We have previously shown that DNA from dying tumor cells may be transferred to living cells via the uptake of apoptotic bodies and may contribute to tumor progression. DNA encoding H-rasV12 and c-myc oncogenes may be transferred to the nucleus of the phagocyte but will only integrate and propagate in p53- and p21-deficient mouse embryonic fibroblasts, whereas normal cells are resistant to transformation. Here, we show that this protective mechanism (activation of p53 and p21 after uptake of apoptotic bodies) is dependent on DNA fragmentation, where inhibition of the caspase-activated DNase in the apoptotic cells, in conjunction with genetic ablation of lysosomal DNase II in the phagocytes, completely blocks p53 activation and consequently allows DNA replication of transferred DNA. We, therefore, suggest that there is a causal relationship between DNA degradation during apoptosis and p53 activation. In addition, we could further show that Chk2−/− cells were capable of replicating the hygR gene taken up from engulfed apoptotic cells, suggesting involvement of the DNA damage response. These data show that the phagocytosing cell is sensing the degraded DNA within the apoptotic cell, hence preventing these genes from being replicated, probably through activation of the DNA damage response. We, therefore, hypothesize that DNase II together with the Chk2, p53, and p21 pathway form a genetic barrier blocking the replication of potentially harmful DNA introduced via apoptotic bodies, thereby preventing transformation and malignant development. (Mol Cancer Res 2006;4(3):187–95)

The genomic integrity of an individual cell is threatened by DNA-damaging agents, such as UV or ionizing radiation, mutagenic chemicals, and endogenous oxygen radicals generated by normal metabolism. Another potential insult to the integrity of the genome is the horizontal transfer of genetic material from dead cells via the uptake of apoptotic bodies. We and others have shown that genetic information may be transferred from dying to living cells via the uptake of apoptotic bodies (1-3). However, transfer of apoptotic DNA to normal fibroblast or endothelial cells results in cell cycle arrest and senescence and is therefore not replicated to its daughter cells (2, 4). The p53 tumor suppressor is a central mediator of cellular responses to DNA damage and other forms of cellular stress (5, 6). Abnormalities of the p53 tumor suppressor gene are thought to be central to the development of a high proportion of human tumors. We have previously shown that mouse embryonic fibroblasts (MEF) deficient in the p53 gene or its transcriptional target p21 differ from normal cells as they are able to replicate DNA taken up from apoptotic bodies (Table 1). Furthermore, apoptotic bodies derived from a H-rasV12– and human c-myc–transformed rat fibrosarcoma were able to transform MEF p53- or p21-deficient cells but not wild-type MEF cells (4, 7). Integration and propagation of the tumor-derived H-rasV12 and human c-myc oncogenes could be verified by fluorescence in situ hybridization analysis of the resulting mouse tumor cells. These findings indicate that horizontal gene transfer between tumor cells may be a driving force of genomic instability and high mutability of tumor cells, and that the p53/p21 pathway protects normal cells against propagation of potentially harmful DNA (8).

Table 1.

Results from Cocultivation of MEFs Together with Apoptotic Tumor Cells

Recipient cellsp53 inductionHygromycin resistantFocus formationTumor
Wild-type MEF − − − 
DNase II−/− −/+ − ND 
Chk2−/− − − − 
p53−/− − 
p21−/− 
Recipient cellsp53 inductionHygromycin resistantFocus formationTumor
Wild-type MEF − − − 
DNase II−/− −/+ − ND 
Chk2−/− − − − 
p53−/− − 
p21−/− 

NOTE: Coculture of apoptotic cells, containing the oncogenes c-myc and H-rasV12 together with the hygR gene, with MEFs will lead to different outcome depending on the genetic background of the cells. Our data indicate that p53, p21, Chk2, and DNase II protect the phagocytosing cells from propagation of incoming DNA.

To further investigate how cells are protected against incoming apoptotic DNA, we hypothesized that fragmented DNA from dead cells may activate the DNA damage response of the phagocyte. Here, we show for the first time that DNA degradation of apoptotic DNA by the caspase-activated DNase (CAD) and especially DNase II enzymes is necessary for activation of the DNA damage pathway within the phagocytosing cell. Furthermore, we provide evidence that activation of the Chk2/p53/p21 pathway by these DNA fragments blocks replication of genomic DNA that has been transferred from apoptotic cells. Chk2 is a critical regulator of p53 functions because cells derived from Chk2-deficient mice have a defective G1-S checkpoint, apoptosis, and the transcriptional induction of p53 target genes, such as p21Waf1, Bax, and Noxa (9, 10). Thus, Chk2 is a key component of a highly conserved DNA damage response signaling pathway, where p53 is one critical target. Here, we show that genetic ablation of either the DNase II or Chk2 gene abolishes the activation of p53 and allows replication of foreign genomic DNA. Therefore, we suggest that the DNase II enzyme together with the Chk2/p53/p21 DNA damage pathway form a protective barrier against horizontally transferred DNA. This genetic barrier may guard the genomic integrity of normal cells and thereby prevent transformation and malignant development.

