SEARCH

SEARCH BY CITATION

Keywords:

  • dATP inhibition;
  • brain distribution and development;
  • DNA repair;
  • double strand breaks (DSBs);
  • non-homologous DNA end joining (NHEJ);
  • V(D)J recombination

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Recent evidence suggests that DNA double strand breaks (DSBs) are introduced in neurons during the course of normal development, and that repair of such DSBs is essential for neuronal survival. Here we describe a non-homologous DNA end joining (NHEJ) system in the adult rat brain that may be used to repair DNA DSBs. In the brain NHEJ system, blunt DNA ends are joined with lower efficiency than cohesive or non-matching protruding ends. Moreover, brain NHEJ is blocked by DNA ligase inhibitors or by dATP and can occur in the presence or absence of exogenously added ATP. Comparison of NHEJ activities in several tissues showed that brain and testis share similar mechanisms for DNA end joining, whereas the activity in thymus seems to utilize different mechanisms than in the nervous system. The developmental profile of brain NHEJ showed increasing levels of activity after birth, peaking at postnatal day 12 and then gradually decreasing along with age. Brain distribution analysis in adult animals showed that NHEJ activity is differentially distributed among different regions. We suggest that the DNA NHEJ system may be utilized in the postnatal brain for the repair of DNA double strand breaks introduced within the genome in the postnatal brain.

Abbreviations
used

ara-CTP, 1-β-d-arabinofuranosylcytosine triphosphate

BSA

bovine serum albumin, DNA-PK, DNA-dependent protein kinase

DNA-PKcs

the catalytic subunit of DNA-PK

dATP

deoxyadenosine triphosphate

dNTPs

deoxyribonucleotides

DSBs

double strand breaks

DTT

dithiothreitol

NEM

N-ethylmaleimide

NHEJ

non-homologous DNA end joining

PMSF

phenylmethylsulfonyl fluoride

SSC

sodium chloride/sodium citrate

SDS

sodium dodecyl sulfate

TAE, 40 mM Tris-acetate, 2 mM EDTA, pH 8.5; V(D)J

variety (diversity) joining

XRCC4

X-ray cross-complementation 4.

DNA double strand breaks (DSBs) are considered to be a critical lesion in the formation of chromosomal aberrations, which if left unrepaired can result in cell death (Critchlow and Jackson 1998). DSBs are caused by pathological agents such as ionizing radiation, oxidative free radicals and anticancer drugs, but are also of physiological importance during cellular processes, such as V(D)J recombination (Weaver 1995). To repair these potentially lethal DNA damages, eukaryotic cells have evolved a variety of sophisticated DNA repair mechanisms related to homologous recombination (Deans et al. 2000; Johnson and Jasin 2001) and non-homologous DNA end joining (NHEJ), also known as illegitimate DNA recombination (Lieber 1998). In fact, the joining stages of V(D)J recombination share the same machinery components as NHEJ (Jeggo et al. 1995; Schwarz and Bartram 1996). These components are known to include DNA-dependent protein kinase (DNA-PK), X-ray cross-complementation 4 (XRCC4) and DNA ligase IV. DNA-PK is itself composed of three subunits: Ku70, Ku80 and DNA-PK catalytic subunit (DNA-PKcs), which are together required for DNA binding and protein kinase activity (Critchlow and Jackson 1998; Lieber 1998).

Interestingly, every component in the DNA-PK-dependent NHEJ pathway has been found in the developing and mature brain (Gao et al. 1998; Gu et al. 2000; Oka et al. 2000; Culmsee et al. 2001), suggesting that DNA end-joining mechanisms may be related to neural functions during development and in the adult. More recently, DNA ligase IV, XRCC4 (Gao et al. 1998; Barnes et al. 1998; Frank et al. 2000; Lee et al. 2000), and the DNA binding subunits of DNA-PK, Ku70 and Ku80 (Gu et al. 2000), were found to be necessary for normal neuronal development in the mouse embryo. However, NHEJ activities in developing and mature brain have not been studied. Several reports studied NHEJ activities in cell-free extracts prepared from Xenopus eggs (Labhart 1999), calf thymus (Johnson and Fairman 1996), mouse testis (Sathees and Raman 1999), and murine or human cell lines (Boe et al. 1995; Daza et al. 1996; Baumann and West 1998). For the present studies, we used an in vitro assay to test NHEJ activity in cell-free protein extracts prepared from adult rat cerebrum. The NHEJ assay detects DNA joining activity in the extracts resulting in intermolecular ligation of a variety of DNA substrates. The brain NHEJ system can process cohesive and non-matching protruding DNA ends as well as blunt ends, although with reduced efficiency. The reaction can be blocked by DNA ligase inhibitors and can occur in the presence or absence of exogenously added ATP. Moreover, we show that cerebral NHEJ is higher in young rats than in older animals and that cortex and hippocampus are among the brain regions showing higher NHEJ activity. Overall, our data support the idea that the mature brain possesses the capacity for the repair of DNA DSBs and that brain NHEJ activity might be of importance for normal brain function.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Subjects

The experiments were performed using male Long Evans rats weighing 270–300 g (2–2.5 months old) that were obtained from Harlan Sprague Dawley Inc. (Indianapolis, IN, USA) or from an in-house colony initiated with animals from the same vendor. Food and water were available at all times and animals were kept on a␣12-h on/12-h off light/dark cycle. Rats were killed by decapitation or during the brain perfusion protocol as described bellow. For the postnatal developmental study, pregnant females were obtained and the pups were killed at 1, 6, 12 and 24 days of age. Older animals were purchased from the vendor. Perfusion was performed to remove all blood traces from the brain with standard saline solution as described previously (Crawley et al. 2000). Briefly, rats were anesthetized with an intraperitoneal injection of pentobarbital (50 mg/kg). The heart was exposed and the right atrium was cut. A cannula was inserted into the left ventricle and perfusion was carried out for 5–10 min with saline solution (0.15 m NaCl). The brain, lung, testis and thymus were obtained from these saline-perfused animals. All experiments, except for the brain distribution studies, which also included the cerebellum, utilized only the rat cerebrum for analysis. The tissues were stored at − 80°C until used.

