Requirement of ATR for maintenance of intestinal stem cells in aging Drosophila

The pS/TQ signals in Drosophila ISCs/EBs increase with age and under the influence of oxidative stress

We first tested whether the pS/TQ signal (immunostaining), a known marker of ATM/ATR activation in several organisms, is detected in Drosophila midgut cells and is modulated with age and by oxidative stress. The pS/TQ signal was detected weakly in Dl⁺ cells (ISCs) in the gut from 10-day-old wild-type flies (Figure 1A a-c, Figure Supplemental S1A). Nonetheless, it increased in an age-dependent manner in Dl⁺ cells in the gut from 20- (Figure 1A d-f) and 45-day-old (Figure 1A g-i) wild-type flies. A strong pS/TQ signal was also detected in Dl⁺ cells of 10-day-old Catⁿ¹/+ and PQ-treated flies (Figure 1A j-o). Quantitative pS/TQ signals in Dl⁺ cells of the gut from 20- and 45-day-old wild type, 10-day-old Catⁿ¹/+, and PQ-treated flies were increased 1.6-, 1.8-, 2.25-, and 6.2-fold, respectively, compared to 10-day-old wild-type flies (Figure 1B and 1C).

An age-related increase in pS/TQ signals was also detected in esg⁺ small cells (ISCs/EBs) of the posterior midgut of 10-, 20-, and 45-day-old esg>GFP flies, in 10-day-old esg>GFP+Catⁿ¹, and PQ-treated flies (Figure 1D). Quantitative pS/TQ signal levels in esg⁺ cells of 20- and 45-day-old wild-type, 10-day-old esg>GFP+Catⁿ¹, and PQ-treated flies were increased 3.2-, 4.2-, 5.1-, and 8.1-fold, respectively, compared to 10-day-old wild-type flies (Figure 1E and 1F). We also analyzed pS/TQ signals in differentiated cells and observed a few increase of the signal in ECs and EEs with age and oxidative stresses (Figure Supplemental S2). These results meant an increased pS/TQ signal in ISCs of the gut affected by age- and oxidative stress.

DNA damage-induced pS/TQ signals in Drosophila ISCs/EBs is more dependent on ATR than on ATM

We determined whether the pS/TQ signal is associated with DNA damage and is a dependable indicator of ATM/ATR activity in the adult Drosophila midgut by means of ionizing radiation (IR) as a DNA damage inducer and by means of flies with an ISC/EB-specific knockdown of ATM, ATR, or both under the esg^ts>GFP genotype. We examined pS/TQ signal strength in esg⁺ cells of the gut from esg^ts>GFP, esg^ts>GFP+ATMi+ATRi, esg^ts>GFP+ATRi, and esg^ts>GFP+ATMi flies 1 h after application of 5 Gy of γ-ray irradiation. Strong pS/TQ signals were detected in ISCs and EBs as well as ECs in irradiated esg^ts>GFP wild-type flies in contrast to the weak pS/TQ signals in the midgut cells of unirradiated flies (Figure 2A a-a'' and b-b''). The increase in the pS/TQ signal that was induced by γ-irradiation indicated that the pS/TQ signal was associated with DNA damage.

In contrast to the signal in wild type esg^ts>GFP flies, the γ-irradiation-induced increase of the pS/TQ signal was greatly reduced in esg⁺ cells (ISC/EB) of esg^ts>GFP+ATMi+ATRi flies with an ISC/EB-specific knockdown of both ATM and ATR (Figure 2A c-d''). The IR-induced increase in pS/TQ signals was also greatly decreased specifically in esg⁺ cells of the gut of esg^ts>GFP+ATRi flies with an ISC/EB-specific ATR knockdown (Figure 2A g-h'') and mildly decreased in esg⁺ cells of the esg^ts>GFP+ATMi gut (Figure 2A e-f'').

We also examined γH2AvD foci in esg⁺ cells of the gut from esg^ts>GFP, esg^ts>GFP+ATMi+ATRi, esg^ts>GEP+ATRi and esg^ts>GFP+ATMi flies 1 h after administration of 5 Gy of γ-irradiation. As expected, strong γH2AvD signals were detected in midgut cells including ISCs/EBs in irradiated esg^ts>GFP wild type flies (Figure 2B a-b''); however, the IR-induced increase in the γH2AvD signal was strongly decreased specifically in esg⁺ cells of the gut from esg^ts>GFP+ATMi+ATRi flies (Figure 2B c-d''). The IR-induced increase in the γH2AvD signal was also greatly decreased specifically in esg⁺ cells of the gut from esg^ts>GFP+ATRi flies (Figure 2B g-h'') and mildly decreased in esg⁺ cells of esg^ts>GFP+ATMi gut specimens (Figure 2B e-f'').

