1. Introduction
The CtIP protein was originally identified on the basis of its association with the CtBP transcriptional co-repressor [
1] and the BRCA1 tumor suppressor [
2,
3]. However, its physiological functions remained obscure until CtIP was recognized as the mammalian ortholog of Sae2 [
4], a factor that promotes end resection of DNA double-strand breaks (DSBs) in the budding yeast
Saccharomyces cerevisiae [
5]. In this capacity, CtIP facilitates the processing of DSB ends by MRN, a trimeric protein complex comprised of the Mre11 nuclease, Rad50, and NBS1 (Xrs1 in budding yeast) [
4,
6,
7,
8]. In eukaryotic cells, DSBs are repaired primarily through either of two pathways: canonical non-homologous end-joining (c-NHEJ) or homology-directed repair (HDR). While c-NHEJ is active throughout the cell cycle, HDR is largely restricted to the S and G2 phases, when sister chromatids are available to serve as templates for homologous recombination. DNA resection favors HDR by converting DSB ends into 3′-single-strand DNA (ssDNA) overhangs, which initiate ATR-dependent checkpoint signaling and, when assembled into a Rad51/ssDNA nucleofilament, can invade a homologous DNA duplex [
6]. In addition to the c-NHEJ and HDR pathways, DSBs can also be repaired in some settings by alternative end ligation (alt-EJ), which entails the annealing of ssDNA microhomologies generated at both DNA ends by short-range resection. Although CtIP is required for DSB repair by alt-EJ in most cell types [
9,
10], it is dispensable for alt-EJ-mediated class switch recombination of the immunoglobulin heavy chain (IgM) gene during B cell development, presumably because the highly repetitive and GC-rich switch region sequences provide optimal substrates for resection-independent alt-EJ [
11].
Given its central role in DSB repair choice, the molecular factors that initiate DNA end resection are of keen interest [
6]. The Mre11 subunit of the MRN complex possesses both an endonuclease activity and a 3′-5′ exonuclease activity. Short-range DNA resection is triggered by Mre11-mediated endonucleolytic cleavage of the 5′-terminated DNA strand at a position internal to the DSB end [
12,
13,
14]. The 3′-5′ exonuclease activity of Mre11 then digests the 5′-strand, moving from the internal nick toward the DSB end. The 3′-ssDNA overhang generated by this process can be further extended (long-range resection) by the Exo1 exonuclease or by the coordinated actions of the DNA2 endonuclease and a RecQ-family helicase [
15,
16]. Importantly, the “clipping” reaction that initiates short-range resection is promoted by CtIP
Sae2, which acts as an allosteric co-factor for the endonuclease activity of Mre11 [
6,
7,
12,
13,
14]. In addition, CtIP also facilitates long-range resection by promoting the recruitment of Exo1 to DSB ends [
17] and stimulating both the exonuclease and helicase activities of the DNA2/BLM complex [
18,
19]. While yeast Sae2
CtIP is only required for the resection of blocked DNA ends, such as the Spo11-induced breaks generated during meiosis, mammalian CtIP
Sae2 is also critical for MRN-mediated processing of unblocked ends [
6]. As such, mammalian CtIP controls a pivotal decision point in the DNA damage response by initiating DSB repair through either the HDR or alt-EJ pathways. In doing so, it also shapes the timing and duration of cellular responses to DSBs. For example, CtIP
Sae2-mediated resection both mitigates ATM
Tel1 signaling by removing the MRN
MRX complex from DSB ends and activates ATR
Mec1 signaling by promoting the formation of ssDNA/RPA filaments [
20,
21,
22]. Given the integral functions of CtIP in DNA resection and cell cycle checkpoint control, it is not surprising that Ctip-null mice undergo early embryonic lethality, and in most settings, cell viability is lost in the absence of CtIP [
7,
23,
24,
25].
