Part 2 ERCC1 Mutations Impede DNA Damage Repair And Cause Liver And Kidney Dysfunction in Patients

Contact:ali.ma@wecistanche.com

Click here to part 1

Part 2

Click to Cistanche NZ for kidney disease

Discussion

A new case of bi-allelic ERCC1 mutations Only two individuals with bi-allelic ERCC1 mutations have been reported to date, both of whom displayed CS-like features. The most severely affected individual (165TOR) had a nonsense variant (Q158X) and an F231L missense variant in the (HhH)2 motif of ERCC1 (Fig. S5 F), displayed growth retardation, developmental failure, and contractures, and died after the first year of life due to pneumonia (Jaspers et al., 2007). The second affected individual (CS20LO) had contractures, microcephaly, and hypertonia and was homozygous for the F231L missense variant and died in the second year of life (Kashiyama et al., 2013).

We report two siblings with bi-allelic ERCC1 mutations—a paternally inherited missense variant (p.R156W; c.466C>T) and a null-allele due to a maternally inherited intragenic deletion. All the remaining ERCC1 protein expressed in cells from these patients carries the R156W amino acid substitution, which disrupts a salt bridge between the positively charged R156 residue and the opposing negatively charged D129 residue located just below the XPA-binding pocket in the central domain of ERCC1. The impact of this substitution is twofold: (1) it strongly diminishes the overall stability of ERCC1 and its binding partner XPF, and (2) it specifically affects the inter-action with XPA (Fig. 9).

Reduced nuclear protein levels of ERCC1 due to partial misfolding

Western blot and immunofluorescence analyses revealed that the overall ERCC1 and XPF protein levels are dramatically reduced to below 20% of WT levels in the fibroblasts from both siblings. This effect is also recapitulated by the bi-allelic KI of the missense mutation (p.R156W; c.466C>T) in the endogenous ERCC1 locus of RPE1-hTERT cells. Interestingly, complete loss of ERCC1 in mice led to death within 4 wk, while increasing protein levels to around 15% of the levels in WT mice increased the life span fivefold (Weeda et al., 1997), suggesting that even low levels of ERCC1 considerably increase the potential for viability. Consistent with our findings, previous studies have shown that missense mutations in the central domain of ERCC1 generally cause destabilization (Sijbers et al., 1996b), indicating that this region is important for protein stability.

The recently reported cryogenic electron microscopy structure of the full-length ERCC1-XPF heterodimer (Jones et al., 2020) reveals that the ERCC1R156 residue is located on the very edge of the heterodimer. We speculate that the increased bulkiness and altered electronic structure of the substitution of an arginine by tryptophan not only disturbs the XPA-binding pocket but also the dimerization interface between the central domain of ERCC1 and the nuclease domain of XPF, causing general destabilization (Fig. S5 F). Of interest, the missense variant XPFR799W is located at the same interface on the XPF side (Fig. S5 F) and also results in destabilization and severely reduced XPF protein levels (Mori et al., 2018). In one affected individual (CALIF1010) who presented with CS-like and segmental progeroid features, the XPFR799W variant was inherited together with an intergenic deletion in the other XPF allele, resulting in a strong NER defect combined with a mild ICL repair defect (Mori et al., 2018), similar to the siblings reported in this study.

Ectopic expression of an inducible version of ERCC1R156W for24 h enabled us to reach WT levels of this mutant protein, which resulted in mislocalization in the cytoplasm, reduced interaction with XPF, and increased interaction with protein-folding chaperones, as detected by MS. These findings all, lend support to the notion that ERCC1R156W is partially misfolded and degraded in patient cells, resulting in an -80% reduction in ERCC1 nuclear protein levels. KI of ERCC1R156W in RPE1-hTERT cells supports this conclusion.

When comparing the ERCC1 and XPF protein levels in PV46LD and PV50LD cells with those in cells from more severely affected individuals (165TOR and CS20LO), we noted that these levels were comparable, or perhaps even lower, in cells from the siblings described in this study (Fig. 3 C and Fig. S3, E and F; Jaspers et al., 2007; Kashiyama et al., 2013). Note that CS20LO is homozygous for ERCC1F231L, while 165TOR is heterozygous and carries an additional premature stop on the other ERCC1 allele (Q158X). These findings suggest that in addition to reduced protein levels, the functionality of the mutant protein that is still expressed and the combination of the two mutated alleles need to be taken into account as a potential modulator of phenotypic severity and expressivity.

ERCC1R156W: Differential impact on ERCC1-dependent DNA repair pathways

Expression of ERCC1R156W-GFP in ERCC1-KO cells at levels similar to ERCC1WT enabled us to disentangle the impact of the amino acid substitution on protein stability from the impact on protein functionality. While the mutant ERCC1R156W protein was severely impaired in its UV-induced interaction with XPA and fish and failed to efficiently localize to sites of UV-induced DNA lesions, we still detected considerable (∼40%) repair activity in UDS assays, clonogenic survivals, and in vitro NER assays. These findings suggest that a mutant ERCC1 protein that only interacts very weakly with the NER complex can still support considerable levels of NER (∼40%) activity inside cells. Together, these findings suggest that the low expression level ofERCC1 (∼20%) combined with the residual GGR activity of the mutant proteins that are still expressed provides sufficient protection to prevent PV46LD and PV50LD from developing full-blown XP. Indeed, residual GGR activity was still detected under more sensitive conditions in UDS assays when compared with fibroblasts from a severely affected and cancer-prone XP-A patient. Nonetheless, both patients could still be at risk for developing skin cancer, and protection from sunlight is therefore advised. The residual activity in PV46LD and PV50LD cells were either similar to or lower than the repair activity we detected in 165TOR or CS20LO cells (Fig. 6 J; Jaspers et al., 2007; Kashiyama et al., 2013). The clinical severity, therefore, does not correlate with NER activity but is likely due to a role of ERCC1 outside NER that is important during development.

