2021 Volume 46 Issue 12 Pages 611-618
The gastrointestinal tract is exposed to a myriad of mutagens, making the DNA damage response (DDR) essential to maintain intestinal homeostasis. In vivo models to study DDRs are necessary to understand the mechanisms of disease development caused by genetic disorders such as colorectal cancer. A double-stranded break (DSB) in DNA is the most toxic type of DNA damage; it can be induced by either X-rays or chemicals, including anticancer agents. If DSBs in DNA cannot be repaired, cells can die by apoptosis to be removed from tissues. Here, we show that the DDRs observed as the phosphorylation of H2AX (γH2AX) and caspase-3-dependent apoptosis-induction are under critical control in the intestine of C57BL mice that were injected intraperitoneally with bleomycin, a natural glycopeptide used clinically as an antitumor agent. We found a significant increase in γH2AX expression 2–6 hr post-treatment in mouse ileum, cecum, and colon tissues by Western blotting and immunostaining. Apoptotic cells were observed after 6–24 hr by terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay and immunofluorescence of active caspase-3. We observed that γH2AX expression and apoptotic cells were distributed in the lower part of the crypt. The experimental protocol described here is a simple procedure that can be used generally as an in vivo intestinal toxicity assay. Our experimental approach provides a useful method for examining the effects of various bioactive compounds on the DDR, which is essential for understanding intestinal homeostasis.
The intestinal tract is inherently exposed to numerous mutagenic factors, such as dietary components, microflora, inflammation, and mitotic stress. Following tissue injury by a mutagenic factor, leukocytes that are recruited to the injury site release free radicals, such as reactive oxygen species (ROS) and nitrogen oxide reactive species, which target DNA, RNA, proteins, and lipids, resulting in damage to healthy neighboring cells (Hussain et al., 2003; Knaapen et al., 2006). Most DNA and tissue damage that is associated with inflammation occurs through the generation of ROS that cause oxidative stress in cells. Furthermore, cells exhibit various responses upon ROS exposure, which leads to DNA lesions and chronic inflammation that is activated by the NFκB signaling pathway (Ferguson, 2010). It has also been reported that several modified DNA products and their intermediates are induced by ROS and free radicals (Cooke et al., 2003).
Different types of DNA damage are sensed, signaled, and repaired by the DNA damage response (DDR) in cells, thereby avoiding apoptosis (Jackson and Bartek, 2009). Of the different types of DNA damage, double-stranded breaks (DSBs) are most hazardous to cells. DSBs are repaired by the non-homologous end joining (NHEJ) pathway, which is a major DNA repair pathway in mammalian cells (Mahaney et al., 2009). The NHEJ factor Ku70/Ku80 heterodimer (Ku) binds to the ends of DSBs independent of the cell cycle phase (Koike, 2002; Koike and Koike, 2008; Mahaney et al., 2009). Subsequently, Ku recruits DNA-PKcs and activates the DNA-dependent kinase complex (Khanna and Jackson, 2001; Koike et al., 1999; Shiloh, 2003). A critical factor in DNA repair is the histone protein H2AX, the mammalian form of which is rapidly phosphorylated at Ser-139 by DDR kinases, including ATM kinase and DNA-PK, to produce γH2AX at nascent DSBs sites (Rogakou et al., 1999; Rothkamm and Löbrich, 2003). γH2AX plays an essential role in the recruitment and accumulation of DDR proteins at DNA ends, resulting in the formation of γH2AX foci, including DNA damage sensors and repair protein complexes. Detection of these foci using anti-γH2AX antibody in immunofluorescence studies has been established as a sensitive assay for DSBs in DNA. The DDR kinases, such as ATM kinase and DNA-PK, also phosphorylate and activate p53, a protein that directs apoptosis to remove epithelial cells with damaged DNA (Barlow et al., 1997; Woo et al., 1998).
Previously, we have investigated inflammation-mediated intestinal tumorigenesis using aryl hydrocarbon (AhR)-deficient mice (Ikuta et al., 2013; Kawajiri et al., 2009). The biological function of AhR has been revealed in the context of immune cell development (Gutiérrez-Vázquez and Quintana, 2018) as well as in the antioxidant signaling pathway (Dietrich, 2016). Additionally, a novel AhR function for AhR in the DDR suggests that establishing a simple experimental system for analyzing the DDR in mouse intestinal tissue is important and essential. (Dittmann et al., 2016; Gronke et al., 2019).
The aim of the current study was to investigate the DDR in the intestinal tracts of normal mice that received intraperitoneal injections with bleomycin (Cornelissen et al., 2011), which induces DSBs in DNA. Intestinal tissues were examined to determine and evaluate the time course of DSBs formation and apoptosis in cells by immunohistochemical analysis. Our findings, and the simple procedures used in this study, can lead to a better understanding of the DDR, which is essential for intestinal homeostasis.
