Huntington’s disease (HD) is an age-dependent neurodegenerative disease. DNA repair pathways have recently been implicated as the most predominant modifiers of age of onset in HD patients. We report that endogenous huntingtin protein directly participates in oxidative DNA damage repair. Using novel chromobodies to detect endogenous human huntingtin in live cells, we show that localization of huntingtin to DNA damage sites is dependent on the kinase activity of ataxia telangiectasia mutated (ATM) protein. Super-resolution microscopy and biochemical assays revealed that huntingtin co-localizes with and scaffolds proteins of the DNA damage response pathway in response to oxidative stress. In HD patient fibroblasts bearing typical clinical HD allele lengths, we demonstrate that there is deficient oxidative DNA damage repair. We propose that DNA damage in HD is caused by dysfunction of the mutant huntingtin protein in DNA repair, and accumulation of DNA oxidative lesions due to elevated reactive oxygen species may contribute to the onset of HD.
Huntington’s disease (HD) is a genetic neurodegenerative disease caused by a CAG-repeat expansion in the first exon of the HTT gene encoding the huntingtin protein, producing an expanded polyglutamine tract in the amino terminus (1). The N17 domain, comprising the first 17 amino acids of huntingtin adjacent to the polyglutamine tract, regulates huntingtin localization, endoplasmic reticulum and the vesicle membrane binding, and aggregation (2–6). This domain also modifies the disease progression in animal models as shown by phospho-mimetic mutants (7) and chemical biology approaches (2,8). Mutant huntingtin is hypo-phosphorylated at serine residues S13 and S16 (2), and restoration of phosphorylation changes the conformation of the protein (2,9). We have recently discovered that a conserved methionine at position eight within N17 (M8) is a reactive oxygen species (ROS) sensor by sulfoxidation, which augments N17 alpha-helical structure and triggers the release of membrane-bound huntingtin from the endoplasmic reticulum, allowing N17 phosphorylation and thereby nuclear retention (4,10). While the accumulation of mutant huntingtin in the nucleus has been associated with pathogenesis in HD brains (11), it is likely that N17-phosphorylated huntingtin performs vital functions in the nucleus in response to ROS.
ROS-induced oxidative stress is suggested to play a key role in selective neuronal loss associated with aging and neurodegeneration, including HD (12–14). Damage to nuclear DNA in HD patient samples and mouse models, in the form of strand breaks and damaged bases, has been extensively reported (15–21), and more recent studies support the involvement of DNA repair pathways in HD age at onset (22,23). Moreover, the DNA damage response protein ataxia-telangiectasia mutated (ATM) contributes to disease progression in the BACHD mouse model (24).
ATM, a serine/threonine kinase rich in Huntingtin, Elongation Factor 3, Protein Phosphatase 2A, mTOR (HEAT) domains, signals through a host of proteins in response to double-stranded DNA breaks. It is also activated by single-stranded breaks (25), chromatin reorganization (26), oxidative stress (27), and has an impaired response to manganese in human HD neuroprogenitor cells (28). Our work on the nuclear localization of huntingtin (2,4,10) led us to hypothesize that dysregulated huntingtin and ATM activities in the nucleus may contribute to DNA repair deficiency in HD. Using model systems that interrogate endogenous human proteins and disease-relevant polyglutamine lengths in clinically isolated cells, we have found that huntingtin co-localizes with ATM at nuclear puncta and at sites of DNA damage. The huntingtin protein that is localized to sites of DNA damage is phosphorylated at the N17 domain, and its recruitment is sensitive to ATM kinase inhibition. Furthermore, we show that huntingtin scaffolds DNA damage response (DDR) proteins and that DNA repair is compromised in HD cells.
We hypothesize that highly metabolically active cells, such as the medium-sized spiny neurons of the striatum primarily affected in HD, may be susceptible to excessive oxidative damage that accumulates with aging due to combined sub-optimal huntingtin function in the DNA repair process and typical age-dependent mitochondrial deficiency and increased ROS. This work provides a mechanism to explain the identification of DNA repair pathways as genetic modifiers of age at onset, as well as readouts for further investigation and drug development strategies.