Despite our findings that ERCC1R156W showed a reduced interaction with SLX4 and was also recruited less efficiently to sites of local ICL induction, further analysis revealed that the ERCC1R156W mutant protein was only mildly affected in supporting ICL repair when ectopically expressed in ERCC1-KO cells. However, the impact on ICL repair was much milder than the impact of the R156W substitution on NER, which, combined with earlier findings that the repair of ICLs requires much less ERCC1 protein than the repair of NER-specific DNA lesions (Jaspers et al., 2007; Sijbers et al., 1996b), may very well explain the mild impact on ICL repair. In patient fibroblasts, which have strongly reduced levels of ERCC1, we did detect moderate sensitivity, as well as increased chromosome breakage upon exposure to the ICL-inducing compound MMC, albeit much milder than cells from an FA patient that were included in parallel. Importantly, neither PV46LD nor PV50LD has displayed any overt signs of FA-like clinical features, suggesting that the ERCC1R156W mutant protein provides sufficient protection against the low endogenous ICL load in these patients.

ERCC1 deficiency causes liver and kidney impairment

Mice that are KO for either ERCC1 or XPF to show severe running (i.e., smaller and weaker than WT mice) and die before the first 4 wk of life due to severe liver failure (McWhir et al., 1993; Sijbers et al., 1996b; Tian et al., 2004; Weeda et al., 1997). Liver-specific expression of ERCC1 in these mice partially corrected the smaller size and extended life span up to 12 wk but also unmasked severe kidney dysfunction and renal failure as a secondary cause of death (Selfridge et al., 2001). KO mice specific for XP or FA do not display liver disease, suggesting that this phenotype is not caused by a NER or ICL repair defect. It is possible that partial redundancy between these two pathways, which is lost in ERCC1-KO mice deficient in both pathways, could explain this phenotype. In line with this idea, hepatocyte and kidney-proximal tubule cells became polyploid in ERCC1-KO mice and accumulated high levels of p53 (McWhir et al., 1993; Selfridge et al., 2001; Weeda et al., 1997), suggesting that accumulation of endogenous DNA damage in these organs may occur. The precise nature of this endogenous DNA lesion, however, remains unclear.

Interestingly, mice with joint inactivation of both NER and ICL repair do not display liver dysfunction, suggesting that redundancy between these DNA repair pathways is not the sole explanation and that an additional function of ERCC1-XPF outside these DNA repair pathways contributes to the liver impairment (Mulderrig and Garaycoechea, 2020). It is possible that ERCC1-XPF is involved in additional DNA repair pathways that deal with endogenous DNA damage. For instance, ERCC1-XPF was recently shown to act upon alternative DNA structures, such as Z-DNA, during which it cooperates with mismatch repair proteins rather than NER or ICL repair factors (McKinney et al., 2020). In addition, ERCC1-XPF was shown to be involved in a sub-pathway of base excision repair together with the RECQ1 helicase (Woodrick et al., 2017).

The two siblings (PV46LD and PV50LD) in this study exhibited a failure to thrive, short stature, and a lack of subcutaneous fat, reminiscent of the small size observed in ERCC1-KO mice (McWhir et al., 1993; Selfridge et al., 2001; Weeda et al., 1997). Moreover, the siblings developed liver impairments with a predominantly cholestatic picture that led to orthotopic liver transplantation before the age of 9 yr. This intervention clearly prevented early death but did not improve the failure to thrive, as evident from post-transplant growth of weight, height, and head circumference well below the third centile.

Of interest, cholestatic liver disease was reported as a common feature in Egyptian CS patients, suggesting that this should be monitored more closely in CS patients from other ethnic backgrounds to see if this is a more common feature than previously recognized (Abdel Ghaffar et al., 2011). Although no liver abnormalities have been described in XP patients, there are some cases of hepatic cytolysis (cell breakdown or bursting) in FA patients (Masserot-Lureau et al., 2012), which may be distinct from the liver abnormalities observed in the ERCC1- deficient siblings. Histological examination of a liver biopsy from the older sibling (PV50LD) revealed changes in hepatocyte morphology with nuclear enlargement and variability, as noted in the ERCC1-deficient mice (Fig. S1 D; N´uñez et al., 2000).

The second cause of death in ERCC1-KO mice with liver-specific expression of ERCC1 was severe kidney failure (Selfridge et al., 2001). Both siblings also display renal dysfunction, with features suggestive of proximal tubular dysfunction leading to progressive kidney impairment. However, tubular function stabilized following the liver transplant, but kidney function requires ongoing monitoring. The renal phenotype of the ERCC1- KO mice—with dilated tubules containing leaked proteinaceous material consistent with a tubulopathy (McWhir et al., 1993; Selfridge et al., 2001; Weeda et al., 1997)—is similar to that of the patients described in our study. Moreover, renal impairment and, in some instances, failure has been reported in combined XP/CS (Ben Chehida et al., 2017; Kondo et al., 2016; Kralund et al., 2013) as well as CS (Funaki et al., 2006; Reiss et al., 1996; Sato et al., 1988), while an association between FA and structural anomalies of the kidney has also been reported (Sathyanarayana et al., 2018). The phenotype in these siblings is distinct from prior reports of individuals with biallelic ERCC1 mutations and also distinct from the known phenotypic entities, CS and XP. Neither individual had features suggestive of FA. There may be parallels between the phenotype of these siblings and a previously reported individual with biallelic ERCC1 mutations (XP2020DC; p.K266X; IVS6-26G>A) who died at the age of 37 (Gregg et al., 2011; Imoto, K., et al. 2007. Patients with defects in the interacting nucleotide excision repair proteins ERCC1 or XPF show xeroderma pigmentosum with late-onset severe neurological degeneration. [Abstract] J. Invest.Dermatol. 127:S92). Although no liver impairment was reported, this patient developed severe brain atrophy at age 15, which developed into progressive neurodegeneration with dementia. Mild brain atrophy has also been noted in both siblings at age 13and 11, indicating that the neurological development of both siblings should be monitored carefully.

A previously reported 15-yr-old boy with bi-allelic XPF mutations (XP51RO; p.R153P; c.458G>C; Jaspers et al., 2007; Niedernhofer et al., 2006) presented with a phenotype with some overlap with the siblings reported here, including photosensitivity without skin cancer, short stature, lack of subcutaneous fat, developmental delay, and renal insufficiency (Niedernhofer et al., 2006). Distinct from the siblings reported here, this boy showed pronounced segmental progeroid features, but a much milder liver picture without cholestasis. The siblings reported here demonstrate that bi-allelic ERCC1 mutations can cause a spectrum of phenotypes, from that typically seen in CS through to a phenotype comprising milder short stature, photosensitivity, and severe liver and kidney impairment, potentially because of a combined strong impact on NER and a simultaneous impact on ICL repair, which may particularly affect these organs.

