ribonucleotide excision repair is essential to prevent ... · tumor biology and immunology...

Tumor Biology and Immunology

Ribonucleotide Excision Repair Is Essential toPrevent Squamous Cell Carcinoma of the SkinBj€orn Hiller1,2, Anja Hoppe1, Christa Haase1, Christina Hiller1, Nadja Schubert1,Werner M€uller3, Martin A.M. Reijns4, Andrew P. Jackson4, Thomas A. Kunkel5,J€org Wenzel6, Rayk Behrendt1, and Axel Roers1

Abstract

Because of imperfect discrimination against ribonucleo-side triphosphates by the replicative DNA polymerases,large numbers of ribonucleotides are incorporated into theeukaryotic nuclear genome during S-phase. Ribonucleo-tides, by far the most common DNA lesion in replicatingcells, destabilize the DNA, and an evolutionarily conservedDNA repair machinery, ribonucleotide excision repair(RER), ensures ribonucleotide removal. Whereas completelack of RER is embryonically lethal, partial loss-of-functionmutations in the genes encoding subunits of RNase H2, theenzyme essential for initiation of RER, cause the SLE-relatedtype I interferonopathy Aicardi-Gouti�eres syndrome. Here,we demonstrate that selective inactivation of RER in mouseepidermis results in spontaneous DNA damage and epider-

mal hyperproliferation associated with loss of hair folliclestem cells and hair follicle function. The animals developedkeratinocyte intraepithelial neoplasia and invasive squa-mous cell carcinoma with complete penetrance, despitepotent type I interferon production and skin inflammation.These results suggest that compromises to RER-mediatedgenome maintenance might represent an important tumor-promoting principle in human cancer.

Significance: Selective inactivation of ribonucleotide exci-sion repair by loss of RNaseH2 in themurine epidermis resultsin spontaneous DNA damage, type I interferon response, skininflammation, and development of squamous cell carcinoma.Cancer Res; 78(20); 5917–26. �2018 AACR.

IntroductionGenome integrity is continuously challenged by amultitude of

different hazards that cause a broad spectrum of different lesionsin DNA (1). Spontaneous hydrolysis, ionizing radiation, exoge-nous chemicals, and endogenous metabolites, including reactiveoxygen species, inflict cumulative damage to DNA (1). Fortunate-ly, various DNA repair pathways revert most of this damage,thereby dramatically slowing down age-related deteriorationof DNA integrity and suppressing the development of cancer (2).In addition to continuous, time-dependent harm to DNA, otherforms of damage are introduced already during replication byerrors the replicative polymerases make, despite their remarkable

fidelity. Such replication errors include incorporation of deoxy-ribonucleotides (dNTP) that are not complementary to the tem-plate, which occurs every 104–106 bases, depending on the DNApolymerase. Most mismatched nucleotides are removed by theproofreading activity that some DNA polymerases possess, whereasthe remaining mismatches are detected and corrected by DNAmis-match repair (MMR), which is directly coupled to the replicationmachinery (3–5). Postreplicative MMR is an important tumor-suppressive principle as genetic defects in MMR proteins causecancer predisposition syndromes, including Lynch syndrome (6)and biallelic mismatch repair deficiency (BMMRD) syndrome (7).

Another form of replication error is the incorporation ofnucleotides containing the correct base, but the wrong sugar. Forexample, discrimination between incoming dNTPs and ribonu-cleotides (rNTP) is imperfect for the three DNA polymerases thatreplicate the vast majority of the undamaged eukaryotic nucleargenome (8), resulting in frequent incorporation of rNMPs intogenomicDNA(9, 10). Theprobability for thismistake is increasedby the far greater cellular abundance of rNTPs compared withdNTPs (8, 11) such that on average, more than 106 rNMPs areincorporated during one round of replication of a mammaliangenome (12). Newly incorporated rNMPs destabilize DNA (10)and pose a major threat to genome integrity due to their reactive20OH group. A highly conserved repair pathway, ribonucleotideexcision repair (RER), ensures their removal (10, 13, 14). LikeMMR, RER may be directly coupled to replication and results inrapid postreplicative repair of rNMPs. The first step in RER isdetection of rNMPs embedded in the DNA double helix byribonuclease H2 (RNase H2). RNase H2 then incises therNMP-containing strand immediately 50 of the rNMP. In vitro,this is followed by strand displacement synthesis by pold or pole,flap cleavage by Fen1 or Exo1, and ligation of the nick (14).

1Institute for Immunology, Medical Faculty Carl Gustav Carus, TU Dresden,Dresden, Germany. 2Gene Center and Department of Biochemistry, Ludwig-Maximilians-Universit€at M€unchen, Munich, Germany. 3Faculty of Biology, Med-icine and Health, University of Manchester, Manchester, United Kingdom. 4MRCHuman Genetics Unit, MRC Institute of Genetics and Molecular Medicine, TheUniversity of Edinburgh, Edinburgh, United Kingdom. 5Genome Integrity andStructural Biology Laboratory, National Institute of Environmental HealthSciences (NIEHS), NIH, Research Triangle Park, North Carolina. 6Departmentof Dermatology and Allergy, University Hospital Bonn, Bonn, Germany.

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

B. Hiller, A. Hoppe, R. Behrendt, and A. Roers contributed equally to this article.

Corresponding Authors: Axel Roers, Institute for Immunology, Medical FacultyCarl Gustav Carus, TU Dresden, Fetscherstr. 74, Dresden 01307, Germany.Phone: 4935-1458-6500; Fax: 4935-1458-6316; E-mail:[email protected]; and Rayk Behrendt, [email protected]

doi: 10.1158/0008-5472.CAN-18-1099

�2018 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 5917

on May 2, 2021. © 2018 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst August 28, 2018; DOI: 10.1158/0008-5472.CAN-18-1099

http://crossmark.crossref.org/dialog/?doi=10.1158/0008-5472.CAN-18-1099&domain=pdf&date_stamp=2018-9-29

http://cancerres.aacrjournals.org/

Interestingly, during their brief transient presence inDNA, rNMPsseem to serve important physiologic functions in DNA metabo-lism (8, 13, 15). They may provide a strand discrimination signalin MMR for the nascent leading strand (16, 17), wherein RNaseH2-mediated incisions, occurring only in the newly synthesizedrNMP-containing strand, may serve as entry points for MMRenzymes. Moreover, nicking by RNase H2 could be crucial forrelaxing leading strand torsional stress (18). rNMPsmight also beincorporated during repair synthesis occurring in noncycling cellsthat contain very high rNTP:dNTP ratios.

RNase H2-deficient cells are incapable of executing normalRER, carry high numbers of rNMPs in their DNA and spontane-ouslymount aDNAdamage response (DDR), reflecting theDNA-destabilizing effects of rNMPs (13, 15, 19). rNMPs alter DNAstructure and thereby potentially affect multiple key DNA trans-actions. rNMPs increase the rate of spontaneous hydrolysis andwere shown to trigger a mutagenic repair pathway in yeastdepending on DNA topoisomerase 1, which, like RNase H2, canincise at single rNMPs, potentially generating unligatable ends inrepetitive DNA that result in short deletion mutations (20).Topoisomerase 1 cleavage at rNTPs embedded in the DNA canalso result in double-strand breaks and replication stress ensuingfrom abnormally long-lived topoisomerase 1-DNA complexes(21). Moreover, bypassing rNMPs in DNA may pose a problemfor replicative polymerases, which may result in replication forkstalling, strand breaks, and replicative stress. In addition togenome protection by RNase H2-mediated removal of singlerNMPs, RNase H2 contributes to genome stability also by reso-lution of R-loops (22), preventing damage resulting from colli-sions of RNA polymerases or replication forks with R-loops.