N17-phosphorylated huntingtin localizes to sites of DNA damage in an ATM-dependent manner
We have shown that huntingtin localizes to chromatin-dependent nuclear puncta, visible upon immunofluorescence with the previously characterized and fully validated antibody against phosphorylated serines 13 and 16 (phospho-N17) (2). Based on the report that inhibition of DNA damage response protein ATM blocks the toxicity of mutant huntingtin and slows disease progression in a mouse model (24), we investigated a possible functional link between nuclear huntingtin and ATM. Co-immunofluorescence with the phospho-N17 and anti-ATM antibodies in wild type human fibroblasts revealed that ATM localizes to phospho-N17 nuclear puncta (Fig. 1A). Super resolution-structured illumination microscopy (SR-SIM) revealed that the puncta are granular in nature (Fig. 1B). Both SR-SIM and wide field deconvolution imaging modalities showed that puncta localize to regions of low-level Hoechst staining. Similar co-localization was observed with other antibodies against ATM and additional cell lines (
We next asked whether this co-localization was functionally relevant during DNA damage. Upon micro-irradiation of DNA with a 405nm laser, N17-phosphorylated huntingtin was strongly recruited to damaged DNA sites occupied by ATM (Fig. 2A). This effect was confirmed in additional cell types (29,30) to fluorescent proteins to generate chromobodies for the visualization of endogenous nuclear huntingtin in live cells. Each intrabody was fused to YFP to generate huntingtin chromobodies 1 and 2 (HCB1 and HCB2). To specifically detect endogenous huntingtin in the nucleus, the chromobodies were fused to a nuclear localization signal (NLS) to generate nucHCB1 and nucHCB2. Upon expression in human retinal pigment epithelial (RPE1) cells, which are immortalized by telomerase activity but not transformed, both nucHCB1, which recognizes the proline-rich domain amino acids 41–81 of huntingtin (30), and nucHCB2, which recognizes amino acids 49-148 (29), localized to irradiation loci (Fig. 2B and Fig. 2A and B, and
Antibodies against phosphorylated N17 and unmodified N17 produce distinct immunofluorescence patterns, suggesting that pools of huntingtin localized to different subcellular locations carry different post-translational modifications. The antibody against unmodified N17 was not enriched at irradiation sites (Fig. 3A, left panel). Thus the pool of huntingtin localized to sites of DNA damage is predominantly phosphorylated at serines 13 and 16 (Fig. 3A, right panel).
Although ATM co-localizes with huntingtin at sites of DNA damage, the N17 domain does not contain the SQ substrate recognition motif for ATM (31), thus it is unlikely that ATM directly phosphorylates N17. However, reduction of ATM levels reduces toxicity in HD models (24), suggesting a functional interaction between ATM and huntingtin. We therefore asked whether ATM kinase activity affects the localization of N17-phosphorylated huntingtin to sites of DNA damage. In the presence of ATM kinase inhibitor KU55933, the anti-phospho-N17 antibody was not enriched at sites of micro-irradiation, despite maintaining signal intensity in the nucleus (Fig. 3B). To test whether ATM inhibition was affecting N17 phosphorylation, we measured the levels of phospho-N17 in total cell lysates from cells treated with KU55933. As shown in Fig. 3C, KU55933 treatment did not affect N17 phosphorylation levels. Thus ATM activity is required for the localization of huntingtin to sites of DNA damage, and the mechanism by which ATM inhibition impedes huntingtin recruitment, or retention, to DNA damage sites does not appear to involve the modulation of N17 phosphorylation levels.
