| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Molecular Biology and Genetics |
Program in Molecular Biology [C. J. B., M. J. V.] and Departments of Radiation Oncology [A. T. M. V.] and Medicine [M. O. D.], Maywood, Illinois 60153
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. In particular, the therapy-related breakpoints appear clustered within the telomeric region of the BCR, a location that colocalizes with a DNAase hypersensitivity site and is also the target for site-specific apoptotic cleavage (10, 11, 12)
. The mechanism controlling such nuclease sensitivity is not well understood but may be linked to the earliest detectable apoptotic fragmentation that occurs at sites of DNA attachment to the nuclear matrix (13
, 14)
. The telomeric region of MLL contains an experimentally defined nuclear matrix attachment site (15)
. These locations are also associated with the ability to undergo transient base unpairing linked to tracts of AT- or ATC-rich DNA (16)
. Apoptotic nucleases may therefore cleave within MLL and other locations such as LINES by targeting specific DNA structures found at the nuclear matrix (13
, 15)
. In this study, we asked whether nuclease cleavage within intron 11 of MLL by the apoptosis effector system might be a separable event from apoptotic cell death and thus act as a trigger for translocation initiation in cells destined to survive.
|
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Detection of MLL Cleavage by Ligation-mediated PCR.
The technique is based on ligating a double-stranded asymmetric adapter to the blunt-ended double strand break inserted into MLL by apoptotic nucleases. The adapter and MLL sequences are then used in a seminested PCR reaction. The size of the product produced enables the location of the initiating DNA double strand break to be determined. Adapter: 5'-GCGGTGACCCGGGAGATCTGA-3' and 3'-CTAGACTTAA-5'; MLL seminested primers: 5'-ATGCCCAAGTCCCTAGACAAAATGGTG-3' and 5'-GTCTGTTCACATAGAGTACAGAGGCAACTA-3'.
The two homologous oligonucleotides are annealed together at a gradually decreasing temperature gradient and ligated to isolated genomic DNA using T4 DNA. Before PCR the reaction is heated to 72°C to dissociate the 11-mer oligomer, leaving a 5' 25-mer overhang. During PCR, Taq polymerase elongates the staggered 25-mer end to create the homologous strand producing a double-stranded 25-mer linker molecule ligated to its terminus. Ligation-mediated PCR is conducted using the linker-ligated DNA as a template, the 25-mer oligonucleotide as a primer, and primers specific to MLL. Amplification of cleaved MLL fragments with the primers used generates a product of 
290 bp, identified by Southern blotting using a cDNA probe covering the MLL BCR. The probe used hybridizes with exon 12 contained within the 290-bp product. The source of the PCR product was confirmed by restriction enzyme digestion.
PARP Cleavage.
Protein was extracted from TK6 cells using conventional techniques and prepared for Western blotting using an anti-PARP (Ab-2) antibody (Oncogene Research, Cambridge, MA). Apoptosis introduces a site-specific cleavage event in PARP that splits the Mr 115,000 molecule into Mr 89,000 and Mr 26,000 fragments.
Field Inversion Gel Electrophoresis Analysis of High Molecular Weight DNA Fragments.
For separation of high molecular weight DNA, cells were immobilized in 1.5% low melting agarose; the plugs produced were then treated with 0.5 mg/ml proteinase K, 10% Sarkosyl, 25 mM EDTA, 10 mM Tris (pH 9.5) at 50°C for 24 h. The plugs were then washed three times in 10 mM EDTA, 1 mM Tris-HCL (pH 8), and separation of high molecular weight DNA performed in 1% agarose in 0.5x Tris-borate EDTA buffer at 100V, 25 mA at 4°C using a Power Inverter PPI-200 (MJ Research). Fragmentation was viewed using a Southern blotting technique with the MLL probe described above.
RT-PCR Analysis of MLL-AF9 Expression.
