| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Hematology |
1 Institute of Hematology and Medical Oncology "L. e A. Seràgnoli", University of Bologna, Via Massarenti No. 9, 40138 Bologna, Italy.
aAuthor for correspondence. Fax 39-051-6364037; e-mail gmartino{at}kaiser.alma.unibo.it.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
Methods: Total mRNAs extracted from 34 AML patients were first analyzed for the presence of juxtamembrane length mutations and tyrosine kinase domain point mutations by a conventional method involving PCR amplification, restriction enzyme digestion, and agarose gel electrophoresis (PCR-RED-AGE). Subsequently, the same patient panel was analyzed by D-HPLC, using specifically designed primers and optimized running temperatures for the length and point mutation analysis.
Results: Thirty-four patients were analyzed by PCR-RED-AGE; 9 were positive for known Flt3 mutations: 6 of 34 (18%) for ITDs in exon 14 and 3 of 34 (9%) for point mutations in exon 20. The same patient panel was analyzed by D-HPLC, and additional nucleotide changes were discovered; in total, 14 sequence variations were identified: 7 of 34 (21%) for ITDs in exon 14; 2 of 34 (6%) for point mutations in exon 20; 1 of 34 (3%) for a new point mutation in exon 16; and 4 of 34 (12%) for polymorphisms in exons 13 and 14. Direct sequencing analysis identified nucleotide alterations in each of the "D-HPLC positives" but in none of the "D-HPLC negatives", yielding a specificity and sensitivity of 100% for D-HPLC-based screening.
Conclusions: This novel D-HPLC-based procedure, which is optimized for identification of new point mutations in the catalytic and regulatory domains of FLT3 receptor, could potentially be useful for studies involving precise genotype determination, which could be critical for selection of innovative AML therapies targeting the FLT3 protein.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
The leukemic blasts of 17–34% of AML patients have small internal tandem duplications (ITDs) in the juxtamembrane (JM) region of the cytoplasmic domain of Flt3 (involving exon 14 and sometimes part of exon 15). Although the locations and lengths of the ITDs vary from sample to sample, the ITDs themselves are always readable in frame (3). In vitro studies have demonstrated that mutant FLT3-ITD receptors are dimerized in a ligand-independent manner, leading the mutant cells to autonomous growth (4). An additional 7% of AML patients exhibit two different kinds of point mutations: Asp835 and Iso836 within the kinase domain of the FLT3 receptor. In these patients, D835/I836 mutant cells are constitutively activated and cause the autonomous proliferation of blasts (as occurs with the mutant ITDs) (5)(6). Recently, a new length mutation in exon 20 of the Flt3 gene (in the A-loop) has been described in two patients with AML (7). The mutation encoded for the insertion of a glycine and a serine between amino acids 840 and 841 of FLT3, which induced constitutive kinase activation.
The traditional approach for diagnostic screening of Flt3 in AML patients involves PCR followed by gel electrophoresis (8)(9). However, this method is time-consuming and requires an additional enzyme digestion step for detection of kinase domain point mutations; furthermore, some very small ITDs go undetected. Denaturing HPLC (D-HPLC), initially described by Oefner and Underhill in 1995 (10), has potential advantages over gel electrophoresis because it is generally more sensitive and allows a high throughput (11). We therefore developed a method for screening of Flt3 mutations with PCR plus D-HPLC and investigated its efficacy in a series of AML patients. Our concerns were related to the determination of novel genetic variations in the regulatory and catalytic domains of FLT3 protein. The characterization of new mutations could lead to a better understanding of the complex mechanism underlying drug efficacy and drug resistance in the treatment of AML.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
pcr amplification, restriction enzyme digestion, gel electrophoresis
For the PCR amplification–restriction enzyme digestion–agarose gel electrophoresis (PCR-RED-AGE) procedure, we directly amplified 1 µL of reverse transcription-PCR product by PCR using 30 pmoles of each primer in a multiplex reaction with the following set of primers: 14F (5'-tgtcgagcagtactctaaaca-3') and 15R (5'-atcctagtaccttcccaaactc-3'), which amplify the JM domain of the gene; and 20F (5'-ccgccaggaacgtgcttg-3') and 20R (5'-gcagacgggcattgcccc-3'), which amplify the region containing the D835 and I836 mutations. The primers were custom synthesized by Invitrogen Life Technology. Each PCR was performed in a 50-µL reaction volume comprising 1x Reaction Buffer, 1.5 mM MgCl2, 200 µM each of the deoxynucleotide triphosphates, and 1.5 U of AmpliTaq Gold (Applied Biosystems). Denaturing, annealing, and extension steps were performed at 95 °C for 30 s, 56 °C for 45 s, and 72 °C for 30 s, respectively, for 40 cycles on a MiniCycler PCR system; the PCR also included an initial 10-min denaturation step at 95 °C and a final extension step at 72 °C for 10 min.