Huntingtin participates in the base excision repair pathway
Huntingtin regulation by ATM implicates it in the process of double-stranded break repair, for which ATM is most well studied. However, ATM is also activated by ROS stress (27), which is the more likely culprit for age-dependent pathology such as that seen in HD (12,14). We recently found that ROS triggers huntingtin nuclear translocation (10) via methionine 8 oxidation within the N17 domain. We therefore investigated a role for huntingtin in the base excision repair (BER) pathway, the mechanism responsible for the repair of most oxidative lesions.
Potassium bromate (KBrO3) is an oxidizing agent that induces DNA base damage and, to a lesser extent, strand breaks (32). We tested the response of endogenous huntingtin to KBrO3 treatment in live RPE1 cells using the nucHCB2 chromobody and found that it redistributed from a diffuse pattern throughout the nucleus to enrichment at euchromatic regions (Fig. 4A), similar to the response by proteins of the BER pathway (33).
In response to oxidative stress, BER proteins are recruited to the insoluble chromatin fraction, which can be visualized by extraction of soluble proteins with cytoskeleton buffer (CSK) prior to fixation (34). We found that in untreated cells, the nucHCB2 chromobody was washed away upon CSK extraction (Fig. 4B), suggesting that the chromobody is predominantly bound to soluble huntingtin under resting conditions. Upon treatment with KBrO3, nucHCB2 was retained after extraction, indicating that soluble huntingtin is recruited to the insoluble chromatin fraction upon oxidative stress.
We observed that huntingtin recruitment to sites of micro-irradiation, which generates base damage as well as single- and double-stranded breaks, is dependent on ATM kinase activity (Fig. 3B). We therefore asked whether ATM activity is also required for huntingtin localization to chromatin upon oxidative base damage caused by KBrO3. We found that the amount of nucHCB2 retained in nuclei after CSK extraction was decreased when cells were pre-incubated with KU55933 (Fig. 4B). This suggests that ATM kinase activity is necessary for huntingtin recruitment, or retention, to sites of BER. As it is unlikely that ATM activity is induced by double-stranded breaks upon KBrO3 treatment (32), this may be explained by alternate mechanisms of ATM activation, such as ROS (27,35), single-strand breaks such as those generated during BER (25), or non-break changes in chromatin structure (26).
We next compared the localization of several endogenous DNA repair proteins with that of huntingtin as measured by immunofluorescence against nucHCB2 and various endogenous huntingtin epitopes. In the absence of oxidative stress, X-ray repair cross-complementing protein 1 (XRCC1), a scaffolding protein that interacts with multiple single-strand break repair enzymes, was found at small nuclear puncta uncorrelated with euchromatic regions, while nucHCB2 was extracted by CSK buffer (Fig. 5, top panel). In response to KBrO3 treatment, the nucHCB2 chromobody was retained at regions of euchromatin occupied by XRCC1 (33), providing further evidence that huntingtin participates in the repair of oxidative DNA damage.
We also performed co-immunofluorescence of nucHCB2, huntingtin, and apurinic/apyrimidinic endonuclease (APE1), which makes single-stranded cuts in DNA during base excision repair. In response to KBrO3, nucHCB2 co-localized with APE1 at euchromatin (Fig. 5, middle panel). The mab2166 monoclonal antibody against huntingtin amino acids 181-810 was excluded from these sites but in some cases surrounded them, suggesting that the epitope is inaccessible to mab2166 within BER sites, perhaps due to an alternate conformation of huntingtin. In a similar experiment, nucHCB2 and phospho-N17 co-localized with single-stranded break marker poly-ADP-ribose (PAR) at euchromatic regions upon oxidative stress (Fig. 5, bottom panel). In contrast, neither nucHCB2 nor mab2166 staining co-localized with the double-stranded break marker γ-H2AX upon KBrO3 treatment (
In response to oxidative stress, BER enzymes are coordinated in multi-protein complexes to facilitate repair (36). We therefore tested whether huntingtin acts as a scaffold for DNA repair proteins. Anti-huntingtin immunoprecipitates from cells treated with hydrogen peroxide contained the BER proteins XRCC1, Flap structure-specific endonuclease 1 (FEN1), APE1, and high mobility group box 1 (HMGB1), as well as phosphorylated ATM (Fig. 6). Thus, in response to oxidative stress, huntingtin localizes to sites of BER (Fig. 5), and interacts with DNA repair proteins in a ROS-dependent manner (Fig. 6), consistent with a scaffolding role for huntingtin in DNA repair.