TK6 cells (1 x 107) were treated with 0.5 µg/ml anti-CD95 antibody and RNA isolated at 0, 6, 24, 48, 72, and 96 h after apoptotic induction using the TriZol extraction protocol (Promega, Madison, WI). Nested RT-PCR was then performed with the Titanium RT-PCR kit (Clontech, Palo Alto, CA). Primers used were as follows: MLL RT-forward primer 1, 5'-GCAAACAGAAAAAAGTGGCTCCCCG-3'; AF9 RT-reverse primer 1–5'-TCACGATCTGCTGCAGAATGTGTCT-3'; MLL RT-forward primer 2, 5'-CCTCCGGTCAATAAGCAGGAGAATG-3'; and AF9 RT-reverse primer 2, 5'-CAGAGTCATTGTCGTTATCCTCCAC-3'.
PCR products were separated on a 2% agarose gel and bands corresponding to the fusion transcript were excised, cloned, and sequenced to confirm the source of the PCR product. In cases where zVAD-fmk was used, cells were pretreated for 30 min with the compound before the addition of anti-CD95 antibody.
Analysis of MLL-AF9 Breakpoint Junctions: Genomic PCR.
TK6 cells (1 x 107) exposed to 0.5 µg/ml of anti-CD95 antibody (Kamiya, Seattle, WA) were incubated for a period of 72 h at 37°C and 5% CO2. At this point, cells were split into eight separate vials and both the DNA and RNA isolated separately. Samples were first screened for the presence of the MLL-AF9 fusion transcript via RT-PCR (described above), all such experiments were positive for MLL-AF9 transcription. Samples were then analyzed for a specific MLL-AF9 genomic fusion using the following primers. MLL forward primer 1 5'-CAAACCAGACCTTACAACTG-3'; AF9 reverse primer 1, 5'-CCCTCCTGTGCAGAGCAT-3'; MLL forward primer 2, 5'-TCGTATATTACAGAAAACGTTTA-3'; and AF9 reverse primer 2, 5'-ACAGGGCTCAGTACTGGAT-3'.
PCR products were separated on a 2% agarose gel and stained with ethidium bromide. Bands corresponding to potential MLL-AF9 translocations were excised, purified with the QiaQuick Gel Purification Kit (Qiagen, Valencia, CA), and cloned into pCR4-TOPO (Invitrogen) and sequenced.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, a discrete PCR product of 290 bp is apparent, 1.5–2 h after exposure of TK6 cells to anti-CD95 antibody. The size of the PCR product shows the location of the DNA break to be at 6768 ± 10 bp, within intron 11 and close to the exon 12 boundary using the notation of Gu et al. (17)
. This specific break site was consistently observed in all apoptotic populations studied. Other PCR products were occasionally seen, indicating cleavage at adjacent sites. However such cleavage was not consistently observed either within or between experiments. These data are notable for demonstrating cleavage at a discrete location, rather than being distributed over an extended region. The site of cleavage is 93 bp 5' to a strong topoisomerase II consensus site and 270 bp 3' from an extensive tract of ATC associated with base unpairing and attachment to the nuclear matrix (Ref. 15
, 16
and Fig. 5
). A small stem loop (2-bp stem; 2-bp loop) structure is located at the site of cleavage, however, the significance of this is unknown. The appearance of this cleavage site is consistent with the earliest time that apoptosis is evident, as shown by PARP cleavage (Fig. 1b)
. Addition of 20 µM zVAD.fmk, 30 min before adding anti-CD95 antibody, eliminated both the LM-PCR product and evidence of apoptosis (Fig. 1, b and c)
. These data confirmed that cleavage of both MLL and PARP were similarly dependent upon the apoptotic program.
|
50 kb. These are proposed to be the result of apoptotic scission at DNA nuclear matrix attachment sites (13
, 14)
. Nuclease attack at these locations is therefore both early and rare, raising the possibility that breaks introduced here have the potential to be rejoined consistent with cell survival. To explore the association of nuclease cleavage of MLL with early apoptosis, 50-kb fragments of DNA were separated from apoptotic TK6 cells by field inversion gel electrophoresis (Fig. 1d)
. The 50-kb fragments were then removed from the gel and analyzed by LM-PCR as above. If the termini of such large fragments coincided with the apoptotic cleavage site in MLL, then an identical 290-bp PCR product will be produced as before. As seen in Fig. 1d
, the PCR reaction confirmed that 6 h after exposure to anti-CD95, the high molecular weight fragments produced contained the MLL apoptotic cleavage site at their termini. Elimination of the linker ligation step gave no detectable products. These data therefore link MLL cleavage with both an early apoptotic event and a specific structure, the nuclear matrix.
