of transcription,^9–13 whereas in cooperation with suppressive factors such as Groucho and mSin3, it may act as a transcriptional repressor.^14–16

The AML1 gene is currently one of the most frequently mutated genes in human leukemia. Of particular interest is the fact that AML1, despite being an oncogene, is disrupted by chromosomal translocations in acute leukemias. Apart from t(8;21), which mainly occurs in M2 AML, AML1 has been found to be disrupted by t(12;21) (TEL-AML1) in childhood B-cell acute lymphoblastic leukemia (B-ALL), and indirectly by inv(16), which represent nearly one-third of all cases of AML and one-quarter of all cases of B-ALL.¹⁷ In addition, t(3;21) (AML1-EVI1) is often found in secondary AML and myelodysplasia.^18,19 Overall, over 50 chromosome translocations and over 20 partner genes involving the AML1 gene have been identified.²⁰ The possible role of AML1 gene overexpression in leukemogenesis has already been highlighted.^21–25 Possible mechanisms include the existence of mutations in the AML1 runt domain^26–28 and the increased number of gene copies in ALL, which usually arise from intrachromosomal amplification of the gene or may be derived from amplifications of chromosome 21.^29–33 The extra copies of AML1 appear to be a frequent karyotypic abnormality in childhood ALL, generally associated with an unfavorable prognosis, but not appearing to be associated with other cytogenetic abnormalities in ALL and presenting a rare and secondary phenomenon in therapy-related AML.³⁴ In the few existing studies that have focused on investigating AML1 gene expression in the acute leukemias of childhood, most cases appear to be characterized by an overexpression of the gene.^25,35 This, along with the fact that AML1 is the most frequent target of chromosomal translocations, the promiscuity of its recombinations, and the phenotypic heterogeneity of the leukemias bearing AML1 alterations, makes it a very interesting candidate in the pathogenicity of childhood leukemias. Notably, it has also been shown that AML1 amplification is implicated in the lineage conversion of childhood ALL to AML,³⁴ thus highlighting the diversity of AML1 involvement in the various leukemic subtypes and the possible implication of AML1 overexpression in the etiology of childhood leukemia.

Here, we investigate the transcript levels of AML1 in the diagnostic samples of a cohort of 100 pediatric acute leukemia patients and in 4 leukemic cell lines, and compare them to those of healthy controls. Our aim is to determine whether aberrant AML1 gene expression is a characteristic of childhood acute leukemias and whether it correlates with specific karyotypic abnormalities—for example, those involving AML1 translocations or AML1 amplifications. In addition, we investigate the prognostic impact of high gene expression in the clinical course of childhood leukemia, through possible correlations with minimal residual disease (MRD) and survival outcome.

Materials and methods

Ethics statement

Informed written consent was given by the parents, and all experiments were conducted in compliance with the international biomedical studies stipulations, with reference to the Declaration of Helsinki of the World Medical Association. The study was reviewed and approved by the Institutional Review Committee for the protection of human subjects at Aghia Sofia Children’s Hospital with reference code 2996/07-02-12.

Patient and control samples

A total of 100 children were diagnosed with acute leukemia (86 with ALL and 14 with AML) at the Hematology/Oncology Unit of Medical School, National and Kapodistrian University of Athens at “Aghia Sofia” Children’s Hospital, Athens, Greece. Bone marrow (BM) samples were collected at diagnosis, and all patients were classified morphologically and immunologically. Patient data were as follows: males: 58, females: 42, median age at diagnosis (years): 6.69 ± 4.80, median male age (years): 6.98 ± 4.56, median female age (years): 6.30 ± 5.14, median ALL age (years): 6.29 ± 4.55, and median AML age (years): 9.14 ± 5.71. Patient clinical data are summarized in Supplementary Table 1. BM samples were also collected from 50 healthy children, on the suspicion of neoplasia, but were found negative for the disease, and were therefore used as negative controls. Of these, only 42 yielded good-quality RNA for subsequent analysis. Four childhood leukemia cell lines were also included in the study as additional samples: REH (pre-B ALL), CCRF-CEM (T-cell ALL), CCRF-SB (T-cell ALL), and THP-1 (acute monocytic leukemia). All cell lines were obtained from the European Collection of Authenticated Cell Cultures (ECACC, UK).

RNA extraction

According to the manufacturer’s protocol, total RNA was extracted from BM leukocytes/myelocytes, using TRIzol reagent (Invitrogen Inc., USA). Extraction of RNA from the four cell lines was performed with the same protocol. RNA was treated with amplification grade DNase I to eliminate any residual genomic DNA from the sample and further purified with the RNeasy^® Micro Kit (Qiagen Inc., Germany). RNA quantification measurements were performed on the NanoDrop^® ND-1000 Spectrophotometer (Nanogen Inc., USA).

