Mutagenesis vol.16 no.1 pp.17–23, 2001
Heritable effects of paternal irradiation in mice on signaling protein kinase activities in F3 offspring
Janet E.Baulch1, Otto G.Raabe and Lynn M.Wiley Institute of Toxicology and Environmental Health, Old Davis Road, University of California, Davis, CA 95616, USA
We evaluated F3 mouse offspring from paternal F0 attenuated 137Cs γ-irradiation (1.0 Gy) for heritable effects on gene products that can modulate cell proliferation rate and that may be markers for genomic instability. The F3 generation was selected for evaluation as a stringent test for heritability of effects from paternal F0 germline irradiation. Male CD1 mice were bred 6 weeks after irradiation so that the fertilizing sperm were type B spermatogonia at the time of irradiation. The resulting F1 males were bred to CD1 females to produce F2 four-cell embryos. The F2 embryos with a radiation history were paired with ‘control’ CD1 four-cell embryos that were heterozygous for the neo transgene. These F2 XY–XY chimeras, consisting of cells derived from both an embryo with a paternal F0 radiation history and a control embryo, were transferred to foster mothers, raised to adulthood and bred to produce F3 offspring. F3 offspring were evaluated for hepatic activities of receptor tyrosine kinase, protein kinase C and MAP kinase and for protein levels of nuclear p53 and p21waf1. All three protein kinase activities were altered and nuclear levels of p53 and p21waf1 protein were higher in the group of offspring that included F3 offspring with a paternal F0 radiation history than in littermates in the neo-positive control group. To our knowledge, this is the first observation in the descendants of paternal germline irradiation of effects on signal protein kinase activities and downstream nuclear target proteins that can influence cell proliferation rates.
Introduction In previous studies we have used the preimplantation embryo chimera assay to demonstrate adverse effects of acute paternal whole-body irradiation on progeny embryos. The assay measures the competitive cell proliferation disadvantage of an embryo with a radiation history when it is challenged by direct cell–cell contact with a normal embryo in an aggregation chimera (Obasaju et al., 1988, 1989; Warner et al., 1991; Straume et al., 1993, 1997; Wiley et al., 1994a,b, 1997; Peters et al., 1996). Competitive cell proliferation disadvantage of progeny embryos has been demonstrated following acute paternal whole body exposure to low LET radiation [x-rays (Obasaju et al., 1989) and γ-rays (Warner et al., 1991)] or high LET radiation (512 MeV/amu 56Fe; Wiley et al., 1994a) with absorbed doses of 0.01–1.0 Gy. The transmission of competitive cell proliferation disadvantage to the F1 generation and its heritability by the F2 generation is highly dependent upon the time interval separating the acute irradiation of the F0 male and conception of his F1 offspring 1To
(Obasaju et al., 1989; Warner et al., 1991). The F1 embryos that are conceived 6–7 weeks after paternal F0 irradiation are the most likely to have competitive cell proliferation disadvantage, and they exhibit these effects following paternal doses as low as 0.005 or 0.01 Gy (Warner et al., 1991). These F1 embryos are produced by sperm that were type B spermatogonia at the time of irradiation. The F1 offspring that are conceived 6–7 weeks after paternal irradiation subsequently transmit competitive cell proliferation disadvantage to the F2 generation (Wiley et al., 1997). Ionizing radiation can induce changes in the activities of a number of protein kinases in the cell signal transduction cascades that modulate cell cycle length, cell differentiation and genome stability. Among these signaling kinases is the calcium/phospholipid-dependent serine/threonine-specific protein kinase C (PKC), which moves from the cytosol to the plasma membrane of irradiated cells (Nakajima and Yukawa, 1996) and increases in activity and in mRNA levels (reviewed in Woloschak et al., 1990). Radiation-induced increases in PKC activity can lead to increased levels of the transcription factor p53 (Kanna and Lavin, 1993) and, as a consequence, increased production of p21waf1, which causes arrest or slowing of the cell cycle (Xiong et al., 1993; el-Deiry et al., 1995). In addition, the autophosphorylation activity of epidermal growth factor (EGF) receptor (Schmidt-Ullrich et al., 1996) and the phosphorylation activity of mitogen activated protein kinases (e.g. MAPK; Stevenson et al., 1994) also increase in response to irradiation and can affect cell cycle length. These alterations in gene expression and signal transduction molecule activities, whether initiated by external signals such as growth factors or cellular stress (e.g. ionizing radiation), can promote genetic reprogramming, which may predispose cells to genomic instability (reviewed in Morgan et al., 1996). Addition to the culture medium of growth factors that bind to insulin-like growth factor 1 (IGF1) receptor prevents competitive cell proliferation disadvantage in embryo chimera assays, possibly by stimulating receptor cross-talk between the IGF1 receptor and the EGF receptor (Peters et al., 1996). Adding growth factors to the culture medium that bind to the EGF receptor can prevent the lag in blastocyst development that is exhibited by irradiated four-cell embryos in vitro (Peters et al., 1996). These observations suggested to us that the cell proliferation and cell differentiation/developmental effects that are exhibited by F1 and F2 embryos with a radiation history might be accompanied by heritable alterations in signal transduction kinase activities that are modulated by growth factor receptor activation. The activities of the signaling kinases PKC, MAPK and receptor tyrosine kinase (RTK) were evaluated to test this hypothesis. EGF RTK activity was investigated because it is stimulated by extracellular growth factors that are produced by preimplantation embryos. These growth factors include transforming growth factor α (TGFα) (Rappolee et al., 1988) and amphiregulin (Tsark et al., 1997). PKC and MAPK
whom correspondence should be addressed. E-mail:
[email protected]
© UK Environmental Mutagen Society/Oxford University Press 2000
17
J.E.Baulch, O.G.Raabe and L.M.Wiley Mice The experimental F0 parental generation was a cohort of 8–12 week old CD1 mice (Charles River, Portage, MI). All mice were maintained under a 14 h light/10 h dark photoperiod. Male mice were irradiated as described below. Female mice were induced to superovulate by an intraperitoneal injection of 5 IU of pregnant mare serum gonadotropin (PMSG) followed 48 h later with intraperitoneal injection of 5 IU of human chorionic gonadotropin (hCG). Superovulated females were placed with F0 males overnight to allow mating. Pregnant females were allowed to deliver the F1 litters. Similar procedures for female superovulation were used prior to mating with F1 males, but in this case the females were killed for the two-cell and four-cell F2 embryos 50–52 h after hCG injection. Male and female C57Bl/6J mice that were homozygous for the neo transgene were purchased from Jackson Laboratories (Bar Harbor, MA). This transgene encodes neomycin phosphotransferase II and was a component of a gene ablation cassette for CD4 (Rahemtulla et al., 1991). The transgene was backcrossed on to the CD1 background and these mice were bred to provide control embryos heterozygous for the neo transgene. Recipient CD1 females (foster mothers) for chimera embryos were treated with an intraperitoneal injection of 0.5 IU of PMSG followed 48 h later with intraperitoneal injection of 0.5 IU of hCG and placed with vasectomized males overnight. Blastocyst-stage chimeric embryos were transferred to the uterus of these pseudopregnant females ~76 h after hCG injection.