Loss of RNase H2 causes genome instability in yeast with highrates of gross chromosomal rearrangements and copy numbervariations (13). Mammalian cells lacking the enzyme displayincreased numbers of strand breaks, activation of DNA damageresponses (DDR), and chromosomal aberrations (12, 23). Abnor-malmitotic chromosome segregation associatedwith this damagewas shown to occur in RNase H2-deficient cells, leading toformation of micronuclei (24, 25). Similarly, DNA damage canlead to herniation of chromatin into the cytosol (26).Micronucleiand herniated chromatin expose large amounts of chromosomalDNA to the cytosolic DNA sensor cGAS that activates STING andthereby triggers an inflammatory response, including upregula-tion of IFN-stimulated genes (ISG) and inflammatory cytokineproduction (24). Unbalanced type I IFN responses are likelyresponsible for chronic inflammation in patients suffering fromthe rare monogenic autoimmune disorder Aicardi-Gouti�eres syn-drome (AGS), which can be caused by partial loss-of-function ofRNase H2 or by mutations in other enzymes involved in nucleicacid metabolism (27, 28).

In mouse models, the effects of targeted inactivation of RNaseH2 in vivo were found to critically depend on the degree to whichRNase H2 activity was reduced. The RNase H2 complex is com-posed of three proteins, RNASEH2A, B, and C, all of which areessential for RNase H2 activity (29, 30). Biallelic null or hypo-morphic mutations in RNase H2 genes that cause massive reduc-tion of activity resulted in embryonic or perinatal lethality asso-ciated with spontaneous DNA damage and activation of DDRs(12, 23, 31). In contrast, a homozygousmutation reducing RNaseH2 activity to about 30% of wild-type levels triggered a sponta-neous low level IFN response, but was not associated with grosspathology (32).

To elucidate the consequences of severe RER deficiency for thehomeostasis of a tissue with rapid cell turnover in adult animals,we bypassed embryonic lethality of ubiquitous RNase H2 defi-ciency by conditional gene inactivation. We demonstrate thatcomplete loss of RNase H2 in skin epithelium results in sponta-neous DNA damage and epidermal hyperproliferation that pro-gresses to skin cancer with 100% penetrance, despite potentspontaneous type I IFN responses, demonstrating that RNaseH2-dependent RER is an important tumor-suppressive principle.

Materials and MethodsMice

Rnaseh2bFL/FL (23), K14-Cre (33), Ifnar1�/� (generated fromIfnar1FL mice (34) by ubiquitous deletion of the loxP-flankedfragment), and Trp53�/� (34) were either on the C57BL/6NJbackground or backcrossed for at least five generations toC57BL/6NJ. Mice were housed at the Experimental Center,Medical Faculty Carl Gustav Carus, TU Dresden, under specificpathogen-free conditions. All procedures were in accordancewith institutional guidelines on animal welfare and wereapproved by the Landesdirektion Dresden (permit number24-1/2013-12, 88/2017).

Quantification of mRNAs by quantitative RT-PCRTotal RNA was isolated using the NucleoSpin Kit (Macherey-

Nagel) according to the manufacturer's instructions. cDNA wasgenerated using the RevertAid H Minus First Strand cDNA Syn-thesis Kit (Thermo Fisher Scientific). Quantification of transcriptswas performed on a Mx3005P QPCR System (Agilent Technol-ogies) using the Maxima SYBR Green/ROX qPCR Master Mix(Thermo Fisher Scientific). The following oligos were used:Oasl1up, 50-GCAATCCACAGCGATATCC-30; Oasl1 down, 50-CAACT-GCTCACTGTCCACGG-30; Ifi44 up, 50-GAGTCACTCATTCTCG-GACTCCGC-30; Ifi44 down, 50-GAGCGGGCATTGAAGTAAGG-GC-30; Viperin up, 50-TTCGCCCGCATCAAAGCCGT-30; Viperindown, 50-AGGGGGCAGCGGAAGTCGAT-30; Rnaseh2b up, 50-AGGTTTCCAGGGACAAGGAAGAGGA-30; Rnaseh2b down, 50-GTCAATGAAGCTGGAGGTTCTGGAAG-30; Tbp1 up, 50-TGACC-CAGATCATGTTTGAGACCTTCA-30; and Tbp1 down, 50-GGAGT-CCATCACAATGCCTGTGG-30. All samples were run in triplicates.qRT-PCR data are displayed as fold change compared withmeansof the respective control groups � SD.

Assay for RNase H2 activityCell lysates were prepared and assayed for specific cleavage of

an 18-bp double-stranded DNA substrate containing a singleribonucleotide in one strand as described previously (27, 29).RNase H2–specific activity was determined by subtracting thecellular activity against a sequence-matched DNA duplex withoutrNTPs. Cell lysate protein concentration was determined andlysates were added to the reaction mix at a final protein concen-tration of 100 ng/mL.

Isolation and flow cytometric analysis of epidermal cellsBack skinwas excised and adipose tissuewas removed. The skin

was placed on a layer of trypsin solution (0.25%, Life Technol-ogies) with the epidermal side facing the trypsin solution. Afterincubation for 2 hours, the epidermis was separated from thedermis using blunt forceps. The epidermis was minced withscalpels and digested in 10 mL trypsin solution (0.25%, Life

Hiller et al.

Cancer Res; 78(20) October 15, 2018 Cancer Research5918




Technologies) for another hour. The digest was stoppedby addingan equal volume of medium [DMEM (2/3) þ HAM's F12 (1/3)]supplemented with 10% calcium-free FBS and the cells werestrained through a40-mmnylonmesh.Cellswere then centrifugedfor 8 minutes at 500� g at room temperature and the pellet wasresuspended in 2 mL FACS buffer. Epidermal stem cell popula-tions were stained according to Jensen and colleagues (35). Forflow cytometry, cells were stained using the following antibodies:anti–CD45-FITC (1:400, eBioscience), anti–CD49f-PE (1:200,eBioscience), anti–CD34-eF660 (1:50, eBioscience), anti–Sca1-PerCP-Cy5.5 (1:200, eBioscience), and anti–CD117-APC-Cy7(1:800, BioLegend). Cell sorting and analysis was performed onan ARIA III cell sorter (BD Biosciences) and data were recordedusing DIVA software (BD Biosciences) and analyzed using FlowJo(FlowJo).

Transcriptome analysisKeratinocytes (CD49fþ) and hematopoietic cells (CD45þ)

were isolated from epidermal skin cell suspensions by FACS(see above). Total RNA was isolated using the RNeasy MiniKitþ (Qiagen). mRNA libraries were prepared and subjected todeep sequencing on an IlluminaHighSeq. Reads were mapped tothe reference genome mm10 (Ensembl Version 75) using gsnap(v.2014-12-06 for analysis of CD45�CD49fþ cells, v.2014-12-17for analysis of CD45þ cells) and genes were counted with fea-turecount v.1.4.6. Differentially expressed genes (DEG) were iden-tified using DESeq2 v1.8.1 (36). Clustering of DEGs into specificpathways was investigated using Kyoto Encyclopedia of Genesand Genomes (http://www.ge-nome.jp/kegg/pathway.html),Reactome (https://reactome.org/) and Interferome V2.0 (www.interferome.org/) databases as well as the Ingenuity PathwayAnalysis (IPA) software (Qiagen). RNA-seq data reported in thisarticle are accessible as a super series from the Gene ExpressionOmnibus database under accession number GSE115005.

Histologic analysisFormalin-fixed and paraffin-embedded skin sections were

deparaffinized and rehydrated, subjected to antigen retrieval(20 minutes 98�C in sodium citrate pH 6.0) and washed threetimes in TBST (1xTBS, 0.1% Triton X-100).