DNA Fragmentation in the Apoptotic Body Determines the Kinetics of p53 Induction in the Phagocyte

Induction of apoptosis will result in condensation of chromatin, and this is later followed by cleavage of the DNA. This DNA fragmentation requires activation of DNases; therefore, we analyzed the role of DNA fragmentation by inhibition of the responsible DNases. One of the DNases responsible for DNA degradation during apoptosis is CAD (11). CAD is normally forming a complex with ICAD, its inhibitor, that is cleaved by caspases during apoptosis, and CAD is thereby activated (12, 13). We transfected rat embryonic fibroblasts (REF) with the H-rasV12, human c-myc oncogenes, and the hygR gene (REFrmh cells) to use as donor cells. To study the role of DNA degradation, these REFrmh cells were transfected with mutant ICAD (ICADLdm) that can not be cleaved by caspases, and DNA fragmentation in the apoptotic cell is thereby prevented (14). Analysis of DNA integrity by DNA fragmentation gels showed that ICADLdm expression efficiently inhibited ladder formation after apoptosis induced by nutrient depletion (Fig. 1A). To study the effect of CAD inhibition on p53 induction in the phagocytosing cell, apoptosis was induced by nutrient depletion in REFrmh or REFrmh-ICADLdm cells, and the resulting apoptotic cells were cocultured with MEF cells. The kinetics of p53 induction was analyzed by Western blot 4, 8, 12, and 24 hours after the initiation of the cocultivation, where the induction of p53 was shifted 4 hours and peaked 12 hours after the addition of the apoptotic REFrmh-ICADLdm cells (Fig. 1B and C).

FIGURE 1.

Inhibition of DNA fragmentation in the dying cells shifts the kinetics of p53 activation. A. Donor REFrmh cells were retrofected with ICAD-Ldm or empty vector. Positive cells were selected with puromycin. Five millions cells were harvested, and the nuclei were pelleted by centrifugation. Fragmentation of DNA in the supernatant was analyzed by agarose gel electrophoresis. The cells were either cultured in growth medium or exposed to nutrient depletion for 24 hours. A 123-bp ladder was used as a reference (M). Nutrient depletion induced DNA fragmentation in the vector-transfected REFrm cells, whereas expression of ICAD-Ldm efficiently inhibited DNA fragmentation. B. Western blot analysis of the kinetics of p53 and p21 induction after feeding MEF cells with either REFrm or REFrm-ICADLdm apoptotic bodies. C. Levels of p53 accumulation in three independent experiments. Bars, SD.

FIGURE 1.

Inhibition of DNA fragmentation in the dying cells shifts the kinetics of p53 activation. A. Donor REFrmh cells were retrofected with ICAD-Ldm or empty vector. Positive cells were selected with puromycin. Five millions cells were harvested, and the nuclei were pelleted by centrifugation. Fragmentation of DNA in the supernatant was analyzed by agarose gel electrophoresis. The cells were either cultured in growth medium or exposed to nutrient depletion for 24 hours. A 123-bp ladder was used as a reference (M). Nutrient depletion induced DNA fragmentation in the vector-transfected REFrm cells, whereas expression of ICAD-Ldm efficiently inhibited DNA fragmentation. B. Western blot analysis of the kinetics of p53 and p21 induction after feeding MEF cells with either REFrm or REFrm-ICADLdm apoptotic bodies. C. Levels of p53 accumulation in three independent experiments. Bars, SD.

Close modal

Inhibition of DNA Fragmentation by CAD and DNase II Inhibits p53 Induction

The inhibition of CAD in the donor cell delayed but did not completely inhibit p53 induction. We therefore investigated the involvement of DNases localized in the lysosomes of the phagocytes. For this purpose, we used Bafilomycin A1, a specific irreversible inhibitor of vacuolar H+/ATPase, resulting in inhibition of acidification in endosomes and in lysosomes and thus indirect inhibition of lyzosomal DNase activity (15). Recipient MEFs were treated with Bafilomycin A1 during the cultivation with apoptotic cells. Inhibition of lysosomal acidification resulted in shifted kinetics of p53 accumulation, which peaked at 12 hours (Fig. 2A and C). To test the effect of combination of inhibiting the CAD and DNase II, we used REFrm-ICADLdm as donors together with Bafilomycin A1. The joint inhibition of CAD in the dying cell and lysosomal acidification in the phagocytosing cell resulted in a further delay in p53 accumulation, which now peaked at 24 hours (Fig. 2B and D).

FIGURE 2.

Inhibition of lysosomal DNA fragmentation delays the induction of p53. DNA fragmentation in the lysosomal was inhibited by treating the recipient cells with Bafilomycin A1 as previously reported by McIlroy et al. (14). A. Western blot analysis of p53 and p21 induction after addition of apoptotic bodies. Cells (right) were treated as described in Materials and Methods with Bafilomycin A1. B. Effect of inhibiting both CAD and lysosomal DNase activity on p53 induction using REFrmh-ICADLdm donor cells together with Bafilomycin A1. C. Levels of p53 accumulation in three independent experiments. D. Levels of p53 accumulation in three independent experiments. Bars, SD.

FIGURE 2.

Inhibition of lysosomal DNA fragmentation delays the induction of p53. DNA fragmentation in the lysosomal was inhibited by treating the recipient cells with Bafilomycin A1 as previously reported by McIlroy et al. (14). A. Western blot analysis of p53 and p21 induction after addition of apoptotic bodies. Cells (right) were treated as described in Materials and Methods with Bafilomycin A1. B. Effect of inhibiting both CAD and lysosomal DNase activity on p53 induction using REFrmh-ICADLdm donor cells together with Bafilomycin A1. C. Levels of p53 accumulation in three independent experiments. D. Levels of p53 accumulation in three independent experiments. Bars, SD.