Preparation of tissue extracts

Protocols were based on those previously described (Baumann and West 1998; Labhart 1999). The brain samples were macerated in liquid nitrogen and suspended in extraction buffer (30 mm HEPES/KOH, pH 7.9, 1 mm EDTA, 2 mm dithiothreitol (DTT), 5 mm MgCl2, 0.5 m KCl, 20% glycerol, 0.5 mm phenylmethylsulfonyl fluoride (PMSF), 1 µg/mL each of leupeptin and aprotinin). After 2 h at 4°C, the extracts were centrifuged at 68 000 g for 2 h in a Beckman SW 41 Ti rotor. The supernatant was dialyzed for at least 5 h against NHEJ reaction buffer (30 mm HEPES/KOH, pH 7.9, 50 mm KCl, 0.2 mm EDTA, 5 mm MgCl2, 0.6 mm DTT, 0.5 mmPMSF, 10% glycerol) and microcentrifuged at 15 000 g for 30 min to remove any insoluble materials. The supernatant was stored at −80°C. Protein concentration was determined using the Bradford method. Protein extracts from lung, testis and thymus were prepared following the same procedures. For gel filtration, cell-free protein extracts from brain (5 mg) were loaded onto a column (15 mL bed volume) of Sephacryl S-200 h (Amersham-Pharmacia Biotech, Piscataway, NJ, USA) pre-equilibrated with NHEJ reaction buffer. Protein was eluted with NHEJ reaction buffer and the protein-containing fractions were assayed for DNA end-joining activity. Protein concentration for each fraction was determined with the Bradford method.

DNA substrates and probes

Plasmid pBluescript SK (Stratagene, La Jolla, CA, USA) was digested with one or two different restriction enzymes to produce cohesive, blunt or non-matching DNA ends. The reaction products were separated on a 1% agarose gel followed by purification of selected restriction fragments with QIAEX II Gel Extraction Kit (QIAGEN, Hilden, Germany). For matching and blunt-end substrates, EcoRI, HindIII, PstI, EcoRV, ScaI and SmaI were chosen to generate linearized plasmid. For non-matching DNA-end substrates, EcoRI/AvaII, BglI/AvaII, EcoRV/AvaII and BglI were used to generate DNA fragments for the NHEJ reactions. The linearized pBluescript SK, obtained by EcoRI digestion, was labeled with [α-32P]dCTP using the RediPrimeTM II Kit (Amersham-Pharmacia Biotech) for random primed labeling. The 32P-labeled DNA probes were then purified with a Nick Column (Amersham-Pharmacia Biotech) and used for Southern-blot detection of the NHEJ reaction products (see below).

DNA end-joining reaction

NHEJ assays were performed following similar conditions as those previously described (Baumann and West 1998; Labhart 1999). A standard in vitro end-joining reaction (50 µL) contained 20 ng DNA substrate with EcoRI-end, 50 µg brain protein extracts (as determined using the Bradford method for protein quantitation), 1 mm ATP and 10 µg bovine serum albumin (BSA) in NHEJ reaction buffer. Reactions were incubated at room temperature (21°C) for 1 h, stopped by heating at 70°C for 10 min, followed by the addition of 1 µL of 10% sodium dodecyl sulfate (SDS) and 5 µL of 1.0 mg/mL proteinase K (Stratagene, La Jolla, CA, USA) and followed with incubation at 37°C for 30 min. The deproteinized reaction mixtures were extracted with phenol/chloroform/isoamyl alcohol (25 : 24 : 1), ethanol precipitated, resuspended in sterile water and stored at − 20°C until used. The DNA ligase inhibitors (p-hydroxycercuribenzoate, N-ethylmaleimide (NEM), 1-β-d-arabinofuranosylcytosine triphosphate (ara-CTP) and distamycin (A) were purchased from Sigma (St Louis, MO, USA) and were added to the extracts just prior to the addition of DNA substrates. All experiments were repeated between three and five times.

Analysis of the joined products with Southern blotting

The DNA end-joining products were separated by electrophoresis in 1% agarose gel with 40 mM Tris-acetate, 2 mM EDTA, pH 8.5, (TAE) buffer at 100 V for 120 min. The agarose gel was treated with 0.25 m HCl for 30 min to depurinate the DNA, which in turn leads to strand cleavage. The length reduction improves the transfer for longer DNA molecules. The HCl solution was poured off and the gel was rinsed with distilled water three times. Then the gel was treated twice with denaturing solution (0.5 m NaOH, 1.5 m NaCl) for 20 min. Such treatment produces single-stranded DNA that is suitable for the hybridization with 32P-labeled DNA probes. The denaturing solution was poured off and the gel was rinsed with distilled water, which was followed by two 20-min neutralization steps using 0.5 m Tris-HCl buffer (pH 7.0) containing 1.5 m NaCl. The denatured DNA molecules were transferred from the gel to a nylon membrane using the capillary method with 20X sodium chloride/sodium citrate (SSC) buffer (3 m NaCl, 0.3 m Na3 citrate, pH 7.0) at room temperature overnight. The DNA-containing nylon membrane was crosslinked with the GS Gene LinkerTM UV Chamber (Bio-Rad, Hercules, CA, USA), and either immediately used or kept dry until used for hybridization. The␣membrane was rinsed in 6X SSC and pre-hybridized with 100 µg/mL salmon sperm DNA in hybridization solution (10.36% polyethylene glycol 8000, 7.25% SDS, 0.233 m NaCl, 0.155 m, pH 7.4 phosphate buffer, 1.55 mm EDTA; Gibco BRL, Rockville, MD, USA) at 68°C for 1 h. After the pre-hybridization, the 32P-labeled linear pBluescript SK probe (7 × 105 cpm/mL, see above) was added and hybridized at the same temperature overnight. After hybridization, the membrane was washed twice with 2X SSC containing 0.1% SDS at room temperature for 5 min; once with 0.2X SSC containing 0.1% SDS at room temperature for 5 min and twice with 0.2X SSC, 0.1% SDS at 68°C for 15 min The membrane was then exposed to X-ray film for autoradiography.

