| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Technical Advances |
From the Clinica di Oncoematologia Pediatrica, Azienda Ospedaliera-Università di Padova, Padova, Italy
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, and C
, and one
for the joining region (JH). We first studied seven BL cell lines and
optimized LD-PCR reaction for analysis of tumor specimens. Five of
seven cell lines were positive for the t(8;14), whereas two
lines derived from endemic BL were negative, as expected. Of 15
biopsies obtained from pediatric BL and subsequently analyzed,
13 (87%) were positive for the translocation detected by LD-PCR and
showed a product ranging in size from 2.0 to 9.5 kb. Cµ region was
involved in 6 cases, C and C in 2 cases each, and
JH in 3 cases. Interestingly, 2 of the tumors positive for JH
showed a second, larger PCR product with the C- and
C-specific primer, respectively. We established that our
LD-PCR method could detect 10-3 BL cells within a
population of hematopoietic cells lacking the translocation. In
conclusion, our LD-PCR method represents a fast, highly
sensitive, and specific tool to study sporadic BL and to detect
minimal disease and residual disease in patients affected by
t(8;14)-positive lymphomas.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Molecular analysis of the genomic sequences involved in this
translocation has demonstrated that the breakpoints occur in different
regions within both the MYC and the IgH genes and give rise to
tumor-specific MYC/IgH rearrangements. On chromosome 8, breakpoints are
usually located upstream of the MYC gene in the endemic (African) BL
and within exon 1 or intron 1 in the sporadic (Caucasian)
BL.5
The breakpoints within the IgH locus are distributed
over a region of at least 100 kb, but they are preferentially located
in the Ig joining region (JH) or in the switch regions (Sµ or S)
in the endemic and in the sporadic BL, respectively.6,7
The MYC/IgH rearrangement is detectable only at the genomic DNA level, because no fusion transcript originates from the chromosomal translocation. Differently, the relevant molecular event generated by the t(8;14) is the deregulation of the translocated MYC gene, which results in the enhancement of its oncogenic potential.4,8-10
Regardless of the endemic or sporadic origin of the BL, the MYC and IgH genes are juxtaposed in divergent orientation on the derivative chromosome 14q+ and the breakpoints on the partner chromosomes are distributed over a large region.1 This is the main reason why, in the past, molecular analysis of t(8;14) was performed by Southern blot technique, whereas polymerase chain reaction (PCR) was limited to selected cases in which the breakpoint could have been well characterized previously.11
The recent availability of more efficient DNA polymerases, which can yield products several kilobasepairs long from genomic DNA target sequences, has allowed the detection of a higher percentage of junctional sequences by using a limited number of primer combinations.12,13 In a series of pediatric BL and B-cell acute leukemia (B-ALL), 11 of 20 cases were positive for the long-distance PCR (LD-PCR) product combining 7 primers for the MYC locus and 5 for the IgH locus in different reactions.13
Because the t(8;14)(q24;q32) represents a unique characteristic of BL
and can be used as a marker for the detection of Burkitt’s cells, we
established a LD-PCR based on the use of one MYC-specific primer,
recognizing a region in exon 2, three primers for the constant IgH
regions Cµ, C, and C, and one for the JH region. The assay was
first used to study 7 Burkitt’s cell lines and then applied to the
analysis of 15 tumor samples obtained from children affected by BL.
We demonstrated that our LD-PCR method, which relies on a limited number of primers, has a high sensitivity and can be used efficaciously for the diagnosis and for the study of bone marrow micrometastases, as well as minimal residual disease.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The BL cell lines BL41, DAUDI, DG75, P3HR1, RAJI, and RAMOS were kindly provided by Dr. R. Dolcetti (CRO, Aviano, Italy), and CA46 cells were a gift of Dr. I. Magrath (National Cancer Institute, Bethesda, MD). They have been previously described.11,14-18 The T-cell leukemia cell line Jurkat and the rhabdomyosarcoma line RH30 were purchased from the American Type Culture Collection (Manassas, VA); the anaplastic large cell lymphoma cell lines SUDHL-1 and KARPAS-299 were purchased from the DSMZ cell bank (Braunschweig, Germany). Cell lines were maintained in RPMI-1640 medium (Life Technologies, Milan, Italy) supplemented with 100 U/ml penicillin, 100 µg/ml streptomycin, and 10% fetal calf serum (Seromed, Berlin, Germany) in humidified atmosphere with 5% CO2 at 37°C. Mycoplasma test performed on each cell line was negative.