Close modal

DNase II has previously been shown to be essential for degradation of apoptotic DNA in the lysosomes of the phagocytosing cell (16); therefore, we tested the effect of genetic ablation of DNase II on p53 activation. MEF cells lacking DNase II activity resulted in a dramatic shift in p53 induction, which occurred at 24 hours compared with DNaseII+/+ cells, where p53 was induced at 8 hours (Fig. 3A, B, D, and E). As a control, DNase II enzyme was retransfected into DNaseII−/− cells, and in these cells, the p53 kinetics was restored (Fig. 3C and F). Immunostaining showed an even more dramatic shift in p53 induction, where p53 induction was detected already at 4 hours in DNaseII+/+ cells compared with 24 hours in DNaseII−/− cells (Fig. 4A and B). Furthermore, no p53 induction could be detected by Western blot analysis when both CAD and DNase II enzyme activities were blocked (Fig. 3A and D). These data were confirmed by immunostaining, where no p53 induction could be detected within the DNase II−/− phagocytes after addition of apoptotic cells lacking functional CAD (Fig. 4A).

FIGURE 3.

Inhibition of CAD and lysosomal DNase II completely blocks p53 activation induced by apoptotic bodies. A. Western blot analysis of p53 activation in DNase II–deficient MEF cells after addition of either REFrmh or REFrmh-ICADLdm. B. Western blot analysis of p53 activation in wild-type MEF cells after addition of either REFrmh or REFrmh-ICADLdm. C. Western blot analysis of p53 induction in DNase II–deficient MEF cells retransfected with DNase II, after addition of either REFrmh or REFrmh-ICADLdm apoptotic cells. D. Levels of p53 accumulation in DNase II−/− MEF cells in three independent experiments. E. Levels of p53 accumulation in wild-type MEF cells in three independent experiments. F. Levels of p53 accumulation in MEF DNase II−/− cells retransfected with DNase II in three independent experiments. Bars, SD.

FIGURE 3.

Inhibition of CAD and lysosomal DNase II completely blocks p53 activation induced by apoptotic bodies. A. Western blot analysis of p53 activation in DNase II–deficient MEF cells after addition of either REFrmh or REFrmh-ICADLdm. B. Western blot analysis of p53 activation in wild-type MEF cells after addition of either REFrmh or REFrmh-ICADLdm. C. Western blot analysis of p53 induction in DNase II–deficient MEF cells retransfected with DNase II, after addition of either REFrmh or REFrmh-ICADLdm apoptotic cells. D. Levels of p53 accumulation in DNase II−/− MEF cells in three independent experiments. E. Levels of p53 accumulation in wild-type MEF cells in three independent experiments. F. Levels of p53 accumulation in MEF DNase II−/− cells retransfected with DNase II in three independent experiments. Bars, SD.

Close modal
FIGURE 4.

Induction of p53 in the phagocyte is dependent on DNA fragmentation in the apoptotic body. A. Wild-type MEF cells. Top, positive control of p53 staining of cells treated with cisplatin for 4 hours. Bottom, induction of p53 in cells at indicated time points after incubation with apoptotic REFrm or REFrm-ICADLdm. B. MEF DNaseII−/− cells. Top, positive control of p53 staining in cells treated with cisplatin for 4 hours. Bottom, induction of p53 in cells at indicated time points after incubation with apoptotic REFrm or REFrm-ICADLdm. Arrows, positive nuclei. Bar, 10 μm.

FIGURE 4.

Induction of p53 in the phagocyte is dependent on DNA fragmentation in the apoptotic body. A. Wild-type MEF cells. Top, positive control of p53 staining of cells treated with cisplatin for 4 hours. Bottom, induction of p53 in cells at indicated time points after incubation with apoptotic REFrm or REFrm-ICADLdm. B. MEF DNaseII−/− cells. Top, positive control of p53 staining in cells treated with cisplatin for 4 hours. Bottom, induction of p53 in cells at indicated time points after incubation with apoptotic REFrm or REFrm-ICADLdm. Arrows, positive nuclei. Bar, 10 μm.

Close modal

Horizontally Transferred DNA Is Replicated in Cells with Inactive DNase II

The lack of p53 induction argued that DNase II–deficient cells may be permissive for replicating genomic DNA transferred from engulfed dying cells.

Apoptotic REFrm, REFrmh, or REFrmh-ICADLdm was added to DNase II−/− cells, and resistant colonies were selected with hygromycin. Cocultivation of DNase II−/− cells with either the REFrmh or REFrmh-ICADLdm resulted in stable propagation of the hygR gene and the growth of hygromycin-resistant colonies, although at different frequencies (Fig. 5A; Table 1). Retransfection of the DNase II gene completely blocked the formation of hygromycin-resistant colonies (Fig. 5B), and p53 induction was detected 8 hours after coculture with apoptotic REFrmh cells (Fig. 3C and F). This emphasize that DNase II plays an essential role in protecting cells from replicating genomic DNA acquired from dying cells.

FIGURE 5.

Genetic inactivation of DNase II allows stable replication of genes transferred from apoptotic bodies. A. DNase II−/− and DNase II+/+ cells, respectively, were cocultivated with REFrmh or REFRmh-ICADLdm apoptotic bodies, and resistant colonies were selected for with hygromycin as described in Materials and Methods. REFrm cells lacking the hygR gene served as a negative control. Columns, number of colonies per 10-cm Petri dish from triplicates; bars, SD. nd, nondetectable. B. DNase II−/− cells retrofected with the mouse DNase II gene were incubated with apoptotic bodies derived from REFrm or REFrmh cells. Colonies per 1,000 cells after 2 weeks of hygromycin selection. C. DNA degradation by CAD does not inhibit propagation of transferred DNA. Wild type, DNase II−/−, or p21−/− cells were cocultivated with REFrmh apoptotic cells, and resistant colonies were selected for with hygromycin as described in Materials and Methods. Apoptosis in the donor cells was induced by nutrient depletion. As indicated, apoptosis was induced by γ-irradiation (150 Gy) in some of the donor cells to induce double-strand breaks. Colonies per 1,000 cells after 2 weeks of hygromycin selection. No hygromycin-resistant colonies were observed, where wild-type MEF cells were used as recipient cells. Interestingly, DNase II−/− MEFs acquired the hygR gene from REFrm apoptotic bodies killed by nutrient depletion, but induction of double-strand breaks by irradiation totally blocked propagation of transferred DNA. In contrast, p21−/− cells were able to acquire the hygR gene independent on whether the donor cells had been irradiated or not.

