BRIEF COMMUNICATION www.jasn.org

A FANCD2/FANCI-Associated Nuclease 1-Knockout Model Develops Karyomegalic Interstitial Nephritis

Rannar Airik,* Markus Schueler,* Merlin Airik,* Jang Cho,* Jonathan D. Porath,* † † ‡ Elina Mukherjee, Sunder Sims-Lucas, and Friedhelm Hildebrandt*

*Department of Medicine, Boston Children’s Hospital, Boston, Massachusetts; †Department of Pediatrics, Children’s Hospital of Pittsburgh of the University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania; and ‡Howard Hughes Medical Institute, Chevy Chase, Maryland

ABSTRACT Karyomegalic interstitial nephritis (KIN) is a chronic interstitial nephropathy charac- mice using ES cells obtained from the terized by tubulointerstitial nephritis and formation of enlarged nuclei in the kidneys KOMP Repository [Fan1tm1a(KOMP)Wtsi] and other tissues. We recently reported that recessive mutations in the gene encod- (Supplemental Figure 1A). Homozygous ing FANCD2/FANCI-associated nuclease 1 (FAN1) cause KIN in humans. FAN1 is a Fan1tm1a/tm1a mice were born at Mendelian major component of the Fanconi anemia–related pathway of DNA damage response ratios and appeared healthy, with no gross (DDR) signaling. To study the pathogenesis of KIN, we generated a Fan1 knockout abnormality in the kidney. Using quantita- mouse model, with abrogation of Fan1 expression confirmed by quantitative RT- tive RT-PCR on kidneys isolated from 2/2 PCR. Challenging Fan1 andwild-typemicewith20mg/kgcisplatincausedAKIin wild-type, Fan1tm1a/+, and Fan1tm1a/tm1a both genotypes. In contrast, chronic injection of cisplatin at 2 mg/kg induced KIN that animals, we show that there was no Fan1 2/2 led to renal failure within 5 weeks in Fan1 mice but not in wild-type mice. Cell culture transcript generated in Fan1tm1a/tm1a kid- 2/2 studies showed decreased survival and reduced colony formation of Fan1 mouse neys (Supplemental Figure 1, B and C), embryonic fibroblasts and bone marrow mesenchymal stem cells compared with wild- suggesting that the mice are Fan1 null (des- 2 2 2 2 type counterparts in response to treatment with genotoxic agents, suggesting that ignated Fan1 / ). We aged Fan1 / mice FAN1 mutations cause chemosensitivity and bone marrow failure. Our data show that for 18 months but observed no histologic 2 2 Fan1 is involved in the physiologic response of kidney tubular cells to DNA damage, changes in Fan1 / kidneys compared 2/2 which contributes to the pathogenesis of CKD. Moreover, Fan1 mice provide a new withwild-type controls (Supplemental Fig- model with which to study the pathomechanisms of CKD. ure 2), suggesting that inactivation of Fan1 does not cause renal histologic abnor- 2/2 J Am Soc Nephrol 27: 3552–3559, 2016. doi: 10.1681/ASN.2015101108 malities in normally aging Fan1 ani- mals within laboratory environment.

CKD is a progressive degenerative con- anemia, congenital abnormalities, dition that leads to renal failure. Wehave susceptibility to leukemia and other Received October 9, 2015. Accepted February 17, previously shown that one of the cancers, and hypersensitivity to inter- 2016. mechanisms underlying CKD is chro- strand DNA crosslinking agents, such as 7,8 Published online ahead of print. Publication date mosomal instability caused by a failure mitomycin C (MMC) and cisplatin. available at www.jasn.org. in DNA interstrand crosslink (ICL) re- Interestingly, mutations in FAN1 have ’ fi 1 Present address: Dr. Rannar Airik, Children sHos- pair in people with FAN1 de ciency. thus been far reported to cause karyo- pital of Pittsburgh of the University of Pittsburgh FAN1 is a downstream effector of the megalic interstitial nephritis (KIN) and Medical Center, Pittsburgh, Pennsylvania. Fanconi anemia (FA) DNA repair path- familial colorectal cancer but not bone Correspondence: Dr. Rannar Airik, Children’s 2 — way. FAN1 interacts with FANCD2 marrow abnormalities a central feature Hospital of Pittsburgh of the University of Pittsburgh and FANCI on ICL-induced DNA dam- of FA.1,9 KIN exhibits features similar to Medical Center, 4401 Penn Avenue, Pittsburgh, PA age response (DDR) and is recruited to the nephronophthisis-related ciliopathies 15224, or Dr. Friedhelm Hildebrandt, Boston Chil- dren’s Hospital, EN561, 300 Longwood Avenue, the sites of ICL lesions, where its func- (NPHP-RCs) for which a mechanism of Boston, MA 02115. Email: [email protected] or Fried- tion is critical for resolving ICLs.3–6 DDR was described.10 [email protected]