Quantitative reverse transcription polymerase chain reaction

Total RNA was prepared from blast cells, and gene expression was investigated with the Plexor™ One-Step quantitative reverse transcription polymerase chain reaction (qRT-PCR) System Kit (Promega Inc., USA) as described previously,³⁶ using the LightCycler2 and LightCycler480 Instruments (Roche Diagnostics GmbH). The data were normalized to the GAPDH (glyceraldehyde 3-phosphate dehydrogenase) housekeeping gene. The sequences of the primers in the 5′ to 3′ direction were as follows: AML1F: 5′-GT-GGG-TAC-GAA-GGA-AAT-GAC-TCA-AA-3′ and AML1R: 5′-GCA-GCG-TGG-TAA-AAG-AAA-TCA-TTG-AG-3′. Amplified sequences included the AML1 transcript variant 1 (Refseq: NM_001754.4) and the AML1 transcript variant 2 (Refseq: NM_001001890.2), which correspond to isoforms AML1c and AML1b, respectively.

Real-time data pre-processing and analysis

Real-time data were collected and pre-processed using the LightCycler Software Version 3.5 and the LightCycler Software version SW1.5 (Roche Diagnostics GmbH) for the LightCycler2 and LightCycler480 instruments, respectively. Data were then imported to the Plexor™ Analysis Software (www.promega.com/plexorresources/) (Promega Inc., USA), and gene expression was obtained in the form of Ct values. Data pre-processing was performed with Microsoft^® Excel.

Statistical analysis

The independent t-test failed to reveal a statistically reliable difference in the mean GAPDH expression levels (Ct values) between the patients and controls (p = 0.349); hence, we accepted GAPDH Ct values as reliable normalization constants. Independent t-tests and one-way analysis of variance (ANOVA) were then performed to compare the mean AML1 expression levels between groups. The Ct value of each gene (for every patient and healthy control) was normalized against the Ct value obtained for the housekeeping gene, GAPDH (i.e. ΔCt_AML1 = Ct of AML1 − Ct of GAPDH), and the normalized ΔCt value was used to compare the genes between patients and controls (e.g. ΔΔCt_AML1 = ΔCt _AML1 of patient 1 − (mean) ΔCt _AML1 of controls). Gene expression (or fold change) was calculated using the ΔΔCt after normalizing to GAPDH (i.e. 2^−ΔΔCt). Patients were regarded as significantly overexpressing a gene if the corresponding fold-change value was >1 (and thus corresponding to a positive log2 value). Comparisons between patient subgroups were regarded significant if p ⩽ 0.05. Log-Rank survival curves were used to estimate significance of overall survival (OS) and leukemia-free time (LFT). Data pre-processing was performed with Microsoft^® Excel. Statistical and data analyses have been performed using the IBM SPSS Statistics Program (IBM Corp. Released 2012. IBM SPSS Statistics for Windows, Version 21.0. Armonk, NY: IBM Corp.). Raw data are included in Supplementary file 1.

Results

qRT-PCR analysis

A qRT-PCR assay was established. AML1 was found to be differentially overexpressed, as compared to the healthy controls, in 54 patients (54% of total patient cohort): 51 patients with ALL (1 with pro-pre-B, 32 with B-common, 15 with pre-B, and 3 with T-cell ALL), 3 patients with AML and one of the four cell lines: CCRF-CEM (T-cell ALL). Some patients revealed a much higher expression ratio (R > 5.0) than the rest (p = 0.000) (Supplementary file 1). Therefore, patients were divided into three groups: normal (healthy) ratio (R ⩽ 1.0), high ratio (1.0 < R ⩽ 5.0), and very high ratio (5.0 < R) (Table 1).

Table 1.

Gene expression ratios (fold change) in childhood leukemia patients.

	AML1 gene expression
	Normal ratio, R ⩽ 1.0	High ratio, 1.0 < R ⩽ 5.0	Very high ratio, 5.0 < R	p-value*
Patients with healthy/normal expression ratio	0.36 ± 0.27 (n = 49)	–	–	–
All patients and cell lines (n = 104)	0.36 ± 0.27 (n = 49)	2.10 ± 0.97 (n = 41)	12.46 ± 6.87 (n = 14)	0.000
Patients and cell lines with ALL (n = 89)	0.38 ± 0.28 (n = 37)	2.07 ± 0.92 (n = 38)	12.46 ± 6.87 (n = 14)	0.000
Patients with pro-pre-B ALL (n = 2)	0.19 (n = 1)	2.09 (n = 1)	–	–
Patients with B-common ALL (n = 51)	0.34 ± 0.30 (n = 19)	2.19 ± 0.99 (n = 24)	14.63 ± 7.39 (n = 8)	0.000
Patients with pre-B ALL (n = 24)	0.57 ± 0.23 (n = 9)	1.75 ± 0.74 (n = 11)	8.54 ± 4.15 (n = 4)	0.000
Cell lines with pre-B ALL (n = 1)	0.34 (n = 1)	–	–	–
Patients with T-cell ALL (n = 7)	0.31 ± 0.25 (n = 4)	2.49 ± 1.17 (n = 2)	17.91 (n = 1)	0.016
Cell lines with T-cell ALL (n = 2)	0.12 (n = 1)	–	5.26 (n = 1)	–
Patients with Burkitt ALL (n = 2)	0.33 ± 0.19 (n = 2)	–	–	–
Patients and cell lines with AML (n = 15)	0.29 ± 0.25 (n = 12)	2.48 ± 1.75 (n = 3)	–	0.000
Patients with AML (n = 14)	0.27 ± 0.25 (n = 11)	2.48 ± 1.75 (n = 3)	–	0.001
Cell lines with AML (n = 1)	0.48 (n = 1)	–	–	–

ALL: acute lymphoblastic leukemia; AML: acute myeloid leukemia.

p-value has been calculated by comparing the high ratio to the healthy (normal) ratio.