DNA repair is deficient in Huntington's disease
We next investigated whether the repair of oxidative DNA damage is impaired in HD. Fibroblasts isolated from an HD patient carrying alleles with 43 and 17 CAG repeats, and the spousal control bearing 21 and 18 CAG repeats, were immortalized by stable expression of human telomerase catalytic subunit (hTERT) (37) to generate TruHD-Q43Q17 and TruHD-Q21Q18 cell lines. hTERT immortalization maintains intact p53 function (37), especially critical to the analysis of the DNA damage response.
Alkaline comet assays revealed that TruHD-Q43Q17 cells exhibited higher basal levels of DNA damage than the spousal control cells, and incurred more damage in response to oxidative stress (Fig. 7A). The recent GWAS of HD age at onset showed that genetic variation enriched in DNA repair pathways was associated with disease onset (22). It is therefore important to note that patient cells are not isogenic and may carry gene variants residing in DNA repair pathways that could influence our results. We therefore confirmed our results in three additional patient cell samples (19). Despite this, our results are consistent with reports of excessive DNA damage in other HD patient samples and isogenic HD models, including HD patient fibroblasts and leukocytes distinct from those used in this study (15,38), postmortem brain tissue (20), transgenic mice and cell lines derived from them (16–19,21), and cell systems of huntingtin fragment overexpression (17).
Huntingtin is recruited to chromatin in response to oxidative stress (Figs 4 and 5). It is therefore possible that the deficient DNA repair seen in HD cells is due to dysregulated recruitment of huntingtin to chromatin. To test this, we compared the recruitment of endogenous huntingtin to chromatin in TruHD-Q43Q17 and TruHD-Q21Q18 cells upon KBrO3 treatment. As was observed with RPE1 cells (Fig. 4), the nucHCB2 bound to endogenous huntingtin in fibroblasts was extracted by CSK buffer in untreated cells and retained at chromatin upon KBrO3 treatment (Fig. 7B). We did not detect a difference in the amount of chromatin-bound huntingtin in TruHD-Q43Q17 cells compared to TruHD-Q21Q18 cells, suggesting that HD cells are not impaired in their ability to recruit huntingtin protein to sites of DNA damage. Since TruHD-Q43Q17 cells express one normal copy of the huntingtin protein, we tested the recruitment of huntingtin to DNA damage in cells bearing two expanded alleles. As shown in Fig. 7C and D, huntingtin protein was competent in its localization to DNA damage induced by micro-irradiation in cells either heterozygous or homozygous (Q50/Q40) for the mutant HTT allele, indicating that polyglutamine-expanded huntingtin protein retains the ability to localize to damaged DNA. In summary, mutant huntingtin is capable of reaching damaged DNA, but DNA repair is deficient in HD cells. It is therefore possible that mutant huntingtin somehow interferes with the normal repair of DNA damage, a concept consistent with the dominance of mutant huntingtin over wild type protein in Huntington's disease.
This study describes the direct localization of full-length, endogenous human huntingtin protein to sites of DNA damage, which is dependent on ATM kinase activity. Further, we report the ROS-dependent interaction of huntingtin with DDR proteins and the deficient repair of oxidative DNA damage in HD patient cells.