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

DNA end joining is present in brain extracts at high activity

To detect DNA end-joining activity in the brain, we used a protocol for an in vitro assay as described previously (Fig. 1) (Labhart 1999; Baumann and West 1998). Cell-free extracts were prepared from the cerebrum of adult rats and plasmid fragments generated with a variety of restriction enzymes were used as substrates in the NHEJ assay. After incubation of brain protein extracts with DNA substrates, the reaction products could be detected as dimers and multimers by Southern blotting (Fig. 1). Figure 2(a) shows the results of NHEJ reactions with DNA ends generated by EcoRI. Only the monomer (substrate DNA; molecular mass, 2961 bp) was detected when the protein extracts were heated to 100°C prior to the addition of DNA ends (lane 1), indicating that religation of the linearized plasmid itself did not occur during incubation. When intact protein was incubated for different reaction times in the presence of the DNA substrates, higher molecular DNA products, corresponding to dimers (5922 bp) and linear multimers, were produced by an increasing number of intermolecular ligation steps. The reaction efficiency increased with longer reaction times, as seen by the production of increasing quantities of higher molecular weight linear multimers. The band corresponding to the small increase in size of the substrate DNA band corresponds to the open circular form of the dimer, which has been reported to occur in other NHEJ systems (Labhart 1999; Sathees and Raman 1999). We also prepared protein extracts from animals that were perfused with saline to remove all the blood from their brains prior to decapitation and the protein samples were subjected to gel filtration prior to analysis. The results showed no difference in DNA end joining between extracts prepared from non-perfused or perfused brain tissue (Fig. 2a, lane 8), suggesting that the observed activity is not a result of contamination with blood cells.

image

Figure 1. Measuring NHEJ activity in cell-free protein extracts from the rat cerebrum. Brain tissue was macerated under liquid nitrogen in a mortar and cell-free protein extracts were prepared by 0.5 m KCl extraction, ultracentrifugation and dialysis using NHEJ buffer. The plasmid was digested with restriction enzymes to produce a variety of linearized fragments with cohesive, non-matching or blunt DNA ends. Selected DNA fragments were incubated individually with the protein extracts at room temperature generally for 1 h. The DNA was then purified and fractionated by agarose gel electrophoresis followed by Southern blotting using the linearized plasmid as probe. The lowest molecular weight bands represent the unmodified substrate DNA. NHEJ results in the generation of higher molecular weight dimers, trimers and multimers as the DNA end-joining products. All the experiments were repeated between three and five times.

Download figure to PowerPoint

image

Figure 2. NHEJ assays with protein extracts from rat cerebrum. (a) The time course of DNA end joining and NHEJ with partially purified protein from perfused rat cerebrum. DNA substrates with cohesive ends generated by digestion of pBluescript SK plasmid with EcoRI were incubated with cell-free extracts from rat brain at room temperature. The reactions were stopped at different time points and the DNA was analyzed as described in Fig. 1. Results showed several DNA molecular sizes corresponding to the substrate or monomer (Mono), the open circular dimer (OC), the dimer (Dim), and several higher molecular weight bands corresponding to multimeric DNA products (Mult). The reaction using protein obtained after brain perfusion with saline solution and gel filtration was incubated for 60 min (lane 8). (b)␣Different DNA substrate concentrations in the DNA end-joining reaction. Different concentrations of DNA substrate with EcoRI end were present in the reaction with standard conditions. The molecular marker (10 µg High DNA MassTM ladder; Gibco BRL) was separated by electrophoresis and transferred to a nylon membrane together with the NHEJ reactions. The lane with the ladder was cut from the rest of the membrane and hybridized with a 32P-labeled probe synthesized using the ladder DNA as template following the same procedures described in Materials and methods. (c) Assessment of cerebral IR on different DNA substrates. Different DNA substrates with cohesive (lanes 1–3), blunt (lanes 4–6) and non-matching (lanes 7–10) ends were generated by digestion of the same plasmid as above with one or two restriction enzymes and incubated with cell-free extracts from rat cerebrum. Lane 1, EcoRI end; lane 2, HindIII end; lane 3, PstI end; lane 4, ScaI end; lane 5, EcoRV end; lane 6, SmaI end; lane 7, BalI end; lane 8, BalI/AvaII end; lane 9, EcoRV/AvaII end; lane 10, EcoRI/AvaII end. All results are representative of between three and five experiments.

Download figure to PowerPoint

We also determined that linear multimers are the predominant products both at low or high DNA/protein ratios (Fig. 2b), contrasting with previously reported NHEJ detected in cell-free extracts from the human lymphocyte cell line, SupT1 (Boe et al. 1995; Daza et al. 1996), which yielded equal quantities of multimeric and circular products with either high or low DNA/protein ratios. We next tested the NHEJ efficiencies for different DNA ends generated by different restriction enzymes. We used EcoRI-, HindIII- and PstI-generated cohesive ends; ScaI-, EcoRV- and SmaI-generated blunt ends; and BglI-, BglI/AvaII-, EcoRV/AvaII- and EcoRI/AvaII-generated non-matching ends for our reactions (Fig. 2b). Results showed that DNA substrates with cohesive, blunt and non-matching protruding ends were all joined with different relative efficiencies, the best being cohesive and non-matching DNA ends (Fig. 2c). However, only trace joining products could be detected for the substrates with blunt DNA ends (Fig. 2c). This is in contrast to a previous study showing efficient DNA end joining for blunt substrates in cell-free extracts from mouse testis (Sathees and Raman 1999). Likewise, another study showed that up to 6% of the blunt ends were joined to form dimer products in extracts from lymphoblastoid cell lines (Baumann and West 1998). The band corresponding to the open circular dimer occurred mostly when cohesive and non-blunt or non-matching ends were used as substrates.