Materials and methods

All procedures performed in this study are in accordance with the ethical standards and were approved by the Human Ethics Committee of the Royal Children’s Hospital, Melbourne, Victoria, Australia (HREC36291C) and the UK National Research Ethics Service Committee North East–Newcastle and North Tyneside 2. Patients were recruited into Undiagnosed Diseases Programs for pediatric patients with presumed “orphan” Mendelian disorders. Saliva or blood samples from the patients and their family members were collected for genomic DNA extraction after written informed consent was given. The parents have granted permission to show unredacted photographs of the affected siblings in Fig. 1 A.

Next-generation sequencing (NGS) DNA extracted from the blood of both affected siblings and unaffected parents was subjected to exome capture and sequencing through Oxford Gene Technologies Ltd. Genomic data were analyzed using a custom-made in-house pipeline using an autosomal recessive disease model. A paternally inherited missense variant in NM_202001.2 (ERCC1), g.45922415G>A, c.466C>T was identified in both affected individuals in exon 4 of 8 of the ERCC1 gene. On initial interrogation of the exome data in both affected individuals, this variant appeared to be in a homozygous state but was present in a heterozygous state in the father and absent in the mother, suggesting a possible deletion on the maternal allele. The missense variant is present in the gnomAD database at a frequency of 0.01% (22 heterozygotes, no homozygotes) and has not been previously reported in affected individuals. The missense variant has been deposited in the LOVD database (http://www.LOVD.nl/ERCC1_000020 and http://www.LOVD.nl/ERCC1_000021) and the ClinVar database (variation ID: 978472).

Detection of the deletion by qPCR

To detect the deletion, genomic DNA was extracted from lymphocytes/epithelial cells, and the relevant region of the ERCC1 gene was amplified in a qPCR. qPCR was performed using probes and specific primers (listed in Table S1) designed using the Universal ProbeLibrary Assay Design Center (Roche) and synthesized by Sigma-Aldrich. The primers were designed to include the deleted region in exon 4 and a region within the gene that is not deleted as of a control (exon 5). A multiplex qPCR using the LightCycler R 480 Probes Master reaction mix (Roche) was performed according to the manufacturer’s protocol to amplify and quantify the ERCC1 gene region, using the CFTR gene (exon 27) as an internal standard (Table S1). The qPCR was performed on the LightCycler R 480 instrument (Roche), and data analysis was performed using the LightCycler R 480 software version 1.5.0 (Roche). American College of Medical Genetics guidelines was followed for interpretation of sequence variation.

Detection of the missense variant by Sanger sequencing

Genomic DNA was isolated by resuspending cell pellets in whole-cell lysate buffer (50 mM KCL, 10 mM Tris, pH 8.0,25 mM MgCl2, 0.1 mg/ml gelatin, 0.45% Tween-20, and 0.45%NP-40) containing 0.1 mg/ml Proteinase K (EO0491; ThermoFisher Scientific) and incubating for 1 h at 56°C followed by 10-min heat inactivation of Proteinase K at 96°C. Fragments of∼1 kb spanning the missense mutations were PCR amplified(sequencing primers are listed in Table S1) followed by Sanger sequencing using either the forward or the reverse primer.

Cell lines

Fibroblast cell lines were established from skin biopsies from both affected individuals and parents. All cell lines (listed in table S2) were cultured at 37°C in an atmosphere of 5% CO2 indium (Thermo Fisher Scientific) supplemented with penicillin/streptomycin (Sigma-Aldrich) and 10% FBS (Bodinco BV). All cell lines were routinely tested for mycoplasma infection.

Fibroblast immortalization with hTERT

The WT (48BR) and PV50LD fibroblasts were immortalized by nucleofection of p Babe-Neo-TERT (Addgene Plasmid #1774)using the Amaxa Nucleofector (program U23). Each electro-poration contained 400,000 fibroblasts and 2.5 µg plasmid DNA. After electroporation, cells were seeded in 25-cm2 tissue culture flasks containing 5 ml DMEM and 10% FBS. After 2 d, the culture medium was replaced with a medium containing 20 µg/ml neomycin.

/streptomycin (Sigma-Aldrich) and 10% FBS (Bodinco BV). All cell lines were routinely tested for mycoplasma infection.

Plasmid constructs

All plasmids used in this study are listed in Table S3. The GFP gene in pERCC1-GFP-N1 (Houtsmuller et al., 1999) was replaced with m Venus (a gift of Joachim Goedhart, Amsterdam, Netherlands) using AgeI and BsrGI. The GFP-NLS gene (Luijsterburg et al., 2017) was inserted into pcDNA5-FRT-TO-Puro. The ERCC1-m Venus cassette was amplified by PCR (see Table S1 for primers) and inserted as a NheI and BspEI fragment into plenty-CW57-TO-GFP digested with NheI and AgeI to generate plenty-CW57-TO-ERCC1WT-mVenus. Overlap PCR was used to introduce the c.466C>T mutations in ERCC1. The overlap PCR product was inserted as a NheI and AgeI fragment to generate plenty-CW57-TO-ERCC1R156W-mVenus. ERCC1WT and ERCC1R156Wwere inserted as NheI and AgeI fragments into pcDNA5-FRT-TOPuro-EGFP-N1 to generate pcDNA5-FRT-TO-Puro-ERCC1WT-EGFPand pcDNA5-FRT-TO-Puro-ERCC1R156W-EGFP. Site-directed mutagenesis was used to introduce point mutations in pMacroBac-XPF-ERCC1WT using the KOD-plus mutagenesis kit (Toyobo)as described in the manufacturer’s protocol, using the primers listed in Table S1. All sequences were verified by Sanger sequencing.

