Montanuy et al. Orphanet Journal of Rare Diseases (2020) 15:170 https://doi.org/10.1186/s13023-020-01437-1 RESEARCH Open Access High content drug screening for Fanconi anemia therapeutics Helena Montanuy1, Cristina Camps-Fajol1,2, Jordi Carreras-Puigvert1,2,3, Maria Häggblad4,5, Bo Lundgren6, Miriam Aza-Carmona7,8, Thomas Helleday4, Jordi Minguillón1,3,9† and Jordi Surrallés1,2,3,9*† Abstract Background: Fanconi anemia is a rare disease clinically characterized by malformations, bone marrow failure and an increased risk of solid tumors and hematologic malignancies. The only therapies available are hematopoietic stem cell transplantation for bone marrow failure or leukemia, and surgical resection for solid tumors. Therefore, there is still an urgent need for new therapeutic options. With this aim, we developed a novel high-content cell- based screening assay to identify drugs with therapeutic potential in FA. Results: A TALEN-mediated FANCA-deficient U2OS cell line was stably transfected with YFP-FANCD2 fusion protein. These cells were unable to form fluorescent foci or to monoubiquitinate endogenous or exogenous FANCD2 upon DNA damage and were more sensitive to mitomycin C when compared to the parental wild type counterpart. FANCA correction by retroviral infection restored the cell line’s ability to form FANCD2 foci and ubiquitinate FANCD2. The feasibility of this cell-based system was interrogated in a high content screening of 3802 compounds, including a Prestwick library of 1200 FDA-approved drugs. The potential hits identified were then individually tested for their ability to rescue FANCD2 foci and monoubiquitination, and chromosomal stability in the absence of FANCA. Conclusions: While, unfortunately, none of the compounds tested were able to restore cellular FANCA-deficiency, our study shows the potential capacity to screen large compound libraries in the context of Fanconi anemia therapeutics in an optimized and cost-effective platform. Keywords: Fanconi anemia, High content screening, Drug repositioning, Cell-based assay Background develop hematological malignancies, such as acute mye- Fanconi anemia (FA) is a rare disease characterized by loid leukemia (AML) or myelodysplastic syndrome bone marrow failure (BMF), high incidence of cancer (MDS), with an accumulative incidence of up to 20% in and congenital malformations. FA has a low incidence of the adulthood. Finally, from the third decade of life FA 1/160,000 and an estimated carrier frequency of 1/200. patients will develop solid tumors, especially head and Up to 80% of FA patients will develop BMF before the neck squamous cell carcinoma with an accumulative in- age of 10. At the second decade of life they will start to cidence of more than 50% beyond their fourth decade of life [1]. There are currently 22 FA genes identified, forming a * Correspondence: Jordi.surralles@uab.cat; jordi.minguillon@uab.cat † Jordi Minguillón and Jordi Surrallés contributed equally to this work. complex pathway that repairs stalled replication forks 1 Department of Genetics and Microbiology, Universitat Autònoma de and inter-strand crosslinked lesions during the S-phase Barcelona, Barcelona, Spain of the cell cycle. FANCA is the most frequently mutated 2 Join Research Unit on Genomic Medicine UAB-Sant Pau, Biomedical Research Institute, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain gene accounting for up to 80% of all FA patients [1, 2]. Full list of author information is available at the end of the article The pathophysiological effects of the lack of a functional © The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Montanuy et al. Orphanet Journal of Rare Diseases (2020) 15:170 Page 2 of 9 FA/BRCA pathway include hypersensitivity to DNA- the crosslink, and finally repair the DNA by homologous damaging agents or aldehydes (from alcohol consump- recombination. FANCD2 monoubiquitination can be tion or cell metabolism), oxidative stress or overexpres- easily seen by fluorescence microscopy. In the absence sion of proinflammatory cytokines, such as TNF-α [3– of DNA damage, FANCD2 is normally located in the 5]. These studies derived in the proposal of candidate nucleus as a diffuse staining, while monoubiquitinated drugs to prevent FA defects, and some of them are cur- FANCD2 is seen as intense spots, or foci, throughout rently being tested in clinical trials, such as quercetin the nucleus, representing stalled replication forks on and metformin [6, 7]. chromatin during S-phase (Fig. 1a left) [2]. In a FANCA- Drug screenings are critical steps in the drug discovery deficient cell line, FANCL is not able to monoubiquiti- strategies, and they can be divided in high throughput nate FANCD2 and FANCI. Thus, FANCD2 localization screenings (HTS), when they are based on specific is seen in a nuclear diffuse pattern, even when cells are mechanisms of action, or high content screenings treated with ICL DNA damaging agents (Fig. 1a middle). (HCS), when the readouts are phenotypic changes. The A drug able to restore FANCL activity in the absence of preferred approach has long been HTS, as