| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
1 Laboratory of Genetics, Center for Cancer Research, National Cancer Institute, and
4 Laboratory of Host Defense, National Institute of Allergy and Infectious Diseases, NIH, Bethesda, Maryland;
2 Institute of Molecular Biology and Tumor Genetics, Munich, Germany;
3 Institute of Mammalian Genetics, Neuherberg, Germany; and
5 Science Applications International Corporation-Frederick, National Cancer Institute, Frederick, Maryland
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Mouse models of human cancer are suitable for studying the relationship between general mutagenesis and oncogenesis under genetically controlled conditions. Among the many advantages of the mouse is the ability to assess mutations in vivo using transgenic shuttle vectors with a bacterial reporter gene. Several transgenic reporter systems for measuring mutation frequency in vivo have been developed (7, 8, 9, 10, 11)
. Two widely used systems rely on the phage 
vector, LIZ (12
, 13)
, or the plasmid vector, pUR288 (14)
. Both vectors contain nontranscribed, neutral reporter genes (lacI in case of LIZ and lacZ in case of pUR288) that have been incorporated into the mouse genome in which mutations occur. The reporter genes are excised from mouse genomic DNA and transferred to Escherichia coli, affording efficient selection, enumeration, and analysis of mutations. Both mutagenesis assays, LIZ/lacI and pUR288/lacZ, are validated and thought to accurately reflect the mutant frequency (MF) in the overall genome (15
, 16) . One notable difference, however, is that the LIZ/lacI assay detects primarily point mutations and small deletions in lacI, whereas the pUR288/lacZ assay additionally detects large deletions in lacZ and intra- and interchromosomal rearrangements of lacZ with mouse genomic sequences.
Elevated mutagenesis may facilitate the development of plasmacytomas (PCTs), malignant tumors of terminally differentiated B lymphocytes. Peritoneal plasmacytomagenesis in mice is dependent on a number of pathogenetic factors (reviewed in Ref. 17 ), of which at least two have the potential to increase general mutagenesis in B cells and plasma cells. The first one is the chronic inflammatory, granulomatous tissue that is formed in response to i.p. injections with pristane, which are required to induce PCT in genetically susceptible BALB/c (C) mice. The mutagenic principle in the pristane granuloma is the resident population of inflammatory cells, mainly macrophages and neutrophils, which are able to mutagenize bystander B/plasma cells; e.g., by releasing oxygen intermediates into the extracellular milieu (18, 19, 20) . The second factor is the chromosomal translocations that result in the deregulated expression of the oncogene c-Myc. Virtually all pristane-induced PCT carry Myc-activating translocations, most commonly T(12;15). Deregulated expression of Myc in consequence of T(12;15) translocation may cause increased mutagenesis by a variety of mechanisms, including (a) uncoupling of DNA replication from cell cycle progression (21 , 22) , (b) down-regulation of genes involved in DNA repair (e.g., members of the GADD family; Ref. 23 ) or antimutagenesis (e.g., BCL2; Ref. 24 ), (c) interference with DNA mismatch repair (25) , and (d) disruption of redox hemeostasis (26, 27, 28) . The involvement of pro-mutagenic factors in PCT development, such as chronic inflammation and deregulated Myc, provided the rationale for the hypothesis that PCTs exhibit a phenotype of increased mutability (mutator phenotype).
Here we used the pUR288/lacZ mutagenesis assay to investigate whether PCTs have a mutator phenotype, and, if so, whether this phenotype may be caused by chronic inflammation and/or deregulated Myc. We thus backcrossed the pUR288 transgene from PCT-resistant C57BL/6 (B6) mice onto PCT-susceptible C mice and induced PCTs in the newly generated C.pUR288 congenics. We found that primary and transplanted PCTs exhibited the mutator phenotype of 5.5-fold elevated MFs in lacZ compared with the background levels in normal B cells. To elucidate potential mechanisms of hypermutability, we designed in vitro model systems for assessing the mutagenic impact of inflammatory cells and activated Myc on primary B cells. The latter involved a new transgenic mouse model of inducible Myc expression, Tet-off/MYC. Although the mutation-enhancing effect of inflammatory cells and deregulated Myc was readily demonstrated, the aggregate of our findings suggested that the mutator phenotype of PCTs is mainly determined by intrinsic features of the tumor cells, presumably including mutator genes whose nature remains to be identified.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
).6
The intentional loss of one copy of the transgene resulted in the deterioration of the performance of the pUR288/lacZ assay by approximately 50%. Thus, only half the number of pUR288 plasmids were excised from genomic DNA of C.pUR288 mice relative to parental B6.pUR288 mice.
|
Induction and Transplantation of PCTs.