Generation of KO cell

lines to generate single KOs, U2OS(FRT) and RPE1-hTERT (PuroR/TP53-dKO) cells were cotransfected with pLV-U6g-PPB encoding a guide RNA from the Leiden University Medical Center/Sigma-Aldrich single-guide RNA (sgRNA) library (see Table S3for plasmids and Table S4 for sgRNA sequences) targeting theERCC1 gene together with an expression vector encoding Cas9-2A-GFP (pX458; Addgene #48138) using Lipofectamine 2000(Invitrogen). Transfected U2OS(FRT) cells were selected on puromycin (1 µg/ml) for 3 d and plated at low density, after which individual clones were isolated. Transfected RPE1-hTERTcells were FACS sorted on BFP/GFP and plated at low density, after which individual clones were isolated. KO clones were verified by Western blot analysis. The absence of Cas9 integration/stable expression was confirmed by Western blot.

Generation of stable cell lines

A single ERCC1 KO clone in the U2OS(FRT) background (see table S2) was used to stably express ERCC1WT-GFP, ERCC1R156W-GFP, or GFP-NLS by cotransfection of pcDNA5-FRT-TO-Puroplasmid encoding these cDNAs (2 µg), together with pOG44plasmid encoding the Flp recombinase (0.5 µg). A polyclonal cell population was obtained after selection on 1 µg/ml puromycin and 4 µg/ml blasticidin S. Expression of the GFP-tagged ERCC1proteins or GFP-NLS was induced by the addition of 2 µg/ml doxycycline for 24 h.

Generation of KI cell lines

To generate homozygous KIs, RPE1-hTERT cells were first treated for 30 min with 1 µM DNA-PK inhibitor (NU7441) to suppress error-prone DSB repair and increase the use of homology-dependent repair. Subsequently, 350,000 cells were resuspended in 20 µl nucleofector buffer (V4XP-3032; Lonza) in the presence of 4.5 µg plasmid containing Cas9 and sgRNA targeting the ERCC1 gene (see Table S3 for plasmids and Table S4 forsgRNA sequences) and 100 µM single-stranded donor DNA-containing the patient mutation as well as silent mutations to destroy the PAM site to prevent recutting by Cas9 and silent mutations to introduce a HindIII restriction site to facilitate screening of KI clones. Cells were electroporated using anAmaxa 4D-X Nucleofector Unit (Lonza) using the EA-104 program. Following electroporation, cells were plated at low density in McCoy medium with 10% FBS. The next day, the medium was replaced with DMEM (10% FBS and 1% penicillin/streptomycin), and cells were allowed to form colonies. Approximately 400colonies were isolated and expanded. Genomic DNA was isolated, and a fragment including the introduced mutation and the introduced restriction site (HindIII) was amplified by nested PCR(see Table S1 for primers). PCR fragments were digested with HindIII (NEB) for 3 h at 37°C and then separated by gel electrophoresis. The clones that had the introduced HindIII restriction site were analyzed by Sanger sequencing (see Table S1 for primers). We obtained two homozygous KI clones among 400isolated clones, which corresponds to a KI efficiency of ∼0.5%.

Lentiviral transduction

for lentiviral transduction, mVenus-ERCC1WT or mVenusERCC1R156W was inserted into lentiviral vector pLenti-CW57-TO. HEK293T was transfected with vectors encoding theseERCC1 fusions, VSV-G, RRE, and REV using JetPEI (SigmaAldrich) to produce the virus. The virus-containing supernatant was collected after 24 h and filtered with a 0.44-µm filter. Primary fibroblasts PV46LD and PV50LD were lentivirally transduced in the presence of polybrene (Sigma-Aldrich). Cells were selected with 15 µg/ml puromycin. Expression of the mVenus-tagged ERCC1 proteins was induced by the addition of 2 µg/ml doxycycline for 24 h.

Clonogenic survival assays

Cells were trypsinized, seeded at low density, and allowed to attach. The next day, cells were either mock-treated or exposed to an increasing dose of UV light (1, 2, 3, and 4 J/m2 of UV-C 266nm), an increasing dose of IR (2, 4, 6, and 8 Gy), or increasing concentrations of MMC (2, 4, 6, and 8 ng/ml). After 7–9 d, the cells were washed with 0.9% NaCl and stained with methylene blue. Colonies of >20 cells were scored. Survival experiments were performed in duplicate and repeated at least twice.

Immunoprecipitation for Co-IP

Cells were mock-treated or UV-C irradiated (20 J/m2) and harvested after 1 h. Cell pellets were lysed for 20 min on ice in EBC-150 buffer (50 mM Tris, pH 7.5, 150 mM NaCl, 0.5% NP-40,2 mM MgCl2, and protease inhibitor cocktail [Roche]) supplemented with 500 U/ml Benzonase Nuclease (Novagen). Cell lysates were incubated for 1.5 h at 4°C with GFP-Trap A beads(Chromotek). The beads were then washed six times with EBC-150 buffer (50 mM Tris, pH 7.5, 150 mM NaCl, 0.5% NP-40,1 mM EDTA, and protease inhibitor cocktail) and boiled inLaemmli-SDS sample buffer.

Western blot

Cells were spun down, washed with PBS, and boiled for 10 min in Laemmli buffer (40 mM Tris, pH 6.8, 3.35% SDS, 16.5% glycerol, 0.0005% Bromophenol Blue, and 0.05 M dithiothreitol). Proteins were separated on 4%–12% Criterion XT Bis-Tris gels(#3450124; Bio-Rad) in NuPAGE MOPSrunning buffer (NP0001-02; Thermo Fisher Scientific) and blotted onto polyvinylidene fluoride membranes (IPFL00010; EMD Millipore). The membrane was blocked with blocking buffer (MB-070-003; Rockland) for 1 h at room temperature. The membrane was then probed with antibodies (listed in Table S5) as indicated. An Odyssey CLx system (LI-COR Biosciences) was used for detection.