The broad substrate selectivity of the reaction was demonstrated by the ability of high concentrations of heterologous DNA to compete for end joining and thereby block NHEJ of the added substrates. As shown in Fig. 3(a), when the linearized plasmid from which the substrate DNA fragments were produced was used as probe to detect the products of NHEJ, a marked reduction of substrate-originated end-joining products was observed in the reaction performed in the presence of herring sperm DNA. Even at the lowest concentration (20 ng), the addition of herring sperm DNA completely blocked the generation of multimeric products. Increasing the concentration of herring sperm DNA resulted in a complete blockade of NHEJ products from the added substrate. To determine whether the dimers and multimers were indeed products of ligation, and whether the original restriction sequences used to produce the DNA ends were restored after ligation, we tested the effects of adding the particular restriction enzymes used to generate the DNA substrate to the NHEJ products (Fig. 3b). The end-joining products for both cohesive (EcoRI) and non-matching (EcoRI/AvaII) DNA substrates were used. When the DNA end-joining products of NHEJ reactions that used EcoRI- or EcoRI/AvaII-generated DNA substrates were incubated with the respective restriction enzymes, most of the dimers and multimers were digested back to the monomer size. The results show that the products generated using cohesive DNA ends were not modified in the system, because the joining products were susceptible to cleavage by the restriction enzyme used to generate the substrate DNA ends (Fig. 3b). In contrast, the fact that the products generated using non-matching DNA ends or with 5′ or 3′ protruding ends could be joined efficiently and redigested by the corresponding restriction enzymes back to the monomer size, an event that required restoration of the original restriction site, suggests that the NHEJ system possesses the capability of modifying the DNA ends prior to ligation. Thus, the reaction involves processing of DNA ends, possibly via a DNA nuclease and polymerase, prior to the final DNA end-joining step. Additional experiments confirmed that the final DNA end-joining step involves a DNA ligase because the general DNA ligase inhibitors,␣p-hydroxycercuribenzoate and NEM effectively blocked the reaction (Figs 4a and b). Other more specific DNA ligase inhibitors, such as ara-CTP and distamycin A, were also effective blockers of brain NHEJ (data not shown).

image

Figure 3. The fidelity of DNA end-joining reactions in brain extracts. (a) DNA end-joining reaction in the presence of increasing concentrations of herring sperm DNA. The EcoRI-generated plasmid fragment was used in the reaction as the substrate. (b) Redigestion after DNA end-joining reactions. DNA substrates with cohesive ends generated with EcoRI and non-matching ends generated with EcoRI and AvaII were used. After completing the end-joining reaction with the respective DNA substrates, joined DNA products were extracted and re-suspended in corresponding restriction enzyme buffers and incubated at 37°C for 2.5 h with or without EcoRI or EcoRI/AvaII, respectively. The products were then purified and analyzed by Southern blotting.

Download figure to PowerPoint

image

Figure 4. Effects of the inhibitors of DNA ligase on DNA end-joining reaction in brain extracts. The reactions were performed in the presence of DNA ligase inhibitors: (a) p-hydroxymercuribenzoate or (b) N-ethylmaleimide, at the concentrations indicated. The EcoRI-generated DNA substrate was used. The experiment was performed three times.

Download figure to PowerPoint

dATP is a strong inhibitor of brain NHEJ

Because of the fact that NHEJ reactions utilizing non-matching DNA ends as substrates probably required repair of DNA end sequences prior to ligation, we examined the effects of the addition of deoxyribonucleotides (dNTPs) to the reaction. Surprisingly, as seen in Fig. 5(a), adding dNTPs blocked brain NHEJ activity. We observed a strong inhibition of the reaction when all four dNTPs were added to the brain NHEJ reaction on a DNA substrate with EcoRI-generated cohesive ends. By testing each deoxynucleotide independently, we determined that only dATP, and not dCTP, dGTP or dTTP, could inhibit the reaction (Fig. 5a). The inhibitory effects of dATP increased in a concentration-dependent way, at a fixed concentration of ATP (1.0 mm) and dATP concentrations ranging from 0.02 to 2 mm dATP (Fig. 5b). We also tested the effects of dATP on NHEJ from non-matching DNA ends (Fig. 5c). As seen, dATP blocked the joining of all the types of DNA substrates tested (Fig. 5c). As the concentrations of ATP and dATP used in our experiments fall within their physiological concentrations, 1.5–4.9 mm and 0.002–0.046 mm, respectively (Traut 1994), it is possible that both nucleotides modulate NHEJ in brain cells.

image

Figure 5. Inhibition of DNA end-joining reaction in brain extracts by dATP. (a) Deoxynucleotide specificity for the inhibition of DNA end-joining reaction in brain extracts. The concentration of each nucleotide was 1 mm, as was that of ATP. (b) DNA end-joining reaction of brain extracts with different concentrations of dATP. (c) dATP inhibits NHEJ activity for different DNA substrates with non-matching ends. Lanes 1–2, BalI end; lanes 3–4, BalI/AvaII end; lanes 5–6, EcoRV/AvaII end; lanes 7–8, EcoRI/AvaII end. For sequences see Fig. 2(c).

Download figure to PowerPoint

ATP-independent NHEJ in the brain

As all reactions were performed in 1 mm ATP, we decided to test whether the reaction could proceed in the absence of ATP. Reactions were all carried out with brain extracts that were dialyzed twice with 100 volumes of NHEJ buffer, which should remove small molecules such as ATP and NAD+ from the extracts (see Materials and methods). To further ensure the removal of all traces of such molecules, we proceeded to perform gel filtration of the extracts through Sephacryl S-200 h columns, and then compared their efficiency for NHEJ in the absence and in the presence of ATP. Remarkably, the gel filtration preparations showed NHEJ activity in the absence of added ATP, although at lower levels than in reactions performed in the presence of ATP (Fig. 6a). Importantly, no ligation activity was observed when proteins were heated to 100°C for 5 min, indicating that the products of ATP-independent NHEJ were not plasmid religation artifacts. The ATP-independent NHEJ reaction joined DNA with different ends including cohesive, non-matching and blunt ends (Fig. 6b). Similar to ATP-dependent NHEJ, DNA substrates with protruding ends were joined with higher efficiency than those with blunt ends. In addition, both dATP and NEM blocked ATP-independent DNA end joining (Fig. 6c). Moreover, according to the results in the second and third lanes shown in Figs 5(b) and 6(c), the ATP-independent reaction showed a higher sensitivity to dATP inhibition than the ATP-dependent reaction. More specifically, dATP completely blocked the ATP-independent reaction at 2 mm, whereas dimer products were still observed in the ATP-dependent reactions carried out with the same concentration of dATP.