PCT were induced in 10-week-old C.pUR288 mice by three i.p. injections of 0.5 ml pristane spaced 2 months apart (30)
. Age-matched inbred C mice were used as controls. Beginning 1 week after the third injection of pristane, the mice were monitored for tumor development by microscopic examination of cytofuge specimens of ascites cells stained according to Wright’s Giemsa. Ascites samples (50–100 µl) were obtained by inserting a 16-gauge hypodermic needle into the peritoneal cavity. The presence of 10 or more, large, hyperchromatic, atypical plasma cells was indicative of incipient PCTs. The mice were scored as tumor-positive when the follow-up cytofuge specimen confirmed the presence of PCT cells. In most cases, this specimen contained more than 100 malignant plasma cells, reflecting the rapid progression of PCTs. Mice with advanced PCTs were sacrificed to obtain tumor material for histological confirmation of the tumor diagnosis. Four primary PCTs (G0) were propagated in vivo for three consecutive generations (G1-G3). This was accomplished by transferring 0.5 ml of tumor cell-bearing ascites into inbred C mice primed with 0.5 ml of pristane 2–3 weeks earlier.
pUR288 Mutagenesis Assay.
The pUR288 assay was performed according to the protocol established by Boerrigter et al. (14)
and Vijg et al. (16)
. Briefly, pUR288 plasmids were excised from genomic DNA by digestion with HindIII and were separated from genomic DNA using LacI repressor protein coupled to magnetic beads. After ligation, plasmids were electroporated into E. coli lacZ- galE- and plated onto phenyl-ß-D-galactopyranoside-supplemented mutant selection plates and 5-bromo-4-chloro-3-indolyl-ß-D-galactopyranoside-supplemented wild-type titration plates. MF was calculated as the ratio of mutants to the total number of colonies. GalE revertants were excluded from determinations of MF. To analyze lacZ mutations, plasmid DNA was extracted from overnight cultures using QIAprep 96 Turbo miniprep kits (Quiagen, Valencia, CA). Point mutations were distinguished from rearrangements (R) by restriction analysis with HindIII, AvaI/PstI, and SacI/PstI (New England Biolabs). Point mutations demonstrated the wild-type pattern of pUR288, whereas R demonstrated a deviant pattern.
Mutagenesis in B Cells Exposed to PECs.
LPS blasts were prepared from the spleens of 3-month-old C.pUR288 mice as described previously (20)
. Briefly, splenic perfusates were depleted of erythrocytes in hypotonic lysis buffer and were seeded at a density of 106 cells/ml in tissue culture flasks. The cells were grown for 48 h in RPMI 1640 supplemented with 50 µg/ml LPS from E. coli 055:B5 (Sigma, St. Louis, MO), 10% FCS, 2.0 mM L-glutamine, 50 µM 2-mercaptoethanol, and penicillin and streptomycin (Life Technologies, Inc.). Peritoneal exudate cells (PECs) were obtained from 3–6-month-old C mice primed with 0.5 ml of pristane 2–3 weeks before harvesting the PECs by peritoneal lavage. In some experiments, PECs were obtained from mice homozygous for a null allele of phoxp47 (31)
or the hypomorphic Gpdxa-m2Neu allele of glucose-6-phosphate dehydrogenase (G6PD; Ref. 32
). Loss of p47phox abrogates NADPH oxidase function, whereas partial deficiency of G6PD interferes with the oxidative burst by limiting the NADPH supply (Fig. 2E)
. The average yield of PECs was 1.5 x 107 per mouse. PECs consisted mainly of neutrophils (
60%) and macrophages (40%) and were at least 95% viable as determined by trypan blue exclusion. To assess PEC-mediated mutagenesis in B cells, the pUR288-harboring LPS blasts were cocultured with PECs (not containing pUR288) at a ratio of 1:2 for 24 h at 37°C. 12-O-tetradecanoylphorbol-13-acetate (TPA) (50 nM; LC Laboratories, Woburn, MA) was added to the cell culture (four applications spaced 6 h apart) to trigger the oxidative burst in PECs. After 24 h of coincubation at 37°C, the cells were harvested and genomic DNA was made for the pUR288 assay. In one experiment, LPS blasts were exposed to 4-nitroquinoline-1-oxide (4-NQO) in the absence of PECs (Fig. 2C)
.