RNA recovery assay (RRS)

30,000 cells were seeded on 12-mm glass coverslips in 24-well plates in DMEM with 1% FBS. After 24 h, cells were irradiated with UV-C at a dose of 6 J/m2 and incubated in a conditioned medium for different time periods (0 h, 3 h, and 18 h). Following incubation, nascent transcripts were labeled by incubating the cells with 400 µM 5-ethynyl-uridine (CLK-N002-10; Jena Bioscience), which was then visualized with a Click-iT mix consisting of 50 mM Tris buffer, pH 8.0, 60 µM Atto Azide (647N-101; ATTO-TEC), 4 mM CuSO4•5H2O, 10 mM L-ascorbic acid (A0278; Sigma-Aldrich), and 0.1 µg/ml DAPI (D1306; Thermo Fisher Scientific) for 1 h. Cells were washed three times for 5 min with PBS and mounted on microscope slides (Thermo Fisher Scientific)using Aqua Polymount (Brunschwig).

Immunofluorescence

For immunofluorescent staining, 250,000 cells were seeded on 18-mm glass coverslips. The next day, cells were fixed with4% paraformaldehyde, followed by permeabilization with0.5% Triton X-100 for 10 min. Cells were treated with 100 glycines in PBS for 10 min to block unreacted aldehyde groups, rinsed with PBS, and equilibrated in wash buffer (PBS containing 0.5% BSA and 0.05% Tween-20; Sigma-Aldrich) for10 min. Antibody steps and washes were in wash buffer. The primary antibodies were incubated for 2 h at room temperature, followed by secondary antibodies for 1 h and DAPI for5 min. Primary and secondary antibodies are listed in TableS5. Cells were incubated with 0.1 µg/ml DAPI and mounted using Aqua Polymount.

γH2AX staining

200,000 cells were seeded on 18-mm glass coverslips in 24-well plates with DMEM (10% FBS and 1% P/S). The next day, the cells were exposed to IR by the YXlon International x-ray generator (200 KV, 4 mA; dose rate 2 Gy/min). At the indicated time points, cells were fixed with 3% paraformaldehyde in PBS for 10 min and stained for γH2AX for 1 h (see above). Images were quantified using a custom-built macro in ImageJ that enabled automatic and objective analysis of the number of foci per cell, as described previously (Typas et al., 2015).

Microscopic analysis of fixed cells

Images of fixed samples were acquired on a Zeiss AxioImagerM2 or D2 wide-field fluorescence microscope equipped with63× PLAN APO (1.4 numerical aperture [NA]) oil-immersion objectives (Zeiss) and an HXP 120 metal-halide lamp used for excitation. Fluorescent probes were detected using the following filters for DAPI (excitation filter: 350/50 nm; dichroic mirror: 400 nm; emission filter: 460/50 nm), Alexa 555 (excitation filter: 545/25 nm; dichroic mirror: 565 nm; emission filter: 605/70 nm), or Alexa 647 (excitation filter: 640/30 nm; dichroic mirror: 660 nm; emission filter: 690/50 nm). Images were recorded using ZEN 2012 software and analyzed in ImageJ.

UV laser micro-irradiation

Cells were grown on 18-mm quartz (UV-C) or glass (UV-A)coverslips and placed in a Chamlide CMB magnetic chamber in which the growth medium was replaced by CO2-independent Leibovitz L-15 medium (Thermo Fisher Scientific). UV-C laser tracks were made using a diode-pumped solid-state 266-mm yttrium Aluminum Garnet laser (average power 5 mW, repetition rate up to 10 kHz, and pulse length 1 ns). Prior to UV-Amicro-irradiation, cells were either sensitized with 6 µM trioxsalen for 1 h to generate ICLs or with 15 µM BrdU for 24 h to generate DSBs. UV-A laser tracks were made by a diode-pumped solid-state 355-nm Yttrium Aluminum Garnet laser (average power 14 mW and repetition rate up to 200 Hz). Both lasers were integrated into a UGA-42-Caliburn/2L Spot Illumination system(Rapp OptoElectronic).

Micro-irradiation was combined with live-cell imaging in an environmental chamber set to 37°C on an all-quartz wide-field fluorescence Zeiss Axio Observer 7 microscope, using a 100× (1.2NA) ultrafilter glycerol-immersion objective (UV-C) or a Plan-Neofluar 63× (1.25 NA) oil-immersion objective (UV-A). The laser system is coupled to the microscope via a TriggerBox, and a neutral density (ND-1) filter blocks 90% of the laser light. AnHXP 120-V metal-halide lamp was used for excitation. Images were acquired in Zeiss ZEN and quantified in ImageJ.

Chromosome breakage assay

2 × 106 fibroblasts were seeded in 175-cm2 tissue culture flasks with DMEM (10% FBS) and cultured at 37°C in the presence or absence of 50 nM MMC. After 48 h, 600 µl of demecolcine (10 µg/µl) was added to each culture flask, and cells were incubated for an additional 30 min at 37°C to enrich for metaphases. Cells were then trypsinized and resuspended in 75 mM KCL and incubated for 20 min at room temperature. The cells were spun down, resuspended in 10 ml fixative (75% methanol and 25% acetic acid), incubated for 30 min at room temperature, and centrifuged. Pellets were resuspended in 10 ml of fixative and incubated for 5 min at room temperature. This step was repeated, and finally, the pellet was resuspended in 0.5–1.0 ml fixative. The cell suspension was dropped on a slide and allowed to dry. Slides were stained for 5 min in a 3% Giemsa solution, rinsed in tap water, and allowed to dry. Slides were coded, and from each coded culture, 50 metaphases were examined for chromosomal damage. After scoring, the slides were decoded, and the results were analyzed as presented (Stoepker et al., 2011).

UDS

180,000 cells were seeded on 18-mm glass coverslips in 12-well plates in DMEM with 1% FBS. After 24 h, cells were locally irradiated through a 5-µm filter with 30 J/m2 UV-C. Cells were subsequently pulse-labeled with 20 µM EdU (VWR) and 1 µM5-fluoro-deoxyuridine (Sigma-Aldrich) for either 1 h or 4 h. After labeling, cells were medium chased with 10 µM thymidine in DMEM without supplements for 30 min and fixed for 15 min with 3.7% formaldehyde in PBS. Cells were permeabilized for20 min in PBS with 0.5% Triton-X100 and blocked in 3% BSA(Thermo Fisher Scientific) in PBS. The incorporated EdU was coupled to Attoazide Alexa Fluor 647 using Click-iT chemistry according to the manufacturer’s instructions (Invitrogen). After coupling, the cells were postfixed with 2% formaldehyde for10 min and subsequently blocked with 100 mM glycine. DNA was denatured with 0.5 M NaOH for 5 min, followed by blocking with 10% BSA for 15 min. Next, the cells were incubated with an antibody against cyclobutane pyrimidine dimers (see Table S5)for 2 h, followed by secondary antibodies for 1 h, and DAPI for5 min. Cells were mounted in Polymount (Brunschwig).