Fig. 1. Procedures used to produce neo–/– F3 offspring and neo⫹/– control littermates from adult F3 XY←→XY chimeras. F0 CD1 male mice were exposed to 1.0 Gy of attenuated 137Cs γ-rays and mated to non-exposed CD1 females during the sixth week after paternal F0 irradiation to produce experimental F1 animals with a radiation history. The F1 animals with a radiation history were mated to unirradiated CD1 females to obtain experimental F2 embryos with a paternal F0 radiation history. These experimental four-cell embryos were paired with unirradiated control CD1 four-cell embryos that were heterozygous for the neo cell lineage marker (neo⫹/–). The resulting chimeras were cultured to the blastocyst stage and transferred to a foster mother. The pups resulting from the transfer were screened by PCR to distinguish the XY←→XY (male) chimeric pups from the XX←→XX and XX←→XY chimeric pups. Mature XY←→XY germline chimeras produce control gametes (neo⫹ and neo–) and gametes with a paternal F0 radiation history (neo–) from the same gonad. By breeding these experimental animals to normal females, control offspring (neo⫹/– and neo–/–) and experimental F3 offspring with a paternal F0 radiation history (neo–/–) were obtained as littermates from the same dam. Gender-matched littermate pairs of neo–/– F3 and neo⫹/– control offspring were evaluated for hepatic PKC, MAPK and GST activities and p21waf1 and p53 protein levels.
activities were investigated because they respond independently to the reactive oxygen intermediates (ROIs) that are produced by ionizing radiation (Stevenson et al., 1994) and because they initiate other signaling pathways that lead to nuclear radiation responses including genomic instability. Due to the extremely small size and limited number of mouse preimplantation embryos, we allowed the embryos to develop to term to provide sufficient amounts of tissue for an evaluation of these signaling molecules. To test rigorously the hypothesis for heritability of changes in signaling protein kinase activities and levels of p53 and p21waf1, we chose to examine the F3 offspring of the F2 generation. Materials and methods Overview In the work described below, the CD1 F1 male mice that were conceived 6 weeks after paternal F0 irradiation were bred at 8 weeks of age to provide F2 four-cell embryos (Figure 1). These F2 four-cell embryos were paired with CD1 control four-cell embryos to make chimeras. The control embryos were heterozygous (⫹/–) for the neo transgenic cell lineage marker. The chimeric embryos were transferred to foster mothers to provide adult F2 XY←→XY germline chimeras which were raised to adulthood and bred to provide the F3 offspring from paternal F0 irradiation and littermate control offspring that were evaluated in the present study.
18
Irradiation and post-irradiation breeding to produce F1 males and their F2 embryos Five CD1 male mice were treated with an absorbed dose of 1.0 Gy from acute whole-body attenuated 137Cs γ-irradiation. Although 0.01 Gy of attenuated 137Cs γ-irradiation has been found to be sufficient for the expression of competitive cell proliferation disadvantage by F1 four-cell embryos, we chose the higher dose of 1.0 Gy in these experiments because of the uncertainty about the persistence of any potential cell signaling effects across successive generations of offspring. Mice were irradiated using calibrated 137Cs γ-ray (0.662 MeV) irradiator (mark I, model 30; J.L.Shepherd & Associates, San Fernando, CA) at a dose rate of 0.14 Gy min–1. The actual exposure for the group of mice was measured using three commercially supplied thermoluminescent dosimeters (TLD-100 LiF powder; Englehard Corp., Harshaw, OH; supplied and read by Radiation Detection Co., Sunnyvale, CA). Air-dose dosimeter measurements were converted to tissue-absorbed dose estimates using the ratio of the energy mass absorption coefficients of γ photons for tissue and air. Irradiated F0 males were subsequently mated to unirradiated, superovulated CD1 females at 7 day intervals for 7 weeks after irradiation with the first matings 4 days after irradiation. The mated females from 6 weeks after paternal F0 irradiation were allowed to deliver their F1 litters. Each litter contained four to six F1 males so that the grouped pool consisted of 21–25 F1 males from a paternal F0 radiation history. Twelve adult F1 males were randomly selected from the pool to provide 10 animals representing the five F0 sires for mating and two back-up animals. Beginning at 8 weeks of age, the F1 males were mated to superovulated females to provide the experimental F2 four-cell embryos from a paternal F0 radiation history. New female CD1 mice were obtained from the vendor for each generation’s matings in order to preclude inbreeding. Construction of F2 XY–XY chimeric mice and their breeding with unirradiated CD1 females to obtain F3 offspring The F2 embryos were tracked with respect to their F1 sire to prevent any given F1 male from dominating the pool of F2 chimeras. These chimeras consisted of one F2 embryo from a paternal F0 radiation history and one control embryo. The control CD1 embryo was heterozygous for the neo transgene (neo⫹/–). Offspring resulting from the transfer of experimental chimeras to recipient females were screened by polymerase chain reaction (PCR) amplification of tail DNA (Peippo and Bredbacka, 1995) to determine the presence of the neo cell lineage marker and the sex of each animal (possible outcomes were XX←→XX, XX←→XY and XY←→XY chimeric mice). We obtained five neo-positive F2 XY←→XY chimeric mice from three different F1 sires with a paternal F0 radiation history. These five F2 chimeras were raised to maturity and bred to superovulated, unirradiated CD1 females to obtain neo–/– offspring, which included the F3 offspring with a paternal F0 irradiation history and some littermate control offspring without radiation history, and neo⫹/– littermate control offspring (Figure 1). Testing of transgenic animals XY←→XY chimeric mice comprised cells from one wild-type CD1 embryo and either one CD1 embryo homozygous for the neo transgene (neo⫹/⫹) or one CD1 embryo heterozygous for the neo transgene (neo⫹/–) were constructed in preliminary experiments. Starting at 7 weeks of age these chimeric male mice were bred once a week for as many as 14 weeks. For each male, the percentage of neo-positive offspring per litter was tracked over time, as a
Heritable effects on signaling kinase activities
Table I. Basal hepatic protein kinase activity levels in 19 day old conventionally bred CD1 mice of the neo⫹/⫹ genotype and the neo–/– genotypea Kinase
RTK PKC MAPK