These results indicated that the pS/TQ signal that was induced by γ-irradiation in the ISCs/EBs of the Drosophila midgut was dependent on ATM/ATR, and the role of ATR was more important than that of ATM.

Next, to address whether the age-related increase of pS/TQ signal is dependent on ATM/ATR, we checked changes of the pS/TQ signal with age in esg⁺ cells of the gut from esg^ts>GFP, esg^ts>GFP+ATMi+ATRi, esg^ts>GFP+ATMi, and esg^ts>GFP+ATRi flies for 3 weeks at 27 °C since most of stem cells (esg⁺) in the posterior midgut were disappeared after 2 weeks at 29 °C. In aged gut specimens, a strong pS/TQ signal was detected in esg⁺ small cells of esg^ts>GFP flies (Figure 3A a-a'', Figure Supplemental S3), but the age-related increase of pS/TQ signals was not detected in esg⁺ small cells of esg^ts>GFP+ATMi+ATRi flies (Figure 3A b-b''). The age-relaed increase of pS/TQ signals was also greatly decreased specifically in esg⁺ cells of the gut specimens from esg^ts>GFP+ATRi flies (Figure 3A d-d'') and mildly decreased in esg⁺ cells of esg^ts>GFP+ATMi gut specimens (Figure 3A c-c''). We also checked changes of the γH2AvD signal with age in esg⁺ cells of gut specimens from esg^ts>GFP, esg^ts>GFP+ATMi+ATRi, esg^ts>GFP+ATMi and esg^ts>GFP+ATRi flies. In gut specimens from aged flies, a strong γH2AvD signal was detected in esg⁺ small cells of esg^ts>GFP flies (Figure 3B a-a''). Nonetheless, the age-related increase of γH2AvD signals was not detected in the gut from esg^ts>GFP+ATMi+ATRi flies (Figure 3B b-b''). The age-related increase of γH2AvD signals was also strongly decreased specifically in esg⁺ cells of the gut from esg^ts>GFP+ATRi flies (Figure 3B d-d'') and mildly decreased in esg⁺ cells of esg^ts>GFP+ATMi gut specimens (Figure 3B c-c'').

In addition, we checked changes of the pS/TQ signal after PQ treatment in esg⁺ cells of the gut from esg^ts>GFP, esg^ts>GFP+ATMi+ATRi, esg^ts>GFP+ATMi, and esg^ts>GFP+ATRi flies. In PQ-treated gut specimens, a strong pS/TQ signal was detected in esg⁺ small cells of esg^ts>GFP flies (Figure 4A a-b''), but the PQ-induced increase of pS/TQ signals was not detected in esg⁺ small cells of esg^ts>GFP+ATMi+ATRi flies (Figure 4A c-d''). The PQ-induced increase of pS/TQ signals was also greatly decreased specifically in esg⁺ cells of the gut specimens from esg^ts>GFP+ATRi flies (Figure 4A g-h'') and mildly decreased in esg⁺ cells of esg^ts>GFP+ATMi gut specimens (Figure 4A e-f''). We also checked changes of the γH2AvD signal after PQ treatment in esg⁺ cells of gut specimens from esg^ts>GFP, esg^ts>GFP+ATMi+ATRi, esg^ts>GFP+ATMi and esg^ts>GFP+ATRi flies. In gut specimens from PQ-treated flies, a strong γH2AvD signal was detected in esg⁺ small cells of esg^ts>GFP flies (Figure 4B a-b''). Nonetheless, the PQ-induced increase of γH2AvD signals was not detected in the gut from esg^ts>GFP+ATMi+ATRi flies (Figure 4B c-d''). The PQ-induced increase of γH2AvD signals was also strongly decreased specifically in esg⁺ cells of the gut from esg^ts>GFP+ATRi flies (Figure 4B g-h'') and mildly decreased in esg⁺ cells of esg^ts>GFP+ATMi gut specimens (Figure 4B e-f'').

These results indicated that the pS/TQ signal that was induced with age and by oxidative stress in the ISCs/EBs of the Drosophila midgut was dependent on ATM/ATR, and the role of ATR was more important than that of ATM.

Taken together, these results indicated that the pS/TQ signal that was induced by DNA damage in the ISCs/EBs of the Drosophila midgut was dependent on ATM/ATR, and the role of ATR was more crucial than that of ATM.