image

Figure 6. ATP-independent DNA end-joining activity is present in brain extracts. (a) The brain NHEJ reaction can process cohesive EcoRI ends and non-matching EcoRI/AvaII ends in the presence or absence of ATP. Reactions were carried out with protein extracts subjected to dialysis and gel filtration. For both types of DNA ends, lanes 1 and 5 show the absence of DNA end-joining activity in protein extracts heated to 100°C prior to the addition of substrate. Moderate activity was observed in the absence of added ATP (lanes 2 and 6), whereas robust activity was seen in the presence of 1 mm ATP (lanes 3 and 7). dATP can block the ATP-independent activities for both EcoRI and EcoRI/AvaII DNA substrates at the concentration of 1 mm dATP (lanes 4 and 8). (b) ATP-independent DNA end-joining for different DNA ends (compare with Fig. 2c). Lane 1, EcoRI end; lane 2, HindIII end; lane 3, PstI end; lane 4, ScaI end; lane 5, EcoRV end; lane 6, SmaI end; lane 7, BalI end; lane 8, BalI/AvaII end; lane 9, EcoRV/AvaII end; lane 10, EcoRI/AvaII end. (c) Effects of inhibitors on the ATP-independent DNA end-joining activity in brain extracts. The inhibitors used are dATP and N-ethylmaleimide (NEM) at the indicated concentrations. DNA ends generated with EcoRI were used as substrates.

Download figure to PowerPoint

Comparison of NHEJ in different tissues

We then compared the NHEJ activity in brain extracts with protein extracts from other tissues including testis, lung and thymus. All experiments were performed in tissues from saline-perfused animals (see Materials and methods). The relative activities were as followed: brain ≥ testis > lung >> thymus (Fig. 7a). When we compared the sensitivities of the NHEJ reactions from different tissues to dATP, we found that only the reactions carried out with brain and testis extracts could be blocked by the deoxynucleotide (Fig. 7b). These results suggest that NHEJ in brain and testis use similar mechanisms, whereas NHEJ in thymus and lung utilize different mechanisms. Figures 7(c) and (d) show that the general DNA ligase inhibitor, NEM, could inhibit NHEJ reactions in protein extracts from all tissues used. As seen in Fig. 7, the NHEJ activity in thymus extracts seemed less efficient than that from brain, testis and lung. More specifically, unlike NHEJ in brain extracts, higher concentration of thymus protein showed lower DNA end-joining activity (Figs 8a and b). These data suggested to us the presence of a natural protein inhibitor of NHEJ in thymus that was not present in the brain extracts, and which restricted thymus activity as compared with brain DNA end joining. Furthermore, adding thymus, but not testis, extracts to the brain extracts prior to initiating an NHEJ reaction had an inhibitory effect on brain DNA end joining, which was lost by heating the thymus extracts to 100°C for 5 min (Fig. 8c). Importantly, increasing the concentration of DNA substrates could rescue brain NHEJ from the inhibitory effects of the thymus extracts (Fig. 8d). The results suggest that a protein inhibitor in thymus extracts blocks brain NHEJ by competing for and sequestering DNA ends.

image

Figure 7. Comparison of NHEJ activities in different tissue extracts obtained from saline perfused animals. All reactions were performed at␣standard conditions except in the presence of NHEJ inhibitors. (a) Standard control (without inhibitor); (b) with 1 mm dATP; (c) with vehicle (2 µL 100% ethanol) for the dissolve of N-ethylmaleimide (NEM); (d) with 100 µm NEM.

Download figure to PowerPoint

image

Figure 8. DNA end-joining activities in brain extracts are inhibited by thymus extracts. (a) DNA end-joining activities with different protein concentrations of brain extracts. (b) DNA end-joining activities with different protein concentrations of thymus extracts. (c) DNA end-joining reactions in 50 µg brain extracts were performed in the presence of different protein concentrations of thymus extracts, or 50-µg testis extracts, or 50-µg thymus extracts preheated at 100°C for 5 min. (d) Effects of high concentrations of DNA substrates on the inhibition of DNA end-joining activities in brain extracts by thymus protein extracts.

Download figure to PowerPoint

Developmental time course and distribution of NHEJ in␣the postnatal brain

To address the physiological significance of brain NHEJ we studied its expression profile and distribution in the postnatal brain. We examined the profile of NHEJ activity at different times during postnatal brain development and found higher levels of ATP-dependent and ATP-independent NHEJ in extracts from younger animals than in those from older adults (Fig. 9). ATP-dependent NHEJ increased at day 6 after birth and reached a maximum at postnatal day 12. A gradual decrease in NHEJ activity was observed with age, at least up to 7 months (210 days) after birth. In the absence of ATP the NHEJ activity was high at 12 and 24 days after birth and also showed a gradual decrease in recombination efficiency with age. In summary, activities were low at birth, peaked at postnatal day 12 and then gradually decreased with age. As all the studies were carried out with cell-free extracts␣prepared from rat cerebrum, we wanted to study the distribution of NHEJ activity in different regions of the adult rat brain. For the brain distribution studies, we compared the levels of NHEJ activity between the cerebellum, cortex, hippocampus, anterior striatum, olfactory bulb and spinal cord (Fig. 10). Although ATP-dependent NHEJ activity could be detected in all the regions tested, slightly higher activities could be consistently detected in cortex, hippocampus and the anterior striatum. In contrast, for the ATP-independent NHEJ reaction the highest activities occurred in extracts prepared from the hippocampus, followed by cerebellum, anterior striatum, olfactory bulb, cortex and spinal cord.

image

Figure 9. DNA end-joining activities during brain development. Brain extracts were prepared from animals at different days after birth as indicated. All extracts were prepared simultaneously and analyzed for NHEJ activity in the same experiment. All reactions were performed with 50 µg of protein as determined by the Bradford method. Reactions were performed in the absence (upper panel) or the presence (lower panel) of added ATP. The reaction time was 30 min.

