| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Fusion Transcripts
1 Johnson & Johnson Research Pty Limited, Australian Technology Park, Eveleigh NSW 1430, Australia.
2 Kanematsu Laboratories, Royal Prince Alfred Hospital, Camperdown, Sydney NSW 2050, Australia.
aAddress correspondence to this author at: Biomedical Building, Level 4, 1 Central Ave., Australian Technology Park, Eveleigh NSW 1430, Australia. Fax 61-2-8396-5811; e-mail tapplega{at}medau.jnj.com.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
fusion transcripts provide a readily accessible marker for diagnosis of acute promyelocytic leukemia (APL) and for monitoring response to therapy. Survival rates are improved by therapies guided by such monitoring. We assessed the potential of DzyNA reverse transcription-PCR (RT-PCR) for measurement of PML/RAR fusion transcripts.
Methods: Parallel single-tube DzyNA RT-PCR protocols were developed to allow real-time fluorescent quantification of PML/RAR fusion transcripts and a low abundance control transcript, normal BCR. Calibration curves, generated using cell line RNA, allowed estimation of these transcripts in RNA from patients with APL at various stages of the disease.
Results: DzyNA RT-PCR calibration curves were linear for both transcripts over a broad range and demonstrated interassay variations of 12% (mean, 658 ng) and 10% (mean, 263 ng), respectively. The protocols detected low concentrations of transcripts and resolved twofold dilutions. PML/RAR mRNA was quantified in 10 patients at diagnosis and in 1 patient over a 7-year period. Monitoring of transcript concentrations effectively reflected the disease course in one patient and demonstrated that an increase in PML/RAR transcripts can be detected 4–6 months before hematologic relapse, with no false-positive results.
Conclusion: DzyNA RT-PCR has potential for use in clinical practice as a tool for diagnosis of APL and forsubsequent monitoring of minimal residual disease and detection of molecular relapse.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(RAR) gene on chromosome 17, which generate novel PML/RAR fusion genes and transcripts. The resulting chimeric proteins inhibit PML-dependent apoptotic pathways and block myeloid differentiation by direct transcriptional inhibition of retinoic acid-responsive genes (1)(2). All patients who harbor PML/RAR transcripts respond to all-trans-retinoic acid (ATRA) differentiation therapy in combination with chemotherapy (3). Although almost all patients achieve complete remission within 1–3 months, 30% of these patients eventually relapse. Fortunately, salvage treatment achieves a second remission in the majority of these patients.
Many studies have shown that detection of PML/RAR transcripts by qualitative nested reverse transcription-PCR (RT-PCR) precedes and predicts hematologic relapse, whereas negative RT-PCR results are associated with long-term remission. The small number of reports that do not appear to follow this trend may reflect failure to collect specimens at critical time points, poor quality RNA, and/or inherent differences in the sensitivities of the specific RT-PCR assays used. The interpretation of molecular monitoring remains a contentious issue and highlights a need to define and standardize assays for detection of molecular relapse. This is particularly important because clinical studies indicate that survival rates can be improved by treating patients at molecular relapse as opposed to frank hematologic relapse (4)(5). Preliminary studies suggest that sensitive, quantitative methods will improve the predictive power of molecular monitoring of APL compared with qualitative methods (6)(7). This report describes the use of DzyNA RT-PCR for quantification of PML/RAR fusion transcripts as a clinical tool for diagnosing and monitoring APL.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. The sequences of the primers appear below. The 5' DzyNA-PCR primers contain (a) a 5' region harboring the antisense to a DNA enzyme (DNAzyme) (8) designed to cleave the generic substrate SubG5-DF (underlined sequence indicates the complement of the arms that hybridize to the SubG5-DF; bold font indicates the complement of the DNAzyme catalytic domain) and (b) a 3' region that is complementary to the target transcript (sequence that is not underlined). Three primers were used for amplification of PML/RAR mRNA. The 5' DzyNA primer, 5PML6/Z2, has a gene-specific portion complementary to exon 6 of PML and has the following sequence: 5'-CACCAAAAGAGAACTGCAATTCGTTGTAGCTAGCCTTTCAGGACCCACAGGAGCGCAGGAGCCCCGTCATAGGA-3'. The two 3' PCR primers, both complementary to exon 3 of RAR, have the following sequences: primer 3RAR3/3, 5'-GGGCACTATCTCTTCAGAAC-3'; primer 3RAR3/5, 5'-TTGTAGATGCGGGGTAGA-3'. Two primers were used for amplification of BCR mRNA. The 5' DzyNA primer, 5BCR14/Z7, has a gene-specific portion complementary to exon 14 of the BCR transcript and has the following sequence: 5'-CACCAAAAGAGAACTGCAATTCGTTGTAGCTAGCCTTTCAGGACCCACAGGAGCGCACTCAGCCACTGGATTTAA-3'. The 3' primer, 3BCR15/2, is complementary to exon 15 of the BCR transcript and has the following sequence: 5'-TCCAGGGTGCAGTACAGA-3'. Primers were synthesized by Geneworks.