FIGURE 5.

Genetic inactivation of DNase II allows stable replication of genes transferred from apoptotic bodies. A. DNase II−/− and DNase II+/+ cells, respectively, were cocultivated with REFrmh or REFRmh-ICADLdm apoptotic bodies, and resistant colonies were selected for with hygromycin as described in Materials and Methods. REFrm cells lacking the hygR gene served as a negative control. Columns, number of colonies per 10-cm Petri dish from triplicates; bars, SD. nd, nondetectable. B. DNase II−/− cells retrofected with the mouse DNase II gene were incubated with apoptotic bodies derived from REFrm or REFrmh cells. Colonies per 1,000 cells after 2 weeks of hygromycin selection. C. DNA degradation by CAD does not inhibit propagation of transferred DNA. Wild type, DNase II−/−, or p21−/− cells were cocultivated with REFrmh apoptotic cells, and resistant colonies were selected for with hygromycin as described in Materials and Methods. Apoptosis in the donor cells was induced by nutrient depletion. As indicated, apoptosis was induced by γ-irradiation (150 Gy) in some of the donor cells to induce double-strand breaks. Colonies per 1,000 cells after 2 weeks of hygromycin selection. No hygromycin-resistant colonies were observed, where wild-type MEF cells were used as recipient cells. Interestingly, DNase II−/− MEFs acquired the hygR gene from REFrm apoptotic bodies killed by nutrient depletion, but induction of double-strand breaks by irradiation totally blocked propagation of transferred DNA. In contrast, p21−/− cells were able to acquire the hygR gene independent on whether the donor cells had been irradiated or not.

Close modal

Transfer of DNA into Chk2−/− Cells

Our data suggest that DNA degradation in the apoptotic cell in some way activates p53, and that this plays an important role in protecting the phagocytosing cell from replicating horizontally transferred DNA. It is, therefore, plausible to believe that the degraded apoptotic DNA is sensed by the DNA damage response of the phagocytosing cell, resulting in the induction of p53. Chk2 has been implicated as a mediator of p53 activation triggered by DNA damage; therefore, we assessed the role of Chk2 in the cellular response to fragmented DNA from apoptotic cells. For this purpose, we tested whether the hygromycin resistance gene (hygR) could be transferred to Chk2−/− MEF cells.

Apoptotic REFrmh cells were cocultured with MEF Chk2+/+ or MEF Chk2−/− cells. Apoptotic REFrm cells lacking the hygR gene (REFrm) were used as a negative control, and recipient MEF p21−/− cells served as a positive control. Selection for hygromycin-resistant cells was started 48 hours after the apoptotic bodies were added to the recipient cells. Resistant colonies, visualized by Coomassie staining, were detected 14 to 16 days later in plates where REFrmh apoptotic bodies had been cocultured with MEF Chk2−/− cells (Fig. 6A; Table 1). No colonies were detected in MEF Chk2+/+ cells cocultured with REFrmh cells or in MEF Chk2−/− cells fed with apoptotic bodies lacking the hygR gene. The frequency of colony formation was counted in triplicates as shown in Fig. 6B, where ∼1 of 10,000 MEF Chk2−/− cells had generated resistance to hygromycin treatment.

FIGURE 6.

The hygR gene is stably transferred to and propagated by MEF Chk2−/− cells after coculture with apoptotic REFrmh cells. A. MEF Chk2−/− cells cocultured with apoptotic REFrmh cells result in colony formation after selection with hygromycin. Apoptotic REFrm cells lacking the hygR gene serve as negative control. Coomassie staining shows colony formation in MEF Chk2−/− cells but not in the MEF Chk2+/+ cells cocultured with REFrmh. No colonies were detected where MEF Chk2+/+ or MEF Chk2−/− cells were cocultured with REFrm cells lacking the hygR gene. MEF p21−/− cells were used as positive control. B. PCR analysis of the presence of the hygR, c-myc, and H-rasV12 genes in the MEF Chk2−/− colonies that developed after coculture with REFrmh cells and selection with hygromycin. C. Analysis of phagocytotic activity of wild type (wt), Chk2−/−, and DNase II−/− MEFs. Apoptotic bodies from REFrm or REFrm-ICADLdm cells killed by nutrient deprivation were added to the indicated MEFs and incubated for 2 hours. The number of ingested apoptotic bodies per engulfing cells was analyzed and quantified by counting total number of ingested bodies per total number of cells. D. Frequency of colony formation in MEF Chk2−/− cells cocultured with apoptotic REFrmh cells. No colony formation was detected in MEF Chk2+/+ cells. Columns, average from three independent experiments; bars, SD. E. MEF Chk2−/− cells cocultured with REFrm apoptotic cells were not tumorigenic. MEF Chk2+/+, MEF Chk2−/−, and p21−/− cells cocultured with apoptotic REFrmh or REF cells were injected in the dorsal s.c. space of severe combined immunodeficient mice. Twelve of 15 mice injected with MEF p21−/− cells cocultured with REFrm cells developed tumors. No tumor development was detected in mice injected with MEF Chk2+/+, MEF Chk2−/−, and p21−/− cells cocultured with REF cells in vitro. From two independent experiments.

FIGURE 6.

