Acute myeloid leukemia (AML) is characterized by a block in differentiation of myeloid cells accompanied by suppression of normal production of blood cells and dissemination of immature cells from their normal hematopoietic environment. In order to gain insight into the mechanisms governing these features of leukemia, we have examined mRNA expression patterns in transgenic mice that express PML-RARA (associated with human acute promyelocytic leukemia, APL) and BCL2. Such mice represent a model system in which these genetic changes combine to initially arrest differentiation in a pre-leukemic state, with acute leukemia arising only upon the acquisition of additional genetic changes. In the present study we identified the set of genes whose transcription is altered in pre-leukemic and leukemic bone marrow in comparison to bone marrow from normal mice. We generated and analyzed three sets of gene expression data from cDNA microarrays: (1) PRE-LEUKEMIA vs. NORMAL (6600 substances), (2) LEUKEMIA vs. NORMAL (6677 substances) and (3) PRE-LEUKEMIA vs. LEUKEMIA (3762 substances). Greater than 90% of substances showed similar expression in the pre-leukemic and leukemic samples. Sixty unique genes were differentially expressed, in a statistically significant manner, with two-fold or more difference in expression, and presenting in all three datasets. Comparison of these 60 revealed genes to a published human AML expression dataset (

Bullinger et al.,
N Engl J Med
,
350
:
1605
,
2004
) identified eight genes with the same pattern of expression in our mouse model and in human APL, and four common genes with opposite expression. For mouse pre-leukemia vs. leukemia as compared to all human AMLs vs. APL, the eight genes that showed a similar change were: COL1A2, COL5A2, EIF5A, LAMR1, LGALS1, NOTCH2, SNN, and ACP5. Of particular interest to our laboratory are two genes: LGALS1 and NOTCH2. Among the identified genes the first is most over-expressed, and the second is most under-expressed in human APL. Next, we identified genes in our dataset co-expressed with LGALS1 or NOTCH2 (correlation coefficient >0.85). Among discovered genes were revealed transcription factors (NKX6-1, GTF3A, ZFP95, SOX11, HOXA4 and more), translation factors (EIF3S6, EIF5A, EIF4A1 and more), differentiation factors (RQCD1, NEUROD2, CREBBP/EP300 inhibitory protein 1 and more), oncogenes (Ski, COUP-TF1 and more) and others. We are currently seeking to identify which of these changes may contribute to AML pathogenesis. Gene profiling using such a comparative approach for analysis will allow better understanding of leukemogenesis, and can also be applied for studies of tumor progression in other malignancies.

Topics:
genes, leukemia, mice, galectin 1, acute promyelocytic leukemia, bcl-2 protein, cancer, dna, complementary, gene expression profiling, laminin receptor-1

Author notes

Corresponding author

2005, The American Society of Hematology
2004
Sign in via your Institution