MS data acquisition

MS was performed essentially as previously described (SalasLloret et al., 2019). All the experiments were performed on an EASY-nLC 1000 system (Proxeon) connected to a Q-Exactive Orbitrap (Thermo Fisher Scientific) through a nano-electrospray ion source. The Q-Exactive was coupled to a 25-cm silica emitter (FS360-75-15-N-5-C25; NewObjective) in-house packed with 1.9 µm C18-AQ beads (Reprospher-DE; Pur; Dr. Manish, Ammerbuch-Entringen, Germany). Samples were run in a 40-min chromatography gradient from 0% to 30% acetonitrile and then increased to 95% acetonitrile before column re-equilibration with a flow rate of 200 NL/min. The mass spectrometer was operated in a data-dependent acquisition mode with a top-seven method and a scan range of 300–1,600 m/z. Full-scan MS spectra were acquired at a target value of 3 × 106 and a resolution of 70,000, and the highercollisional dissociation tandem mass spectra (MS/MS) were recorded at a target value of 105, and with a resolution of 35,000, an isolation window of 2.2 m/z, and normalized collision energy of 25%. The minimum automatic gain control target was 104. The maximum MS1 and MS2 injection times were 250 and 120 ms, respectively. The precursor ion masses of scanned ions were dynamically excluded from MS/MS analysis for 30 s. Ions with charge 1 and higher than 6 were excluded from triggering MS2 analysis.

MS data analysis All raw data were analyzed using MaxQuant (version 1.6.6.0) as described previously (Tyanova et al., 2016a). We performed the search against an in silico–digested UniProt reference proteome for Homo sapiens, including canonical and isoform sequences (May 27, 2019). Database searches were performed according to standard settings with the following modifications. Digestion with Trypsin/P was used, allowing four missed cleavages. Oxidation (M) and acetyl (protein N-term) were allowed as variable modifications with a maximum number of three. Carbamido-methyl (C) was disabled as a fixed modification. Label-free quantification (LFQ) was enabled, not allowing Fast LFQ. Max-Quant output data were further processed using the Perseus computational platform (v 1.6.6.0; Tyanova et al., 2016b). LFQ intensity values were log2 transformed, and potential contaminants and proteins identified by site only or reverse peptides were removed. Samples were grouped in experimental categories, and proteins not identified in four out of four replicates in at least one group were also removed. Missing values were imputed using normally distributed values with a 1.8 downshift (log2) and a randomized 0.3 width (log2) considering whole matrix values. Statistical analysis (t-tests) was performed to determine which proteins were significantly enriched. Volcano plots were generated, and statistical analysis output tables were further processed in Microsoft Excel. The MS proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE (Perez-Riverol et al., 2019) partner repository with the dataset identifier PXD017940.

Purification of recombinant ERCC1-XPF

Baculovirus production was performed as described using the pMacroBac-His-ERCC1/XPF-HA constructs (Enzlin and Sch¨ arer, 2002). ERCC1WT and ERCC1R156W were coexpressed with XPFWT from a single baculovirus in Sf9 insect cells. The heterodimers were purified over nickel affinity, size exclusion, and heparin chromatography as described (Enzlin and Sch¨ arer, 2002). ERCC1-XPF was eluted from the heparin column at around 600 mM NaCl. Protein concentrations ranged from 0.1 to 0.2 mg/ml.

Nuclease assay

A stem-loop substrate (GCCAGCGCTCGG(T)22CCGAGCGCTGGC)containing fluorescent dye Cy5 (50 pmol) at 39 termini (IDT)was annealed in a 100-µl annealing buffer (10 mM Tris, pH 7.5, and 50 mM NaCl) by heating at 95°C for 5 min and slowly cooling down over a period of 2 h. 200 fmol of annealed substrate was used in 20-µl reactions containing 25 mM HEPES, pH8.0, 2 mM MgCl2, 10% glycerol, 0.5 mM β-mercaptoethanol, 0.1mg/ml BSA, 40 mM NaCl, and 0–40 nM of protein. The reactions were incubated at 30°C for 30 min and stopped by adding 10 µl90% formamide/10 mM EDTA. After heating at 95°C for 5 min and cooling on ice, 15 µl of each sample was loaded on a 15%denaturing polyacrylamide gel. Gels were run at 30 mA for30 min and bands were visualized by fluorescence by Typhoon RGB(Amersham Biosciences).

In vitro NER assay

XPF-deficient (XP2YO) cell extracts and the plasmid containing site-specific dG-AAF lesion were generated as previously described (Gillet et al., 2005; Shivji et al., 1999). For each reaction, 2 µl of repair buffer (200 mM Hepes-KOH, 25 mM MgCl2, 2.5 mM DTT, 10 mM ATP, 110 mM phosphocreatine, and 1.8 mg/ ml BSA, final pH 7.8), 0.2 µl of creatine phosphokinase buffer (2.5 mg/ml creatine phosphokinase from rabbit muscle [SigmaAldrich], 10 mM glycine, pH 9.0, and 50% glycerol), 3 µl of XPF-deficient cell extract, NaCl (to a final concentration of 70 mM), and purified ERCC1-XPF proteins in a total volume of 9 µl were prewarmed at 30°C for 10 min. 1 µl plasmid containing dG-AAF (50 ng/µl) was added to each reaction, and the reactions were incubated at 30°C for 45 min. The reaction mixture was then cooled on ice for 5 min, followed by the addition of 0.5 µl of 1 µM complementary strand d(GGGGCATGTGGCGCCGGTAATAGC TACGTAGCTC), and the reaction mixture was denatured by heating at 95°C for 5 min. Following 15 min of annealing at room temperature, 1 µl sequence mix (containing 0.13 U of sequence and 2.0 µCi [α-32P] dCTP for each reaction) was added. After preincubation at 37°C for 3 min, 1.2 µl deoxyribonucleotide triphosphate mixture (50 µM dCTP, 100 µM dTTP, 100 µM dATP, and 100 µM dGTP) was added. The reaction mixture was incubated at 37°C for 12 min, and the reaction was stopped by adding 8 µl loading dye (80% formamide and 10 mM EDTA). Samples were heated at 95°C for 5 min, cooled on ice, and loaded onto a 14% denaturing polyacrylamide gel. Gels were run at45 W for 2.5 h and bands visualized using a PhosphorImager(Typhoon RGB).