|
A single generic DzyNA substrate, SubG5-DF, was used to monitor amplification of both BCR and PML/RAR mRNA in parallel reactions. SubG5-DF was labeled with the quencher 4-(4'-dimethylaminophenylazo)benzoic acid and the reporter 6-carboxyfluorescein, which were located on either side of the DNAzyme cleavage site within the substrate. SubG5-DF is a chimeric oligonucleotide containing both RNA (lower case) and DNA bases (uppercase) and has the following sequence: 5'-CACCAAAAGAGAACTGCAATguTTCAGGACCCACAGGAGCG-3'. The substrate is designed such that the bond between the gu ribonucleotides is cleaved by DNAzymes generated during DzyNA RT-PCR. A 3' phosphate group was added to prevent extension by DNA polymerase during PCR. The substrate was synthesized by Trilink.
templates for rt-pcr
Both total RNA and cDNA were used as templates in DzyNA experiments. The human APL cell line NB4 (ACC207; German Collection of Microorganisms and Cell Cultures), which harbors the PML/RAR fusion gene, was used as a positive control. RNA from NB4 was diluted in peripheral blood lymphocytes (PBLs) from healthy volunteers for construction of calibration curves for estimating expression in patients’ samples. The human cell lines A549 (ATCC CCL-185) and MEG-01 (ATCC CRL-2021) were used as negative controls in all other experiments. Additional experiments to assess the linear range of the assay were conducted using the pTL2 plasmid, which contains PML/RAR cDNA (kindly donated by Prof. Chambon, Universite Louis Pasteur, Paris, France). Bone marrow (BM) samples, collected from consenting APL patients before or during therapy, had been referred to the Kanematsu Laboratories for PML/RAR RT-PCR. Total RNA was extracted from human BM and cell lines by TRIzol reagents (part no. 15596-026; Life Technologies) and the QIAamp RNA Blood Mini Kit (part no. 52304; Qiagen), respectively.
amplification and detection
All procedures were performed in physically isolated, positive-pressure rooms dedicated to PCR set-up, extraction, nested PCR, or thermocycling to avoid contamination with PCR product. Thermocycling and monitoring of fluorescence during PCR and data analysis were carried out using the ABI PRISM® 7700 Sequence Detection System (Applied Biosystems) with MicroAmp optical 96-well reaction plates and optical caps (Applied Biosystems). All DzyNA RT-PCR reactions contained 10 mM Tris (Ambion), 75 mM KCl (pH 8.3 at 25 °C), 6 mM MgCl2, 300 µM each deoxynucleotide triphosphate, 7.5 mM (NH4)2SO4, 0.2 µM SubG5-DF, 0.62 µM 6-carboxy-X-rhodamine passive reference dye, 1.25 U of AmpliTaq Gold DNA polymerase (Applied Biosystems), 2.5 U of reverse transcriptase (RNase H minus M-MLV; Promega Corporation), and 10 units of rRNasin (Promega Corporation) in nuclease-free water (Ambion) per 25-µL reaction. Reactions for analysis of PML/RAR transcripts contained 0.02 µM 5PML6/Z2, 0.06 µM 3RAR3/5, and 0.14 µM 3RAR3/3, and an additional 10 U of reverse transcriptase (RNase H minus M-MLV). Reactions for analysis of BCR transcripts contained 0.02 µM 5BCR14/Z7 and 0.2 µM 3BCR15/2. All reactions were performed in duplicate. Additional control reactions were performed in parallel and contained all reaction components with the following changes; no-template control mixtures lacked template RNA, and negative control mixtures contained RNA from A549, MEG-01, or healthy PBLs only. Thermocycling conditions were 50 °C for 89 min, 95 °C for 10 min, 10 cycles of 95 °C for 15 s and 65 °C minus 1 °C/cycle for 1 min, and 50 cycles of 95 °C for 15 s and 55 °C for 1 min.
