Jul 24, 1989 - irradiated female RF/J mice (The Jackson Laboratory, Bar. Harbor, Maine) between 4 ... provided by Donna George (personal communication).
Vol. 10, No. 1
MOLECULAR AND CELLULAR BIOLOGY, Jan. 1990, p. 405-408 0270-7306/90/010405-04$02.00/0 Copyright © 1990, American Society for Microbiology
Neutron Radiation Can Activate K-ras via a Point Mutation in Codon 146 and Induces a Different Spectrum of ras Mutations than Does Gamma Radiation STEVEN R. SLOAN, ELIZABETH W. NEWCOMB, AND ANGEL PELLICER* Department of Pathology and Kaplan Cancer Center, New York University School of Medicine,
New York, New York 10016 Received 24 July 1989/Accepted 3 October 1989
Neutron radiation is known to produce tumors in animals and cause ceH transformation. We have developed protocol to efficiently induce thymic lymphomas in RF/J mice by a single acute dose of neutron irradiation. Activated ras genes were detected in 17% (4 of 24) of the tumors analyzed. One of the tumors contained a K-ras gene activated by a point mutation in codon 146. Activating ras mutations at position 146 have not been previously detected in any known human or animal tumors. The spectrum of ras mutations detected in neutron radiation-induced thymic lymphomas was different from that seen in thymic lymphomas induced by gamma radiation in the same strain of mice. These results may have important implications for the mechanisms by which different types of radiation damage DNA.
a
in animals (23, 24). As yet, however, activation of specific oncogenes in neutron-induced tumors or cell transformants has not been described. Biologically significant consequences of the different actions of neutron and gamma radiation can be studied by comparing the activating point mutations of ras oncogenes that result from tumors induced by the different radiation treatments. To determine an optimal dose of neutron radiation that would induce thymic lymphomas with high frequency, we irradiated female RF/J mice (The Jackson Laboratory, Bar Harbor, Maine) between 4 and 7 weeks old with a single dose of 0.8, 1.0, 2.0, or 3.0 Gy of whole-body 0.44-MeV neutron radiation. Neutrons were produced from a Van de Graaff particle accelerator. Uniform whole-body neutron irradiation was delivered to mice restrained in Plexiglas cylinders and rotated around the neutron source. To correct for possible fluctuations, after one-half of the dose was administered, the tubes were rotated by 1800. The complete dose was administered in a period of 2 to 6 h. Mice were sacrificed when they appeared visibly ill and had palpable lymph nodes. The only effective dose was 1.0 Gy. All of the mice given 2.0 and 3.0 Gy of neutrons died within 2 weeks from radiation sickness. Only one of seven mice that received 0.8 Gy by neutrons developed a thymic lymphoma. Of 34 mice that received 1.0 Gy of neutrons, 24 developed thymic lymphomas, a 71% incidence. The latency of tumor induction averaged 170 23 days (mean standard deviation) (range, 134 to 230 days). Two assays were used to detect the presence of activated ras genes in the neutron-induced thymic lymphomas. Tumor DNA was screened in the nude mouse tumor assay (20) for the presence of transforming activity and independently screened for the presence of specific ras mutations by differential oligonucleotide hybridization to DNA amplified by the polymerase chain reaction (PCR) (17, 22). Six neutron-induced thymic lymphomas scored positive in the nude mouse assay. We wanted to compare the development of the 6 positive tumors with the development of the 18 negative tumors. The mean latency of the six neutroninduced lymphomas that were positive in the nude mouse assay was less than the mean latency of lymphomas that
Oncogene activation is frequently associated with the development of both human tumors and tumors in animal model systems (2). ras is mutated in approximately 20% of all human tumors, making it the most commonly activated oncogene (18). ras comprises a family of genes, three of which, H-, K-, and N-ras, have been implicated in carcinogenesis in mammals. Each ras gene codes for a similar protein of molecular mass 21 kilodaltons that is known as p21 (1). Activation of ras oncogenes has been shown to occur via point mutations in a limited number of codons. These codons are believed to code for regions of p21 that are important for the regulation of p21 activity. ras genes activated by somatic mutations in vivo have been