| Edited by Hiroshi Iwasaki. Sanggyu Lee: Corresponding author. E-mail: slee@knu.ac.kr |
Recurring chromosome rearrangements are a common feature of hematopoietic malignancies. In acute myeloid leukemia (AML), the MLL (mixed-lineage leukemia) gene located at human chromosome band 11q23 is frequently involved in reciprocal chromosome translocation with other genes, which results in a break in MLL and the partner genes, and leads to formation of a new fusion gene. The fusion gene contains 5’ MLL joined to the 3’ part of the partner gene (Thirman et al., 1993; Rubnitz et al., 1996). Nearly 40 different partner genes, many of which are transcription factors, have been identified as being involved in this translocation (Daser and Rabbitts, 2004). Of these genes, the AF9 gene located at human chromosome band 9p22 is known as one of the most common partner genes with MLL. Chromosome translocation t(9;11)(p22;q23) leads to form the MLL-AF9 fusion protein that functions as a facilitator of cell growth directly (Nakamura et al., 1993; Pession et al., 2003).
The availability of human leukemia cell lines as a self-renewing resource of accessible and manipulable living cells has contributed significantly to a better understanding of the pathophysiology of hematopoietic tumors (Drexler et al., 2000). For researching acute myeloid leukemia carrying t(9;11)(p22;q23), the cell lines with cytogenetically diagnosed t(9;11)(p22;q23) have been widely used in laboratories. Mono Mac 6 (MM6), one of these cell lines, is a human acute monocytic leukemia cell line (AML-M5; FAB-classification) which was originally established from the peripheral blood of a 64-year-old male patient with monoblastic leukemia in 1988 (Ziegler-Heitbrock et al., 1988). The MM6 cell line have a complex karyotype like hypotetraploid, including 2 copies of normal chromosome 9 and 11 as well as 2 copies of the t(9;11) translocated forms (MacLeod et al., 1993; Super et al., 1995). In addition, the MM6 cell line exhibits a phagocytosis of antibody-coated erythrocytes in 80% of the cells, and it is known to exhibit the phenotypic and functional features of mature monocytes (Ziegler-Heitbrock et al., 1988). Therefore, the MM6 cell line has been applied for various experiments as a model of monocytes and t(9;11) AML cells.
In addressing the genome response to various stimuli on cell lines, microarray-based approaches have been widely used. The microarray-based approaches can only detect the known genes or ESTs. However, SAGE (Serial Analysis of Gene Expression) can detect not only relatively rare transcripts but also novel transcripts, regardless of the expression level.
In this study, SAGE was performed to determine the gene expression profile of the MM6 cell line. This study is an attempt to determine whether the gene expression pattern of the cell line is different from the primary human t(9;11) cells that have a same chromosome translocation, in an aim to provide a new insight into the AML related studies.
Mono Mac 6 cell line was obtained from DSMZ (Braunschweig, Germany). Mono Mac 6 Cells were cultured in an RPMI 1640 medium supplemented with 10% FBS, 100 unit/ml penicillin, and 100 μg/ml streptomycin in a humidified 5% CO2 atmosphere.
SAGE libraries were constructed following SAGE protocol (Lee et al., 2001). Briefly, total RNA and mRNA were purified from MM6 cells. Double-strand cDNA were synthesized, and 3' cDNA were purified using NlaIII digestion. SAGE tags were released from 3' cDNA for cancatemerization and cloning into the pZero vector. Sequencing reactions for SAGE clones were performed with ABI Big-Dye 3.1 kit. SAGE tag sequences were collected with an ABI3730 sequencer. Sequences passed Phred20 were used for SAGE tag extraction. SAGE tags were extracted from the sequences using SAGE 2000 software.
SAGE tags were matched to SAGEmap database updated on February 19, 2008 (http://www.ncbi.nlm.nih.gov/SAGE/). For the SAGE tags shared by multiple genes, only the single-matched SAGE tags were selected. To determine the differentially expressed genes between the primary human t(9;11) and MM6 cells, SAGE data were analyzed using IDEG6 (http://telethon.bio.unipd.it/bioinfo/IDEG6_form/). For statistical analysis, a general Chi-squared test, with the significance threshold set as 0.05, was performed. The differentially expressed genes between the primary human and MM6 cells were selected based on p values < 0.05 and over 4-fold expression change.
For the comparison with MM6, the SAGE data of primary human t(9;11) cells were obtained from the paper published by Lee et al. (2006). The data of three t(9;11) AML-M5/5a (FAB classification) cells were integrated and extracted randomly to make one expression profile. As a result, total 65,307 tags having unique 36,465 tags were constructed.
For the functional classification of the differentially expressed genes between the primary human t(9;11) and MM6 cells, EASE (version 2.0) software (http://david.niaid.nih.gov/david/ease.htm) was used for gene ontology analysis. EASE can perform a statistical analysis of gene categories in a gene list to find those that are most overrepresented, either because of under- or over-expression. This enabled the “biological process” for the analyzed genes able to be defined.
58,472 SAGE tags were collected from the MM6 cells and 14,661 unique SAGE tags were identified. These unique SAGE tags were matched to the reference database (SAGEmap), which showed that 74% of the tags were matched to known transcripts (Table 1). Generally, the tags of high copies accounted for high percentage in matching tags and the majority of the novel tags were concentrated in the low-abundance class.