data analysis
Sequence Detection Software (Applied Biosystems) was used to monitor the increase in 6-carboxyfluorescein fluorescence at 530 nm after cleavage of the substrate by DNAzymes. Analysis was performed using correction for the 6-carboxy-X-rhodamine included in the DzyNA RT-PCR mixtures. A cycle threshold value (Ct) was determined for each sample corresponding to the cycle when fluorescence exceeded a defined baseline signal threshold (
Rn) within the log phase of product accumulation. Baseline settings for analysis were in the range of 3–25 cycles. Calibration curves were generated by plotting the log of the amount of template against the Ct value. Quantification of the amounts of RNA in reactions containing unknown numbers of PML/RAR or BCR transcripts were estimated from the calibration curves.
calibration curves and analysis of patient samples
Calibration curves for quantification of PML/RAR concentrations in patients’ specimens were constructed using 500 ng of total RNA, comprising twofold dilutions of NB4 RNA in RNA from healthy PBLs over the range of 100% (500 ng of NB4) to 1.56% (7.8 ng of NB4) for diagnosis or to 0.1% (488 pg of NB4) for monitoring minimal residual disease. Calibration curves were constructed for estimating BCR expression in patients, using total RNA from NB4 diluted 10-fold in nuclease-free water (from 500 ng to 50 pg per 25-µL reaction). The amount of total RNA used when quantifying both transcripts in patients was 50 ng for all samples, with the exception that 500 ng was used for PML/RAR analysis of samples collected after treatment. Quantitative data were expressed as ng-equivalents of the total NB4 RNA and as a ratio of disease transcripts to control transcripts (RDC), given as a percentage:
 |
compared with BCR analysis. Limits for quantification and inclusion of data were set to establish whether patient RNA was of sufficiently high quality and to eliminate potential false negatives. Samples were deemed "detectable" when one, but not both duplicates were quantifiable by DzyNA RT-PCR and the presence of fusion transcripts could be confirmed by two-step nested PCR. Potential false negatives were excluded if a 50-ng sample contained <10 ng of BCR equivalents of NB4 RNA.