a molecular FANCA would be able to again monoubiquitinate target is identified by basic research and further explored FANCD2/FANCI, restoring FANCD2 ability to form nu- by research centers or the pharmaceutical industry, clei foci upon DNA damage induction (Fig. 1a right). screening hundreds of thousands of molecules. Pheno- For our cell-based system we used the fast growing typic approaches, however, require the identification of a and genetically amendable U2OS cell line where we specific cellular function that may be defective in a spe- knocked-out FANCA gene expression by TALEN (Tran- cific disease, of which the target is usually unknown, but scription Activator-like Effector Nuclease) gene editing has been very useful to find new drugs for rare diseases [10]. FANCA-deficient clones were verified by sequen- [8]. Compounds are usually screened in cells to correct cing, FANCA expression and FANCD2 ubiquitination an abnormal function, and these assays are optimized in (Fig. 1b and Figure Suppl. 1). Additionally, the two inde- technological facilities, such as microscopy or cell cy- pendent clones selected were more sensitive to mitomy- tometry coupled with robotic arms, in order to allow for cin C treatment in survival assays than the wild type cell high-content screenings with libraries of thousands to line (Fig. 1c). One of the clones was stably transfected tens of thousands of compounds. with YFP-FANCD2 fluorescent fusion protein, in order In this study we present a cell-based system with a to easily detect FANCD2 localization by fluorescence FANCA-defective background, in which we monitored microscopy. The FANCA-deficient YFP-FANCD2 cell the FA/BRCA pathway activity analyzing FANCD2 foci line expressed only the non-ubiquitinated form (Fig. 1d, by fluorescence microscopy. We performed a high con- left panel) and was unable to form fluorescent FANCD2 tent screening in search for drugs that could rescue foci when cells were treated with hydroxyurea (HU, Fig. FANCA deficiency, and among the 3802 compounds 1e, middle panel). When the cell line was infected with a tested we included FDA-approved drugs to check for FANCA-expressing retrovirus, endogenous FANCD2 drug repositioning candidates. We identify benefits and and YFP-FANCD2 could be monoubiquitinated and drawbacks of using this cell-based system for potential FANCD2 foci could be seen again (Fig. 1d and e, right future screenings in the Fanconi anemia field. panels). Results High content screening FA-cell based system to monitor FANCD2 foci formation After testing different conditions (Fig. 2a and data not Our objective was to create a cellular system based on shown), we seeded 1500 cells per well in 384 well-plates fluorescence, to monitor the activation of the Fanconi where we previously dispensed the drug collection. After anemia pathway (Fig. 1a). With this purpose, we devel- 24 h cells were treated with hydroxyurea, and 24 h later oped a FANCA deficient cell line expressing fluores- plates were fixed, washed, images collected and proc- cently tagged FANCD2, as a central protein that essed. From the collected images around 100–200 nuclei requires functional upstream FA proteins (FANCA, per well were counted with CellProfiler 2.0 software (six FANCC and FANCG among them, whose mutations images/well taken with the 40x objective). 384 well plate represent the majority of FA patients) to be activated by layout allowed to test 147 compounds/plate (two plates monoubiquitination. Covalent bonds between both DNA for replicates), with two concentrations per plate (1 and strands (interstrand-crosslinks, ICLs) in S-phase induce 10 μM), totaling 52 plates for the 3802 drugs from 4 dif- stalled replication forks and the activation of the E3 ubi- ferent libraries (Fig. 2b and c). Figure 2e shows a repre- quitin ligase FANCL [9]. FANCL then monoubiquiti- sentative plate with the percentage of nuclei with 5 foci nates FANCD2 and FANCI, which will bind to the DNA or more (black bars, left panel), along with the number lesion to recruit downstream proteins that will process of nuclei counted (green dots, right panel). Negative and Montanuy et al. Orphanet Journal of Rare Diseases (2020) 15:170 Page 3 of 9 Fig. 1 FA cell-based system setup. a Upon DNA damage causing ICLs on DNA, FA/BRCA pathway is activated, FANCD2 ubiquitinated, and fluorescent FANCD2 foci can be seen on the microscope (left). Deficient FA/BRCA pathway is not able to ubiquitinate FANCD2 and no FANCD2 foci can be detected (middle). A drug able to overcome FANCA deficiency on FA/BRCA pathway can be addressed with this cell-based system. FANCD2 foci can be clearly identified from the nucleus FANCD2 background signal (right). b Lack of FANCA expression of two positive clones after TALEN-mediated gene targeting. Cells were treated with MMC at 10 and 33 nM for 24 h, lyzed and FANCA expression and FANCD2 ubiquitination analyzed by Western blot. Vinculin was used as a loading control. c Survival assay of FANCA-deficient U2OS clones. 