Correlation analysis between gene expression and leukemic subtypes

A statistically significant difference in gene expression was detected between the two main leukemia subtypes, ALL and AML, with the ALL group exhibiting significantly higher transcript levels of AML1 as compared to the other group (p = 0.000) (Figure 1). This was somewhat expected, as only 3 out of the 14 AML patients appeared to overexpress AML1 as compared to controls. When examining gene expression with respect to the various immunophenotypic subgroups in ALL, no significant correlation was found between abnormal gene expression and any of the subgroups. However, in the B-common and pre-B subgroups, the majority (approximately 1/3) of the patients appeared to overexpress the gene. Gene expression was not examined with respect to the various immunophenotypic subgroups in AML due to the small number of patients in each subgroup (1–3 patients). Similarly, no significant difference in abnormal gene expression was detected among the various cytogenetic subgroups, in both ALL and AML. In the AML subtype, overexpression of AML1 was observed only in patients with karyotypic abnormalities of the gene, that is, carrying the AML1-ETO translocation, inv(16), and extra copies of the gene (Supplementary Table 1). However, in childhood ALL, it appeared that AML1 aberrant gene expression did not exclusively correlate with specific karyotypic abnormalities, such as those involving AML1 translocations or AML1 amplifications. Although the majority of patients with TEL-AML1 (13/15), as well as the majority of patients with extra copies of AML1 (18/27), exhibited high transcript levels of the gene, abnormally high gene expression was also observed in other subgroups, such as those with BCR-ABL1 translocation (2/3 patients), with MLL rearrangements (2/9 patients), and with extra copies of genes other than AML1 (7/15 patients) (Supplementary Table 1).

Figure 1.

AML1 gene expression in leukemia patients.

Correlation analysis between gene expression and MRD levels

Patients with MRD on day 33 (d33) and/or failure to treatment are presented in Table 2. Failure to treatment is defined as relapse or death, events that are sometimes preceded by refractory disease. Overall, 27 patients (27%) had positive MRD on d33. A total of 12 patients had failure to treatment (i.e. relapsed or died) but only 9 of them had concurrent positive MRD on d33. Therefore, a total of 30 patients had MRD on d33 and/or failure to treatment. Eleven (11) of the 30 patients in Table 2 appear to overexpress AML1, whereas the t-test failed to reveal a statistically reliable difference in AML1 gene expression levels between patients with positive MRD and patients with negative MRD on d33 (p = 0.181), or between patients who relapsed and patients who did not relapse (p = 0.668).

Table 2.

Patients with MRD (d33) and/or failure to treatment.