The relationship between human CtIP and yeast Sae2 was originally recognized on the basis of a short amino acid homology between the carboxy-terminal sequences of both proteins (
Figure S1) [
4,
7]. CtIP/Sae2 orthologs that function in DNA resection and share this C-terminal Sae2-like homology domain have since been found throughout the eukaryotic kingdom [
26,
27,
28]. A striking feature of the Sae2-like domain is the presence of two highly conserved phosphorylation sites, one of which serves as a substrate for cyclin-dependent kinases (CDKs) [
29,
30] and the other for certain phosphoinositide 3-kinase-related protein kinases (PI3KKs), such as ATR and ATM [
17,
31]. Phosphorylation of the CDK site (T847 in human CtIP) is essential for CtIP to promote the MRN clipping reaction that initiates DNA resection [
13,
29,
30]. Mechanistically, T847-phosphorylated CtIP induces clipping by binding the FHA/BRCT domains of NBS1, which in turn allows NBS to elicit the endonuclease activity of Mre11 [
32]. Likewise, phosphorylation of the corresponding residue of Sae2 (S267) also promotes clipping by the MRX complex in yeast [
33]. Accordingly, mutations that ablate this site (e.g., CtIP-T847A or Sae2-S267A) cause genomic instability and severely impair resection-dependent cellular functions such as HDR and ATR-dependent checkpoint activation [
29,
30]. Interestingly, the viability of Ctip-null mice can be restored by ectopic expression of the wildtype but not the T847A-mutant human CtIP protein [
34]. In light of this observation, it is possible that animal development is dependent on CDK-mediated CtIP-T847 phosphorylation and the MRN clipping reaction that it sets in motion.
The C-terminal Sae2-like homology domain of CtIP also possesses a highly conserved phosphorylation motif (T859 in human CtIP and T855 in mouse Ctip) recognized by the ATR and ATM kinases. In response to DNA damage, ATR/ATM-mediated phosphorylation of CtIP-T859 facilitates the stable association of CtIP with DSB-containing chromatin and promotes optimal DNA resection and HDR repair of DSBs [
17,
31]. Although ATR/ATM phosphorylation of this site also promotes DNA resection in B lymphocytes, it is not required for programmed immunoglobulin gene rearrangements such as V(D)J recombination and class switch recombination [
25,
35]. Here we show that mice bearing a missense mutation that eliminates this phosphorylation site (Ctip-T855A) develop normally but are hypersensitive to genotoxic stress. Moreover, cells from these mice display marked defects in genotoxin-induced DNA resection, HDR repair of DSBs, and maintenance of chromosome stability. These results indicate that ATR/ATM phosphorylation of Ctip-T855 enhances DNA resection in response to genotoxic stress, but unlike CDK phosphorylation of Ctip-T847 DNA, it is dispensable for normal animal development.
2. Materials and Methods
2.1. Generation of CtipT855A Mice
All mice were housed in an AAALAC-accredited facility at Columbia University Irving Medical Center and studied according to protocols approved by the Columbia University Institutional Animal Care and Use Committee. To generate mice harboring the
CtipT855A allele, a knock-in targeting vector including
Ctip exons 16–19 was constructed by inserting a neomycin-resistance gene cassette flanked by loxP sites (loxP-PGK-neo-loxP) into intron 18 and replacing the T855 codon (ACT) of exon 18 with a codon for alanine (GCT) (
Figure S2a,b). KV1 embryonic stem (ES) cells [
36] were then electroporated with the targeting vector. Properly recombined neomycin-resistant ES clones were identified by both Southern analysis and the presence of the desired T855A coding mutation confirmed by nucleotide sequence analysis of a 565 base-pair DNA fragment generated by PCR amplification with a forward Ctip-19 (TCAGTAAAATGCCACTCTGG) and reverse Ctip-O (TGCCCACTTTTGAAGGCACTGAGTC) oligonucleotide. Two independent clones of