2F8 and 2G7 clones and WT cells were treated with a MMC dose curve from 1 to 100 nM. 72 h later cells were collected and counted. Graph shows mean survival percentage respect non-treated cells +/− SEM of at least three independent experiments with similar results. d 2F8 FANCA-deficient U2OS clone was stably transfected with YFP-FANCD2 (see materials and methods). FANCA-deficient cells (left lanes) and FANCA corrected cells (right lanes) were treated with 2 mM HU for 24 h, then lyzed and FANCA, endogenous FANCD2 and exogenous YFP-FANCD2 expression analyzed by Western blot. Vinculin was used as a loading control. e Fluorescence (green) of YFP-FANCD2 stably transfected on WT (left), FANCA-deficient (middle), and FANCA corrected (right) U2OS cell lines was analyzed by microscopy after treatment with 2 mM HU for 24 h. Bottom images show overlay of YFP-FANCD2 fluorescence with DAPI (nuclei staining) Montanuy et al. Orphanet Journal of Rare Diseases (2020) 15:170 Page 4 of 9 Fig. 2 High content screening setup. a FANCA-deficient (KO) and corrected cells stably transfected with YFP-FANCD2 were plated at different concentrations, left untreated or treated 24 h with 2 mM HU. Fluorescent images were obtained on Image Express Microscopy and nuclear counts or nuclear foci detected with CellProfiler 2.0 software (see materials and methods). Grey bars (left legend) show percentage of cells with 5 or more nuclear foci. Green dots (right legend) show nuclei counted from 6 average images captured for each well. Graph shows mean +/− SD of three replicates. b Library collections used in the screening. c Drug layout on 384 well plates used in the screening. Violet wells, 1 μM drug treatment, white wells, 10 μM. Green wells, positive controls (top six wells, FANCA-deficient cell line, bottom six wells, FANCA-corrected cell line). Grey wells that come from the plate edges where left empty to avoid sample evaporation. d Percentage of nuclei with 5 or more foci from control samples on each plate. e Representative graph from a 384 well plate with percentage of cells with foci (black bars, left legend) analyzed with CellProfiler 2.0 software, in parallel with nuclear count (green dots, right legend) to show potential drug toxicity positive controls (FANCA-deficient and corrected cells, and maintained cell number with respect the control (in- respectively) displayed a good signal respect the back- dicating no or low toxicity). Unfortunately, all the posi- ground (Fig. 2d), and a Z’ factor of 0.4 (see materials tive candidates once manually revised were false and methods). Although for non-cell based, high positives, as all the compounds displayed high autofluo- throughput assays, a Z’ factor above 0.5 is recom- rescence or different FANCD2 distribution, not compat- mended, cell-based assays are intrinsically more complex ible with nuclear foci (data not shown). These results in their read-out, and in these cases Z’ factor values of indicated that none of the drugs tested were able to re- 0.4 still provide a good signal window [11, 12]. cover FANCD2 foci in our FANCA-deficient cell-based system. We then searched for drugs that could have a Hit validation lower but detectable activity, by filtering out which drugs Once we performed the screening, collected the images could induce between 2 to 5% of nuclei with 5 or more and processed them, we selected compounds that in- FANCD2 foci. We selected 6 potential compounds that duced at least 5% nuclei with 5 or more FANCD2 foci, could induce FANCD2 foci in a small percentage of the Montanuy et al. Orphanet Journal of Rare Diseases (2020) 15:170 Page 5 of 9 FANCA-deficient cells: roflumilast, a phosphodiesterase- line used in the screening (Fig. 3a, b and data not 4 (PDE-4) inhibitor; 5-iodo-A-85380, an agonist for the shown). We also checked for FANCD2 monoubiquitina- α4β2 and α6β2 nicotinic acetylcholine receptors; ezeti- tion by Western blot, and as seen in Fig. 3c (and data mibe, a lipid-lowering compound; forskolin, a labdane not shown), neither of the candidates were able to res- diterpene used in research to increase cyclic AMP cue the lack of FANCD2 ubiquitination. Finally, we (cAMP); SANT-1, a sonic hedgehog signaling inhibitor; wanted to see if these molecules were able to reduce and AG555, an EGFR kinase inhibitor. In the follow-up genome instability in a FA-deficient cell line. FA cells ac- studies we additionally used 8-Bromo-cAMP, a bromi- cumulate genomic instability, both spontaneously and nated derivative of cAMP, that has a similar effect as for- induced by DNA-damaging agents such as hydroxyurea, skolin [13]. Unexpectedly, we were unable to validate mitomycin C or diepoxybutane. This genomic instability FANCD2 foci formation by these drugs in the same cell is used in diagnostics in chromosome fragility tests [14], Fig. 3 Hit validation. a FANCD2 foci detection from YFP-FANCD2 fluorescence from corrected (left lanes), FANCA-deficient (KO, middle lanes), and FANCA-deficient U2OS cell line, incubated with one of the hits (right lanes). Cells were treated with selected hits at 10 μM, 4–6 h later treated with 2 mM HU, and left 16–24 h. Cells were then fixed and immunostained with FANCD2 antibody (see