A knockdown of ATM/ATR in ISCs/EBs affects maintenance and proliferation of midgut ISCs

To understand the role of the age-related increase in pS/TQ signals (ATM/ATR activity) in midgut ISCs, we characterized the effects of a cell type-specific persistent knockdown of ATM, ATR or both in the posterior midgut. This knockdown was expressed in ISCs/EBs using esg^ts>GFP flies kept at 29 °C for 7 days.

Surprisingly, most esg⁺ cells in the posterior midgut of esg^ts>GFP+ATMi+ATRi and esg^ts>GFP+ATRi flies disappeared after culture at 29 °C for 2 weeks (Figure 5A, yellow dotted region) and in the gut of esg^ts>GFP+ATMi flies after culture at 29 °C for 3 weeks (data not shown). In the gut of esg^ts>GFP+ATRi, esg^ts>GFP+ATMi, and esg^ts>GFP+ATMi+ATRi flies kept at 29 °C for 7 days, the Dl levels in esg⁺ cells esg^ts>GFP+ATMi+ATRi and esg^ts>GFP+ATRi were reduced markedly, and esg⁺ cells were larger and more spherical, compared to their counterparts in the gut of esg^ts>GFP flies (Figure 5B). Under the same conditions, the shape of esg⁺ cells in the gut of esg^ts>GFP+ATMi flies was similar to that of the control (Figure 5B). The total number of cells in the same area decreased in the gut of esg^ts>GFP+ATMi+ATRi and esg^ts>GFP+ATRi flies (Figure 5C), but the ratio of esg⁺ cells to total cells was unchanged compared to the control (Figure 5D). The ratio of Dl⁺ cells (a marker of ISCs) to esg⁺ cells in the gut of esg^ts>GFP+ATMi+ATRi and esg^ts>GFP+ATRi flies was reduced to 2.31% and 10.65%, respectively, whereas that in the control was 41% (Figure 5E). The ratio Dl⁺ cells to esg⁺ cells in the gut of esg^ts>GFP+ATMi flies was 40.38% (Figure 5E). These results indicated that ATM and ATR were required for ISC maintenance, and the dependence on ATR was stronger than the dependence on ATM.

To determine the roles of ATM and ATR in ISC proliferation, we first analyzed the pattern and ratio of 5-ethynyl-2′-deoxyuridine (EdU)⁺ cells to esg⁺ cells in the gut with an ISC/EB-specific knockdown of ATM, ATR, or both. When ATM, ATR, or both were repressed in ISCs/EBs of esg^ts>GFP flies after culture at 29 °C for 7 days, the number of EdU⁺ cells was significantly decreased compared to the control (Figure 5F). The ratio of EdU⁺ to esg⁺ cells in the gut of 7-day-old esg^ts>GFP+ATMi+ATRi, esg^ts>GFP+ATRi, and esg^ts>GFP+ATMi flies kept at 29 °C was 6.9%, 7.1%, and 16.6%, respectively, whereas that in the control was 41.51% (Figure 5G). These results indicated that ATM and ATR were necessary for DNA replication in ISCs/EBs, and the role of ATR was more important than that of ATM.

Next, we examined the effects of an ISC/EB-specific knockdown of ATM, ATR, or both on the increase of ISC proliferation by PQ treatment. In gut specimens from PQ-treated flies, the number of proliferating cells increased in control esg^ts>GFP flies, but the PQ-induced increase of ISC proliferation was strongly attenuated in gut specimens with an ISC/EB-specific knockdown of both ATM and ATR (Figure 5H). The PQ-induced increase of ISC proliferation was also strongly attenuated in the gut specimens with an ISC/EB-specific knockdown of ATR, while being weakly reduced if at all in the gut specimens with an ISC/EB-specific knockdown of ATM (Figure 5H). These results indicated that ATM and ATR in ISC/EB were required for the PQ-induced increase in ISC proliferation, and the role of ATR was greater than that of ATM. Furthermore, to confirm the role of ATR in maintenance and proliferation of midgut ISCs, we generated clones homozygous for the null allele mei-41 encoding ATR using the mosaic analysis with a repressible cell marker (MARCM) method. Clones generated by MARCM method were allowed to grow for 7 days and stained for the ISC marker Delta (Figure 5I and Figure Supplemental S4). The number of cells per one clone in the posterior midgut was counted. The size of mei-41 null mutant clones was markedly reduced compare to that of control (Figure 5J). These results indicated that ATR was required for maintenance and proliferation of midgut ISCs.