Online supplemental material

Fig. S1 shows laboratory values of liver and kidney functions in sibling 1 and sibling 2. Fig. S2 shows genomic sequencing data confirming the missense mutation and the intragenic deletion in the ERCC1 gene. Fig. S3 shows the protein expression of ERCC1 and XPF in patient fibroblasts. Fig. S4 shows purified recombinant ERCC1 and XPF proteins and Co-IP of ERCC1WT and ERCC1R156W. Fig. S5 shows microscopy images of UDS, the γH2AX foci, and the location of patient mutations in ERCC1-XPF. Table S1 contains the sequences of all primers and probes used in this study. Table S2 lists all cell lines tested in this study. Table S3 lists all plasmids involved in this study. Table S4 provides the study’s sgRNA sequences and Table S5 shows the antibodies. Table S6 provides the number of cells used for quantification in the figures. Data S1 displays all the uncropped Western blot files.

Acknowledgments

The authors acknowledge both siblings and their parents for contributing to this paper, Sylvie Noordermeer (Leiden University Medical Center [LUMC], Leiden, Netherlands) for the thepLenti-CW57-TO-GFP lentiviral expression vector and advice on generating KI cells, Joachim Goedhart (University of Amsterdam, Amsterdam, Netherlands) for the mVenus cDNA, Bert van derKooij (LUMC, Leiden, Netherlands) for the DNA-PK inhibitor, Wim Vermeulen and Anja Rams (Erasmus MC, Rotterdam, Netherlands) for providing 165TOR cells, Tomoo Ogi (University of Nagoya, Japan) for providing CS20LO cells, Alan Lehman (University of Sussex, UK) for providing CSL16NG and CSL16NGhTERT cells, and Joost Schimmel (LUMC, Leiden, Netherlands)for advice on generating KI cells.

This work was funded by a Leiden University Medical CenterResearch Fellowship and a Nederlandse Organisatie voor Wetenschappelijk Onderzoek VIDI grant (ALW.016.161.320) to M.S.Luijsterburg, a European Research Council starting grant(310913) to A.C.O. Vertegaal, a KWF Kankerbestrijding YoungInvestigator grant (11367) to R. Gonz´alez-Prieto, a KWF Kankerbestrijding grant (VU 2013-5983) to M.A. Rooimans, andgrants from the Korean Institute for Basic Science (IBS-R022-A1)and the National Cancer Institute (P01CA092584) to O.D.Sch arer.

Author contributions: K. Apelt generated KO cells, KI cells, and stable cell lines; performed Western blot and immunostainings for ERCC1 and XPF expression, lentiviral transductions, clonogenic survivals (UV, MMC, IR), Co-IP experiments for Western blot analysis, UDS experiments, Co-IP experiments for MS, andγH2AX stainings; and wrote the paper. A. Kragten performedUDS, RRS, and immunostainings after local UV. A.P. Wondergem generated KO and KI cells and performed UDS experiments, Western blot, and Sanger sequencing to confirm the missense mutation. S.M. White provided counseling for the patients, analyzed exome data, performed biopsies, generated primary fibroblasts, and wrote the clinical patient description. S. Lunke and B.T. Wilson generated NGS data. S. Pantaleoand D. Flanagan performed qPCR to validate the deletion. C.Quinlan wrote the clinical description of the kidney. W. Hardikarwrote the clinical description of the liver. M.A. Rooimans andR.M.F. Wolthuis generated hTERT-immortalized 48BR andPV50LD fibroblasts and performed the MMC-induced chromosome breakage assays. R. Gonzalez-Prieto and A.C.O. Vertegaal analyzed all MS experiments. W.W. Wiegant and H. vanadium performed clonogenic survivals in U2OS cells (IR). J.-E.Yeo and H.S. Kim purified recombinant proteins and performed the nuclease assay and the in vitro NER assays. O.D. Sch¨arersupervised in vitro NER work contributed to the structural interpretation of the R156W allele and edited the paper. M.S. Luijsterburg supervised the project and wrote the paper.

References

Abdel Ghaffar, T.Y., E.S. Elsobky, and S.M. Elsayed. 2011. Cholestasis in patients with Cockayne syndrome and suggested modified criteria for clinical diagnosis. OrphanetJ. Rare Dis. 6:13. -1172-6-13

Abdullah, U.B., J.F. McGouran, S. Brolin, D. Ptchelkine, A.H. El-Sagheer, T. Brown, and P.J. McHugh. 2017. RPA activates the XPF-ERCC1 endonuclease to initiate the processing of DNA interstrand crosslinks. EMBO J. 36: 2047–2060.

Adair, G.M., R.L. Rolig, D. Moore-Faver, M. Zabelshansky, J.H. Wilson, and R.S. Nairn. 2000. Role of ERCC1 in the removal of long non-homologous tails during targeted homologous recombination. EMBO J. 19:5552–5561.

Ahmad, A., A.R. Robinson, A. Duensing, E. van Drunen, H.B. Beverloo, D.B. Weisberg, P. Hasty, J.H. Hoeijmakers, and L.J. Niedernhofer. 2008. ERCC1-XPF endonuclease facilitates DNA double-strand break repair. Mol. Cell. Biol. 28:5082–5092.

Ahmad, A., J.H. Enzlin, N.R. Bhagwat, N. Wijgers, A. Raams, E. Appledoorn, A.F. Theil, J.H. J Hoeijmakers, W. Vermeulen, N.G. J Jaspers, et al. 2010. Mislocalization of XPF-ERCC1 nuclease contributes to reduced DNA repair in XP-F patients. PLoS Genet. 6:e1000871.