CtipT855A-neo/+ ES cells were injected into C57BL/6J blastocysts for the production of germline-transformed mice. The loxP-PGK-neo-loxP cassette was then excised from the targeted allele (
Figure S2c) by mating chimeric male
CtipT855A-neo/+ mice with females carrying a ubiquitously expressed
Cre transgene (EIIa-Cre; B6.FVB-Tg(EIIa-cre)C5379Lmgd/J; strain 003724, (Jackson Laboratory, Bar Harbor, ME, USA). Offspring heterozygous for the desired
CtipT855A allele (
Figure S2d) were identified by PCR amplification of tail DNA with a forward Ctip-20 (5′-GCCTGGCTTGGCATTCAGATTTCA-3′) and reverse Ctip-T (5′-ATGCCATAGACCTCATAGTC-3′) oligonucleotide, which generates distinct DNA fragments for the wildtype
Ctip+ (338 base pairs) and mutant
CtipT855A (248 base pairs) alleles. The
CtipT855A/+ mice were then backcrossed twice with pure C57Bl/6J mice (Jackson Laboratory stock 000664) to yield animals that were approximately 94% C57BL/6J (N3 backcrossed), and the mouse embryonic fibroblast (MEF) and embryonic stem (ES) cell lines were generated using mice on this background. All bred mice and all derived cell lines were (1) genotyped by PCR amplification as described above to distinguish the
Ctip+ and
CtipT855A alleles and (2) subjected to nucleotide sequence analysis to confirm the presence of the T855A codon on the
CtipT855A allele (
Figure S2e). To evaluate IR sensitivity, five-week-old littermates of
Ctip+/+,
CtipT855A/+, and
CtipT855A/T855A mice were subjected to whole-body radiation (7.5 Gy) inside the turnable chamber of an MKI-30 SN 1172 Cesium-137 irradiator and then monitored for survival for 3 months. The
CtipT855A mouse strain is available from Jackson Laboratory (strain # 037396).
2.2. Generation of Cell Lines
Mouse embryonic fibroblasts (MEFs) were derived from the progeny of a
CtipT855A/+ intercross by obtaining embryos from pregnant mothers at 13.5 days post-fertilization. Primary MEFs were cultured and immortalized with SV40 large T antigen as described [
37], and the individual
Ctip+/+,
CtipT855A/+, and
CtipT855A/T855A MEF clones used in this study are listed in
Table S1. The isogenic pairs of
Ctip+/+ and
CtipS326A/S326A MEFs were described previously [
38]. Embryonic stem (ES) cells were derived from day E3.5 blastocysts and cultured as described previously [
37]. Independent subclones of
Ctip+/+ (A, B, and C) and
CtipT855A/T855A (D, E, and F) ES cells possessing a DR-GFP recombination reporter integrated into the Pim1 locus were then generated using the p59xDR-GFP6 plasmid [
39]. The isogenic pair of
Brca1+/+ and
Brca1S1598F/S1598F ES subclones with a Pim1-integrated DR-GFP reporter were described previously [
40].
2.3. Clonogenic Survival Assays
For cell survival assays, immortalized MEFs were seeded on 6-well plates at 1000 cells/well and evaluated in triplicates for each experimental condition, as described previously [
37]. Thus, at 48 h after plating, cells were subjected to varying doses of ionizing radiation (0, 2, 4, 6, or 8 Gy) using an Atomic Energy of Canada Gammacell 40 Cesium unit or ultraviolet radiation (0, 0.34, 0.68, 1.36, 2.04, or 4.08 J/m
2) using a UVGL-58 lamp (short-wave UVC calibrated with UVX-25 radiometer), or exposed to mitomycin C (0, 50, 100, 200, or 800 ng/mL) for 4 h, hydroxyurea (0, 0.06, 0.26, 1.02, or 4.10 µM) for five hours, neocarzinostatin (0, 25, 50, 75, 100, or 200 ng/mL) for one hour, camptothecin (0, 0.05, 0.10, 0.20, 0.40, or 1.00 µM) for one hour, or etoposide (0, 0.1, 0.3, 0.8, 2.0, or 5.0 µM) for one hour. Treated and mock-treated cells were then washed twice with PBS and cultured in fresh media. At 5–7 days post-treatment, the cells were stained with 0.2% crystal violet in a 50% methanol solution, and the surviving colonies (containing > 50 cells) were counted. For olaparib treatment, MEFs were exposed to varying concentrations (0, 0.064, 0.16, 0.4, 1.0, or 2.5 µM) at 24 h after plating, and the media were replaced every 48 h with fresh media containing the corresponding concentrations of olaparib until the time of cell harvest, 6–8 days after initial drug treatment.