materials and methods). Upper images show FANCD2 localization and bottom images show overlay of FANCD2 with DAPI. White arrows on left lanes show cells with FANCD2 foci. b Graph quantification of cells with 5 or more foci, from the experiment performed in a (and data not shown). Results show mean percentage of cells with foci +/− SEM from three independent experiments with similar results. c FANCA-deficient U2OS cells were treated with SANT-1 at 10 μM, and or MMC, for 24 h. Cells were lyzed and FANCD2 monoubiqutination analyzed by Western blot. Vinculin was used as a loading control. Similar results were obtained from the other hits testes (data not shown). d Spontaneous chromosomal fragility analysis in FANCA-deficient lymphoblastoid cell lines. Graphs show gates from cell cytometry experiments for micronuclei production (left) and G/M cell cycle arrest (right) in Wild type (WT, upper), FANCA-deficient (KO, middle) or FANCA-deficient lymphoblastoid cells stimulated with 10 μM 8-Bromo-AMPc (bottom). e Graph showing mean +/− SEM spontaneous micronuclei production after stimulation of selected hits (at 1 or 10 μM), from at least three independent experiments as in d Montanuy et al. Orphanet Journal of Rare Diseases (2020) 15:170 Page 6 of 9 and it has also been reported in micronucleus produc- as they did not fit the typical nuclei foci pattern. How- tion and single-cell gel electrophoresis (comet assay) ever, potential molecules with real activity but autofluor- studies [15, 16]. We adapted the micronucleus assay to escent would be discarded, a true limitation of our cell- cell cytometry (Figure Suppl. 2) [17], where we could based system.The 6 candidates with apparently mild ac- count thousands of events in a short period of time, hav- tivity detected in the HCS could not be later confirmed ing more sensitivity to detect mild changes. As seen in in validation studies, including FANCD2 foci formation Fig. 3d and e, in basal conditions FANCA-deficient lym- by immunohistochemistry, chromosome fragility and phoblastoid cells had close to two time more micronu- FANCD2 monoubiquitination. Nevertheless, we cannot cleus than wild type cells. When treated with candidate rule out the possibility of a FANCL-independent hits, the micronucleus levels in these cells were not sig- FANCD2 activity, as it has recently been reported that nificantly reduced. However, some drugs did have a mild FANCD2 chromatin recruiting is partially independent reduction effect, such as roflumilast, forskolin or SANT- of monoubiquitination [22]. Interestingly, we did see an 1, all three at 10 μM concentration (Fig. 3e). effect on micronuclei reduction in some candidates (roflumilast and forskolin, Fig. 3e), which would suggest Discussion these drugs may reduce chromosome instability, al- So far, drug screenings that target the FA/BRCA path- though this effect was mild. The potential applicability way published in the literature have been focused in of these results requires further validation and searching for inhibitors [18, 19]. Instead, our cell-based investigation. system was focused in the FA/BRCA-pathway recovery, Our cell-based system was designed with the objective in a FANCA-deficient background that yields FANCL to correct FA/BRCA pathway activity in the absence of ubiquitin ligase inactive, so FANCD2 cannot be monou- FANCA protein. However, there is the possibility that biquitinated nor localize to the nucleus in stalled replica- drug candidates could impact positively in the cellular tion forks. Thus, by looking for FANCD2 foci formation, phenotype, more than in the pathway itself, in FA- to our knowledge, this is the first report to screen for deficient cells. As an example, micronucleus formation, drugs with the aim to recover FA/BRCA pathway activity increased in buccal mucosa of FA-patients and blood in FA-deficient cells. cells in murine models [23, 24], is a general biomarker It is extremely difficult to conduct a complete drug of chromosome instability, and drugs able to reduce this discovery process, which normally takes 15 years or induction could be good therapeutic candidates. Another more, with highly elevated costs [20]. As the potential example could be monitoring other phenotypic changes economic return would be very limited in a rare disease, seen in FANCA-deficient cells, such as mitochondrial a drug repositioning strategy has been explored in some activity [25]. cases, such as the recent identification of the antifungal ciclopirox to treat congenital erythropoietic porphyria Conclusions [21]. In the FA field itself, repositioning approaches for We developed and validated a novel FA/BRCA-pathway BMF or DNA-damage prevention have actually been de- recovery cell-based assay to screen thousands of com- scribed, and positive results have led to clinical trials pounds in a HCS approach. Our results suggest that at such as with quercetin (NCT01720147 and the conditions tested, none of the FDA-approved drugs NCT03476330, www.clinicaltrials.gov) [6] and metfor- can rescue FA/BRCA pathway activity in the absence of min (NCT03398824) [7]. In our