Download figure to PowerPoint

image

Figure 10. The distribution of DNA end-joining activities in rat brain. The six regions of the central nervous system, cortex, cerebellum, hippocampus, striatum, olfactory bulb and spinal cord, were obtained from the perfused rat, and protein extracts were prepared from those regions. All extracts were prepared simultaneously and analyzed for NHEJ activity in the same experiment. All reactions were performed with 50 µg of protein as determined by the Bradford method. Reactions were performed in the absence (upper panel) or the presence (lower panel) of added ATP.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

The DNA end-joining reaction in brain extracts is a complex NHEJ system

We are the first to describe a DNA end-joining system for DSBs in the postnatal rat brain, which might be of importance for normal brain function. Our findings suggest that the NHEJ capability of the brain is possibly mediated by multiple protein components rather than by a single-step DNA ligation reaction. The reaction produces multimeric ligation products, which can be digested to the monomer size using the restriction enzyme initially, used to generate the DNA ends (Fig. 3b). When matching DNA ends were used as substrates, these were not modified in the system to allow for ligation, as shown by the fact that the DNA joining products were susceptible to cleavage by the restriction enzyme used to generate the substrate DNA ends (Fig. 3b). The DNA substrates with non-matching DNA ends, or with 5′ or 3′ protruding ends, could be joined efficiently and redigested by the same restriction enzymes used initially to produce the ends (Figs 2c and 3b). Therefore, the DNA ends must have been modified and then joined to generate the original restriction site. For example, one DNA substrate was produced by EcoRI and AvaII thereby generating two different 5′ ends with EcoRI (AATT-) and AvaII (-CAG) recognition sequences (Fig. 2c). This substrate could not only be used to generate dimers via ligation of the matching EcoRI ends, but was also used in the system to produce multimeric products that could only be formed by utilizing the AvaII-generated ends (Figs 2c and 3b). In order for such a reaction to occur, the middle base A of one of the AvaII protruding DNA ends must be replaced by a T to allow for ligation and reconstitution of the original AvaII recognition site. Indeed, when we incubated such joining products with EcoRI/AvaII we obtained a DNA product of the same size as the original substrate (Fig. 3b). Similar mechanisms would have to be used for substrates with 5′ or 3′ overhangs such as those produced by BalI/AvaII digests (Fig. 2c). Thus, the brain NHEJ system is probably a multiprotein system involving DNA nuclease and polymerase activities and cannot be classified as a single-step DNA ligation process. Recent reports indicated that both DNA nucleases and polymerases are involved in NHEJ to repair DSBs. MreII is a double-strand DNA 3′[RIGHTWARDS ARROW]5′ exonuclease and single-strand DNA endonuclease that can interact with RAD50 and Nbs1 proteins, thereby forming a complex that is involved in both homologous and non-homologous DNA end joining in mammalian cells (Petrini 1999). In addition, several DNA polymerases have been found to take part in NHEJ reactions (Pospiech et al. 2001). However, the DNA nuclease and polymerase involved in the brain NHEJ system remain to be identified.

Several of our results suggest that the predominant ligase present in the brain extracts is DNA ligase IV. First, only trace products of DNA end joining could be observed when blunt DNA ends were used as substrates in the brain NHEJ system, whereas high-end joining activities were shown for the cohesive or non-matching DNA ends (Fig. 2). Previous reports demonstrated that purified DNA ligase IV had no joining activity for blunt DNA ends (Robins and Lindahl 1996; Chen et al. 2000). The trace joining products for blunt ends in our system may be contributed by other DNA ligases, such as DNA ligase III (Tomkinson and Levin 1997), which is present in the brain at low levels. Second, we show that cell-free protein extracts prepared from adult rat brain exhibit both ATP-dependent and ATP-independent DNA end-joining activity (Fig. 6). Correspondingly, previous reports indicated that DNA ligase IV displays low, but detectable, ATP-independent ligation activity (Robins and Lindahl 1995) because of the covalently linked enzyme-AMP intermediate in the reactions. Similar to the purified DNA ligase IV, the brain enzyme could be also present as an endogenous enzyme-adenylate complex in the extracts, rendering DNA end joining independent of exogenously added ATP. Together, our data suggests that DNA ligase IV may be involved in the brain NHEJ system.

Previous investigations have shown that dATP can block DNA ligase I and II activities (David and Chapeville 1976; Bhat and Grossman 1986). However, it is not clear whether dATP has an inhibitory effect on DNA ligase IV. Because DNA ligase IV shares homology within the catalytic site with DNA ligase I and III (Tomkinson and Levin 1997; Robin and Lindahl 1996), it is indeed possible that it might also be suppressed by the deoxynucleotide. We also found that high concentrations of ATP could not rescue the reaction from the inhibition of dATP (data not shown), suggesting that dATP may bind to the enzyme at a site other than the ATP binding site. Thus, it is possible that dATP can bind with high affinity to a regulatory site of a key NHEJ factor, perhaps DNA ligase IV, thereby causing structural changes that ultimately block the reaction.

A DNA recombination system may be present in the postnatal brain

The physiological importance of the DNA DSBs generated during embryonic development in the brain remains to be determined. It has been suggested that DNA breaks in developing neurons may be produced as part of a somatic recombination system utilized to generate neuronal diversity as V(D)J recombination in lymphocytes generates antigen receptor diversity (Chun and Schatz 1999). In studies looking at apoptosis-related DNA fragmentation, DSBs could be detected widely during normal development in the embryonic and postnatal cortex, but few DSBs were observed in the adult (Blaschke et al. 1996; Thomaidou et al. 1997). Specifically, very high numbers of DNA DSBs were found in the brain at postnatal day 1, were reduced by postnatal day 5 and were almost undetectable at postnatal day 9 (Blaschke et al. 1996). Indeed, detection of DNA DSBs was inversely related to the timing of NHEJ activity detected by us within the postnatal brain. In our studies, we found that the cerebral NHEJ activity was low at birth, showed increased levels at postnatal day 6, reached its highest levels at postnatal day 12 and then gradually decreased along with age, a time course similar to that of increases in DNA ligase activity in the rat cerebellum (Nakaya et al. 1977). We can suggest that the early increase in DNA end-joining activity after birth may be responsible for the repair of DSBs generated during embryonic development and help maintain genomic stability in the adult brain. Moreover it is possible that, as suggested previously (Gilmore et al. 2000), a portion of the DSBs detected in the developing cortex by a sensitive in situ DNA end-labeling technique (Blaschke et al. 1996) could be generated by somatic recombination events, whereas the majority of the detected DSBs are most likely related to apoptosis. Several reports studying the time course of neuronal apoptosis during postnatal brain development have found peaks of cell death between days 4 and 14 after birth, depending on the brain structure examined (Marti et al. 1997; Chu et al. 2000). One possible role for NHEJ in the embryonic and postnatal brain could be to aid in the selection of surviving neurons, which would be those that effectively repair their DNA DSBs in a process that may involve rearrangement of specific genes. A recent study of somatic mutation during development using a transgenic mouse model dependent on a reporter gene for chromosomal integration found that increasing mutations, related to genomic rearrangement events, occurred in the brain between birth and 4–6 months of age, but not in the aging brain (Dolle et al. 1997). Importantly, DNA ligase IV and XRCC4 have been shown to be essential for neurogenesis, and mutant mice lacking these factors exhibit late embryonic lethality (Gao et al. 1998). Histological examination of the mutant brains has revealed extensive apoptotic death of post-mitotic neurons in the embryos. As the only known function of DNA ligase IV and XRCC4 is related to NHEJ, it is likely that this DNA end-joining reaction is critical for normal brain development. Other factors related to DNA DSB repair by homologous recombination (XRCC2) or in base excision repair (DNA polymerase β) have also been shown to be essential for neurogenesis (Deans et al. 2000; Sugo et al. 2000).