neo–/– genotype
neo⫹/⫹ genotype
n
basal activityb
n
basal activityb
5 5 5
0.81 ⫾ 0.03 0.40 ⫾ 0.04 1.46 ⫾ 0.05
5 5 5
0.81 ⫾ 0.04 0.40 ⫾ 0.04 1.50 ⫾ 0.07
a
No significant differences for all three kinases in comparison of CD1 mice of the neo⫹/⫹ genotype and CD1 mice of the neo–/– genotype. b Mean picomoles of 32P per nanomole of substrate per 20 µg protein ⫾ SEM. measure of stability of the transgenic cell lineage marker in the chimeric germline. We found that the neo transgene was selected against over time in the germline of the neo⫹/⫹ chimeric males. As a result of this chimeric drift, the germline neo marker was lost over time in neo⫹/⫹ chimeric males and eventually no neo-positive offspring were observed in any litters (data not shown). In contrast, the percentage of neo-positive offspring did not change over time in litters sired by neo⫹/– chimeric males. Based on these observations, the neo transgene was only used in the heterozygous state in this study. As a result, only half of the control offspring carried the neo marker and could be identified. The remaining, neo–/–, control offspring were indistinguishable from the neo–/– F3 offspring with the paternal F0 radiation history and were grouped together in the analysis. To evaluate the effect of the neo marker on signaling protein kinase activity, the mean activities of RTK, PKC and MAPK were evaluated in liver tissues from neo⫹/⫹ and neo–/– offspring obtained from conventional matings. Five CD1 female mice of each genotype were induced to superovulate by intraperitoneal injection of 5 IU of PMSG followed 48 h later with intraperitoneal injection of 5 IU of hCG. Superovulated females were placed with CD1 males of the same genotype (neo⫹/⫹ or neo–/–) overnight to allow mating. At 19 days of age, the neo⫹/⫹ and neo–/– offspring obtained from these matings were killed and their livers were frozen in liquid nitrogen. One liver from each of the 10 litters was selected for analysis so that both sexes were represented and matched between the corresponding neo⫹/⫹ and neo–/– groups of offspring. The mean activities of hepatic RTK, PKC and MAPK were not different and were comparable to the kinase activity levels in the neo⫹/– group of control offspring that were produced by the F2 XY←→XY chimeras (Table I). These data suggest that the presence of neo had no effect on the activities of the protein kinases that were evaluated for the offspring produced by the F2 XY←→XY chimeras. Evaluation of liver:body weight ratios At 19 days of age, the F3 offspring were killed and the mean liver weight:body weight ratios were calculated as a measure of hepatomegaly. Because the litters differed in the number of offspring, it was necessary to normalize liver size against body weight as body weight can vary considerably according to the number of offspring that are being nursed by a given dam. The livers were then frozen in liquid nitrogen. Protein isolation For all protein assays, cytosolic and nuclear protein extracts were prepared as described in Enan and Matsumura (1995b). The stored liver tissue was weighed and homogenized in a volume of TEDG buffer (pH 7.4) proportional to that weight. The buffer was composed of 25 mM Tris, 1 mM EDTA, 1 mM dithiothreitol (DTT), 10% glycerol and protease inhibitors [1 mM phenylmethylsulfonyl fluroide (PMSF) and 1 µg/ml aprotinin]. Nuclei were pelleted by centrifugation. Purified nuclear protein extracts were resuspended in buffer C (20 mM HEPES, 0.42 M NaCl, 1 mM EDTA, 25% glycerol, pH 7.9) containing protease inhibitors. Purified cytosolic protein extracts were resuspended in TEDG buffer containing protease inhibitors. Protein concentrations were measured by Bradford assay (Bio-Rad, Hercules, CA). Enzyme assays RTK, PKC and MAPK activities and p53 and p21waf1 protein levels were measured using protein isolated from stored liver tissues. RTK, PKC and MAPK activities were measured in two series of assays using the phosphocellulose paper method (Enan and Matsumura, 1995a). Littermate pairs of the same sex, some male and some female, were evaluated from each of three F2 germline chimeric sires. Preliminary experiments showed no sex differences in F3 offspring for RTK, PKC and MAPK activities or in p53 and p21waf1 protein levels (data not shown). Consequently, data from male and female
littermate pairs were pooled. Each activity measurement was repeated three times for a given animal. The first series of assays was conducted soon after completion of the experiment and the second series of assays was performed ~9 months later to verify the results. We also evaluated the possibility that the presence of the neo transgenic marker in the control offspring might have influenced the outcome of these enzyme assays even though this gene’s product, neomycin phosphotransferase II, has no known natural substrate in mammalian cells. The total cytosolic PKC activity was measured in the presence and absence of 40 µg of the peptide substrate histone type III-S (Enan and Matsumura, 1993). The assay buffer (20 mM Tris-HCl, pH 7.5) contained 1.83 mM CaCl2, 2 µg dioctanylglycerol, 5 µg phosphtidylserine, 5 µM ATP, 1.5 mM MgCl2, 2 mM DTT, 2 mM EGTA and 18.5 kBq [γ-32P]ATP. The overall PKC activity was defined as the difference between the enzyme activity in the presence and of the peptide substrate and in that in its absence, calculated based on the specific reactivity of the 32P and the amount spotted on the phosphocellulose paper. The EGF receptor was immunoprecipitated with an antibody to phosphotyrosine (anti-pTyr) (Wu et al., 1993). The RTK activity in the immunocomplex was then measured in assay buffer (50 mM HEPES pH 7.4, 10 µM Na3VO4, 10 mM MnCl2, 18.5 kBq [γ-32P]ATP and 10 µM ‘cold’ ATP) that contained 200 µM of the specific substrate peptide RR-SRC (Enan and Matsumura, 1996). The specificity and purity of each immunoprecipitate was tested using western blotting and by using non-specific IgG in parallel for all assays. The cytosolic MAPK activity was measured in assay buffer (12.5 mM MOPS pH 7.2, 7.5 mM MgCl2, 0.5 mM EGTA, 0.05 mM NaF, 2 mM DTT, 0.05 mM Na3VO4) (Enan and Matsumura, 1995b). The phosphorylation reaction was carried out in the presence of 18.5 kBq [γ-32P]ATP, 10 µM ‘cold’ ATP and 250 µM MAP kinase substrate peptide (APRTPGGRR) (Upstate Biotechnology, Lake Placid, NY; catalog no. 12-125). Activity was calculated based on the specific reactivity of the 32P and the amount spotted on the phosphocellulose paper. Glutathione S-transferase (GST) activity was measured by spectrophotometric assays that were performed with aromatic substrate to measure enzyme activity (Habig et al., 1974). Assays were performed at 25°C in 0.1 M potassium phosphate buffer at pH 6.5 with a final reaction volume of 1 ml. The reaction mixture contained 1 mM glutathione, 1 mM 1-chloro-2,4dinitrobenzene and 20 µg of cytosolic protein and was followed by observing an increase in absorbance at 340 nm for 2 min. GST activity assays were repeated three times for each animal. Tissues from five same-sex littermate pairs representing three F2 sires were analyzed. Measurements of nuclear p53 and p21waf1 Two types of assay were used to measure hepatic nuclear p53 and p21waf1 protein levels. The first set of assays utilized immunoprecipitation and western blotting of nuclear protein and was performed soon after completion of the experiment. In these assays p53 and p21waf1 protein amounts were each normalized to the total starting amount of protein prior to immunoprecipitation. Nuclear protein from the same three littermate pairs whose cytosolic proteins were used in kinase assay 1 was used in this assay. In the second set of assays nuclear protein was electrophoresed without immunoprecipitation and p53 and p21waf1 proteins were each normalized to the amount of α-tubulin in the same lane. This second set of assays was conducted ~4 years after the tissues were initially stored. Tissues from six same-sex littermate pairs were used in this assay, but not all provided sufficient protein for detection of both p53 and p21waf1 proteins by western analysis (see Table V). For the first set of assays, 100 µl of each nuclear protein preparation (2 µg/µl) was incubated with mouse monoclonal agarose-conjugated IgG for p53 (Pab240; Santa Cruz Biotech, Santa Cruz, CA; catalog no. sc-99) and p21waf1 (Santa Cruz BioTech, catalog no. sc-397) in 200 µl of Tris buffer, pH 7.5 and 50 µl of lysis buffer (20 mM Tris pH 7.5, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM β-glycerol phosphate, 1 mM Na3VO4, 1 µg/ml aprotinin and 1 mM PMSF) overnight at 4°C. After the addition of 15 µl of protein A/G plus agarose (Santa Cruz BioTech, catalog no. sc-2003), samples were incubated for 1 h at 4°C with gentle mixing. The immunoprecipitate was collected at maximum speed using an Eppendorf microcentrifuge at 4°C for 10 min. After two washes with 0.5 ml of lysis buffer and one wash with PBS, the pellet was suspended in 30 µl of PBS and mixed with 15 µl of 3⫻ loading buffer [187.5 mM Tris buffer pH 6.8, 6% sodium dodecyl sulfate (SDS), 30% glycerol and 0.3% bromophenol blue]. After 5 min of heating at 95°C, the samples were centrifuged and the entire volume of each supernatant was electrophoresed on a 10% SDS–polyacrylamide gel and then western blotted. The p53 and p21waf1 protein bands were visualized using non-radioactive SuperSignal West Pico Chemiluminescent Substrate (Pierce, IL; catalog no. 34080) and anti-mouse IgG horseradish peroxidase (HRP)-linked whole antibody (Amersham–Pharmacia, catalog no. NA 931). Non-specific IgG was
19
J.E.Baulch, O.G.Raabe and L.M.Wiley
Table II. Outcome of mating five F2 chimeric male mice to unirradiated CD1 females Breeding pair 1 2 3 4c 5c
neo–/– offspringa
neo⫹/– control offspring
male
female
male
female
2 1 0 4b 7
2 4 2 6 4
2b 2 0 0 0
7 3 3 0 0
total offspring per litter 13 10 5 10 11
This group includes both neo–/– F3 offspring from paternal radiation history and neo–/– control offspring. b One animal died before reaching 19 days of age. c These litters were not germline chimeras (they had no neo⫹/– offspring), so they are not included in any biochemical analyses.
Table III. Liver and body weight (mean ⫾ SEM) and liver weight:body weight ratio for F3 and control offspringa Offspring
n
Liver weight
Body weight
Ratio
neo⫹/– control neo–/– b
16 11
0.264 ⫾ 0.010 0.272 ⫾ 0.015
6.680 ⫾ 0.140 6.580 ⫾ 0.162
0.039 ⫾ 0.001 0.041 ⫾ 0.002
a Liver weights and body weights in grams were evaluated when the offspring were 19 days old. b The neo–/– group may contain both neo–/– F offspring from the paternal 3 F0 radiation history and neo–/– control offspring.
a
used in parallel as a negative control. Western blot images were scanned by a CCD camera (Sony, Japan) interfaced with a computer. Each protein band was analyzed densitometrically using AMBIS Image Acquisition and Analysis version 4.0 imaging software (AMBIS, San Diego, CA). Both of the presumptive p53 bands were scanned and summed for p53 protein quantification. In the second set of assays, 30 µg of nuclear protein (2 µg/µl) was loaded into each lane of a 10% SDS–polyacrylamide gel and electrophoresed. p53 and p21 protein standards (Oncogene catalog no. WB21 and Santa Cruz BioTech catalog no. 4078WB, respectively) were also run on each gel as a positive control. The proteins were then electroblotted to a polyvinylidene difluoride (PVDF) membrane (Bio-Rad). p21waf1 was detected by incubating the membrane with purified p21 antibody (Santa Cruz BioTech, catalog no. SC-6246) at a dilution of 1/1000 in TBS buffer (150 mM NaCl, 10 mM Tris– HCl, pH 8) containing 0.1% Tween-20 (TBST) and 5% milk overnight at 4°C. HRP-linked anti-mouse IgG was used as the secondary antibody at a dilution of 7/10 000 in TBST and 5% milk for 1 h at room temperature. SuperSignal West Pico Chemiluminescent Substrate was again used to visualize the protein bands. The same PVDF membrane was probed for p53 (diluted 15/10 000; Santa Cruz BioTech, catalog no. SC-99) then α-tubulin (diluted 1/1000; Oncogene, catalog no. CP06) by first incubating the membrane in standard stripping buffer (2% SDS, 62.5 mM Tris–HCl, 100 mM βmercaptoethanol, pH 6.8) for 15 min at 50°C, followed by two 10 min TBST washes, blocking with 5% milk in TBST for 1 h at room temperature and reprobing with the appropriate primary antibody overnight at 4°C. Visualization was performed as described for p21waf1. Western blot images were scanned and analyzed densitometrically as described previously. The optical density (OD) of each p53 or p21waf1 protein band was normalized to its respective α-tubulin protein band. Western blots were repeated three times for each animal to obtain an average OD ⫾ SEM. Statistical analyses Differences between animals or groups were compared by analysis of variance (ANOVA) or Student’s t-test. The appropriate t-test of mean values was performed depending on whether a significant difference in variance was detected using F-tests. Student’s t-test calculations were facilitated using INSTAT computer software (Graph Pad Software, San Diego, CA). ANOVA calculations were facilitated using SuperANOVA computer software (Abacus Concepts, Inc., Berkeley, CA). All statistical tests utilized P ⬍ 0.05 significance levels.