The amplified product was digested with EcoRV (5 U) for 1 h at 37 °C in a 20-µL total reaction volume containing 0.1 g/L bovine serum albumin in 1x Buffer D (Promega). After enzyme digestion, 20 µL of amplified PCR products was electrophoresed on a 3% agarose gel, stained with ethidium bromide, and photographed under ultraviolet light. Wild-type samples for the JM domain yielded a single 366-bp product, whereas samples with ITDs yielded an extra band of larger size. In addition, samples displaying a completely digested product (with two smaller bands of 68 and 46 bp) corresponded to the wild-type form for this fragment, whereas the presence of a partially digested product (with a larger band of 114 bp as well as the two smaller bands) indicated that a point mutation abolished the EcoRV recognition sites.
experimental design of d-hplc analysis
All samples, either positive or negative by the PCR-RED-AGE method for both types of mutations, were studied using the WAVETM DNA Fragment Analysis System (Transgenomic Ltd.) for D-HPLC.
The genomic fragment to be studied (total length, 1485 bases; from exon 11 to exon 24; GenBank accession no. NM_004119) has been divided into four subfragments of optimal length for D-HPLC analysis (PCR products shorter than 150 bp have too similar melting points; the recommended length is 200–500 bp); fragments were PCR-amplified using a new set of primers designed at the Primer3 web site (http://www-genome.wi.mit.edu/cgi-bin/primer/primer3_www.cgi). Selected primers were as follow:
PCR was performed in a 50-µL reaction volume comprising 1x Reaction Buffer, 1.5 mM MgCl2, 200 µM each of the deoxynucleotide triphosphates, and 1.5 U of AmpliTaq Gold (Applied Biosystems), using the primers above specified. The touchdown PCR protocol was used, and cycling conditions were as follow: denaturation step at 95 °C for 10 min followed by 16 touchdown cycles involving denaturation at 95 °C for 30 s, annealing for 60 s, and primer extension at 72 °C for 75 s, with a 0.5 °C decrease in annealing temperature per cycle starting at 64.3 °C for F491 and F471 or 66.3 °C for F465 and F402. An additional 24 cycles were performed at the annealing temperature specific for each pair of primers (53.5 °C for F491, 54.5 °C for F471 and F402, and 55.5 °C for F465). A final hybridization step was performed starting at 95 °C and reducing by 1.5 °C/min to 25 °C.
Using WAVEmaker software, Ver. 4.1.40 (Transgenomic), we calculated the melting curves to select optimal temperatures for study of the various fragments of FLT3 receptor protein. For detection of point mutations, we tried a set of temperatures within a range spanning 55.5–59.8 °C. We loaded 8-µL aliquots of crude PCR samples (preheated for 10 min at 96 °C and then gradually reannealed for 10 min at room temperature) on a preheated C18 reversed-phase column at the following selected temperatures: F491 at 55.5, 57.6, and 58.7 °C; F471 at 55.8, 57.5, and 59.8 °C; F465 at 56.6 and 57.4 °C; F402 at 56.2, 57.8, and 58.2 °C. DNA was eluted from the column by a linear acetonitrile gradient in 0.1 mmol/L triethylamine acetate buffer (TEAA; Transgenomic) at a constant flow rate. The gradient was formed by mixing buffer A (0.1 mmol/L TEAA) and buffer B (0.1 mmol/L TEAA containing 250 mL/L acetonitrile). Eluted DNA was detected by the absorbance at 260 nm, and each sample run took 8 min.
For detection of ITDs, after an initial denaturation step (96 °C for 10 min), we loaded 8 µL of PCR product into the autosampler of the automated D-HPLC system, under nondenaturing conditions (at 50 °C). The use of D-HPLC for the detection of ITDs is based on the variations in length of ITD-containing PCR products. To ensure that homozygous mutations could not escape D-HPLC detection, for all samples to be studied we prepared a mixture of PCR products from wild-type control and patient DNA.
direct dna sequencing
Samples exhibiting an abnormal PCR-RED-AGE or D-HPLC profile were confirmed by direct sequencing using ABI PRISM 377 DNA sequencing (Applied Biosystems). In these cases, the amplified band [for tyrosine kinase domain (TKD) mutations] or the ITD band (for JM length mutations) was purified from the agarose gel, and 50 ng of precipitated DNA was sequenced using the Big Dye Terminator Cycle Sequencing Kit (Applied Biosystems). In addition, crude PCR products from patients scored as "negative" were also directly sequenced to exclude the presence of undetected genetic variations. Sequences were compared with the wild-type sequence [GenBank accession no. NM_004119 (mRNA)].
|
Results and Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
). The optimized conditions presented in this work validate the possibility that D-HPLC could readily differentiate mutants from the wild type for both known and unknown nucleotide variations.