MLL-AF9 Translocations Are Produced by the Apoptotic Program.
We have previously shown that exposure of TK6 cells to either irradiation or anti-CD95 antibody triggered translocations to a range of chromosome partners, initiated at a location adjacent to the site of apoptotic cleavage shown here (11)
. None of the MLL translocation partners identified by this inverse PCR approach had any recognized role in human leukemia. For this reason, we were interested in identifying translocations initiated by apoptosis with known pathological consequence and so chose to search for MLL-AF9 translocations. This and MLL-AF4 are two of the most commonly observed translocations isolated from patients with AML and acute lymphoblastic leukemia, respectively (2)
. TK6 cells were exposed to 0.5 µg/ml anti-CD95 antibody, and RNA was extracted using standard techniques. RT-PCR was performed using primers designed to detect the message resulting from ligation of chromosome breaks within intron 11 of MLL and intron 4 of AF9, the latter a common partner in MLL translocations and which contains similar structural motifs to that found at the MLL cleavage site (18)
. All combinations of such breaks, after ligation, will generate an MLL-AF9 message of characteristic size(s) after exon splicing occurs. Cell samples were taken from 6 to 96 h after treatment, and the presence of a PCR product of the size expected for an MLL-AF9 fusion was observed (Fig. 2a)
. To confirm that the PCR products obtained were the result of MLL-AF9 gene fusion, they were excised from the gel, cloned into pCR4-TOPO, expanded in liquid culture, and sequenced. All clones obtained were positive for the MLL-AF9 fusion. Two RT-PCR products were observed in the MM6 cell line used as a reference and also in some experiments. These products were also sequenced and found to be splice variants, involving the presence or absence of exon 11 as reported before (Fig. 2
; Refs. 19
, 20
). The reason for the variable splicing in some experimental samples is unknown. Repeating the exposure of TK6 cells to anti-CD95 antibody in the presence of the irreversible inhibitor of caspase activity, Z-DEVD-fmk, eliminated the MLL-AF9 PCR product. These data indicate that the apoptotic process was involved in the initial DNA strand breaks in MLL and, probably AF9, that were subsequently ligated together permitting transcription of the resulting MLL-AF9 fusion gene (Fig. 1b)
.
|
, no MLL-AF9 transcripts were observed when 1.25 x 105 cells or less were analyzed. This sets a limit on the frequency of MLL-AF9 translocations production after exposure of TK6 cells to anti-CD95 at 1:1.2 x 105.
Cells Expressing Apoptosis-induced MLL-AF9 Translocations Are Able to Divide.
A critical question in this study was the capacity of individual cells to survive site-directed apoptotic cleavage. The rationale used was to detect cells that had divided after an MLL-AF9 translocation by the presence of two or more cells containing sequence identical MLL-AF9 breakpoint junctions in the same population of cells.