Auerbach, A.D. 2009. Fanconi anemia and its diagnosis. Mutat. Res. 668:4–10.

Ben Chehida, A., N. Ghali, R. Ben Abdelaziz, F. Ben Moussa, and N. Tebib. 2017. Renal Involvement in 2 Siblings With Cockayne Syndrome. Iran. J. Kidney Dis. 11:253–255.

Biggerstaff, M., D.E. Szymkowski, and R.D. Wood. 1993. Co-correction of the ERCC1, ERCC4, and xeroderma pigmentosum group F DNA repair defects in vitro. EMBO J. 12:3685–3692.

-2075.1993.tb06043.xBogliolo, M., B. Schuster, C. Stoepker, B. Derkunt, Y. Su, A. Raams, J.P. Trujillo, J. Minguillón, M.J. Ramı´rez, R. Pujol, et al. 2013. Mutations in ERCC4, encoding the DNA-repair endonuclease XPF, cause Fanconi anemia. Am. J. Hum. Genet. 92:800–806.

Ceccaldi, R., P. Sarangi, and A.D. D’Andrea. 2016. The Fanconi anemia pathway: new players and new functions. Nat. Rev. Mol. Cell Biol. 17: 337–349.

de Laat, W.L., E. Appeldoorn, N.G. Jaspers, and J.H. Hoeijmakers. 1998a. DNA structural elements required for ERCC1-XPF endonuclease activity. J. Biol. Chem. 273:7835–7842.

de Laat, W.L., A.M. Sijbers, H. Odijk, N.G. Jaspers, and J.H. Hoeijmakers. 1998b. Mapping of interaction domains between human repair proteins ERCC1 and XPF. Nucleic Acids Res. 26:4146–4152.

nar/26.18.4146

de Vries, A., C.T. van Oostrom, F.M. Hofhuis, P.M. Dortant, R.J. Berg, F.R. de Gruijl, P.W. Wester, C.F. van Kreijl, P.J. Capel, H. van Steeg, and S.J. Verbeek. 1995. Increased susceptibility to ultraviolet-B and carcinogens of mice lacking the DNA excision repair gene XPA. Nature. 377:169–173.

DiGiovanna, J.J., and K.H. Kraemer. 2012. Shining a light on xeroderma pigmentosum. J. Invest. Dermatol. 132:785–796.

Enzlin, J.H., and O.D. Schrer. 2002. The active site of the DNA repair endonuclease XPF-ERCC1 forms a highly conserved nuclease motif. EMBO J. 21:2045–2053.

Funaki, S., S. Takahashi, H. Murakami, K. Harada, and H. Kitamura. 2006. Cockayne syndrome with recurrent acute tubulointerstitial nephritis. Pathol. Int. 56:678–682.

.02029.x

Gillet, L.C., J. Alzeer, and O.D. Schrer. 2005. Site-specific incorporation of N- (deoxyguanosine-8-yl)-2-acetylaminofluorene (dG-AAF) into oligonucleotides using modified ‘ultra-mild’ DNA synthesis. Nucleic Acids Res. 33:1961–1969.

Gregg, S.Q., A.R. Robinson, and L.J. Niedernhofer. 2011. Physiological consequences of defects in ERCC1-XPF DNA repair endonuclease. DNA Repair (Amst.). 10:781–791.

Guo, C., Y. Nakazawa, L. Woodbine, A. Bjorkman, M. Shimada, H. Fawcett, N. Jia, K. Ohyama, T.S. Li, Y. Nagayama, et al. 2015. XRCC4 deficiency in human subjects causes a marked neurological phenotype but no overt immunodeficiency. J. Allergy Clin. Immunol. 136:1007–1017.

Helfricht, A., P.E. Thijssen, M.B. Rother, R.G. Shah, L. Du, S. Takada, M. Rogier, J. Moritz, H. IJspeert, C. Stoepker, et al. 2020. Loss of ZBTB24 impairs nonhomologous end-joining and class-switch recombination in patients with ICF syndrome. J. Exp. Med. 217:e20191688.

Houtsmuller, A.B., S. Rademakers, A.L. Nigg, D. Hoogstraten, J.H. Hoeij- makers, and W. Vermeulen. 1999. Action of DNA repair endonuclease ERCC1/XPF in living cells. Science. 284:958–961. h

science.284.5416.958

Jaspers, N.G., A. Raams, M.C. Silengo, N. Wijgers, L.J. Niedernhofer, A.R. Robinson, G. Giglia-Mari, D. Hoogstraten, W.J. Kleijer, J.H. Hoeij- makers, and W. Vermeulen. 2007. The first reported patient with human ERCC1 deficiency has Cerebro-oculofacio-skeletal syndrome with a mild defect in nucleotide excision repair and severe developmental failure. Am. J. Hum. Genet. 80:457–466.

Joenje, H., A.B. Oostra, M. Wijker, F.M. di Summa, C.G. van Berkel, M.A. Rooimans, W. Ebell, M. van Weel, J.C. Pronk, M. Buchwald, and F. Arwert. 1997. Evidence for at least eight Fanconi anemia genes. Am. J. Hum. Genet. 61:940–944.

Jones, M., F. Beuron, A. Borg, A. Nans, C.P. Earl, D.C. Briggs, A.P. Snijders, M. Bowles, E.P. Morris, M. Linch, and N.Q. McDonald. 2020. Cryo-EM structures of the XPF-ERCC1 endonuclease reveal how DNA-junction engagement disrupts an auto-inhibited conformation. Nat. Commun. 11: 1120.

Karikkineth, A.C., M. Scheibye-Knudsen, E. Fivenson, D.L. Croteau, and V.A. Bohr. 2017. Cockayne syndrome: Clinical features, model systems, and pathways. Ageing Res. Rev. 33:3–17.002

Kashiyama, K., Y. Nakazawa, D.T. Pilz, C. Guo, M. Shimada, K. Sasaki, H. Fawcett, J.F. Wing, S.O. Lewin, L. Carr, et al. 2013. Malfunction of nuclease ERCC1-XPF results in diverse clinical manifestations and causes Cockayne syndrome, xeroderma pigmentosum, and Fanconi anemia. Am. J. Hum. Genet. 92:807–819.