Huntingtin responds to oxidative damage
Using modified huntingtin-specific intrabodies (29,30), we found that endogenous huntingtin in live cells was recruited to sites of DNA damaged by irradiation. The chromobodies, which bind soluble huntingtin under resting conditions, also redistributed to insoluble regions of euchromatin upon oxidative stress, much like the classic behaviour of BER proteins (33,34). This redistribution of soluble huntingtin is regulated by the redox-sensing function of methionine eight (M8) within the N17 domain recently reported by our group (10). We reported that in response to ROS, the N17 alpha helix undergoes a conformational change, releasing huntingtin from the ER and enhancing S13/S16 phosphorylation and huntingtin nuclear translocation. The current study extends this understanding of the ROS-dependent nucleo-cytoplasmic shuttling of huntingtin to include its direct localization to sites of DNA damage.
The oxidation event that changes N17 structure is allosterically transduced along the total huntingtin protein, in a mechanism proposed in the past by us (9) and recently shown by electron microscopy of full-length huntingtin (39) as well as for other HEAT-repeat proteins (40). Like ATM, huntingtin is rich in HEAT repeats, which mediate intra- and inter-molecular interactions to impart scaffolding function (41). In light of its ROS-dependent interaction with DDR proteins, we propose that huntingtin is a scaffolding protein for the DNA repair machinery that changes conformation according to cellular ROS levels via N17 (M8) ROS-sensing (10).
Accumulation of DNA damage due to age-associated ROS has been suggested as a pathogenic mechanism for several neurodegenerative disorders including Alzheimer’s and Parkinson’s diseases (12–14). This may also be the case in HD if huntingtin has a functional scaffolding role in the DNA damage response that is dysregulated upon polyglutamine expansion, thus preventing the timely resolution of age-related oxidative DNA damage.
In HD, this may be further compounded by the somatic CAG expansion observed in brain regions most heavily affected by the disease (42). Somatic CAG instability is driven by ROS and results from improper BER and mismatch repair (MMR) (43). Enzymes that participate in BER and MMR modulate the process of CAG expansion in HD (43,44) and also modify disease progression (22,23). In recent studies dosing the Hdh(CAG150) model with a potent anti-oxidant, the disease was attenuated, as was CAG expansion (45). It is therefore possible that age-associated ROS, in conjunction with dysregulated mutant huntingtin function in DNA repair, may lead to somatic DNA instability in the HTT gene. This would result in a temporal cascade of progressive polyglutamine expansion and thus more dysfunctional huntingtin protein as HD progresses.
Phenotypes observed in the BACHD mouse model, bearing a 97-repeat CAA-CAG-encoded huntingtin gene designed not to expand, suggest that the somatic repeat instability may not drive neuropathogenesis in HD (46). However, the excessive repeat lengths necessary to elicit disease phenotypes in mouse models, from either synthetic sources or rare, severely expanded juvenile HD individuals, may in fact mimic the somatically expanded pathogenic huntingtin seen in typical HD patients at the age of onset. If so, the pre-manifest therapeutic window to prevent CAG expansion to the pathogenic length may not be represented in some existing mouse models.
ATM regulates the huntingtin DNA damage response
In addition to the pool of soluble huntingtin that responds to ROS, a pool of huntingtin phosphorylated at N17 exists at chromatin-dependent, detergent-resistant nuclear puncta, which we now show also contain the DNA repair protein ATM. ATM inhibition is protective in models of HD (24), and we show here that it blocks the recruitment, or retention, of huntingtin to sites of DNA damage. As expanded huntingtin protein is capable of localizing to DNA damage sites, it could be speculated that inhibiting the localization of mutant huntingtin to damage sites is beneficial, for example, if mutant huntingtin slows or impairs the resolution of the repair process. This mechanism is consistent with the recent report that HD fibroblasts exhibit delayed clearance of DDR proteins after gamma irradiation (15). Mutant huntingtin is also associated with retarded clearance of nuclear actin stress rods (47) and cytoplasmic stress bodies (48), and it is possible that its numerous inappropriate protein-protein interactions (49) disrupt the timely resolution of stress response pathways.