The hygR gene is stably transferred to and propagated by MEF Chk2−/− cells after coculture with apoptotic REFrmh cells. A. MEF Chk2−/− cells cocultured with apoptotic REFrmh cells result in colony formation after selection with hygromycin. Apoptotic REFrm cells lacking the hygR gene serve as negative control. Coomassie staining shows colony formation in MEF Chk2−/− cells but not in the MEF Chk2+/+ cells cocultured with REFrmh. No colonies were detected where MEF Chk2+/+ or MEF Chk2−/− cells were cocultured with REFrm cells lacking the hygR gene. MEF p21−/− cells were used as positive control. B. PCR analysis of the presence of the hygR, c-myc, and H-rasV12 genes in the MEF Chk2−/− colonies that developed after coculture with REFrmh cells and selection with hygromycin. C. Analysis of phagocytotic activity of wild type (wt), Chk2−/−, and DNase II−/− MEFs. Apoptotic bodies from REFrm or REFrm-ICADLdm cells killed by nutrient deprivation were added to the indicated MEFs and incubated for 2 hours. The number of ingested apoptotic bodies per engulfing cells was analyzed and quantified by counting total number of ingested bodies per total number of cells. D. Frequency of colony formation in MEF Chk2−/− cells cocultured with apoptotic REFrmh cells. No colony formation was detected in MEF Chk2+/+ cells. Columns, average from three independent experiments; bars, SD. E. MEF Chk2−/− cells cocultured with REFrm apoptotic cells were not tumorigenic. MEF Chk2+/+, MEF Chk2−/−, and p21−/− cells cocultured with apoptotic REFrmh or REF cells were injected in the dorsal s.c. space of severe combined immunodeficient mice. Twelve of 15 mice injected with MEF p21−/− cells cocultured with REFrm cells developed tumors. No tumor development was detected in mice injected with MEF Chk2+/+, MEF Chk2−/−, and p21−/− cells cocultured with REF cells in vitro. From two independent experiments.

Close modal

The mouse origin of the colonies was verified by mouse-specific MHC staining (data not shown). Colonies from MEF Chk2−/− cells cocultured with REFrmh cells were harvested, and DNA was isolated from these cells. The presence of the hygromycin resistance gene in these cells was shown by PCR analysis (Fig. 6B). In addition, transfer of H-rasV12 and human c-myc could also be detected by PCR. These data show that Chk2-deficient cells are capable of propagating DNA from apoptotic bodies.

MEFs deficient for either the p53 or p21 genes are transformed by the uptake of apoptotic REFrmh cells in vitro and form tumors when injected into severe combined immunodeficient mice (4, 7). Because we could detect transfer of the c-myc and H-rasV12 oncogenes to the Chk2−/− cells by PCR, we tested whether this resulted in a tumorigenic phenotype. MEF Chk2−/− cells were cultured with apoptotic REFrm cells or nontransformed REF cells for 10 days before injection into severe combined immunodeficient mice. Twelve of 15 animals of the positive control, MEF p21−/− cells cocultured with apoptotic REFrm cells, formed tumors within 1 week and were euthanized due to tumor size within 30 days after injection. In contrast, no tumor development was detected in any of the mice injected with MEF Chk2−/− cells cocultured with apoptotic REFrmh cells (Fig. 6E; Table 1). These results show that Chk2-deficient cells are able to salvage and replicate DNA from apoptotic bodies but are not transformed by the transferred H-rasV12 and c-myc oncogenes.

Apoptotic Bodies Induce Chk2-Dependent Accumulation of p53 and p21 Proteins

Following its activation, Chk2 phosphorylates not only p53 but a set of effector molecules, including Brca1, E2F, pml, and Plk3, that are involved in checkpoint control (17). However, our previous data clearly indicate an essential role of the p53/p21 pathway in controlling replication of apoptotic DNA. We, therefore, assessed whether p53 is activated by the addition of apoptotic bodies and whether this induction was Chk2 dependent. Addition of apoptotic bodies to wild-type MEF cells resulted in rapid engulfment (Fig. 6C) and induction of p53 protein 8 hours later in the phagocytosing cells (Fig. 7A and D). Induction of the p21 target gene was also detected with similar kinetics. In contrast, coculture of MEF Chk2−/− with apoptotic bodies did not yield any detectable accumulation of p53 or p21 (Fig. 7B and D), although the frequency of phagocytosed dead cells did not differ. The p53/p21 pathway was intact in the Chk2-deficient cells, as both proteins could be induced after UV irradiation. Furthermore, the increased p53 and p21 protein levels were not derived from the engulfed apoptotic donor cells, as no positive signal could be detected in the apoptotic cells when cultured alone (Fig. 7C).

FIGURE 7.

Uptake of apoptotic bodies induces Chk2-dependent activation of p53. Apoptosis was induced in REFrm cells by nutrient depletion and added to MEF Chk2+/+ or Chk2−/− cells. Induction of p53/p21 by γ or UV irradiation served as positive controls. A. Induction of p53 and p21 was analyzed by Western blot in Chk2+/+ MEF cells after cultivation with apoptotic REFrm cells for 4, 8, 12, and 24 hours. Induction of both p53 and p21 was detected after 8 hours. B. In contrast, although UV irradiation triggered a readily detectable p53 response, addition of apoptotic bodies did not induce p53 or p21 in MEF Chk2−/− cells. C. No positive signal could be detected in lysates from apoptotic bodies cultured alone during the same incubation times. D. Quantification of p53 accumulation in three independent experiments. Bars, SD. E. Chk2-dependent accumulation of p53 in the nuclei of cells phagocytosing apoptotic bodies. Top, 4′,6-diamidino-2-phenylindole (DAPI) staining of cells cultured with apoptotic bodies. Middle, total p53 staining. Bottom, composite of 4′,6-diamidino-2-phenylindole + p53 stainings. Note that the apoptotic bodies stain negative for p53, whereas the recipient MEF Chk+/+ cells are positive. p53 staining was not detectable in Chk2−/− cells. Arrows, engulfed apoptotic bodies. Bar, 10 μm.