Mutations in genes encoding for FA pro- Tostudy thepathophysiologyofKIN,we Copyright © 2016 by the American Society of teins cause FA, which features aplastic first generated homozygous mutant Fan1 Nephrology

3552 ISSN : 1046-6673/2712-3552 JAmSocNephrol27: 3552–3559, 2016 www.jasn.org BRIEF COMMUNICATION

Our previous work suggested that abnormalities, such as occasional tubu- cisplatin had indistinguishable histology 2 2 individuals with mutations in FAN1 lar casts (Supplemental Figure 3B), from wild–type and untreated Fan1 / mightbemoresusceptibletoenviron- whereas no recovery was observed in control mice (Figure 2, C–E). In contrast, 2 2 2 2 mental genotoxins, which contribute to cisplatin–treated Fan1 / kidneys Fan1 / mice treated with cisplatin the renal pathology of KIN.1 Indeed, in (Supplemental Figure 3D). Together, showed the characteristic features of 2 2 vitro studies with FAN1-deficient cells our data show that Fan1 / mice are KIN—formation of karyomegalic nuclei have shown cellular hypersensitivity more susceptible to cisplatin toxicity in the proximal tubule, presence of in- to DNA interstrand crosslinking agents and that they fail to recover from cisplatin- terstitial infiltrate, loss of brush borders, cisplatin and MMC.1,3–6 We, therefore, induced AKI in contrast to wild-type and tubular dilation (Figure 2F). In ad- 2 2 hypothesized that treatment of Fan1 / littermates. dition, they presented with segmental mice with cisplatin would induce DNA We next hypothesized that adminis- thickening of the tubular basement 2 2 ICL lesions and that repair in Fan1 / tering a reduced dose of cisplatin would membrane and frequent thickening mice would be deficient, leading to avoid the acute toxicity of cisplatin and of the Bowman’scapsule(Figure2F, KIN. Cisplatin is one of the most widely allow us to investigate a more subtle and Supplemental Figure 4), a characteristic used cancer therapeutics, but its use is chronic effect of drug toxicity on the feature of NPHP-RC. 2 2 limited because of severe nephrotoxicity, kidney function in Fan1 / animals. We To further characterize the extent of 2 2 with about 25%–35% of patients next administered a dose of 2 mg/kg renal fibrosis in Fan1 / kidneys, we displaying a transient decline in weekly. No mortality was observed in performed Masson trichrome staining. 2 2 renalfunctionafterasingledoseof Fan1 / mice at this dose initially. We Although we observed a strong blue 2 2 cisplatin.11 Itstoxicityarisesfrom monitored the mice for signs of general staining of collagen fibers in Fan1 / production of reactive oxygen species toxicity by weighing them weekly. By the kidneys (Figure 3D), no staining was 2 2 and generation of ICLs.12,13 We first third week, Fan1 / mice in the cisplatin detected in the kidneys from cisplatin– 2 2 2 2 asked whether Fan1 / mice were treatment group had reduced body injected wild–type or Fan1 / control more susceptible to AKI than wild- weight compared with treated wild– mice (Figure 3, A–C). Thus, our data on 2 2 type animals by injecting them with type mice (Figure 2A); 5 weeks after ini- the renal histology of Fan1 / mice 2 2 2 2 cisplatinatadoseof20mg/kg(n=8 tiating cisplatin treatment, Fan1 / mice show that the mouse model of Fan1 / 2 2 per experimental group). Fan1 / had lost 24% of their initial body recapitulates the human FAN1 deficiency mice showed 100% mortality in response weight and appeared moribund. In phenotypes of KIN and interstitial fibrosis to cisplatin treatment by postinjection day contrast, wild-type mice in the cisplatin in the kidney. 7, with some mice dying already on day 2 treatment group showed normal age– It has been suggested that karyomegaly 2 2 (Figure 1A). Fan1 / mice that survived related increases in body weight is caused by impaired progression of the until day 7 appeared severely dehydrated (106%) and appeared healthy, indicat- cell cycle, which is characterized by re- and cachectic, having lost 30% of their ing that the administered dose was duced levels of the cell proliferation body weight (Figure 1B). In contrast, not sufficient to cause cachexia in marker Ki67 and an increase in S–phase wild-type mice treated with cisplatin wild–type animals (Figure 2A). Like- marker proliferating cell nuclear antigen 2 2 (20 mg/kg) showed 10% weight loss wise, untreated Fan1 / mice showed in karyomegalic nuclei.14 However, these with no mortality (Figure 1B). no weight loss (Figure 2A). observations have not been confirmed To examine the renal histology of Individuals with mutations in the by other groups.15 We investigated the cisplatin-injected mice, we performed FAN1 gene have progressive renal failure levels of Ki67 by immunofluorescence 2 2 histologic analysis of kidney sections and increased levels of BUN.14 We mea- in Fan1 / kidneys. Neither wild–type 2 2 2 2 using hematoxylin and eosin staining sured the BUN levels in Fan1 / mice (Figure 3E) nor Fan1 / control kidneys on postinjection days 3 and 7. Both wild- after they had been treated for 5 weeks (Figure 3G) contained Ki67-positive 2 2 type and Fan1 / mice had character- with cisplatin. There was a steep increase nuclei in the kidneys, indicative of very 2 2 istic signs of AKI on the third day in in the BUN levels in Fan1 / mice low levels of cell proliferation in adult response to cisplatin administration (ninefold over wild–type treated mice), kidneys. In contrast, kidneys from 2 2 compared with uninjected control although wild-type mice receiving cisplatin wild-type (Figure 3F) and Fan1 / 2 2 mice (Figure 1, C–F), and the affected injections and untreated Fan1 / mice mice (Figure 3H) that were treated kidneys contained dilated tubules filled had BUN levels indistinguishable from with cisplatin contained Ki67-positive with massive protein casts (Figure 1, D those of the wild-type mice (Figure 2B). nuclei, suggesting that cisplatin induces 2 2 and F). In addition, Fan1 / kidneys The range of BUN increases was indica- cell cycle activity in the renal tubular 2 2 showed overt sloughing of the brush tive of severe renal failure in Fan1 / epithelium, probably as a repair re- borders, vacuolization, and blebbing mice. To examine the renal histologic sponse.16 The number of Ki67-positive 2 2 2 2 of the proximal tubules (Figure 1F). At changes in Fan1 / mice, we performed nuclei was notably higher in Fan1 / day 7, cisplatin–treated wild–type mice periodic acid–Schiff staining. Kidneys kidneys compared with in wild-type showed signs of repair and only a few fromwild-typemicetreatedwith kidneys (Figure 3, F and H). Moreover,