For quantification of phosphorylated histoneH2A.X (gH2A.X),sections were blocked for 1 hour (10% goat serum in TBST) andincubated at 4�C overnight with a phospho-histone H2A.X(pSer139) antibody [Cell Signaling Technology, 1:50 in 1% goatserum/tris-buffered saline with 0.05% Tween20 (TBST)]. Next,sections were washed (TBST) and incubated with a goat anti–rabbit-AF488 antibody (1:500, Thermo Fisher Scientific) for 4hours at room temperature in the dark. After washing, nuclei werecounterstained with DAPI in themounting solution. Images wererecorded on a Zeiss Axiovert ApoTome II and analyzed using Zensoftware (Zeiss). gH2A.X foci in at least 220 keratinocyte nuclei ofthe interfollicular epidermis per section were counted.

For quantification of Ki67-positive cells, endogenous peroxi-dase activity was blocked (3% H2O2, 20 minutes in the dark),unspecific epitopes were blocked (PBS 10% BSA, 0.5% Triton X-100 for 1 hour at room temperature) and sections were washedand incubated at 4�C overnight with anti-Ki67-Biotin(eBioscience, 1:100 in PBS 1% BSA, 0.05% Triton X-100). Afterwashing, sections were incubated for 1 hour at room temperaturewith horseradish peroxidase–conjugated Streptavidin (Dako,1:300, PBS, 1% BSA, 0.05% Triton X-100). After addition of DAB

mix (Vector Labs) and counterstaining with hematoxylin, Ki67-positive keratinocytes were counted in interfollicular epidermis.

For quantification of apoptotic cells positive for activatedcaspase-3, antigen retrieval was performed in PBS with1 mmol/L EDTA, 0.05% Tween 20, pH 8.0 for 20 minutes in apressure cooker), sections were blocked (PBS, 0.1%Tween 20, 5%normal goat serum), and incubated with rabbit anti–active-caspase-3 AF835 (R&D, 1:600) at 4�C overnight, followed bywashing and incubation for 1 hour at room temperature withgoat anti-rabbit AF488 (ab150077, Abcam, 1:500). Nuclei werestained with DAPI. Images were recorded on a Keyence BZ-X710microscope and positive cells were counted in interfollicularepidermis and follicular infundibula.

Histologic evaluation was performed by a professional der-mato-histopathologist (J. Wenzel, Bonn, Germany) in a blindedfashion.

Statistical analysisUnless stated otherwise, significance was calculated by

unpaired, two-sided Student t test, ��, P < 0.001; ��, P < 0.01;�, P < 0.05.

ResultsComplete loss of RNaseH2 in the epidermis results in epithelialhyperproliferation and loss of stem cells

We inactivated theRnaseh2b gene selectively in the epidermis bycrossing Rnaseh2bFLOX mice (23) to the K14-Cre line that deletesloxP-flanked DNA in all basal cells of skin epithelium includinghair follicles (23, 33). Rnaseh2bFL/FLK14-Cre mice ("Rnaseh2EKO"mice) showed strong reduction of Rnaseh2b transcripts and RNaseH2 activity in total epidermis or FACS-purified epidermal cells, asexpected (Supplementary Fig. S1A–S1C). The animals were con-spicuous already few days after birth, showing significant hyper-pigmentation, most prominently of ears, snout, tails, and paws(Fig. 1A), a known sign of ongoing DNA damage responses inmouse epidermis (37, 38). Althoughwhiskers were hypomorphicand reduced in number already in young mice (Fig. 1A), the furinitially appeared macroscopically normal. However, Rnaseh2EKO

mice began to lose hair at about 12 weeks and were almostcompletely nude by 20 weeks (Fig. 1A).

Histologically,Rnaseh2EKO epidermiswas normal at the age of 5weeks except for discrete hyperkeratosis (thickening of the corni-fied layer). In older animals (10–13 weeks and 18–21 weeks),focal thickening of the epithelium and moderate hyperkeratosissuggested increased epithelial proliferation (Fig. 1B). Immunos-taining for the cell-cycle marker Ki67 (Fig. 1B) revealed increasednumbers of proliferating epithelial cells in Rnaseh2EKO comparedwith control skin in all age groups tested.More rapid proliferationwas paralleled by a strong increase in the frequency of apoptotickeratinocytes as detected by staining for active caspase-3 (Fig. 1Cand D). Loss of hair follicles did not account for the almostcomplete boldness of Rnaseh2EKO mice as substantial numbers offollicles were still present even at 30weeks of age (SupplementaryFig. S1D). In their active (anagen) phase, hair follicles grow toextend well into the dermal adipose tissue paralleled by thicken-ing of the adipose tissue, whereas resting (telogen) follicles areusually confined to the dermal collagen layer. Numbers of anagenfollicles were not different from control numbers at 5 weeks ofage. At 10–13 weeks, we found no follicles extending into theadipose layer in control skin, indicating a synchronous resting

Loss of RNase H2 Causes Skin Cancer

www.aacrjournals.org Cancer Res; 78(20) October 15, 2018 5919



http://www.ge-nome.jp/kegg/pathway.html

https://reactome.org/

www.interferome.org/

www.interferome.org/


phase of the hair cycle in this part of the body. In contrast,all Rnaseh2EKO mice featured numerous "active" follicles atthis timepoint, demonstrating disturbed hair cycle regulation(Fig. 1E). This was also reflected by increased thickness ofRnaseh2EKO skin at 10 to 13 weeks as determined by skin foldmeasurements (Supplementary Fig. S1E). Loss of hair folliclefunction in older Rnaseh2EKO mice suggested exhaustion of epi-thelial regenerative capacity. In accordance with this notion, wefound that the number of hair follicle stem cells, in particularbulge stem cells (CD49fþCD34þSca1lo) was strongly reduced inolder mutants compared with controls, as determined by flowcytometric analysis of epidermal single cell suspensions (Fig. 1F;Supplementary Fig. S1F). In contrast, other undifferentiated

epidermal cell populations were not reduced in mutant skin (Fig.1F; Supplementary Fig. S1F).

Collectively, loss of RNase H2 in the epidermis results in morerapid epithelial cell turnover due to enhanced epithelial cellproliferation and apoptosis, associated with reduction of hairfollicle stem cell numbers and loss of hair follicle function.

Spontaneous type I IFN response and inflammation inRnaseh2EKO skin

Histology and flow cytometry revealed enhanced leukocyteinfiltration of Rnaseh2EKO skin (Fig. 2A and B), which was accen-tuated around hair follicles (Fig. 2A). Inflammatory cell invasioninto the epithelium was associated with degeneration of the

Figure 1.

Epithelial hyperproliferation and stem cell exhaustion in mice lacking RNase H2 in the epidermis (Rnaseh2EKO). A, Macroscopic phenotype of a 15-week-oldRnaseh2EKO mouse and a control (Rnaseh2bFL/FLCre-negative) littermate. B, Representative histology of back skin from a 12-week-old Rnaseh2EKO and a control(Rnaseh2bFL/FLCre-negative) littermate. Left, hematoxylin-eosin (H&E) staining showing focal epithelial thickening and hyperkeratosis (thickening of cornified layer)of mutant epidermis (see Supplementary Material for detailed histopathologic findings); right, Ki67 immunostaining showing increased proliferation of mutantepidermis. Scale bars, 40 mm. Graph shows numbers of Ki67þ interfollicular epidermal cells per high power field (40�); mean � SD; � , P ¼ 0.014 (5 weeks);�� ,P¼0.0039 (10–12weeks); ��,P¼0.0007 (18–21weeks); � ,P¼0.0101 (30weeks).C,Detection of apoptotic cells in interfollicular epidermis by immunostaining foractive caspase-3 in back skin sections of Rnaseh2EKO mice and control (Rnaseh2bFL/FLCre-negative) littermates. Representative images are shown. Scalebar, 50mm.D,Quantification of active caspase-3 permmepidermal length (n¼6–10/group and timepoint);means�SD, �� ,P¼0.0073 (10–13weeks); �� ,P<0.0001(18–21 weeks). E, Quantification of "active" hair follicles (i.e., follicles extending beyond the dermal collagen into the dermal adipose layer, thus including folliclesin anagen II–VI and early catagen stages) per low power (10�) field in hematoxylin and eosin–stained back skin sections of Rnaseh2EKO mice and control(Rnaseh2bFL/FLCre-negative) littermates. Means � SD; � , P ¼ 0.0391 (10–13 weeks); � , P ¼ 0.0335 (18–21 weeks). F, Flow cytometric quantification of epithelialstem cell populations in back skin of Rnaseh2EKO mice and control (Rnaseh2bFL/FLCre-negative) littermates. See Supplementary Fig. S1 for gating. SCs, stemcells; HFSC, hair follicle stem cells; IFE, interfollicular epidermis. Mean and SD; �� ,P¼0.0011 (suprabasal bulge SCs, 20weeks); �,P¼0.0205 (bulge SCs, 10–12weeks);�� , P ¼ 0.0033 (bulge SCs, 20 weeks); � , P ¼ 0.0239 (IFEþ infundibulum SCs, 20 weeks).