2.4. Alkaline Comet Assays
To assess DNA damage using the alkaline comet assay, immortalized MEFs were seeded onto 12-well plates and treated the following day with either 10 Gy ionizing radiation (IR), 1 µM camptothecin (CPT) for one hour, 2 mM hydroxyurea (HU) for five hours, or 40 ng/mL mitomycin C (MMC) for 16 h. Immediately following genotoxic exposure, 4 × 105 cells were mixed with 50 µL of 0.5% low-melting agarose/PBS, and the mixture was pipetted onto 2-well glass slides (Trivegen/Bio-Techne, Minneapolis, MN, USA). Once the agarose mixture had solidified, the slides were incubated with pH 10.0 lysis buffer (10 mM Tris-HCl, 2.5 M NaCl, 100 mM EDTA, and 1% Triton X-100) at 4 °C overnight while protected from light. The following day, the slides were equilibrated in pre-chilled (4 °C) electrophoresis buffer (300 mM NaOH, 1 mM EDTA) for 20 min. After electrophoresis in a horizontal chamber (Fisher Scientific, Waltham, MA, USA) for 40 min at 20 V (constant volts) at 300 mA, the slides were incubated twice in water for 5 min, fixed in 70% ethanol for 5 min, left to dry at 37 °C for 40 min, stained with SyBr Gold fluorescent dye (1:30,000 in 10 mM Tris-HCl pH 8.0, 0.1 mM EDTA) for 30 min, rinsed with water, and dried at room temperature. The stained comets were then imaged on an Eclipse 80i fluorescent microscope (Nikon, Meville, NY USA) with a CoolSNAP HQ2 camera (Telegyne Photometrics, Tucson, AZ, USA) at 10× magnification. Comet tail moment values were determined using CometScore Software Version 1.5. At least 75 tails were analyzed per experimental condition. Apoptotic cells (a small comet head and a very large comet tail) were excluded from the analysis.
2.5. T-FISH Assays
To assess chromosomal abnormalities by telomere fluorescent in situ hybridization (T-FISH), primary MEFS (passage 3 or earlier) were plated on 0.2% gelatin-coated plates and allowed to attach overnight. Upon reaching exponential growth at 48 h, the cells were treated (or mock-treated) with either 40 ng/mL MMC for 16 h or 1 µM CPT for 1 h. After drug exposure, Karyomax Colcemid (Thermo Fisher Scientific, Waltham, MA, USA) was added to each plate to a final concentration of 100 ng/mL, and the cells were incubated for an additional 4 h. The cells were then harvested, treated in hypotonic buffer (0.03 M Sodium citrate) at 37 °C for 25 min, fixed in a methanol/acetic acid (3:1) solution, and dropped onto glass microscope slides to obtain metaphase spreads. The telomeres were then stained with the Cy3-PNA probe (PNA Bio, Thousand Oaks, CA, USA), and the DNA was counterstained with DAPI-containing mounting media (Vectashield; Vector Laboratories, Newark, CA, USA). The T-FISH metaphase spreads were imaged on an Axio Imager Z2 fluorescent microscope with Coolcube1 camera (Zeiss, Hebron, KY, USA), and Metafer software version 3.10.6 (Metasystems, Newton, MA, USA) was used to automatically locate metaphases at 10× magnification and automatically capture images at 63× magnification (Cytogenetics Shared Resource, HICCC). The captured metaphases were then analyzed using Isis fluorescent imaging system software version 3.10.6 (Metasystems).