The existence of DNA recombination systems in the brain is a greatly disputed issue (for a recent review see Chun and Schatz 1999; Gilmore et al. 2000; Peña de Ortiz and Arshavsky 2001). Although several studies suggest that a DNA recombination system may exist in the brain, the recombinase responsible for such function remains to be found. Similar to the joining stages of V(D)J recombination, the NHEJ system in the brain can ligate random DNA sequences, thus generating recombinant DNA products. In V(D)J recombination, the recombining DNA segments are obtained by RAG1/2 endonuclease mediated site-specific cleavage (Weaver 1995; Schwarz and Bartram 1996). Although the activity addressed in our studies corresponds only to the DNA joining stage in recombination, we could suggest that the brain possesses the potential capacity for DNA recombination as well as for DNA repair (Brooks et al. 1996; Brooks 1998). In fact, both RAG1 and, more recently, RAG2 have been shown to be expressed in the vertebrate central nervous system (Chun and Schatz 1999; Frippiat et al. 2001; Jessen et al. 2001). It is important to mention that although NHEJ activity could be detected in other tissues, our results suggest that there are differences between the brain system and that of tissues high in V(D)J-recombination activity such as the thymus (Nussenzweig 1998). Finally, the differential distribution of both ATP-dependent and ATP-independent NHEJ in the brain could point to important regional differences in DNA recombination/repair activity in the mature central nervous system. However, it must be emphasized that whether the brain NHEJ system described here is related to DNA repair or to somatic recombination in the postnatal brain remains to be determined.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