Six-week-old male C57BL mice were purchased from CLEA Japan Inc. (Tokyo, Japan). The mice were housed in a specific pathogen-free facility under a 12 hr light cycle and given free access to water and chow (Funabashi Farm, Chiba, Japan) in the animal house of the Saitama Cancer Center. The mice had a one-week quarantine period after delivery. To induce DSBs in DNA, mice received an intraperitoneal injection of 100 μg of bleomycin (#203408, Merck, Darmstadt, Germany) (Cornelissen et al., 2011) dissolved in water. To remove the intestines for our study, the mice were euthanized by cervical dislocation at 0, 1, 2, 6, 24, 48 hr after ip injection (n = 1 at each time point). Intestinal tissue was separated at the anterior and posterior of the cecum. The 5-cm segment at the end of the small intestine was isolated as the ileum, and the 3-cm segment below the cecum was used as colonic tissue. This series of experiments was repeated once so that a total of twelve mice were studied. This study was given prior approval by the Institutional Animal Committee and was conducted following the Saitama Cancer Center animal experimentation regulations.
Western blottingIntestinal tissues were lysed in buffered sodium dodecyl sulfate (SDS) (Ikuta et al., 2013; Kawajiri et al., 2009), and the supernatants from lysed samples were separated by electrophoresis on 15% SDS-polyacrylamide gels, followed by transfer to nitrocellulose membranes. Following incubation with 4% bovine serum albumin (BSA) in Tris-buffered saline (20 mM Tris-HCl, 150 mM NaCl, pH 7.5) to block nonspecific binding, the membranes were incubated with anti-γH2AX antibody (#9718, Cell Signaling Technology, Beverly, MA, USA) or anti-actin (A2066, Sigma-Aldrich, St. Louis, MO, USA) antibody, followed by incubation with an appropriate horseradish peroxidase (HRP)-conjugated secondary antibody. Detection was performed using an Immobilon Western Chemiluminescence HRP substrate (Millipore, Burlington, MA, USA). Blots corresponding to the proteins of interest were visualized using an ImageQuant LAS 4000 instrument (GE Healthcare, Chicago, IL, USA).
ImmunohistochemistryFreshly excised tissues from mice were frozen in Tissue-Tek OCT compound (Sakura Fine Technical Co., Ltd., Tokyo, Japan) and sectioned into 5 μm thick sections using a cryostat microtome (CM1950, Leica Microsystems, Wetzlar, Germany). Sectioned tissues were fixed with 4% formaldehyde diluted in phosphate buffered saline, followed by immersing in 100% methanol for 5 min at −30°C, then blocked with 4% BSA supplemented with 0.3% Triton X-100 for 10 min. The antibodies used for immunostaining were as follows: anti-γH2AX antibody (rabbit monoclonal, #9718, 1:1000 dilution), anti-cleaved caspase-3 antibody (rabbit polyclonal, #9661, Cell Signaling Technology, 1:200 dilution), anti-β-catenin antibody (rabbit polyclonal, C2206, Sigma-Aldrich, 1:1000 dilution), FITC-conjugated anti-EpCAM antibody (#118207, Biolegend, San Diego, CA, USA, 1:1000 dilution), and FITC-conjugated anti-CD45 antibody (#147709, Biolegend, 1:1000 dilution). After an overnight incubation with primary antibody, tissue sections were incubated with rhodamine-labeled anti-rabbit IgG (AP182R, Merck, 1:1000 dilution) for 1 hr at room temperature. Apoptotic cells were examined by double staining with immunofluorescence using anti-caspase-3 antibody, followed by terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) using an in situ Apoptosis Detection kit (#MK500, Takara Bio, Shiga, Japan), according to the manufacturer’s instructions. For quantitative analysis, cells were counted (100–600 cells) to show the rate of positive cell (%).
Using immunoblotting, we performed a series of time-course experiments to analyze the DDR induced by intraperitoneal injections of bleomycin in tissues of the ileum, cecum, and colon (Fig. 1A). In each intestinal region, we found a significant increase in γH2AX expression at 2 hr post-treatment compared with non-treated animals. This increase in expression was maintained for up to 6 hr and returned to basal levels after 48 hr (Fig. 1B). Tissues from each intestinal region exhibited a similar γH2AX expression pattern.
Western blot and densitometric analysis of γH2AX. After injection of bleomycin, mice were sacrificed at the indicated times to isolate intestinal tissues. (A) γH2AX protein levels in the ileum, cecum, and colon were determined and (B) quantitated by ImageJ. Actin was used as a loading control. Values are expressed as means ± S. D. of three experiments. *, significant increase compared to non-treated (P < 0.05, Student t-test).