Hiller et al.





outermost epithelial cells of the hair follicle (Fig. 2A), an inflam-mation pattern ("interface dermatitis") typical of skin lesions inSLE (39).

Because RNase H2 deficiency was shown to be associated withspontaneous activation of type I IFN responses (31, 32), wequantified transcript levels of type I IFN-inducible genes inRnaseh2EKO and control skin by qRT-PCR analysis. ISG mRNAswere strongly increased in total epidermis and in purifiedkeratinocytes from young and older mutant mice (Fig. 2C; Sup-plementary Fig. S2A). Comparison of transcriptomes of leuko-cytes from control versusRnaseh2EKO skin showedupregulation ofmultiple ISGs in the cells from themutants, indicating exposure totype I IFN (Fig. 2D; Supplementary Table S1).

Type I IFNs were shown to regulate stem cell function in thehematopoietic system (40). We therefore addressed what role thechronic IFN response inRnaseh2EKO epidermis played and crossedRnaseh2EKO mice to a type I IFN receptor knock out (Ifnar1�/�)line (41; Supplementary Fig. S2B). Hyperpigmentation, hairloss (Fig. 2E), hyperkeratosis and focal epidermal thickening

(Supplementary Fig. S2C), and inflammatory leukocyte infiltra-tion (Fig. 2F) of Rnaseh2EKO skin were notmitigated by additionalinactivation of type I IFN signaling. Rnaseh2EKOIfnar1�/� miceshowed a reduction of hair follicle stem cell numbers comparedwith controls (Fig. 2G), similar to IFNAR-competent Rnaseh2EKO

mice (Fig. 1E).Collectively, the skin inflammation of Rnaseh2EKO mice is

associated with type I IFN production; however, leukocyte infil-tration, epidermal hyperproliferation, and loss of hair folliclefunction occur independent of type I IFN.

Loss of RNase H2 in the epidermis results in spontaneous DNAdamage and skin cancer

Spontaneous DNA damage and activation of DNA damageresponses was shown in RNase H2-deficient embryos and cells(12, 23). Moreover, the hyperpigmented skin of Rnaseh2EKOmiceis suggestive of chronic DNA damage responses in the epidermis(37, 38). Immunostaining of skin sections for gH2AX and 53BP1repair foci indeed revealed increased numbers of gH2AX and

Figure 2.

Loss of RNase H2 in the epidermis results in spontaneous type I IFN response and inflammation. A, Representative Rnaseh2EKO back skin section (age 10–12 weeks)immunostained for CD45 (purple) for identification of hematopoietic cells (see Supplementary Materials and Methods for detailed histopathologic findings).Original magnification, �200. B, Quantification of CD45þ leukocytes in single cell suspensions prepared from the epidermal compartment of back skin from10–12 week-old Rnaseh2EKO (EKO) mice (n ¼ 10) and control (Rnaseh2bFL/FLCre-negative) littermates (n ¼ 9). Mean and SD; ��, P < 0.0001. C, Comparisonof ISG transcript levels determined by qRT-PCR in RNA extracted from total epidermis of 10–12 and 20-week-old Rnaseh2EKO (EKO) mice to the meanobtained for control (Rnaseh2bFL/FLCre-negative) littermates (10–12 wk, n ¼ 6; 20 wk, n ¼ 3), which was set to 1 (dotted line).10–12 weeks: �� , P ¼ 0.0018 (Oasl1);� , P ¼ 0.0105 (Ifi44); � , P ¼ 0.0162 (Viperin); 20 weeks: � , P ¼ 0.0242 (Oasl1); �� , P ¼ 0.0061 (Ifi44); � , P ¼ 0.0246 (Viperin). D, Comparison of geneexpression profiles of FACS-purified CD45þ leukocytes from epidermis of 12-week-old Rnaseh2EKO (EKO) mice (n ¼ 7) and control (Rnaseh2bFL/FLCre-negative)littermates (n ¼ 6) based on RNA sequencing. Black, significantly (P < 0.05) deregulated genes; red, significantly deregulated ISGs. E, Macroscopicphenotype of an 11-week-old Ifnar1�/�Rnaseh2EKO mouse (note hyperpigmentation of ears, tail, and paws, and almost complete hair loss) and a control(Ifnar1-/-Rnaseh2bFL/FLCre-negative) littermate. F, Quantification of CD45þ leukocytes in the epidermis of 10–12 week-old Ifnar1�/�Rnaseh2EKO (EKO) mice (n ¼ 7)and control (Rnaseh2bFL/FLCre-negative) littermates (n ¼ 7). Mean and SD; �� , P ¼ 0.0009. G, Flow cytometric quantification of hair follicle stem cellpopulations in back skin of 10–12 week-old Ifnar1�/�Rnaseh2EKO mice (n ¼ 7) and control (Ifnar1�/�Rnaseh2bFL/FLCre-negative) littermates (n ¼ 7). SeeSupplementary Fig. S1F for gating. Mean and SD; � , P ¼ 0.0209; �� , P < 0.0001.






53BP1 foci in Rnaseh2EKO compared with control epidermal cells(Fig. 3A; Supplementary Fig. S3A), consistent with increasedfrequencies of double-strand breaks in RNase H2-deficient yeast(21). In line with this finding, transcriptome analysis of flowcytometrically purified Rnaseh2EKO keratinocytes showed upregu-lation of several p53-inducible genes compared with control cells(Fig. 3B; Supplementary Table S2), indicating ongoing DNAdamage responses.

Starting from week 12, Rnaseh2EKO mice develop chroniculcerations of the skin, most frequently in the dorsal neckarea, affecting most animals by the age of 1 year (Fig. 3C). AllRnaseh2EKO mice had to be sacrificed between 23 and 55weeks of age because of these ulcerations, and/or because oflarge tumors in various locations, often in the mandibularregion close to the ear (Fig. 3C). Histology revealed (Fig. 3D;Supplementary Table S3) that all ulcerations occurred inneoplastic skin that was classified keratinocyte intraepithelialneoplasia (KIN; ref. 42), that is, malignant epidermal growththat has not yet breached the epithelial basement membrane,

also called "carcinoma in situ". In three of these cases, thetumor had focally broken through the basement membraneand was thus classified invasive squamous cell carcinoma(SCC). All macroscopic tumors histologically proved to beinvasive SCC, except for two, which were KIN producing largemasses of cornified material (Fig. 3D; Supplementary TableS3). Figure 3E shows that 100% of Rnaseh2EKO mice devel-oped cancer of at least the KIN stage by about 50 weeks of age.By this time, 60% of the animals had already progressed toinvasive SCC (Fig. 3E). Additional inactivation of type I IFNsignaling resulted in a slight (albeit insignificant) accelerationof neoplastic skin ulceration (Supplementary Fig. S3B), mostlikely reflecting loss of stimulation of antitumor immunity bytype I IFN.