Patient no.	Patient/cell-line code	Gender	Diagnosis (Dx)	Age at Dx (years)	Blast count (%)	WBC (cells/uL)	Immunophenotype	Karyotype	Cytogenetics	Additional cytogenetics	MRD (d33)	Failure to treatment	Overexpression
5	O-48	M	ALL	14.96	78.9	92,960	B-common	45,XY,t(9;22)(q34;q11.2),der(9)t(9;13)(p12;q1)del(11)(q13qter),13³/46,sl,+mar⁶/45,sl1,mar,add(22)(p11.1)³/88,sl x 2,-17,-Y³	BCR/ABL		−	Relapse/death	+
6	86	M	ALL	1.75	85.9	6460	B-common	47,XY,+21⁸/46,XY¹²		Trisomy 21 and four copies of AML1	+		+
9	61	F	ALL	6.05	85.9	10,670	B-common	46,XX,t(9;22)(q34;q11),i(9)(q10)³/46,XX³	BCR/ABL		+	Relapse/death	−
13	O-57	M	ALL	7.33	42	46,000	B-common	87,<4n>,XY,+Y,+1,+1,+2,+2,+3,+3,+4,+4,+5,+5,+6,+6,+7,+7,+8,+8,+9,+9,+10,+11,+11,+12,+12,+13,+14,+15,+16,+16,+17,+17,+18,+18,+19,+19,+20,+20,+21,+21,+22,+22⁵/46,XY⁵	TEL/AML1		−	Relapse	+
23	143	M	ALL	4.76	95.7	108,300	B-common	46,XY,add(19)(q13)⁴/46,XY³	TEL/AML1		+	Relapse/death	−
28	148	M	ALL	3.62	88.9	42,560	B-common	52,XY,+4,+6,+10,+17,+21,+21⁵/46,XY²		Extra copies of AML1	+	Relapse/death	−
35	186	M	ALL	9.67	86	1900	B-common	N/A		Extra copies of ABL and AML1	+	−	+
36	188	M	ALL	9.67	86	1900	B-common	57,XY,+X,+1,del(1)(p32p36),+4,+5,+6,+8,+9,+10,+21,+21,+22⁹/46,XY⁶		Extra copies of PBX1, AML1, BCR, and ABL1	+	Refractory disease/death	+
37	195	M	ALL	12	97	42,080	B-common	46,XY²⁰		One copy of TEL	+	−	−
40	207	M	ALL	8.25	85	420,000	B-common	48,XY,t(9;22)(q34;q11),+21,+der(22) t(9;22)(q34;q11)¹²	BCR/ABL	Three copies of AML1	+	Relapse	+
42	232	F	ALL	4.17	90	3160	B-common	64,XX,+X,ins(1;?)(q21;?),+2,+4,+5,+8,+9,+10,+11,+12,+14,+15,+16,+18,+18,+21,+21,+22³/46,XX⁸		Extra copies of TEL, AML1, MLL, and BCR	+	−	−
43	235	F	ALL	2.17	82	3000	B-common	46,XX⁵	TEL/AML1		+	−	+
47	249	M	ALL	5 years 4 months	99	100,000	B-common	53,XY,+X,del(1)(p32;p36),+1,+1,+6+10,+14,+21,+21³/46,XY¹²		Extra copies of AML1 and PBX1	+	−	−
48	252	F	ALL	1 year 2 months	97	11,200	B-common	56,XX,+X,+4,+6,+9,+10,+14,+17,+18,+21,+21¹²/46,XX⁸		Extra copies of AML1 and ABL1	+	−	+
67	209	F	ALL	16.83	96	1400	pre-B	46,XX²²		Three copies of E2A	+	−	−
72	230	F	ALL	2.17	87	82,600	pre-B	60,XX,+X,+4,+5,+6,+9,+10,+11,+14,+14,+16,+17,+18,+21⁴/46,XX¹²			+	−	−
73	238	F	ALL	14.5	88	19,000	pre-B	46,XX,dup(21)(q)¹²/46,XX³		Five copies of AML1 (97.4%)	+	−	+
76	278	M	ALL	1	100	160,000	pre-B	46,XY²⁰			+	−	−
81	108	M	ALL	2.96	84.4	143,500	Cortical T	46,XY,del(6)(q12q22),i(9)(q10q10),t(11;14)(p13;q11)⁷/46,XY⁸		TCRA/D rearrangement and three copies of ABL	+	Relapse	−
83	229	M	ALL	8.5	83	547,100	Cortical T	46,XY²⁰			−	Relapse	−
87	80	M	AML	9.51	22.4	3390	M1	48,XY,add(3)(p25),+8,inv(16)(p13q22),+22²/47,sl,-8¹¹/47,sdl1,+8,-22⁷	inv16		+	Refractory disease/death	+
88	174	F	AML	3.86	90	7400	M1	46,XX⁴			+	−	−
89	208	F	(MDS→)AML	15.17	40	1210	M1	46,XX,t(2;11)(q31;p15)²⁰	MLL		+	−	−
91	261	M	AML	15.67	89	24,810	M2/M5	46,XY,inv(16)(p13q22)¹⁵/46,XY⁵	CBFβ/MYH11		+	Relapse	−
92	268	M	AML	13.17	100	34,170	M3	46,XY,t(15;17)(q22;q21)¹⁶	PML/RARA		+	−	−
93	257	F	AML	13.17	99	1070	M3	46,XX,t(15;17)(q22;q21)¹⁴/46,XX¹	PML/RARA		+	−	−
96	154	M	AML	11.44	89.7	232,300	M5	47,XY,+8,der(9)t(9;11)(p10;q10)t(9;11)(p22;q23),der(11)t(11;17)(q10;p10),der(17)t(9;17)(p10;q10,t(9;11)(p22;q23)¹⁵	MLL		+	Refractory disease/death	−
97	215	F	(MBL→)AML	11.83	60	3980	M5	46,XX,t(2;11)(q31;p15)²⁰	MLL		+	−	−
99	183	F	AML	1.42	79	2300	M7	48,XX,+8,+21²/46,XX⁵		Extra copies of CEP8 and AML1	+	−	+
100	221	M	(NBL→)AML	10.75	50	8050	Therapy-related	46,XY,t(11;19)(q23;p13.3)¹⁵	MLL		+	−	−

MRD: minimal residual disease; d33: day 33 (of therapy); M: male; F: female; ALL: acute lymphoblastic leukemia; AML: acute myeloid leukemia; NBL: neuroblastoma; “+”: positive; “−”: negative.

Correlation analysis between gene expression and OS/LFT

A 3-year survival period has been calculated for all the patients (100) enrolled in the study (Supplementary file 1). LFT has been calculated as the event-free or disease-free survival, that is, the time to relapse. No significant correlation appeared to exist between high AML1 gene expression and OS (p = 0.970) or between high AML1 gene expression and LFT, that is, time to relapse (p = 0.973) (Figure 2).

Figure 2.

Overall survival (OS) and leukemia-free time (LFT) analyses with respect to AML1 gene expression.