confirmation of results with nested pcr
All nested PCR mixtures contained 1 µL of PCR product from the DzyNA RT-PCR, 10 mM Tris, 75 mM KCl (pH 8.3 at 25 °C), 2 mM MgCl2, 300 µM each deoxynucleotide triphosphate, 7.5 mM (NH4)2SO4, 1.25 U of AmpliTaq Gold DNA polymerase, 0.02 µM 3RAR3/3 (5'-TCTCTTCAGAACTGCTGCTC-3'), and 0.02 µM 5PML1/1 (5'-AAGTGAGGTCTTCCTGCCCAA-3') in nuclease-free water per 25-µL reaction. Thermocycling was performed on a Perkin-Elmer 9600 programmed for 95 °C for 10 min, 5 cycles of 95 °C for 15 s and 70 °C minus 1 °C/cycle for 30 s, and 25 cycles of 95 °C for 15 s and 60 °C for 30 s. The 89-bp product was visualized on a 5% Nusieve® GTG® agarose gel (BMA) stained with ethidium bromide at 250 µg/L (Amresco).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
0.99) over six orders of magnitude (1 µg to 1 pg), based on 10-fold dilutions of NB4 total RNA. (The amplification and calibration curves for BCR are available as a data supplement accompanying the online version of this article, at http://www.clinchem.org/content/vol48/issue8/). Similarly, the assay targeting PML/RAR exhibited good linearity (R2 0.98) over five orders of magnitude (106–10 copies), based on 10-fold dilutions of pTL2 plasmid. The linear range and ability to detect low concentrations of PML/RAR transcripts were confirmed by analysis of patient RNA serially diluted in RNA from healthy PBLs. Calibration curves were linear (R2 0.99) over five orders of magnitude, and fusion transcripts were detected in patient RNA that had been diluted 1:105 (total volume, 25 µL). Analysis of the same dilutions with the BCR DzyNA RT-PCR assay confirmed equal amounts of total RNA in all patient dilutions. In comparison, PML/RAR transcripts could be detected only in 1:104 dilutions of NB4 RNA diluted with A549 RNA (total volume, 25 µL), indicating relatively low PML/RAR expression in the NB4 cell line (R2 0.98).
Analysis of NB4 RNA diluted twofold with RNA from MEG-01 demonstrated the capacity for DzyNA RT-PCR to discriminate between small differences in expression (Fig. 2
). The PML/RAR assay resolved twofold dilutions with high precision (R2 = 0.992) across the range of 100–0.1% NB4 RNA, corresponding to 500 ng to 488 pg of NB4 equivalents. Both the PML/RAR and the BCR DzyNA RT-PCR assays were specific for target RNA and exhibited low interassay imprecision. The CV for the Ct values (n = 18) was 2.7–3.9% for the BCR assay (50 pg to 500 ng of NB4 RNA) and 1.2–4.0% for the PML/RAR assay (16–500 ng of NB4 RNA). Estimates of unknowns from seven experiments of the BCR and PML/RAR calibration curves demonstrated CVs of 12% (mean, 658 ng) and 10% (mean, 263 ng), respectively.
|
analysis of apl specimens
The PML/RAR and BCR DzyNA RT-PCR assays were used to analyze total RNA extracted from 10 patients at the time of diagnosis of APL. At diagnosis, PML/RAR was expressed at an average of 109 ng of NB4 equivalents per 50 ng of patient total RNA (range, 36–275 ng-equivalents), and BCR was expressed at an average of 194 ng of NB4 equivalents per 50 ng of patient total RNA (range, 103–378 ng-equivalents). The RDC ranged from 33% to 105%, and values were reproducible in duplicate experiments performed on different days.
DzyNA RT-PCR was also used to retrospectively monitor the concentrations of PML/RAR in one patient for whom serial samples and the clinical history were available for a 7-year period (Fig. 3
). Relative concentrations of PML/RAR were high in BM samples taken at disease presentation (points 1 and 2). The patient was treated with ATRA; 6 weeks later (point 3), the patient’s BM demonstrated some abnormal morphology but had reverted to a normal karyotype. Retrospective analysis of this sample by DzyNA RT-PCR showed that PML/RAR expression had decreased, but was still quantifiable despite the absence of cytologic evidence of the t(15;17) translocation. The patient was then treated with cytosine arabinoside and idarubicin and achieved remission (point 4). Fusion transcripts were undetectable by DzyNA RT-PCR during this remission period. Eleven months later (point 5), the patient was still in remission as assessed by BM morphology and cytogenetics, but our retrospective analysis showed that PML/RAR transcript numbers had begun to increase. Six months later (points 6 and 7), the clinician diagnosed hematologic relapse on the basis of abnormal BM morphology and the presence of the t(15;17) translocation in 25% of metaphases. Analysis of these BM samples by DzyNA RT-PCR showed concentrations of PML/RAR similar to those observed at disease presentation. At this time, ATRA therapy was re-initiated, and the patient successfully achieved a second remission (point 8). The patient received additional therapy with idarubicin and remained in hematologic remission for 3.5 years (points 8–13). During this period, PML/RAR transcripts were either undetectable or detectable but not expressed at quantifiable concentrations. Over the next 18 months, the numbers of PML/RAR transcripts again increased to quantifiable values (point 14) in a BM sample that had normal morphology and no evidence of the t(15;17) translocation by either standard cytogenetics or fluorescence in situ hybridization analysis of 300 cells. Four months later (points 15 and 16), the patient had clinical evidence of a second hematologic relapse. Further treatment with combined ATRA and idarubicin followed by arsenic consolidation therapy achieved a third remission in this patient.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
fusion transcripts, thus providing a highly specific marker for APL. The second assay quantifies BCR transcripts, which allow normalization of data by serving as an internal control for the quality of clinical specimens.