|
. The transgene consisted of an MluI-PvuI fragment obtained from construct, pmyc-tet (BC266), described in greater detail elsewhere (33)
. The Tet-off/MYC contains the tetracycline-controlled transactivator, tTA, a minimal MYC promoter, and the protein-encoding portion of the human MYC gene (exons 2 and 3). The expression of MYC is activated on withdrawal of the tetracycline analogue, doxicycline (Tet-off). Tet-off/MYC mice were crossed with B6.pUR288 mice to develop a system in which mutagenesis in primary B cells can be studied under conditions of inducible Myc expression. B220+ splenocytes from Tet-off/MYC-pUR288 bitransgenic mice were fractionated by MACS (Miltenyi Biotec) using B220 (CD45R) VS+ columns. Cell recovery relative to total splenocytes was 30%. The purity of MACS-sorted B cells was 
90% by fluorescence-activated cell sorting. The pUR288-harboring B cells were cocultured in the presence of doxicycline (MYC off) or absence of doxicycline (MYC on). Myc expression was monitored by Western blotting. In some experiments, 4-NQO was added to induce heightened levels of mutagenesis compared with the pro-mutagenic environment of standard cell culture. Cells were harvested after 3 and 27 h to prepare genomic DNA for the pUR288 assay.
|
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, squares). Pristane-treated C mice showed a similar incidence [7 (35%) of 20] and onset of PCTs (Fig. 1A
, circles). These findings indicated that strain C.pUR288 is as susceptible to PCTs as strain C.
Compromising the performance of the pUR288/lacZ mutagenesis assay is another risk when generating pUR288 congenics. The lacZ assay is dependent on the integrity of the pUR288 transgene, which can be altered by deletions and other genetic rearrangements that result in the loss of individual gene copies from the concatemer. Loss of copy number affects assay performance by reducing the efficiency with which reporter genes can be excised from mouse genomic DNA and transferred to E. coli. This problem was encountered in a previous study involving the B6 to C transfer of the LIZ transgene, in which a severe drop in phage packaging efficiency was noted at backcross generations N4 and N5 (37)
. To avoid similar problems during the B6 to C transfer of the pUR288 transgene, the performance of the pUR288/lacZ assay was monitored at each backgross generation (N1–N6) to select breeders with a properly functioning mutagenesis assay. The average plasmid rescue efficiency at generation N7 (used for the tumor induction study described above) was 9.0 x 103/µg DNA, 50% lower than in parental B6 mice. The drop in rescue efficiency was expected because, unlike B6 mice, C mice contained only one copy of pUR288. Despite the reduction in assay performance, strain C.pUR288 was suitable for determining MF in PCTs using the lacZ reporter gene assay.
Elevated MFs in Primary PCTs.
The determination of the MF in transgenic reporter genes in tumors and corresponding normal and premalignant tissues provides important information about genomic stability control during oncogenesis. To evaluate whether pristane-induced PCTs exhibit genomic instability, the MF in lacZ was compared with that in B cells and peritoneal granulomas. This comparison was meaningful because B-lymphocytes are the precursors of PCTs and the inflammatory granuloma is the tissue site in which PCTs develop (Fig. 1B)
. The mean MF in primary PCTs (40.3 x 10-5; n = 5) was 5.5 times higher than the mean MF in MACS-sorted B220+ splenocytes (B cells; 7.36 x 10-5). This indicated that PCTs have a phenotype of moderate hypermutability. The MF in B220+ splenocytes was 2.2 times higher than in B220- splenocytes (non-B cells), consistent with a previous result using LIZ (38)
. The relatively high MF of splenic B cells may be related to their high proliferative activity in vivo, considering that increased proliferation facilitates mutagenesis (39)
. The MF in the granulomas (17.3 x 10-5) was 2.3 times lower than the MF in the PCTs, but 3.7-fold higher compared with the mesentery (4.71 x 10-5). The mesentery, a fatty tissue, was included as a control for the granulomas because it serves as the matrix for granuloma development in pristane-treated mice. The increased MF in the granulomas relative to the mesentery suggested that resident granuloma cells (mainly phagocytes, such as macrophages and neutrophils) accumulate somatic mutations by self-inflicted mutagenesis, possibly by releasing reactive oxygen intermediates. It further suggested that the granulomas provide a mutagenic environment for infiltrating B cells, which may become targets of phagocyte-mediated oxidative damage (Fig. 1B
, top). However, the contribution of B cells to the elevated MF in the granulomas remained unclear, because mutations in B cells were not distinguished from mutations in granuloma cells (both cell types carried the lacZ reporter gene).