Results Five F2 offspring from three F1 sires with a paternal F0 radiation history were both XY← XY chimeras and positive for the neo transgene in tail DNA. Three of these males were germline chimeras and were derived from the three different F1 sires. They produced litters with both neo–/– offspring, a group that included the F3 offspring with a paternal F0 radiation history and some control offspring, and neo⫹/– offspring that could be unequivocally identified as controls (Table II). The five XY← XY chimeras were mated once to provide a pool of 49 offspring. Of these 49 offspring, 32 were offspring with a possible paternal F0 germline radiation history (neo–/–) and 20
Table IV. Basal liver protein kinase activity in F3 and control offspringa Kinase
RTK PKC MAPK
Assay 1
Assay 2
neo⫹/– control offspring
neo⫹/– control offspringb
neo⫹/– control offspring
neo–/– offspringb
0.76 ⫾ 0.03 0.36 ⫾ 0.01 1.81 ⫾ 0.01
0.48 ⫾ 0.02c 0.86 ⫾ 0.01c 0.75 ⫾ 0.00c
NDe 0.31 ⫾ 0.01 1.67 ⫾ 0.01
NDe 0.76 ⫾ 0.01c 0.54 ⫾ 0.01c
a
Values represent mean picomoles of 32P per nanomole of substrate per 20 µg protein ⫾ SEM (three mice per kinase per group). b The neo–/– group may contain both neo–/– F offspring from the paternal 3 F0 radiation history and neo–/– control offspring. c neo–/– F offspring significantly different from neo⫹/– control offspring 3 (P ⬍ 0.001; two-tailed Student’s t-test). d Data from assay 2 were obtained from additional animals after samples had been frozen for 9 months. e Not determined.
17 were littermate control offspring (neo⫹/–) (Table II). One of the 32 neo–/– offspring and one of the 17 neo⫹/– offspring died before 19 days of age. The relative numbers of neo⫹/– and neo–/– offspring varied between litters, probably as a result of the random allocation of cells from the two different embryonic populations to the chimeric germline rather than differences in the two embryonic cell populations per se. A substantial number of these chimeras would be required to resolve this point (see Falconer and Avery, 1978); such studies are beyond the scope of the present investigation. Evaluation of liver:body weight ratios Animals from the three litters that had both neo⫹/– and neo–/– offspring were evaluated for hepatomegaly, a gross indicator of non-specific toxicity or stress that correlates with many adverse health effects. Using nested ANOVAs where the genotype (neo⫹/– or neo–/–) was the main effect and the sire was the nested effect, the neo–/– offspring, which included the F3 offspring with paternal F0 radiation history, did not exhibit significant differences in mean liver weights, mean body weights or mean ratios of liver weight:body weight in comparison with neo⫹/– control offspring. Significant sire effects were, however, observed for mean body weights and mean liver weights (P ⬍ 0.05, nested ANOVA). Enzyme activity assays The results from assay 1 show significant differences between neo–/– offspring and neo⫹/– control offspring in the activity of all three protein kinases (P ⬍ 0.001, two-tailed t-test; Table IV). The second assay performed 9 months later confirmed the differences in PKC and MAPK activities between these two groups of offspring (P ⬍ 0.001, two-tailed t-test; Table IV). The neo–/– offspring and the neo⫹/– control offspring were also compared for the activity of GST. The mean ⫾ SEM
Heritable effects on signaling kinase activities
Fig. 2. Representative western immunoblot of hepatic p21waf1 (a) and p53 (b) protein levels in a littermate paired neo–/– F3 offspring and neo⫹/– control offspring after immunoprecipitation of the target proteins.
Table V. Hepatic levels of p53 and p21waf1 protein in F3 and control offspring detected by western immunoblots: protein assay 1 (n ⫽3 in each group) Mean ISEM densitometry reading/ µg protein
p53 p21waf1
neo⫹/– control offspring
neo–/– offspringa
762 ⫾ 35 499 ⫾ 23
3810 ⫾ 237 2275 ⫾ 136
Fold increase
P value
5.00 4.56
⬍⬍0.001 ⬍⬍0.001
Fig. 3. Representative direct western immunoblot of hepatic p21waf1 (a) and p53 (b) protein levels in neo–/– F3 offspring and neo⫹/– control offspring from two different litters. (c) α-Tubulin. Lanes 1 and 3, and lanes 2 and 4, are littermate-paired neo⫹/– control offspring and neo–/– F3 offspring.