The creation of a transcriptionally active MLL-AF9 fusion can use any break that occurs within the 3.5-kb intron 11 of MLL and the 40-kb intron 4 of AF9. To reduce this wide range of possible fusion junctions to manageable levels, a single set of nested primers was used to prospectively amplify a restricted region of MLL and AF9 genomic fusions. The system was designed to include the documented apoptotic cleavage site in MLL and a location in AF9 that has been shown to be a site for MLL ligation in multiple clinical samples (12)
. Because of the documented frequency of the MLL-AF9 translocations (1:105) and the restrictive conditions for PCR amplification, a PCR product found in two or more cells is more likely to have arisen after cell duplication than be independently produced events. This possibility may be confirmed by sequencing. The location of the PCR primers used ensures that any translocations observed will be from the same regions of MLL and AF9 involved in MLL-AF9 fusions in clinical material. To carry out the experiment, 4 x 106 cells were treated with 0.5 µg/ml anti-CD95 antibody that kills 60–75% of cells by 24 h. The surviving cells are able to recover both their starting cell number and viability (>95%) 3 days after treatment. Therefore, 3 days after treatment, cells were placed into eight different tubes, each tube contained 0.5 x 106 cells of which 4 cells would be predicted to contain a de novo MLL-AF9 translocation (0.5 x 106/125,000). This experiment was repeated until PCR products from MLL-AF9 fusion were observed in two or more tubes. The results of 2 of 10 such screening experiments are shown in Figs. 3, a and b
. For the experiment in Fig. 3a
, the top band of the two detected was shown to be a sequence identical MLL-AF9 breakpoint junction. The lower band was apparently the result of a false-priming reaction. In the case of Fig. 3b
, individual PCR reactions for each lane were carried out on different days to ensure that cross contamination did not occur. Here the major band detected was also a unique, duplicated MLL-AF9 fusion. These results were consistent with the prediction of the experimental design in that if two cell samples from the same population were positive for the MLL-AF9 genomic breakpoint, they would most likely be duplications because of the predicted rarity of individual breakpoint junctions. The breakpoint junctions observed had therefore most likely arisen from the duplication of a single progenitor cell, indicating the capacity for cells containing MLL-AF9 fusion junctions, triggered by apoptosis, to divide. The DNA sequence of the identical paired duplications shown in Fig. 3, a and b
, were different, both from each other and the MLL-AF9-positive cell line (MM6) used as a positive control (Fig. 4)
.
|
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Activation of apoptosis with anti-CD95 antibody, unlike genotoxic agents, introduces DNA breaks only as a consequence of apoptotic activation. The site of attack is linked to the nuclear matrix attachment region, associated with the generation of high molecular weight fragments (50 kb), indicating that cleavage here is a rare event (Fig. 1d)
. Second, activation of apoptosis is sufficient to provide DNA strand breaks in both MLL and AF9 that are subsequently fused as biologically active MLL-AF9 translocations, detected by the presence of MLL-AF9 transcripts (Fig. 2a)
. Finally, cells containing an apoptosis-induced MLL-AF9 translocation are able to divide, providing evidence of biological relevance (Fig. 3)
. A controversial aspect of these data are the implication that cells entering the chromatin cleavage stage of apoptosis retain the capacity to divide. The conventional interpretation of apoptosis is that once triggered, the process continues until the inevitable death of the cell (21 , 22) . This linear view of apoptosis is being challenged by data illustrating the intimate association between the apoptotic and cell growth regulatory pathways (23) . As an example, the oncoprotein Myc is involved in the regulation of clonal expansion in addition to promoting apoptosis through facilitating cytochrome c release (24) . Of specific relevance here, direct microscopic observation of the nematode Caenorhabditis elegans, has shown that cells are able to survive the morphological expression of apoptosis in the absence of phagocytic signals (25 , 26) . Cells that express the early marker of apoptotic commitment, annexin V, also retain the ability to divide (27) . Others have shown that caspases, in particular the executioner apoptotic protease caspase-3, may be found in T cells that are not undergoing apoptosis (28) . These data provide indirect support for the contention that cells may retain viability into the chromatin cleavage phase of apoptosis, providing the damage introduced does not overwhelm a cells capacity to repair.
The data presented raise a number of issues relevant to leukemogenesis. In experiments involving anti-CD95 treatment, TK6 cells containing detectable MLL-AF9 fusion transcripts were observed at a frequency of 1:105. These data indicate that the generation of fusion genes of known pathological significance is a relatively common event. Thus, both MLL and AF9 are frequent targets for cleavage and inappropriate ligation. The normal execution of apoptosis will remove the overwhelming majority of cells carrying these fusion genes and the need for a second (or more) genetic events to trigger leukemogenesis will additionally suppress transformation of these cells (9)
. Although this study specifically addressed cleavage introduced into the MLL gene by apoptosis, the AF9 gene also contains a nuclease sensitive site that exhibits structural similarities with MLL, including the relationship of nuclear matrix attachment sites and topoisomerase II consensus sequences (18)
. Proapoptotic agents are also able to introduce site-specific cleavage into the TEL and AML1 genes within EBV-immortalized human lymphocytes, and this also results in the generation of TEL-AML1 fusion transcripts (29)
. Significantly, lymphocytes taken from the peripheral blood of normal individuals also undergo MLL cleavage on exposure to a range of apoptotic agents (10)
. Thus MLL cleavage is not restricted to malignant cells.