Next, we examined the expression of γH2AX in different cell types by immunohistochemical analysis using coincubation with anti-EpCAM (Fig. 2A) or anti-CD45 (Fig. 2B) to identify epithelial cells and leukocytes, respectively. A semi-quantitative analysis of positive signals in γH2AX was performed to show changes throughout the time-course and site distribution (Supplemental Fig. 1A, B). For the ileum, bleomycin injection increased γH2AX levels in the nuclei. γH2AX levels reached a maximum after 2 hr, and expression levels returned to basal levels by 24 hr. Meanwhile, EpCAM or CD45 localized to the cell membrane. We also found that γH2AX was distributed mainly in the intestinal crypts and that some cells in the lamina propria exhibited double staining with anti-γH2AX and anti-CD45 antibodies (Fig. 2B). Similarly, we observed notable expression of γH2AX at 1–2 hr in the cecum (Fig. 3A) and at 2 hr in the colon (Fig. 3B), which was localized in the lower region of the crypts. In contrast, γH2AX expression was rarely observed in the upper cryptic region.
Immunohistochemical analysis of γH2AX in the ileum. Mice were sacrificed after bleomycin treatment at the indicated times. Cryostat sections were incubated with an anti-γH2AX antibody (red). Immunostaining was performed using anti-EpCAM antibody (A) or FITC-anti-CD45 (B) antibody (green). Cells that exhibit double staining for anti-γH2AX and anti-CD45 are indicated with arrows. 4′,6-Diamidino-2-phenylindole (DAPI) was used to counterstain nuclei (blue). Scale bar = 40 μm (A), 50 μm (B). Each experiment was repeated twice, and comparable results were obtained.
Immunohistochemical analysis of γH2AX in the cecum (A) and colon (B). Cryostat sections were incubated with anti-γH2AX antibody (red) and anti-EpCAM antibody (green). DAPI was used to counterstain nuclei (blue). Scale bar = 50 μm (A), 20 μm (B). Each experiment was repeated twice, and similar results were obtained.
We then evaluated apoptosis in bleomycin-treated cells using a TUNEL assay. Tissue sections were incubated with an anti-β-catenin antibody to identify epithelial cells rather than using fluorescein isothiocyanate (FITC)-conjugated anti-EpCAM. In each intestinal region, we observed that TUNEL-positive cells were rarely detected at 0 or 2 hr post-treatment. However, we found marked evidence of apoptosis at 6 hr after treatment in the lower part of the crypts. Furthermore, there were some TUNEL-positive cells at 24 hr post-treatment that returned to basal levels at 48 hr (Fig. 4, Supplemental Fig. 2). We also evaluated apoptosis by immunohistochemical staining using anti-cleaved (active) caspase-3 antibody (Fig. 5). This staining revealed partial co-localization of a cleaved caspase-3-positive signal with a TUNEL-positive signal at 6 hr. Neither signal was detected after 2 hr. Since γH2AX expression was still detected at 6 hr in the ileum and the cecum; double staining with TUNEL was performed at this time point and no double-stained cells were found (Supplemental Fig. 3). Although there are subtle differences between tissue types, γH2AX expression was observed 1–6 hr after administration and apoptosis was detected 6–24 hr after administration.
Distribution of apoptotic cells examined by TUNEL assays. Mice were sacrificed after bleomycin treatment at the indicated times to isolate the ileum, cecum, and colon. Frozen sections from each intestinal tissue were incubated with an anti-β-catenin antibody (red) followed by TUNEL analysis (green). DAPI was used to counterstain nuclei (blue). Scale bar = 40 μm.
Distribution of apoptotic cells examined by anti-cleaved caspase-3 antibody and TUNEL assays. Cryostat tissue sections were examined by immunostaining with anti-cleaved caspase-3 antibody (red) and by TUNEL assays (green). DAPI was used to counterstain nuclei (blue). Cells that double-stained with anti-caspase-3 and TUNEL are indicated by arrows (6 hr). Mice were sacrificed at the indicated times after bleomycin treatment. Scale bar = 30 μm.
The primary purpose of this study was to develop a simple experimental method to investigate the DDR in intestinal epithelia. We administered bleomycin to mice by intraperitoneal injections. The large surface area of the peritoneal cavity, considerable blood supply, and lymphatic transport facilitated the rapid absorption of the drug, resulting in its effective transfer from the peritoneal cavity to systemic circulation (Kuzlan et al., 1997; Al Shoyaib et al., 2020). This method of administration established a simple experimental system for evaluating DDR activation throughout the intestine.