2/2 J Am Soc Nephrol 27: 3552–3559, 2016 Fan1 Model of Renal Karyomegaly 3553 BRIEF COMMUNICATION www.jasn.org

2 2 and survival of Fan1 / cells compared with in wild–type control cells in re- sponse to these genotoxins (Figure 4, A and B). Bone marrow failure in patients with FA includes abnormalities in the bone marrow mesenchymal stem cell (BM- MSC) niche.17,18 Thus, we investigated whether cisplatin administration to 2 2 Fan1 / mice induces bone marrow abnormalities in the mesenchymal stem cell compartment. We isolated 2 2 BM-MSCs from wild-type and Fan1 / mice and examined their invitro clonogenic potential.19 The average number of colonies formed per mouse (cells were isolated from two femurs per mouse) in control and cisplatin-injected animals indicates that the BM-MSCs from cisplatin–treated 2 2 Fan1 / mice had severely depressed clonogenic ability compared with those from wild–type cisplatin–injected mice (Figure 4C) (P,0.001). Bone marrow 2 2 cells from the Fan1 / control group had the same colony formation capa- bility as cells from wild–type control mice (Figure 4C). Together, our find- ingssuggestthatFan1functionis critical for bone marrow maintenance by eliminating ICLs. 2 2 Figure 1. Fan1 / mice show lethality in response to treatment with cisplatin at 20 mg/kg. Phosphorylation of H2AX (gH2AX) 2 2 (A) Survival curves of wild-type and Fan1 / mice in response to cisplatin administration. is a signature of ATM/ATR–regulated 2/2 Wild-type mice did not show mortality in response to cisplatin. However, all Fan1 DDR pathway activation. We next exam- mice were either dead or severely moribund and requiring euthanasia by day 7 after ined whether there is a difference be- cisplatin (20 mg/kg) administration. Survival is expressed as the percentage of the tween gH2AX levels in wild-type versus number of live mice subjected to cisplatin treatment. (B) Mice were weighed at days 2 2 Fan1 / kidneys on treatment with 0 and 7 of cisplatin (20 mg/kg) administration. No differences in body weight were 2 2 observed between untreated wild–type and Fan1 / mice. There was a small but sig- cisplatin. Indeed, we observed increased – nificant weight loss in cisplatin–treated wild–type mice compared with control mice. In levels of gH2AX in cisplatin treated 2 2 2/2 contrast, cisplatin–treated Fan1 / mice had dramatic weight loss compared with Fan1 kidneys, whereas gH2AX levels control mice. *P,0.05; **P,0.01; ***P,0.001. (C–F) Histology of kidneys from wild- in cisplatin–treated wild–type kidneys 2 2 type and Fan1 / animals 3 days after cisplatin (20 mg/kg) administration. (C) Untreated did not differ from the levels of untreated 2 2 2 2 wild–type and (E) Fan1 / kidneys appear histologically normal. (D) Wild-type kidneys wild–type and Fan1 / kidneys (Figure treated with cisplatin display a few protein casts in the proximal tubules (asterisks). (F) In 4D). Together, it shows a prolonged and 2/2 contrast, Fan1 kidneys treated with cisplatin have characteristic signs of AKI, in- unquenched DDR activity in cisplatin– 2 2 cluding sloughing of the proximal tubule brush borders, tubular dilations, and presence treated Fan1 / kidneys, a likely cause of massive protein casts in the proximal tubules (asterisks). ctrl, Control; wt, wild type. for CKD.20,21 Scale bars, 30 mm. 2 2 Our finding that Fan1 / mice de- velop bone marrow abnormalities on measuring the area of Ki67-positive nuclei To examine the effect of various treatment with cisplatin suggests that 2 2 revealed that they were significantly en- genotoxins on Fan1 / cell survival, FAN1 mutations may underlie previ- 2 2 larged in Fan1 / kidneys compared we isolated mouse embryonic fibroblasts ously unrecognized susceptibility for 2 2 with in wild-type kidneys (Figure 3I). (MEFs) from wild-type and Fan1 / bone marrow failure in humans. This This result is consistent with induction embryos and treated them with MMC warrants analysis of FAN1 mutations in of karyomegaly caused by impaired cell or diepoxybutane (DEB). Indeed, we ob- patients with bone marrow failure. Thus 2 2 cycle regulation in Fan1 / kidneys.3 served reduced colony formation ability far, FAN1 deficiency has been clinically