most frequently detected in codons 12 and 61 and less frequently in codons 13 and 117 (1, 21). Animal model systems have provided a means by which oncogene activation can be studied in tumors in which the inducing agent and genetic background are well controlled (8, 16, 21). We have been studying ras oncogene activation in murine thymic lymphomas induced either by the chemical carcinogen N-nitroso-N-methylurea or by gamma radiation and have found activated K- and N-ras genes in a significant proportion of the tumors (7, 20). In this study, we wanted to determine the extent of ras oncogene activation in neutroninduced thymic lymphomas in RF/J mice and compare it with our previous analysis of gamma ray-induced thymic lymphomas in the same strain of mouse. Gamma and neutron radiation are two types of ionizing radiation that deliver energy to target molecules in very different ways (13). Unlike gamma radiation, neutron radiation is composed of particles that can directly interact with the nuclei of intracellular target molecules. Gamma rays, on the other hand, transfer their energy to orbital electrons of various intracellular molecules, creating free radicals that chemically react with DNA. Although both are ionizing radiations, neutron irradiation results in denser clusters of ionized molecules and atoms than does gamma irradiation. Neutron irradiation has been shown to transform cells in vitro (15, 19) and produce tumors, including thymic lympho*
mas,
±
Corresponding author. 405
±
406
NOTES
MOL. CELL. BIOL. TABLE 1. Primers used in PCR'
Gene
Exon
Primer
Sequence
K-ras
1
5' 3'
2
5' 3' 5' 3' 5' 3' 5' 3' 5' 3' 5' 3'
5'-TTTTATTGTAAGGCCTGCTG-3' 5'-TTTACAAGCGCACGCAACTG-3' 5'-CCAGACTGTGTTTCTCCCTT-3' 5'-TGCCAACTTTCTTATTCAAC-3' 5'-GTTTCTTCCCCAGAGAACAA-3' 5'-TCGTCAACACCTGTCTTGTC-3' 5'-TACAATGCAGAGAGTGGAGG-3' 5'-ACTTACCAGATTACATTATA-3' 5'-TATATTTCAGGGTGTTGACG-3' 5'-GTGTGCCTTAAGAAAGAGTA-3' 5'-GCCCTTGAAGTATTGTAGGT-3' 5'-CAAAGTGAGGATAAGGGCCA-3' 5'-CACCACCACCTCCTCACTCT-3' 5'-GCACTTGTTCTGGCCTCCAC-3'
3
4a
4b N-ras
1
2
a The PCR reaction was performed as described previously (22). The sequences of primers for exons 3, 4a, and 4b of K-ras were derived from published sequences (10, 12). Primer sequences for both N-ras exons and exon 1 of K-ras were obtained from intronic regions that were previously sequenced by I. Guerrero in this laboratory (unpublished data). Primer sequences for exon 2 of K-ras were obtained from intronic sequences provided by Donna George (personal communication).
scored negative (152 + 9 days versus 178 ± 23 days; P < 0.025). Hence, transforming activity detected in the nude mouse assay may be related to genes that are associated with the more rapid development of some neutron-induced tumors. Southern blots were used to determine whether additional copies of the H-, K-, or N-ras gene were present in nude mouse tumor DNA. Amplification of ras gene sequences in DNA from a nude mouse tumor indicates that the transforming activity responsible for the nude mouse tumor is associated with that particular gene (11, 20). This analysis demonstrated that DNAs from three neutron-induced thymic lymphomas contained transforming genes identified as Kras, DNA from one thymic lymphoma contained transforming activity associated with N-ras, and transforming activities detected in two thymic lymphomas were not associated with the presence of K-, H-, or N-ras sequences. For each primary thymic lymphoma in which transforming activity was associated with a particular ras gene, all of the nude mouse tumors independently derived from that thymic lymphoma showed amplification of the same ras gene. In the second assay, tumors were tested for the presence of ras mutations by differential oligonucleotide hybridization to PCR-amplified DNA. In this technique, exons 1 and 2 of K- and N-ras from each neutron-induced tumor were amplified by PCR, using the primers listed in Table 1. The resulting amplified DNA samples were slot blotted. The slot blots were hybridized to 32P-labeled oligonucleotides specific for each relevant point mutation. After the filters were washed, autoradiography revealed that no labeled probe remained bound to slots containing PCR-amplified normal brain DNA. Presence of labeled probe bound to a tumor sample after washing indicated that the tumor sample contained a sequence that was exactly complementary to the oligonucleotide probe and therefore contained that ras point mutation. This differential oligonucleotide hybridization technique was used to screen all of the neutron-induced thymic lymphomas for all possible point mutations in codons 12, 13, and 61 of K- and N-ras that could lead to amino acid substitutions. This analysis revealed that 3 of 24 neutron-induced thymic lymphomas contained mutated ras genes. Two lym-