It is clear from clinical data that the induction of therapy-related leukemia harboring MLL rearrangements is strongly linked to the presence of topoisomerase II inhibitors such as etoposide (5
, 6)
. Such agents are also activators of the apoptotic program, which alone is sufficient to create the leukemogenic fusion gene without directly affecting topoisomerase II (Fig. 2)
. However cytotoxic agents used in therapy that activate apoptosis but do not target topoisomerase II are not associated with MLL rearrangements in related leukemias (5)
. These clinical data and the proximity of a strong (all 10 required bases are present in both the top and bottom strand) topoisomerase II consensus site immediately 3' to the site of apoptotic cleavage and a functional link to the same motif in the cleavage of AF9 all argue for the specific involvement of topoisomerase II in fusion gene creation (15)
. In addition, the breakpoints within both MLL and AF9 are in close proximity to sites of nuclear matrix attachment, as determined by experiment (15
, 18)
. Analysis of the DNA sequence surrounding the cleavage site in MLL shows a region enriched in ATC sequences 270-bp 5' to the cleavage site (Fig. 5)
. Such regions are known to adopt a stable base unpaired configuration, a DNA arrangement that is strongly associated with nuclear matrix binding (16)
. We have shown in Fig. 1d
that MLL cleavage is linked to excision of large DNA loops, a process initiated at the nuclear matrix (14)
. It is proposed that immobilizing MLL DNA at the nuclear matrix is sufficient to create a nuclease sensitive target that is functionally similar to those targeted in early apoptotic fragmentation (14)
. Inactivation of the adjacent topoisomerase II site by drugs such as etoposide traps the preexisting nuclease target between these two fixed points, potentially enhancing the nuclease sensitivity of the target by stabilizing DNA secondary structure (Fig. 5)
. Using a prostate cancer cell line model, others have suggested that the proximity of nuclear matrix and topoisomerase II motifs provides a stable platform for chromosomal exchange. Here, the high mobility group protein, HMGI(Y), is proposed to stabilize DNA into a form that promotes DNA exchange, which requires the presence of both base unpaired DNA and topoisomerase II (30)
.
The survival of cells that have initiated the apoptotic program and have inherited proleukemogenic translocations may be linked to errors in the regulation of apoptotic execution. Such failures are predicted to be highest during embryogenesis and chemotherapy exposure where the amount of apoptosis and risk of initiating leukemia are both increased (31 , 32) . Additionally, these data provide an explanation for the observation that fusion gene transcripts may be detected in a significant number of normal individuals (33) . Such transcripts may be produced by the routine operation of the apoptotic program, but this normally occurs in cells destined to die and therefore, in most cases, is without clinical consequence. A critical question therefore is to determine what factors might promote the survival of cells that have engaged the effector stage of the apoptotic program. A possible pathway may be found in the naturally occurring inhibitor of apoptosis proteins (34 , 35) . These proteins, of which the X-linked member XIAP appears the most potent, specifically bind to activated caspase-3 and have been shown capable of suppressing apoptosis and increasing cell survival (36 , 37) . In particular, XIAP exerts activity in cells where caspase-3 and therefore caspase-3-dependent nucleases such as CAD are already functional (38) .