|
We first used a classic multiplex-PCR-based method. Using two different set of primers (14F/15R and 20F/20R; see Materials and Methods), we screened a panel of 34 AML patients based on fragments amplified by a multiplex-PCR procedure. We then performed a restriction enzyme digestion of 20 µL of PCR product using EcoRV. Finally, each PCR-digested product was electrophoresed in a 3% agarose gel. Fig. 2
shows representative examples of band patterns commonly obtained by multiplex-PCR in the presence of an ITD (patient A) or a TKD (patient B) mutations, alongside the wild-type profile (control lane). PCR-RED-AGE identified 9 sequence variations: 6 of 34 (18%) for an ITD in exon 14 and 3 of 34 (9%) for point mutations in exon 20.
|
Taking advantage of our knowledge of the Flt3 gene and its aberrations, we defined the primers and temperatures for D-HPLC mutational analysis. The new set of primers (see Materials and Methods) yielded fragments of optimal length to be used in D-HPLC mutational analysis. The entire coding sequence, from exon 11 to exon 24, was screened in 34 patients by D-HPLC. Having established the DNA fragment sequences to be studied, we determined the melting temperatures to be used during D-HPLC screening (see Materials and Methods for details). WAVEmaker software (Ver. 4.1.40) was able to predict melting temperatures to maximize the resolution of heteroduplex and homoduplex peaks and visualize the position of melting subdomains with respect the whole sequence fragment. Mutations previously detected by PCR-RED-AGE were used as positive controls to validate the melting domains elicited for blinded mutational analysis. One wild-type sample was used as a negative control; the chromatogram from each tested patient was overlaid with the wild-type profile, and samples with an extra peak were scored as positive. Fig. 3A
(left panel) shows a representative D-HPLC chromatogram for fragment F_402; the elution profile for the 402-bp PCR product showed a single peak at 4.2 min (blue trace) corresponding to the wild-type sequence. By comparison, the mutant PCR product showed a double peak (gray trace): the first, at 4.1 min, corresponding to the heteroduplex fraction, which has a shorter retention time; the second, at 4.2 min, corresponding to the homoduplex product (Fig. 3B
).
|
For identification of ITDs, the PCR products were also studied by D-HPLC under nondenaturing conditions at 50 °C; this temperature enables specific screening for this mutation because of the difference in length between the wild-type and mutant alleles. Each of the PCR products with ITDs produced a distinct profile with respect to the control wild-type sequence (Fig. 4B
). ITDs produced three major peaks: one (named wtHM) at 14.69 min, corresponding to the wild-type homoduplex; a second (named itdHM) at 15.55 min, corresponding to the ITD homoduplex, whose retention time is longer because of its larger size; and a third (named HT) at 13.76 min, representing an artifact heteroduplex complex generated during the initial denaturation step at 96 °C (see Materials and Methods). To assess the sensitivity of the D-HPLC method, which is variable from one study to other, we also performed dilution experiments using known quantities of amplified wild-type and mutant Flt3 PCR products (ratio wild-type/mutant >1:1, 10:1, 50:1, and 100:1) for either ITD or point mutations; the results indicated that D-HPLC could detect the minority DNA species present in concentrations as low as 1–5% (data not shown).
|
By D-HPLC we identified 14 sequence variations [7 of 34 (21%) with an ITD in exon 14; 2 of 34 (6%) with point mutations in exon 20; 1 of 34 (3%) with a new point mutation in exon 16; and 4 of 34 (12%) with polymorphisms in exons 13 and 14], including all those identified by PCR-RED-AGE (see Tables 1
and 2
for details). Among the 3 of 34 patients who appeared to be positive for Flt3 point mutations by the PCR-RED-AGE method, only 2 were positive by D-HPLC (see Table 1
, patient 19). Direct sequencing confirmed the validity of the D-HPLC analysis. The false positive found by gel electrophoresis can be attributed to incomplete enzyme digestion. All seven samples that tested positive for Flt3 ITD mutations by D-HPLC were confirmed for the presence of a specific aberration by direct PCR product sequencing (Fig. 4A
; representative ITD sequence). This result suggests that D-HPLC analysis of length mutations offers the advantage of higher sensitivity, especially for short insertions, because one patient who was scored negative by PCR-gel electrophoresis was positive by D-HPLC (Table 1
, patient 29). Five unknown nucleotide changes were identified by D-HPLC; direct sequencing analysis identified nucleotide alterations in each of the "D-HPLC positives" but none of the "D-HPLC negatives", yielding a specificity and sensitivity of 100% for D-HPLC-based screening of FLT3 receptor.