phomas contained different mutations in codon 12 of K-ras (GGT-sTGT and GGT--GTT), and one lymphoma contained a mutation in codon 61 of N-ras (CAA-sAAA) (results summarized in Table 2). DNA from these three thymic lymphomas had induced ras-positive nude mouse tumors. Differential oligonucleotide hybridization confirmed that these nude mouse tumors contained the same ras mutations as did their respective thymic lymphomas. These results confirmed the sensitivity of the nude mouse assay for detection of activated ras oncogenes, since all neutroninduced thymic lymphomas containing detectable ras mutations scored positive in the nude mouse assay. One additional neutron-induced lymphoma and its three independent nude mouse tumors positive by Southern blot analysis for K-ras were negative for all possible codon 12, 13, and 61 point mutations. To determine the mutation in this activated ras gene, each of the five coding exons of K-ras (1, 2, 3, 4A, and 4B) from one of the nude mouse tumors was separately amplified by PCR. Each of the amplified exons was then sequenced. Only one mutation was detected, an ACA-for-GCA substitution in codon 146 (Fig. 1). This mutation, which results in the substitution of the amino acid threonine for the normal alanine, has never been detected in any tumor or in any system reported so far. Differential hybridization of oligonucleotides to amplified DNA samples confirmed that the mutation in codon 146 was present not only in the three nude mouse tumors derived from the neutron-induced thymic lymphoma but also in the thymic lymphoma itself (Fig. 1). Transformation of NIH 3T3 mouse fibroblasts is a biological assay that is frequently used to assess the activity of ras mutations. The activity of the 146 mutation in the NIH 3T3 focus-forming assay was determined by transfecting DNA from one of the three positive nude mouse tumors into NIH 3T3 cells as described elsewhere (20). The focus-forming activity of the 146 mutation was compared with the activity of DNA from two nude mouse tumors, one containing a K-ras gene mutated in codon 12 (GGT-*GAT) and the other containing a K-ras gene mutated in codon 13 (GGT--GAT). Densitometric analysis of Southern blots showed that the nude mouse tumor containing the K-ras gene mutated in codon 146 had more copies of integrated K-ras gene than did the other tumors used in the focus-forming assay. Whereas the negative control (normal murine thymus DNA) yielded no foci, DNA from the nude mouse tumor with the activating mutation at K-ras position 146 yielded 0.09 focus per ,ug of DNA. This was less than the number of foci induced by DNA from nude mouse tumors activated by a mutation in K-ras codon 12 (0.29 focus per ,ug) or K-ras codon 13 (0.18 focus per ,ug). This result indicates that the mutation in codon 146 of K-ras has weaker transforming efficiency than do other activating mutations, especially when compared with the most common K-ras codon 12 mutation in this system. This weaker transforming activity was also demonstrated by the fact that the nude mouse tumors containing TABLE 2. ras gene point mutationsa Gene
Codon
Mutation
GGT--IGT (Gly- Cys) GGT--GTT (Gly-WVal) GCA--ACA (Ala--Thr) N-ras CAA--*AAA (Gln--Lys) a Screened by PCR amplification, oligonucleotide hybridization, and DNA K-ras
sequencing (see text).
K12 K12 K146 N61
0t-
NOTES
VOL. 10, 1990
C'
PCR-amplified DNA. None of these thymic lymphomas contained the mutation in codon 146 (data not shown). The evidence presented in this report indicates that the mutation in codon 146, replacing threonine for alanine, activates the transforming potential of the c-K-ras gene. The fact that DNA from the neutron-induced lymphoma containing the mutation in codon 146 generated three independent nude mouse tumors containing additional copies of K-ras demonstrates that the transforming activity segregated with the K-ras gene in these tumors. Focus formation in NIH 3T3 cells verified the transforming ability of the activated gene. Furthermore, sequencing data showed that no other mutations were present in any other coding regions of the K-ras
~.