3554 Journal of the American Society of Nephrology J Am Soc Nephrol 27: 3552–3559, 2016 www.jasn.org BRIEF COMMUNICATION

doses of environmental genotoxins, such as food-derived genotoxins, genotoxic drugs, chemotherapeutics, etc.,mayinsid- iously lead to accumulation of irreparable DNA damage and KIN in humans. This is not without a precedent. For example, 2 2 2 2 Fanca / and Fancc / mice do not reca- pitulate the spontaneous bone marrow disease of patients with FA.23,24 However, when challenged with genotoxic agents or inflammatory signals, these mice precipitated bone marrow failure.23,24 Together, these studies show that environ- mental factors are critical contributors to the onset of these diseases. 2 2 In conclusion, we show that Fan1 / mice provide a new model to study the pathomechanisms of drug-induced CKD 2 2 and NPHP-RC. We show that Fan1 / mice are highly sensitive to genotoxic in- sults, which lead to KIN and bone marrow abnormalities.

CONCISE METHODS

Mouse Breeding and Maintenance The experimental protocol was reviewed and approved by the Animal Care Commit- tee of Boston Children’s Hospital. Targeted Fan1tm1a(KOMP)Wtsi ES cells were obtained from KOMP and injected into blastocysts 2 2 to generate Fan1-transgenic mice. Chimeric Figure 2. Chronic administration of low doses of cisplatin causes KIN in Fan1 / mice; mice were prepared by blastocyst micro- 6- to 8-week-old mice were weekly given low doses of cisplatin (2 mg/kg). Renal his- tology and function were analyzed 5 weeks later. (A) Chronic cisplatin administration injection and bred with C57BL/6J mice to 2 2 resulted in dramatic weight loss in treated Fan1 / mice but not in wild-type or un- obtain germline transmission. Founders 2 2 treated Fan1 / mice. Weight loss became apparent already 3 weeks after cisplatin were bred into a 129SvImJ congenic line was administered. (B) BUN values as a measure of renal function in control and cisplatin- for at least five generations before experi- treated mice after 5 weeks of treatment. Chronic administration of cisplatin did not ments were performed. Wild–type litter- 2 2 affect kidney function in wild-type or untreated Fan1 / mice compared with control mate mice were used as controls for Fan1 2 2 mice. In contrast, Fan1 / mice showed a ninefold increase in BUN values in response 2 2 mutant mice. to cisplatin (wild type, 40.461.4 mg/dl; Fan1 / ,47.467.9 mg/dl; wild-type cisplatin, 2 2 35.266.5 mg/dl; Fan1 / cisplatin, 324.7617.1 mg/dl). **P,0.01. (C–F) Histology 2/2 Cisplatin Injection of kidneys from wild-type and Fan1 animals after 5 weeks of cisplatin treatment. 2/2 2 2 Eight-week-old or 6-month-old Fan1 Although cisplatin–treated wild–type and untreated Fan1 / kidneys appear histologically 2/2 mice were injected intraperitoneally with 2 normal, Fan1 kidneys treated with cisplatin (2 mg/kg) display the characteristic fea- tures of KIN, presence of interstitial infiltrate, and karyomegalic nuclei (green arrows). or 20 mg/kg cisplatin, respectively (Teva Notice the thickening of the tubular basement membranes (red arrows). CTRL, Control; Parenteral Medicines, Inc.), diluted in normal wt, wild type. saline solution. Control mice received normal saline injections of equivalent volumes. Mouse body weight at the time of injection ranged associated with microcephaly together in determining the disease onset in indi- from 25 to 40 g. with other developmental defects, KIN, viduals with FAN1 deficiency who are 1,9,22 and colorectal carcinoma. Our predisposed to genetic injury. On the Biochemical Analyses 2/2 work also sheds light on the possible in- basis of our studies in Fan1 mice, On the day of euthanasia, blood was imme- volvement of environmental genotoxins we propose that exposure to subclinical diately collected, and serum levels of BUN