Table VI. Hepatic levels of p53 and p21waf1 protein in F3 and control offspring detected by direct western blots: protein assay 2a
a
The neo–/– group may contain both neo–/– F3 offspring from the paternal F0 radiation history and neo–/– control offspring. b Two-tailed Student’s t-test.
hepatic GST activities were similar in the two groups of offspring: 125.2 ⫾ 3.17 and 124.60 ⫾ 4.35 nmol/min/µg of protein, respectively (five animals per group, each tested in triplicate). Measurements of nuclear p53 and p21waf1 p53 protein levels increase in response to ionizing radiations as a result of PKC activity. Because the group of offspring that included F3 animals with paternal F0 radiation history had higher PKC activity in the liver than did control offspring (Table IV), an increase in p53 levels was expected. In the first series of assays with immunoprecipitation and western blotting, we observed that the neo–/– offspring had five-fold higher levels of nuclear hepatic p53 protein than the neo⫹/– control offspring (P ⬍⬍ 0.001, two-tailed t-test; Table V, Figure 2a). p21waf1 was measured because this gene product is a transcriptional target of activated p53 (discussed in Okorokov et al., 1997) and p21waf1 causes cell cycle delays. The neo–/– offspring had more than four times as much p21waf1 protein as the neo⫹/– control offspring (P ⬍⬍ 0.001, two-tailed t-test; Table V, Figure 2b). To confirm the initial observations, the measurement of nuclear p53 and p21waf1 protein levels was repeated in a second series of assays using conventional western analysis procedures instead of immunoprecipitation of the target protein (Figure 3). Additionally, α-tubulin was used as an internal control to which p53 and p21waf1 protein amounts were normalized (Figure 3c). Despite the occurrence of protein degradation between the times of the two assays, similar trends in the relative amounts of p53 and p21waf1 proteins were observed in livers from neo–/– offspring and neo⫹/– control offspring. Even though there was considerable variability among the samples, as shown in Figure 3, the neo–/– offspring that included the F3 animals with paternal F0 radiation history had
Mean ISEM densitometry readingb/ µg protein neo⫹/– control offspring p53 p21waf1
6875 (n ⫽ 2359 (n ⫽
⫾ 2139 4) ⫾ 798 3)
Fold increase
P value
2.19
0.04
2.21
0.15
neo–/– offspringa 15035 ⫾ 2823 (n ⫽ 6) 5205 ⫾ 1862 (n ⫽ 4)
a
Data obtained from additional animals after tissue had been frozen for 4 years. b Normalized to α-tubulin. c The neo–/– group may contain both neo–/– F offspring from the paternal 3 F0 radiation history and neo–/– control offspring. d One-tailed Student’s t-test.
more than twice as much α-tubulin-normalized p53 and p21waf1 as neo⫹/– control offspring (P ⫽ 0.04 and P ⫽ 0.15, respectively, one-tailed t-test; Table VI). Discussion The results of this experiment with CD1 mice suggest that F3 offspring from a paternal F0 germline irradiation of type B spermatogonia had important differences in markers of cell signaling when compared with littermates with no germline radiation history. This heritable cellular response to a paternal F0 germline irradiation included alterations in the activities of signaling kinases that are modulated by activation of growth factor receptors (RTK, MAPK and PKC). These changes were associated with increases in the cell cycle protein p21waf1 and in p53 protein, which is involved in p21waf1 regulation. The experimental design, which utilized germline chimeras as F2 fathers, minimized variability due to maternal heterogeneity and animal husbandry because F3 offspring with a radiation history and control offspring were obtained from sperm generated in the same gonad and both types of offspring were born in the same litter. The neo transgenic marker identified control 21
J.E.Baulch, O.G.Raabe and L.M.Wiley
offspring. A limitation of the design was our inability to use the neo marker in the homozygous state because of chimeric drift in the germline. Because the neo marker was used in the heterozygous state, the neo⫹/– control group could be positively identified, but the neo–/– group was assumed to be composed of some control animals as well as the F3 offspring with an F0 radiation history. This experimental design was considered acceptable because the presence of control animals in the group of neo–/– offspring would only diminish any difference in biological endpoints that was due to the F0 radiation history. It is not clear why the neo transgenic cell lineage marker was selected against in the germline of chimeric males. The neo-containing CD4 gene ablation cassette has no effect on fertility, reproduction or embryo development (Rahemtulla et al., 1991). In our laboratory, preliminary experiments showed that no competitive cell proliferation disadvantage was observed in preimplantation embryo chimeras when either neo⫹/⫹ or neo⫹/– CD1 embryos were paired with wild-type CD1 embryos (data not shown). In the present study, no effect on signaling protein kinase activities was observed when neo⫹/⫹ animals were compared with neo–/– animals (Table I). We have also used conventional multi-generation studies with CD1 mice to compare offspring from paternal F0 radiation history with offspring from shamirradiated concurrent controls. In these experiments, we observed differences between the animals with a radiation history and the controls in liver:body weight ratios, in hepatic PKC and MAPK activities and in p53 and p21waf1 protein levels (unpublished observations). Taken together, these published and unpublished data support our conclusion that the biological and biochemical differences observed between the neo–/– animals and the neo⫹/– animals in the present study are the result of a cellular response to paternal F0 germline irradiation in a subset of the neo–/– offspring and not a response to the neo transgene. The group of neo–/– offspring that included F3 animals with a paternal F0 radiation history did not have altered liver:body weight ratios or hepatic GST activity in comparison with neo⫹/– littermate controls. The GST results suggest that the other observed effects were not a result of an unrecognized acute exposure to chemical toxicants (McGuire et al., 1996) or ionizing radiations (Berhane et al., 1994; Richardson and Siemann, 1994) experienced by the offspring shortly before they were killed. Although it is not clear whether GST is functionally associated with signal transduction cascades, it is stimulated by the presence of ROIs and products of lipid peroxidation as a proximal step in the cellular radiation response (see Berhane et al., 1994; Richardson and Siemann, 1994). No difference in the mean body weights or liver weights of 19 day old offspring from the two groups was observed, even though the adult F1 offspring from a paternal F0 radiation history in the present experiment had significantly decreased mean body weights in comparison with concurrent controls (male, Wiley et al., 1997; female, unpublished data). In another experiment we observed that adult F2 males from a paternal F0 irradiation history also had significantly decreased mean body weights in comparison with concurrent control males (unpublished data). No significant correlation between body weight and radiation history was observed prior to weaning when animals from paternal F0 irradiation were compared with concurrent controls in these other studies (unpublished data). When somatic cells are exposed to acute irradiation with x22