Collectively, we show that RNase H2-deficient hyperprolifera-tive epidermis features spontaneous DNA damage as demonstrat-ed by increased numbers of repair foci and increased transcriptlevels of p53-inducible genes, and progresses to cancer in 100%ofthe animals within the first year of life.

Figure 3.

Loss ofRNaseH2 in the epidermis results in spontaneousDNAdamage and skin cancer.A, Immunostaining of gH2AX repair foci (green) in back skin of 10–13week-oldRnaseh2EKO (EKO)mice (n¼ 7) and control (Rnaseh2bFL/FLCre-negative) littermates (n¼ 6). Left, representative result for EKO and control (DAPI counterstaining).Scale bars, 5 mm. Right, quantification of gH2AX foci per nucleus. At least 220 nuclei (interfollicular epidermis) were counted per animal. Mean and SD.�� , P < 0.0001 (one-way ANOVA). B, Comparison of gene expression profiles of FACS-purified keratinocytes from epidermis of 9–12 week-old Rnaseh2EKO (EKO)mice (n ¼ 3) and control (Rnaseh2bFL/FLCre-negative) littermates (n ¼ 3) based on RNA sequencing. Black, significantly (P < 0.05) deregulated genes;red, significantly deregulated p53-inducible genes. C, Incidence of macroscopic lesions (ulcerations and tumors) in Rnaseh2EKO mice. Graph indicates thetimepoints at which euthanasia was required for these lesions. D, Representative histology of the lesions represented in C, Top, hematoxylin and eosin–stainedsection of Rnaseh2EKO skin adjacent to ulceration showing characteristic features of KIN, that is, cancer that has not yet broken through the basementmembrane. Scale bars, 200 mm, left; 100 mm, right. Ki67 immunostaining (bottom right) demonstrates proliferating Ki67þ cells in all strata of the epitheliumas is characteristic of KIN. Scale bar, 40 mm. Macroscopic tumor in the genital area of a 36-week-old Rnaseh2EKO mouse histologically (bottom left) proved tobe invasive SCC (Supplementary Table S3, animal 62166). Scale bar, 200 mm. See Supplementary Materials and Methods for detailed histopathologic findings.E, Incidence of invasive SCC and incidence of cancer of all stages (KIN and/or SCC) in Rnaseh2EKO mice. � , P ¼ 0.0131 (log-rank test).

Hiller et al.





Additional loss of p53 in Rnaseh2EKO mice enhances theepidermal IFN response and accelerates hyperproliferation andcarcinogenesis

To determine the effect of p53-dependent DNA damage re-sponses on the Rnaseh2EKO phenotype, we crossed Rnaseh2EKO

to Trp53�/� mice (34) and observed potent effects of Trp53gene dose. Rnaseh2EKO mice heterozygous for the Trp53null

allele showed ameliorated hyperpigmentation comparedwith p53-competent controls (Supplementary Fig. S4A) andno sign of alopecia until the age of 40 weeks. All Rnaseh2EKO

Trp53þ/� mice that were allowed to age had to be sacrificedbecause of macroscopic ulcerations or skin tumors before theage of 40 weeks. Histology showed that these lesions were

invasive SCC of various stages and grades (SupplementaryTable S4; Fig. 4A).

Biallelic loss of Trp53 completely reverted the hyperpigmenta-tion of Rnaseh2EKO (Fig. 4B; Supplementary Fig. S4A). Althoughno generalized alopecia was observed, all of these animals devel-oped inflamed lesions associated with circumscribed loss of hairand extensive scratching in the neck area, starting at about 12weeks of age (Fig. 4B). Massive pruritus and erosions of the neckskin invariably required euthanasia few weeks later. In all cases,histology identified KIN of different stages as the cause of thelesions (Fig. 4C; Supplementary Tables S5 and S6). Ki67 immu-nostaining showed themassive proliferation of cells in all strata ofthe epithelium typical of KIN (Fig. 4D). Inactivation of p53

Figure 4.

Additional loss of p53 in Rnaseh2EKO mice enhances the epidermal IFN response and accelerates hyperproliferation and cancerogenesis. A, Incidence ofinvasive SCC (left) and of cancer of all stages (KIN and/or SCC; right) in Rnaseh2EKO mice that were Trp53WT/- or Trp53�/�. Cancer incidence of Trp53WT/WT

Rnaseh2EKO mice from Fig. 3E is shown in light gray for comparison. �� , P < 0.0001, both (log-rank test). B, Macroscopic phenotype of a 10-week-oldTrp53�/�Rnaseh2EKOmouse and an 11-week-old control (Trp53�/�Rnaseh2bFL/FLCre-negative). Note normal pigmentation (tail and paws) of themutant, in contrastto Trp53wtwtRnaseh2EKO mice (Fig. 1A). Except for an inflamed, hyperkeratotic lesion in the neck with circumscribed hair loss, the fur is largely normal. C,Representative hematoxylin and eosin–stained neck skin section of a Trp53�/�Rnaseh2EKO mouse sacrificed for severe itching and erosions. Note typicalfeatures of KIN, including massive thickening of epidermis and severe dysplasia of the epithelium (irregular nuclear shape; numerous mitotic figures also inupper strata). See Supplementary Materials and Methods for detailed histopathologic findings. Scale bar, left, 50 mm; inset, 20 mm. D, Immunostaining of neckskin sections of 10–12 week-old Trp53�/�Rnaseh2EKO mice (n ¼ 4) and control (Trp53�/�Rnaseh2bFL/FLCre-negative) littermates (n ¼ 8) for Ki67. Representativeimage of Trp53�/�Rnaseh2EKO skin. Scale bar, 50 mm. Graph shows numbers of Ki67þ epidermal cells per randomly placed high power (40�) field. Mean and SD.�� , P < 0.0001. E, Quantification of gH2AX repair foci in immunostained back skin of 10–13 week-old p53�/�Rnaseh2EKO (EKO) mice (n ¼ 4) and control(Trp53�/�Rnaseh2bFL/FLCre-negative) littermates (n ¼ 4). Number of foci was counted in at least 220 nuclei (interfollicular epidermis and follicles) per animal.See Supplementary Fig. S4B for images. Mean and SD. �� , P < 0.0001 (one-way ANOVA). F, Flow cytometric quantification of epithelial stem cell populationsin back skin of 10–12 week-old Trp53�/�Rnaseh2EKO mice (n ¼ 6) and control (Trp53�/�Rnaseh2bFL/FLCre-negative) littermates (n ¼ 5). See SupplementaryFig. S1F for gating. Mean and SD. �� , P ¼ 0.0001. G, Comparison of ISG transcript levels determined by qRT-PCR in RNA extracted from total epidermis (left)or FACS-purified keratinocytes (right) of 10–12 week-old Trp53þ/þRnaseh2EKO and Trp53�/�Rnaseh2EKO mice with the mean obtained for the respectivecontrol group, which was set to 1. Left, �� , P ¼ 0.0005 (Oasl1), �� , P < 0.0001 (Ifi44), �� , P ¼ 0.0066 (Viperin); right, �� , P ¼ 0.0042 (Oasl1; p53 deficiencyalone did not cause a spontaneous IFN response; see Supplementary Fig. S4D.)






signaling resulted in survival of more cells with a higherdamage load as demonstrated by immunostaining for repairfoci (Fig. 4E; Supplementary Fig. S4B) as compared with p53-competent Rnaseh2EKO skin (Fig. 3A). Flow cytometric analysis ofRnaseh2EKOTrp53�/� skin revealed that the loss of hair folliclestem cells caused by epidermal RNase H2 deficiency was pre-vented by lack of p53. Interestingly, hair follicle stem cell numbersof Rnaseh2EKOTrp53�/� epidermis ranged about 3-fold highercompared with control numbers (Fig. 4F).