Discussion

In this study, we have found AML1 to be differentially overexpressed in just over half of the patients (54%) and in one of the four cell lines, as compared to the healthy controls. Although abnormal gene expression appears to associate with both pediatric ALL and AML, a statistically significant difference in gene expression was denoted between the two groups, with the ALL patients exhibiting significantly higher transcript levels of AML1. In addition, none of the AML patients exhibited very high transcript levels of AML1 (R > 5.0), as opposed to ALL patients (Table 1). This observation could be due to the fact that there were far more ALL patients with karyotypic abnormalities of the AML1 gene, such as rearrangements or amplifications, than AML patients. Nonetheless, it has been shown that the AML1a transcript variant of AML1 induces the development of ALL, but not AML, when transplanted into lethally irradiated mice.²⁴

In our childhood ALL samples, abnormally high transcript levels of the gene have been observed at various maturation stages of B-cell ALL (pro-pre-B, B-common, and pre-B) and in T-cell ALL (Table 1). Based on this observation, it is difficult to suggest an implication of AML1 overexpression in the leukemic transformation of B- and T-lymphoid progenitors in childhood, as promiscuous transcriptional expression of several lineage-affiliated genes has been shown to precede lineage commitment of multipotent progenitor cells without altering their biological potential.³⁷ Furthermore, a very high ratio of gene expression (R > 5.0) has been observed in T-cell leukemia, as well as in B-cell leukemia with B-common and pre-B maturation (Table 1). While AML1 has been shown to be involved in thymocyte differentiation via control of cell proliferation,³⁸ with AML1 overexpression influencing the transition of T-cells from the CD4-CD8- to the CD4+CD8+ maturation stage,^24,39,40 this is the first report documenting abnormally high transcript levels in childhood B-cell ALL with B-common and pre-B cell maturation. Our results suggest an association of AML1 overexpression in the leukemic B-cell transformation, especially in the relatively immature B-common subtype, where the highest transcript levels are recorded (R = 30.47) (Table 1). This association of AML1 upregulation with the various immunophenotypic subtypes of ALL could reflect the many critical roles of the gene in normal hematopoietic development. Since AML1 is essential and constitutively expressed in all lineages of the hematopoietic system, it is possible that abnormal changes in its transcriptome would also reflect in lymphoid-cell associated malignancy, thus appearing significantly higher in the relatively immature B-common ALL subtype.

In our childhood AML samples, AML1 overexpression has been observed only in patients with cytogenetic abnormalities of the gene, that is, a patient with t(8;21) (AML1-ETO), a patient with inv(16), and a patient with extra copies of AML1. The rest of the AML cases in our patient cohort (11/14) and the THP-1 cell line, who carried no or other cytogenetic aberrations (such as MLL rearrangements and PML-RARA fusions), displayed normal transcript levels of AML1 as compared to controls (Supplementary Table 1). Therefore, our results are in agreement with the general notion that abnormal gene expression in AML is associated with karyotypic abnormalities of the AML1 gene.¹⁷ However, in the ALL samples, aberrantly high transcript levels of the AML1 gene, as compared to controls, did not appear to correlate exclusively with karyotypic abnormalities involving AML1 translocations or amplifications. Instead, AML1 overexpression has been observed in a broader range of cytogenetic subgroups, such as those with BCR-ABL1 translocation, with MLL rearrangements, and with extra copies of genes other than AML1 (Supplementary Table 1). While it is difficult to validate a reason for this observation, one possible explanation might be that in the ALL patients with no apparent cytogenetic abnormalities of AML1, abnormally high levels of the gene might be related to overexpression of the AML1a isoform. Since AML1a is known to interfere with the function of the other two variants, AML1b and AML1c (collectively called AML1), by exhibiting higher DNA sequence affinity for target genes and therefore by antagonizing the binding sites of AML1,⁴¹ it is possible that apart from the normal function of AML1 in hematopoiesis, AML1a also distorts the function of AML1 in leukemogenesis. In addition, it has already been proposed that in cases of AML1 rearrangement or mutation, it is the lack of the transactivation domain that triggers leukemogenesis.⁴² Coincidentally, the protein encoded by AML1a, as opposed to the proteins encoded by AML1b and AML1c (i.e. AML1), lacks a transactivation domain and hence it is possible that it disrupts the normal function of AML1 in a manner similar to that of the leukemia-associated fusion proteins.²⁴ Another possibility is that the AML1 gene interacts with other genes, also critical for both normal hematopoiesis and leukemogenesis. Such an example is the recently unveiled physical and functional interaction of AML1 with MLL, not only regulating the epigenetic status of essential cis-regulatory elements for hematopoietic genes but also with possible implications in the epigenetic regulation of malignant hematopoiesis.⁴³