Tumor specimens were obtained at diagnosis from unselected children with BL enrolled in the non-Hodgkin’s lymphoma trial of the Italian Association of Pediatric Hematology and Oncology (AIEOP). Each patient was identified by a number to protect confidentiality. Tissue was frozen and stored at -80°C until used.
DNA Extraction
High molecular weight genomic DNA was prepared from cultured cell lines and frozen tissue using the QIAamp Tissue Kit (Qiagen, Hilden, Germany) following instructions of the manufacturer.
LD-PCR
Quality of genomic DNA and adequacy of sample for the
amplification of long DNA fragments were tested in each case by using a
set of primers of the Human tPA Control Primer Set (Boehringer
Mannheim, Mannheim, Germany), which yield PCR products of 4.8, 9.3, and
15 kb, respectively. To detect the translocation involving the MYC and
the IgH genes we used 1 primer for the MYC gene (MYC/04)12
combined, alternatively, with 4 primers for the IgH locus: 3 primers
for the constant regions (Cµ/03, C/02, C/01)12
and
1 for the joining region (JH).13
Primers for both genes
represent the antisense strands in reverse direction, due to the
head-to-head orientation of MYC and IgH genes. Primer sequences are
reported in Table 1
.
|
Southern Blotting
After gel electrophoresis, PCR products were transferred onto positively charged nylon membrane (Boehringer Mannheim) and subsequently hybridized to a MYC-specific oligonucleotide probe (5'-TCGCTCTGCTGCTGCTGCTGG-3') recognizing a sequence of MYC exon 2 and previously described as MYC/06.12 It was labeled with digoxigenin by using the Dig oligonucleotide 3'-end Labeling Kit (Boehringer Mannheim) following the manufacturer’s instructions. Hybridization signals were detected by chemiluminescence (CDP-Star, Boehringer Mannheim) using an alkaline phosphatase-conjugated anti-digoxigenin antibody (Antidigoxigenin-AP Fab fragments, Boehringer Mannheim), according to the suggestions of the manufacturer. Membrane was exposed to Hyperfilm (Amersham, Milan, Italy) for 30 seconds to 5 minutes.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Based on the structure of the genes involved in the
t(8;14)(q24;q32) translocation and on data in the
literature,12,13
we selected a limited number of primers
bracketing the breakpoints and specific for exon 2 of the MYC gene and
for the joining and constant regions of the IgH gene (Table 1)
. Figure 1
is a schematic representation of the
location of each of the 5 primers with respect to the MYC and the IgH
genes. Primers and reaction conditions were first tested in 7
Burkitt’s cell lines with cytogenetically proven t(8;14), using in
different reactions the same MYC-specific primer in combination with
each of the primers recognizing a selected IgH region. In 5 of the cell
lines (BL41, CA46, DG75, RAJI, RAMOS) we obtained a PCR product ranging
in size from 1.5 to 7.6 kb (Table 2
and
Figure 2A
). The cell lines DAUDI and
P3HR1, obtained from endemic BL and known to have the breakpoint on
chromosome 8 about 170–190 kb upstream of MYC gene,19
were negative, as expected. Among the 5 BL cell lines positive for the
LD-PCR product, we found that BL41 and CA46 cells were positive with
the C/01 primer and DG75 cells were positive with the primer
specific for the JH region, whereas RAJI and RAMOS were positive when
using the C/02 and Cµ/03 primer, respectively. In this manner we
not only identified the region near the IgH gene breakpoint in
each cell line, but at the same time established a positive control for
each pair of primers to be used in the study of pediatric BL specimens.
|
|
|
Southern blot analysis was performed on all of the positive cell lines
using an oligonucleotide probe complementary to a MYC exon 2
sequence,12
which should be contained in all of the
amplified genomic sequences. All of the products of the LD-PCR were
recognized by the probe, thus demonstrating the specificity of the
assay (Figure 2B)
.
Sensitivity of LD-PCR
To determine the sensitivity of the method, increasing dilutions
of BL41 cells were prepared in T-lymphoblastic leukemia cells (Jurkat)
followed by DNA extraction. LD-PCR was then performed using the MYC and
C primers. The PCR product was detectable in ethidium
bromide-stained gel at BL41 cell concentration as low as
10-3, although the signal was weak (Figure 3A)
. In an attempt to increase the
sensitivity of the method, PCR products were transferred to a nylon
membrane and hybridized to the digoxigenin-labeled MYC/06
oligonucleotide probe. The signal, which appeared better defined than
in the ethidium-stained gel, was clearly visible at the same dilution
of 10-3 BL41 cells (Figure 3B)
.
|
Tumor specimens obtained from 15 unselected children with BL were studied. DNA extracted from the samples was suitable for LD-PCR analysis because amplification of a 4.8-kb fragment of the tPA gene was achieved in all cases.