Alternatively, mutant huntingtin dysregulation in the DNA repair process could result in the observed accumulation of excessive DNA damage in HD models and patient samples, leading to prolonged ATM activation (24). ATM mediates apoptosis via p53, a major regulator of the DNA damage response that directly regulates transcription of the HTT gene (50) and has been implicated in HD pathogenesis (51,52). Thus the benefits of ATM inhibition in HD models (24) may be the result of not only inhibiting mutant huntingtin recruitment to DDR complexes, but also mitigating the downstream effects of dysregulated DNA repair caused by the expanded protein.
Our data support the pursuit of ATM as a therapeutic target. In addition, we found that the huntingtin localized to sites of damage is phosphorylated at N17, and we previously showed that mutant huntingtin is hypo-phosphorylated at this domain (2). The requirement of N17 phosphorylation for huntingtin function in DNA repair would explain the robust beneficial effects of pharmacological or genetic restoration of N17 phosphorylation (2,7,8), and justifies augmentation of mutant huntingtin N17 phosphorylation as a valid sub-target for therapeutic intervention in Huntington's disease.
Materials and Methods
All reagents were from Sigma Aldrich unless otherwise stated.
Antibodies against huntingtin phosphorylated at serines 13 and 16 of the N17 domain (anti-phospho-N17) and unmodified at serines 13 and 16 (anti-N17) were previously characterized and validated (2). Anti-ATM antibodies were from from Santa Cruz Biotechnology. Additional antibodies used in this study: mab2166 (Millipore); FITC-conjugated anti-GFP, anti-XRCC1 for western, anti-HMGB1, anti-GAPDH, and anti-γH2AX (Abcam); anti-XRCC1 for immunofluorescence and anti-FEN1 (Bethyl Laboratories); anti-APE1 (Novus Biologicals); anti-PAR mAb 10H (Enzo Life Sciences). Secondary antibodies against rabbit, mouse, and goat IgG, conjugated to Alexa Fluor 488, 555, 594, or Cy5, were from ThermoFisher Scientific. Anti-rabbit IgG and anti-mouse IgG HRP conjugates were from Abcam. Catalogue numbers for primary antibodies are listed in the supplemental experimental procedures.
Primary human fibroblasts from the Coriell Institute for Medical Research Biorepositories were cultured in Minimum Essential Media supplemented with 15% fetal bovine serum and 1% Glutamax at 37 °C in a 5% CO2 incubator. Fibroblasts used in this study: GM02153 (wild type female, clinically normal widow of affected husband), ND30013 (male HD patient bearing 43/17 CAG repeats, 54 years of age at sampling), ND30014 (wild type spousal control to ND30013 bearing 21/18 CAG repeats, 52 years of age at sampling), GM04857A (female homozygous patient bearing 50/40 CAG repeats, 23 years of age at sampling), ND33391 (wild type female bearing 19/17 CAG repeats), ND30626 (male HD patient bearing 41/17 CAG repeats, 62 years of age at sampling), GM09197 (male juvenile HD patient bearing 180 CAG repeats, 6 years of age at sampling). ND30013 and ND30014 cell lines were immortalized by transduction with hTERT-expressing lentivirus (Hung et al, manuscript in preparation) to generate the TruHD-Q43Q17 and TruHD-Q21Q18 cell lines, which were cultured as above. The hTERT-immortalized retinal pigment epithelial cell line, RPE1 (ATCC), was cultured in DMEM/F12 1:1 media supplemented with 10% FBS and 0.01 mg/ml hygromycin B at 37 °C in a 5% CO2 incubator. HEK 293 cells were cultured in alpha-MEM containing 10% FBS at 37 °C in a 5% CO2 incubator. STHdh cells were cultured in DMEM media supplemented with 10% FBS. All media and supplements were from ThermoFisher Scientific.