Several experiments were performed to establish the characteristics of DzyNA RT-PCR. Results showed that both the BCR and PML/RAR DzyNA RT-PCR assays were linear over six and five orders of magnitude, respectively. Both assays demonstrated a good limit of detection. The BCR transcript, expressed at low to moderate abundance (unpublished data), was detected in 1 pg of total RNA by the DzyNA RT-PCR assay. The PML/RAR DzyNA assay detected as few as 10 copies of plasmid DNA and could still detect transcripts in an APL patient sample diluted 105-fold with RNA from healthy PBLs. PML/RAR mRNA was detected in 500 ng of NB4 diluted 104-fold, but not 105-fold. This observation suggests that expression is relatively low in the NB4 cell line and agrees with other authors, who estimated that 1 µg of NB4 RNA contains 
104 copies of the fusion transcript (12). Together these observations suggest that DzyNA RT-PCR assays can detect transcripts approaching the theoretical limit of detection and are therefore suitable for clinical studies that include monitoring low transcript numbers associated with minimal residual disease. DzyNA RT-PCR also allowed discrimination between small differences (twofold) in expression of PML/RAR with high precision. Finally, both the PML/RAR and the BCR DzyNA RT-PCR assays were specific for the target transcripts, and the low interassay imprecision, which was comparable to the imprecision for other real-time PCR methodologies (13), demonstrated the high reproducibility of both assays.
In the past, protocols for analysis of PML/RAR in APL have provided only qualitative data and have been labor-intensive and prone to contamination. The standard qualitative RT-PCR procedure involves transcription of RNA to cDNA, followed by two rounds of PCR and gel electrophoresis. In contrast, DzyNA RT-PCR provides a rapid, single-tube quantitative assay with real-time fluorescent detection. The potential for false positives associated with contamination by PCR product is eliminated because all the reaction components are present in a single, sealed vessel. Quantitative protocols for analysis of PML/RAR in APL have only recently been published (5)(6)(7)(12), and initial studies support the hypothesis that quantitative analysis will be more informative than qualitative monitoring of APL. The radiolabelled competitive PCR approach reported by Tobal et al. (6) is well suited to laboratory research, but it is not practical for clinical studies. By comparison, real-time quantitative assays such as TaqMan (14), Beacons (15), or DzyNA RT-PCR (9) are highly suited to clinical testing. The ABI 7700 provides a 96-well format platform fully automated for amplification and data acquisition and is a convenient method for simultaneous analysis of calibrators and patient samples. At present, the TaqMan protocols published for PML/RAR use two steps whereby cDNA synthesis is performed separately before amplification (7)(16). DzyNA RT-PCR is currently the only protocol that allows detection and quantification of PML/RAR transcripts directly from total RNA in a one-step reaction. DzyNA RT-PCR has all of the characteristics necessary to provide a basis for high-throughput analysis of large numbers of clinical specimens as a test for either molecular diagnosis or monitoring of APL or other diseases.
analysis of apl specimens
To date, the majority of APL studies have not included an appropriate internal control for RNA quality. This control is essential because RNA is highly labile, and degradation can produce false-negative results. This problem is exacerbated by the difficulty in ensuring appropriate transport of clinical samples before arrival at the testing laboratory. Quantification of expression of an internal control transcript is a more informative measure of sample integrity than mere qualitative detection. Quantification allows normalization of the data and also allows the parameters to be defined to exclude potential false-negative results. BCR was chosen as the internal control because it fulfills several criteria desirable for control transcripts: it is expressed at concentrations similar to those for the target transcript (at diagnosis) and does not have any known pseudogenes (17). BCR transcripts degrade at a rate similar to that of PML/RAR transcripts over a 24-h period under conditions mimicking transport and storage of patient samples (data not shown). Furthermore, we observed no correlation between BCR and PML/RAR expression (P <0.0005), suggesting that its expression does not change with disease stage.