rays or γ-rays, there is an interaction between radiation photons and the plasma membrane (and associated water molecules). This interaction results in the production of ROIs and in lipid peroxidation (Nakajima et al., 1996) and is followed by tyrosine phosphorylations (Uckun et al., 1993) including those on EGF receptors through autophosphorylation by RTK (Schmidt-Ullrich et al., 1996). MAPK can be directly activated by radiation-produced ROIs independently of PKC activation, although PKC and MAPK activation may both occur within acutely irradiated cells (see Stevenson et al., 1994). PKC activation can lead to increases in p53 protein in acutely irradiated cells (Kanna and Lavin, 1993) and the presence of p53 can down-regulate MAPK activities (see Fukasawa and Woude, 1997). In addition, RTK activity generally correlates with MAPK activity (Bulton et al., 1991; Crews et al., 1992). Although the cell signaling responses to acute low LET irradiations are generally transient (see Uckun et al., 1993; Wilson et al., 1993), acute irradiations can lead to long-term consequences that persist across many cell divisions. These consequences include persisting oxidative stress (Clutton et al., 1996) and an increased incidence of chromosomal anomalies (i.e. ‘genomic instability’; reviewed by Lee et al., 1994 and Morgan et al., 1996; see also Clutton et al., 1996; Watson et al., 1996). Although persisting oxidative stress can lead to increased activities of cytosolic PKC (Emerit, 1994), the normal GST activity levels that were exhibited by the F3 offspring suggest that persisting oxidative stress might not cause the observed effects on cell signaling. We hypothesize that ‘genomic instability’ is an important factor mediating the heritability of adverse effects following paternal irradiation. Although there was no evidence of chromosomal abnormalities in the F3 offspring from paternal F0 radiation history, the definition of genomic instability can be broadened to include additional genetic effects such as point mutations that might not necessarily lead to gross aneuploidy. Evidence from other laboratories suggests that the p53 protein can bind to damaged DNA through interaction of its C-terminus tail with the staggered ends of single-stranded DNA (ssDNA) that result from double-stranded DNA (dsDNA) breaks (Selivanova et al., 1996). The ssDNA-binding domain of p53 appears to be directly phosphorylated by PKC and this phosphorylation activates binding of p53 to consensus sequences in target genes, including the gene encoding p21waf1 (Hupp and Lane, 1994). These observations are consistent with the interpretation that the nuclear p53 protein we detected in F3 offspring was functional and that its elevated levels contributed to the elevated levels of p21waf1 protein. This interpretation must be tentative because we did not evaluate the F3 offspring for DNA damage per se. To conclude, we hypothesize that irradiation of the paternal F0 type B spermatogonia initiated DNA-affecting events that evolved to produce the phenotype we describe here for the F3 offspring. Future studies on cell lines established from additional F3 offspring from paternal F0 radiation history will include experiments to test the hypothesis for ‘genomic instability’ and to identify the functional relationships between the elevated p53 protein levels and changes in signaling kinase activities in these F3 offspring. Acknowledgements The authors wish to acknowledge the expert technical assistance of Marie Suffia and George Withers in the production of the chimeric F2 XY← →XY males and of Richard Lum and Zhong Wu in the biochemical analyses and
Heritable effects on signaling kinase activities Dr James W. Overstreet for his critical review of this manuscript. Work supported by NIH RO1 ES06516 to L.M.W. and J.W.O. and NIEHS grant 5 T32 ES07059 to J.E.B.
References Berhane,K.M., Widersten,M., Engstrom,A., Kozarich,J.W. and Mannervik,B. (1994) Detoxification of base propenals and other α,β-unsaturated aldehyde products of radical reactions and lipid peroxidation by human glutathione transferases. Proc. Natl Acad. Sci. USA, 91, 1480–1484. Brown,C.R., Doxsey,S.J., White,E. and Welch,W.J. (1994) Both viral (adenovirus E1B) and cellular (hsp70, p53) components interact with centrosomes. J. Cell. Physiol., 160, 47–60. Bulton,T.G. et al. (1991) ERKs: a family of protein-serine/threonine kinases that are activated and tyrosine phosphorylated in response to insulin and NGF. Cell, 65, 663–675. Clutton,S.M., Townsend,K.M.S., Walker,C., Ansell,J.D. and Wright,E.G. (1996) Radiation-induced genomic instability and persisting oxidative stress in primary bone marrow cultures. Carcinogenesis, 17, 1633–1639. Crews,C.M., Alesandrini,A. and Erikson,R.L. (1992) The primary structure of MEK, a protein kinase that phosphorylates the ERK gene product. Science, 258, 478–480. el-Deiry,W.S. et al. (1995) Topological control of p21WAF1/CIP1 expression in normal and neoplastic tissues. Cancer Res., 55, 2910–2919. Emerit,I. (1994) Reactive oxygen species, chromosome mutation, and cancer: possible role of clastogenic factors in carcinogenesis. Free Radical Biol. Med., 16, 99–109. Fukasawa,K. and Woude,G.F. (1997) Synergy between the mos/mitogenactivated protein kinase pathway and loss of p53 function in transformation and chromosome instability. Mol. Cell. Biol., 17, 506–518. Fukasawa,K., Choi,T., Kuriyama,R., Rulong,S. and Woude,G.F. (1996) Abnormal centrosome amplification in the absence of p53. Science, 271, 1744–1747. Enan,E. and Matsumura,F. (1993) Activation of phosphoinositide/protein kinase C pathway in rat brain tissue by pyrethroids. Biochem. Pharmacol., 45, 703–710. Enan,E. and Matsumura,F. (1995a) Evidence for a second pathway