Mimicking the Mutagenic Effect of the Inflammatory Granuloma in Vitro.
To measure mutations in B cells exposed to inflammatory granuloma cells, we coincubated LPS-stimulated B220+ splenocytes from C.pUR288 mice in vitro with pristane-elicited PECs (macrophages and neutrophils) from normal C mice. Mutations in the B cells (positive for lacZ) were enumerated, whereas mutations in PECs (negative for lacZ) were not (Fig. 2A)
. The coincubation of B cells and PECs for 24 h at a ratio of 1:2 resulted in a MF of 20.1 x 10-5 in the B cells, a 3.2-fold increase compared with B cells without PECs (6.2 x 10-5). Activation of the oxidative burst in PECs (four applications of 50 nM TPA spaced 6 h apart) resulted in a small additional increase in the MF to 23.2 x 10-5, but this was not significant. Treatment of B cells with TPA in the absence of PECs caused an 1.5-fold increase in MF (10.3 x 10-5) relative to background levels in untreated B cells (Fig. 2B)
. These results were consistent with previous ones using the LIZ assay (20)
. The magnitude of PEC-induced mutagenesis was impressive compared with that induced by the powerful UV-mimetic mutagen, 4-NQO, which generates oxidative stress and DNA adduct formation through intracellular redox cycling (40)
. TPA-stimulated PECs and 25 nM 4-NQO were roughly equipotent (Fig. 2C)
. The addition of the antioxidant catalase (2 units/µl) reduced the mutagenic potency of PECs by 75% (Fig. 2D
, column 3), but no further reduction in MF was achieved by adding 1 mM pyruvate on top of the catalase (Fig. 2D
, column 4). PECs obtained from mice with genetic defects in the oxidative burst (Fig. 2E)
were less efficient in causing mutations in the lacZ reporter gene than were normal PECs (Fig. 2F)
. Total loss of NADPH oxidase function (phoxp47 null mice; Ref. 31
) resulted in a drop of mutant levels by 35% on TPA stimulation (Fig. 2F
, column 4) and 41% without TPA stimulation (Fig. 2F
, column 2). Partial deficiency in glucose-6-phosphate dehydrogenase (G6PD) caused a reduction in MF by 51% in TPA-stimulated PECs (Fig. 2F
, column 5) and 49% in unstimulated PECs (Fig. 2F
, column 3). These findings agreed with evidence that PEC-induced mutagenesis is complex and involves both oxidative and nonoxidative mechanisms (41)
.
Mimicking the Mutagenic Effect of Deregulated Myc in Vitro.
Peritoneal PCTs carry chromosomal translocations that result in the deregulated expression of Myc, which may promote neoplastic plasma cell transformation by facilitating genomic instability in B cells (42, 43, 44)
. To assess the potential role of deregulated Myc as a catalyst of B-cell mutagenesis, a transgenic mouse model of conditional Myc expression was developed. The model, designated Tet-off/MYC, uses the tetracycline-responsive regulatory system of gene expression (45)
to overexpress a human Myc gene in all tissues, including B cells, on withdrawal of the tetracycline analogue, doxicycline (Fig. 3A)
. Fig. 3B
(top) shows the time course of Myc protein expression in Tet-off/MYC-pUR288 bitransgenic B cells on withdrawal of doxicycline. The increase in Myc reached a maximum after 3 h and lasted for 27 h. A similar peak of Myc expression at 2–4 h was also observed in SV40 large T-transformed Tet-off/MYC-pUR288 fibroblasts used as controls (Fig. 3B
, bottom).