2.6. Western Blotting
For Western blotting, MEFs were seeded at 0.3 to 0.5 × 10
6 cells per 100-mm plate. At 48 h, the exponentially growing cells were treated with 1 µM CPT for 1 h or for the time points 0, 20, 60, and 90 min. Treated and mock-treated cells were then harvested, lysed in 5× packed cell volumes of LS lysis buffer (10 mm Hepes, pH 7.6, 0.25 M NaCl, 0.1% NP40, 5 mM EDTA, 10% glycerol, EDTA-free complete proteases inhibitor cocktail (Roche, Indianapolis, IN, USA), 1 mM DTT, and 50 mM NaF) on ice for 10 min, and spun at maximum speed in a microcentrifuge for 10 min at 4 °C. The supernatants were then collected as cell lysates, and 50 μg protein aliquots of each lysate were fractionated by electrophoresis through 6.5% polyacrylamide gels in running buffer (25 mM Trizma base, 192 mM glycine, 0.1% SDS, pH 8.3) and electroblotted onto an Amersham Protran 0.45 µm nitrocellulose membrane (GE Healthcare Life Sciences, Chicago, IL, USA) in transfer buffer (25 mM Tris-HCl pH 7.6, 190 mM glycine, 20% methanol, 0.04% SDS) at 22 volts. After blocking with 10% milk in TBS-T buffer (20 mM Tris-HCl pH 7.6, 0.137 M NaCl, 0.1% Tween 20) for 5 min on a shaking platform, each membrane was incubated for 2 h at room temperature with one of the following primary antibodies in 2% milk/TBS-T: Brca1 rabbit polyclonal (1:2000) [
40], Ctip 14-1 mouse monoclonal (1:50) [
41], phospho-KAP-1 (S824) rabbit polyclonal (1:2000; Bethyl, Houston, TX, USA), KAP-1 rabbit polyclonal (1:20,000; Bethyl, Houston, TX, USA), phospho-Chk1 (S345) rabbit polyclonal (1:1000; Cell Signaling, Danvers, MA, USA), Chk1 (G4) mouse monoclonal (1:500; Santa Cruz, Dallas, TX, USA), phospho-RPA2 (S4/S8) rabbit polyclonal (1:2000; Bethyl, Houston, TX, USA), RPA2 rabbit polyclonal (1:20,000; Bethyl, Houston, TX, USA), and
α-tubulin mouse monoclonal DM1A (1:10,000; Millipore, Burlington, MA USA). After incubation with the appropriate secondary antibody (GE Healthcare, Chicago, IL USA, NA934 Donkey anti-rabbit HRP at 1:10,000 dilution or Sigma, Burlington, MA USA, A5278 goat anti-mouse HRP at 1:10,000 dilution), each membrane was developed by chemiluminescence using either the SuperSignal West Pico substrate or the SuperSignal West Dura substrate (ThermoFischer Scientific, Waltham, MA USA).