|
|
Four patients exhibited silent nucleotide changes (Table 1
): three in exon 13 (nt 1740) and one in exon 14 (nt 1899). The pathogenic significance of these silent mutations is unclear, although it may be speculated that single base-pair substitutions can cause significant structural alterations in mRNA or introduce cryptic splice sites that may potentially alter the final transcript. Notably, for one patient (Table 1
, patient 15), we detected a nucleotide change (C>T at nt 2096; GenBank accession no. NM_004119) in exon 16 of the Flt3 gene, corresponding to the first catalytic domain in the FLT3 receptor protein, which produced the amino acid change Ala680Val. Molecular characterization of this new point mutation is ongoing; although two tyrosine kinase proteins (ABL and IRK) naturally present this amino acid substitution, the amino acid sequence NH2-NLLGACT-COOH in which we found the mutation, and it is extremely conserved with respect to other known tyrosine kinase receptors, such as vascular endothelial growth factor receptor, c-KIT, and platelet-derived growth factor receptor (BLASTP analysis; data not shown). Moreover, RNA from four donors was analyzed by D-HPLC to screen for the presence of the new polymorphisms described above for AML patients; although a larger donor panel should be studied to rule out the possibility that these sequence variations could also be present in healthy blood cells, we were unable to detect such variations in our small control group.
Because it is broadly accepted that point or length mutations near the ATP binding site can abolish the therapeutic activity of tyrosine kinase inhibitors, further studies are necessary to determine the chemotherapeutic consequences of this type of mutation. Moreover, it should be noted that we could not identify any commercially available restriction enzyme that cleaved at the position of the mutated nucleotide in either the wild-type or mutant sequence; therefore, although restriction enzyme digestion has played and continues to play a major role in analyzing the genetic changes in cancer, the application of D-HPLC may enhance the screening in cancer genetics.
To the best of our knowledge, this is the first description of the use of an innovative technique, D-HPLC, for genotyping mutational events in the Flt3 gene. D-HPLC is a reversed-phase ion-pair HPLC method that is very sensitive for detection of DNA sequence variation. With respect to conventional PCR analysis, the method described provides a more rapid (PCR product is loaded directly on the D-HPLC column, and up to 196 samples could be screened in a single analytical run), sensitive, and specific automated system for analysis of PCR-generated products. Detection is based on differences in the retention of perfectly matched homoduplexes and heteroduplexes containing one mismatched base pair. More importantly, this method has been used for identification of new point mutations occurring in the catalytic and regulatory domains of FLT3 protein. Obviously, this scanning method requires direct DNA sequencing to be performed to determine the precise sequence abnormality.
To date, Genescan analysis has played a major role in analyzing the mutational events in the Flt3 gene; this procedure also allows high sample throughput with acceptable sensitivity. In fact, using the Genescan method, Thiede et al. (12) defined a mutant/wild-type ratio, in 121 FLT3-ITD+ patients analyzed, that would represent an independent prognostic factor for worse overall and disease-free survival. However, it should be noted that the D-HPLC software quantification tool also allows accurate measurements of the peak height and area, which are indicative of the dosage; we therefore can replace gel electrophoresis quantification of fluorescently labeled products.
In recent years, a series of studies supported the idea that FLT3 is potentially a good pharmacologic target in AML and that therapeutic efficacy and tolerability could strongly depend on the presence of a specific genetic variation in the Flt3 gene (13)(14)(15). Mutational D-HPLC profiles provide a "fingerprint" to reveal the presence of sequence alterations. Future therapeutic approaches (16) may involve the design of novel molecules targeting the underlying molecular defects of the FLT3 receptor protein: precise genotype determination would be critical for the selection of such therapies (17)(18)(19). The D-HPLC method described here could be of use for such studies, but taking into account the costs for this technology, we recommend this application for large-scale mutation screening to prospectively identify subsets of patients for whose tumors the importance of the kinase is established.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  R. W. Stam, M. L. den Boer, P. Schneider, M. Meier, H. B. Beverloo, and R. Pieters D-HPLC analysis of the entire FLT3 gene in MLL rearranged and hyperdiploid acute lymphoblastic leukemia Haematologica, November 1, 2007; 92(11): 1565 - 1568. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Palmisano, T. Grafone, E. Ottaviani, N. Testoni, M. Baccarani, and G. Martinelli NPM1 mutations are more stable than FLT3 mutations during the course of disease in patients with acute myeloid leukemia Haematologica, September 1, 2007; 92(9): 1268 - 1269. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Pfeifer, B. Wassmann, A. Pavlova, L. Wunderle, J. Oldenburg, A. Binckebanck, T. Lange, A. Hochhaus, S. Wystub, P. Bruck, et al. Kinase domain mutations of BCR-ABL frequently precede imatinib-based therapy and give rise to relapse in patients with de novo Philadelphia-positive acute lymphoblastic leukemia (Ph+ ALL) Blood, July 15, 2007; 110(2): 727 - 734. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