Taken together, these results indicated that ATM and ATR in ISCs/EBs were required for ISC maintenance and proliferation, and the role of ATR was bigger than that of ATM.

Fly stock

Fly stocks were maintained at 25 °C on standard feed in the approximate 12 h/12 h light/dark cycle. The feed consisted of 79.2% of water, 1% of agar, 7% of cornmeal, 2% of yeast, 10% of sucrose, 0.3% of bokinin and 0.5% of propionic acid. To avoid larval overpopulation in all vials, 50–60 adult flies per vial were transferred to new food vials every 2–3 days for a period of 50–60 days or longer. We used the following transgenic RNAi strains of flies: UAS-ATM-RNAi²²⁵⁰² (#22502, VDRC, Vienna, Austria), UAS-ATM-RNAi¹⁰⁸⁰⁷⁴ (#108074, VDRC), UAS-ATR-RNAi¹¹²⁵¹ (#11251, VDRC), and UAS-ATR-RNAi¹⁰³⁶²⁴ (#103624, VDRC). The esg-GAL4,UAS-GFP/CyO flies were provided by the Drosophila Genetic Resource Center (DGRC, Kyoto, Japan). Temperature-inducible ISC/EB-specific esg-Gal80^ts flies [17] were obtained from B. Ohlstein. Catalase mutant (Catⁿ¹) flies were provided by the Bloomington Drosophila Stock Center (BDSC). Oregon-R flies served as the wild type. The esg^ts>GFP flies were the result of a cross of the Oregon-R males and esg-GAL4,UAS-GFP,tub-Gal80^ts/CyO (esg^ts) females. The esg^ts>GFP+ATMi+ATRi flies were obtained by crossing the UAS-ATRi/UAS-ATRi;UAS-ATMi/UAS-ATMi males and esg^ts females. For MARCM, we used P{ry[+t7.2]=hsFLP}1, P{w[+mC]=tubP-GAL80}LL1 w[*] P{ry[+t7.2]=neoFRT}19A; P{w[+mC]=UAS-mCD8::GFP.L}LL5 (#5134, BDSC), y[1] w[*]; P{w[+mC]=tubP-GAL4}LL7/TM3, Sb[1] Ser[1] (#5138, BDSC), P{ry[+t7.2]=neoFRT}19A; ry[506] (#1709, BDSC), and y[1] w[*] P{w[+mC]=lacW}G0221a P{lacW}mei-41[G0221b], l(1)G0221[G0221] P{neoFRT}19A/FM7c; P{ey-FLP.N}5 (#111381, DGRC). The results described in this study were obtained using female flies.

MARCM

Wild-type (WT) MARCM clones were induced by heat-shock for 45 min in 37 °C water bath in female flies of the genotype hsFLP, tubP-GAL80, neoFRT 19A/neoFRT 19A; UAS-mCD8::GFP/+; tubP-GAL4/+. The mei-41 MARCM clones were induced as described above in the genotypes hsFLP, tubP-GAL80, neoFRT 19A/mei-41[G0221b] neoFRT 19A; UAS-mCD8::GFP/+; tubP-GAL4/+ [37]. To assess cell lineage, markers were analyzed at 7 day after induction. Clonal composition was counted from posterior region in 9-10 midguts of each genotype.

Immunochemical analysis

An intact adult gut was dissected and fixed at room temperature. For staining with an anti-green fluorescent protein (GFP) antibody, the gut was fixed for 1 h in 4% formaldehyde (Sigma-Aldrich, St. Louis, MO, USA). For staining with anti-γH2AvD and anti-Dl antibodies, the whole-gut specimens were fixed for 30 min in 4% paraformaldehyde (Electron Microscopy Science, USA), dehydrated for 5 min in 50%, 75%, 87.5%, and 100% methanol, and rehydrated for 5 min in 50%, 25%, and 12.5% methanol in PBST (0.1% Triton X-100 in phosphate-buffered saline) for postfixing. Then, the specimens were washed with PBST and incubated overnight with a primary antibody at 4 °C. After washing with PBST, we incubated the specimens for 1 h with secondary antibodies at 25 °C, washed the specimens again in PBST, mounted them on Vectashield (Vector Laboratories, Burlingame, CA, USA), and analyzed them under a Zeiss Axioskop 2Plus microscope (Carl Zeiss Inc., Göttingen, Germany). For quantitative analysis of pS/TQ and γH2AvD, the images were processed in Photoshop (Adobe Systems Incorporated, San Jose, CA, USA). Wild-type cells and esg⁺, Delta protein⁺ (DI⁺), pS/TQ⁺, and EdU⁺ cells were counted in a 0.06 × 0.04-cm area of the posterior midgut. PH3⁺ cells were counted in the entire midgut.