Compared with p53-competent Rnaseh2EKO mice, leukocyteinfiltration was more pronounced in Rnaseh2EKOTrp53�/� skin(Supplementary Fig. S4C). A potential reason for enhanced skininflammation of these animals might be potentiated cytokineresponses aswe found strongly increased ISGmRNA levels in totalepidermis and in keratinocytes ofTrp53�/� as comparedwithp53-proficient Rnaseh2EKO mice (Fig. 4G), demonstrating that p53-dependent DNA damage responses control the intensity ofthe STING-induced IFN response mounted by RNase H2-defi-cient keratinocytes. Although we have shown that type I IFNwas not causing the inflammation of Rnaseh2EKO skin (Fig. 2F),STING-mediated, NFkB-dependent proinflammatory cytokineexpression, triggered in parallel to the IFN response, could beresponsible for the enhanced inflammation of Ifnar1-deficientRnaseh2EKO skin.

In summary, cancer development in Rnaseh2EKO mice is accel-erated by additional absence of p53. This finding demonstratesthat p53-dependent elimination of epithelial cells with highdamage load potently antagonizes oncogenic transformation ofRNase H2-deficient epidermis.

DiscussionDNA polymerases do not replicate genomic DNA without

mistakes. They incorporate nucleotides carrying the wrong baseor wrong sugar at substantial rates. To ensure genome integrity,two DNA repair pathways operate on the newly synthesizedstrand to correct these replication errors. MMR, which removesnucleotides that do not base-pair correctly with the templatestrand, and RER, which removes rNTPs, that are incorporatedinto the genomic DNA at a rate of 1 every 7,600 nucleotides (12)and threaten genome stability. Humans and mice with geneticdefects causing loss ofMMRactivity are viable but exhibit stronglyincreased cancer risk (6, 7, 43). Genetic defects of RER leading toinability to remove rNTPs fromgenomicDNA result in embryoniclethality in the mouse (12, 23) and most likely also in humans,as partial loss-of-function mutations of the genes encodingRNase H2, the enzyme essential for initiation of RER, can resultin severe disease and biallelic complete null alleles have not beenfound (44).

We bypassed embryonic lethality of global RER deficiency byconditional inactivation of the Rnaseh2b gene only in the epider-mis of the skin and observed spontaneous DNA damage andepithelial hyperproliferation, resulting in spontaneous cancerdevelopment in 100% of the animals within the first year of life.Thus, we showed that RER is essential to prevent malignanttransformation.

Although the epithelium proliferated rapidly, it featured a highrate of keratinocyte apoptosis induced by p53 activation. Cancerdevelopment driven by epidermal RER deficiency was acceleratedupon additional loss of one allele of theTrp53 gene. InRnaseh2EKO

mice harboring two Trp53 null alleles, the skin of neck and back

transformed intoone large confluent carcinoma in situbetween10and 20 weeks of age. It seems likely that this process would haveaffected the entire skin and would have rapidly progressed toinvasive SCC with increasing age, if not for the fact that theanimals had to be euthanized for severe itch and erosions.

Carcinogenesis triggered by defective RNase H2 seems torequire reduction of RNase H2 activity to low levels. Mice inwhich RNase H2 activity was reduced to 30% of control levelsbecause of a homozygous Rnaseh2b partial loss-of-functionmuta-tion showed spontaneous activation of type I IFN responses butno sign of cancer by the age of 1 year (32). Likewise, mice thatcarried a biallelic knock in of a Rnaseh2a partial loss-of-functionmutation, reducing RNase H2 activity to few percent of normallevels, and thatwere rescued fromperinatal lethality by additionalinactivation of STING were not reported to develop neoplasticdisease (31). However, investigation of cancer incidence in alarger cohort of older animals was hampered by the fact thatonly a small fraction of these mice survived into adulthood (31).

We observed accelerated proliferation of Rnaseh2EKO epider-mis resulting in epidermal hyperplasia and thickening of thecornified layer (hyperkeratosis) despite greatly increased ratesof keratinocyte apoptosis. An important inducer of apoptosisin Rnaseh2EKO skin was p53, because additional loss of p53resulted in significant further acceleration of epithelial hyper-proliferation. These rapid cell divisions in response to kerati-nocyte genome damage ensuing from RER deficiency likelyreflect a response program that the epidermis executes uponexposure to multiple forms of DNA damage, aiming atimproved light protection by increasing thickness of epidermisand cornified layer, associated with enhanced melanin produc-tion, as we also observed in our animals. This response isactivated, for example, upon overexposure to UV irradiation(45). Enhanced epidermal proliferation upon genome damagealso results in elimination of damaged cells through rapidepithelial turnover, which represents an alternative keratino-cyte disposal pathway in addition to apoptosis (46). Damagedbasal cells arrested in late S-phase, detach from the basementmembrane and get eliminated into the cornified layer (46).

Accelerated epithelial turnover in Rnaseh2EKO skin was associ-ated with disturbed hair cycle regulation with progressive loss ofhair follicle stem cells and hair follicle function. This might be aresult of exhaustion of this stem cell population due to increaseddemand for differentiated keratinocytes, but may also reflect lossof hair follicle stem cells through apoptosis or terminal differen-tiation and transepidermal elimination. The latter was describedto be the major cause for physiologic hair follicle aging and forpremature hair follicle deterioration enforced by ionizing irradi-ation or genetic defects of nucleotide excision repair (47). Otherepidermal stem cell populations were not reduced in numbers,likely reflecting differential responses to genome damage causedby defective RER. Specific responses of different stem cell popula-tions to genome damage are documented in various tissues (48).Hair follicle stem cells were found to be more resistant to celldeath induction by ionizing irradiation compared with otherepidermal stem cell populations because of their capability torapidly repair double-strand breaks by nonhomologous endjoining and high level expression of the antiapoptotic proteinBcl2 (49). Epidermis-specific inactivation of homologous recom-bination repair by conditional knock out of BRCA1, however,resulted in selective depletion of hair follicle stem cells (50),similar to the loss of hair follicle stem cells in our Rnaseh2EKO

Hiller et al.





animals. In both mouse lines, loss of hair and hair follicle stemcells are largely rescued by additional inactivation of the Trp53gene (50), indicating p53-dependent cell death induction by RERdeficiency in these cells. Additional absence of p53 in Rnaseh2EKO

mice not only restored stem cell numbers to control levels, butresulted in three-fold higher hair follicle stem cell numberscompared with RER competent controls, likely reflecting preneo-plastic expansion of this population.

Rapid progression to cancer in 100% of the animals withepidermis-specific loss of RER activity, even despite intact p53signaling, raises the question of whether mutations impair-ing RER are relevant for human cancer. However, the studiesof cancer incidence in patients with AGS carrying defects ofRNASEH2 genes are hampered by their reduced life span (51).

Germline mutations compromising RER as well as somaticdefects of RER acquired by growing tumor clones could affecttumor biology in two opposite ways. On one hand, DNA damageensuing from unrepaired rNTPs causes genomic instability, thuspromoting carcinogenesis and tumor progression. On the otherhand, high rates of tumor cell death or senescence induction bypersistent DNA lesions will antagonizemalignant growth and canconfer a competitive disadvantage to tumor cells. The net effect,tumor promotion or suppression, likely depends on the degree towhich RER activity is reduced. A window of low RER activityadvantageous for tumor growth may exist. The net effect of RERdeficiency on cancer biology likely also depends on tumor type, asdifferent cell types and cells of the same type but differentdifferentiation stage show specific responses to DNA damageranging from apoptosis and senescence to enforced differentia-tion (48, 52). In addition to detection of damage by nuclearsensors likely occurring in RNase H2-deficient cells, micronucleusformation associated with RER deficiency and subsequent micro-nuclear envelope rupturewere shown to lead to frequent exposureof chromosomalDNA to the cytosolicDNAsensor cGAS, resultingin the activation of STING (24). STING is also activated uponDNA damage by herniation of chromatin into the cytosol (26).The potent type I IFN response we observed in Rnaseh2EKO skinmost likely reflects activation of the cGAS-STING axis. Potentia-tion of this IFN response in the absence of functional p53 stronglysuggests that genome damage is responsible for STING activationin our animals, as lack of p53 allows cells with high damage loadproducing IFN to survive longer. STING can trigger differentialresponses, including senescence or apoptosis as well as antiviraland inflammatory cytokine production stimulating immune acti-vation, depending on signal strength and cell type. RER-deficientepidermis invariably develops cancer despite robust STING acti-vation and production of type IFN, which can strongly boostantitumor immunity (53). Intriguingly, STING-mediated innateimmune responses can also promote neoplastic growth, as STINGactivation by micronuclear DNA in chromosomally unstable

cancer cells was recently shown to drive metastasis by activationof noncanonical NF-kB signaling (54).