To evaluate mutagenesis in B cells under conditions of inducible Myc expression, MACS-sorted LPS-stimulated bitransgenic B220+ splenocytes were cultured in the presence or absence of doxicycline before preparing genomic DNA for the lacZ mutagenesis assay. Some cell cultures were exposed to oxidative cell culture stress and others were treated with 4-NQO to generate additional genotoxic stress (Fig. 3C)
. Treatment with 4-NQO resulted in a significantly higher MF in Myc-overexpressing B cells (MYC on) compared with normal B cells (MYC off) at 27 h (88.6 x 10-5 versus 41.8 x 10-5), indicating that deregulated Myc can sensitize B cells to undergo elevated mutagenesis in the presence of a strong mutagen. The difference was much smaller at 3 h (20.8 x 10-5 versus 18.6 x 10-5), suggesting that the mutation-enhancing effect of overexpressed Myc was dependent on time (cell division) to fix the mutations. Under conditions of relatively mild oxidative damage associated with cell culture stress, the MF in Myc-overexpressing B cells was lower relative to normal B cells: 11.9 x 10-5 versus 14.7 x 10-5 at 3 h and 17.2 x 10-5 versus 20.0 x 10-5 at 27 h. This result demonstrated that overexpressed Myc, on its own, is not a mutator gene in B cells.
Myc-dependent changes in lacZ mutant levels in fibroblasts were strikingly similar to those observed in B cells (Fig. 3D)
. In the presence of 4-NQO, the MF was higher (2.4-fold) in Myc-overexpressing fibroblasts relative to normal fibroblasts (Fig. 3D
, compare columns 4 and 2). In the absence of 4-NQO, the MF in Myc-overexpressing fibroblasts was lower (by 18%) than in normal fibroblasts (compare columns 3 and 1). The matching results in two different cell types (B cells and fibroblasts) indicated that deregulated Myc might act as a catalyst of chemically induced mutagenesis in many tissues. Furthermore, the mutation-enhancing ability of Myc might be fully compensated when genotoxic stress is low.
Mutant Levels and Patterns of Mutations in Transplanted PCTs.
Transgenic reporter genes are useful to evaluate whether in vivo propagation of primary tumors leads to run-away genomic instability (13
, 46, 47, 48, 49)
. To investigate this possibility, we serially transplanted (G1-G3) four primary PCTs (G0) into pristane-primed C mice and harvested cells 4–6 weeks posttransplantation. Pristane treatment of recipient mice was required for the outgrowth of transferred tumor cells, which are as dependent on the microenvironment of the pristane granuloma as are primary PCTs (50)
. The MF in G1 to G3 PCTs was remarkably stable (42.3–47.0 x 10-5; Fig. 4A
) and only slightly higher than the MF in primary PCTs (40.3 x 10-5). This demonstrated that mutant levels in PCTs were not determined by the length of stay in the granuloma (35 ± 7 days for transplanted PCTs versus 180 ± 45 days for primary PCTs). Furthermore, unlike primary PCTs in which mutations in granuloma cells may have contributed to the overall MF in the tumors, only mutations in tumor cells were measured in the transplanted PCTs. This added confidence to the contention that serial propagation of PCTs was not characterized by elevated genomic instability and that PCT mutant levels were governed mainly by intrinsic properties of the tumor cells.
|
]. R mutations were less prevalent in B cells exposed to PECs [44 (63.7%) of 69], in B cells from the Tet-off/MYC experiment [490 (54.5%) of 868], and in control tissues [145 (51%) of 284]. The prevalence of R mutations in normal tissues of C.pUR288 mice (51%) was higher than that reported for B6.pUR288 mice (42–45%; Refs. 14
, 52
), which may reflect genotypic differences in genomic stability control between strains C and B6. One of the genes that may contribute to the apparent proclivity of C mice to undergo R mutations is the weak allele of Prkdc encoding the catalytic subunit of the DNA double-strand break repair protein, DNA-PK, a key player in DNA double-strand break repair (53)
. The weak efficiency allele of DNA-PK has recently been shown to be associated with delayed genomic instability in B cells (54)
and increased lymphoma susceptibility (55)
. |
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