FIGURE 7.

Uptake of apoptotic bodies induces Chk2-dependent activation of p53. Apoptosis was induced in REFrm cells by nutrient depletion and added to MEF Chk2+/+ or Chk2−/− cells. Induction of p53/p21 by γ or UV irradiation served as positive controls. A. Induction of p53 and p21 was analyzed by Western blot in Chk2+/+ MEF cells after cultivation with apoptotic REFrm cells for 4, 8, 12, and 24 hours. Induction of both p53 and p21 was detected after 8 hours. B. In contrast, although UV irradiation triggered a readily detectable p53 response, addition of apoptotic bodies did not induce p53 or p21 in MEF Chk2−/− cells. C. No positive signal could be detected in lysates from apoptotic bodies cultured alone during the same incubation times. D. Quantification of p53 accumulation in three independent experiments. Bars, SD. E. Chk2-dependent accumulation of p53 in the nuclei of cells phagocytosing apoptotic bodies. Top, 4′,6-diamidino-2-phenylindole (DAPI) staining of cells cultured with apoptotic bodies. Middle, total p53 staining. Bottom, composite of 4′,6-diamidino-2-phenylindole + p53 stainings. Note that the apoptotic bodies stain negative for p53, whereas the recipient MEF Chk+/+ cells are positive. p53 staining was not detectable in Chk2−/− cells. Arrows, engulfed apoptotic bodies. Bar, 10 μm.

Close modal

Immunofluorescence staining confirmed the data from the Western blot analysis, showing that cells containing apoptotic bodies were p53 positive (Fig. 7E). The positive staining was only localized in the nuclei of the phagocytosing cells, as no staining was detected in the ingested apoptotic bodies. As expected from the Western blot results, the Chk2−/− recipient cells that had engulfed apoptotic bodies remained p53 negative, as indicated by arrows. In conclusion, engulfment of apoptotic cells trigger Chk2-dependent p53 accumulation and p21 activation in the phagocytosing cell.

The role of apoptosis in healthy individuals is to eliminate unwanted and potentially harmful cells. These apoptotic cells are rapidly cleared by macrophages or neighboring cells in vivo (18). We have previously shown that DNA from dying cells is transferred to living cells via the uptake of apoptotic bodies and may contribute to tumor progression. Here, we provide evidence that the apoptotic DNA degraded in the lysosomes by DNase II activates the DNA damage response of the phagocytosing cells and thus prevents propagation of potentially pathologic DNA acquired by engulfment of apoptotic cells.

In this study, we found that DNase II plays an essential role in the activation of p53 and the consequent inhibition of replicating transferred DNA. DNase II has been shown to be responsible for degradation of DNA from apoptotic bodies in macrophages where genetic inactivation of DNase II inhibits DNA degradation of engulfed dead cells, as shown by terminal deoxynucleotidyl transferase–mediated nick end labeling staining, resulting in accumulation of apoptotic bodies in the cytoplasm of macrophages (14). Furthermore, DNase II acts as a lysosomal barrier to transfection with nonviral expression vectors (19). This effect was specific for transfection methods that introduce DNA into cells by endocytosis but not other methods (e.g., electroporation). Our data indicate that the lysosomal degradation of DNA from the engulfed apoptotic cell is critical for the activation of p53 in the phagocytosing cell. Furthermore, joint inhibition of CAD and DNase II totally blocked p53 activation. These data argue in favor for the hypothesis that apoptotic DNA, similar to transfected DNA, may enter the nucleus via the endosomal/lysosomal pathway. We propose that the DNase II–cleaved DNA from engulfed apoptotic cells triggers a DNA damage response in the nucleus of the phagocyte (Fig. 8). We have indeed shown local colocalization of the DNA damage markers MRE11 and gamma H2AX with apoptotic DNA labeled with bromodeoxyuridine in the nucleus of the recipient cell.4

4

J. Ehnfors, unpublished data.

FIGURE 8.

Schematic figure of our hypothesis. We suggest that the Chk2/p53/p21 DNA damage pathway in the phagocytosing cell, together with the DNase II enzyme, form a genetic barrier against horizontally transferred DNA, blocking replication of foreign DNA from engulfed apoptotic cells. This genetic barrier may guard the genomic integrity of normal cells.

FIGURE 8.

Schematic figure of our hypothesis. We suggest that the Chk2/p53/p21 DNA damage pathway in the phagocytosing cell, together with the DNase II enzyme, form a genetic barrier against horizontally transferred DNA, blocking replication of foreign DNA from engulfed apoptotic cells. This genetic barrier may guard the genomic integrity of normal cells.

Close modal

The inhibition of the DNA degradation blocked induction of p53, and consequently, DNA transferred from apoptotic cells (hygR gene) could be replicated. Retransfection of the DNase II gene into the DNase II−/− cells restored the protection against replication of salvaged DNA, and no hygR colonies could be detected (Fig. 5B). In these retransfected cells, accumulation of p53 could be detected with a similar kinetics as wild-type MEF cells after coculture with apoptotic cells (Fig. 3C and F). It is also interesting to note that DNA was transferred to DNase II−/− recipient cells even when cocultured with apoptotic bodies with DNA degraded by CAD activity, whereas induction of double strand breaks using γ-irradiation completely blocked DNA transfer (Fig. 5A and C). A possible explanation to these apparently contradictory data is that the qualitative difference of the generated DNA fragments may affect the cellular response. CAD cleaves DNA generating, blunt-ended, double-stranded fragments carrying 5′-phosphate and 3′-hydroxyl groups, whereas lysosomal degradation mediated by DNase II generates 5′-hydroxyl and 3′-phosphate ends (20, 21). An alternative explanation could be that DNA cleaved by CAD in the serum-starved donor cells triggers a weaker/later p53 response in the DNase II−/− recipient cells compared with DNA with double-strand breaks in the γ-irradiated donor cells. A delayed or weak p53 response could result in recipient cells becoming more susceptible to horizontally transferred DNA.