Of our 27 MRD-positive patients, only 9 were found to exhibit abnormally high expression levels of AML1, as compared to controls (Table 2). As a result, no statistically significant differences in gene expression were obtained between the MRD-positive patient group and the MRD-negative group (p = 0.181). This signifies that AML1 overexpression cannot be used to distinguish between positive and negative MRD status and, hence, does not have the potential to determine the response to therapy in childhood leukemia patients. In addition, no significant correlation appeared to exist between aberrantly high AML1 gene expression and dismal OS (p = 0.970), or relapse (p = 0.973) in childhood leukemias (Figure 2). In other words, in our patient cohort, the abnormally high transcript levels of AML1 do not show a direct relationship with disease prognosis. This could be either due to an absence of interference of the AML1 gene product with the effects of the treatment regimen that was used or due to a balanced effect of AML1 on the leukemic cells. For example, AML1 elicits signaling pathways that protect the leukemic cell, while at the same time restrict its clonogenic potential.^44,45 Specifically, wild-type AML1 represses nuclear factor-kappa B (NF-κB) signaling,⁴⁶ which is expected to limit the capacity of the cells expressing AML1 to generate malignant subclones, as abnormal functioning of the NF-κB signaling network is essential for malignant cell dissemination in tissues.⁴⁷ As a consequence, wild-type AML1 could compensate for part of the disrupted negative feedback signals and restrict the leukemogenic potential of the clones that express it.

The multitude of chromosome translocations and partner genes involving the AML1 gene, along with the dramatic biological consequences related to AML1 deregulation in hematological malignancies,^20,48 as well as its recently unveiled involvement in the malignant transformation of non-hematopoietic tissues such as epithelial breast cells,^49–51 constitute irrefutable evidence that the gene plays a central role in the gene regulatory network that governs cellular homeostasis and self-renewal. Over the years, it has also become apparent that many important regulators of hematopoietic stem cell development are also implicated in the abnormal self-renewal capacity of the leukemic stem cells. Of particular interest is the microRNA (miRNA)-based deregulation of AML1 gene expression, as miRNAs have the ability to affect AML1 transcriptional dosage by differentially regulating the level of AML1 splice variants through targeting of specific 3′ untranslated regions (UTRs). More specifically, the AML1α splice variant bears a very short 3′UTR, whereas isoforms AML1b and AML1c contain a long 3′UTR. While mir-27 can target the 3′UTR of both short and long AML1 isoforms, mir-17, which is often found upregulated in leukemias and lymphomas, can target the 3′UTR of only the longer isoforms.^52–55 In other words, miRNAs have the ability to produce different biological responses by differentially regulating AML1 transcription. It is through the identification of such regulators, and of their downstream targets, that we hope to reveal how deregulated transcription on key developmental pathways leads to the malignant phenotype and to prospectively identify new targets for successful therapy against malignancy. Study limitations should be considered carefully as they could pose obstacles toward the scope of the investigation. Sample size and availability, as well as the choice of appropriate controls, are of greatest importance, as not only are there different types of cells in a patient sample but also similarities in the transcriptional profile of essential hematopoietic genes between normal hematopoiesis and leukemia.

Conclusion

Collectively, our study shows that abnormally high AML1 gene expression characterizes a variety of leukemic subtypes of childhood acute leukemias, both immunophenotypically and cytogenetically. Statistically significantly higher transcripts of the gene were detected in the ALL group as compared to the AML group, and at various maturation stages of B-cell ALL and in T-cell ALL. While AML1 overexpression has been observed in T-cell differentiation previously, this is the first report documenting an aberrantly high gene expression in childhood B-cell ALL with B-common and pre-B cell maturation. In addition, in ALL but not in AML, overexpression appears to correlate with cytogenetic abnormalities other than those known to involve AML1. However, as seen in our patient cohort, the abnormally high transcript levels of AML1 at the diagnosis of childhood leukemias do not show a direct relationship with disease prognosis, nor do they appear to have the potential to determine the response to therapy. In summary, abnormal activation of AML1 appears to be a characteristic of the leukemias that carry cytogenetic abnormalities additional to those that engage the AML1 gene, or that are reported as showing a “normal” karyotype. This highlights the unpredictable diversity of AML1 gene involvement in the etiology of childhood leukemia that remains to be unraveled.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Meyers

Downing

Hiebert

SW.

Identification of AML-1 and the (8;21) translocation protein (AML-1/ETO) as sequence-specific DNA-binding proteins: the runt homology domain is required for DNA binding and protein-protein interactions. Mol Cell Biol 1993; 13: 6336–6345.

Okuda

van Deursen

Hiebert

. AML1, the target of multiple chromosomal translocations in human leukemia, is essential for normal fetal liver hematopoiesis. Cell 1996; 84: 321–330.

Wang

Stacy

Binder

. Disruption of the Cbfa2 gene causes necrosis and hemorrhaging in the central nervous system and blocks definitive hematopoiesis. Proc Natl Acad Sci USA 1996; 93: 3444–3449.

Takakura

Watanabe

Suenobu

. A role for hematopoietic stem cells in promoting angiogenesis. Cell 2000; 102: 199–209.

Chen

Yokomizo

Zeigler

. RUNX1 is required for the endothelial to haematopoietic cell transition but not thereafter. Nature 2009; 457: 887–891.

Boisset

van Cappellen

Andrieu-Soler

. In vivo imaging of haematopoietic cells emerging from the mouse aortic endothelium. Nature 2010; 464: 116–120.