RPE1 cells were transfected with Turbofect (ThermoFisher Scientific) according to manufacturer’s specifications. Fibroblasts were transfected using the 4D-Nucleofector device (Lonza) with the Amaxa SG Cell Line X Kit L and the NIH3T3 program. The nucHCB1 and nucHCB2 vectors were generated as described in the supplemental experimental procedures. The pH2B mCherry-IRES-puro2 vector (53) was a gift from Daniel Gerlich (Addgene plasmid # 21045).
Cells were grown in glass-bottom tissue culture dishes to approximately 80% confluence prior to the indicated treatments. Cells were fixed in methanol at −20 °C for 10 min then washed with wash buffer (50 mM Tris-HCl, pH 7.5, 150mM NaCl, 0.1% Triton X-100) and blocked for 1 h at room temperature with blocking buffer (wash buffer + 2% FBS). Cells were incubated with primary antibodies diluted in blocking buffer for either 1 h at room temperature or overnight at 4 °C before washing and incubation with secondary antibodies diluted in blocking buffer for 30 min at room temperature. After washing, nuclei were stained with Hoechst (0.2 μg/ml in PBS) for 5 min at room temperature, washed, and imaged in PBS. Cells grown for SR-SIM imaging were seeded on to #1.5 coverslips, fixed and stained as above, and mounted on to glass slides using frame slide chambers (Bio-Rad).
Extraction of soluble proteins and quantification of nuclear intensity
Extraction of soluble proteins was carried out by incubating cells with the cytoskeleton (CSK) buffer (100mM NaCl, 300mM glucose, 10 mM PIPES pH 6.8, 3 mM MgCl2, 0.5% Triton X-100) on ice for 5 min prior to fixation and immunofluorescence. In RPE1 cells, the nuclear intensity of endogenous huntingtin staining was measured by immunofluorescence with FITC-anti-GFP to detect nucHCB2, and cells were imaged by wide field epifluorescence microscopy as described below. In hTERT-immortalized fibroblasts, nucHCB2 was co-transfected with histone H2B-mCherry and cells were fixed with 4% paraformaldehyde for 15 min at room temperature. Cells were imaged by confocal microscopy as described below. Nuclear intensity in each case was quantified using a homemade MATLAB script as described in the supplemental experimental procedures.
Induction of DNA damage
For micro-irradiation experiments, cells were grown overnight in glass-bottom dishes then stained with NucBlue (ThermoFisher Scientific) for 15 min at 37 °C in a 5% CO2 incubator. Media was replaced with Hanks’ Balanced Salt Solution (HBSS) (ThermoFisher Scientific) immediately prior to micro-irradiation. Cells were irradiated using the Nikon C2+ confocal system described below, equipped with an InVivo Scientific environmental chamber held at 37 °C. Regions of interest (∼20 pixels each) were irradiated at a scan speed of 1/16 frames per second (512×512) using a Coherent OBIS 405nm diode laser set to 100% power. XY-coordinates were recorded, and cells were incubated at 37 °C for the indicated periods prior to fixation and immunofluorescence. XY-coordinates were then revisited to image the irradiated cells. For live cell imaging, RPE1 were cells transfected with huntingtin-specific chromobodies then subjected to micro-irradiation as described above. During the incubation period, cells were imaged at 5-min intervals using wide field epifluorescence microscopy as described below. To induce oxidative DNA damage, medium was replaced with either HBSS (mock) or HBSS + 40mM KBrO3, or HBSS + 100μM H2O2 and cells were processed as indicated in the figure legends.