2.7. Immunofluorescence Microscopy
For immunofluorescent microscopy of phospho-RPA2 and gH2AX foci, 105 MEFs were seeded onto poly-l-lysine (Sigma)-coated coverslips. After 48 h, the cells were treated or mock-treated with 1 µM camptothecin (Sigma) for 1 h and then incubated on ice for 5 min with pre-extraction buffer (100 mM Pipes, pH 6.8, 2 mM EGTA, 1 mM MgCl2, 0.5% Triton X-100) and 3 min with strip buffer (100 mM Tris-HCl, pH 7.4, 10 mM NaCl, 3 mM MgCl2, 1% Tween 20, 0.25% sodium deoxycholate). After fixing with 4% paraformaldehyde (w/v) in PBS for 15 min, the cells were incubated in serum-free DMEM for 5 min, washed twice with PBS, and permeabilized with 1% Triton X-100 in net gel (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 5 mM EDTA, 0.05% NP40, 0.25% Gelatin IV (bloom 75, type B), 0.02% sodium azide) at 4 °C for 10 min. After washing twice with net gel, the coverslips were blocked for 15 min with 1% BSA in PBS. For immunofluorescent staining, the cells were first incubated with either phospho-RPA2 (T21) rabbit polyclonal antibody 1:100 (Abcam, Waltham, MA USA) or phospho-H2AX (Ser139) mouse monoclonal antibody 1:200 (Millipore, Burlington, MA USA) in a humidified chamber for 40 min, washed three times with 1% BSA/PBS, and then incubated for 40 min in 1:1000 dilutions of, respectively, goat anti-rabbit Alexa 488 or goat anti-mouse Alexa 568 (Molecular Probes, Waltham, MA USA). After staining, the coverslips were washed twice with 1% BSA/PBS, rinsed with PBS and then water, and mounted with Vectashield hard set mounting media containing 4′, 6-diamidino-2-phenylindole (DAPI; Vector Laboratories, Newark, CA, USA). The cells were imaged on a Nikon Ti Eclipse inverted confocal microscope with a Nikon A1 Plus camera at 60× magnification. For automated quantification of the pRPA2 and gH2AX foci, slides were stained with phospho-RPA2 (T21) rabbit polyclonal antibody 1:100 (Abcam Waltham, MA USA) or phospho-H2AX (S139) rabbit polyclonal antibody 1:200 (Cell Signaling, Danvers, MA, USA) as primary antibodies and incubated with 1:1000 diluted secondary antibodies of goat anti-rabbit Alexa 488 (Molecular Probes, Waltham, MA USA) as described above. Imaging was carried out using the Metafer software version 4 (Metasystems, Newton, MA USA), and 2000 cells were counted on each slide from three independent experiments.
2.8. DR-GFP and Single-Molecule Analysis of Resection Tracks (SMART) Assays
Double-strand DNA break repair by homologous recombination was evaluated in ES cell subclones harboring a single DR-GFP reporter integrated into the
Pim1 locus [
39] as described previously [
37]. For each experimental condition, 50,000 cells were analyzed by Flow Jo (version 10) software. For SMART assays [
42], MEF cells were seeded on 1 × 100 mm plates at 0.25 to 0.3 × 10
6 cells per plate. The next day, the cells were provided with fresh media containing 10 µM BrdU (Sigma) and cultured for 24 h. The cells were then either irradiated (10 Gy, Gammacell 40, cesium-137) and harvested 1 h later or treated with 1 µM CPT for 1 h. The cells were then embedded in low-melting agarose (Bio-Rad, Hercules, CA USA), followed by DNA extraction using the FiberPrep DNA extraction Kit (Genomic Vision, Bagneux, France). To stretch the DNA fibers, ComdiCoverslips (Genomic Vision) were dipped into the DNA solution for 15 min and pulled out at a constant speed (250 µm/s) using the FiberComb machine (Genomic Vision). Coverslips were baked for 2 h at 60 °C and incubated with anti-BrdU mouse monoclonal antibody 1:1000 (GE) in BlockAid (Molecular Probe) buffer for 1 h. The coverslips were washed three times in PBS with 0.05% Tween 20, incubated with anti-mouse Alexa 594 antibody 1:1000 in BlockAid buffer for 40 min, washed three times in PBS with 0.02% Tween 20, rinsed in water, and dehydrated in 70%, 90%, and 100% ethanol, respectively. The dried coverslips were mounted with Prolong Gold mounting media (Invitrogen) containing 1:15,000 YOYO1 (Invitrogen), and images of the DNA fibers were collected on an Eclipse 80i fluorescent microscope (Nikon) with a CoolSNAP HQ2 camera (Telegyne Photometrics, Tucson, AZ, USA) at 20× magnification and recorded using NIS ELEMENTS Nikon software version 4. DNA fiber lengths were measured by Fuji Image J, and a minimum of 230 fibers were analyzed for each sample.