2/2 J Am Soc Nephrol 27: 3552–3559, 2016 Fan1 Model of Renal Karyomegaly 3555 BRIEF COMMUNICATION www.jasn.org

2 2 Figure 3. KIN is characterized by renal tubular fibrosis and Ki67 positivity in Fan1 / mice. (A–D) Masson trichrome staining was used to 2 2 examine kidney fibrosis. No fibrosis was observed in (A) wild–type, (B) Fan1 / , or (C) cisplatin–treated wild–type control mice. (D) 2 2 In contrast, cisplatin–treated Fan1 / animals developed renal fibrosis (blue staining indicates collagen deposits). (E–H) Ki67 2 2 immunostaining (red) shows cell cycle activation in the renal epithelium of (F) cisplatin–treated wild–type and (H) Fan1 / kidneys but not in

3556 Journal of the American Society of Nephrology J Am Soc Nephrol 27: 3552–3559, 2016 www.jasn.org BRIEF COMMUNICATION

tubules were stained with FITC–conjugated lotus tetragonolobus agglutinin lectin (Vector Laboratories, Burlingame, CA). Secondary antibodies were donkey anti– rabbit Alexa Fluor 594 (Molecular Probes) and goat anti–rabbit IgG-HRP (sc-2004; Santa Cruz Biotechnology, Santa Cruz, CA). Samples were mounted in ProlongGold (Molecular Probes), and images were captured on a Leica TSC 5SPX Confocal Microscope (Leica Microsystems, Buffalo Grove, IL).