Assessing contributions of RER defects to human cancerrequires studies correlating frequencies of partial loss-of-functionmutations in RNASEH2 genes, either in the germline or the tumortissue of the patients, in particular cancer entities. Increasedfrequencies of RNASEH2B partial loss-of-function mutationswere found in the germline of patients with sporadic prostatecancer compared with control cohorts (55). RNASEH2B partialloss-of-function mutations were also detected at higher frequen-cies in nontumor DNA of patients with glioma with familialcancer predisposition and cosegregated with manifestation ofcancer in some of the families (55). These findings provide firsthints that compromised RER activity can be associated withhuman cancer.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: W. M€uller, T.A. Kunkel, R. Behrendt, A. RoersDevelopment of methodology: B. Hiller, A. Hoppe, C. Haase, C. HillerAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): B. Hiller, A. Hoppe, C. Haase, C. Hiller,M.A.M. Reijns, A.P. Jackson, J. Wenzel, R. BehrendtAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): B. Hiller, A. Hoppe, C. Haase, M.A.M. Reijns,T.A. Kunkel, J. Wenzel, R. BehrendtWriting, review, and/or revision of the manuscript: N. Schubert, W. M€uller,M.A.M. Reijns, T.A. Kunkel, J. Wenzel, R. Behrendt, A. RoersAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases):Study supervision: R. Behrendt, A. RoersOthers: Assisted with enzymatic assays (A.P. Jackson)

Discussion of the experiments at planning stage and discussions of theresults (W. M€uller)

AcknowledgmentsWe thank Tobias H€aring for expert technical assistance.The study was

supported by the Else Kr€oner-Fresenius Foundation (grant no. 2015_A100 toA. Roers and R. Behrendt), German Research Foundation (DFG; grant no.Ro2133/6-1 in the setting of Clinical Research Unit KFO249 to A. Roers), andCRTD seed grant (CRTD-FZ 111 to A. Roers). R. Behrendt is supported by theAicardi-Gouti�eres Syndrome Americas Association. Work in the laboratoryof A.P. Jackson was supported by the Medical Research Council (MRC,U127580972).

The costs of publication of this articlewere defrayed inpart by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received April 13, 2018; revised July 10, 2018; accepted August 22, 2018;published first August 28, 2018.

References1. Hoeijmakers JH. DNA damage, aging, and cancer. N Engl J Med 2009;361:

1475–85.2. Lindahl T, Wood RD. Quality control by DNA repair. Science 1999;286:

1897–905.3. Ganai RA, Johansson E. DNA Replication-A Matter of Fidelity. Mol Cell

2016;62:745–55.4. Kunkel TA. Evolving views of DNA replication (in)fidelity. Cold Spring

Harb Symp Quant Biol 2009;74:91–101.5. Kunkel TA, Erie DA. Eukaryotic mismatch repair in relation to DNA

replication. Annu Rev Genet 2015;49:291–313.

6. Lynch HT, Snyder CL, Shaw TG, Heinen CD, Hitchins MP. Milestones ofLynch syndrome: 1895–2015. Nat Rev Cancer 2015;15:181–94.

7. Wimmer K, Etzler J. Constitutional mismatch repair-deficiencysyndrome: have we so far seen only the tip of an iceberg? HumGenet2008;124:105–22.

8. Nick McElhinny SA, Watts BE, Kumar D, Watt DL, Lundstr€om EB, BurgersPM, et al. Abundant ribonucleotide incorporation into DNA by yeastreplicative polymerases. Proc Natl Acad Sci U S A 2010;107:4949–54.

9. Joyce CM. Choosing the right sugar: how polymerases select a nucleotidesubstrate. Proc Natl Acad Sci U S A 1997;94:1619–22.






10. NickMcElhinny SA, KumarD, Clark AB,Watt DL,Watts BE, Lundstr€omEB,et al. Genome instability due to ribonucleotide incorporation into DNA.Nat Chem Biol 2010;6:774–81.

11. Traut TW. Physiological concentrations of purines and pyrimidines. MolCell Biochem 1994;140:1–22.

12. Reijns MA, Rabe B, Rigby RE, Mill P, Astell KR, Lettice LA, et al. Enzymaticremoval of ribonucleotides fromDNA is essential for mammalian genomeintegrity and development. Cell 2012;149:1008–22.

13. Williams JS, Lujan SA, Kunkel TA. Processing ribonucleotides incorporatedduring eukaryoticDNA replication.Nat RevMolCell Biol 2016;17:350–63.

14. Sparks JL, Chon H, Cerritelli SM, Kunkel TA, Johansson E, Crouch RJ, et al.RNaseH2-initiated ribonucleotide excision repair.MolCell 2012;47:980–6.

15. Cerritelli SM, Crouch RJ. The balancing act of ribonucleotides in DNA.Trends Biochem Sci 2016;41:434–45.

16. Lujan SA, Williams JS, Clausen AR, Clark AB, Kunkel TA. Ribonucleotidesare signals for mismatch repair of leading-strand replication errors.Mol Cell 2013;50:437–43.

17. Ghodgaonkar MM, Lazzaro F, Olivera-Pimentel M, Artola-Bor�an M, CejkaP, Reijns MA, et al. Ribonucleotides misincorporated into DNA act asstrand-discrimination signals in eukaryotic mismatch repair. Mol Cell2013;50:323–32.

18. Williams JS, Clausen AR, Lujan SA, Marjavaara L, Clark AB, Burgers PM,et al. Evidence that processing of ribonucleotides inDNAby topoisomerase1 is leading-strand specific. Nat Struct Mol Biol 2015;22:291–7.

19. Caldecott KW. Molecular biology. Ribose–an internal threat to DNA.Science 2014;343:260–1.

20. Kim N, Huang SN, Williams JS, Li YC, Clark AB, Cho JE, et al. Mutagenicprocessing of ribonucleotides in DNA by yeast topoisomerase I. Science2011;332:1561–4.

21. Huang SN,Williams JS, AranaME, Kunkel TA, Pommier Y. TopoisomeraseI-mediated cleavage at unrepaired ribonucleotides generates DNA double-strand breaks. EMBO J 2017;36:361–73.

22. Chon H, Sparks JL, Rychlik M, Nowotny M, Burgers PM, Crouch RJ, et al.RNase H2 roles in genome integrity revealed by unlinking its activities.Nucleic Acids Res 2013;41:3130–43.

23. Hiller B, Achleitner M, Glage S, Naumann R, Behrendt R, Roers A. Mam-malianRNaseH2 removes ribonucleotides fromDNA tomaintain genomeintegrity. J Exp Med 2012;209:1419–26.

24. Mackenzie KJ, Carroll P,MartinCA,MurinaO, Fluteau A, SimpsonDJ, et al.cGAS surveillance ofmicronuclei links genome instability to innate immu-nity. Nature 2017;548:461–5.

25. Bartsch K, Knittler K, Borowski C, Rudnik S, Damme M, Aden K, et al.Absence of RNase H2 triggers generation of immunogenic micronucleiremoved by autophagy. Hum Mol Genet 2017;26:3960–72.