2.9. Statistical Analyses
Statistical analyses were conducted on PRISM software version 6 (GraphPad), and significant differences were labeled with one, two, three, or four asterisks for p < 0.05, p < 0.01, p < 0.001, or p < 0.0001. The Mann–Whitney test was used to ascertain significant differences between the survival of distinct mouse cohorts, as well as the lengths of DNA fragments generated in the comet and SMART assays. The paired Student’s t test was used for all other statistical determinations.
4. Discussion
CtIP
Sae2 promotes efficient resection of DSB ends through multiple mechanisms [
6,
7]. First, by allosterically activating the MRN endonuclease, CtIP
Sae2 triggers the “clipping” reaction that initiates short-range DNA resection. Second, CtIP
Sae2 facilitates subsequent long-range resection by recruiting and activating addition factors, such as Exo1 and the DNA2/BLM complex. The 3′-single-strand DNA (ssDNA) overhang generated by CtIP-mediated resection can then serve as a platform for the assembly of Rad51/ssDNA nucleofilaments, a critical step in the DNA damage response required for both ATR-dependent checkpoint signaling and homology-directed repair (HDR) of DNA breaks.
Given the central role of CtIP
Sae2 in the DNA damage response, the evolutionary conservation of its primary amino acid sequence seems rather modest. For example, recognizable homology between mammalian CtIP and yeast Sae2 is largely restricted to their C-terminal sequences, and within this “Sae2-like” domain, only a few amino acids are well conserved across the phylogenetic spectrum [
4]. Of these, the most striking are two residues that serve as substrates for CDK-like and PI3KK-like kinases, respectively. Early studies showed that phosphorylation of the CDK site (T847 in human CtIP;
Figure S1) promotes DNA resection [
29,
30] and subsequent work uncovered a specific role for CtIP-T847 phosphorylation in the MRN-mediated clipping reaction that initiates short-range resection [
12,
13,
14]. The other highly conserved phosphorylation site is a substrate for DNA damage-inducible members of the phosphatidylinositol 3-kinase-related kinase (PI3KK) family, including the mammalian ATR and ATM proteins and their yeast orthologs Mec1 and Tel1 [
6]. Although mutations of this residue, such as human CtIP-S859A and
S. cerevisiae Sae2-T279A, severely impair genotoxin-induced DNA resection in vivo [
29,
30], the precise molecular mechanisms by which phosphorylation of this site promotes resection have not yet been elucidated. Of note, the ability of yeast Sae2 to trigger the clipping reaction in vitro is abrogated by mutation of its CDK phosphorylation site (S267A), but not its PI3KK site (T279A) [
33]. In vitro studies with human components have shown that mutation of the PI3KK phosphorylation site (T859A) does reduce the ability of wild-type CtIP to stimulate the clipping reaction (from 43-fold to 15-fold), but to a significantly lesser degree than mutation of the CDK site (from 43-fold to 3-fold) [
44]. Nonetheless, the precise molecular mechanisms by which PI3KK phosphorylation of CtIP at this site promotes DNA resection remain unclear.
Consistent with studies of human CtIP [
17,
31], we observed that mouse cells lacking Ctip-T855 phosphorylation (
CtipT855A/T855A) are severely impaired for both genotoxin-induced DNA resection and homology-dependent repair (HDR) of DSBs. Nonetheless, in the absence of acute stress,
CtipT855A/T855A mice are largely indistinguishable from their wild-type and heterozygous (
CtipT855A/+) littermates in terms of viability, growth, fertility, and longevity. The modest phenotype of
CtipT855A/T855A mice contrasts with that of Ctip-null mice, which invariably die during early embryonic development [
7,
23,
24]. Interestingly, the viability of Ctip-null mice can be restored by ectopic expression of human CtIP, but not by a mutant lacking the CDK site (T847A) [
34]. Thus, although phosphorylation of the CDK site and the ATR/ATM site are both necessary for efficient CtIP-mediated resection in response to genotoxic stress, only the former is required for normal animal development. Indeed, a similar phenomenon is observed within the B lymphocyte lineage, where B cell proliferation is dependent on phosphorylation of Ctip at its CDK site but not its PI3KK site [
25,
34].