Total RNA was extracted from BM collected at disease presentation and after treatment for APL. DzyNA RT-PCR detected PML/RAR transcripts in the RNA of 9 of 10 patients collected at disease presentation, thus confirming the clinical diagnosis of APL. The BCR assay demonstrated that the RNA from the remaining patient was degraded and unsuitable for quantification. Experiments performed on different days showed that estimates of the transcript numbers, as well as the ratios of the transcripts (RDCs) were reproducible. The results demonstrated that, in most cases, 50 ng of total RNA is sufficient for relative quantification of BM samples collected at disease presentation. There was complete concordance between results obtained by DzyNA-PCR and those obtained by standard multistep nested qualitative RT-PCR (data not shown). Considerable variation was observed in the relative transcript numbers of PML/RAR in RNA specimens collected from patients at presentation, with values ranging from approximately one third to equal abundance of BCR (RDC, 35–105%). It is unknown whether the concentrations of transcripts at diagnosis are of prognostic significance, and such a correlation is being investigated in ongoing studies.
DzyNA RT-PCR was also used to retrospectively monitor PML/RAR transcript numbers in one patient over a 7-year period (Fig. 3
). The results showed a close correlation between the numbers of PML/RAR transcripts and the clinical stage of disease. Transcript numbers were high at disease presentation and during hematologic relapse and were undetectable or low and nonquantifiable during remission. The results also show that quantification of PML/RAR by DzyNA RT-PCR can predict imminent relapse 4–6 months before morphologic or cytologic symptoms. We suggest that that the presence of PML/RAR in a BM specimen would justify analysis of a second BM aspirate 1 month later. Furthermore, we propose that molecular relapse could be defined as two positive results, where the transcript numbers in the second are higher than those in the first. Because studies indicate significantly better outcomes for patients treated at molecular relapse as opposed to hematologic relapse, the ability to accurately predict relapse could allow clinicians to tailor therapies for individual patients.
DzyNA RT-PCR is now being used to analyze specimens from patients enrolled in a clinical trial being conducted by the Australasian Leukemia and Lymphoma Group. This international multicenter trial is studying 100 individuals, with APL patients receiving combined treatment with ATRA and idarubicin over 3 years. The clinical significance of PML/RAR transcript numbers is being assessed with a view to enable early and reliable prediction of relapse so that clinicians can stratify patients according to risk and tailor therapies accordingly.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
, retinoic acid receptor; ATRA, all-trans-retinoic acid; RT-PCR, reverse transcription-PCR; DNAzyme, DNA enzyme; PBL, peripheral blood lymphocyte; BM, bone marrow; Ct, threshold cycle; Rn, change in fluorescent signal; and RDC, ratio of disease to control. 
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
transcripts with a sensitive competitive RT-PCR method. Leukemia 2001;15:1060-1065.[ISI][Medline]
[Order article via Infotrieve]
mRNA using real-time reverse-transcription polymerase chain reaction. J Mol Diagn 2001;3:141-149.-positive acute promyelocytic leukemia patients: real time quantification of minimal residual disease. A pilot study. Leukemia 2001;15:1696-1700.[ISI][Medline]
[Order article via Infotrieve]
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  T. L. Applegate, H. J. Iland, E. Mokany, and A. V. Todd Molecular Monitoring of Acute Promyelocytic Leukemia by DzyNA Reverse Transcription-PCR Clin. Chem., October 1, 2002; 48(10): 1858 - 1860. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