screening of 3802 com- FANCA. The same is true for other thousands of com- pounds we included 1200 FDA-approved drugs pounds known to reflect chemical diversity. However, (Prestwick library). However, we were not able to find even that we did not observe FANCD2 foci recovery, good candidates which recovered FANCD2 foci induc- other positive effects such as in reduction of genomic in- tion in our FA-deficient cell line. Thus, our results dis- stability, oxidative stress or apoptosis cannot be dis- card a therapeutic potential of FDA approved drugs to carded. That would be the case of some drugs, such as treat FA, in terms of upstream FA/BRCA-pathway activ- quercetin and metformin, recently described in the lit- ity. Additionally, around 5% of the drugs led to false erature and currently in clinical trials, or NAC, dasatinib positive hits, as some had a high autofluorescence, and and TGF-β inhibitors, also reported earlier [26–28]. others changed FANCD2 distribution, not similar to foci spots. An example was merbromin, which emits strong Methods autofluorescence (data not shown). These false positives Reagents could be a drawback of the system, taking into account Ethidium Monoazide Bromide (EMA, ref. e1374) Opti- that all molecules with green autofluorescence will be MEM (ref. 31,985–070), BCA Protein Quantification detected as positive hits. Fortunately, false positives in Assay (ref. 23,227), SYTOX® Green (ref. S7020) and our assay were quickly filtered-out by visual inspection, Hoechst® (ref. H1399) were from ThermoFisher Montanuy et al. Orphanet Journal of Rare Diseases (2020) 15:170 Page 7 of 9 Scientific. IGEPAL (ref. I8896), DAPI (ref. D9542), die- detected cells were sorted in a FACSJazz cell sorter (Bec- poxybutane (ref. 202533–16), hydroxyurea (ref. H8627– ton Dickinson) and individual cells seeded in 96 well 56), mitomycin C (ref. M-0503), β-mercaptoethanol (ref. plates. Grown clones were then tested by Western blot for M3148), triton X-100 (ref. T8787), BSA (ref. A7906), FANCA expression, and positive clones with no FANCA Complete Mini EDTA-free Protease Inhibitor Cocktail were sequenced to identify mutations/deletions. For fluor- (ref. 04693159001), PhosSTOP Phosphatase Inhibitor escent FANCD2 stable transfection, 5 × 105 U2OS WT or Cocktail Tablets (ref. 04906837001), roflumilast FANCA-deficient cells were seeded on 6 well plates and (SML1099), 5-iodo-A-85380 (SML0023) and 8-bromo- transfected with pEAK8-YFP-FANCD2 plasmid [30] with cAMP (B7880) were from Sigma-Aldrich. Ezetimibe lipofectamin reagent (Invitrogen), following manufac- (HY-17376) and forskolin (HY-15371) were from Med- turer’s instructions. Fluorescent cells were sorted, and se- chem Express. RNase A (ref. 12,091,021), Lipofectamine® lection was repeated two more times to maintain stable 2000 (ref. 11,668,019) and glycerol (ref. 15,514–011) fluorescence. To correct FANCA-deficient YFP-FANCD2 were from Invitrogen. RPMI (ref. L04495) and DMEM U2OS cells, a FANCA expressing retroviral plasmid was (ref. L0104) mediums, non-essential amino acids (ref. infected in these cells, and growth sensitivity checked by X0557), fetal bovine serum (FBS, ref. S181B), sodium 10 nM MMC treatment for 3–5 days. pyruvate (ref. L0642) and penicillin-streptomycin anti- biotic (ref. L0022) were from Biowest. Plasmocin (ref. ant-mpt) was from Invivogen. RIPA 10X lysis buffer (ref. Survival assay 20–188) was from Millipore. Benzonase nuclease (ref. For survival assays 1 × 105 cells were seeded in 6 well 70,746) was from Novagen. Antibodies rabbit anti- plates in duplicates, 24 h later were treated with a MMC FANCD2 (ab2187) and mouse anti-Vinculin (ab18058) dose curve from 1 to 100 nM and left in culture for 72 h. were from Abcam. Rabbit anti-FANCA (F3165) was Cells were then trypsinized and counted in a Z series from sigma. SANT-1 (sc-203,253), AG555 (sc203500), Coulter Counter (Beckman-Coulter). Goat anti-rabbit IgG-HRP (sc-2004) and goat anti- mouse IgG-HRP (sc-2005) secondary antibodies were from Santa Cruz. Goat anti-rabbit Alexa-Fluor488 was Western blot from ThermoFischer. Paraformaldehyde (ref. SC281692) 2 × 105 cells were seeded on 6 well plates, 24 h later were was from Santa Cruz, vectashield antifade vector (ref. H- treated with hydroxyurea, MMC or left untreated, and 1000) was from Vector Laboratories. 24 h later cells were lyzed and SDS-PAGE and blotting with indicated antibodies performed as previously de- Cell lines scribed [31]. WT and FANCA-deficient lymphoblastoid cell lines were cultured in RPMI medium with 20% FBS heat inac- tivated, penicillin-streptomycin, sodium pyruvate, nones- FANCD2 foci sential amino acids and β-Mercaptoethanol. U2OS cell 2 × 105 cells from YFP-FANCD2 stably transfected WT, lines (WT, U2OS-FANCAKO-YFP-FANCD2 and U2OS- FANCA-deficient or corrected U2OS cell lines were FANCAKO-YFP-FANCD2-FANCAcorrected) were cul- plated in coverslips in 6 well plates. After 4–6 h cells tured in DMEM medium with 10% FBS heat inactivated were treated with selected drugs or left untreated for 24 with plasmocin antibiotic. All cell lines were cultured at h. Cells