2
3
qm
2 3
FIG.
1. Analysis of codon 146 ras mutation. Slot blot analysis performed essentially as described previously (17), and DNA sequencing was performed as described by Higuchi et al. (14). (A) Sequence of a portion of the exon 3 of K-ras of normal RF/J brain DNA (left) and nude mouse tumor DNA (right). The normal was
sequence in codons 145 to 147,
with TCAACAAAG in the nude of a slot blot of K-ras
exon
3
TCAQjCAAAG, mouse tumor.
amplified
has been
replaced
(B) Autoradiograph
DNA of the three nude
mouse
tumors derived from one neutron-induced
thymic lymphoma (left slot of row 1 and right two slots of row 2). The tumor represented in the middle slot of row 2 is the one whose sequence is shown in panel A. Negative controls consisting of DNA from RF/J brain and nude mouse
tumors
derived from other ras-activated neutron-induced
thymic lymphomas are in the left slot of row 2 and the three slots of row 3, respectively. The slot blot was hybridized to a 32P-labeled oligonucleotide specific for the codon 146 mutation (TTGTC1TTTGT TGAGOTCTC) that was sequenced in panel A. (C) Autoradiograph of a slot blot of K-ras exon 3 amplified DNA from six neutroninduced tumors (the three slots in each of rows 1 and 2) and normal RF/J brain DNA (left slot of row 3) hybridized to the 32P-labeled oligonucleotide specific for the K-ras 146 mutation. The only positive sample is DNA from the neutron-induced tumor that induced the nude mouse tumors containing the codon 146 mutation (row 2, left slot).
the K-ras gene mutated in codon 146 had
a longer mean containing ras gene activated in codon 12 or 61 (21.7 ±t 3.5 days versus 16.5 ±+ 1.6 days; P < 0.005). In a recent report, an in vitro-mutagenized H-ras
latency
than did tumors
with
gene
a
mutation
in
codon
146
also
showed
biological activity than did a gene with a codon (9). Similar results were previously reported for
lower
12 mutation some
N-ras
codon 13 mutations (3). To compare the effects of gamma and neutron irradiation
damage
to codon
146,
we
analyzed DNA from 24
neutron-
induced tumors and DNA from 25 thymic lymphomas induced
by
gamma radiation in RF/J mice for the presence of
the 146 mutation. These gamma radiation-induced
lymphopreviously been screened for mutations in codons 12, 13, and 61 (7). These same tumor DNA samples were screened by differential oligonucleotide hybridization to
mas
had
407
gene. Codons 145 to 147 of p21 are well conserved, suggesting an important function for this region. A Ser-Ala-Lys consensus sequence is present in N-, K-, and H-ras genes in mammals and in the ras genes of drosophila and chickens, among others (1). Structural analogies with other G proteins and X-ray analysis of crystallized p21 predict that this region of p21 interacts with the purine structure of GTP and GDP (6). Furthermore, random in vitro mutagenesis has produced an H-ras mutant in codon 146 with increased guanine nucleotide exchange rates (9). This H-ras mutation transforms NIH 3T3 cells with weaker potency than do other mutations at codon 12 or 61. Neutron radiation, as opposed to gamma radiation, leads to a diversity of ras point mutations. In thymic lymphomagenesis, gamma radiation led to 24% (9 of 37) ras activation in tumors (7). Of the ras-activated tumors, 89% (eight of nine) were K-ras activated, and 87.5% (seven of eight) of these had a substitution of GAT for GGT in codon 12 (7). Although neutrons result in a lower incidence of ras mutations (4 of 24 [17%]), they have not led to a preponderance of any one particular mutation and have not resulted in the most common mutation seen in gamma radiation-induced tumors. A comparison of the frequency of the codon 12 GAT-for-GGT substitution in K-ras-activated tumors induced by gamma radiation and neutron radiation produces a P value of less than 0.05 (chi-square test). Several explanations might account for the different spectra of ras mutations produced by different forms of radiation. The preponderance of one mutation resulting from gamma radiation was probably not due solely to biological selection for that mutation, since many other mutations are biologically active, as shown in this report. Relative inefficiency in repairing a gamma radiation-induced lesion at one particular base might