in the action mechanism of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). Biochem. Pharmacol., 49, 249–261. Enan,E. and Matsumura,F. (1995b) Regulation by 2,3,7,8-tetrachlorodibenzop-dioxin (TCDD) of the DNA binding activity of transcriptional factors via nuclear protein phosphorylation in guinea pig adipose tissue. Biochem. Pharmacol., 50, 1199–1206. Enan,E. and Matsumura,F. (1996) Identification of C-Src as an integral component of the cytosolic Ah receptor complex, transducing the signal of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) through the protein phosphorylation pathway. Biochem. Pharmacol., 52, 1599–1612. Falconer,D.S. and Avery,P.J. (1978) Variability of chimeras and mosaics. J. Embryol. Exp. Morphol., 43, 195–219. Habig,W.H., Pabet,M.J. and Jakoby,W.B. (1974) Glutathione S-transferases: the first enzymatic step in mercapturic acid formation. J. Biol. Chem., 249, 7130–7139. Hupp,T.R. and Lane,D.P. (1994) Allosteric activation of latent p53 tetramers. Curr. Biol., 4, 865–875. Lee,J.M., Abrahamson,J.L. and Bernstein,A. (1994) DNA damage, oncogenesis and the p53 tumor-suppressor gene. Mutat. Res., 307, 573–581. McGuire,S., Daggett,D., Bostad,E., Schroeder,S., Siegel,F. and Kornguth,S. (1996). Cellular localization of glutathione S-transferases in retinas of control and lead-treated rats. Invest. Ophthalm. Visual Sci., 37, 833–842. Morgan,W.F., Day,J.P., Kaplan,M.I., McGhee,E.M. and Limoli,C.L. (1996) Genomic instability induced by ionizing radiation. Radiat. Res., 146, 247–258. Nakajima,T. and Yukawa,O. (1996) Radiation-induced translocation of protein kinase C through membrane lipid peroxidation in primary cultured rat hepatocytes. Int. J. Radiat. Biol., 70, 473–480. Obasaju,M.F., Wiley,L.M., Oudiz,D.J., Miller,L., Samuels,S.J., Chang,R.J. and Overstreet,J.W. (1988) An assay using embryo aggregation chimeras for the detection of non-lethal changes in X-irradiated mouse preimplantation embryos. Radiat. Res., 113, 289–299. Obasaju,M.F., Wiley,L.M., Oudiz,D.J., Raabe,O.G. and Overstreet,J.W. (1989) A chimera embryo assay reveals a decrease in embryonic cellular proliferation induced by sperm from X-irradiated male mice. Radiat. Res., 118, 246–256. Okorokov,A.L., Ponchel,F. and Milner,J. (1997) Induced N- and C-terminal cleavage of p53: a core fragment of p53, generated by interaction with damaged DNA, promotes cleavage of the N-terminus of full-length p53,
whereas ssDNA induced C-terminal cleavage of p53. EMBO J., 16, 6008–6017. Peippo,J. and Bredbacka,P. (1995) Sex-related growth rate differences in mouse preimplantation embryos in vivo and in vitro. Mol. Reprod. Dev., 40, 56–61. Peters,J.M., Tsark,E.C. and Wiley,L.M. (1996) Radiosensitive target in the mouse embryo chimera assay: implications that the target involves autocrine growth factor function. Radiat. Res., 145, 722–729. Rahemtulla, A. et al. (1991) Normal development and function of SC8⫹ cells but markedly decreased helper cell activity in mice lacking CD4. Nature, 353, 180–184. Rappolee,D.A., Brenner,C.A., Schultz,R., Mark,D. and Werb,Z. (1988) Developmental expression of PDGF, TGF-α, and TGF-β genes in preimplantation mouse embryos. Science, 241, 1823–1825. Richardson,M.E. and Siemann,D.W. (1994) Thiol-related mechanism of resistance in a murine tumor model. Int. J. Radiat. Oncol. Biol. Physics, 29, 387–392. Schmidt-Ullrich,R.K., Valerie,K., Fogleman,P.B. and Walters,J. (1996) Radiation-induced autophosphorylation of epidermal growth factor receptor in human malignant mammary and squamous epithelial cells. Radiat. Res., 145, 81–85. Selivanova,G., Iotsova,V., Kiseleva,E., Strom,M., Bakalkin,G., Grafstrom,R.C. and Wiman,K.G. (1996) The single-stranded DNA end binding site of p53 coincides with the C-terminal regulatory region. Nucleic Acids Res., 24, 3560–3567. Stevenson,M.A., Pollock,S.S., Colman,C.N. and Calderwood,S.K. (1994) X-irradiation, phorbol esters, and H2O2 stimulate mitogen-activated protein kinase activity in NIH-3T3 cells through the formation of reactive oxygen intermediates. Cancer Res., 54, 12–15. Straume,T., Raabe,O.G., Walsh,K.J. and Wiley,L.M. (1993) Inherited effects from irradiated mouse immature oocytes detected in aggregation embryo chimeras. Mutat. Res., 287, 243–251. Straume,T., Raabe,O.G., Walsh,K.J. and Wiley,L.M. (1997) Transmitted proliferation disadvantage from mouse oocytes labeled in vivo with 3H-thymidine: radiosensitive target considerations. Mutat. Res., 374, 11–19. Tsark,E.C., Adamson,E.D., Withers,G.E. and Wiley,L.M. (1997) Expression and function of amphiregulin during murine preimplantation development. Mol. Reprod. Dev., 47, 271–283. Uckun,F., Schieven,G.L., Tuel-Ahlgren,L.M., Dibrdik,I., Fyers,D.E., Ledbetter,J.A. and Song,C.W. (1993) Tyrosine phosphorylation is a mandatory proximal step in radiation-induced activation of the protein kinase C signaling pathway in human B-lymphocyte precursors. Proc. Natl Acad. Sci. USA, 90, 252–256. Warner,P., Wiley,L.M., Oudiz,D.J., Overstreet,J.W. and Raabe,O.G. (1991) Paternally inherited effects of gamma radiation on mouse preimplantation development detected by the chimera assay. Radiat. Res., 128, 48–58. Watson,G.E., Lorimore,S.A. and Wright,E.G. (1996) Long-term in vivo transmission of α-particle-induced chromosomal instability in murine haemopoietic cells. Int. J. Radiat. Biol., 69, 175–182. Wiley,L.M., Van Beek,M.E.A.B. and Raabe,O.G. (1994a) Embryonic effects transmitted by male mice irradiated with 512 MeV/amu 56Fe nuclei. Radiat. Res., 138, 373–385. Wiley,L.M., Raabe,O.G., Khan,R. and Straume,T. (1994b) Radiosensitive target in the early mouse exposed to very low doses of ionizing radiation. Mutat. Res., 309, 83–92. Wiley,L.M., Baulch,J.E., Raabe,O.G. and Straume,T. (1997) Impaired cell proliferation in mice that persists across at least two generations after paternal irradiation. Radiat. Res., 148, 145–151. Wilson,R.E., Taylor,S.L., Atherton,G.T., Johnston,D., Waters,C.M. and Norton,J.D. (1993) Early response gene signaling cascades activated by ionising radiation in primary human B cells. Oncogene, 8, 3229–3237. Woloschak,G.E., Chang-Liu,C.M. and Shearin-Jones,P. (1990) Regulation of protein kinase C by ionizing radiation. Cancer Res., 50, 3963–3967. Wu,J., Dent,P., Jelinek,T., Wolfman,A., Weber,M.J. and Sturgill,T.W. (1993) Inhibition of the EGF-activated MAP kinase signaling pathway by adenosine 3,5-monophosphate. Science, 262, 1065. Xiong,Y., Hannon,G.J., Zhang,H., Casso,D., Kobayashi,R. and Beach,D. (1993) p21 is a universal inhibitor of cyclin kinases. Nature, 366, 701–704. Received on March 21, 2000; accepted on August 18, 2000
23