In 13 tumors (87%) we detected the t(8;14)(q24;q32) by LD-PCR, whereas
2 specimens were negative. The PCR fragments ranged in size between 2.0
and 9.5 kb (Figure 4A
and Table 2
). The
IgH region involved in the translocation was C in 2 cases (B01,
B07), C in 2 cases (B13, B15), Cµ in 6 cases (B02, B04, B05, B10,
B11, B12) and JH in 1 case (B08). In cases B09 and B14 we obtained two
different PCR products for each sample, one using the JH primer and a
larger one using the C/01and C/02, respectively (Table 2
and
Figure 5
).
|
|
. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In particular, by using 7 primers specific for the MYC gene and 5 for the IgH locus, zur Stadt et al could detect the t(8;14) in approximately 50% of children with BL or B-ALL. Their method implied that, on average, 20 different primer pairs were used to study each sample.13
In this work we established a simplified and highly efficient LD-PCR
approach for the detection of the t(8;14)(q24;q32) using fewer primers
and focused our attention exclusively on BL. The MYC/04 primer was
chosen because it recognizes a sequence in exon 2 that is external to
the region of the gene where the breakpoints occur more often (exon 1
and intron 1). Cµ/03, C/02, and C/01 primers were selected
because they recognize constant regions of the IgH locus in the
proximity of the switch regions most often involved in the
translocation. Furthermore, because it was reported that in European
BL, similarly to the endemic BL, the joining region could be often
involved,21
we decided to use also the JH primer. The
decision to use only four pairs of primers was supported by the
preliminary observation that our LD-PCR method allowed the
amplification of large fragments of genomic DNA, up to 12 to 15 kb.
Initially we characterized 7 BL cell lines cytogenetically positive for
the t(8;14). The CA46 cell line was positive with the C/01 primer,
as expected by the reported location of the breakpoint within S
region;22
RAJI was positive for the C/02 primer
confirming the breakpoint in the region S23,24
and
RAMOS cell line was positive with the Cµ/03 primer, as expected by
the description of the breakpoint within Sµ region.11
BL41 and DG75, for which no detailed information existed in the
literature regarding the exact region of the IgH breakpoint, were
characterized by an involvement of C and JH regions, respectively.
The analysis of BL cell lines allowed us also to establish a positive
control for each pair of primers, which we used on clinical specimens
(Table 2)
, and to optimize reaction conditions. The cell lines DAUDI
and P3HR1 originated from endemic BL were negative, as expected, since
the breakpoint on chromosome 8 is far 5' of the MYC gene in this group
of BL.19
LD-PCR analysis of 15 tumor samples obtained from children with BL showed that 13 (87%) were positive for the t(8;14): although obtained on a small series of patients, this represents the highest sensitivity of such an approach compared with the data for other approaches reported in the literature.13 Whether the absence of patients with B-ALL in our series, compared with previous reports,13 might have any relevance for the results of our study, although unlikely, remains to be demonstrated.
In 6 of 15 cases cytogenetic analysis was performed successfully. For the remaining patients we do not have cytogenetic data, either because only fixed tumor tissue was available or because the quality of the fresh tissue was inadequate for cytogenetic studies.
One of the two tumors negative by LD-PCR was cytogenetically positive for the translocation; in this case, failure to detect the recombined sequences might be due to an unusual breakpoint in the MYC gene (ie, 3' of the exon 2 region recognized by the MYC primer), to a distant breakpoint in the IgH region, or to a loss of the IgH recognition site of the IgH primers used. Unfortunately, because of the limited amount of tissue available, a Southern blot analysis was not possible. The second negative case was that of a patient with histologically typical BL localized to the Waldeyer ring who had surgery for a suspect benign tonsillar hypertrophy and for whom no cytogenetic analysis was planned.
Interestingly, the Cµ region was the most frequently involved in our patients, as previously reported for BL occurring in North America.7 Moreover, 3 of 13 positive cases had a breakpoint within or close to the JH region.