account for the high frequency of a gamma radiation-induced mutation at that base. In Escherichia coli, the efficiency of the UvrABC excision repair system is known to be affected by the neighboring base sequence (5). In mammalian cells, lesions caused by gamma radiation are often effectively repaired (4), but it is not known whether repair occurs with equal efficiency at all sites. It has been suggested that densely ionizing radiations such as neutrons produce clusters of lesions in DNA molecules that would be difficult to repair (25). Alternatively, neutron-induced lesions might be more difficult to repair than gamma ray-induced lesions because of qualitative differences in the types of lesions. If either hypothesis were correct, neutron-induced lesions at most sites would be more difficult to repair than gamma ray-induced lesions at those sites. Hence, neutroninduced mutations would be detected in several different sites, whereas gamma ray-induced mutations would be detected only in those sites that are relatively refractory to repair. Also, neutrons would be more likely to create various
408
NOTES
types of damage throughout the genome that could be tumongenic. Most DNA damage due to gamma radiation is mediated through the production of free radicals (4). The conformation of DNA and DNA-protein interactions could protect many regions of DNA from damage created by free radicals but would be less effective in protecting from direct damage produced by neutrons. The second base of ras codon 12 could be vulnerable to attack by free radicals and hence increase the appearance of a mutation in this base in gamma radiation-induced tumors. This report describes a novel mutation in a K-ras-activated oncogene affecting codon 146, a very well conserved region in the p21 molecule that has been previously found to affect nucleotide binding (9) and is in a loop that interacts with the purine in the X-ray diffraction model (6). This mutation has weak transforming activity in the NIH 3T3 focus-forming assay but was readily detectable in the nude mouse assay. The spectrum of ras mutations in thymic lymphomas induced by neutrons is very different from the one observed in the same mouse strain with gamma radiation. This result provides experimental evidence at the DNA level which relates the different physical properties of neutron and gamma ionizing radiation with different activating mutations in ras oncogenes. We thank S. Marino of the Radiological Research Accelerator Facility at Nevis Laboratories, Columbia University, for performing the dosimetry and operating the accelerator used to irradiate the mice. We also thank B. Goldschmidt for oligonucleotide synthesis, J. M. Kahn for excellent technical assistance, Natalie Little for preparing the manuscript, and G. W. Teebor for discussions and critical reading of manuscript. This investigation was supported by Public Health Service grants CA36327, ES03847, and CA40533 from the National Institutes of Health and by Public Health Service training grant 5 T32 GM07308 from the National Institute of General Medical Sciences. A.P. is a scholar of the Leukemia Society of America.
1. 2. 3.
4.
5.
6.
7.
LITERATURE CITED Barbacid, M. Ras genes. 1987. Annu. Rev. Biochem. 56: 779-827. Bishop, J. M. 1987. The molecular genetics of cancer. Science 235:305-311. Bos, J. L., D. Toksoz, C. J. Marshall, M. Verlaan-de Veries, G. H. Veeneman, A. J. van de Eb, J. H. van Boom, J. W. G. Janssen, and A. C. M. Steenvoorden. 1985. Amino-acid substitutions at codon 13 of the N-ras oncogene in human acute myeloid leukemia. Nature (London) 315:726-730. Breimer, L. H. 1988. Ionizing radiation-induced mutagenesis. Br. J. Cancer 57:6-18. Burns, P. A., A. J. E. Gordon, K. Kunsmann, and B. W. Glickman. 1988. Influence of neighboring base sequence on the distribution and repair of N-ethyl-N-nitrosourea-induced lesions in Escherichia coli. Cancer Res. 48:4455-4458. DeVos, A. M., L. Tong, M. V. Milburn, P. M. Matias, J. Jancarik, S. Noguchi, S. Nishimura, K. Miura, E. Ohtsuka, and S. Kim. 1988. Three-dimensional structure of an oncogene protein: catalytic domain of human c-H-ras p21. Science 239: 888-893. Diamond, L. E., I. Guerrero, and A. Pellicer. 1988. Concomitant
MOL. CELL. BIOL.