were then treated with 2 mM hydroxyurea and 37 °C and 5% of CO2. FANCA gene was knocked-out in 24 h later were washed, fixed with 4% PFA in PBS for U2OS cell line by TALEN gene editing (Transposagen 15 min and permeabilized with 0.5% TX-100 solution in Biopharmaceuticals), a more specific alternative than PBS for 10 min, DAPI stained and foci imaged with CRISPR/CAS9 system, with fewer off-targets [10]. We fluorescence microscopy (Leica Microsystems). For im- used surrogate reporters for clone selection optimization, munofluorescence, YFP-FANCD2 stably transfected as provided by the manufacturer and described elsewhere FANCA-deficient or corrected cell lines were plated in [29]. Briefly, 3 × 105 cells where seeded on 6 well plates. coverslips in 6 well plates. After the experiment, fixed TALEN plasmids where designed to target EXON1 of and permeabilized cells were blocked for 15 min with 5% FANCA gene, being the target sequence 5′ TAGGCG BSA and 0.05% Tween-20 in PBS (blocking solution). CCAAGGCCATGT ccgactcgtgggtccc GAACTCCGCC Primary antibody was incubated in blocking solution TCGGGCCA 3′ (uppercase letters show forward and re- overnight at 4 °C in humidity. After washing, samples verse binding regions and the lowercase letters the cutting were incubated with secondary antibody in blocking so- region, see Supplementary Figure 1). Cells were trans- lution for 20 min at 37 °C. Samples were then washed fected with surrogate-FANCA and forward/reverse twice with blocking solution and deionized water, DAPI TALEN-FANCA plasmids. Once green fluorescence was stained and foci imaged by fluorescence microscopy. Montanuy et al. Orphanet Journal of Rare Diseases (2020) 15:170 Page 8 of 9 High content screening the basis of DNA content. Several gates were needed in High content screening of 3802 compounds was con- order to separate micronuclei from apoptotic chromatin ducted at Science for Life Laboratory, (Karolinska Insti- and debris (see Supplementary Figure 3). tutet, Stockholm, Sweden). Libraries used (chemical properties in Supplementary Figure 2) were: Prestwick li- Supplementary information brary of FDA approved drugs (1200 compounds); Selleck Supplementary information accompanies this paper at https://doi.org/10. 1186/s13023-020-01437-1. library, including FDA approved and 378 kinase inhibi- tors (1368 compounds); Tocris library (1119 com- Additional file 1: Supplementary Figure 1. FANCA exon 1 sequence pounds) and compounds from Thomas Helleday lab from U2OS clones 2F8 (A) and 2G7 (B). Forward and reverse primers are (115 compounds). These molecules were dissolved in shown in blue and yellow, respectively. Bold letters show first codon of the FANCA open reading frame in the exon (ATG). Violet filled boxes DMSO at 10 mM concentration, and dispensed to 384 show targeted TALEN sequence. Open black boxes show deleted regions well plates with Echo® 550 Liquid Handler (Labcyte) at of 2F8 (A, 7 bp) and 2G7 (B, 83 bp) clones. two different final concentrations (1 and 10 μM, see Fig. Additional file 2: Supplementary Figure 2. Library chemical 2c layout), sealed and stored at 4 °C. Next day, cells were properties. Average chemical properties from the 4 libraries used: A and seeded in 50 μl of medium using Multidrop™ Combi dis- B, AlogP (partition coefficient, measures the hydrophobicity/lipophilicity); C, HBA (number of hydrogen bond acceptors); D, HBD (number of penser (ThermoFisher Scientific) and left in the incuba- hydrogen bond donors); E, Heavy Atom Count (number of non-hydrogen tor. The following day cells were treated with 2 mM atoms); F, MW (molecular weight); G, RB (number of rotatable bonds). hydroxyurea for 24 h. Cells were then fixed with 4% PFA Additional file 3: Supplementary Figure 3. Cell cytometry gate in PBS, stained with 2 μ M Hoechst and stored in dark- analysis for micronuclei assay in FANCA-deficient LFA55 lymphoblastoid cell line. A) Forward scattering (FSC) versus side scattering (SSC). A gate is ness at 4 °C in PBS with 0.05% sodium azide. created to include events with 1/100 nuclei size (FSC) and 1/100 nuclei Images were taken with ImageXpress Micro XLS granularity (SSC). B) SYTOX histogram. In this gate events with less fluor- escence than 1/100 of the G1 population are discarded. C) SYTOX gate to microscope (Molecular Devices) through a VALet™ ro- plot width versus area to discard doublets and select single events. D) botic arm (ThermoFisher Scientific). Six images/well SYTOX vs SSC gate. Nuclei and events with a granularity and fluorescence were taken for nuclei and FANCD2 foci counting, with a 1/100 of the G1 are selected. E) SYTOX vs FSC. Nuclei and events with a size and fluorescence 1/100 of the G1 are selected. F) EMA vs DAPI. By 40x objective. For FANCD2 foci detection and nuclei discarding EMA positive events, data from apoptotic or necrotic cells is counting, a specific pipeline was prepared with CellProfi- removed. G) FSC vs SSC. After all gate selection, FSC vs SSC dot plot ler 2.0 software (Broad Institute). Using proper positive show a clear distinction between nuclei (upper right box) and micronu- clei (MN, lower left box). MN are defined as events between 1/100 to 1/ and negative controls, image analysis gave a Z’ factor of 10 fluorescence from that of 2 N nuclei. H) SYTOX histogram. Nuclei from 0.4. Z’ factor reflects the separation magnitude between G can be further analyzed to see cell cycle status (selected region shows the positive and the negative controls. After the initial 1/2 G2/M cells). hit identification, positive hits were manually checked to discard auto fluorescent compounds. Acknowledgments The authors want to thank the Chemical Biology Consortium Sweden (CBCS), and the Laboratories for Chemical Biology at Karolinska Institutet (LCBKI) for Micronucleus analysis by cell cytometry their help with the chemical screening. Chromosomal fragility and G2/M cell cycle analysis were carried out using WT and FANCA-deficient lympho- Authors’ contributions HM and JCP designed and performed the experiments. CC performed the blastoid cell lines and was previously described [17]. 6 × experiments. MAC designed FANCA TALEN-targeted plasmids. TH reviewed 104 cells were seeded in triplicates in 96 well plates, and the experiments and the manuscript. JM and JS designed and analyzed the 4–6 h after were treated with candidate hits at 1 or experiments and wrote the manuscript. The authors read and approved the final manuscript. 10 μM. 24 h later cells were treated with 0.1 μg/ml DEB. 72 h later cells were harvested and counted in order to Funding check for proper proliferation. Cells were then incubated Dr. Surrallés laboratory is supported by the ICREA-Academia program, the Spanish Ministry of Health (projects FANCOSTEM and FANCOLEN), the Span- with 25 μg/ml EMA (in PBS + 2% FBS) to stain necrotic ish Ministry of Economy and Competitiveness (projects CB06/07/0023 and and late-stage apoptotic cells. After 20 min of photoacti- SAF2015–64152-R), the European Commission (EUROFANCOLEN project vation at 4 °C, EMA excess was removed by adding PBS- HEALTH-F5–2012-305421 and P-SPHERE COFUND project), the Fanconi Anemia Research Fund Inc., and the “Fondo Europeo de Desarrollo Regional” 2% FBS. After EMA labelling, cells were exposed to lysis (FEDER). CIBERER is an initiative of the Instituto de Salud Carlos III, Spain. solution 1 (0.3 μ l/ml IGEPAL, 0,582 mg/ml NaCl, 1 mg/ ml RNAse A, 1 mg/ml sodium citrate, and 0.2 μM Availability of data and materials The datasets used and/or analyzed during the current study are available SYTOX Green). After one hour at room temperature, from the corresponding author on reasonable request. cells were diluted with lysis solution 2 (85.6 mg/ml su- crose, 15 mg/ml citric acid, 0.2 μM SYTOX Green). After Ethics approval and consent to participate The study design was approved by the appropriate ethics review board. 30 min at room temperature, samples were stored at 4 °C until its analysis 24 h later in a BD FACSCalibur Consent for publication flow cytometer. 20,000 live event cells were acquired on The authors give consent for publication of the manuscript. Montanuy et al. Orphanet Journal of Rare Diseases (2020) 15:170 Page 9 of 9 Competing interests cytogenetic damage and cytotoxicity. Mutat Res Toxicol Environ Mutagen. The authors declare no competing financial interests. 2007;630:78–91. 18. Chirnomas D, Taniguchi T, de la Vega M, Vaidya AP, Vasserman M, Hartman Author details A-R, et al. Chemosensitization to cisplatin by inhibitors of the Fanconi 1 Department of Genetics and Microbiology, Universitat Autònoma de anemia/BRCA pathway. Mol Cancer Ther. 2006;5:952–61. Barcelona, Barcelona, Spain. 2Join Research Unit on Genomic Medicine 19. Jacquemont C, Simon JA, D’Andrea AD, Taniguchi T. Non-specific chemical UAB-Sant Pau, Biomedical Research Institute, Hospital de la Santa Creu i Sant inhibition of the Fanconi anemia pathway sensitizes cancer cells to cisplatin. Pau, Barcelona, Spain. 3Centro de Investigación Biomédica en Red de Mol Cancer. 2012;11:26. Enfermedades raras, Barcelona, Spain. 4Division of Translational Medicine and 20. Pushpakom S, Iorio F, Eyers PA, Escott KJ, Hopper S, Wells A, et al. Drug Chemical Biology, Science for Life Laboratory, Department of Molecular repurposing: progress, challenges and recommendations. Nat Rev Drug Biochemistry and Biophysics, Karolinska Institutet, Stockholm, Sweden. Discov. 2019;18(1):41-58. https://doi.org/10.1038/nrd.2018.168. 5 Currently at Division of Genome Biology, Science for Life Laboratory, 21. Urquiza P, Laín A, Sanz-Parra A, Moreno J, Bernardo-Seisdedos G, Dubus P, Department of Molecular Biochemistry and Biophysics, Karolinska Institutet, et al. Repurposing ciclopirox as a pharmacological chaperone in a model of Stockholm, Sweden. 6Department of Biochemistry and Biophysics, SciLifelab, congenital erythropoietic porphyria. Sci Transl Med. 2018;10:eaat7467. Stockholm University, Stockholm, SE, Sweden. 7Institute of Medical and 22. Chen X, Bosques L, Sung P, Kupfer GM. A novel role for non-ubiquitinated Molecular Genetics and Skeletal dysplasia multidisciplinary Unit, Hospital FANCD2 in response to hydroxyurea-induced DNA damage. Oncogene. Universitario La Paz, Universidad Autónoma de Madrid, IdiPaz, Madrid, Spain. 