RNA Extraction and Quantitative RT-PCR 2 RNA was isolated from wild–type, Fan1+/ , 2 2 and Fan1 / adult mouse kidneys using the RNeasy Mini Kit (Qiagen, Germantown, 2 2 MD), and reverse transcription was performed Figure 4. Fan1 is critical for resistance to genotoxins and bone marrow survival. (A) Fan1 / MEFs have reduced clonogenic ability in response to treatment with MMC compared with using Superscript III (Invitrogen, Carlsbad, 2 2 – wild-type MEFs. **P,0.01. (B) Fan1 / MEFs have reduced survival in response to treat- CA). Quantitative real time PCR was carried ment with DEB compared with wild-type MEFs. (C) Clonogenic survival assays were carried out using TaqMan primers (Life Technologies, 2 2 out with BM-MSCs isolated from the wild-type, Fan1 / control (saline solution), and wild– Carlsbad, CA) and run on a StepOne Real- 2 2 type Fan1 / –treated (2 mg/kg cisplatin) mice. Colony formation ability of bone marrow Time PCR System (Applied Biosystems, 2 2 cells from noninjected Fan1 / mice was comparable with that of cells from noninjected Foster City, CA). Data were normalized to wild–type mice (P=0.87). Cisplatin reduced the colony formation ability of wild-type cells Gapdh and B2m. compared with cells from noninjected animals. However, cisplatin injection caused a much 2/2 stronger repression of colony formation of Fan1 bone marrow cells compared with cells Western Blotting – – 2/2 from cisplatin treated wild type animals and nontreated Fan1 animals. *P=0.03; Kidney tissues were lysed in RIPA Lysis Buffer ***P,0.001; ****P,0.001. (D) Whole-kidney lysates from control and cisplatin–treated 2 2 (Pierce, Rockford, IL) and homogenized mouse kidneys analyzed for DDR pathway activity. Fan1 / kidneys show elevated levels of with a douncer. Cleared tissue lysates were phosphorylation of H2AX 7 days after a single cisplatin injection of 10 mg/kg, a sign of persistent DDR pathway activity. CTRL, Control; wt, wild type. produced by centrifugation of the resulting samples at 16,0003g for 30 minutes at 4°C. Gel electrophoresis of tissue lysates were measured using a Urea Assay Kit software (National Institutes of Health, Be- was performed using the NuPAGE System (ab83362; Abcam, Inc., Cambridge, MA). thesda, MD). (Invitrogen). Samples were resolved on 4%–12% Bis-Tris gels in 3-(N-morpholino) Histologic and Immunofluorescence Generation of MEFs propanesulfonic acid buffer and transferred Analyses of Kidney MEFs were established from wild–type and to a nitrocellulose membrane that was then 2 2 Kidneys were fixed with 10% formalin for Fan1 / E13.5 embryos and cultured probed for the protein of interest using anti- 24 hours. Kidneys were sectioned at 5-mm in DMEM with 10% FBS and penicillin/ bodies diluted in Tris-buffered saline containing thicknessandstainedwithhematoxylin streptomycin. 5% milk and 0.1% Tween-20 (Sigma-Aldrich). and eosin, periodic acid–Schiff, or Masson trichrome reagent for histologic examina- Antibodies Statistical Methods tion. For immunostaining, sections were Primary antibodies used were as follows: rabbit The t test was used to compare data between blocked in 10% donkey serum/1% BSA polyclonal anti–Ki67 (ab15580; Abcam, Inc.), two groups. Significance was determined at and permeabilized in 0.1% Tween-20 rabbit polyclonal H2AX (ab11175; Abcam, P,0.05, and P,0.05, P,0.01, and P,0.001 (Sigma-Aldrich, St. Louis, MO). Analysis of Inc.), and polyclonal anti-gH2AX (9718; Cell are indicated. Where appropriate, data were nuclear size was performed using ImageJ Signaling Technology, Danvers, MA). Proximal presented as means6SEMs.

2 2 2 2 (E) nontreated wild–type and (G) Fan1 / kidneys. (H) However, the number of proliferating cells (Ki67 positive) is higher in Fan1 / 2 2 kidneys. Furthermore, a majority of the Ki67-positive nuclei in Fan1 / kidneys are enlarged, representing karyomegalic nuclei. Notice a partial 2 2 2 2 loss of LTA–positive brush borders (green arrows) from Fan1 / kidneys. (I) Area of Ki67-positive nuclei in wild-type versus Fan1 / kidneys 2 2 treated with cisplatin. The mean size of Ki67-positive nuclei in Fan1 / kidneys is significantly larger compared with that in wild-type kidneys (60866341.9 versus 39246180.0 pixels). ctrl, Control; LTA, lotus tetragonolobus agglutinin; px, pixels; wt, wild type. ****P,0.001.

2/2 J Am Soc Nephrol 27: 3552–3559, 2016 Fan1 Model of Renal Karyomegaly 3557 BRIEF COMMUNICATION www.jasn.org