75%) in the tumor cells. To better understand this mutator phenotype, we evaluated the mutagenic consequences of inflammation and the deregulated Myc expression in B cells with the help of two dedicated in vitro model systems. The first model used the coincubation of B cells and PECs to mimic the mutagenic environment of the inflammatory granuloma. It showed that macrophages and neutrophils are able to mutagenize neighboring B cells, in part by releasing reactive oxygen intermediates. The second model used the Tet/off-MYC transgene of inducible Myc expression to mimic the effects of deregulated Myc on lacZ mutant levels in primary B cells. It demonstrated that deregulated Myc sensitizes B cells to undergo increased mutagenesis on treatment with 4-NQO. Although our studies with these model systems established a clear link between the natural history of pristane-induced PCT (inflammation and Myc) and elevated mutations in B cells, they did not suggest that the mutator phenotype of PCT was actually caused by inflammation or Myc. Some considerations leading to the alternative conclusion that the hypermutability of PCT is a complex phenotype that may be governed in part by obscure mutator genes in the tumor cells will be discussed in the following. A brief one-day coincubation of PECs with proliferating B cells in vitro resulted in a 3.7-fold increase in lacZ mutant levels in the B cells. The average increase in lacZ mutant levels in primary PCT was higher (5.4-fold) than in PEC-exposed B cells, yet the additional increase seemed negligible if one considers that the tumor cells accumulated mutations over an average period of 180 days (mean tumor onset). The steady, daily accumulation of 3.7-fold increases in MF over 180 days would lead, in theory, to a 666-fold increase in mutant levels in PCT, but this was obviously not the case. It thus appears that inflammatory cells have a much larger potential to cause mutations in B cells than is reflected by lacZ MF in plasma cell tumors. The great majority of B cells with mutations in lacZ may be eliminated in vivo, possibly by p53-dependent mechanisms of apoptosis, an important safeguard against Myc-driven B-cell tumors (56 , 57) . A recent study on MFs in pUR288 mice, deficient in the non-homologous end-joining pathway of DNA double-strand break repair (Ku80 null) but proficient for p53, supported the view that genomic integrity can be maintained in the face of increased mutagenesis, provided that the checkpoints of apoptosis are operational (58) . The remarkable stability of MF during serial propagation of PCT also argues against the hypothesis that chronic inflammation is the main determinant of lacZ mutant levels in the tumors. MFs in transplanted PCT (42.3–47.0 x 10-5) and primary PCT (40.3 x 10-5) were essentially the same, even though the transplanted PCT had a significantly shorter exposure to the mutagenic environment of the pristane granuloma (35 ± 7 days) than the transplanted PCT (180 ± 45 days). The apparent stability of lacZ MF in serially transplanted PCT was strong evidence that the hypermutability of PCT is neither determined by a high mutation rate in PCT nor by persistent exogenous damage inflicted by inflammatory cells. The hypermutability may be transient, e.g., as a consequence of acute up-regulation of Myc after chromosomal translocation, or caused by intrinsic features of the tumor cell, such as mutator genes.
The overexpression of Myc in Tet/off-Myc-pUR288 bitransgenic B cells doubled 4-NQO-induced mutations compared with B cells without overexpressed Myc. The mutation-enhancing effect of Myc can be added to a growing body of evidence that deregulated Myc can function as a mutator gene (reviewed in Refs. 43
, 44
). Some mechanisms by which Myc induces genetic instability and increased mutagenesis are currently emerging (21
, 22
, 24
, 26, 27, 28)
. However, our finding that overexpressed Myc in Tet/off-Myc-pUR288 B cells exposed to oxidative cell culture stress (59
, 60)
in the absence of 4-NQO was not accompanied by elevated mutagenesis indicated that deregulated Myc is not a mutator gene per se. In fact, the lacZ mutant levels in Myc-overexpressing B cells were lower than in the controls: 11.9 x 10-5 versus 14.7 x 10-5 after 3 h of cell culture and 17.2 x 10-5 versus 20.0 x 10-5 after 27 h. Lack of DNA damage was not the underlying reason, because normal B cells in cell culture (MYC off) showed a clear upward trend in lacZ MF over time: 9.66 x 10-5 at 0 h, 14.7 x 10-5 at 3 h, and 20.0 x 10-5 at 27 h. These results indicated that the impact of deregulated Myc on MF in B cells is variable and dependent on the type and intensity of the mutagenic attack. The pro-mutagenic potential of Myc may be fully restrained under conditions of moderate oxidative stress, possibly because of Myc-dependent adaptive responses that involve the up-regulation of key components of oxidative stress control, such as thioredoxin peroxidase, glutathione transferase, and lactate dehydrogenase. Lactate dehydrogenase is clearly activated in mouse plasma cell tumors (Supplementary Fig. 2
),6
leading to increased production of pyruvate, a scavenger of oxygen radicals (23
, 61)
. The ability of Myc to positively regulate DNA repair genes (62)
, including NBS1 (63)
involved in DNA double-strand break repair, may further contribute to harnessing the detrimental effects of Myc on the genome. These considerations suggest that although Myc may be involved in the PCT mutator phenotype, Myc is unlikely to be the sole cause for this phenotype.