2016;35:22–34. 8 Centro de Investigación Biomédica en Red de Enfermedades Raras, Madrid, 23. Ramírez MJ, Minguillón J, Loveless S, Lake K, Carrasco E, Stjepanovic N, et al. Spain. 9Genetics Department, Hospital de la Santa Creu i Sant Pau, Barcelona, Chromosome fragility in the buccal epithelium in patients with Fanconi Spain. anemia. Cancer Lett. 2020;472:1–7. 24. Montanuy H, Martínez-Barriocanal A, Casado JA, Rovirosa L, Ramírez MJ, Received: 28 January 2020 Accepted: 15 June 2020 Nieto R, et al. Gefitinib and afatinib show potential efficacy for Fanconi anemia-related head and neck cancer. Clin Cancer Res. 2020; clincanres. 2020;26(12):3044–57. https://doi.org/10.1158/1078-0432.CCR-19-1625. 25. Bottega R, Nicchia E, Cappelli E, Ravera S, De Rocco D, Faleschini M, et al. References Hypomorphic FANCA mutations correlate with mild mitochondrial and 1. Minguillón J, Surrallés J. Therapeutic research in the crystal chromosome clinical phenotype in Fanconi anemia. Haematologica. 2018;103:417–26. disease Fanconi anemia. Mutat Res Genet Toxicol Environ Mutagen. 2018; 26. Ponte F, Carvalho F, Porto B. Protective effect of acetyl-l-carnitine and α- 836(Pt A):104–8. https://doi.org/10.1016/j.mrgentox.2018.05.012. lipoic acid against the acute toxicity of diepoxybutane to human 2. Bogliolo M, Surrallés J. Fanconi anemia: a model disease for studies on lymphocytes. Toxicology. 2011;289:52–8. human genetics and advanced therapeutics. Curr Opin Genet Dev. 2015;33: 27. Anur P, Yates J, Garbati MR, Vanderwerf S, Keeble W, Rathbun K, et al. p38 32–40. MAPK inhibition suppresses the TLR-hypersensitive phenotype in FANCC- 3. Cappelli E, Ravera S, Vaccaro D, Cuccarolo P, Bartolucci M, Panfoli I, et al. and FANCA-deficient mononuclear phagocytes. Blood. 2012;119:1992–2002. Mitochondrial respiratory complex I defects in Fanconi anemia. Trends Mol 28. Zhang H, Kozono DE, O’Connor KW, Vidal-Cardenas S, Rousseau A, Hamilton Med. 2013;19:513–4. A, et al. TGF-β inhibition rescues hematopoietic stem cell defects and bone 4. Du W, Erden O, Pang Q. TNF-α signaling in Fanconi anemia. Blood Cells Mol marrow failure in Fanconi Anemia. Cell Stem Cell. 2016;18:668–81. Dis. 2014;52:2–11. 29. Kim H, Um E, Cho S-R, Jung C, Kim H, Kim J-S. Surrogate reporters for 5. Garaycoechea JI, Crossan GP, Langevin F, Daly M, Arends MJ, Patel KJ. enrichment of cells with nuclease-induced mutations. Nat Methods. 2011;8: Genotoxic consequences of endogenous aldehydes on mouse 941–3. haematopoietic stem cell function. Nature. 2012;489:571–5. 30. Pace P, Johnson M, Tan WM, Mosedale G, Sng C, Hoatlin M, et al. FANCE: 6. Li J, Sipple J, Maynard S, Mehta PA, Rose SR, Davies SM, et al. Fanconi the link between Fanconi anaemia complex assembly and activity. EMBO J. anemia links reactive oxygen species to insulin resistance and obesity. 2002;21:3414–23. Antioxidants Redox Signal. 2012;17:1083–98. 31. Hernández G, Ramírez MJ, Minguillón J, Quiles P, Ruiz De Garibay G, Aza- 7. Zhang Q-S, Tang W, Deater M, Phan N, Marcogliese AN, Li H, et al. Carmona M, et al. Decapping protein EDC4 regulates DNA repair and Metformin improves defective hematopoiesis and delays tumor formation phenocopies BRCA1. Nat Commun. 2018;9(1):967. in Fanconi anemia mice. Blood. 2016;128:2774–84. 8. Bellomo F, Medina DL, De Leo E, Panarella A, Emma F. High-content drug Publisher’s Note screening for rare diseases. J Inherit Metab Dis. 2017;40:601–7. Springer Nature remains neutral with regard to jurisdictional claims in 9. Meetei AR, de Winter JP, Medhurst AL, Wallisch M, Waisfisz Q, van de Vrugt published maps and institutional affiliations. HJ, et al. A novel ubiquitin ligase is deficient in Fanconi anemia. Nat Genet. 2003;35:165–70. 10. Boettcher M, McManus MT. Choosing the right tool for the job: RNAi, TALEN, or CRISPR. Mol Cell. 2015;58:575–85. 11. Birmingham A, Selfors LM, Forster T, Wrobel D, Kennedy CJ, Shanks E, et al. Statistical methods for analysis of high-throughput RNA interference screens. Nat Methods. 2009;6(8):569–75. https://doi.org/10.1038/nmeth.1351. 12. Zhang Z, Guan N, Li T, Mais DE, Wang M. Quality control of cell-based high- throughput drug screening. Acta Pharm Sin B. 2012;2:429–38. 13. Alasbahi RH, Melzig MF. Forskolin and derivatives as tools for studying the role of cAMP. Pharmazie. 2012;67:5–13. 14. Castella M, Pujol R, Callen E, Ramirez MJ, Casado JA, Talavera M, et al. Chromosome fragility in patients with Fanconi anaemia: diagnostic implications and clinical impact. J Med Genet. 2011;48:242–50. 15. Maluf SW, Erdtmann B. Genomic instability in Down syndrome and Fanconi anemia assessed by micronucleus analysis and single-cell gel electrophoresis. Cancer Genet Cytogenet. 2001;124:71–5. 16. Francies FZ, Wainwright R, Poole J, De Leeneer K, Coene I, Wieme G, et al. Diagnosis of Fanconi Anaemia by ionising radiation- or mitomycin C- induced micronuclei. DNA Repair (Amst). 2018;61:17–24. 17. Bryce SM, Bemis JC, Avlasevich SL, Dertinger SD. In vitro micronucleus assay scored by flow cytometry provides a comprehensive evaluation of