26. Gluck S, Guey B, GulenMF,Wolter K, Kang TW, SchmackeNA, et al. Innateimmune sensing of cytosolic chromatin fragments through cGASpromotessenescence. Nat Cell Biol 2017;19:1061–70.

27. Crow YJ, Leitch A, Hayward BE, Garner A, Parmar R, Griffith E, et al.Mutations in genes encoding ribonuclease H2 subunits cause Aicardi-Goutieres syndrome andmimic congenital viral brain infection. Nat Genet2006;38:910–6.

28. Crow YJ, Manel N. Aicardi-Goutieres syndrome and the type I interfer-onopathies. Nat Rev Immunol 2015;15:429–40.

29. Reijns MA, Bubeck D, Gibson LC, Graham SC, Baillie GS, Jones EY, et al.The structure of the human RNase H2 complex defines key interactioninterfaces relevant to enzyme function and human disease. J Biol Chem2011;286:10530–9.

30. Chon H, Vassilev A, DePamphilis ML, Zhao Y, Zhang J, Burgers PM, et al.Contributions of the two accessory subunits, RNASEH2B and RNASEH2C,to the activity and properties of the human RNase H2 complex. NucleicAcids Res 2009;37:96–110.

31. Pokatayev V, Hasin N, Chon H, Cerritelli SM, Sakhuja K, Ward JM, et al.RNase H2 catalytic core Aicardi-Goutieres syndrome-related mutantinvokes cGAS-STING innate immune-sensing pathway in mice. J Exp Med2016;213:329–36.

32. Mackenzie KJ, Carroll P, Lettice L, Tarnauskait _e �Z, Reddy K, Dix F, et al.Ribonuclease H2 mutations induce a cGAS/STING-dependent innateimmune response. EMBO J 2016;35:831–44.

33. Hafner M, Wenk J, Nenci A, Pasparakis M, Scharffetter-Kochanek K, SmythN, et al. Keratin 14Cre transgenicmice authenticate keratin 14 as anoocyte-expressed protein. Genesis 2004;38:176–81.

34. Jacks T, Remington L, Williams BO, Schmitt EM, Halachmi S, BronsonRT, et al. Tumor spectrum analysis in p53-mutant mice. Curr Biol 1994;4:1–7.

35. Jensen KB, Driskell RR,Watt FM. Assaying proliferation and differentiationcapacity of stem cells using disaggregated adult mouse epidermis.Nat Protoc 2010;5:898–911.

36. Love MI, Huber W, Anders S. Moderated estimation of fold change anddispersion for RNA-seq data with DESeq2. Genome Biol 2014;15:550.

37. Chang CH, Kuo CJ, Ito T, Su YY, Jiang ST, Chiu MH, et al. CK1alphaablation in keratinocytes induces p53-dependent, sunburn-protective skinhyperpigmentation. Proc Natl Acad Sci U S A 2017;114:E8035–44.

38. Cui R,WidlundHR, Feige E, Lin JY,WilenskyDL, Igras VE, et al. Central roleof p53 in the suntan response and pathologic hyperpigmentation. Cell2007;128:853–64.

39. Wenzel J, Tuting T. An IFN-associated cytotoxic cellular immune responseagainst viral, self-, or tumor antigens is a common pathogenetic feature in"interface dermatitis". J Invest Dermatol 2008;128:2392–402.

40. Essers MA, Offner S, Blanco-Bose WE, Waibler Z, Kalinke U, Duchosal MA,et al. IFNalpha activates dormant haematopoietic stem cells in vivo. Nature2009;458:904–8.

41. Kamphuis E, Junt T, Waibler Z, Forster R, Kalinke U. Type I interferonsdirectly regulate lymphocyte recirculation and cause transient blood lym-phopenia. Blood 2006;108:3253–61.

42. Fernandez FiguerasMT. Fromactinic keratosis to squamous cell carcinoma:pathophysiology revisited. J Eur Acad Dermatol Venereol 2017;31Suppl2:5–7.

43. Lee K, Tosti E, Edelmann W. Mouse models of DNA mismatch repair incancer research. DNA Repair (Amst) 2016;38:140–6.

44. Rice G, Patrick T, Parmar R, Taylor CF, Aeby A, Aicardi J, et al. Clinical andmolecular phenotype of Aicardi-Goutieres syndrome. Am J Hum Genet2007;81:713–25.

45. Sterenborg HJ, de Gruijl FR, van der Leun JC. UV-induced epidermalhyperplasia in hairless mice. Photodermatol 1986;3:206–14.

46. Stout GJ, Westdijk D, Calkhoven DM, Pijper O, Backendorf CM, WillemzeR, et al. Epidermal transit of replication-arrested, undifferentiated kerati-nocytes in UV-exposed XPC mice: an alternative to in situ apoptosis. ProcNatl Acad Sci U S A 2005;102:18980–5.

47. Matsumura H,Mohri Y, BinhNT, Morinaga H, FukudaM, ItoM, et al. Hairfollicle aging is driven by transepidermal elimination of stem cells viaCOL17A1 proteolysis. Science 2016;351:aad4395.

48. Blanpain C, Mohrin M, Sotiropoulou PA, Passegu�e E. DNA-damageresponse in tissue-specific and cancer stem cells. Cell Stem Cell 2011;8:16–29.

49. Sotiropoulou PA, Candi A, Mascr�e G, De Clercq S, Youssef KK, Lapouge G,et al. Bcl-2 and accelerated DNA repair mediates resistance of hair folliclebulge stem cells to DNA-damage-induced cell death. Nat Cell Biol2010;12:572–82.

50. Sotiropoulou PA, Karambelas AE, Debaugnies M, Candi A, Bouwman P,Moers V, et al. BRCA1 deficiency in skin epidermis leads to selective lossof hair follicle stem cells and their progeny. Genes Dev 2013;27:39–51.

51. Crow YJ, Chase DS, Lowenstein Schmidt J, Szynkiewicz M, Forte GM,Gornall HL, et al. Characterization of human disease phenotypes associ-ated with mutations in TREX1, RNASEH2A, RNASEH2B, RNASEH2C,SAMHD1, ADAR, and IFIH1. Am J Med Genet A 2015;167A:296–312.

52. Lans H, VermeulenW. Tissue specific response to DNA damage: C. elegansas role model. DNA Repair (Amst) 2015;32:141–8.

53. Zitvogel L, Galluzzi L, Kepp O, SmythMJ, Kroemer G. Type I interferons inanticancer immunity. Nat Rev Immunol 2015;15:405–14.

54. Bakhoum SF, Ngo B, Laughney AM, Cavallo JA, Murphy CJ, Ly P, et al.Chromosomal instability drives metastasis through a cytosolic DNAresponse. Nature 2018;553:467–72.

55. Beyer U, Brand F, Martens H, Weder J, Christians A, Elyan N, et al. RareADAR and RNASEH2B variants and a type I interferon signature in gliomaand prostate carcinoma risk and tumorigenesis. Acta Neuropathol2017;134:905–22.


Hiller et al.




2018;78:5917-5926. Published OnlineFirst August 28, 2018.Cancer Res Björn Hiller, Anja Hoppe, Christa Haase, et al. Cell Carcinoma of the SkinRibonucleotide Excision Repair Is Essential to Prevent Squamous

Updated version

10.1158/0008-5472.CAN-18-1099doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2018/08/28/0008-5472.CAN-18-1099.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/78/20/5917.full#ref-list-1

This article cites 53 articles, 16 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/78/20/5917.full#related-urls

This article has been cited by 4 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/78/20/5917To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-18-1099

http://cancerres.aacrjournals.org/content/suppl/2018/08/28/0008-5472.CAN-18-1099.DC1

http://cancerres.aacrjournals.org/content/78/20/5917.full#ref-list-1

http://cancerres.aacrjournals.org/content/78/20/5917.full#related-urls

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerres.aacrjournals.org/content/78/20/5917


ribonucleotide excision repair is essential to prevent ... · tumor biology and immunology...

Documents