Further, we have described a series of time-course experiments in mice to evaluate γH2AX expression, which is a marker of DDR activation triggered by DSBs in DNA. Based on Western blot analysis, we observed increased γH2AX expression levels as early as 1 hr post-bleomycin administration in the ileum, cecum, and colon. Levels returned to basal values 24–48 hr post-treatment. Immunostaining results showed a time course nearly identical to the Western blot analysis, with some differences between intestinal tissue types. The observed changes in γH2AX expression levels suggest that bleomycin-induced DSBs activate the DDR in the intestine and that some of the DSBs may be repaired by a DNA repair pathway. Furthermore, we observed that most of the cells immunoreactive with anti-γH2AX antibody were of epithelial origin and were localized in the lower portion of the crypts, an area that is characterized by active cell proliferation. We hypothesized that cells with proliferative capacities such as tissue stem cells or transient amplifying cells are repaired to establish and maintain normal tissue function. Differentiated cells, on the other hand, are not repaired and are eventually extruded from the surface of epithelia into the gut lumen. A previous study reported that DNA damage caused by irradiation is repaired more efficiently in intestinal crypt base stem cells compared with differentiated cells (Hua et al., 2012). In addition, the study demonstrated that BRCA1 and RAD1, which are markers of homologous recombination, as well as DNA-PK, which is required for the NHEJ pathway, were rapidly incorporated in order to repair foci in crypt base columnar cells and transient amplified cells, but not in differentiated cells, following irradiation. Thus, our protocol for the detection of γH2AX provides a suitable method for investigating DNA damage and activating DDR in the intestine.
We also evaluated apoptotic cells using TUNEL assays and immunofluorescence with anti-cleaved caspase-3. Although not detected until 2 hr post-treatment in TUNEL assays, we observed maximum signal levels at 6 hr after treatment that returned to basal levels at 48 hr. It is well known that TUNEL-positive cells are apoptotic cells, as are cells with DSBs under certain conditions (Takata et al., 2013). However, in our experiments, results from TUNEL assays did not reflect the formation of DSBs in DNA as indicated by anti-γH2AX because TUNEL did not respond up to 2 hr, despite strong expression of γH2AX. In addition, immunostaining with anti-γH2AX and TUNEL assay did not generate double staining signal in tissue samples treated for 6 hr. Active caspase-3-positive cells were detected mainly after 6 hr along with TUNEL-positive cells. Therefore, the TUNEL assay detects apoptosis rather than DSBs in DNA under our experimental conditions. The degree of overlap between TUNEL assays and cleaved caspase-3 was different among tissue types. In the cecum, most TUNEL-positive cells were also positive for caspase-3. Alternatively, in the colon, only a few cells exhibited overlap of these signals. We speculate that the difference between the colon and the other two tissues, cecum and ileum, might depend on the difference of caspase-3 activity among the three tissues, although further studies are required to clarify this point. Altogether, our findings suggest that the formation of DSBs in DNA that is triggered by bleomycin treatment might induce γH2AX foci for DNA repair mainly in the lower part of the crypt at 2–6 hr after treatment. Subsequently, cells that fail to repair DNA damage might undergo caspase-3-dependent apoptosis after 6 hr to be excluded from tissue. However, further analysis is required to explain the discrepancy between TUNEL and anti-cleaved caspase-3 assays in each tissue studied.
It is well known that bleomycin causes cellular toxicity through ROS production (Allawzi et al., 2019). As a physiological stimulus, the generation of ROS in gastrointestinal tract inflammation (Aviello and Knaus, 2017) induces DDR in the area of the lower crypt. In circumstances when the DDR is not effective, cells with DNA mutations may survive and proliferate to become cancer-initiating cells. Frick et al. (2018) demonstrated upregulation of oxidative stress in colon tissue in IL10-knockout mice. In addition, γH2AX-positive nuclei are increased in inflamed and dysplastic tissue samples, predominantly in the basal crypt region and the intermediate zone. In this study, we observed that apoptotic cells were localized in the lower part of the crypts but were not included in the differentiated cell layer that corresponded with γH2AX localization. In cells with severe DNA damage, activation of p53 can result in apoptosis to eliminate such critically injured cells. Arai et al. (1996) investigated the spatial relationship between p53 and apoptosis after irradiation in rat intestine. They reported that accumulation of p53, as well as apoptotic cells, were restricted to the crypt base. Altogether, we speculate the presence of a zone sensitive to stimulation that triggers DNA damage to protect stem cells and progenitor cells in the epithelia in the intestine.
In this study, we investigated the DDR in the mouse intestine after intraperitoneal injection of bleomycin. Our findings suggest that bleomycin-triggered DDR is different not only between proliferative cells and differentiated cells in the same tissue, but also among the three intestinal tissues examined. The experimental procedure used in this study is a very simple in vivo toxicity assay. Our experimental approach can provide a useful procedure for examining the effects of various bioactive compounds on the DDR, which is essential for intestinal homeostasis.
This work was supported by the Grants-in-Aid for Scientific Research, number 20K07437. The authors would like to thank Enago (www.enago.jp) for the English language review.
Conflict of interestThe authors declare that there is no conflict of interest.