At present, we do not know why Ctip
T855A/T855A cells are more sensitive to some (e.g., IR) but not other (e.g., CPT) forms of genotoxic stress. For example, given that the CPT-induced one-ended DSBs that arise during DNA replication are normally repaired by HDR, a process dependent on CtIP-mediated resection, it seemed surprising that these cells are relatively resistant to CPT treatment. Conversely, since IR-induced two-ended DSBs are largely repaired by canonical NHEJ, a process that is independent of CtIP-mediated resection, the IR hypersensitivity of Ctip
T855A/T855A was also unexpected. Conceivably, these results might reflect differing genotoxic sensitivities at certain stages of cell cycle progression. The elevated expression of CtIP protein in the S and G2 phases of cell cycle progression [
41], together with the CDK-dependent phosphorylation of CtIP-T847 [
29,
30], likely ensure that DNA resection is robust during the cell cycle stages in which HDR occurs. Nonetheless, CtIP-mediated resection has also been observed in G1 cells, where CDK activity is limited [
45,
46,
47,
48]. Indeed, when G1 cells are exposed to ionizing radiation, CtIP-T847 can instead be phosphorylated by the Plk3 kinase, allowing a resection-dependent mode of non-homologous end-joining (NHEJ) to repair a subset of DSBs with slower kinetics than canonical NHEJ [
45,
46]. However, since CtIP protein levels and CDK activity are both markedly lower in G1 cells, the DNA resection required for this mode of repair may be more dependent on genotoxin-induced Ctip-T855 phosphorylation during the G1 phase than during the S phase. If so, then this dependency might explain why Ctip
T855A/T855A cells display sensitivity to IR but not CPT exposure. At the same time, we cannot dismiss the possibility that the observed pattern of genotoxin sensitivity is determined by as yet undefined resection-independent functions of Ctip-T855 phosphorylation.
Despite the relatively normal phenotype of unstressed
CtipT855A/T855A mice, their hypersensitivity to ionizing radiation (
Figure 1a) implies that phosphorylation of Ctip-T855 does have physiological relevance when animals face environmental challenges that elicit certain forms of DNA damage. Interestingly, Ctip-T855 phosphorylation is also required in biological settings where the NHEJ pathway of DSB repair is compromised, such as for the survival of Xrcc4/Tp53-deficient mice, cytokine activation of Xrcc4/Tp53-deficient B cells, and lymphomagenesis in
DNA-PKcs−/−Tp53−/− mice [
11,
35]. Thus, Ctip-T855 phosphorylation appears to be relevant whether DNA damage is elicited exogenously, by genotoxin exposure, or endogenously due to persistent DSB repair defects. Taken together, these observations suggest that phosphorylation of the PI3KK site serves to enhance resection in response to acute genotoxic stress, while phosphorylation of the CDK site—and thus induction of the MRN-mediated clipping reaction—is essential for cell viability and animal development even in the absence of stress.
In addition to its central role in DNA resection, CtIP has also been implicated in other cellular processes, such as transcriptional co-repression, cell cycle regulation, and chromosome segregation [
7]. Nonetheless, the embryonic lethality of mice expressing Ctip lacking the conserved CDK site (T847A) [
34] suggests that DNA resection—and more particularly, the clipping reaction—may be the CtIP function most important for normal animal development. Given that DNA resection is severely abrogated in vivo by loss of either the CDK site (T847) or the neighboring PI3KK site (T859), it is surprising that mice lacking the PI3KK site display such a modest phenotype. Elucidating the molecular mechanisms by which phosphorylation of the conserved PI3KK site promotes resection may help to resolve this enigma.