In conclusion, mouse PCTs exhibit the increased mutability of a lacZ reporter gene, which is associated with, but presumably not caused by, chronic inflammation and activated Myc. The elevated lacZ mutant levels can be added to previously reported manifestations of genomic instability in PCTs, e.g., cytogenetic aberrations (64)
including Myc-activating chromosomal translocations (65)
, activation of intracisternal A particles (66)
, and DNA mutations due to deficiencies in transcription-coupled nucleotide excision repair (67)
and DNA double-strand break repair (53
, 54)
. Thus far, none of these changes has been elucidated with regard to the underlying molecular mechanism, and the relationship to deregulated Myc has also remained unclear. On the other hand, it has been shown that Myc-dependent mouse B-cell and plasma cell tumors can be associated with different mutator phenotypes. For example, Burkitt-like lymphomas of -MYC-pUR288 bitransgenic mice exhibited the same propensity to rearrangement mutations in lacZ as PCTs did, but demonstrated, in contrast to PCTs, a very small increase in lacZ MF. These findings illustrate the persistent difficulty of distinguishing mutator phenotypes associated with Myc from mutator phenotypes caused by Myc. Our newly developed model of mutation assessment in pUR288 transgenic mice and pUR288-Tet-off/MYC bitransgenic mice may be useful to elucidate this distinction.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org).
Requests for reprints: Siegfried Janz, Laboratory of Genetics, National Cancer Institute, Building 37, Room 3140A, Bethesda, Maryland 20892-4256. Phone: (301) 496-2202; Fax: (301) 402-1031; E-mail: sj4s{at}nih.gov
6 Supplementary data: http://canceres.aacrjournals.org. 
Received 8/20/03. Revised 10/ 9/03. Accepted 11/ 6/03.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
supF transgenic mice. Environ. Mol. Mutagen., 28: 459-464, 1996.[CrossRef][Medline]
ZAP/lacI shuttle vector. Environ. Mol. Mutagen., 18: 316-321, 1991.[Medline]
transgene target. Proc. Natl. Acad. Sci. USA, 93: 9073-9078, 1996.
/c-myc transgenic mouse model of accelerated hepatocarcinogenesis. J. Biol. Chem., 273: 15846-15853, 1998.LIZ mice. Cancer Res., 59: 3621-3626, 1999.lacZ mice. Cancer Res., 59: 2587-2592, 1999./c-myc transgenic mouse model of accelerated hepatocarcinogenesis. Lab. Investig., 83: 643-654, 2003.[Medline]
This article has been cited by other articles:
|
|
 |
|
  S. Ray, K. R. Atkuri, D. Deb-Basu, A. S. Adler, H. Y. Chang, L. A. Herzenberg, and D. W. Felsher MYC Can Induce DNA Breaks In vivo and In vitro Independent of Reactive Oxygen Species. Cancer Res., July 1, 2006; 66(13): 6598 - 6605. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
2 test). MF calculation was based on a total of 784 mutants (range, 40–299 mutants) scored among 2.1 x 106 wild-type colonies (range, 2.64 x 105 to 5.59 x 105). D, inhibition of PEC-mediated mutagenesis in B cells by the antioxidants, catalase [C (column 3)] or catalase plus pyruvate [C+P (column 4)]. Above the columns, residual lacZ MF (%) relative to TPA-stimulated PECs without antioxidants (100%, column 1). MF calculation was based on a total of 179 mutants (range, 20–76) scored among 8.07 x 105 wild-type colonies (range, 122 x 103 to 272 x 103). E, genetic defects in oxidative burst leading to diminished mutagenic potential of PECs. PECs deficient in p47phox, an essential component of active NADPH oxidase, are unable to produce superoxide (O2-) in the NADPH oxidase reaction. O2- is further reduced to the highly reactive hydroxyl radical (not shown) and reacts with nitric oxide (NO) to form peroxynitrite (ONOO-), which is highly toxic compared with O2- and NO 
