| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
1
Department of Medical Sciences and
2
Department of Genetics and Pathology, Rudbeck Laboratory, Uppsala University, 75185 Uppsala, Sweden.
a Author for correspondence. Fax 46-18-553601; e-mail gisela.barbany{at}medsci.uu.se
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: We used an oligo(dT)-coated manifold support to capture the mRNA directly from the cell lysate. After reverse transcription, the cDNA was eluted from the manifold support, and BCR-ABL and GAPDH mRNAs were quantified in real time using the TaqMan fluorogenic detection system.
Results: The detection limit of the method was one positive K562 cell among 105 negative cells. GAPDH was chosen as a reference gene based on the low variation between samples from different stages of the disease and the low signal in the absence of reverse transcription. The day-to-day variation of the method (CV) was 32%. In 43 blood samples from 13 CML patients, mRNA quantification agreed well with cytogenetic data.
Conclusions: The proposed procedure constitutes a reproducible and sensitive BCR-ABL mRNA quantification method and is suitable to monitor minimal residual disease in CML patients.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The advent of methods to monitor DNA amplification reactions in real time, such as the 5'-nuclease assay (9), has made mRNA quantification by RT-PCR simpler and more accurate. The 5'-nuclease assay takes advantage of the 5'-nuclease activity of Taq DNA polymerase to cleave a dual-labeled probe hybridizing to the amplified fragment during the extension phase (10). The cleavage reaction separates the two fluorophores, abolishing fluorescence resonance energy transfer, and producing increased fluorescence. The increase in fluorescence is proportional to the target accumulation and can be measured in real time (11). The fractional amplification cycle at which fluorescence exceeds baseline fluorescence is called the threshold cycle (CT). The CT value is recorded as a measure of the number of target molecules in the amplification reaction.
RNA typically is isolated through extraction with acid guanidinium-phenol-chloroform (12), a method that is relatively labor-intensive and requires the use of hazardous chemicals. Simultaneous processing of numerous samples is difficult and time-consuming, limiting the applicability of RT-PCR in routine laboratories. We recently developed an oligo(dT)-coated manifold support that allows mRNA to be isolated directly from cell lysates by hybridization followed by transfer through the different washing and enzymatic steps of the assay with minimal pipetting (13).
In the present report, we describe a streamlined and rapid method to quantify BCR-ABL transcripts, where the mRNA in cell lysates is captured onto a manifold solid support and quantified in real time using the TaqMan fluorogenic detection system. The whole procedure can be completed in a few hours. We used the method to follow BCR-ABL transcript expression in peripheral blood from 13 CML patients, comparing the results to cytogenetic data from bone marrow smears.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). Patients 1 and 2 received autologous stem cell
transplantation (ASCT) during chronic phase, and the remaining patients
received allogeneic bone marrow transplantation (BMT) during chronic
phase.
|
After informed consent, peripheral blood was collected from CML patients in EDTA tubes, and mononuclear cells were separated by Ficoll density gradient centrifugation. The cells were washed in phosphate-buffered saline and counted. Viability was assessed by trypan blue exclusion. Where indicated, erythrocytes were removed by lysis with NH4Cl, and the remaining cells were washed in phosphate-buffered saline before nucleated cells were lysed. Leukapheresis samples were collected from patients who had undergone ASCT, and were stored frozen. Cells were washed once in phosphate-buffered saline before lysis in extraction buffer.
The cell line K562, carrying the BCR-ABL translocation, was used as a positive control, and the lymphoblastoid cell line BSM was used as a negative control.
preparation of the oligo(dT)-coated manifold supports
The procedure has been described in detail elsewhere
(13). The manifold supports, shaped so that individual
prongs fit the wells of a microtiter plate, were sonicated in 950
mL/L ethanol. Oligo(dT)-cellulose (Amersham Pharmacia Biotech)
was rinsed repeatedly in triethylamine before being mixed with
triethylamine to obtain a slurry. The manifold supports were immersed
for 2 s in the slurry, washed once in ethanol and once in water,
allowed to air dry, and stored at 4 °C until used.