North

Stacy

Matheny

. RUNX1 is expressed in adult mouse hematopoietic stem cells and differentiating myeloid and lymphoid cells, but not in maturing erythroid cells. Stem Cells 2004; 22: 158–168.

Lutterbach

Hiebert

SW.

Role of the transcription factor AML-1 in acute leukemia and hematopoietic differentiation. Gene 2000; 245: 223–235.

Zhang

Hetherington

Meyers

. CCAAT enhancer-binding protein (C/EBP) and AML1 (CBF alpha2) synergistically activate the macrophage colony-stimulating factor receptor promoter. Mol Cell Biol 1996; 16: 1231–1240.

10.

Westendorf

Yamamoto

Lenny

. The t(8;21) fusion product, AML-1-ETO, associates with C/EBP-alpha, inhibits C/EBP-alpha-dependent transcription, and blocks granulocytic differentiation. Mol Cell Biol 1998; 18: 322–333.

11.

Selvamurugan

Chou

Pearman

. Parathyroid hormone regulates the rat collagenase-3 promoter in osteoblastic cells through the cooperative interaction of the activator protein-1 site and the runt domain binding sequence. J Biol Chem 1998; 273: 10647–10657.

12.

Kim

Sieweke

Ogawa

. Mutual activation of Ets-1 and AML1 DNA binding by direct interaction of their autoinhibitory domains. EMBO J 1999; 18: 1609–1620.

13.

Mao

Frank

Zhang

. Functional and physical interactions between AML1 proteins and an ETS protein, MEF: implications for the pathogenesis of t(8;21)-positive leukemias. Mol Cell Biol 1999; 19: 3635–3644.

14.

Aronson

Fisher

Blechman

. Groucho-dependent and -independent repression activities of Runt domain proteins. Mol Cell Biol 1997; 17: 5581–5587.

15.

Levanon

Goldstein

Bernstein

. Transcriptional repression by AML1 and LEF-1 is mediated by the TLE/Groucho corepressors. Proc Natl Acad Sci USA 1998; 95: 11590–11595.

16.

Lutterbach

Westendorf

Linggi

. A mechanism of repression by acute myeloid leukemia-1, the target of multiple chromosomal translocations in acute leukemia. J Biol Chem 2000; 275: 651–656.

17.

Rubnitz

Crist

WM.

Molecular genetics of childhood cancer: implications for pathogenesis, diagnosis, and treatment. Pediatrics 1997; 100: 101–108.

18.

Nucifora

Birn

Espinosa

3rd . Involvement of the AML1 gene in the t(3;21) in therapy-related leukemia and in chronic myeloid leukemia in blast crisis. Blood 1993; 81: 2728–2734.

19.

Rubin

Larson

Anastasi

. t(3;21)(q26;q22): a recurring chromosomal abnormality in therapy-related myelodysplastic syndrome and acute myeloid leukemia. Blood 1990; 76: 2594–2598.

20.

De Braekeleer

Douet-Guilbert

Morel

. RUNX1 translocations and fusion genes in malignant hemopathies. Future Oncol 2011; 7: 77–91.

21.

Kurokawa

Tanaka

. Overexpression of the AML1 proto-oncoprotein in NIH3T3 cells leads to neoplastic transformation depending on the DNA-binding and transactivational potencies. Oncogene 1996; 12: 883–892.

22.

Fujii

Hayashi

Niki

. Overexpression of AML1 renders a T hybridoma resistant to T cell receptor-mediated apoptosis. Oncogene 1998; 17: 1813–1820.

23.

Osato

Ito

Increased dosage of the RUNX1/AML1 gene: a third mode of RUNX leukemia?

Crit Rev Eukaryot Gene Expr 2005; 15: 217–228.

24.

Liu

Zhang

. Overexpression of an isoform of AML1 in acute leukemia and its potential role in leukemogenesis. Leukemia 2009; 23: 739–745.

25.

Mikhail

Serry

Hatem

. AML1 gene over-expression in childhood acute lymphoblastic leukemia. Leukemia 2002; 16: 658–668.

26.

Osato

Asou

Abdalla

. Biallelic and heterozygous point mutations in the runt domain of the AML1/PEBP2alphaB gene associated with myeloblastic leukemias. Blood 1999; 93: 1817–1824.

27.

Preudhomme

Warot-Loze

Roumier

. High incidence of biallelic point mutations in the Runt domain of the AML1/PEBP2 alpha B gene in Mo acute myeloid leukemia and in myeloid malignancies with acquired trisomy 21. Blood 2000; 96: 2862–2869.

28.

Taketani

Taki

Takita

. AML1/RUNX1 mutations are infrequent, but related to AML-M0, acquired trisomy 21, and leukemic transformation in pediatric hematologic malignancies. Genes Chromosomes Cancer 2003; 38: 1–7.

29.

Niini

Kanerva

Vettenranta

. AML1 gene amplification: a novel finding in childhood acute lymphoblastic leukemia. Haematologica 2000; 85: 362–366.

30.