 View Details | Table 1 Distribution of the SAGE tags from MM6 cells |
To check highly expressed genes in MM6 cells, the tags with more than 100 copies were arranged (Table 2). There were 68 SAGE tags, and all the tags except two were matched to known genes or EST. Of these, 33 tags were represented to the ribosomal proteins and 7 were transcribed loci similar to protein sequences. There were a variety of genes among the remaining 26 genes. Some were the housekeeping genes such as GAPDH and FTH1 and some were functional genes related with chemotaxis such as PPIA and PPIB (Xu et al., 1992; Yurchenko et al., 2001) (Table 2). Although the highly expressed genes were helpful for determining the character of the MM6 cells, because the SAGE tag data of MM6 only show the expression level of the individual transcript, in order to obtain more meaningful data, further analysis is required. Accordingly, we performed the comparison analysis between MM6 cells and the primary human t(9;11) AML cells that have the same chromosome translocation t(9;11).
 View Details | Table 2 Genes highly expressed in MM6 cells |
To examine whether the pattern of gene expression of the Mono Mac 6 cell line is different from that of primary cells from bone marrow samples with t(9;11) AML patients (the SAGE tags data were from Lee et al., 2006), the SAGE tags from both cells were compared via statistical and bioinformatical analysis. Among the 884 unique SAGE tag obtained from the analysis, p values < 0.05 and 4-fold expression change, 693 (78%) tags were increased in MM6 cells and 191 (22%) were decreased. And 830 (94%) were matched to the SAGE map database and 54 (6%) had no match. To classify the 830 matched genes, the SAGE tags were analyzed by the DAVID Functional Annotation Tool (http://david.abcc.ncifcrf.gov/). 491 out of 644 over-expressed genes had DAVID ID. Majority of the genes were related with biosynthetic and metabolic processes on the GOTERM_BP_3 category (Table 3A). On the other hand, 136 out of the 186 under-expressed genes were matched to DAVID IDs, and the result of grouping the genes showed that many genes were involved in biosynthetic and metabolic processes, similar to the over-expressed genes (Table 3B). Collectively, although there were a number of the alternatively expressed genes between MM6 and primary cells, the majority of the genes belonged to similar biological process. That implies the gene expression of MM6 cells is generally similar to that of primary AML cells.
 View Details | Table 3 Classification of the alternatively expressed genes |
Then, in order to find out the additional feature of MM6, we performed an in-depth study. To examine which biological pathways were strengthened or weakened in MM6 cells, the over-expressed or under-expressed genes were applied to the biocarta pathway category on the DAVID Functional Annotation Tool. Interestingly, four genes involved in Erk1/Erk2 MAPK pathway were exhibited to be over-expressed in MM6 cells. The pathway is known to govern the growth, proliferation, differentiation and survival of many, if not all, cell types and cause various diseases, including cancer when deregulated (Orton et al., 2005). Especially the pathway including RAS is known to be implicated as a key component of proliferative drive in AML (Bowen et al., 2005). On our data, HRAS was expressed only in MM6 cells (Table 4). According to Lin’s group (1996) (Lin et al., 1996), AML patients with positive HRAS in complete remission were at risk for early relapse and, thus, it was suggested that HRAS might be involved in the process of leukemogenesis and progression in AML. In this cue, it is possible to expect that immortal trait of MM6 cells is highly associated with the expression of HRAS. On the other hand, in the primary AML cells, since there was no expression of the gene, we presumed that other members of RAS family might be expressed in the cells. As a result of our investigation, however, although NRAS tag (GCACTGTACT) and KRAS tag (GTCACTCTCC) were exhibited in the SAGE tag data, they are not seemed to be meaningful data because of their too high p-values (Table 4). So, as there are RAS-positive and -negative AML cells, it is thought that the primary human t(9;11) cells assayed by SAGE were extracted from the AML patients having a less expression of RAS.
 View Details | Table 4 The genes involved in Erk1/Erk2 MAPK signaling pathway |
In addition to HRAS gene, other three genes involved in the pathway were also significantly over-expressed in MM6 cells. They were platelet-derived growth factor receptor, alpha polypeptide (PDGFRA), MAP kinase interacting serine/threonine kinase 2 (MKNK2), and a member of the ribosomal S6 kinase, RPS6KA1 (Table 4). Except these four genes, we investigated the expression level of the other members implicated in the pathway. Most of the genes had a tendency to express similarly in both cells or more in MM6. GRB2 that mediates between receptor and RAS (Lowenstein et al., 1992) and SRF that is a transcription factor involved in the pathway (Spencer and Misra, 1999) were especially over-expressed in MM6, although p value of the genes was more than 0.05 (Table 4). Consequently, these data suggest that MM6 cells have a strong Erk1/Erk2 MAPK pathway relatively to the primary AML cells.
In summary, we assayed the expression of MM6 cell line using SAGE in order to determine if the cell line is proper to study AML with t(9;11)(p22;q23). The comparative analysis between MM6 cells and primary human t(9;11) AML cells shows that there was not much distinctive difference of gene expression between two cells. However, the several component genes in Erk1/Erk2 MAPK signaling pathway, including HRAS, were over-expressed in MM6 cells. Therefore, the results indicate that MM6 cell line can be a good model for researching HRAS- positive t(9;11) AML. Based on this result, MM6 cell line is expected to be applied to the correlation study between the expression of HRAS and leukemia.
This study was supported by a research grant from Ministry of Education, Science and Technology (NBM2300812), Republic of Korea.
|
Index