We have previously shown that normal cells are able to take up DNA from apoptotic bodies but are subsequently cell cycle arrested, thus preventing DNA replication. Here, we show that uptake of apoptotic cells results in p53 induction of the phagocytosing cell, and that this is not only due to functional DNases but also to functional Chk2, as no detectable accumulation of p53 or p21 could be detected in Chk2-deficient cells. MEF cells lacking the Chk2 gene have lost the cell cycle checkpoint, as these cells could acquire and replicate the hygR gene from apoptotic cells (Table 1). Interestingly, we could also detect transfer of the H-rasV12 and c-myc oncogenes that, however, did not result in a transformed or a tumorigenic phenotype, as the hygromycin-resistant colonies became senescent when propagated in vitro (data not shown). This is in contrast to p53−/− or p21−/− recipient cells that readily form tumors after uptake of oncogenic DNA (Table 1). These findings indicated that Chk2 may primarily respond to the incoming DNA but is not protecting against tumor transformation. Because the Chk2-deficient cells contain an intact p19ARF gene, it is likely that the transferred oncogenes activate p53 via the p19ARF pathway and thus prevents transformation.

During its lifetime, the somatic cell is exposed to genotoxic stress, such as alkylating agents, oxygen radicals, and ionizing radiation. To this list of genomic insults, we would like to add DNA transfer from apoptotic bodies. The uptake of DNA may be a mere consequence of inadequate DNA degradation in the lysosomes, which is then transported into the nucleus. This may pose a serious threat to the integrity of the genome and may result in neoplastic transformation. We speculate that one role of the molecular pathway, starting with lysosomal DNA degradation leading to the Chk2-dependent p53 activation, is to protect the individual cell from propagating DNA taken up from apoptotic cells (Fig. 8). Malfunction of the checkpoint control may cause accumulation of genetic alterations, ultimately leading to cancer.

Cell Lines

REFs and MEFs from DNase II−/− p21−/−, Chk2−/−, and Chk2+/+ mice were grown in DMEM with glutamin, penicillin/streptomycin, and 10% fetal bovine serum (4, 10, 21). REFrmh cells containing the H-rasV12, human c-myc, and the hygR gene have been described (4). The REFrmh-ICADLdm (12) and transfection of DNase II into MEF DNase II−/− cells were generated as previously described (22), selected in puromycin (5 and 1 μg/mL, respectively; Sigma-Aldrich, St. Louis, MO).

Gene Transfer Experiments

Recipient MEFs (1 × 106) of indicated genetic backgrounds were plated in 10-cm Petri dishes. Apoptosis was induced in REFrmh cells by nutrient depletion, previously described (4), and verified by Hoechst 33258 and Annexin V staining. Five million apoptotic cells were added to each 10-cm Petri dish containing the MEF recipients. Forty-eight hours after addition of the apoptotic donor cells, the tissue culture medium was changed. Cells were grown in the presence of hygromycin (200 μg/mL; Sigma-Aldrich) to select for the uptake of the hygR gene in MEF cells. The resulting colonies were visualized by Coomassie staining.

PCR Analysis

DNA from hygromycin-resistant colonies was isolated with Qiaamp Blood kit (Qiagen, Chatsworth, CA). PCR analysis was done with specific primers and conditions for the human c-myc, the H-rasV12, and the hygR genes as previously described (4).

Tumor Growth

MEF Chk2+/+, MEF Chk2−/−, or MEF p21−/− cells (1 × 106) cocultured with apoptotic REFrmh cells in vitro for 10 days, as indicated in the figure legends, were injected in the dorsal s.c. space of 6- to 8-week-old severe combined immunodeficient mice. Tumor growth was examined by palpation. Animals that did not develop tumors within 3 months were scored as negative.

Western Blot

For analysis of p53 induction by Western blot, cells were harvested [100 mmol/L Tris (pH 8), 150 mmol/L NaCl, 1% NP40, protease inhibitors cocktail] at 4, 8, 12, and 24 hours after the addition of apoptotic bodies. Protein concentration was measured (A595, Bradford method), and samples were loaded (12% polyacrylamide gel, Bio-Rad, Richmond, CA). Transfer to Protean filter (Schleicher and Schuell, Keene, NH). Block with 5% milk for 1 hour at room temperature before incubation with primary antibody (anti-p53 FL-393, 1:400, Santa Cruz Biotechnology, Inc., Santa Cruz, CA and anti-p21, 1:200, BD PharMingen, San Diego, CA) overnight at 4°C or anti-β-actin (Sigma-Aldrich; 1:3,000) at room temperature for 1 hour, secondary antibody (anti-mouse-horseradish peroxidase or anti-rabbit-horseradish peroxidase NA 931 and NA 934V, 1:5,000; Amersham Pharmacia Life Science, Piscataway, NJ) for 2 hours at room temperature. The membrane was developed using detection system according to protocol of the manufacturer (Santa Cruz Biotechnology). MEF Chk2+/+ cells (1 × 106) were plated on 10-cm Petri dishes; 50 nmol/L of Bafilomycin A1 (Sigma-Aldrich) was added 2 hours before addition of the apoptotic REFrmh and REFrmh-Ldm cells. Cells were then cocultured with or without Bafilomycin A1 and analyzed by Western blot at 4, 8, 12, and 24 hours as described above.