This region is frequently involved in endemic BL, but was also reported in European BL. In a series of Italian and Spanish BL, 6 of 12 cases had the involvement of the JH region, but almost all of them were adults.21 In the pediatric series of zur Stadt, 3 of 11 positive children had the breakpoint in the JH region.13
With regard to the 3 cases positive with the JH primer, case B09 and
B14 also showed a PCR product with C/01 and C/02, respectively.
This was not reported in previous studies and has different possible
explanations. The PCR products obtained with the JH primer, in both
cases, were smaller (2.5 and 2.0 kb) than those obtained with the
C/01 and C/02 primers (8 and 9 kb, respectively). This
observation, together with the knowledge that the DNA region
between the JH and the C/01 or the C/02 primers is far
above the size limit of amplification for the current LD-PCR
techniques, suggest that there might be a deletion between the joining
and the C regions in case B09 or between the joining and C
regions in case B14. This hypothesis is supported by the absence of any
PCR product using primers for regions located between JH and C or
C, respectively (Figure 5)
. Alternatively, one could speculate that
two different clones, harboring a different t(8;14) translocation,
might coexist in the same tumor or that two different t(8;14)
translocations affected both alleles within the same cell clone.
Because it is unlikely that in a tumor cell there might be more than
one t(8;14) and because the rearrangements of the IgH locus with DNA
deletion is a normal event in the process leading to immunoglobulin
production in B lymphocytes, the first hypothesis appears the more
suitable to explain our findings.
Given the fact that the LD-PCR product detecting the t(8;14) is
specific for each individual tumor, it represents a marker that could
be used to monitor micrometastases at diagnosis or during treatment in
patients with BL. In our hands this technique was capable of detecting
10-3 BL cells when mixed in vitro
with other leukemic cells (Figure 3, A and B)
. Thus, it has the
potential to be applied for the determination of minimal residual
disease in the bone marrow, as well as for the discrimination of tumor
cells inside a residual mass during or after treatment. Although
validation of these methods and of their clinical relevance for BL of
childhood requires a larger series of patients (particularly in the
setting of the current high-dose-intensity chemotherapy regimens, which
can cure the great majority of pediatric BL), we believe their use
could improve the management of this disease.
As a general consideration regarding the sensitivity and applicability of the LD-PCR technique for the detection of t(8;14) in BL, it should be clear that other methods, such as cytogenetic and FISH analyses, can be used. Cytogenetics has certainly the advantage of detecting not only the t(8;14), but also the variant translocations t(2;8) and t(8;22) typical of BL4 and any other chromosomal aberrations. Classic cytogenetics is hampered by a low mitotic index or by a poor quality of metaphase spreads in approximately 20% of the cases even in very experienced laboratories.25 FISH analysis, on the other hand, is suitable to demonstrate the t(8;14)(q24;q32), it can give some information on the distribution of the translocation within a tumor cell population on a cell-by-cell basis and, most importantly, it can be performed on fresh as well as on fixed tissue. It has the limitation, though, that the breakpoints on the two chromosomes cannot be characterized so precisely as by LD-PCR, and it might suffer from excess false positive or false negative results, depending on the criteria used to interpret the results.26 A disadvantage common to both cytogenetic and FISH analyses, compared to LD-PCR techniques, is that they are not suitable for minimal residual disease studies.
Ideally, any BL case should be studied by means of classic and molecular cytogenetics and LD-PCR, which represent two complementary methods in the genetic characterization of the tumor.
In conclusion, we have shown that using LD-PCR for the detection of t(8;14)(q24;q32) based on a limited number of primer combinations can be useful in the study of sporadic BL and has the potential for application to the assessment of micrometastases and minimal residual disease.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported by AIL (Associazione Italiana contro le Leucemie) and by MURST (Ministero Università Ricerca Scientificae Tecnologica) 9706153160–006/1997. K. Basso is recipient of a fellowship from AIL (sezione di Treviso).
Accepted for publication July 15, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
switch region. Mol Cell Biol 1985, 5:501-509 1 genes in a Burkitt lymphoma revealing a third exon in the c-myc oncogene. Nature 1983, 304:135-139[Medline]
This article has been cited by other articles:
 |
|
 |
|
  L. Mussolin, M. Pillon, V. Conter, M. Piglione, L. Lo Nigro, P. Pierani, C. Micalizzi, S. Buffardi, G. Basso, L. Zanesco, et al. Prognostic Role of Minimal Residual Disease in Mature B-Cell Acute Lymphoblastic Leukemia of Childhood J. Clin. Oncol., November 20, 2007; 25(33): 5254 - 5261. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