Colony Formation Assay S, Ramaswami G, Dixon J, Burkhalter F, 11. Pabla N, Dong Z: Cisplatin nephrotoxicity: BM-MSC were cultured as described.19 Spoendlin M, Moch H, Mihatsch MJ, Mechanisms and renoprotective strategies. – Briefly, total numbers of isolated bone Verine J, Reade R, Soliman H, Godin M, Kidney Int 73: 994 1007, 2008 Kiss D, Monga G, Mazzucco G, Amann K, 12. Wang LC, Gautier J: The Fanconi anemia marrow cells from two femurs per mouse Artunc F, Newland RC, Wiech T, pathway and ICL repair: Implications for were seeded in 6-cm dishes and cultured for Zschiedrich S, Huber TB, Friedl A, Slaats cancer therapy. Crit Rev Biochem Mol Biol 7 days. Colonies were stained with crystal GG, Joles JA, Goldschmeding R, Washburn 45: 424–439, 2010 violet solution and counted. J, Giles RH, Levy S, Smogorzewska A, 13. Marullo R, Werner E, Degtyareva N, Moore B, Hildebrandt F: FAN1 mutations cause Altavilla G, Ramalingam SS, Doetsch PW: karyomegalic interstitial nephritis, linking Cisplatin induces a mitochondrial-ROS re- DNA Damage Sensitivity Assay chronic kidney failure to defective DNA sponse that contributes to cytotoxicity de- MEFs were plated in a six-well plate in damage repair. Nat Genet 44: 910–915, pending on mitochondrial redox status and triplicate at a density of 53104 cells per 2012 bioenergetic functions. PLoS One 8: e81162, ’ well. Immediately after plating, MMC or 2. D Andrea AD: Susceptibility pathways in 2013 Fanconi’s anemia and breast cancer. N Engl J 14. Spoendlin M, Moch H, Brunner F, Brunner W, DEB was added at a final concentration of Med 362: 1909–1919, 2010 Burger HR, Kiss D, Wegmann W, Dalquen P, – 0or0.5mM for MMC or 0 1 mg/ml for 3.LiuT,GhosalG,YuanJ,ChenJ,HuangJ: Oberholzer M, Thiel G, Mihatsch MJ: DEB. After 8 days of culture, surviving FAN1 acts with FANCI-FANCD2 to promote Karyomegalic interstitial nephritis: Further colonies (MMC) and cell numbers (DEB) DNA interstrand cross-link repair. Science support for a distinct entity and evidence for a – – were counted. Colony number after 329: 693 696, 2010 genetic defect. Am J Kidney Dis 25: 242 252, 4. Kratz K, Schöpf B, Kaden S, Sendoel A, 1995 MMC or cell number after DEB treatment Eberhard R, Lademann C, Cannavó E, 15. Baba F, Nanovic L, Jaffery JB, Friedl A: was normalized to cell number in the Sartori AA, Hengartner MO, Jiricny J: De- Karyomegalic tubulointerstitial nephritis—a untreated sample to give the percentage ficiency of FANCD2-associated nuclease case report. Pathol Res Pract 202: 555–559, of survival. KIAA1018/FAN1 sensitizes cells to inter- 2006 strand crosslinking agents. Cell 142: 77–88, 16. Megyesi J, Safirstein RL, Price PM: Induction 2010 of p21WAF1/CIP1/SDI1 in kidney tubule 5. MacKay C, Déclais AC, Lundin C, Agostinho cells affects the course of cisplatin-induced – ACKNOWLEDGMENTS A, Deans AJ, MacArtney TJ, Hofmann K, acute renal failure. J Clin Invest 101: 777 Gartner A, West SC, Helleday T, Lilley DM, 782, 1998 Rouse J: Identification of KIAA1018/FAN1, a 17. Li Y, Chen S, Yuan J, Yang Y, Li J, Ma J, Wu X, We thank Thomas L. Saunders, Elizabeth DNA repair nuclease recruited to DNA Freund M, Pollok K, Hanenberg H, Goebel Hughes, Keith Childs, Galina Gavrilina, and damage by monoubiquitinated FANCD2. WS, Yang FC: Mesenchymal stem/pro- – Debra Vanheyningen for preparation of the Cell 142: 65 76, 2010 genitor cells promote the reconstitution of 6. Smogorzewska A, Desetty R, Saito TT, exogenous hematopoietic stem cells in embryonic stem cell mouse chimeras from – tm1a(KOMP)Wtsi Schlabach M, Lach FP, Sowa ME, Clark Fancg-/- mice in vivo. Blood 113: 2342 Fan1 embryonic stem cell clone AB, Kunkel TA, Harper JW, Colaiácovo 2351, 2009 EPD0736_4_C08 and the Transgenic Animal MP, Elledge SJ: A genetic screen iden- 18. Mantelli M, Avanzini MA, Rosti V, Ingo DM, Model Core of the University of Michigan’s tifies FAN1, a Fanconi anemia-associated Conforti A, Novara F, Arrigo G, Boni M, Biomedical Core Facilities. nuclease necessary for DNA inter- Zappatore R, Lenta E, Moretta A, Acquafredda strand crosslink repair. Mol Cell 39: 36–47, G, de Silvestri A, Cirillo V, Cicchetti E, Algeri Core support was also provided by the 2010 M, Strocchio L, Vinti L, Starc N, Biagini S, Diabetes Research and Training Center and 7. Butturini A, Gale RP, Verlander PC, Adler- Sirleto P, Bernasconi P, Zuffardi O, Maserati E, the O’Brien Renal Core Center, which are Brecher B, Gillio AP, Auerbach AD: Hema- Maccario R, Zecca M, Locatelli F, Bernardo funded by National Institutes of Health tologic abnormalities in Fanconi anemia: An ME: Comprehensive characterization of (NIH) grants DK20572 and P30-DK08194, International Fanconi Anemia Registry study. mesenchymal stromal cells from patients with Blood 84: 1650–1655, 1994 Fanconi anaemia. Br J Haematol 170: 826– respectively. This research was supported 8. Auerbach AD: Fanconi anemia diagnosis and 836, 2015 by NIH grants DK099434 (to R.A.) and the diepoxybutane (DEB) test. Exp Hematol 19. Soleimani M, Nadri S: A protocol for isolation DK068306 (to F.H.). F.H. is an Investigator of 21: 731–733, 1993 and culture of mesenchymal stem cells from the Howard Hughes Medical Institute and the 9. Seguí N, Mina LB, Lázaro C, Sanz- mouse bone marrow. Nat Protoc 4: 102–106, Warren E. Grupe Professor. Pamplona R, Pons T, Navarro M, Bellido F, 2009 López-Doriga A, Valdés-Mas R, Pineda M, 20. Chaki M, Airik R, Ghosh AK, Giles RH, Chen R, Guinó E, Vidal A, Soto JL, Caldés T, Durán Slaats GG, Wang H, Hurd TW, Zhou W, M, Urioste M, Rueda D, Brunet J, Balbín M, Cluckey A, Gee HY, Ramaswami G, Hong CJ, DISCLOSURES Blay P, Iglesias S, Garré P, Lastra E, Hamilton BA, Cervenka I, Ganji RS, Bryja V, Sánchez-Heras AB, Valencia A, Moreno V, Arts HH, van Reeuwijk J, Oud MM, Letteboer None. Pujana MA, Villanueva A, Blanco I, SJ, Roepman R, Husson H, Ibraghimov- Capellá G, Surrallés J, Puente XS, Valle L: Beskrovnaya O, Yasunaga T, Walz G, Eley L, Germline mutations in FAN1 cause Sayer JA, Schermer B, Liebau MC, Benzing T, REFERENCES hereditary colorectal cancer by impairing Le Corre S, Drummond I, Janssen S, Allen SJ, DNA repair. Gastroenterology 149: 563–566, Natarajan S, O’Toole JF, Attanasio M, 1. Zhou W, Otto EA, Cluckey A, Airik R, Hurd 2015 Saunier S, Antignac C, Koenekoop RK, Ren TW, Chaki M, Diaz K, Lach FP, Bennett 10. Lans H, Hoeijmakers JH: Genome stability, H, Lopez I, Nayir A, Stoetzel C, Dollfus H, GR, Gee HY, Ghosh AK, Natarajan S, progressive kidney failure and aging. Nat Massoudi R, Gleeson JG, Andreoli SP, Thongthip S, Veturi U, Allen SJ, Janssen Genet 44: 836–838, 2012 Doherty DG, Lindstrad A, Golzio C, Katsanis