K- and N-ras gene point mutations in clonal murine lymphomas. Mol. Cell. Biol. 8:2233-2236. 8. Duran-Reynals, M. L., F. Lilly, A. Bosch, and K. J. Blank. 1978. The genetic basis of susceptibility to leukemia induction in mice by 3-methylcholathrene applied percutaneously. J. Exp. Med. 147:459-469. 9. Feig, L. A., and G. M. Cooper. 1988. Relationship among guanine nucleotide exchange, GTP hydrolysis, and transforming potential of mutated ras proteins. Mol. Cell. Biol. 8:2472-2478. 10. George, D. L., A. F. Scott, S. Trusko, B. Glick, E. Ford, and D. J. Dorney. 1985. Structure and expression of amplified cKiras gene sequences in Y1 mouse adrenal tumor cells. EMBO J. 4:1199-1203. 11. Guerrero, I., P. Calzada, A. Mayer, and A. Pelficer. 1984. A molecular approach to leukemogenesis: mouse lymphomas contain an activated c-ras oncogene. Proc. Natl. Sci. USA 81: 202-205. 12. Guerrero, I., A. Villasante, V. Corces, and A. Pellicer. 1985. Loss of the normal N-ras allele in a mouse thymic lymphoma induced by a chemical carcinogen. Proc. Natl. Acad. Sci. USA 82:7810-7814. 13. Hail, E. J. 1988. Radiobiology for the radiologist, 3rd ed., p. 10-15. J. B. Lippincott, Philadelphia. 14. Higuchi, R., C. H. von Beroldingen, G. F. Sensabaugh, and H. A. Erlich. 1988. DNA typing from single hairs. Nature 332:543546. 15. Hill, C. K., B. A. Carnes, A. Han, and M. M. Elkind. 1985. Neoplastic transformation is enhanced by multiple low doses of fission spectrum neutrons. Radiat. Res. 102:404 410. 16. Ishizaka, S. T., and F. Lilly. 1987. Genetically determined resistance to 3-methylcholanthrene-induced lymphoma is expressed at the level of bone marrow-derived cells. J. Exp. Med. 166:565-570. 17. Liu, E., B. Hjelie, R. Morgan, F. Hecht, and J. M. Bishop. 1987. Mutations of the Kirsten-ras proto-oncogene in human preleukemia. Nature (London) 330:186-188. 18. McKay, I. A., H. Paterson, R. Brown, D. Toksoz, C. J. Marshall, and A. Hall. 1986. N-ras and human cancer. Anticancer Res. 6:483-490. 19. Miller, R. C., D. J. Brenner, C. R. Geard, K. Komatsu, S. A. Marino, and E. J. Hall. 1988. Oncogenic transformation by fractionated doses of neutrons. Radiat. Res. 114:589-598. 20. Newcomb, E. W., J. J. Steinberg, and A. Pellicer. 1988. Ras oncogenes and phenotypic staging in N-methylnitrosourea- and gamma-irradiation-induced thymic lymphomas in C57BL/6J mice. Cancer Res. 48:5514-5521. 21. Reynolds, S. H., S. J. Stowers, R. M. Patterson, R. R. Maronpot, S. A. Aaronson, and M. W. Anderson. 1987. Activated oncogenes in B6C3F1 mouse liver tumors: implications for risk assessment. Science 237:1309-1316. 22. Saiki, R. K., S. Scharf, F. Faloona, K. B. Mullis, G. T. Horn, H. A. Erlich, and N. Arnheim. 1985. Enzymatic amplification of beta-globin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia. Science 230:1350-1354. 23. Ullrich, R. L., M. C. Jernigan, G. E. Cosgrove, L. C. Satterfield, N. D. Bowles, and J. B. Storer. 1976. The influence of dose and dose rate on the incidence of neoplastic disease in RFM mice after neutron irradiation. Radiat. Res. 68:115-131. 24. Upton, A. C., M. L. Randolph, J. W. Conklin, M. A. Kastenbaum, M. Slater, G. S. Melville, Jr., F. P. Conte, and J. A. Sproul, Jr. 1970. Late effects of fast neutrons and gamma-rays in mice as influenced by the dose rate of irradiation: induction of neoplasia. Radiat. Res. 41:467-491. 25. Ward, J. F. 1985. Biochemistry of DNA lesions. Radiat. Res.
104(Suppl. 8):S103-S111.