rna isolation and cDNA synthesis
Cells were counted and lysed in extraction buffer containing 100
mmol/L Tris-HCl (pH 7.9), 10 mmol/L EDTA, 5 mmol/L
dithiothreitol, 500 mmol/L lithium chloride, and 10 g/L
lithium dodecyl sulfate to a density of 107
cells/mL. Viscosity was reduced by repeated passage through a needle
(0.7 x 50 mm). The manifold supports were presoaked in extraction
buffer for 5 min and submerged in 50 µL of cell lysate. mRNA was
captured by hybridization to the supports on a shaking platform for 30
min at room temperature. The supports were subsequently washed eight
times in 80 µL of 10 mmol/L Tris-HCl (pH 7.5), 0.1 mol/L NaCl, and 1
mmol/L EDTA; transferred to fresh microtiter wells containing 50 µL
of 50 mmol/L Tris-HCl (pH 8.3), 75 mmol/L KCl, 3 mmol/L
MgCl2, 10 mmol/L dithiothreitol, 0.5 mmol/L
dNTPs, 0.25 g/L bovine serum albumin, 2.5 µmol/L random DNA
hexamers (Pharmacia Biotech), 25 units of
HPRITM ribonuclease inhibitor (Amersham Life
Science), and 200 U of M-MLV reverse transcriptase (Amersham Life
Science); and incubated at 37 °C for 1 h. The cDNA was eluted
from the manifold support by denaturation at 95 °C for 5 min in 50
µL of water and amplified immediately or stored at -20 °C.
primers and probes
Primers and 5'-nuclease probes were designed using Primer Express
(PE Biosystems) software (Table 2
). Two different forward primers were selected for the
BCR-ABL transcript, one located in exon b2 and one located
in exon b3 of the BCR gene. The 5'-nuclease probe was
located in exon 2 of the ABL gene. The reference
ß-ACTIN cDNA was amplified with intron-spanning
primers and the probe from the TaqMan ß-ACTIN
control reagent kit (PE Biosystems). Glyceraldehyde-3-phosphate
dehydrogenase (GAPDH) cDNA was amplified with the TaqMan
GAPDH control reagents kit (PE Biosystems).
|
pcr amplification
cDNA (10 µL) was transferred to an optical microtiter well
containing 25 µL of buffer A (PE Biosystems); 2 mmol/L
MgCl2; 0.2 mmol/L dGTP, dATP, and dCTP; 0.4
mmol/L dUTP; 0.5 µmol/L each amplification primer for
BCR-ABL; 0.1 µmol/L TaqMan probe; 1.25 U of
AmpliTaq GoldTM (PE Biosystems); and 0.25 U of
Amperase UNGTM (PE Biosystems). The reactions
were incubated at 50 °C for 2 min and 95 °C for 10 min,
and cycled 45 times between 95 °C for 15 s and 60 °C for 1
min in the ABI Prism 7700 (PE Biosystems). For the reference genes, PCR
amplification was started with 5 µL of cDNA.
construction of the plasmid calibrators
cDNA was prepared from a patient with a b2a2 translocation.
BCR-ABL and GAPDH sequences were amplified with
the described PCR primers. The amplification products were cloned into
the pCRII vector (Invitrogen) and sequenced. Known amounts of each
plasmid were linearized with HindIII and EcoRV,
respectively. Serial dilutions (10-fold) representing 5 to 5 x
105 copies of pCRBCR-ABL and 4 to 4 x
105 copies of pCRGAPDH were prepared from the
cleaved plasmids, aliquoted, and stored at -20 °C.
measurement of bcr-abl mRNA in patient samples
Serial dilutions from 4 to 4 x 105
copies of pCRGAPDH and 5 to 5 x 105 copies
of pCRb2a2 were amplified in duplicate. A calibration curve was
derived by plotting the CT values obtained
for each dilution against the logarithm of the plasmid copy number. The
mean slopes of the calibration curves for the two genes were -3.42 for
GAPDH and -3.6 for BCR-ABL, with interassay CVs
of 6% and 4%, respectively. The calibration curves show a strong
linear correlation, with correlation coefficients
(r2) between 0.97 and 0.997.
BCR-ABL and GAPDH sequences were amplified in duplicate from the patient samples, and the copy numbers of both genes were calculated with help of the respective calibration curves. The estimated amount of BCR-ABL mRNA was normalized by dividing by the amount of GAPDH mRNA to compensate for variations in quantity or quality of starting mRNA as well as for differences in reverse transcriptase efficiency. The normalized values were multiplied by the constant 104.