Antisera

The following primary antibodies (diluted in PBST) were used in our experiments: mouse anti-Dl, mouse anti-Arm (Developmental Studies Hybridoma Bank, Iowa City, IA, USA), 1:200; mouse anti-GFP and rabbit anti-GFP (Molecular Probes, Eugene, OR, USA), 1:1000; rabbit anti-γH2AvD (Rockland, Gilbertsville, PA, USA), 1:2000; rabbit anti-pS/TQ (Cell Signaling Technologies, Danvers, MA, USA), 1:1000; rabbit anti-phospho-histone H3 (PH3, Millipore, Billerica, MA, USA), 1:1000; mouse anti-γ-tubulin (Sigma-Aldrich, St. Louis, MO, USA), at 1:1000 dilution. The following secondary antibodies (diluted in PBST) were used: a goat anti-rabbit fluorescein isothiocyanate (FITC) conjugate (Cappel, Solon, OH, USA), 1:400; a goat anti-rabbit Cy3 conjugate (Jackson ImmunoResearch, West Grove, PA, USA), 1:400; a goat anti-mouse FITC conjugate (Jackson ImmunoResearch), 1:400; and a goat anti-mouse Cy3 conjugate (Jackson ImmunoResearch), 1:400. We also used 4′,6-diamidino-2-phenylindole (DAPI, Molecular Probes) for staining, at a 1:1000 dilution.

γ-Irradiation

Adult flies were irradiated using a γ-ray machine [¹³⁷Cs, 21.275 tBq (575 Ci)] at the dose-rate of 2.55 Gy/min. After irradiation at the 5 Gy dose, the irradiated flies and unirradiated control fly vials were maintained at 25 °C [24].

PQ treatment

Flies were starved for 2 h prior to placement in empty in 10 mM PQ with 5% sucrose. After incubation at 25 °C for 18-20 h, whole-gut specimens were excised and analyzed by immunostaining.

5-Ethynyl-2′-deoxyuridine (EdU) incorporation

For Edu labeling, flies were starved for 2 h before being placed in vials containing 2.5 × 3.75 cm filter paper soaked in 100 μM EdU (Invitrogen, Grand Island, NY, USA) with 5% sucrose for 24 h. The gut was excised and fixed with 4% formaldehyde for 20 min. EdU incorporation was quantified using an EdU Alexa Fluor 488 Heat Shock Assay (Click-iT, Invitrogen) according to the manufacturer's instructions. After EdU incorporation, the whole-gut specimens were subjected to anti-Dl antibody staining as described previously [17, 38].

Analysis of pS/TQ signals

Fluorescent images were captured using the same exposure time in each experiment by means of a Zeiss AxioSkop 2 Plus microscope using the AxioVision Rel 4.8 software (Carl Zeiss Inc.). Individual pS/TQ signals could not be distinguished accurately in the gut of young flies. Therefore, Adobe Photoshop CS5.1, extended version (Adobe System Incorporated), was used to measure the fluorescence level of pS/TQ. Individual merged fluorescent images were split in three channels, and the specific channel for pS/TQ was used to analyze the fluorescence levels. Fluorescence levels were measured within the nucleus using the DAPI channel based on the boundaries defined using the Photoshop CS5 quick-selection tool. Mean fluorescence was analyzed after subtraction of the mean fluorescence of the background region (from three spots excluding the nuclear portion in the posterior midgut). The mean fluorescence of the background region was not significantly different from that for secondary-antibody-only controls (data not shown). The mean fluorescence of the background was used to determine the fluorescence threshold. The data were analyzed in Sigma Plot 10.0 (Systat Software Inc., San Jose, CA, USA). Fluorescence measurements are presented in arbitrary units. We excluded the pS/TQ signals of mitotic cells; these signals were strong in the gut of not only young but also aged flies (Figure Supplemental S1B).

Quantitative analysis of the cell number

For quantitative analysis of PH3-positive cells, we counted the number of such cells in the whole gut. After that, the total numbers of cells and GFP-, Dl-, or EdU-positive cells were evaluated in a visual field of the posterior midgut.

Statistical analysis

Quantified data are expressed as the mean ± SE. Significance testing was conducted using Student's t test.