Wide field epifluorescence microscopy was done on a Nikon Eclipse Ti wide field epifluorescent inverted microscope using the PLAN APO 60×/1.4 oil objective, PLAN FLUOR 40×/0.6 dry objective, and PLAN APO 20×/0.75 dry objective and Spectra X LED lamp (Lumencor). Images were captured using a Hamamatsu Orca-Flash 4.0 CMOS camera. Image acquisition and deconvolution of Z-stacks were done with NIS-Elements Advanced Research version 4.30 64-bit acquisition software (Nikon). Super-resolution imaging was done on the Nikon N-SIM super-resolution microscope system attached to a Nikon Eclipse Ti inverted microscope using an APO TIRF 100×/1.49 oil objective, and 405nm, 488nm, and 561nm lasers (Coherent Inc). Images were captured using a Hamamatsu Orca-Flash 4.0 CMOS camera and acquired with the NIS-Elements software version 4.50. For micro-irradiation experiments, the 405nm laser was part of the Nikon C2+ confocal system attached to a Nikon Eclipse Ti inverted microscope, using a PLAN APO 60×/1.4 oil objective. The confocal system used for quantification of nuclear intensity in hTERT-immortalized fibroblasts was the Nikon A1+ confocal system, equipped with GaAsP detector, attached to a Nikon Eclipse Ti inverted microscope, using a PLAN APO 20×/0.75 dry objective.
Cell lysis and co-immunoprecipitation
HEK 293 cells were lysed in RIPA buffer (50 mM Tris-HCl pH 8.0, 150 mM NaCl, 1% NP-40, 0.25% sodium deoxycholate, 1 mM EDTA, protease and phosphatase inhibitors (Roche) and lysates were sonicated using the Branson Sonifier 250 for 2 × 15 pulse cycles at output power setting 3 and 10% duty cycle. Sonicated lysates were cleared at 17000 x g for 10 min at 4 °C and supernatants were analysed by western blotting or used in co-immunoprecipitation assays as described below. For western blotting, proteins were separated by SDS-PAGE on a 4–20% gradient gel (Bio-Rad) and transferred to a PVDF membrane (Millipore). Membranes were blocked in TBST (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.1% Tween-20) containing 5% skim milk powder for 1 h then probed with primary antibodies in the same buffer overnight at 4 °C. Membranes were washed six times with TBST-2.5% milk then probed with secondary antibodies for 30 min at room temperature. After washing as above, membranes were incubated with ECL (Millipore) and imaged with a DNR MicroChemi chemiluminescence detector. For co-immunoprecipitation, lysates were pre-cleared with protein G-sepharose beads and input samples acquired prior to the addition of mab2166 with rotation overnight at 4 °C. Lysates were then incubated with protein G-sepharose with rotation for 1 h at 4 °C and supernatants discarded. Beads and associated proteins were washed three times with RIPA lysis buffer then denatured in SDS loading buffer at 95 °C for 5 min. Solubilized proteins were analysed by western blotting as above.
Alkaline comet assay
Cells treated as indicated were harvested by incubation with 20 mM EDTA in PBS for 5 min at 37 °C, resuspended in low melting point agarose and plated on agarose-coated slides. After agarose solidification for 15 min at 4 °C, slides were incubated in lysis buffer (10 mM Tris-HCl pH 10, 2.5M NaCl, 0.1M EDTA, 1% Triton X-100) for 2 h at 4 °C, then in alkaline buffer (0.3M NaOH, 1 mM EDTA) for 30 min at 4 °C. Slides were electrophoresed in alkaline buffer for 30 min at 300mA, washed 3 × 2 min in cold dH2O, then fixed in cold 70% ethanol for 5 min. After drying of agarose, DNA was stained with Vista Green DNA dye (Cell Biolabs) for 15 min at room temperature, washed with dH2O, and dried. Images were acquired on the EVOS FL Auto system (ThermoFisher Scientific) using a PLAN FLUOR 10×/0.3 objective. Images were analysed with the OpenComet plugin for ImageJ (54) and tail moment (tail % DNA x tail length) was calculated for 50–100 cells per condition.
The authors would like to thank Rohan Philip for assistance with structured illumination microscopy, Nicholas Caron for initiating the huntingtin-specific chromobody project, and Amber Southwell for providing the Happ1 intrabody.
Conflict of Interest Statement. None Declared.
This work was supported by operating grants to R.T. from the Huntington Society of Canada, the Canadian Institutes of Health Research (CIHR MOP-119391), the Krembil Family foundation, and CHDI Inc.