Real-time PCR analysis of the apoptosis related genes in ATRA treated APL t(15;17) patients

All-trans retinoic acid (ATRA) treatment of the acute promyelocytic leukemia (APL) have subsequently resulted in cell apoptosis, but the molecular mechanism of this effect remains elusive. In order to understand a possible involvement of genes regulating apoptotic signal pathways, expression levels of bcl2, bax, dapk1, myc, bad, wt1, and mcl genes were analyzed during ATRA treatment in five APL patients with t (15;17) using Real- time PCR (LightCycler). Two samples from each patient were compared to each other: primary diagnostic sample and a sample taken at remission. Effect of the ATRA treatment was demonstrated by the concomitant induction of cd14 and il1β genes in four patients. Also other apoptosis related genes were found down-regulated in general but especially the down regulated levels of wt1 and bax attract attention. Result suggested that ATRA dependent apoptosis of APL was under the control of both internal and external pathways without relationships to the amount of the blast populations. Ratio of bcl2 to bax may be more important for this regulation than the ratio of bcl2 to bad. Either bcl2 family or less known apoptosis related genes as wt1 will still be required to further studies in this setting.


Introduction
All-trans retinoic acid (ATRA) dependent differentiation induction is used in the treatment of the acute promyelocytic leukemia (APL) successfully. It has been clearly demonstrated in pre-clinical studies that cells induced to differentiate subsequently die via apoptosis (Martin et al., 199l;Gillis et al., 1995;Lo Coco et al., 1998).
Apoptosis might be induced by external signals at the surface of the cell. Fas and the TNF receptors transmit a signal to the cytoplasm that leads to activation of caspase 8. Myc expression can activate this signalling pathway prematurely (Hoffman et al., 2002). In the second mechanism, a group of bcl2 family proteins protect or initiate apoptosis. The proand anti-apoptotic bcl2 proteins can make heterodimers and the ratio of these determines the execution of cell death. Among these members bcl2 is anti-apoptotic while bax and bad are pro-apoptotic (Oltvai et al., 1993;Adams and Cory, 1998;Srivastava et al., 1999;Mitchell et al., 2000;White et al., 2001).
Mcl1 is also described as a differentiation-related mitochondrial anti-apoptotic factor related to erk signal transduction pathway. There may be other genes in different pathways leading to apoptosis which are not yet fully characterized. Among these genes, dapk1 is a potential mediator cell death, and over expression of wt1 may lead to apoptosis (Deiss et al., 1995;Townsend et al., 1999;Mitchell et al., 2000).
Apoptosis has been traditionally studied by characteristic cleavage of DNA into nucleosomal fragments. Unfortunately, this method is unable to help in studying different genes simultaneously and not very sensitive.
In our study we analysed the expression of the apoptosis related genes during the ATRA treatment in five APL patients with t(15;17) using an extremly sensitive technique known as quantitative Real-Time PCR (LightCycler, Roche Diagnostics GmbH, Germany). We have used SYBR Green I dye binding method and determined the expression levels of seven apoptosis related genes (bcl2, bax, mcl1, dapk1, myc, bad and wt1). Differentiation effect of the ATRA Real-tim e PCR analysis of the apoptosis related genes in ATRA treated APL t(15;17) patients treatment was aimed to demonstrate by cd14 and IL1β genes. Two samples from each patient were compared to each other: primary diagnostic sample in diagnosis and a sample taken at complete remission.

M aterials and M ethods
Our study was based on bone marrow aspirates from 5 APL patients having t(15;17) and treated with ATRA (45 mg/m 2 , daily). Among of these patients, 4 were  in adult (average age: 27) and 1 was in childhood age (2). Diagnosis of APL is based on FAB criteria's. All of the patients were positive for PML-RAR alpha fusion transcripts detected by RT-PCR (Table 1) The samples were obtained from the University of Istanbul and all the treatments have been performed between 1998-2001. Mononuclear cells from bone marrow aspirates were isolated by ficoll-hypaque centrifuga-tion. Total RNA was extracted by using guanidium thiocyanate-phenol-chloroform extraction method as previously described (Chomznsky et al., 1987). RNA samples were treated with DNase I and quantitative Real-time PCR was performed as we described previously (Savli et al., 2002;Savli et al., 2003). Primer sequences of the selected genes have been shown in Table 2. On line fluorescence curves of  PCR amplifications and melting curve analysis were shown in Figure 1. Standart curve analysis of the serial dilutions of the RPS9 housekeeping gene was shown in Figure 2.
Obtained gene expression values were normalized using a housekeeping gene in both patient and healthy control groups. For this aim rps9 was used which belongs to a gene family accepted more reliable than either of the classical housekeeping genes, in human and mouse malignant cell lines (Bhatia et al., 1994;Zhong et al., 1999). A new software tool was used, named REST (relative expression software tool), and compared treated and non-treated samples of each patient. The mathematical model used is based on the PCR efficiencies and the crossing point deviation between the samples (Pfaffl et al., 2002).

R esults and D iscussion
Specific results were indicated that induction of the apoptosis related genes takes place in parallel with the induction of differentiation (Table 3). This is demonstrated by the concomitant induction of cd14 and IL1β genes in four of the five patients while myc was down-regulated. Myc is a well known oncogene who has many functions in the cell cycle and these findings were suggested that external pathway of the apoptosis is involved in the regulation of ATRA dependent affect on APL (Cole et al., 1986;Schumacher et al., 2001).
It has been informed that ATRA can induce differentiation and reduce intracellular bcl2 levels without altering the susceptibility to drug-induced apoptosis and ATRA seems to increase chemo sensitivity by down regulation of bcl-2 in cell line (Bruel et al., 1995;Hu et al., 1996;Ketley et al., 1997).
ATRA can also down regulate bcl2 expression in native AML blasts for a subset of patients independent of their FAB classification (Pisani et al., 1997).
Here we demonstrated that bcl2 gene levels were down-regulated in four of the five patients and these findings were in concordance with the previous literature. Pro and anti-apoptotic bcl2 proteins can make heterodimers where the ratio determines the sensitivity of leukemic cells to apoptosis. In some reports the bcl2 to bax ratio has been inversely related with drug induced apoptosis in vitro and clinical response to chemotherapy. (Martin et al., 1991;Banker et al., 1997;Decaudin et al., 1997;Pepper et al., 1997;Adams and Cory, 1998;Meijerink et al., 1998;Hoffman et al., 2001).
We found that bcl2 to bad ratios were 3.09, 0.04, 15, 114, and 1.5 while bcl2 to bax ratios were 16, 160, 17.6, 1.6, and 23.2 after the ATRA treatment, for patients 1 to 5 respectively. This suggests that the ratio of bcl2 to bax may be more important for regulation of ATRA dependent apoptosis in this setting than the ratio of bcl2 to BAD.
Mcl1 gene ratios were found under-expressed in three patients (1, 3, and 4) while only slightly upregulated in patients 2 and 5. Also, dapk1 expressions were down regulated in patients 1, 2 and 3 but the ratios were very close to normal expression levels. On the other hand, wt1 gene levels were extremely down-regulated in patients 2, 3, 4.
ATRA dependent apoptosis of APL cells were observed under the control of the both internal and external pathways without relationship to the amount of the blast populations. Our data was confirming the hypothesis that the ratio of bcl2 to bax determines the ATRA dependent apoptotic response. Either bcl2 family or other less known apoptosis related genes as wt1 will still be required to study in larger groups.