These data provide evidence for the existence of a novel mechanism of leukemogenesis based on site-specific targeting of apoptotic nucleases and the aberrant behavior of apoptotic effector processes. The substantial knowledge of the pathways that regulate the apoptotic program should enable the design of strategies to eliminate or suppress this mechanism for the creation of pathogenic fusion genes. The simplest route for such manipulation would be techniques that enhance the efficiency of apoptotic execution, eliminating potentially aberrant cells. Of wider significance, these data indicate that additional deleterious events linked to the accumulation of DNA double strand breaks such as mutations or loss of heterozygosity may also be triggered by inappropriate activation of the apoptotic program and may similarly be subject to manipulation for patient benefit.
|
FOOTNOTES |
|---|
1 This work was supported by a VA Merit Grant (to A. T. M. V.) and a gift from the Wolf family (to M. O. D.). 
2 To whom requests for reprints should be addressed, at Loyola University Medical Center, 2160 South First Avenue, Building 112, Maywood, IL 60153.
3 The abbreviations used are: MLL, mixed lineage leukemia; AML, acute myelogenous leukemia; BCR, breakpoint cluster region; PARP, poly(ADP-ribose) polymerase; IAP, inhibitor of apoptosis protein; LM-PCR, ligation-mediated PCR; RT-PCR, reverse transcription-PCR; CAD, caspase-activated DNase.
Received 8/ 1/02. Accepted 1/17/03.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. W. Robinson, N.-K. V. Cheung, C. P. Kolaris, S. C. Jhanwar, J. K. Choi, N. Osheroff, and C. A. Felix Prospective tracing of MLL-FRYL clone with low MEIS1 expression from emergence during neuroblastoma treatment to diagnosis of myelodysplastic syndrome Blood, April 1, 2008; 111(7): 3802 - 3812. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. B. van Waalwijk van Doorn-Khosrovani, J. Janssen, L. M. Maas, R. W.L. Godschalk, J. G. Nijhuis, and F. J. van Schooten Dietary flavonoids induce MLL translocations in primary human CD34+ cells Carcinogenesis, August 1, 2007; 28(8): 1703 - 1709. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Azarova, Y. L. Lyu, C.-P. Lin, Y.-C. Tsai, J. Y.-N. Lau, J. C. Wang, and L. F. Liu From the Cover: Roles of DNA topoisomerase II isozymes in chemotherapy and secondary malignancies PNAS, June 26, 2007; 104(26): 11014 - 11019. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. S. Hars, Y. L. Lyu, C.-P. Lin, and L. F. Liu Role of Apoptotic Nuclease Caspase-Activated DNase in Etoposide-Induced Treatment-Related Acute Myelogenous Leukemia. Cancer Res., September 15, 2006; 66(18): 8975 - 8979. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. C. Briggs, K. E. Shults, L. A. Flye, S. A. McClintock-Treep, M. H. Jagasia, S. A. Goodman, F. I. Boulos, J. W. Jacobberger, G. T. Stelzer, and D. R. Head Dysregulated Human Myeloid Nuclear Differentiation Antigen Expression in Myelodysplastic Syndromes: Evidence for a Role in Apoptosis. Cancer Res., May 1, 2006; 66(9): 4645 - 4651. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. R. Mistry, C. A. Felix, R. J. Whitmarsh, A. Mason, A. Reiter, B. Cassinat, A. Parry, C. Walz, J. L. Wiemels, M. R. Segal, et al. DNA Topoisomerase II in Therapy-Related Acute Promyelocytic Leukemia N. Engl. J. Med., April 14, 2005; 352(15): 1529 - 1538. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. D. Aplan "You break it, you fix it" Blood, March 1, 2005; 105(5): 1843 - 1844. [Full Text] [PDF]
|
|
|
|
|
|
J. Libura, D. J. Slater, C. A. Felix, and C. Richardson Therapy-related acute myeloid leukemia-like MLL rearrangements are induced by etoposide in primary human CD34+ cells and remain stable after clonal expansion Blood, March 1, 2005; 105(5): 2124 - 2131. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Yu and A. Gabriel Reciprocal Translocations in Saccharomyces cerevisiae Formed by Nonhomologous End Joining Genetics, February 1, 2004; 166(2): 741 - 751. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. G. Gilliland, C. T. Jordan, and C. A. Felix The Molecular Basis of Leukemia Hematology, January 1, 2004; 2004(1): 80 - 97. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