3558 Journal of the American Society of Nephrology J Am Soc Nephrol 27: 3552–3559, 2016 www.jasn.org BRIEF COMMUNICATION

N, Pape L, Abboud EB, Al-Rajhi AA, Lewis Hildebrandt F: Renal-retinal ciliopathy gene haematopoietic stem cells. Cell Cycle 14: RA, Omran H, Lee EY, Wang S, Sekiguchi JM, Sdccag8 regulates DNA damage response 2734–2742, 2015 Saunders R, Johnson CA, Garner E, Vanselow signaling. JAmSocNephrol25: 2573–2583, 24. Carreau M, Gan OI, Liu L, Doedens M, K, Andersen JS, Shlomai J, Nurnberg G, 2014 McKerlie C, Dick JE, Buchwald M: Bone Nurnberg P, Levy S, Smogorzewska A, Otto 22. Trujillo JP, Mina LB, Pujol R, Bogliolo M, marrow failure in the Fanconi anemia group EA, Hildebrandt F: Exome capture reveals Andrieux J, Holder M, Schuster B, C mouse model after DNA damage. Blood ZNF423 and CEP164 mutations, linking renal Schindler D, Surrallés J: On the role of 91: 2737–2744, 1998 ciliopathies to DNA damage response sig- FAN1 in Fanconi anemia. Blood 120: 86– naling. Cell 150: 533–548, 2012 89, 2012 21. Airik R, Slaats GG, Guo Z, Weiss AC, Khan N, 23. Kaschutnig P, Bogeska R, Walter D, Lier A, Ghosh A, Hurd TW, Bekker-Jensen S, Huntscha S, Milsom MD: The Fanconi This article contains supplemental material online Schrøder JM, Elledge SJ, Andersen JS, anemia pathway is required for efficient at http://jasn.asnjournals.org/lookup/suppl/doi:10. Kispert A, Castelli M, Boletta A, Giles RH, repair of stress-induced DNA damage in 1681/ASN.2015101108/-/DCSupplemental.

2/2 J Am Soc Nephrol 27: 3552–3559, 2016 Fan1 Model of Renal Karyomegaly 3559