Because sample size and quality vary between individual samples, interpreting negative or weakly positive results may be difficult. We thus calculated the absolute limit of detection of BCR-ABL mRNA for each individual sample. The limit of detection was calculated by dividing the theoretically lowest number of BCR-ABL molecules detectable by PCR, i.e., one molecule, by the number of GAPDH molecules detected in that particular sample, and multiplying by 104.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
reproducibility of sample preparation
To determine the variability in binding capacity of the manifold
supports and in the sample preparation procedure, mRNA from cell
lysates was captured on individual prongs of different manifold
supports. RNA from lysates corresponding to 2 x
104 and 2 x 105 K562
cells was immobilized on 6 and 10 individual prongs, respectively.
After reverse transcription, the cDNA was amplified with primers and
probes specific for GAPDH. Fig. 1
shows the individual CT values and
the average of the CT values for the different
RNA preparations. The SD of the CT value was 0.6
and 0.7 cycles for samples containing 2 x
104 and 2 x 105
cells, respectively.
|
sample handling
Successful amplification of RNA from blood samples required the
elimination of the erythrocytes before cell lysis (data not shown).
Erythrocytes were removed either by Ficoll separation or by hypotonic
lysis with NH4Cl, and the remaining cells were
counted in an hemocytometer before being lysed for RNA capture.
Isolation of leukocytes from whole blood by Ficoll gradient
centrifugation consistently gave a lower CT value
than red cell lysis with hypotonic NH4Cl (Fig. 2
). In additional experiments, mononuclear cells were
purified from blood by Ficoll separation before mRNA isolation on the
manifold supports.
|
sensitivity and specificity of the assay
To estimate the sensitivity of the method, K562 cells were mixed
with variable numbers of cells from the BCR-ABL-negative
lymphoblastoid cell line BSM to give ratios ranging from 1:1 to
1:106. mRNA corresponding to 500 000 cells of
the different dilutions was bound to the manifold supports and
processed through the different steps of the assay. In 50% of the
cases, BCR-ABL mRNA could be quantified in cDNA prepared
from samples containing 1 K562 cell in 105 BSM
cells. In the remaining 50% of the cases, only one of the duplicate
samples of this dilution was positive for the dilution containing 1
K562 cell in 105 BSM cells. Both primer
combinations (b2a2 and b3a2) gave the same detection limit (1 K562 cell
in 105 BSM cells), and the slopes of the curves
were very similar (Fig. 3
). These results indicate that it is possible to use one
primer/probe set to amplify either fusion variant.
|
No BCR-ABL products were obtained from cDNA prepared from BSM cells or mononuclear cells from healthy individuals or when reverse transcription was omitted.
choice of reference gene
Three reference genes were investigated,
ß-ACTIN, GAPDH, and BCR
in samples from 12 different individuals and two cell lines to
determine which gene showed the smallest variability among samples and
where genomic DNA contributed the least to the signal. All samples were
processed in duplicate, and the reverse transcription enzyme was
omitted in every second sample. All three genes could be amplified in
the absence of the reverse transcription reaction. The
CT values were on average 11 cycles higher in the
case of ß-ACTIN and GAPDH and 8
cycles in the case of BCR, compared with the corresponding
samples treated with reverse transcriptase (Fig. 4
). Variations in reference gene expression among different
individuals were very similar for the three genes tested, as shown by
the similar SD values. In subsequent experiments, GAPDH was
used as a reference gene.
|
day-to-day imprecision
The overall reproducibility of the quantification assay was
estimated by analyzing six different samples, starting from the RNA
isolation, on four different occasions. The expression of
BCR-ABL and GAPDH mRNAs was calculated by
comparison with their respective calibration curves, and the normalized
BCR-ABL values were calculated. Fig. 5
shows the means and SDs for the different samples. The mean CV
for the BCR-ABL/GAPDH ratio was 32% (range,
20–44%).
|
analysis of transcript expression in patient samples
Lysates (50 µL) from leukocytes obtained from patient blood
samples were incubated with the manifold supports and processed as
described in Materials and Methods. Fig. 6
shows the BCR-ABL mRNA expression normalized to
GAPDH mRNA expression for two different CML patients who
received 
-interferon (-IFN) + hydroxyurea (HU) followed by
ASCT. BCR-ABL mRNA expression in patient 1 decreased from 22
arbitrary units to below the detection limit after ASCT. The patient
remained PCR-negative for 1 year after ASCT, with one transitional
PCR-positive analysis in the 6 months follow-up sample. In an analysis
90 months after ASCT, BCR-ABL expression had increased to
~25 arbitrary units, although the patient remained in clinical and
cytogenetic remission. Patient 2 did not respond clinically to ASCT.