Dal Cin

Atkins

Ford

. Amplification of AML1 in childhood acute lymphoblastic leukemias. Genes Chromosomes Cancer 2001; 30: 407–409.

31.

Streubel

Valent

Lechner

. Amplification of the AML1(CBFA2) gene on ring chromosomes in a patient with acute myeloid leukemia and a constitutional ring chromosome 21. Cancer Genet Cytogenet 2001; 124: 42–46.

32.

Busson-Le Coniat

Nguyen Khac

Daniel

. Chromosome 21 abnormalities with AML1 amplification in acute lymphoblastic leukemia. Genes Chromosomes Cancer 2001; 32: 244–249.

33.

Penther

Preudhomme

Talmant

. Amplification of AML1 gene is present in childhood acute lymphoblastic leukemia but not in adult, and is not associated with AML1 gene mutation. Leukemia 2002; 16: 1131–1134.

34.

Podgornik

Debeljak

Zontar

. RUNX1 amplification in lineage conversion of childhood B-cell acute lymphoblastic leukemia to acute myelogenous leukemia. Cancer Genet Cytogenet 2007; 178: 77–81.

35.

Harewood

Robinson

Harris

. Amplification of AML1 on a duplicated chromosome 21 in acute lymphoblastic leukemia: a study of 20 cases. Leukemia 2003; 17: 547–553.

36.

Adamaki

Lambrou

Athanasiadou

. Implication of IRF4 aberrant gene expression in the acute leukemias of childhood. PLoS ONE 2013; 8: e72326.

37.

Miyamoto

Akashi

Lineage promiscuous expression of transcription factors in normal hematopoiesis. Int J Hematol 2005; 81: 361–367.

38.

Sato

Ito

Nunomura

. Requirement of transcription factor AML1 in proliferation of developing thymocytes. Immunol Lett 2003; 89: 39–46.

39.

Basecke

Feuring-Buske

Brittinger

. Transcription of AML1 in hematopoietic subfractions of normal adults. Ann Hematol 2002; 81: 254–257.

40.

Hayashi

Abe

Watanabe

. Overexpression of AML1 transcription factor drives thymocytes into the CD8 single-positive lineage. J Immunol 2001; 167: 4957–4965.

41.

Tanaka

Ogawa

. An acute myeloid leukemia gene, AML1, regulates transcriptional activation and hemopoietic myeloid cell differentiation antagonistically by two alternative spliced forms. Leukemia 1997; 14(2): 299–302.

42.

Tanaka

Ogawa

. An acute myeloid leukemia gene, AML1, regulates hemopoietic myeloid cell differentiation and transcriptional activation antagonistically by two alternative spliced forms. EMBO J 1995; 14: 341–350.

43.

Koh

Wang

. RUNX1 meets MLL: epigenetic regulation of hematopoiesis by two leukemia genes. Leukemia 2013; 27: 1793–1802.

44.

Hyde

Zhao

Alemu

. RUNX1 is required for hematopoietic defects and leukemogenesis in Cbfb-MYH11 knock-in mice. Leukemia 2015; 29: 1771–1778.

45.

Zhou

Ching

Chng

WJ.

Aberrant nuclear factor-kappa B activity in acute myeloid leukemia: from molecular pathogenesis to therapeutic target. Oncotarget 2015; 6: 5490–5500.

46.

Nakagawa

Shimabe

Watanabe-Okochi

. AML1/RUNX1 functions as a cytoplasmic attenuator of NF-κB signaling in the repression of myeloid tumors. Blood 2011; 118: 6626–6637.

47.

Vlahopoulos

Cen

Hengen

. Dynamic aberrant NF-κB spurs tumorigenesis: a new model encompassing the microenvironment. Cytokine Growth Factor Rev 2015; 26: 389–403.

48.

Lam

Zhang

DE.

RUNX1 and RUNX1-ETO: roles in hematopoiesis and leukemogenesis. Front Biosci 2012; 17: 1120–1139.

49.

Janes

KA.

RUNX1 and its understudied role in breast cancer. Cell Cycle 2011; 10: 3461–3465.

50.

Scheitz

Tumbar

New insights into the role of Runx1 in epithelial stem cell biology and pathology. J Cell Biochem 2013; 114: 985–993.

51.

Chimge

Frenkel

The RUNX family in breast cancer: relationships with estrogen signaling. Oncogene 2013; 32: 2121–2130.

52.

Ota

Tagawa

Karnan

. Identification and characterization of a novel gene, C13orf25, as a target for 13q31-q32 amplification in malignant lymphoma. Cancer Res 2004; 64: 3087–3095.

53.

Dixon-McIver

East

Mein

. Distinctive patterns of microRNA expression associated with karyotype in acute myeloid leukaemia. PLoS ONE 2008; 3: e2141.

54.

Chen

. Aberrant overexpression and function of the miR-17-92 cluster in MLL-rearranged acute leukemia. Proc Natl Acad Sci USA 2010; 107: 3710–3715.

55.

Rossetti

Sacchi

RUNX1: a microRNA hub in normal and malignant hematopoiesis. Int J Mol Sci 2013; 14: 1566–1588.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.34 MB

0.39 MB