Fragmentation Assay

Apoptosis was induced (nutrient depletion) for 24 hours in 5 × 106 REFrmh cells with or without ICAD-Ldm. DNA fragmentation was analyzed as described (2).

Immunofluorescence

MEF Chk2+/+ and MEF Chk2−/− cells were plated in chamber slides (Falcon, Lincoln, NJ) to adhere overnight. Nutrient depleted apoptotic cells were added and incubated for 4, 8, 16, and 24 hours, respectively. For total p53 staining, cells were fixed in methanol/acetone (1:1) for 10 minutes blocked with 5% horse serum and stained with primary antibody (anti-p53 FL-393, 1:100; Santa Cruz Biotechnology). Incubation for 30 minutes at room temperature with anti-rabbit FITC-conjugated secondary antibody (DAKO, Carpinteria, CA; 1:40), following mounting media containing 4′,6-diamidino-2-phenylindole (Vector Laboratories, Burlingame, CA).

Grant support: Swedish Cancer Society, Swedish Society of Medicine, Cancerföreningen Stockholm, Lennander's Foundation, and Karolinska Institutet.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1
de la Taille A, Chen MW, Burchardt M, Chopin DK, Buttyan R. Apoptotic conversion: evidence for exchange of genetic information between prostate cancer cells mediated by apoptosis.
Cancer Res
1999
;
59
:
5461
–3.
2
Holmgren L, Szeles A, Rajnavölgyi E, et al. Horizontal transfer of DNA by the uptake of apoptotic bodies.
Blood
1999
;
93
:
3956
–63.
3
Spetz AL, Patterson BK, Lore K, Andersson J, Holmgren L. Functional gene transfer of HIV DNA by an HIV receptor-independent mechanism.
J Immunol
1999
;
163
:
736
–42.
4
Bergsmedh A, Szeles, A, Henriksson M, et al. Horizontal transfer of oncogenes by uptake of apoptotic bodies.
Proc Natl Acad Sci U S A
2001
;
98
:
6407
–11.
5
Lane DP. Cancer. p53, guardian of the genome.
Nature
1992
;
358
:
15
–6.
6
Levine AJ. p53, the cellular gatekeeper for growth and division.
Cell
1997
;
88
:
323
–31.
7
Bergsmedh A, Szeles A, Spetz AL, Holmgren L. Loss of the p21(Cip1/Waf1) cyclin kinase inhibitor results in propagation of horizontally transferred DNA.
Cancer Res
2002
;
62
:
575
–9.
8
Holmgren L, Bergsmedh A, Spetz AL. Horizontal transfer of DNA by the uptake of apoptotic bodies.
Vox Sang
2002
;
83
Suppl 1:
305
–6.
9
Hirao A, Cheung A, Duncan G, et al. Chk2 is a tumor suppressor that regulates apoptosis in both an ataxia telangiectasia mutated (ATM)-dependent and an ATM-independent manner.
Mol Cell Biol
2002
;
22
:
6521
–32.
10
Takai H, Naka K, Okada Y, et al. Chk2-deficient mice exhibit radioresistance and defective p53-mediated transcription.
EMBO J
2002
;
21
:
5195
–205.
11
Enari M, Sakahira H, Yokohama H, Okawa K, Iwamatsu A, Nagata S. A caspase-activated DNase that degrades DNA during apoptosis, and its inhibitor ICAD.
Nature
1998
;
391
:
43
–50.
12
Sakahira H, Enari M, Nagata S. Cleavage of CAD inhibitor in CAD activation and DNA degradation during apoptosis.
Nature
1998
;
391
:
96
–9.
13
Sakahira H, Takemura Y, Nagata S. Enzymatic active site of caspase-activated DNase (CAD) and its inhibition by inhibitor of CAD.
Arch Biochem Biophys
2001
;
388
:
91
–9.
14
McIlroy D, Tanaka M, Sakahira H, et al. An auxiliary mode of apoptotic DNA fragmentation provided by phagocytes.
Genes Dev
2000
;
14
:
549
–58.
15
Bowman EJ, Siebers A, Altendorf K. Bafilomycins: a class of inhibitors of membrane ATPases from microorganisms, animal cells, and plant cells.
Proc Natl Acad Sci U S A
1988
;
85
:
7972
–6.
16
Nagata S, Nagase H, Kawane K, Mukae N, Fukuyama H. Degradation of chromosomal DNA during apoptosis.
Cell Death Differ
2003
;
10
:
108
–16.
17
Motoyama N, Naka K. DNA damage tumor suppressor genes and genomic instability.
Curr Opin Genet Dev
2004
;
14
:
11
–6.
18
Savill J, Fadok V. Corpse clearance defines the meaning of cell death.
Nature
2000
;
407
:
784
–8.
19
Howell DP, Krieser RJ, Eastman A, Barry A. Deoxyribonuclease II is a lysosomal barrier to transfection.
Mol Ther
2003
;
8
:
957
–63.
20
Rosl F. A simple and rapid method for detection of apoptosis in human cells.
Nucleic Acids Res
1992
;
20
:
5243
.
21
Kawane K, Fukuyama H, Kondoh G, et al. Requirement of DNase II for definitive erythropoiesis in the mouse fetal liver.
Science
2001
;
292
:
1546
–9.
22
Troyanovsky B, Levchenko T, Mansson G, Matvijenko O, Holmgren L. Angiomotin: an angiostatin binding protein that regulates endothelial cell migration and tube formation.
J Cell Biol
2001
;
152
:
1247
–54.