The increases in both normalized BCR-ABL mRNA expression and
numbers of Philadelphia-positive cells also reflect this fact.
|
The rest of the patients received allogeneic BMT, patients 3–6 from an
unrelated donor and patients 7–13 from a relative. In this group of
patients, a pretransplantation sample was available for patients 4, 5,
and 6. The normalized BCR-ABL mRNA expression at diagnosis
was on average 805 arbitrary units (range, 230–3200; Fig. 7
). After BMT, BCR-ABL mRNA expression dropped
dramatically, between 1000- and 10 000-fold, over a period of 3–6
months and became undetectable in patient 5, 3 months after BMT (Fig. 7
), and in patient 6, 9 months after BMT (Fig. 7
). In three additional
patients analyzed 30–48 months after BMT, the normalized
BCR-ABL mRNA expression was below the detection limit (data
not shown). After allogeneic BMT (6–12 months), BCR-ABL
mRNA expression was always below 5 arbitrary units. The patients
receiving allogeneic BMT remained in hematological and cytogenetic
remission during the period studied, except patients 7 and 10. Both
patients were transiently Philadelphia chromosome positive (1 cell in
50) in a bone marrow sample 12 and 24 months after BMT,
respectively.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We propose a streamlined RNA isolation method where poly(A)+ mRNA is isolated directly from the cell lysate by binding to an oligo(dT)-coated manifold support. Manifold supports are devices composed of sets of prongs that project into a corresponding set of reaction wells. The surfaces of the prongs may be modified to allow the binding of biomolecules so that these can be processed in a set of reactions and loaded on detection instruments.
The use of manifold supports in molecular diagnostics has been useful for genotyping by the oligonucleotide ligation assay or minisequencing, and for the detection of mutations by sequencing [reviewed in Ref. (14)]. Because the RNA binds to a solid support, multiple samples can be manipulated in parallel with minimal effort. Moreover, the reduction in the number of sample processing and pipetting steps diminishes the risk for uncontrolled variables associated with the quality of the RNA. Currently, instability of RNA limits the use of RT-PCR to research laboratories with trained personnel. The use of manifold supports greatly simplifies mRNA isolation, making RT-PCR-based analysis more accessible to routine laboratories. The method can also be adapted to other oligo(dT)-coated solid supports, such as paramagnetic beads. We found that 25 µL of Dynabeads (Dynal) had a binding capacity similar to that of individual prongs in our manifold support (data not shown).
Both ß-ACTIN and GAPDH are known to have pseudogenes that can give rise to a PCR product that interferes with the quantification of these RNAs. The presence of genomic DNA in the RNA preparation thus represents a potential source of error when quantifying the reference gene. The difference in CT values (11 cycles) that we found between the samples amplified with and without reverse transcriptase shows that genomic DNA contamination of the mRNA bound to the manifold support is negligible.
We found that a single primer/probe set was adequate to amplify the two most common BCR-ABL fusion variants, which simplifies the assay and eliminates the need to analyze the fusion variant of individual patients. This result is in agreement with two recent reports where all fusion variants of BCR-ABL could be quantified using a single set of primers (15)(16).
We also examined how sample preparation affected the sensitivity of the assay. It was necessary to remove erythrocytes to obtain a positive PCR, probably because of the known inhibitory effect of hemoglobin (17). Peripheral blood mononuclear cell purification by Ficoll density separation gave lower CT values for both BCR-ABL and GAPDH mRNAs, presumably because of better RNA quality.