This work was supported by the National Institutes of Health (SPO grants NIGMS-MBRS SOGGMO 8102–26S1 and NINDS-SNRP U54 NS39405; UPR grant NCRR-RCMI 2G12RRO3641). We thank Drs Walter Stuehmer, David Sweatt and Carlos Santiago for their helpful comments on this manuscript. We also thank Dr Carmen S. Maldonado-Vlaar, Dr Jianpeng Wang and Lily Álvarez for collaboration with the brain perfusion protocol.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • Barnes D. E., Stamp G., Rosewell I., Denzel A. and Lindahl T. (1998) Targeted disruption of the gene encoding DNA ligase IV leads to lethality in embryonic mice. Curr. Biol. 8, 13951398.
  • Baumann P. and West S. C. (1998) DNA end-joining catalyzed by␣human cell-free extracts. Proc. Natl Acad. Sci. USA 95, 1406614070.
  • Bhat R. and Grossman L. (1986) Purification and properties of two DNA ligases from human placenta. Arch. Biochem. Biophys. 244, 801812.
  • Blaschke A. J., Staley K. and Chun J. (1996) Widespread programmed cell death in proliferative and postmitotic region of the fetal cerebral cortex. Development 122, 11651174.
  • Boe S. O., Sodroski J., Helland D. E. and Farnet C. M. (1995) DNA end-joining in extracts from human cells. Biochem. Biophys. Res. Commun. 215, 987993.
  • Brooks P. J. (1998) Detection of excision nuclease in cell-free extracts from the adult mammalian brain. Mutat. Res. 408, 3746.
  • Brooks P. J., Marietta C. and Goldman D. (1996) DNA mismatch repair and DNA methylation in adult brain neurons. J. Neurosci. 16, 939945.
  • Chen L., Trujillo K., Sung P. and Tomkinson A. E. (2000) Interactions of the DNA ligase IV-XRCC4 complex with DNA ends and the DNA-dependent protein kinase. J. Biol. Chem. 275, 2619626205.
  • Chu T., Hullinger H., Schilling K. and Oberdick J. (2000) Spatial and temporal changes in natural and target deprivation-induced cell death in the mouse inferior olive. J. Neurobiol. 1, 1830.
  • Chun J. and Schatz D. G. (1999) Rearranging views on neurogenesis: neuronal death in the absence of DNA end-joining proteins. Neuron 22, 710.
  • Crawley J. N., Gerfen C. R., McKay R., Rogawski M. A., Sibley D. R. and Skolnick P. (2000) Current Protocols in Neuroscience, 2nd edn, 1.6.11.6.7. John Wiley & Sons Inc., New York.
  • Critchlow S. E. and Jackson S. P. (1998) DNA end-joining from yeast to man. Trends Biochem. Sci. 23, 394398.
  • Culmsee C., Bondada S. and Mattson M. P. (2001) Hippocampal neurons of mice deficient in DNA-dependent protein kinase exhibit increased vulnerability to DNA damage, oxidative stress and excitotoxicity. Brain Res. Mol. Brain Res. 87, 257262.
  • David J. C. and Chapeville F. (1976) Inhibition of chick embryo DNA ligase by dATP: its use in enzyme purification. Biochem. Biophys. Res. Commun. 71, 579583.
  • Daza P., Reichenberger S., Gottlich B., Hagmann M., Feldmann E. and Pfeiffer P. (1996) Mechanisms of nonhomologous DNA end-joining in frogs, mice and men. Biol. Chem. 377, 775786.
  • Deans B., Griffin C. S., Maconochie M. and Thacker J. (2000) Xrcc2 is required for genetic stability, embryonic neurogenesis and viability in mice. EMBO J. 19, 66756685.
  • Dolle M. E. T., Giese H., Hopkins C. L., Martus H.-J., Hausdorff J.␣ M. and Vijg J. (1997) Rapid accumulation of genome rearrangements in liver but not in brain of old mice. Nat. Genet. 17, 431434.
  • Frank K. M., Sharpless N. E., Gao Y., Sekiguchi J. M., Ferguson D. O., Zhu C., Manis J. P., Horner J. and DePinho R. (2000) A and, Alt F. W. DNA ligase IV deficiency in mice leads to defective neurogenesis and embryonic lethality via the p53 pathway. Mol. Cell 5, 9931002.
  • Frippiat C., Kremarik P., Ropars A., Dournon C. and Frippiat J. P. (2001) The recombination-activating gene 1 of Pleurodeles waltl (urodele amphibian) is transcribed in lymphoid tissues and in the central nervous system. Immunogenetics 52, 264275.
  • Gao Y., Sun Y., Frank K. M., Dikkes P., Fujiwara Y., Seidl K. J., Sekiguchi J. M., Rathbun G. A., Swat W., Wang J., Bronson R. T., Malynn B. A., Bryans M., Zhu C., Chaudhuri J., Davidson L., Feirni R., Stamato T., Orkin S. H., Greenberg M. E. and Alt F. W. (1998) A critical role for DNA end-joining proteins in both lymphogenesis and neurogenesis. Cell 95, 891902.
  • Gilmore E. C., Nowakowski R. S., Caviness V. S. Jr and Herrup K. (2000) Cell birth, cell death, cell diversity and DNA breaks: how do they all fit together? Trends Neurosci. 23, 100105.
  • Gu Y., Sekiguchi J., Gao Y., Dikkes P., Frank K., Ferguson D., Hasty P. and Chun J. (2000) and, Alt F. W. Defective embryonic neurogenesis in Ku-deficient but not DNA-dependent protein kinase catalytic subunit-deficient mice. Proc. Natl Acad. Sci. USA 97, 26682673.
  • Jeggo P. A., Taccioli G. E. and Jackson S. P. (1995) Menage a trois: double strand break repair, V(D)J recombination and DNA-PK. Bioessays 17, 949957.
  • Jessen J. R., Jessen T. N., Vogel S. S. and Lin S. (2001) Concurrent expression of recombination activating genes 1 and 2 in zebrafish olfactory sensory neurons. Genesis 29, 156162.
  • Johnson A. P. and Fairman M. P. (1996) The identification and characterization of mammalian protein involved in the rejoining of DNA double-strand breaks in vitro. Mut. Res. 364, 103116.
  • Johnson R. D. and Jasin M. (2001) Double-strand-break-induced homologous recombination in mammalian cells. Biochem. Soc. Trans. 29, 196201.DOI: 10.1042/0300-5127:0290196
  • Labhart P. (1999) Ku-dependent nonhomologous DNA end joining in Xenopus egg extracts. Mol. Cell. Biol. 19, 25852593.
  • Lee Y., Barnes D. E., Lindahl T. and McKinnon P. J. (2000) Defective neurogenesis resulting from DNA ligase IV deficiency requires Atm. Genes Dev. 14, 25762580.
  • Lieber M. R. (1998) Pathological and physiological double-strand breaks, roles in cancer, aging, and the immune system. Am. J. Pathol. 153, 13231332.
  • Marti M. J., James C. J., Oo T. F., Kelly W. J. and Burke R. E. (1997) Early developmental destruction of terminals in the striatal target induces apoptosis in dopamine neurons of the substantia nigra. J.␣Neurosci. 17, 20302039.
  • Nakaya N., Sawasaki Y., Teraoka H., Nakajima H. and Tsukada K. (1977) Changes of DNA ligase in developing rat brain. J. Biochem. 81, 15751577.
  • Nussenzweig M. C. (1998) Immune receptor editing: revise and select. Cell 95, 875878.
  • Oka A., Takashima S., Abe M., Araki R. and Takeshita K. (2000) Expression of DNA-dependent protein kinase catalytic subunit and Ku80 in developing human brains: implication of DNA-repair in neurogenesis. Neurosci. Lett. 292, 167170.
  • Peña de Ortiz S. and Arshavsky Y. I. (2001) DNA recombination␣is␣a possible mechanism in declarative memory. J. Neurosci. Res. 63, 7281.
  • Petrini J. H. (1999) The mammalian Mre11-Rad50-nbs1 protein complex: integration of functions in the cellular DNA-damage response. Am. J. Hum. Genet. 64, 12641269.
  • Pospiech H., Rytkonen A. K. and Syvaoja J. E. (2001) The role of DNA polymerase activity in human non-homologous end joining. Nucl. Acids Res. 29, 32773288.
  • Robins P. and Lindahl T. (1996) DNA ligase IV from Hela cell nuclei. J.␣Biol. Chem. 271, 2425724261.
  • Sathees C. R. and Raman M. J. (1999) Mouse testicular extracts process DNA double-strand breaks efficiently by DNA end-to-end joining. Mut. Res. 433, 113.
  • Schwarz K. and Bartram C. R. (1996) V (D) J recombination pathology. Adv. Immunol. 61, 285362.
  • Sugo N., Aratani Y., Nagashima Y., Kubota Y. and Koyama H. (2000) Neonatal lethality with abnormal neurogenesis in mice deficient in DNA polymerase β. EMBO J. 19, 13971404.
  • Thomaidou D., Mione M. C., Cavanagh J. F. and Parnavelas J. G. (1997) Apoptosis and its relation to the cell cycle in the developing cerebral cortex. J. Neurosci. 17, 10751085.
  • Tomkinson A. E. and Levin D. S. (1997) Mammalian DNA ligases. Bioesssays 19, 893901.
  • Traut T. W. (1994) Physiological concentration of purines and pyrimidines. Mol. Cell. Biochem. 140, 122.
  • Weaver D. T. (1995) V (D) J recombination and double-strand break repair. Adv. Immunol. 58, 2985.