Because clinical specimens are collected remote from the site of analysis, differences in time from sample collection to RNA isolation may lead to variable degradation of RNA in the sample, with variability in assay sensitivity as a consequence. We calculated the absolute detection limit of every sample, defined as one single copy of BCR-ABL divided by the number of copies of GAPDH for a particular sample, multiplied by the constant 104. This value should constitute a valuable help in interpreting results by reflecting the size and quality of the sample. The absence of BCR-ABL amplification can only be interpreted as a true negative when amplification of the reference gene reveals a positive result. In addition, positive samples where the normalized BCR-ABL mRNA expression falls below the detection limit are to be interpreted with caution.
The interassay imprecision (CV) was low (4–6%) for calibrators but high for patient samples (mean, 32%). This implies that the precision of the present method is insufficient to analyze small differences in BCR-ABL mRNA expression between samples.
The normalized BCR-ABL expression measurements from 13 CML patients correlated well with clinical and cytogenetic data. Recently, several groups have reported analysis of CML patients using real-time RT-PCR (15)(18)(19)(20) and found that real-time PCR was suitable for following the kinetics of BCR-ABL mRNA in CML. The sensitivity of our method is comparable to that of Preudhomme et al. (20) and Eder et al. (19), indicating that the manifold support has sufficient capacity to monitor minimal residual disease, where only a small minority of the cells are positive for BCR-ABL fusion mRNA.
Typically, the normalized BCR-ABL values for untreated
patients were between 250 and 1000 arbitrary units. Patients that show
a partial response (-IFN + HU and ASCT) had BCR-ABL
values between 100 and 10 arbitrary units. After allogeneic BMT,
BCR-ABL mRNA fell dramatically to <5 arbitrary units.
Additional studies will be necessary to determine what threshold
BCR-ABL mRNA expression is critical and predictive of
relapse.
In conclusion, the combination of a manifold support to isolate mRNA in a fast and reproducible manner, together with the advantages of real-time RT-PCR is promising as a sensitive and accurate method to study the kinetics of BCR-ABL mRNA expression, and is suitable to monitor minimal residual disease in CML patients.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
-IFN, -interferon; and HU, hydroxyurea. 
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  T. Nakanishi, K. Shiozawa, B. A. Hassel, and D. D. Ross Complex interaction of BCRP/ABCG2 and imatinib in BCR-ABL-expressing cells: BCRP-mediated resistance to imatinib is attenuated by imatinib-induced reduction of BCRP expression Blood, July 15, 2006; 108(2): 678 - 684. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. L. Nashed, K. W. Rao, and M. L. Gulley Clinical Applications of BCR-ABL Molecular Testing in Acute Leukemia J. Mol. Diagn., May 1, 2003; 5(2): 63 - 72. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Barbany, M. Hoglund, B. Simonsson, and the Swedish CML Group Complete Molecular Remission in Chronic Myelogenous Leukemia after Imatinib Therapy N. Engl. J. Med., August 15, 2002; 347(7): 539 - 540. [Full Text] [PDF]
|
|
|
|
 |
|
 H. Chisaka, E. Morita, K. Murata, N. Ishii, N. Yaegashi, K. Okamura, and K. Sugamura A transgenic mouse model for non-immune hydrops fetalis induced by the NS1 gene of human parvovirus B19 J. Gen. Virol., February 1, 2002; 83(2): 273 - 281. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Bagg Chronic Myeloid Leukemia: A Minimalistic View of Post-Therapeutic Monitoring J. Mol. Diagn., February 1, 2002; 4(1): 1 - 10. [Full Text] [PDF]
|
|
|
|
 |
|
 T. Masui, R. Hosotani, S. Tsuji, Y. Miyamoto, S. Yasuda, J. Ida, S. Nakajima, M. Kawaguchi, H. Kobayashi, M. Koizumi, et al. Expression of METH-1 and METH-2 in Pancreatic Cancer Clin. Cancer Res., November 1, 2001; 7(11): 3437 - 3443. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. Foy and H. C. Parkes Emerging Homogeneous DNA-based Technologies in the Clinical Laboratory Clin. Chem., June 1, 2001; 47(6): 990 - 1000. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, normalized
BCR-ABL expression; 
, absolute detection limit for a
particular sample. Results from the cytogenetic analysis are indicated
as percentage of Philadelphia-positive (Ph+) cells in
bone marrow smears. Ph-, no Philadelphia-positive
metaphases.